We trained two Random Forest (RF) machine learning models for cloud mask and cloud thermodynamic-phase detection using spectral observations from Visible Infrared Imaging Radiometer Suite (VIIRS) on board Suomi National Polar-orbiting Partnership (SNPP). Observations from Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) were carefully selected to provide reference labels. The two RF models were trained for all-day and daytime-only conditions using a 4-year collocated VIIRS and CALIOP dataset from 2013 to 2016. Due to the orbit difference, the collocated CALIOP and SNPP VIIRS training samples cover a broad-viewing zenith angle range, which is a great benefit to overall model performance. The all-day model uses three VIIRS infrared (IR) bands (8.6, 11, and 12 μm), and the daytime model uses five Near-IR (NIR) and Shortwave-IR (SWIR) bands (0.86, 1.24, 1.38, 1.64, and 2.25 μm) together with the three IR bands to detect clear, liquid water, and ice cloud pixels. Up to seven surface types, i.e., ocean water, forest, cropland, grassland, snow and ice, barren desert, and shrubland, were considered separately to enhance performance for both models. Detection of cloudy pixels and thermodynamic phase with the two RF models was compared against collocated CALIOP products from 2017. It is shown that, when using a conservative screening process that excludes the most challenging cloudy pixels for passive remote sensing, the two RF models have high accuracy rates in comparison to the CALIOP reference for both cloud detection and thermodynamic phase. Other existing SNPP VIIRS and Aqua MODIS cloud mask and phase products are also evaluated, with results showing that the two RF models and the MODIS MYD06 optical property phase product are the top three algorithms with respect to lidar observations during the daytime. During the nighttime, the RF all-day model works best for both cloud detection and phase, particularly for pixels over snow and ice surfaces. The present RF models can be extended to other similar passive instruments if training samples can be collected from CALIOP or other lidars. However, the quality of reference labels and potential sampling issues that may impact model performance would need further attention.
. © 2020 Copernicus GmbH. All rights reserved." "57217170189;","A Random Forest Approach to Identifying Young Stellar Object Candidates in the Lupus Star-forming Region",2020,"10.3847/1538-3881/ab72ac","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086570011&doi=10.3847%2f1538-3881%2fab72ac&partnerID=40&md5=1f2b63174bee7e3cee3c539749ea04bb","The identification and characterization of stellar members within a star-forming region are critical to many aspects of star formation, including formalization of the initial mass function, circumstellar disk evolution, and star formation history. Previous surveys of the Lupus star-forming region have identified members through infrared excess and accretion signatures. We use machine learning to identify new candidate members of Lupus based on surveys from two space-based observatories: ESA's Gaia and NASA's Spitzer. Astrometric measurements from Gaia's Data Release 2 and astrometric and photometric data from the Infrared Array Camera on the Spitzer Space Telescope, as well as from other surveys, are compiled into a catalog for the random forest (RF) classifier. The RF classifiers are tested to find the best features, membership list, non-membership identification scheme, imputation method, training set class weighting, and method of dealing with class imbalance within the data. We list 27 candidate members of the Lupus star-forming region for spectroscopic follow-up. Most of the candidates lie in Clouds V and VI, where only one confirmed member of Lupus was previously known. These clouds likely represent a slightly older population of star formation. © 2020. The American Astronomical Society. All rights reserved.." "57190382179;57201377332;6701404111;6701363691;","Land use/land cover mapping using multitemporal sentinel-2 imagery and four classification methods-A case study from Dak Nong, Vietnam",2020,"10.3390/RS12091367","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085246606&doi=10.3390%2fRS12091367&partnerID=40&md5=1ec5552140b0d12a4853ca695175c8f1","Information on land use and land cover (LULC) including forest cover is important for the development of strategies for land planning and management. Satellite remotely sensed data of varying resolutions have been an unmatched source of such information that can be used to produce estimates with a greater degree of confidence than traditional inventory estimates. However, use of these data has always been a challenge in tropical regions owing to the complexity of the biophysical environment, clouds, and haze, and atmospheric moisture content, all of which impede accurate LULC classification. We tested a parametric classifier (logistic regression) and three non-parametric machine learning classifiers (improved k-nearest neighbors, random forests, and support vector machine) for classification of multi-temporal Sentinel 2 satellite imagery into LULC categories in Dak Nong province, Vietnam. A total of 446 images, 235 from the year 2017 and 211 from the year 2018, were pre-processed to gain high quality images for mapping LULC in the 6516 km2 study area. The Sentinel 2 images were tested and classified separately for four temporal periods: (i) dry season, (ii) rainy season, (iii) the entirety of the year 2017, and (iv) the combination of dry and rainy seasons. Eleven different LULC classes were discriminated of which five were forest classes. For each combination of temporal image set and classifier, a confusion matrix was constructed using independent reference data and pixel classifications, and the area on the ground of each class was estimated. For overall temporal periods and classifiers, overall accuracy ranged from 63.9% to 80.3%, and the Kappa coefficient ranged from 0.611 to 0.813. Area estimates for individual classes ranged from 70 km2 (1% of the study area) to 2200 km2 (34% of the study area) with greater uncertainties for smaller classes. © 2020 by the authors." "57215000725;23012792800;6507322779;8839237600;","Near-daily discharge estimation in high latitudes from Sentinel-1 and 2: A case study for the Icelandic Þjórsá river",2020,"10.1016/j.rse.2020.111684","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079616729&doi=10.1016%2fj.rse.2020.111684&partnerID=40&md5=a975d646f6bb71c849a37353935da14f","Climate change is a threat to many high-latitude regions. Changing patterns in precipitation intensity and increasing glacial ablation during spring and summer have major influence on river dynamics and the risk of widespread flooding. To monitor these rapid events, more frequent discharge observations are necessary. Having access to near-daily satellite based discharge observations is therefore highly beneficial. In this context, the recently launched Sentinel-1 and 2 satellites promise unprecedented potential, due to their capacity to obtain radar and optical data at high spatial (10 m) and high temporal (1–3 days) resolutions. Here, we use both missions to provide a novel approach to estimate the discharge of the Þjórsá (Thjórsá) river, Iceland, on a near-daily basis. Iceland, and many other high-latitude regions, are affected by frequent cloud-cover, limiting the availability of cloud-free optical Sentinel-2 data. We trained a Random Forest supervised machine learning classifier with a set of Sentinel-1 backscatter metrics to classify water in the individual Sentinel-1 images. A Sentinel-2 based classification mask was created to improve the classification results. Second, we derived the river surface area and converted it to the effective width, which we used to estimate the discharge using an at-a-station hydraulic geometry (AHG) rating curve. We trained the rating curve for a six-month training period using in situ discharge observations and assessed the effect of training area selection. We used the trained rating curve to estimate discharge for a one-year monitoring period between 2017/10 and 2018/10. Results showed a Kling-Gupta Efficiency (KGE) of 0.831, indicating the usefulness of dense Sentinel-1 and 2 observations for accurate discharge estimations of a medium-sized (200 m width) high-latitude river on a near-daily basis (1.56 days on average). We demonstrated that satellite based discharge products can be a valuable addition to in situ discharge observations, also during ice-jam events. © 2020 Elsevier Inc." "57191430389;56210720700;57204436627;57211576350;57203553986;26026749200;7003361863;7403247998;7101846027;16308514000;","Stratocumulus cloud clearings: Statistics from satellites, reanalysis models, and airborne measurements",2020,"10.5194/acp-20-4637-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083902538&doi=10.5194%2facp-20-4637-2020&partnerID=40&md5=1c943dc4c0d1800f2336d7e1389e7519","This study provides a detailed characterization of stratocumulus clearings off the US West Coast using remote sensing, reanalysis, and airborne in situ data. Ten years (2009 2018) of Geostationary Operational Environmental Satellite (GOES) imagery data are used to quantify the monthly frequency, growth rate of total area (GRArea), and dimensional characteristics of 306 total clearings. While there is interannual variability, the summer (winter) months experienced the most (least) clearing events, with the lowest cloud fractions being in close proximity to coastal topographical features along the central to northern coast of California, including especially just south of Cape Mendocino and Cape Blanco. From 09:00 to 18:00 (PST), the median length, width, and area of clearings increased from 680 to 1231, 193 to 443, and ~ 67000 to ~ 250000 km2, respectively. Machine learning was applied to identify the most influential factors governing the GRArea of clearings between 09:00 and 12:00 PST, which is the time frame of most rapid clearing expansion. The results from gradient-boosted regression tree (GBRT) modeling revealed that air temperature at 850 hPa (T850), specific humidity at 950 hPa (q950), sea surface temperature (SST), and anomaly in mean sea level pressure (MSLPanom) were probably most impactful in enhancing GRArea using two scoring schemes. Clearings have distinguishing features such as an enhanced Pacific high shifted more towards northern California, offshore air that is warm and dry, stronger coastal surface winds, enhanced lower-tropospheric static stability, and increased subsidence. Although clearings are associated obviously with reduced cloud fraction where they reside, the domain-averaged cloud albedo was actually slightly higher on clearing days as compared to non-clearing days. To validate speculated processes linking environmental parameters to clearing growth rates based on satellite and reanalysis data, airborne data from three case flights were examined. Measurements were compared on both sides of the clear cloudy border of clearings at multiple altitudes in the boundary layer and free troposphere, with results helping to support links suggested by this study s model simulations. More specifically, airborne data revealed the influence of the coastal low-level jet and extensive horizontal shear at cloud-relevant altitudes that promoted mixing between clear and cloudy air. Vertical profile data provide support for warm and dry air in the free troposphere, additionally promoting expansion of clearings. Airborne data revealed greater evidence of sea salt in clouds on clearing days, pointing to a possible role for, or simply the presence of, this aerosol type in clearing areas coincident with stronger coastal winds. © 2020 Copernicus GmbH. All rights reserved." "41762272700;57195920156;56707193700;57211466726;","Introduction of digital technologies in the enterprise",2020,"10.1051/e3sconf/202015904004","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084044657&doi=10.1051%2fe3sconf%2f202015904004&partnerID=40&md5=64d5622265e329e6ef88a0c8f25c3b10","Digitalization of an enterprise is necessary to increase the efficiency and sustainability of its functioning through dramatic changes in the quality of management, both technological processes and decision-making processes at all levels of management, based on modern methods of production and further use of information about the state and prediction of possible changes in managed elements and subsystems. For the purposes of designing a digital transformation of an enterprise, it is necessary to develop a classification of digital technologies according to the criterion of accessibility and expediency of their implementation in the enterprise. Thus, key digital technologies are grouped into three groups: basic technologies are technologies without which digital transformation of an enterprise is impossible (cloud technologies, wireless technologies, paperless technologies, etc.); critical technologies are technologies that provide a complete digital transformation of the enterprise (big data, cloud computing, unmanned technologies, etc.); breakthrough technologies - technologies that realize the transition from analog to a digital enterprise (artificial intelligence, neural networks, distributed data registry, machine learning, etc.) © The Authors, published by EDP Sciences, 2020." "55850401500;56422261200;16038426300;6602762273;","Riparian trees genera identification based on leaf-on/leaf-off airborne laser scanner data and machine learning classifiers in northern France",2020,"10.1080/01431161.2019.1674457","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074007989&doi=10.1080%2f01431161.2019.1674457&partnerID=40&md5=a21f3aaf8a3ec2c764e7021d5b6bc1b3","Riparian forests are valuable environments delivering multiples ecological services. Because they face both natural and anthropogenic constraints, riparian forests need to be accurately mapped in terms of genera/species diversity. Previous studies have shown that the Airborne Laser Scanner (ALS) data have the potential to classify trees in different contexts. However, an assessment of important features and classification results for broadleaved deciduous riparian forests mapping using ALS remains to be achieved. The objective of this study was to estimate which features derived from ALS data were important for describing trees genera from a riparian deciduous forest, and provide results of classifications using two Machine Learning algorithms. The procedure was applied to 191 trees distributed in eight genera located along the Sélune river in Normandy, northern France. ALS data from two surveys, in the summer and winter, were used. From these data, trees crowns were extracted and global morphology and internal structure features were computed from the 3D points clouds. Five datasets were established, containing for each one an increasing number of genera. This was implemented in order to assess the level of discrimination between trees genera. The most discriminant features were selected using a stepwise Quadratic Discriminant Analysis (sQDA) and Random Forest, allowing the number of features to be reduced from 144 to 3–9, depending on the datasets. The sQDA-selected features highlighted the fact that, with an increasing number of genera in the datasets, internal structure became more discriminant. The selected features were used as variables for classification using Support Vector Machine (SVM) and Random Forest (RF) algorithms. Additionally, Random Forest classifications were conducted using all features computed, without selection. The best classification performances showed that using the sQDA-selected features with SVM produced accuracy ranging from 83.15% when using three genera (Oak, Alder and Poplar). A similar result was obtained using RF and all features available for classification. The latter also achieved the best classification performances when using seven and eight genera. The results highlight that ML algorithms are suitable methods to map riparian trees. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group." "57218168269;24460797900;7005403704;25225898300;57188835107;8602636100;7003877068;36675777000;57217677002;35185338100;55790076000;57218170381;","Physical properties of the star-forming clusters in NGC 6334",2020,"10.1051/0004-6361/201935699","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086125242&doi=10.1051%2f0004-6361%2f201935699&partnerID=40&md5=d1e1f98aa3e388d9479cbe3a4564e724","Aims. We aim to characterise certain physical properties of high-mass star-forming sites in the NGC 6334 molecular cloud, such as the core mass function (CMF), spatial distribution of cores, and mass segregation. Methods. We used the Atacama Large Millimeter/sub-millimeter Array (ALMA) to image the embedded clusters NGC 6334-I and NGC 6334-I(N) in the continuum emission at 87.6 GHz. We achieved a spatial resolution of 1300 au, enough to resolve different compact cores and fragments, and to study the properties of the clusters. Results. We detected 142 compact sources distributed over the whole surveyed area. The ALMA compact sources are clustered in different regions. We used different machine-learning algorithms to identify four main clusters: NGC 6334-I, NGC 6334-I(N), NGC 6334-I(NW), and NGC 6334-E. The typical separations between cluster members range from 4000 au to 12 000 au. These separations, together with the core masses (0.1-100 M· ), are in agreement with the fragmentation being controlled by turbulence at scales of 0.1 pc. We find that the CMFs show an apparent excess of high-mass cores compared to the stellar initial mass function. We evaluated the effects of temperature and unresolved multiplicity on the derived slope of the CMF. Based on this, we conclude that the excess of high-mass cores might be spurious and due to inaccurate temperature determinations and/or resolution limitations. We searched for evidence of mass segregation in the clusters and we find that clusters NGC 6334-I and NGC 6334-I(N) show hints of segregation with the most massive cores located in the centre of the clusters. Conclusions. We searched for correlations between the physical properties of the four embedded clusters and their evolutionary stage (based on the presence of H II regions and infrared sources). NGC 6334-E appears as the most evolved cluster, already harbouring a well-developed H II region. NGC 6334-I is the second-most evolved cluster with an ultra-compact H II region. NGC 6334-I(N) contains the largest population of dust cores distributed in two filamentary structures and no dominant H II region. Finally, NGC 6334-I(NW) is a cluster of mainly low-mass dust cores with no clear signs of massive cores or H II regions. We find a larger separation between cluster members in the more evolved clusters favouring the role of gas expulsion and stellar ejection with evolution. The mass segregation, seen in the NGC 6334-I and NGC 6334-I(N) clusters, suggests a primordial origin for NGC 6334-I(N). In contrast, the segregation in NGC 6334-I might be due to dynamical effects. Finally, the lack of massive cores in the most evolved cluster suggests that the gas reservoir is already exhausted, while the less evolved clusters still have a large gas reservoir along with the presence of massive cores. In general, the fragmentation process of NGC 6334 at large scales (from filament to clump, i.e. at about 1 pc) is likely governed by turbulent pressure, while at smaller scales (scale of cores and sub-fragments, i.e. a few hundred au) thermal pressure starts to be more significant. © 2020 ESO." "56422909100;57205183542;57201261818;57211978863;57211983696;57218174561;57210827853;8632797000;","An updated moving window algorithm for hourly-scale satellite precipitation downscaling: A case study in the Southeast Coast of China",2020,"10.1016/j.jhydrol.2019.124378","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075561162&doi=10.1016%2fj.jhydrol.2019.124378&partnerID=40&md5=0a86827ffb33c11cd0c3d7c561c56b16","Accurate gridded precipitation products with both finer tempo-spatial resolutions are critical for various scientific communities (e.g., hydrology, meteorology, climatology, and agriculture). Downscaling on coarse satellite-based rainfall estimates is an optimal approach to obtain such datasets. The Integrated Multi-satellitE Retrivals for Global Precipitation Measurement (GPM) (IMERG) data provides the “best” satellite-based precipitation estimates at half-hourly/0.1° scales, while its spatial resolution is still coarse for certain hydrometeorology research. To acquire hourly downscaled precipitation estimates based on IMERG, there are two great challenges: (1) limited rainfall-related environmental variables (0.01°×0.01°, hourly) used to downscale the IMERG data; and (2) far few rainfall pixels used for regressing traditional relationships between precipitation and environmental variables. In this case, most traditional or commonly-used regression/empirical models and the state-of-art machine learning algorithms are not suitable to cater these requirements. Therefore, we proposed a new strategy to obtain hourly downscaled precipitation estimates based on IMERG called Geographically Moving Window Weight Disaggregation Analysis (GMWWDA). Additionally, we explored multiple cloud properties, including cloud effective radius (CER), cloud top height (CTH), cloud top temperature (CTT) and cloud optical thickness (COT), as covariates to downscale IMERG using GMWWDA method, and concluded as follows: (1) the downscaled results (0.01° × 0.01°, hourly) based on the above mentioned cloud properties outperformed the original IMERG data; (2) the downscaled results based on CER performed better than those based on CTH, CTT and COT, respectively; (3) the accuracy of satellite-based precipitation products pose significant effects on those of the downscaled results. This study provides a great potential solution for generating satellite-based precipitation dataset with finer spatio-temporal resolutions. © 2019" "57199457111;57209342660;57202461555;57203961945;56421194700;55455270400;55574709900;57190952440;57216593098;56159151300;36462695500;7402170368;","Estimating forest stock volume in Hunan Province, China, by integrating in situ plot data, Sentinel-2 images, and linear and machine learning regression models",2020,"10.3390/RS12010186","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083997153&doi=10.3390%2fRS12010186&partnerID=40&md5=14d9db74b5883060d3c9a5e1e5121a0a","The forest stock volume (FSV) is one of the key indicators in forestry resource assessments on local, regional, and national scales. To date, scaling up in situ plot-scale measurements across landscapes is still a great challenge in the estimation of FSVs. In this study, Sentinel-2 imagery, the Google Earth Engine (GEE) cloud computing platform, three base station joint differential positioning technology (TBSJDPT), and three algorithms were used to build an FSV model for forests located in Hunan Province, southern China. The GEE cloud computing platform was used to extract the imagery variables from the Sentinel-2 imagery pixels. The TBSJDPT was put forward and used to provide high-precision positions of the sample plot data. The random forests (RF), support vector regression (SVR), and multiple linear regression (MLR) algorithms were used to estimate the FSV. For each pixel, 24 variables were extracted from the Sentinel-2 images taken in 2017 and 2018. The RF model performed the best in both the training phase (i.e., R2 = 0.91, RMSE = 35.13 m3 ha-1, n = 321) and in the test phase (i.e., R2 = 0.58, RMSE = 65.03 m3 ha-1, and n = 138). This model was followed by the SVR model (R2 = 0.54, RMSE = 65.60 m3 ha-1, n = 321 in training; R2 = 0.54, RMSE = 66.00 m3 ha-1, n = 138 in testing), which was slightly better than the MLR model (R2 = 0.38, RMSE = 75.74 m3 ha-1, and n = 321 in training; R2 = 0.49, RMSE = 70.22 m3 ha-1, and n = 138 in testing) in both the training phase and test phase. The best predictive band was Red-Edge 1 (B5), which performed well both in the machine learning methods and in the MLR method. The Blue band (B2), Green band (B3), Red band (B4), SWIR2 band (B12), and vegetation indices (TCW, NDVI_B5, and TCB) were used in the machine learning models, and only one vegetation index (MSI) was used in the MLR model. We mapped the FSV distribution in Hunan Province (3.50 x 108 m3) based on the RF model; it reached a total accuracy of 63.87% compared with the ocial forest report in 2017 (5.48 x 108 m3). The results from this study will help develop and improve satellite-based methods to estimate FSVs on local, regional and national scales. © 2020 by the authors." "55353441700;33768158800;6505760587;","Fusion of sentinel-1 with official topographic and cadastral geodata for crop-type enriched LULC mapping using FOSS and open data",2020,"10.3390/ijgi9020120","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85081254350&doi=10.3390%2fijgi9020120&partnerID=40&md5=3f2f8ca6de52c7c1b29a62203c5d9a3e","Accurate crop-type maps are urgently needed as input data for various applications, leading to improved planning and more sustainable use of resources. Satellite remote sensing is the optimal tool to provide such data. Images from Synthetic Aperture Radar (SAR) satellite sensors are preferably used as they work regardless of cloud coverage during image acquisition. However, processing of SAR is more complicated and the sensors have development potential. Dealing with such a complexity, current studies should aim to be reproducible, open, and built upon free and open-source software (FOSS). Thereby, the data can be reused to develop and validate new algorithms or improve the ones already in use. This paper presents a case study of crop classification from microwave remote sensing, relying on open data and open software only. We used 70 multitemporal microwave remote sensing images from the Sentinel-1 satellite. A high-resolution, high-precision digital elevation model (DEM) assisted the preprocessing. The multi-data approach (MDA) was used as a framework enabling to demonstrate the benefits of including external cadastral data. It was used to identify the agricultural area prior to the classification and to create land use/land cover (LULC) maps which also include the annually changing crop types that are usually missing in official geodata. All the software used in this study is open-source, such as the Sentinel Application Toolbox (SNAP), Orfeo Toolbox, R, and QGIS. The produced geodata, all input data, and several intermediate data are openly shared in a research database. Validation using an independent validation dataset showed a high overall accuracy of 96.7% with differentiation into 11 different crop-classes. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/)." "55939190800;","A robust deep learning approach for spatiotemporal estimation of Satellite AOD and PM2.5",2020,"10.3390/rs12020264","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85081079391&doi=10.3390%2frs12020264&partnerID=40&md5=313d1c0e312093219c85c580204b81a4","Accurate estimation of fine particulate matter with diameter =2.5 μm (PM2.5) at a high spatiotemporal resolution is crucial for the evaluation of its health effects. Previous studies face multiple challenges including limited ground measurements and availability of spatiotemporal covariates. Although the multiangle implementation of atmospheric correction (MAIAC) retrieves satellite aerosol optical depth (AOD) at a high spatiotemporal resolution, massive non-random missingness considerably limits its application in PM2.5 estimation. Here, a deep learning approach, i.e., bootstrap aggregating (bagging) of autoencoder-based residual deep networks, was developed to make robust imputation of MAIAC AOD and further estimate PM2.5 at a high spatial (1 km) and temporal (daily) resolution. The base model consisted of autoencoder-based residual networks where residual connections were introduced to improve learning performance. Bagging of residual networks was used to generate ensemble predictions for better accuracy and uncertainty estimates. As a case study, the proposed approach was applied to impute daily satellite AOD and subsequently estimate daily PM2.5 in the Jing-Jin-Ji metropolitan region of China in 2015. The presented approach achieved competitive performance in AOD imputation (mean test R2: 0.96; mean test RMSE: 0.06) and PM2.5 estimation (test R2: 0.90; test RMSE: 22.3 μg/m3). In the additional independent tests using ground AERONET AOD and PM2.5 measurements at the monitoring station of the U.S. Embassy in Beijing, this approach achieved high R2 (0.82-0.97). Compared with the state-of-the-art machine learning method, XGBoost, the proposed approach generated more reasonable spatial variation for predicted PM2.5 surfaces. Publically available covariates used included meteorology, MERRA2 PBLH and AOD, coordinates, and elevation. Other covariates such as cloud fractions or land-use were not used due to unavailability. The results of validation and independent testing demonstrate the usefulness of the proposed approach in exposure assessment of PM2.5 using satellite AOD having massive missing values. © 2020 by the authors." "55578822100;15844957700;56337405600;","Next-Generation Artificial Intelligence Techniques for Satellite Data Processing",2020,"10.1007/978-3-030-24178-0_11","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075504732&doi=10.1007%2f978-3-030-24178-0_11&partnerID=40&md5=257bee2f35e72fcc76f0da6861d4b5df","In this chapter, we have tried to cover majority of the artificial intelligence (AI) techniques that has contributed to the remote sensing community in the form of satellite data processing, right from the basics to advanced level. A wide variety of applications and enormous amount of satellite data growing exponentially has critical demands in speedup, cost cutting, and automation in its processing while maintaining the accuracy. We have started with the need of AI techniques and evolution made for revolutionary changes in remote sensing and other areas. Subsequently, the traditional ML techniques and its limitations, advancements, and need of introducing DL in various applications are reviewed with what is the present requisites and expectation from AI community to overcome the issues and meet the upraised demands by emerging applications. We concluded that ML and DL technology should integrate with big data technologies and cloud computing to meet the future needs. © 2020, Springer Nature Switzerland AG." "35247685500;","Comparison of multi-seasonal Landsat 8, Sentinel-2 and hyperspectral images for mapping forest alliances in Northern California",2020,"10.1016/j.isprsjprs.2019.11.007","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074895939&doi=10.1016%2fj.isprsjprs.2019.11.007&partnerID=40&md5=4455cb8c3ff6720b82ff084131bc1073","The current era of earth observation now provides constellations of open-access, multispectral satellite imagery with medium spatial resolution, greatly increasing the frequency of cloud-free data for analysis. The Landsat satellites have a long historical record, while the newer Sentinel-2 (S2) satellites offer higher temporal, spatial and spectral resolution. The goal of this study was to evaluate the relative benefits of single- and multi-seasonal multispectral satellite data for discriminating detailed forest alliances, as defined by the U.S. National Vegetation Classification system, in a Mediterranean-climate landscape (Sonoma County, California). Results were compared to a companion analysis of simulated hyperspectral satellite data (HyspIRI) for the same study site and reference data (Clark et al., 2018). Experiments used real and simulated S2 and Landsat 8 (L8) data. Simulated S2 and L8 were from HyspIRI images, thereby focusing results on differences in spectral resolution rather than other confounding factors. The Support Vector Machine (SVM) classifier was used in a hierarchical classification of land-cover (Level 1), followed by alliances (Level 2) in forest pixels, and included summer-only and multi-seasonal sets of predictor variables (bands, indices and bands plus indices). Both real and simulated multi-seasonal multispectral variables significantly improved overall accuracy (OA) by 0.2–1.6% for Level 1 tree/no tree classifications and 3.6–25.8% for Level 2 forest alliances. Classifiers with S2 variables tended to be more accurate than L8 variables, particularly for S2, which had 0.4–2.1% and 5.1–11.8% significantly higher OA than L8 for Level 1 tree/no tree and Level 2 forest alliances, respectively. Combining multispectral bands and indices or using just bands was generally more accurate than relying on just indices for classification. Simulated HyspIRI variables from past research had significantly greater accuracy than real L8 and S2 variables, with an average OA increase of 8.2–12.6%. A final alliance-level map used for a deeper analysis used simulated multi-seasonal S2 bands and indices, which had an overall accuracy of 74.3% (Kappa = 0.70). The accuracy of this classification was only 1.6% significantly lower than the best HyspIRI-based classification, which used multi-seasonal metrics (Clark et al., 2018), and there were alliances where the S2-based classifier was more accurate. Within the context of these analyses and study area, S2 spectral-temporal data demonstrated a strong capability for mapping global forest alliances, or similar detailed floristic associations, at medium spatial resolutions (10–30 m). © 2019 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)" "57209285216;56278282600;57209616506;","Verification of two machine learning approaches for cloud masking based on reflectance of channel IR3.9 using Meteosat Second Generation over Middle East maritime",2019,"10.1080/01431161.2019.1624863","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85068262019&doi=10.1080%2f01431161.2019.1624863&partnerID=40&md5=e3a268082747fad192facc6fe811c519","Most cold channels of Meteosat Second Generation (MSG) satellites can distinguish between the sea and ice cloud tops, except for the IR3.9 channel because of the close reflectance and radiance values of the IR3.9 channel for maritime, low-level cloud and ice cloud tops. In this article, we introduce and evaluate two machine learning methods for cloud masking of Spinning Enhanced Visible and Infrared Imager (SEVIRI) images in the day and night that use the reflectance value of the IR3.9 channel. We reached a good correlation by comparing the results of the modelled cloud masking of Meteosat satellite images with MODIS (Moderate Resolution Imaging Spectroradiometer) and CLM (Cloud Mask product of EUMETSAT) images in a way that the coefficient of determination (R2) value was 92.34%, 89.91% and 83.69%, 78.23% in the cold season and 90.17%, 87.09% and 80.37%, 76.48% in the warm season, respectively, using the CHAID (chi-squared automatic interaction detection) decision tree and RBF (radial basis function) neural network approaches. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group." "57211316093;55574230638;57208802674;","Convective/stratiform precipitation classification using ground-based doppler radar data based on the K-nearest neighbor algorithm",2019,"10.3390/rs11192277","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85073475434&doi=10.3390%2frs11192277&partnerID=40&md5=4c7521e7799af47afa88e35a7d48be73","Stratiform and convective rain types are associated with different cloud physical processes, vertical structures, thermodynamic influences and precipitation types. Distinguishing convective and stratiform systems is beneficial to meteorology research and weather forecasting. However, there is no clear boundary between stratiform and convective precipitation. In this study, a machine learning algorithm, K-nearest neighbor (KNN), is used to classify precipitation types. Six Doppler radar (WSR-98D/SA) data sets from Jiangsu, Guangzhou and Anhui Provinces in China were used as training and classification samples, and the 2A23 product of the Tropical Precipitation Measurement Mission (TRMM) was used to obtain the training labels and evaluate the classification performance. Classifying precipitation types using KNN requires three steps. First, features are selected from the radar data by comparing the range of each variable for different precipitation types. Second, the same unclassified samples are classified with different k values to choose the best-performing k. Finally, the unclassified samples are put into the KNN algorithm with the best k to classify precipitation types, and the classification performance is evaluated. Three types of cases, squall line, embedded convective and stratiform cases, are classified by KNN. The KNN method can accurately classify the location and area of stratiform and convective systems. For stratiform classifications, KNN has a 95% probability of detection, 8% false alarm rate, and 87% cumulative success index; for convective classifications, KNN yields a 78% probability of detection, a 13% false alarm rate, and a 69% cumulative success index. These results imply that KNN can correctly classify almost all stratiform precipitation and most convective precipitation types. This result suggests that KNN has great potential in classifying precipitation types. © 2019 by the authors." "57195412150;57205353363;7401526171;55331455800;7005052907;57208765667;6701549596;35509463200;56012593900;","Conditional generative adversarial networks (cGANs) for near real-time precipitation estimation from multispectral GOES-16 satellite imageries-PERSIANN-cGAN",2019,"10.3390/rs11192193","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85073408806&doi=10.3390%2frs11192193&partnerID=40&md5=ee563f287877925796524d0b1da5cda2","In this paper, we present a state-of-the-art precipitation estimation framework which leverages advances in satellite remote sensing as well as Deep Learning (DL). The framework takes advantage of the improvements in spatial, spectral and temporal resolutions of the Advanced Baseline Imager (ABI) onboard the GOES-16 platform along with elevation information to improve the precipitation estimates. The procedure begins by first deriving a Rain/No Rain (R/NR) binary mask through classification of the pixels and then applying regression to estimate the amount of rainfall for rainy pixels. A Fully Convolutional Network is used as a regressor to predict precipitation estimates. The network is trained using the non-saturating conditional Generative Adversarial Network (cGAN) and Mean Squared Error (MSE) loss terms to generate results that better learn the complex distribution of precipitation in the observed data. Common verification metrics such as Probability Of Detection (POD), False Alarm Ratio (FAR), Critical Success Index (CSI), Bias, Correlation and MSE are used to evaluate the accuracy of both R/NR classification and real-valued precipitation estimates. Statistics and visualizations of the evaluation measures show improvements in the precipitation retrieval accuracy in the proposed framework compared to the baseline models trained using conventional MSE loss terms. This framework is proposed as an augmentation for PERSIANN-CCS (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Network- Cloud Classification System) algorithm for estimating global precipitation. © 2019 by the authors." "57148128800;6602369340;7004959591;6602211600;","USING NEURAL NETWORKS to DETECT OBJECTS in MLS POINT CLOUDS BASED on LOCAL POINT NEIGHBORHOODS",2019,"10.5194/isprs-annals-IV-2-W7-17-2019","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084637582&doi=10.5194%2fisprs-annals-IV-2-W7-17-2019&partnerID=40&md5=f574691bf7a6115ddbcce0ef6a734c04","This paper presents an approach which uses a PointNet-like neural network to detect objects of certain types in MLS point clouds. In our case, it is used for the detection of pedestrians, but the approach can easily be adapted to other object classes. In the first step, we process local point neighborhoods with the neural network to determine a descriptive feature. This is then further processed to generate two outputs of the network. The first output classifies the neighborhood and determines if it is part of an object of interest. If this is the case, the second output determines where it is located in relation to the object center. This regression output allows us to use a voting process for the actual object detection. This processing step is inspired by approaches based on implicit shape models (ISM). It is able to deal with a certain amount of incorrectly classified neighborhoods, since it combines the results of multiple neighborhoods for the detection of an object. A benefit of our approach as compared to other machine learning methods is its low demand for training data. In our experiments, we achieved a promising detection performance even with less than 1000 training examples. © 2019 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences. All rights reserved." "36243762400;57209469105;55234747900;19337612500;","Development and validation of a supervised machine learning radar Doppler spectra peak-finding algorithm",2019,"10.5194/amt-12-4591-2019","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071753493&doi=10.5194%2famt-12-4591-2019&partnerID=40&md5=d3441d7b1ef09d9020043aa9f338a1fd","In many types of clouds, multiple hydrometeor populations can be present at the same time and height. Studying the evolution of these different hydrometeors in a time-height perspective can give valuable information on cloud particle composition and microphysical growth processes. However, as a prerequisite, the number of different hydrometeor types in a certain cloud volume needs to be quantified. This can be accomplished using cloud radar Doppler velocity spectra from profiling cloud radars if the different hydrometeor types have sufficiently different terminal fall velocities to produce individual Doppler spectrum peaks. Here we present a newly developed supervised machine learning radar Doppler spectra peak-finding algorithm (named PEAKO). In this approach, three adjustable parameters (spectrum smoothing span, prominence threshold, and minimum peak width at half-height) are varied to obtain the set of parameters which yields the best agreement of user-classified and machine-marked peaks. The algorithm was developed for Ka-band ARM zenith-pointing radar (KAZR) observations obtained in thick snowfall systems during the Atmospheric Radiation Measurement Program (ARM) mobile facility AMF2 deployment at Hyytiälä, Finland, during the Biogenic Aerosols - Effects on Clouds and Climate (BAECC) field campaign. The performance of PEAKO is evaluated by comparing its results to existing Doppler peak-finding algorithms. The new algorithm consistently identifies Doppler spectra peaks and outperforms other algorithms by reducing noise and increasing temporal and height consistency in detected features. In the future, the PEAKO algorithm will be adapted to other cloud radars and other types of clouds consisting of multiple hydrometeors in the same cloud volume. © 2019 Copernicus GmbH. All rights reserved." "57206924573;57209398568;57209691961;","Cloud identification and classification from high spectral resolution data in the far infrared and mid-infrared",2019,"10.5194/amt-12-3521-2019","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85068450158&doi=10.5194%2famt-12-3521-2019&partnerID=40&md5=f4cbe18aa76ba93e25ed0670659b3204","A new cloud identification and classification algorithm named CIC is presented. CIC is a machine learning algorithm, based on principal component analysis, able to perform a cloud detection and scene classification using a univariate distribution of a similarity index that defines the level of closeness between the analysed spectra and the elements of each training dataset. CIC is tested on a widespread synthetic dataset of high spectral resolution radiances in the far- and mid-infrared part of the spectrum, simulating measurements from the Fast Track 9 mission FORUM (Far-Infrared Outgoing Radiation Understanding and Monitoring), competing for the ESA Earth Explorer programme, which is currently (2018 and 2019) undergoing industrial and scientific Phase A studies. Simulated spectra are representatives of many diverse climatic areas, ranging from the tropical to polar regions. Application of the algorithm to the synthetic dataset provides high scores for clear or cloud identification, especially when optimisation processes are performed. One of the main results consists of pointing out the high information content of spectral radiance in the far-infrared region of the electromagnetic spectrum to identify cloudy scenes, specifically thin cirrus clouds. In particular, it is shown that hit scores for clear and cloudy spectra increase from about 70 % to 90 % when far-infrared channels are accounted for in the classification of the synthetic dataset for tropical regions. © Author(s) 2019." "55574821400;57195530501;57207107574;21741875100;57213194211;57198137843;56071662500;6701728686;57189843313;57213187691;57200272141;","Applying Machine Learning to Earth Observations in A Standards Based Workflow",2019,"10.1109/IGARSS.2019.8898032","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077684387&doi=10.1109%2fIGARSS.2019.8898032&partnerID=40&md5=6eafffde1b2a76e5bff01fcc4fa8b3a5","Earth Observations (EO) enable scientific research, such as the study of meteorology and climate, ecosystems and forests, hydrology and marine life. Applications of EO help protect populations from disasters and improve life in intelligent cities. Increasingly, Machine Learning techniques are seen as key to solve these complex multidisciplinary problems. The scale and dimensionality of data involved often require the definition of processing chains, or workflows. Standards can facilitate the composition, sharing, execution and discovery of these workflows and applications, making them more useful. This paper presents three applications based on Deep Learning: a tree species classifier, a car detector and a flood detector. These applications rely on software containers to package ML framework and algorithms, as well as on workflows to process EO data. We found that these practices allow improved reuse and deployment of research assets in infrastructures. We also note the strong discriminative capabilities of Deep Learning on smaller datasets and the difficulty of gen-eralization to other methods of sensing or regions of interest. © 2019 IEEE." "57192947904;34869963500;6603354695;","Domain Adaptation of Landsat-8 and Proba-V Data Using Generative Adversarial Networks for Cloud Detection",2019,"10.1109/IGARSS.2019.8899193","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076112397&doi=10.1109%2fIGARSS.2019.8899193&partnerID=40&md5=66c070099c71f0ae9a4abd1dba92fa21","Training machine learning algorithms for new satellites requires collecting new data. This is a critical drawback for most remote sensing applications and specially for cloud detection. A sensible strategy to mitigate this problem is to exploit available data from a similar sensor, which involves transforming this data to resemble the new sensor data. However, even taking into account the technical characteristics of both sensors to transform the images, statistical differences between data distributions still remain. This results in a poor performance of the methods trained on one sensor and applied to the new one. In this this work, we propose to use the generative adversarial networks (GANs) framework to adapt the data from the new satellite. In particular, we use Landsat-8 images, with the corresponding ground truth, to perform cloud detection in Proba-V. Results show that the GANs adaptation significantly improves the detection accuracy. © 2019 IEEE." "22980018800;21742642500;24081888700;","Reconstruction of Cloud Vertical Structure With a Generative Adversarial Network",2019,"10.1029/2019GL082532","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85068180615&doi=10.1029%2f2019GL082532&partnerID=40&md5=0129a56da6c98b3418f6a862312fd8a9","We demonstrate the feasibility of solving atmospheric remote sensing problems with machine learning using conditional generative adversarial networks (CGANs), implemented using convolutional neural networks. We apply the CGAN to generating two-dimensional cloud vertical structures that would be observed by the CloudSat satellite-based radar, using only the collocated Moderate-Resolution Imaging Spectrometer measurements as input. The CGAN is usually able to generate reasonable guesses of the cloud structure and can infer complex structures such as multilayer clouds from the Moderate-Resolution Imaging Spectrometer data. This network, which is formulated probabilistically, also estimates the uncertainty of its own predictions. We examine the statistics of the generated data and analyze the response of the network to each input parameter. The success of the CGAN in solving this problem suggests that generative adversarial networks are applicable to a wide range of problems in atmospheric science, a field characterized by complex spatial structures and observational uncertainties. ©2019. American Geophysical Union. All Rights Reserved." "56326408200;57203324978;55739545700;57209326925;57209326451;7102410570;57190186331;55716531300;","Reconstruction of ocean color data using machine learning techniques in polar regions: Focusing on off Cape Hallett, Ross Sea",2019,"10.3390/rs11111366","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85067395831&doi=10.3390%2frs11111366&partnerID=40&md5=7065de2352e9d6c9ca9e9490aa744aa5","The most problematic issue in the ocean color application is the presence of heavy clouds, especially in polar regions. For that reason, the demand for the ocean color application in polar regions is increased. As a way to overcome such issues, we conducted the reconstruction of the chlorophyll-a concentration (CHL) data using the machine learning-based models to raise the usability of CHL data. This analysis was first conducted on a regional scale and focused on the biologically-valued Cape Hallett, Ross Sea, Antarctica. Environmental factors and geographical information associated with phytoplankton dynamics were considered as predictors for the CHL reconstruction, which were obtained from cloud-free microwave and reanalysis data. As the machine learning models used in the present study, the ensemble-based models such as Random forest (RF) and Extremely randomized tree (ET) were selectedwith 10-fold cross-validation. As a result, both CHL reconstructions fromthe twomodels showed significant agreement with the standard satellite-derived CHL data. In addition, the reconstructed CHLs were close to the actual CHL value even where it was not observed by the satellites. However, there is a slight difference between the CHL reconstruction results from the RF and the ET, which is likely caused by the difference in the contribution of each predictor. In addition, we examined the variable importance for the CHL reconstruction quantitatively. As such, the sea surface and atmospheric temperature, and the photosynthetically available radiation have high contributions to the model developments. Mostly, geographic information appears to have a lower contribution relative to environmental predictors. Lastly, we estimated the partial dependences for the predictors for further study on the variable contribution and investigated the contributions to the CHL reconstruction with changes in the predictors. © 2019 by the authors." "56282082000;27567950000;6604053069;","Estimating and modeling of spatial variability of volumetric attributes of a Nigerian heavy oil and bitumen deposit",2019,"10.2118/198859-MS","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084015654&doi=10.2118%2f198859-MS&partnerID=40&md5=51d77b8eb28093b26266d911bbb23f64","Deposits of heavy oil and natural bitumen have been long-discovered in the Dahomey basin south-western Nigeria. However, inconsistency in estimates of volumes of hydrocarbon contained in these deposits has inhibited commercial interest in the deposits. The inconsistency is attributable to the little or no consideration for spatial variability in those studies. This work is therefore motivated by the need for spatially-coherent geomodels leading to reliable volumetric estimates. An existing database of porosity, depth-to-top and thickness attributes of a section of the deposits located at Agbabu is the subject of this work. This work conducted exploratory spatial data analysis (ESDA) as well as empirical variogram estimation, interpretation and modeling of the attributes. Here, the estimation and interpretation of empirical variogram faced a number of challenges with potentials to render the estimates uninterpretable, unstable and inconsistent with geologic information. These include presence of spatial outlier data, clusteredness of variogram clouds, data paucity, and irregular distribution of point-pairs on variogram clouds. Spatial outliers were either removed or correlated with existing geologic information. The clusteredness issues were resolved using a machine-learning – aided variogram estimation technique recently proposed. Variogram cloud binning approach was deployed to handle irregular distribution of point-pairs. In attempting to deploy an automatic fitting algorithm, cases of insufficient empirical points leading to lack of convergence were encountered. Such cases were resolved by adopting a combination of manual and automatic fitting approaches. Ultimately, this work presents a three-dimensional anisotropic (zonal) porosity variogram model and two-dimensional anisotropic (geometric) models for the depth-to-top and thickness variograms. These models are suitable inputs to spatial interpolation algorithms in generating maps of these volumetric attributes. Copyright 2019, Society of Petroleum Engineers." "57190173611;55458058500;7402317480;57213185612;","An improved automatic pointwise semantic segmentation of a 3D urban scene from mobile terrestrial and airborne LiDAr point clouds: A machine learning approach",2019,"10.5194/isprs-annals-IV-4-W8-139-2019","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077959523&doi=10.5194%2fisprs-annals-IV-4-W8-139-2019&partnerID=40&md5=1407526b397e77da313f1e7c442d045d","Automatic semantic segmentation of point clouds observed in a 3D complex urban scene is a challenging issue. Semantic segmentation of urban scenes based on machine learning algorithm requires appropriate features to distinguish objects from mobile terrestrial and airborne LiDAR point clouds in point level. In this paper, we propose a pointwise semantic segmentation method based on our proposed features derived from Difference of Normal and the features “directional height above” that compare height difference between a given point and neighbors in eight directions in addition to the features based on normal estimation. Random forest classifier is chosen to classify points in mobile terrestrial and airborne LiDAR point clouds. The results obtained from our experiments show that the proposed features are effective for semantic segmentation of mobile terrestrial and airborne LiDAR point clouds, especially for vegetation, building and ground classes in an airborne LiDAR point clouds in urban areas. © Authors 2019. CC BY 4.0 License." "57189220493;6701598631;6603372854;","Automatic detection of objects in 3D point clouds based on exclusively semantic guided processes",2019,"10.3390/ijgi8100442","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075520183&doi=10.3390%2fijgi8100442&partnerID=40&md5=414aabba8b837dd61f37ad2eb86addc2","In the domain of computer vision, object recognition aims at detecting and classifying objects in data sets. Model-driven approaches are typically constrained through their focus on either a specific type of data, a context (indoor, outdoor) or a set of objects. Machine learning-based approaches are more flexible but also constrained as they need annotated data sets to train the learning process. That leads to problems when this data is not available through the specialty of the application field, like archaeology, for example. In order to overcome such constraints, we present a fully semantic-guided approach. The role of semantics is to express all relevant knowledge of the representation of the objects inside the data sets and of the algorithms which address this representation. In addition, the approach contains a learning stage since it adapts the processing according to the diversity of the objects and data characteristics. The semantic is expressed via an ontological model and uses standard web technology like SPARQL queries, providing great flexibility. The ontological model describes the object, the data and the algorithms. It allows the selection and execution of algorithms adapted to the data and objects dynamically. Similarly, processing results are dynamically classified and allow for enriching the ontological model using SPARQL construct queries. The semantic formulated through SPARQL also acts as a bridge between the knowledge contained within the ontological model and the processing branch, which executes algorithms. It provides the capability to adapt the sequence of algorithms to an individual state of the processing chain and makes the solution robust and flexible. The comparison of this approach with others on the same use case shows the efficiency and improvement this approach brings. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/)." "6505585734;","Machines’ fault detection and tolerance using big data management",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074587141&partnerID=40&md5=76daadcbcef928b88e04af290742e6a9","Internet of Things (IoT) is one of significant topics in business and in research field. Modern societies are adopting IoT as a necessity rather than a luxury in their daily life style. IoT defined as devices and machines are connected to data repositories which mainly reside in the Cloud. One of the most challenging issues related to IoT environments is the huge amount of data, referred to as Big Data. Data generated from IoT environments are generally stored and processed in cloud infrastructure, then it consumed (in analyzed forms) as input to feed the same IoT environments. IoT and Big Data are two faces of the same coin. This article discusses IoT aspects related to big data in cloud, proposes a Big Data Management Center (BDMC) that utilizes accumulated big data transmitted form machines sensors .BDMC can anticipate faults that might occur in machines, besides it can decide which fault can be tolerated and which fault resulted in severe damage in that machine. This prediction will give us advantage in advance to overcome problems before they occur. This article will also go through important aspects relate to IoT such as “storage as cloud”, “wireless connectivity” and data storage using Hadoop format. Finally, this article presents a case study on fault detection on real vehicles tests, then experimental results is provided. © International Research Publication House." "36491173000;7202269863;55313567200;35612130700;","Using UAV for automatic lithological classification of open pit mining front",2019,"10.1590/0370-44672018720122","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85069743611&doi=10.1590%2f0370-44672018720122&partnerID=40&md5=37ccfcf51f3d5cee0a6c6a03e0d70000","Mine planning is dependent on the natural lithologic features and on the definition of their limits. The geological model is constantly updated during the life of the mine, based on all the information collected so far, plus the knowledge developed from the exploration stage up to the mine closure. As the mine progresses, the amount of available data increases, as well as the experience of the geological modeller and mine planner who deliver the short, medium, and long-term plans. This classical approach can benefit from the automation of the geological mapping on the mining faces and outcrops, improving the speed of repetitious work and avoiding exposure to intrinsic dangers like mining equipment, falling rocks, high wall proximity, among others. The use of photogrammetry to keep up with surface mining activities boarded in UAVs is a reality and the automated lithological classification using machine learning techniques is a low-cost evolution that might present accuracies above 90% of the contact zones and lithologies based on the automated dense point cloud classification when compared to the manual (or reality) classified model. © 2019, Escola de Minas. All rights reserved." "41461833800;57209197440;","Implementing the real-time data environment for the oil and gas digital transformation",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85066780065&partnerID=40&md5=4e962086b1ef030096523e6c9c5d5980","Real-time operations have long provided a crucial decision support and collaboration capability that enhances decision making during the well construction process. Operators, drilling contractors, and service companies use real-time capabilities to improve operations service quality, create efficiencies, understand formation geology, and enhance overall reservoir knowledge. Historical applications have provided domain experts with real-time information to support less-experienced wellsite personnel, provide advice and guidance, and conduct remotely performed operations. However, the information was typically centered on a single well and often with a discrete job-focused execution mindset, even when the same provider delivered multiple services. The industry focus on “digital transformation” creates new opportunities for the real-time environment. The promise of real-time analytics and machine learning increases the potential value of having access to all petrotechnical data. To be successful, these systems should simultaneously integrate data across two dimensions: all services provided throughout a single asset’s life cycle and all wells across an entire basin. While engineers use historical real-time data for offset analysis during their next well design, data typically remain filed away, relatively inaccessible for large-scale analytics. This paper presents a case history on the development and implementation of an enterprise real-time data environment in support of the digital transformation. Key objectives for the environment are to ensure data accessibility for analytics modeling while maintaining the digital coherence necessary during the data collection and distribution processes across multiple services and wells. Functional capabilities include a multi-well, real-time historian, automated data distribution to a big data environment for full-scale analytics, the capability to run analytics models at the edge, and interconnectivity with other field systems for more context-enriched datasets. Key design principles for success are the use of an open architecture to accelerate innovation and a cloud-first, microservices design to maximize implementation flexibility. Key development learnings are also discussed. © 2019 Offshore Mediterranean Conference (OMC). All rights reserved." "23481467500;23491866500;","A single-tree processing framework using terrestrial laser scanning data for detecting forest regeneration",2019,"10.3390/rs11010060","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85059951414&doi=10.3390%2frs11010060&partnerID=40&md5=ddf5e0d43930a30490417285279458cf","Direct assessment of forest regeneration from remote sensing data is a previously little-explored problem. This is due to several factors which complicate object detection of small trees in the understory. Most existing studies are based on airborne laser scanning (ALS) data, which often has insufficient point densities in the understory forest layers. The present study uses plot-based terrestrial laser scanning (TLS) and the survey design was similar to traditional forest inventory practices. Furthermore, a framework of methods was developed to solve the difficulties of detecting understory trees for quantifying regeneration in temperate montane forest. Regeneration is of special importance in our montane study area, since large parts are declared as protection forest against alpine natural hazards. Close to nature forest structures were tackled by separating 3D tree stem detection from overall tree segmentation. In support, techniques from 3D mathematical morphology, Hough transformation and state-of-the-art machine learning were applied. The methodical framework consisted of four major steps. These were the extraction of the tree stems, the estimation of the stem diameters at breast height (DBH), the image segmentation into individual trees and finally, the separation of two groups of regeneration. All methods were fully automated and utilized volumetric 3D image information which was derived from the original point cloud. The total amount of regeneration was split into established regeneration, consisting of trees with a height > 130 cm in combination with a DBH < 12 cm and unestablished regeneration, consisting of trees with a height < 130 cm. Validation was carried out against field-based expert estimates of percentage ground cover, differentiating seven classes that were similar to those used by forest inventory. The mean absolute error (MAE) of our method for established regeneration was 1.11 classes and for unestablished regeneration only 0.27 classes. Considering the metrical distances between the class centres, the MAE amounted 8.08% for established regeneration and 2.23% for unestablished regeneration. © 2019 by the authors." "57193851407;57204945981;7004242319;8573340700;","A Versatile Method for Ice Particle Habit Classification Using Airborne Imaging Probe Data",2018,"10.1029/2018JD029163","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85058022375&doi=10.1029%2f2018JD029163&partnerID=40&md5=8987464a75d3f5988ccaa33eff02a697","A versatile method to automatically classify ice particle habit from various airborne optical array probes is presented. The classification is achieved using a multinomial logistic regression model. For each airborne probe, the model determines the particle habit (among six classes) based on a large set of geometrical and textural descriptors extracted from the two-dimensional image of a particle. The technique is applied and evaluated using three probes with significantly different specifications: the high volume precipitation spectrometer, the two-dimensional stereo probe, and the cloud particle imager. Performance and robustness of the method are assessed using standard machine learning tools on the basis of thousands of images manually labeled for each of the considered probes. The three classifiers show good performance characterized by overall accuracies and Heidke skill scores above 90%. Depending on the application and user preferences, the classification scheme can be easily adapted. For a more precise output, intraclass subclassification can be achieved in a nested fashion, illustrated here with columnar crystals and aggregates. A comparative study of the classification output obtained with the three probes is presented for two aircraft flight periods selected when the three probes were operating together. Results are globally consistent in term of proportions of habit identified (once blurry and partial images have been automatically discarded). A perfect agreement is not expected as the three considered probes are sensitive to different particle size range. ©2018. American Geophysical Union. All Rights Reserved." "7402962962;","Outer space and cyber space: Meeting et in the cloud",2018,"10.1017/S1473550416000318","https://www.scopus.com/inward/record.uri?eid=2-s2.0-84980338898&doi=10.1017%2fS1473550416000318&partnerID=40&md5=951d98ef9ba52c38cd0d9ecec98a7ed5","What justifies the astrobiologist's search for post-biological or machine-intelligence in outer space? Four assumptions borrowed from transhumanism (H+) seem to be at work: (1) it is reasonable to speculate that life on Earth will evolve in the direction of post-biological intelligence; (2) if extraterrestrials have evolved longer than we on Earth, then they will be more scientifically and technologically advanced; (3) superintelligence, computer uploads of brains, and dis-embodied mind belong together; and (4) evolutionary progress is guided by the drive toward increased intelligence. When subjected to critical review, these assumptions prove to be weak. Most importantly, evolutionary biologists do not support the idea that evolution is internally directed toward increased intelligence. Without this assumption, justifying the search for ET more intelligent than earthlings is anaemic. Nevertheless, one can still hope that in the near future we will be communicating with new neighbours in the Milky Way. Can sheer hope inspire science? © 2016 Cambridge University Press." "56673446700;55362143700;56783246900;23005893600;","Classification of image matching point clouds over an urban area",2018,"10.1080/01431161.2018.1452069","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85054810160&doi=10.1080%2f01431161.2018.1452069&partnerID=40&md5=b0318c442fe84cc14f3ea0c69ddd745d","Airborne laser scanning (ALS) and image matching are the two main techniques for generating point clouds for large areas. While the classification of ALS point clouds has been well investigated, there are few studies that are related to image matching point clouds. In this study, point clouds of multiple resolutions from high-resolution aerial images (ground sampling distance, GSD, of 6 cm) over the city of Vienna were generated and investigated with respect to point density and processing time. Three different study sites with various urban structures are selected from a bigger dataset and classified based on two different approaches: machine learning and a traditional operator-based decision tree. Classification accuracy was evaluated and compared with confusion matrices. In general, the machine learning method results in a higher overall accuracy compared to the simple decision tree method, with accuracies of 87% and 84%, respectively, at the highest resolution. At lower-resolution levels (GSDs of 12 cm and 24 cm), the overall accuracy of machine learning drops by 4% and that of the simple decision tree by 7% for each level. Classifying rasterized data instead of the original point cloud resulted in an accuracy drop of 5%. Thus, using machine learning on point clouds at the highest available resolution is suggested for classification of urban areas. © 2018 Informa UK Limited, trading as Taylor & Francis Group." "55787237000;49964767000;","Efficient cloud Resource Scaling based on prediction approaches",2018,"10.14419/ijet.v7i3.2.14563","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079962183&doi=10.14419%2fijet.v7i3.2.14563&partnerID=40&md5=058e3f1105aec6afedf975b0f7bbb0a2","Resource Scaling is one of the important job in cloud environment while adapting resource configurations due to elasticity mechanism. In the view of cloud computing, resource scaling mechanism hold the assurance of QoS (Quality of Service), So, one of the key challenging task in cloud environment is, resource scaling. Effective scaling mechanism gives an optimal solutions for computational problems while achieving QoS and avoiding SLA (Service Level Agreement) violations. To enhance resource scaling mechanism in cloud environment, predicting future workload to the each application in different manners like number of physical machines, number of virtual machines, number of requests and resource utilization etc., is an essential step. According to the prediction results, resource scaling can be done in the right time, while preventing QoS dropping and SLA violations. To achieve efficient resource scaling, proposed approach lease advantages of fuzzy time series and machine learning algorithms. The proposed approach is able to reach effective resource scaling mechanism with better results. © 2018 Authors." "57194005615;57215463665;57202322668;57202320981;57202320607;","Handover forecasting in 5G using machine learning",2018,"10.14419/ijet.v7i2.31.13401","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85047855373&doi=10.14419%2fijet.v7i2.31.13401&partnerID=40&md5=1e57b2659dbefcac187ba588a8e2e431","Communication plays a major role in human's life. Without network communication can't be done, to achieve proper and effective communication different generations of networks are introduced. Each generation has its own features and perspective of communication, but till now there is no network properly makes people to communicate. Many researches says 5G network will rule the network world as it satisfies all the effective network goals. This paper is proposed to obtain all the goals of a communication network by making proper handover with the help of machine learning. Here we have used two main algorithms to make our 5G handover process by clustering and classifying. Clustering is a process of making the datasets into single units of every users and classification is a process of classifying user's clustered datasets into common path using prediction and forecasting. For clustering we are using K-means and for classification we are using Random Forest algorithm. By using the algorithms the datasets which is being predicted and forecasted is stored in the cloud. Here cloud technology is used as a platform for developing datasets associated with internet. 5G network adapts to any form technology easier and here we have used all the essential technologies under machine learning. This paper deals with all the above methodologies effectively with newer combinations of algorithms along with proper solutions. © 2018 Authors." "16305516000;56192121900;57201487640;","Harmonic extension on the point cloud",2018,"10.1137/16M1098747","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85045054370&doi=10.1137%2f16M1098747&partnerID=40&md5=4f8df65d9daaff21e416814eb142c4b7","In this paper, we consider the harmonic extension problem, which is widely used in many applications of machine learning. We formulate the harmonic extension as solving a Laplace–Beltrami equation with Dirichlet boundary condition. We use the point integral method (PIM) proposed in [Z. Li, Z. Shi, and J. Sun, Commun. Comput. Phys., 22 (2017), pp. 228–258; Z. Shi and J. Sun, Res. Math. Sci., to appear; Z. Li and Z. Shi, Multiscale Model. Simul., 14 (2016), pp. 874–905] to solve the Laplace–Beltrami equation. The basic idea of the PIM method is to approximate the Laplace equation using an integral equation, which is easy to discretize from points. Based on the integral equation, we found that the traditional graph Laplacian method (GLM) fails to approximate the harmonic functions near the boundary. One important application of the harmonic extension in machine learning is semisupervised learning. We run a popular semisupervised learning algorithm by Zhu, Ghahramani, and Lafferty [Machine Learning, Proceedings of the Twentieth International Conference (ICML 2003), 2003, Washington, DC, 2003, ARAI Press, Menlo Park, CA pp. 912–919] over a couple of well-known datasets and compare the performance of the aforementioned approaches. Our experiments show the PIM performs the best. We also apply PIM to an image recovery problem and show it outperforms GLM. Finally, on the model problem of the Laplace–Beltrami equation with Dirichlet boundary, we prove the convergence of the point integral method. © 2018 Society for Industrial and Applied Mathematics." "7006792019;48661217200;23027278800;57200600874;7102811204;","Feature extraction and machine learning for the classification of active cropland in the Aral Sea Basin",2017,"10.1109/IGARSS.2017.8127326","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85041836347&doi=10.1109%2fIGARSS.2017.8127326&partnerID=40&md5=a0732bcaf557128e13a5d2c882e8302e","Agricultural monitoring over large areas such as the irrigation zones of the Aral Sea Basin (ASB) needs Remote Sensing imagery with a high temporal resolution and a large swath width. Satellite sensors such as MODIS or the upcoming Sentinel-3 suits these requirements. A clear discrimination of active cropland coverage however is impossible using low spatial resolution imagery due to mixed pixel effects with other landcover types. In this paper we utilized Landsat 8 data and present the derivation of the active cropland in the ASB for 3 consecutive years (2014-2016). Each observed year comprises two growing seasons (spring and summer) with 40 Landsat scenes in each season, with a total of 240 Landsat scenes. Due to the lack of adequate in-situ data for remote agricultural regions, a feature extraction method based on image segmentation and k-means clustering is applied to delineate samples of agricultural fields which are forwarded to machine learning algorithms for image classification including. A simple markovian logic is used to introduce prior distribution values the following observation year. The results clearly show, that the extraction of active cropland is possible without the need for time series data, which is limited due to clouds and acquisition date irregularity. Model transferability tests were performed to assess machine learning inference on unseen data with training datasets from adjacent or non-adjacent scenes. The extraction of active cropland for each season also yielded secondary information such as the assessment of land use intensities and crop rotation patterns. © 2017 IEEE." "22235457800;7403531523;52264136000;7004364155;","Determination of CERES TOA fluxes using machine learning algorithms. Part I: Classification and retrieval of CERES cloudy and clear scenes",2017,"10.1175/JTECH-D-16-0183.1","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85032746166&doi=10.1175%2fJTECH-D-16-0183.1&partnerID=40&md5=ecf69b5675c0551235ccb0203aca6fc7","Continuous monitoring of the earth radiation budget (ERB) is critical to the understanding of Earth's climate and its variability with time. The Clouds and the Earth's Radiant Energy System (CERES) instrument is able to provide a long record of ERB for such scientific studies. This manuscript, which is the first of a two-part paper, describes the new CERES algorithm for improving the clear/cloudy scene classification without the use of coincident cloud imager data. This new CERES algorithm is based on a subset of the modern artificial intelligence (AI) paradigm called machine learning (ML) algorithms. This paper describes the development and application of the ML algorithm known as random forests (RF), which is used to classify CERES broadband footprint measurements into clear and cloudy scenes. Results from the RF analysis carried using the CERES Single Scanner Footprint (SSF) data for January and July are presented in the manuscript. The daytime RF misclassification rate (MCR) shows relatively large values (> 30%) for snow, sea ice, and bright desert surface types, while lower values (< 10%) for the forest surface type. MCR values observed for the nighttime data in general show relatively larger values for most of the surface types compared to the daytime MCR values. The modified MCR values show lower values (< 4%) for most surface types after thin cloud data are excluded from the analysis. Sensitivity analysis shows that the number of input variables and decision trees used in the RF analysis has a substantial influence on determining the classification error. © 2017 American Meteorological Society." "37003444300;55496988000;57218522431;55638787200;56430227700;","Improved online sequential extreme learning machine for simulation of daily reference evapotranspiration",2017,"10.24850/j-tyca-2017-02-12","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85030551493&doi=10.24850%2fj-tyca-2017-02-12&partnerID=40&md5=b6286a0fc06b4ce201639d2c900af0f3","The traditional extreme learning machine has significant disadvantages, including slow training, difficulty in selecting parameters, and difficulty in setting the singularity and the data sample. A prediction model of an improved Online Sequential Extreme Learning Machine (IOS-ELM) of daily reference crop evapotranspiration is therefore examined in this paper. The different manipulation of the inverse of the matrix is made according to the optimal solution and using a regularization factor at the same time in the model. The flexibility of the IOS-ELM in ET0 modeling was assessed using the original meteorological data (Tmax, Tm, Tmin, n, Uh, RHm, φ, Z) of the years 1971–2014 in Yulin, Ankang, Hanzhong, and Xi’an of Shaanxi, China. Those eight parameters were used as the input, while the reference evapotranspiration values were the output. In addition, the ELM, LSSVM, Hargreaves, Priestley-Taylor, Mc Cloud and IOS-ELM models were tested against the FAO-56 PM model by the performance criteria. The experimental results demonstrate that the performance of IOS-ELM was better than the ELM and LSSVM and significantly better than the other empirical models. Furthermore, when the total ET0 estimation of the models was compared by the relative error, the results of the intelligent algorithms were better than empirical models at rates lower than 5%, but the gross ET0 empirical models mainly had 12% to 64.60% relative error. This research could provide a reference to accurate ET0 estimation by meteorological data and give accurate predictions of crop water requirements, resulting in intelligent irrigation decisions in Shaanxi." "56012593900;56701578200;35509463200;57199699111;8308107100;9744927700;57207556152;24468168900;","Deep learning for very high-resolution imagery classification",2017,"10.4324/9781315371740","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85051813632&doi=10.4324%2f9781315371740&partnerID=40&md5=f44bc29eff1aaeba3177dd20727f0693","Very high-resolution (VHR) land cover classification maps are needed to increase the accuracy of current land ecosystem and climate model outputs. Limited studies are in place that demonstrate the state-of-the-art in deriving VHR land cover products [1-4]. Additionally, most methods heavily rely on commercial softwares that are difficult to scale given the area of study (e.g., continents to globe). Complexities in present methods relate to (1) scalability of the algorithm, (2) large image data processing (compute and memory intensive), (3) computational cost, (4) massively parallel architecture, and (5) machine learning automation. VHR satellite data sets are of the order of terabytes and features extracted from these data sets are of the order of petabytes. This chapter demonstrates the use of a scalable machine learning algorithm using airborne imagery data acquired by the National Agriculture Imagery Program (NAIP) for the Continental United States (CONUS) at an optimal spatial resolution of 1 m [5]. These data come as image tiles (a total of quarter million image scenes with ~ 60 million pixels) that are multispectral in nature (red, green, blue, and near-infrared [NIR] spectral channels) and have a total size of ~ 60 terabytes for an individual acquisition over CONUS. Features extracted from the entire data set would amount to ~ 8 -10 petabytes. In the proposed approach, a novel semiautomated machine learning algorithm rooted on the principles of “deep learning” is implemented to delineate the percentage of canopy tree cover. In order to perform image analytics in such a granular system, it is mandatory to devise an intelligent archiving and query system for image retrieval, file structuring, metadata processing, and filtering of all available image scenes. This chapter showcases an end-to-end architecture for designing the learning algorithm, namely deep 114belief network (DBN) (stacked restricted Boltzmann machines [RBMs] as an unsupervised classifier) followed by a backpropagation neural network (BPNN) for image classification, a statistical region merging (SRM)-based segmentation algorithm to perform unsupervised segmentation, and a structured prediction framework using conditional random field (CRF) that integrates the results of the classification module and the segmentation module to create the final classification labels. In order to scale this process across quarter million NAIP tiles that cover the entire CONUS, we provide two architectures, one using the National Aeronautics and Space Administration (NASA) high-performance computing (HPC) infrastructure [6,7] and the other using the Amazon Web Services (AWS) cloud compute platform [8]. The HPC framework describes the granular parallelism architecture that can be designed to implement the process across multiple cores with low-to-medium memory requirements in a distributed manner. The AWS framework showcases use-case scenarios of deploying multiple AWS services like the Simple Storage Service (S3) for data storage [9], Simple Queuing Service (SQS) [10] for coordinating the worker nodes and the compute-optimized Elastic Cloud Compute (EC2) [11] along with spot instances for implementing the machine learning algorithm © 2017 by Taylor & Francis Group, LLC." "55100412100;11140546100;57203059237;","Constrained, multi-objective, steamflood injection redistribution optimization, using a cloud-distributed, metamodel-assisted, memetic optimization algorithm",2017,"10.2118/186010-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85050390048&doi=10.2118%2f186010-ms&partnerID=40&md5=0326de0f7be8b3edd43331af42644a11","A cloud-distributed optimization algorithm applicable to large scale, constrained, multiobjective, optimization problems, such as steamflood redistribution, is presented. The proposed algorithm utilizes the so-called Metamodel Assisted Evolutionary Algorithm (MAEA) as its algorithmic basis. MAEAs use a generic implementation of an evolutionary algorithm as their main optimization engine and advanced machine learning techniques as metamodels. Metamodels are utilized through the application of an inexact pre-evaluation phase during the optimization, which substantially decreases the evaluations of the problem specific forward model. Additionally, a unification of global search (GS) and local search (LS) is achieved via the use of Lamarckian learning principles applied on top of a MAEA creating, in essence, a Metamodel Assisted Memetic Algorithm (MAMA). MAMAs profit from the abilities of MAEAs to explore the most promising regions of the design space without being trapped in local optima while also utilizing the efficiency of deterministic methods to further refine promising solutions located during GS. At the end of each EA generation, the most promising members of the current populations are selected to undergo LS using a gradient-based method. Further, integration with scalable cloud-distributed computing allows MAMAs (CD-MAMA) to perform rapid and simultaneous evaluation of tens of thousands of operating plans. The proposed algorithm has been statistically validated on two mathematical test cases and, subsequently, used to optimize a field undergoing steamflood under two different oil-price scenarios. Thus, demonstrating that, cloud-distributed MAMAs coupled with efficient reservoir models, allow for steamflood injection redistribution optimization in affordable, by industry, wallclock times (hours). For the field in question production comes from poorly consolidated sands within the Antelope Shale member of the Miocene Monterey formation with porosity averaging 30%, permeability averaging 2,000 mD and net thicknesses typically between 50 and 300 feet. Structural dip is steep at approximately 60 degrees. The reservoirs are shallow, with depths ranging from 200 - 600 feet TVD. Oil gravity is approximately 13° API. Reservoir pressures are well below bubble point and average 50 - 100 PSI. The field has about 200 producers and 30 injectors, producing about 2000 bpd of oil and injecting 8000bpd of steam. The field has been under operation for about 7 years, with most of the continuous steamflood starting about 3.5 years ago. Copyright 2017, Society of Petroleum Engineers." "6602459839;57211129420;8699469900;","Deep data: Discovery and visualization Application to hyperspectral ALMA imagery",2016,"10.1017/S1743921317000175","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020016036&doi=10.1017%2fS1743921317000175&partnerID=40&md5=489378aa21e2b752e8765fa6c80b161b","Leading-edge telescopes such as the Atacama Large Millimeter and sub-millimeter Array (ALMA), and near-future ones, are capable of imaging the same sky area at hundreds-to-thousands of frequencies with both high spectral and spatial resolution. This provides unprecedented opportunities for discovery about the spatial, kinematical and compositional structure of sources such as molecular clouds or protoplanetary disks, and more. However, in addition to enormous volume, the data also exhibit unprecedented complexity, mandating new approaches for extracting and summarizing relevant information. Traditional techniques such as examining images at selected frequencies become intractable while tools that integrate data across frequencies or pixels (like moment maps) can no longer fully exploit and visualize the rich information. We present a neural map-based machine learning approach that can handle all spectral channels simultaneously, utilizing the full depth of these data for discovery and visualization of spectrally homogeneous spatial regions (spectral clusters) that characterize distinct kinematic behaviors. We demonstrate the effectiveness on an ALMA image cube of the protoplanetary disk HD142527. The tools we collectively name NeuroScope are efficient for Big Data due to intelligent data summarization that results in significant sparsity and noise reduction. We also demonstrate a new approach to automate our clustering for fast distillation of large data cubes. © Copyright International Astronomical Union 2017." "7006802750;6602337949;6505862984;7202201947;7201527458;14031427400;","Automating the estimation of various meteorological parameters using satellite data and machine learning techniques",2002,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-0036028201&partnerID=40&md5=6ffced3f546df96f93f91360d5bf672f","Satellite data from various sensors and platforms are being used to develop automated algorithms to assist in U.S. Navy operational weather assessment and forecasting. Supervised machine learning techniques are used to discover patterns in the data and develop associated classification and parameter estimation algorithms. These methods are applied to cloud classification in GOES imagery, tropical cyclone intensity estimation using SSM/I data, and cloud ceiling height estimation at remote locations using appropriate geostationary and polar orbiting satellite data in conjunction with numerical weather prediction output and climatology. All developed algorithms rely on training data sets that consist of records of attributes (computed from the appropriate data source) and the associated ground truth." "57204526072;57219866769;57212363342;26026749200;","Comparative Assessment of Snowfall Retrieval From Microwave Humidity Sounders Using Machine Learning Methods",2020,"10.1029/2020EA001357","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85095979478&doi=10.1029%2f2020EA001357&partnerID=40&md5=28fc29b84e8d694595c75dbe2c11b1c2","Accurate quantification of snowfall rate from space is important but has remained difficult. Four years (2007–2010) of NOAA-18 Microwave Humidity Sounder (MHS) data are trained and tested with snowfall estimates from coincident CloudSat Cloud Profiling Radar (CPR) observations using several machine learning methods. Among the studied methods, random forest using MHS (RF-MHS) is found to be the best for both detection and estimation of global snowfall. The RF-MHS estimates are tested using independent years of coincident CPR snowfall estimates and compared with snowfall rates from Modern-Era Retrospective analysis for Research and Applications Version 2 (MERRA-2), Atmospheric Infrared Sounder (AIRS), and MHS Goddard Profiling Algorithm (GPROF). It was found that RF-MHS algorithm can detect global snowfall with approximately 90% accuracy and a Heidke skill score of 0.48 compared to independent CloudSat samples. The surface wet bulb temperatures, brightness temperatures at 190 GHz, and 157 GHz channels are found to be the most important features to delineate snowfall areas. The RF-MHS retrieved global snowfall rates are well compared with CPR estimates and show generally better statistics than MERRA-2, AIRS, and GPROF products. A case study over the United States verifies that the RF-MHS estimated snowfall agrees well with the ground-based National Center for Environmental Prediction (NCEP) Stage-IV and MERRA-2 product, whereas a relatively large underestimation is observed with the current GPROF product (V05). MHS snowfall estimated based on RF algorithm, however, shows some underestimation over cold and snow-covered surfaces (e.g., Greenland, Alaska, and northern Russia), where improvements through new sensors or retrieval techniques are needed. ©2020. The Authors." "57194554240;14519510100;43561035200;43561518700;56684747900;","A hybrid data balancing method for classification of imbalanced training data within google earth engine: Case studies from mountainous regions",2020,"10.3390/rs12203301","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092930622&doi=10.3390%2frs12203301&partnerID=40&md5=9c7b2e0955ced4519f475b763bf4cf8a","Distribution of Land Cover (LC) classes is mostly imbalanced with some majority LC classes dominating against minority classes in mountainous areas. Although standard Machine Learning (ML) classifiers can achieve high accuracies for majority classes, they largely fail to provide reasonable accuracies for minority classes. This is mainly due to the class imbalance problem. In this study, a hybrid data balancing method, called the Partial Random Over-Sampling and Random Under-Sampling (PROSRUS), was proposed to resolve the class imbalance issue. Unlike most data balancing techniques which seek to fully balance datasets, PROSRUS uses a partial balancing approach with hundreds of fractions for majority and minority classes to balance datasets. For this, time-series of Landsat-8 and SRTM topographic data along with various spectral indices and topographic data were used over three mountainous sites within the Google Earth Engine (GEE) cloud platform. It was observed that PROSRUS had better performance than several other balancing methods and increased the accuracy of minority classes without a reduction in overall classification accuracy. Furthermore, adopting complementary information, particularly topographic data, considerably increased the accuracy of minority classes in mountainous areas. Finally, the obtained results from PROSRUS indicated that every imbalanced dataset requires a specific fraction(s) for addressing the class imbalance problem, because different datasets contain various characteristics. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57192257968;7006813492;7005734687;7004563021;","See the forest and the trees: Effective machine and deep learning algorithms for wood filtering and tree species classification from terrestrial laser scanning",2020,"10.1016/j.isprsjprs.2020.08.001","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089223435&doi=10.1016%2fj.isprsjprs.2020.08.001&partnerID=40&md5=d849c8012eddc1006d689bcc5fbb482b","Determining tree species composition in natural forests is essential for effective forest management. Species classification at the individual tree level requires fine-scale traits which can be derived through terrestrial laser scanning (TLS) point clouds. A generalizable species classification framework also needs to decouple seasonal foliage variation from deciduous species, for which wood filtering is applicable. Different machine learning and deep learning models are feasible for wood filtering and species classification. We investigated 13 machine learning and deep learning classifiers for 9 species, and 15 classifiers for filtering wood points from TLS plot scans. Each classifier was evaluated using the criteria of mean Intersection over Union accuracy (mIoU), training stability and time cost. On average, deep learning classifiers outperformed machine learning classifiers by 10% and 5% in terms of wood and species classification mIoU, respectively. PointNet++ provided the best species classifier, with the highest mIoU (0.906), stability, and moderate time cost. Among wood classifiers, UNet achieved the top mIoU (0.839) while ResNet-50 was recommended for rapid trial and error testing. Across the classifications, the factors of input resolution, attributes and features were also analyzed. Hot zones of species classification with PointNet++ were visualized to indicate how AI interpret species traits. © 2020" "55636630900;56724707600;6506773247;35192452600;","Mapping mangrove forest using Landsat 8 to support estimation of land-based emissions in Kenya",2020,"10.1007/s40808-020-00778-x","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086737208&doi=10.1007%2fs40808-020-00778-x&partnerID=40&md5=f62a1d2054da056b1d752ff8192955b4","The Kenyan coast is constantly under persistent cloud cover which hinders mapping using optical images.Up-to-date land-cover information in such areas is sometimes missing from national mapping initiatives.This study uses a computed composite image based on a mean of cloud and shadow free Function of Mask masked multi-temporal Landsat 8 images acquired during long-dry season in a pilot area. We test the effectiveness of the composite to map mangrove forest using random forest (RF) and support vector machines (SVM) machine learning algorithms integrated with context from Markov random fields (MRF(s)).MRFs was chosen because it is computationally efficient hence can be scaled out nationally. The MRF frameworks are compared to pixel-based classification using threefold independent validation samples. SVM–MRFs and RF–MRFs methods have the highest overall accuracy compared to pixel-based classification.However, visual assessment of predicted land-cover using aerial photograghs established that SVM–MRFs framework corresponded well to land-cover in the study area. This framework also managed to map classes with limited ground reference data better than RF–MRFs. Generally, context in both techniques played a discriminative role especially in heterogeneous regions. Therefore, scaling out this approaches would go a long way in generating mangrove forest map inventory in persistent cloud cover regions which is useful for land-based emission estimation. © 2020, Springer Nature Switzerland AG." "57216639492;6602209960;55470317400;36015861500;","Efficient Training of Semantic Point Cloud Segmentation Via Active Learning",2020,"10.5194/isprs-annals-V-2-2020-243-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090845790&doi=10.5194%2fisprs-annals-V-2-2020-243-2020&partnerID=40&md5=685f3acb71db43ebd275aa00a938712f","With the development of LiDAR and photogrammetric techniques, more and more point clouds are available with high density and in large areas. Point cloud interpretation is an important step before many real applications like 3D city modelling. Many supervised machine learning techniques have been adapted to semantic point cloud segmentation, aiming to automatically label point clouds. Current deep learning methods have shown their potentials to produce high accuracy in semantic point cloud segmentation tasks. However, these supervised methods require a large amount of labelled data for proper model performance and good generalization. In practice, manual labelling of point clouds is very expensive and time-consuming. Active learning can iteratively select unlabelled samples for manual annotation based on current statistical models and then update the labelled data pool for next model training. In order to effectively label point clouds, we proposed a segment based active learning strategy to assess the informativeness of samples. Here, the proposed strategy uses 40% of the whole training dataset to achieve a mean IoU of 75.2% which is 99.1% of the accuracy in mIoU obtained from the model trained on the full dataset, while the baseline method using same amount of data only reaches 69.6% in mIoU corresponding to 90.9% of the accuracy in mIoU obtained from the model trained on the full dataset. © 2020 Copernicus GmbH. All rights reserved." "57218546375;56263897000;56922261600;57117329200;55207690500;35552595100;","Building extraction using orthophotos and dense point cloud derived from visual band aerial imagery based on machine learning and segmentation",2020,"10.3390/RS12152397","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089524467&doi=10.3390%2fRS12152397&partnerID=40&md5=b4d9ab1dee9a5f54feb4474ca2e229c6","Urban sprawl related increase of built-in areas requires reliable monitoring methods and remote sensing can be an efficient technique. Aerial surveys, with high spatial resolution, provide detailed data for building monitoring, but archive images usually have only visible bands. We aimed to reveal the efficiency of visible orthophotographs and photogrammetric dense point clouds in building detection with segmentation-based machine learning (with five algorithms) using visible bands, texture information, and spectral and morphometric indices in different variable sets. Usually random forest (RF) had the best (99.8%) and partial least squares the worst overall accuracy (~60%). We found that >95% accuracy can be gained even in class level. Recursive feature elimination (RFE) was an efficient variable selection tool, its result with six variables was like when we applied all the available 31 variables. Morphometric indices had 82% producer's and 85% user's Accuracy (PA and UA, respectively) and combining them with spectral and texture indices, it had the largest contribution in the improvement. However, morphometric indices are not always available but by adding texture and spectral indices to red-green-blue (RGB) bands the PA improved with 12% and the UA with 6%. Building extraction from visual aerial surveys can be accurate, and archive images can be involved in the time series of a monitoring. © 2020 by the authors." "57126877800;57215848943;57217155505;57200555819;55469293700;","DANCE-NET: Density-aware convolution networks with context encoding for airborne LiDAR point cloud classification",2020,"10.1016/j.isprsjprs.2020.05.023","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086503837&doi=10.1016%2fj.isprsjprs.2020.05.023&partnerID=40&md5=d34e9ce6e8055c89f41ef75d493a4157","Airborne LiDAR point cloud classification has been a long-standing problem in photogrammetry and remote sensing. Early efforts either combine hand-crafted feature engineering with machine learning-based classification models or leverage the power of conventional convolutional neural networks (CNNs) on projected feature images. Recent proposed deep learning-based methods tend to develop new convolution operators which can be directly applied on raw point clouds for representative point feature learning. Although these methods have achieved satisfying performance for the classification of airborne LiDAR point clouds, they cannot adequately recognize fine-grained local structures due to the uneven density distribution of 3D point clouds. In this paper, to address this challenging issue, we introduce a density-aware convolution module which uses the point-wise density to reweight the learnable weights of convolution kernels. The proposed convolution module can approximate continuous convolution on unevenly distributed 3D point sets. Based on this convolution module, we further develop a multi-scale CNN model with downsampling and upsampling blocks to perform per-point semantic labeling. In addition, to regularize the global semantic context, we implement a context encoding module to predict a global context encoding and formulated a context encoding regularizer to enforce the predicted context encoding to be aligned with the ground truth one. The overall network can be trained in an end-to-end fashion and directly produces the desired classification results in one network forward pass. Experiments on the ISPRS 3D Labeling Dataset and 2019 Data Fusion Contest Dataset demonstrate the effectiveness and superiority of the proposed method for airborne LiDAR point cloud classification. © 2020 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)" "7401931279;","Potential of Large-Scale Inland Water Body Mapping from Sentinel-1/2 Data on the Example of Bavaria’s Lakes and Rivers",2020,"10.1007/s41064-020-00111-2","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086029687&doi=10.1007%2fs41064-020-00111-2&partnerID=40&md5=e1e10e6d28329bbdaccf288598b41194","The mapping of water bodies is an important application area of satellite-based remote sensing. In this contribution, a simple framework based on supervised learning and automatic training data annotation is shown, which allows to map inland water bodies from Sentinel satellite data on large scale, i.e. on state level. Using the German state of Bavaria as an example and different combinations of Sentinel-1 SAR and Sentinel-2 multi-spectral imagery as inputs, potentials and limits for the automatic detection of water surfaces for rivers, lakes, and reservoirs are investigated. Both quantitative and qualitative results confirm that fully automatic large-scale inland water body mapping is generally possible from Sentinel data; whereas, the best result is achieved when all available surface-related bands of both Sentinel-1 and Sentinel-2 are fused on a pixel level. The main limitation arises from missed smaller water bodies, which are not observed in bands with a resolution of about 20 m. Given the simplicity of the proposed approach and the open availability of the Sentinel data, the study confirms the potential for a fully automatic large-scale mapping of inland water with cloud-based remote sensing techniques. © 2020, The Author(s)." "57132509000;23003259600;57189378782;16481561000;55597324400;8721610200;56463389100;57052137700;56375286600;55710921100;57208273241;57131646800;56068376200;","Construction of a virtual PM2.5 observation network in China based on high-density surface meteorological observations using the Extreme Gradient Boosting model",2020,"10.1016/j.envint.2020.105801","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085507158&doi=10.1016%2fj.envint.2020.105801&partnerID=40&md5=b5cce83aca2f79b6b73fa50796103d8d","With increasing public concerns on air pollution in China, there is a demand for long-term continuous PM2.5 datasets. However, it was not until the end of 2012 that China established a national PM2.5 observation network. Before that, satellite-retrieved aerosol optical depth (AOD) was frequently used as a primary predictor to estimate surface PM2.5. Nevertheless, satellite-retrieved AOD often encounter incomplete daily coverage due to its sampling frequency and interferences from cloud, which greatly affect the representation of these AOD-based PM2.5. Here, we constructed a virtual ground-based PM2.5 observation network at 1180 meteorological sites across China using the Extreme Gradient Boosting (XGBoost) model with high-density meteorological observations as major predictors. Cross-validation of the XGBoost model showed strong robustness and high accuracy in its estimation of the daily (monthly) PM2.5 across China in 2018, with R2, root-mean-square error (RMSE) and mean absolute error values of 0.79 (0.92), 15.75 μg/m3 (6.75 μg/m3) and 9.89 μg/m3 (4.53 μg/m3), respectively. Meanwhile, we find that surface visibility plays the dominant role in terms of the relative importance of variables in the XGBoost model, accounting for 39.3% of the overall importance. We then use meteorological and PM2.5 data in the year 2017 to assess the predictive capability of the model. Results showed that the XGBoost model is capable to accurately hindcast historical PM2.5 at monthly (R2 = 0.80, RMSE = 14.75 μg/m3), seasonal (R2 = 0.86, RMSE = 12.28 μg/m3), and annual (R2 = 0.81, RMSE = 10.10 μg/m3) mean levels. In general, the newly constructed virtual PM2.5 observation network based on high-density surface meteorological observations using the Extreme Gradient Boosting model shows great potential in reconstructing historical PM2.5 at ~1000 meteorological sites across China. It will be of benefit to filling gaps in AOD-based PM2.5 data, as well as to other environmental studies including epidemiology. © 2020 The Author(s)" "57216464399;55212747200;56472102300;57216460815;56463389100;36070629400;57217650709;","A robust segmentation framework for closely packed buildings from airborne LiDAR point clouds",2020,"10.1080/01431161.2020.1727053","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083579745&doi=10.1080%2f01431161.2020.1727053&partnerID=40&md5=595e1a9129ff07dad2a2fd1b9d90502c","Urban villages (UVs) are commonly found in many Asian cities. These villages contain many closely packed buildings constructed decades ago without proper urban planning. There is a need for those buildings to be identified and put into statistics. In this paper, we present a segmentation framework that invokes multiple machine learning techniques and point cloud/image processing algorithms to segment individual closely packed buildings from large urban scenes. The presented framework consists of two major segmentation processes. The framework first filters out the non-ground objects from the point cloud, then it classified them by using the Random Forest classifier to isolate buildings from the entire scene. After that, the building point clouds will be segmented based on several building attribute analysis methods. This is followed by using the Random Sample Consensus (RANSAC) plane filtering method to expand the space between two closely packed buildings, so that the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering technique can be used to more accurately segment each individual building from the closely packed building areas. Two airborne Light Detection and Ranging (LiDAR) datasets collected in two different cities with some typical closely packed buildings were used to verify the proposed framework. The results show that the framework can effectively identify the closely packed buildings with unified structures from large airborne LiDAR datasets. The overall segmentation accuracy reaches 84% for the two datasets. The proposed framework can serve as a basis for analysis and segmentation of closely packed buildings with a more complicated structure. © 2020, © 2020 Informa UK Limited, trading as Taylor & Francis Group." "57219411340;56276541300;55974891000;57189520569;57209799235;8835983800;","Learning vector quantization neural network for surface water extraction from Landsat OLI images",2020,"10.1117/1.JRS.14.032605","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092674089&doi=10.1117%2f1.JRS.14.032605&partnerID=40&md5=c67878c68d11e08c36cc76bf92513763","There is a growing concern over surface water dynamics due to an increased understanding of water availability and management with current climate trends. Remote sensing has now become an effective means of water extraction due to the availability of an enormous amount of data with diverse spatial, spectral, and temporal resolutions. However, water extraction from optical remote sensing data is associated with several major difficulties, such as the applicability of the extraction method over large areas and complex environments; shadow contamination from clouds, buildings, and mountains; and disclosure of shadowed water and exclusion of floating and submerged plants. To address these difficulties, a learning vector quantization (LVQ) neural network-based method was proposed and implemented to extract water using Landsat 8 imageries. This method is capable of separating water from clouds, build-up areas, shadows, and shadowed water by the ideal input of bands 1 to 7 and normalized difference vegetation index. This model learns water across Sri Lanka. Eight OLI scenes were tested, and the performance was compared with five widely used machine learning algorithms: support vector machine, K-nearest neighbor, discriminant analysis, combination of modified normalized difference water index and modified fuzzy clustering method, and K-means clustering methods. This method performed the best, achieving overall accuracies and the kappa coefficients between 97.8% and 99.7% and between 0.96 and 0.99, respectively. Results have demonstrated robustness, consistency, and preciseness in various dark surfaces, noisiest water environments, and highly water scarce scenes. LVQ revealed a good generalizing ability to detect all types of water with less amount of training samples. This method can be easily adaptable for other sensors and global water to support water resource studies. © The Authors. Published by SPIE under a Creative Commons Attribution 4.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI." "35179029000;57202511009;57202513399;56269844200;7006360842;","Deep reinforcement learning for the management of software-defined networks and network function virtualization in an Edge-IoT architecture",2020,"10.3390/su12145706","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088836863&doi=10.3390%2fsu12145706&partnerID=40&md5=18e4e88674d12b371cea2bcbe24d4e5b","The Internet of Things (IoT) paradigm allows the interconnection of millions of sensor devices gathering information and forwarding to the Cloud, where data is stored and processed to infer knowledge and perform analysis and predictions. Cloud service providers charge users based on the computing and storage resources used in the Cloud. In this regard, Edge Computing can be used to reduce these costs. In Edge Computing scenarios, data is pre-processed and filtered in network edge before being sent to the Cloud, resulting in shorter response times and providing a certain service level even if the link between IoT devices and Cloud is interrupted. Moreover, there is a growing trend to share physical network resources and costs through Network Function Virtualization (NFV) architectures. In this sense, and related to NFV, Software-Defined Networks (SDNs) are used to reconfigure the network dynamically according to the necessities during time. For this purpose, Machine Learning mechanisms, such as Deep Reinforcement Learning techniques, can be employed to manage virtual data flows in networks. In this work, we propose the evolution of an existing Edge-IoT architecture to a new improved version in which SDN/NFV are used over the Edge-IoT capabilities. The proposed new architecture contemplates the use of Deep Reinforcement Learning techniques for the implementation of the SDN controller. © 2020 by the authors." "57209655847;57204570329;57210343076;57217728843;57210343863;55315026400;9277159100;","Automated 3D reconstruction using optimized view-planning algorithms for iterative development of structure-from-motion models",2020,"10.3390/rs12132169","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087546220&doi=10.3390%2frs12132169&partnerID=40&md5=67dc55d346e11be144c7a74bd4c80582","Unsupervised machine learning algorithms (clustering, genetic, and principal component analysis) automate Unmanned Aerial Vehicle (UAV) missions as well as the creation and refinement of iterative 3D photogrammetric models with a next best view (NBV) approach. The novel approach uses Structure-from-Motion (SfM) to achieve convergence to a specified orthomosaic resolution by identifying edges in the point cloud and planning cameras that ""view"" the holes identified by edges without requiring an initial model. This iterative UAV photogrammetric method successfully runs in various Microsoft AirSim environments. Simulated ground sampling distance (GSD) of models reaches as low as 3.4 cm per pixel, and generally, successive iterations improve resolution. Besides analogous application in simulated environments, a field study of a retired municipal water tank illustrates the practical application and advantages of automated UAV iterative inspection of infrastructure using 63% fewer photographs than a comparable manual flight with analogous density point clouds obtaining a GSD of less than 3 cm per pixel. Each iteration qualitatively increases resolution according to a logarithmic regression, reduces holes in models, and adds details to model edges. © 2020 by the authors." "55236333500;55602536800;35186195400;57195959155;57192094610;8684670500;","Machine learning algorithms to predict tree-related microhabitats using airborne laser scanning",2020,"10.3390/rs12132142","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087544650&doi=10.3390%2frs12132142&partnerID=40&md5=c3cc526a714606d8406481340d0a79d6","In the last few years, the occurrence and abundance of tree-related microhabitats and habitat trees have gained great attention across Europe as indicators of forest biodiversity. Nevertheless, observing microhabitats in the field requires time and well-trained staff. For this reason, new efficient semiautomatic systems for their identification and mapping on a large scale are necessary. This study aims at predicting microhabitats in a mixed and multi-layered Mediterranean forest using Airborne Laser Scanning data through the implementation of a Machine Learning algorithm. The study focuses on the identification of LiDAR metrics useful for detecting microhabitats according to the recent hierarchical classification system for Tree-related Microhabitats, from single microhabitats to the habitat trees. The results demonstrate that Airborne Laser Scanning point clouds support the prediction of microhabitat abundance. Better prediction capabilities were obtained at a higher hierarchical level and for some of the single microhabitats, such as epiphytic bryophytes, root buttress cavities, and branch holes. Metrics concerned with tree height distribution and crown density are the most important predictors of microhabitats in a multi-layered forest. © 2020 by the authors." "57204706166;36627288300;57203752278;7003582919;56519829700;57217530999;24825059000;24825059000;56414357900;7006481399;36100657400;23989037500;23989037500;","Information content of jwst nirspec transmission spectra of warm neptunes",2020,"10.3847/1538-3881/ab9176","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087346234&doi=10.3847%2f1538-3881%2fab9176&partnerID=40&md5=da95a5c6e9d823cec49e0b72ac2aa239","Warm Neptunes offer a rich opportunity for understanding exo-atmospheric chemistry. With the upcoming James Webb Space Telescope (JWST), there is a need to elucidate the balance between investments in telescope time versus scientific yield. We use the supervised machine-learning method of the random forest to perform an information content (IC) analysis on a 11-parameter model of transmission spectra from the various NIRSpec modes. The three bluest medium-resolution NIRSpec modes (0.7-1.27 μm, 0.97-1.84 μm, 1.66-3.07 μm) are insensitive to the presence of CO. The reddest medium-resolution mode (2.87-5.10 μm) is sensitive to all of the molecules assumed in our model: CO, CO2, CH4, C2H2, H2O, HCN, and NH3. It competes effectively with the three bluest modes on the information encoded on cloud abundance and particle size. It is also competitive with the low-resolution prism mode (0.6-5.3 μm) on the inference of every parameter except for the temperature and ammonia abundance. We recommend astronomers to use the reddest medium-resolution NIRSpec mode for studying the atmospheric chemistry of 800-1200 K warm Neptunes; its corresponding high-resolution counterpart offers diminishing returns. We compare our findings to previous JWST IC analyses that favor the blue orders and suggest that the reliance on chemical equilibrium could lead to biased outcomes if this assumption does not apply. A simple, pressure-independent diagnostic for identifying chemical disequilibrium is proposed based on measuring the abundances of H2O, CO, and CO2. © 2020. The Author(s). Published by the American Astronomical Society" "57208411135;35576373500;55234984500;35098639300;23110503000;15081795000;57217492400;57217489905;","Wheat Area Mapping in Afghanistan Based on Optical and SAR Time-Series Images in Google Earth Engine Cloud Environment",2020,"10.3389/fenvs.2020.00077","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087306998&doi=10.3389%2ffenvs.2020.00077&partnerID=40&md5=a40a9eb808b6884de49516a04af61e9d","Wheat is cultivated on more than 2.7 million hectares in Afghanistan annually, yet the country is dependent on imports to meet domestic demand. The timely estimation of domestic wheat production is highly critical to address any potential food security issues and has been identified as a priority by the Ministry of Agriculture Irrigation and Livestock (MAIL). In this study, we developed a system for in-season mapping of wheat crop area based on both optical (Sentinel-2) and synthetic aperture radar (SAR, Sentinel-1) data to support estimation of wheat cultivated area for management and food security planning. Utilizing a 2010 Food and Agriculture Organization (FAO) cropland mask, wheat sown area for 2017 was mapped integrating decision trees and machine learning algorithms in the Google Earth Engine cloud platform. Information from provincial crop calendars in addition to training and validation data from field-based surveys, and high-resolution Digitalglobe and Airbus Pleiades images were used for classification and validation. The total irrigated and rainfed wheat area were estimated as 912,525 and 562,611 ha, respectively for 2017. Province-wise accuracy assessments show the maximum accuracy of irrigated (IR) and rainfed (RF) wheat across provinces was 98.76 and 99%, respectively, whereas the minimum accuracy was found to be 48% (IR) and 73% (RF). The lower accuracy is attributed to the unavailability of reference data, cloud cover in the satellite images and overlap of spectral reflectance of wheat with other crops, especially in the opium poppy growing provinces. While the method is designed to provide estimation at different stages of the growing season, the best accuracy is achieved at the end of harvest using time-series satellite data for the whole season. The approach followed in the study can be used to generate wheat area maps for other years to aid in food security planning and policy decisions. © Copyright © 2020 Tiwari, Matin, Qamer, Ellenburg, Bajracharya, Vadrevu, Rushi and Yusafi." "57217281424;57217281247;16068804700;6507509159;","Generating elevation surface from a single RGB remotely sensed image using deep learning",2020,"10.3390/rs12122002","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087000238&doi=10.3390%2frs12122002&partnerID=40&md5=9ed85b9d4bff82ed6c458fa5d78c05c5","Generating Digital Elevation Models (DEM) from satellite imagery or other data sources constitutes an essential tool for a plethora of applications and disciplines, ranging from 3D flight planning and simulation, autonomous driving and satellite navigation, such as GPS, to modeling water flow, precision farming and forestry. The task of extracting this 3D geometry from a given surface hitherto requires a combination of appropriately collected corresponding samples and/or specialized equipment, as inferring the elevation from single image data is out of reach for contemporary approaches. On the other hand, Artificial Intelligence (AI) and Machine Learning (ML) algorithms have experienced unprecedented growth in recent years as they can extrapolate rules in a data-driven manner and retrieve convoluted, nonlinear one-to-one mappings, such as an approximate mapping from satellite imagery to DEMs. Therefore, we propose an end-to-end Deep Learning (DL) approach to construct this mapping and to generate an absolute or relative point cloud estimation of a DEM given a single RGB satellite (Sentinel-2 imagery in this work) or drone image. The model has been readily extended to incorporate available information from the non-visible electromagnetic spectrum. Unlike existing methods, we only exploit one image for the production of the elevation data, rendering our approach less restrictive and constrained, but suboptimal compared to them at the same time. Moreover, recent advances in software and hardware allow us to make the inference and the generation extremely fast, even on moderate hardware. We deploy Conditional Generative Adversarial networks (CGAN), which are the state-of-the-art approach to image-to-image translation. We expect our work to serve as a springboard for further development in this field and to foster the integration of such methods in the process of generating, updating and analyzing DEMs. © 2020 by the authors." "57191851405;","Coupled online learning as a way to tackle instabilities and biases in neural network parameterizations: General algorithms and Lorenz 96 case study (v1.0)",2020,"10.5194/gmd-13-2185-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084740461&doi=10.5194%2fgmd-13-2185-2020&partnerID=40&md5=e8e1286f7da138978f140bd5ee1b6d5b","Over the last couple of years, machine learning parameterizations have emerged as a potential way to improve the representation of subgrid processes in Earth system models (ESMs). So far, all studies were based on the same three-step approach: first a training dataset was created from a high-resolution simulation, then a machine learning algorithm was fitted to this dataset, before the trained algorithm was implemented in the ESM. The resulting online simulations were frequently plagued by instabilities and biases. Here, coupled online learning is proposed as a way to combat these issues. Coupled learning can be seen as a second training stage in which the pretrained machine learning parameterization, specifically a neural network, is run in parallel with a high-resolution simulation. The high-resolution simulation is kept in sync with the neural network-driven ESM through constant nudging. This enables the neural network to learn from the tendencies that the high-resolution simulation would produce if it experienced the states the neural network creates. The concept is illustrated using the Lorenz 96 model, where coupled learning is able to recover the ""true"" parameterizations. Further, detailed algorithms for the implementation of coupled learning in 3D cloud-resolving models and the super parameterization framework are presented. Finally, outstanding challenges and issues not resolved by this approach are discussed.
. © Author(s) 2020." "57216972010;55719407900;56415300800;57216969459;56720614400;","Mapping rice paddy based on machine learning with sentinel-2 multi-temporal data: Model comparison and transferability",2020,"10.3390/rs12101620","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085583302&doi=10.3390%2frs12101620&partnerID=40&md5=ac14d222ed7eeaa8f56a2ed7e308fd5f","Rice is an important agricultural crop in the Southwest Hilly Area, China, but there has been a lack of efficient and accurate monitoring methods in the region. Recently, convolutional neural networks (CNNs) have obtained considerable achievements in the remote sensing community. However, it has not been widely used in mapping a rice paddy, and most studies lack the comparison of classification effectiveness and efficiency between CNNs and other classic machine learning models and their transferability. This study aims to develop various machine learning classification models with remote sensing data for comparing the local accuracy of classifiers and evaluating the transferability of pretrained classifiers. Therefore, two types of experiments were designed: local classification experiments and model transferability experiments. These experiments were conducted using cloud-free Sentinel-2 multi-temporal data in Banan District and Zhongxian County, typical hilly areas of Southwestern China. A pure pixel extraction algorithm was designed based on land-use vector data and a Google Earth Online image. Four convolutional neural network (CNN) algorithms (one-dimensional (Conv-1D), two-dimensional (Conv-2D) and three-dimensional (Conv-3D_1 and Conv-3D_2) convolutional neural networks) were developed and compared with four widely used classifiers (random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM) and multilayer perceptron (MLP). Recall, precision, overall accuracy (OA) and F1 score were applied to evaluate classification accuracy. The results showed that Conv-2D performed best in local classification experiments with OA of 93.14% and F1 score of 0.8552 in Banan District, OA of 92.53% and F1 score of 0.8399 in Zhongxian County. CNN-based models except Conv-1D provided more desirable performance than non-CNN classifiers. Besides, among the non-CNN classifiers, XGBoost received the best result with OA of 89.73% and F1 score of 0.7742 in Banan District, SVM received the best result with OA of 88.57% and F1 score of 0.7538 in Zhongxian County. In model transferability experiments, almost all CNN classifiers had low transferability. RF and XGBoost models have achieved acceptable F1 scores for transfer (RF = 0.6673 and 0.6469, XGBoost = 0.7171 and 0.6709, respectively). © 2020 by the authors." "57200795580;22941140200;55885604100;","Feature selection for airborne LiDAR data filtering: a mutual information method with Parzon window optimization",2020,"10.1080/15481603.2019.1695406","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076344149&doi=10.1080%2f15481603.2019.1695406&partnerID=40&md5=64141d135c23cd9eb64fa12a4ce9aaa1","Filtering is one of the key steps for Digital Elevation Model (DEM) generation from airborne Light Detection and Ranging (LiDAR) data. Machine-learning-based filters have emerged as a class of filtering algorithms in recent years. Most existing studies mainly focus on feature generation due to limited available features a point cloud possesses. More than 30 features have been described in the existing literature. But most generated features are based on geometric information of points. Several redundant and irrelevant features may not necessarily improve the filtering accuracy. Hence, this paper proposes a feature-selection method using minimal-Redundancy-Maximal-Relevance (mRMR) combined with Parzen window optimization to deal with both discrete and continuous features. An optimal/suboptimal feature subset is constructed for machine-learning filters in various landscapes. Experimental results based on AdaBoost show that height-related features, particularly height itself, are of the greatest significance in both urban and rural scenes. Moreover, different subsets can be selected from the datasets of the two landscapes by our feature-selection strategy, which increases the data relevance for describing each geographical landscape. This study provides guidelines for the selection of optimal/suboptimal features for point cloud filtering based on machine-learning algorithms. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group." "57194612091;57195695111;55362143700;23005893600;52263795700;57216910936;57216910905;7003276131;57216911316;23397202200;15059097100;","Generalized sparse convolutional neural networks for semantic segmentation of point clouds derived from tri-stereo satellite imagery",2020,"10.3390/RS12081289","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085257464&doi=10.3390%2fRS12081289&partnerID=40&md5=f0f82bc389ed2df4fba4ff7e57aee7cb","We studied the applicability of point clouds derived from tri-stereo satellite imagery for semantic segmentation for generalized sparse convolutional neural networks by the example of an Austrian study area. We examined, in particular, if the distorted geometric information, in addition to color, influences the performance of segmenting clutter, roads, buildings, trees, and vehicles. In this regard, we trained a fully convolutional neural network that uses generalized sparse convolution one time solely on 3D geometric information (i.e., 3D point cloud derived by dense image matching), and twice on 3D geometric as well as color information. In the first experiment, we did not use class weights, whereas in the second we did. We compared the results with a fully convolutional neural network that was trained on a 2D orthophoto, and a decision tree that was once trained on hand-crafted 3D geometric features, and once trained on hand-crafted 3D geometric as well as color features. The decision tree using hand-crafted features has been successfully applied to aerial laser scanning data in the literature. Hence, we compared our main interest of study, a representation learning technique, with another representation learning technique, and a non-representation learning technique. Our study area is located in Waldviertel, a region in Lower Austria. The territory is a hilly region covered mainly by forests, agriculture, and grasslands. Our classes of interest are heavily unbalanced. However, we did not use any data augmentation techniques to counter overfitting. For our study area, we reported that geometric and color information only improves the performance of the Generalized Sparse Convolutional Neural Network (GSCNN) on the dominant class, which leads to a higher overall performance in our case. We also found that training the network with median class weighting partially reverts the effects of adding color. The network also started to learn the classes with lower occurrences. The fully convolutional neural network that was trained on the 2D orthophoto generally outperforms the other two with a kappa score of over 90% and an average per class accuracy of 61%. However, the decision tree trained on colors and hand-crafted geometric features has a 2% higher accuracy for roads. © 2020 by the authors." "57054207300;55780256400;","Integrating landsat-8 and sentinel-2 time series data for yield prediction of sugarcane crops at the block level",2020,"10.3390/RS12081313","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084642257&doi=10.3390%2fRS12081313&partnerID=40&md5=f7b5109e380d54bf42c52c8ae03c654c","Early prediction of sugarcane crop yield at the commercial block level (unit area of a single crop of the same variety, ratoon or planting date) offers significant benefit to growers, consultants, millers, policy makers, crop insurance companies and researchers. This current study explored a remote sensing based approach for predicting sugarcane yield at the block level by further developing a regionally specific Landsat time series model and including individual crop sowing (or previous seasons' harvest) date. For the Bundaberg growing region of Australia this extends over a five months period, from July to November. For this analysis, the sugarcane blocks were clustered into 10 groups based on their specific planting or ratoon commencement date within the specified five months period. These clustered or groups of blocks were named 'bins'. Cloud free (<20%) satellite data from the polar-orbiting Landsat-8 (launched 2013), Sentinel-2A (launched 2015) and Sentinel-2B (launched 2017) sensors were acquired over the cane growing region in Bundaberg (area of 32,983 ha), from the growing season starting in July 2014, with the average green normalised difference vegetation index (GNDVI) derived for each block. The number of images acquired for each season was defined by the number of cloud free acquisitions. Using the Simple Linear Machine Learning (ML) algorithm, the extracted Landsat derived GNDVI values for each of the blocks were converted to Sentinel GNDVI. The average GNDVI of each 'bin' was plotted and a quadratic model was fitted through the time series to identify the peak growth stage defined as the maximum GNDVI value. The model derived maximum GNDVI values for each of the bins were then regressed against the average actual yield (t·ha-1) achieved for the respective bin over the five growing years, producing strong correlations (R2 = 0.92 to 0.99). The quadratic curves developed for the different bins were shifted according to the specific planting or ratoon date of an individual block allowing for the peak GNDVI value of the block to be calculated, regressed against the actual block yield (t·ha-1) and the prediction of yield to be made. To validate the accuracies of the 10 time series algorithms representing each of the 10 bins, 592 individual blocks were selected from the Bundaberg region during the 2019 harvest season. The crops were clustered into the appropriate bins with the respective algorithm applied. From a Sentinel image acquired on the 5 May 2019, the prediction accuracies were encouraging (R2 = 0.87 and RMSE = 11.33 (t·ha-1)) when compared to actual harvested yield, as reported by the mill. The results presented in this paper demonstrate significant progress in the accurate prediction of sugarcane yield at the individual sugarcane block level using a remote sensing, time-series based approach. © 2020 by the authors." "57216790729;57216787681;57205495136;14053992900;55848021900;48261068400;50560994200;57216786213;55273559900;","Retrieval of aerodynamic parameters in rubber tree forests based on the computer simulation technique and terrestrial laser scanning data",2020,"10.3390/RS12081318","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084638859&doi=10.3390%2fRS12081318&partnerID=40&md5=90da3ce53ed63eb29857da3067ab3241","Rubber trees along the southeast coast of China always suffer severe damage from hurricanes. Quantitative assessments of the capacity for wind resistance of various rubber tree clones are currently lacking. We focus on a vulnerability assessment of rubber trees of different clones under wind disturbance impacts by employing multidisciplinary approaches incorporating scanned points, aerodynamics, machine learning and computer graphics. Point cloud data from two typical rubber trees belonging to different clones (PR107 and CATAS 7-20-59) were collected using terrestrial laser scanning, and a connection chain of tree skeletons was constructed using a clustering algorithm of machine learning. The concept of foliage clumps based on the trunk and first-order branches was first proposed to optimize rubber tree plot 3D modelling for simulating the wind field and assessing the wind-related parameters. The results from the obtained phenotypic traits show that the variable leaf area index and included angle between the branches and trunk result in variations in the topological structure and gap fraction of tree crowns, respectively, which are the major influencing factors relevant to the rubber tree's capacity to resist hurricane strikes. The aerodynamics analysis showed that the maximum dynamic pressure, wind velocity and turbulent intensity of the wind-related parameters in rubber tree plots of clone PR107 (300 Pa, 30 m/s and 15%) are larger than that in rubber tree plots of clone CATAS-7-20-59 (120 Pa, 18 m/s and 5%), which results in a higher probability of local strong cyclone occurrence and a higher vulnerability to hurricane damage. © 2020 by the authors." "55694702700;26423907800;56960514500;56368863100;57208054851;","Harmonized landsat 8 and sentinel-2 time series data to detect irrigated areas: An application in Southern Italy",2020,"10.3390/RS12081275","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084588520&doi=10.3390%2fRS12081275&partnerID=40&md5=7c8cb13264f96429dca8af2c17bca9b6","Lack of accurate and up-to-date data associated with irrigated areas and related irrigation amounts is hampering the full implementation and compliance of theWater Framework Directive (WFD). In this paper, we describe the framework that we developed and implemented within the DIANA project to map the actual extent of irrigated areas in the Campania region (Southern Italy) during the 2018 irrigation season. For this purpose, we considered 202 images from the Harmonized Landsat Sentinel-2 (HLS) products (57 images from Landsat 8 and 145 images from Sentinel-2). Such data were preprocessed in order to extract a multitemporal Normalized Difference Vegetation Index (NDVI) map, which was then smoothed through a gap-filling algorithm. We further integrated data coming from high-resolution (4 km) global satellite precipitation Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN)-Cloud Classification System (CCS) products. We collected an extensive ground truth in the field represented by 2992 data points coming from three main thematic classes: bare soil and rainfed (class 0), herbaceous (class 1), and tree crop (class 2). This information was exploited to generate irrigated area maps by adopting a machine learning classification approach. We compared six different types of classifiers through a cross-validation approach and found that, in general, random forests, support vector machines, and boosted decision trees exhibited the best performances in terms of classification accuracy and robustness to different tested scenarios. We found an overall accuracy close to 90% in discriminating among the three thematic classes, which highlighted promising capabilities in the detection of irrigated areas from HLS products. © 2020 by the authors." "57192504939;57218174545;56181390800;57202231708;57200511212;","Convective Clouds Extraction from Himawari-8 Satellite Images Based on Double-Stream Fully Convolutional Networks",2020,"10.1109/LGRS.2019.2926402","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082885920&doi=10.1109%2fLGRS.2019.2926402&partnerID=40&md5=c8b5406d9ff30a9740eed2e05158380e","Auto-extraction of convective clouds is of great significance. Convective clouds often bring heavy rain, strong winds, and other disastrous weather. Early warning of convection can effectively reduce loss. Using remote sensing images, we can get large-scale cloud information, which provides many effective methods for convective clouds detection. In this letter, we proposed a novel method to extract convective clouds. We introduce a novel deep network using only 1\times 1 convolution (3ONet) to extract the spectral characteristics. We then combine a 3ONet with the symmetrical dense-shortcut deep fully convolutional networks (SDFCNs) with a double-stream fully convolutional network to extract convective clouds. In the experiment, we used 12 000 Himawari-8 satellite image patches to verify the proposed framework. Experimental results with 0.5882 mean intersection over union (mIOU) pointed out the proposed method can extract convective clouds effectively. © 2004-2012 IEEE." "57215001908;35316130500;55882914500;34968949700;12144303500;23667564700;","A multi-resolution air temperature model for France from MODIS and Landsat thermal data",2020,"10.1016/j.envres.2020.109244","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079681488&doi=10.1016%2fj.envres.2020.109244&partnerID=40&md5=1a0e37aacb8aea9703499533265aa401","Understanding and managing the health effects of ambient temperature (Ta) in a warming, urbanizing world requires spatially- and temporally-resolved Ta at high resolutions. This is challenging in a large area like France which includes highly variable topography, rural areas with few weather stations, and heterogeneous urban areas where Ta can vary at fine spatial scales. We have modeled daily Ta from 2000 to 2016 at a base resolution of 1 km2 across continental France and at a 200 × 200 m2 resolution over large urban areas. For each day we predict three Ta measures: minimum (Tmin), mean (Tmean), and maximum (Tmax). We start by using linear mixed models to calibrate daily Ta observations from weather stations with remotely sensed MODIS land surface temperature (LST) and other spatial predictors (e.g. NDVI, elevation) on a 1 km2 grid. We fill gaps where LST is missing (e.g. due to cloud cover) with additional mixed models that capture the relationship between predicted Ta at each location and observed Ta at nearby weather stations. The resulting 1 km Ta models perform very well, with ten-fold cross-validated R2 of 0.92, 0.97, and 0.95, mean absolute error (MAE) of 1.4 °C, 0.9 °C, and 1.4 °C, and root mean square error (RMSE) of 1.9 °C, 1.3 °C, and 1.8 °C (Tmin, Tmean, and Tmax, respectively) for the initial calibration stage. To increase the spatial resolution over large urban areas, we train random forest and extreme gradient boosting models to predict the residuals (R) of the 1 km Ta predictions on a 200 × 200 m2 grid. In this stage we replace MODIS LST and NDVI with composited top-of-atmosphere brightness temperature and NDVI from the Landsat 5, 7, and 8 satellites. We use a generalized additive model to ensemble the random forest and extreme gradient boosting predictions with weights that vary spatially and by the magnitude of the predicted residual. The 200 m models also perform well, with ten-fold cross-validated R2 of 0.79, 0.79, and 0.85, MAE of 0.4, 0.3, and 0.3, and RMSE of 0.6, 0.4, and 0.5 (Rmin, Rmean, and Rmax, respectively). Our model will reduce bias in epidemiological studies in France by improving Ta exposure assessment in both urban and rural areas, and our methodology demonstrates that MODIS and Landsat thermal data can be used to generate gap-free timeseries of daily minimum, maximum, and mean Ta at a 200 × 200 m2 spatial resolution. © 2020 Elsevier Inc." "57203220499;55576827900;56674284600;","Generalization considerations and solutions for point cloud hillslope classifiers",2020,"10.1016/j.geomorph.2020.107039","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077919094&doi=10.1016%2fj.geomorph.2020.107039&partnerID=40&md5=d46d20aa106baf8c869298420a83d4d2","Point cloud classifiers have the potential to rapidly perform landscape characterization for a variety of applications. The generalization (i.e., transferability to new sites) of such classifiers could improve their accessibility and usefulness for both engineers and researchers alike, but guidelines for classifier generalization are lacking in the literature. This study develops and applies a Random Forest machine learning classifier for Terrestrial Laser Scanning (TLS) point clouds, and generalizes the classifier to point clouds from several different locations. The classifier is trained to identify basic hillslope topographic features, including vegetation, soil, talus, and bedrock using multi-scale geometric features of the point cloud. Four rock and soil slopes in western Colorado were scanned using TLS. Generalization experiments were performed testing point density, occlusion, and between-site domain variance factors, and all factors showed a significant influence on generalization accuracy. Several methods for improving classifier generalization accuracy were tested and compared, including combining training data from multiple sites, imposing probability thresholds, and a Domain Adaptation methodology known as Active Learning. It was found that incorporating data from multiple sites resulted in improved generalization accuracy, but in most cases the largest improvements in accuracy were associated with adding new training data from the target site. In this case, using Active Learning resulted in significant accuracy improvements with an over 90% reduction in the number of added training points. The results suggest that scanning characteristics are important factors in classifier generalization accuracy, but their effects can be mitigated by using the techniques described herein. © 2020 Elsevier B.V." "57202218489;57205157831;57192717976;15060638500;","Review of empirical solar radiation models for estimating global solar radiation of various climate zones of China",2020,"10.1177/0309133319867213","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85073822617&doi=10.1177%2f0309133319867213&partnerID=40&md5=7e5e3de90ff384db8f067e947aea2561","Utilizing solar energy requires accurate information about global solar radiation (GSR), which is critical for designers and manufacturers of solar energy systems and equipment. This study aims to examine the literature gaps by evaluating recent predictive models and categorizing them into various groups depending on the input parameters, and comprehensively collect the methods for classifying China into solar zones. The selected groups of models include those that use sunshine duration, temperature, dew-point temperature, precipitation, fog, cloud cover, day of the year, and different meteorological parameters (complex models). 220 empirical models are analyzed for estimating the GSR on a horizontal surface in China. Additionally, the most accurate models from the literature are summarized for 115 locations in China and are distributed into the above categories with the corresponding solar zone; the ideal models from each category and each solar zone are identified. Comments on two important temperature-based models that are presented in this work can help the researchers and readers to be unconfused when reading the literature of these models and cite them in a correct method in future studies. Machine learning techniques exhibit performance GSR estimation better than empirical models; however, the computational cost and complexity should be considered at choosing and applying these techniques. The models and model categories in this study, according to the key input parameters at the corresponding location and solar zone, are helpful to researchers as well as to designers and engineers of solar energy systems and equipment. © The Author(s) 2019." "57218571650;57199227774;","Diagnosis of Parkinson disease using sensor data and machine learning approach in mobile cloud",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089606302&partnerID=40&md5=e602e5daee888cb3fb8987fe25f2c8e3","In reality, the utilization of sensors is developing each day to improve the personal satisfaction by giving medicinal services data on clinical diagnostics. There are various sensors like detecting advance sensors, ""electronic gadgets"" physical sensors have been effectively shown in the field of biomedical applications because of sensor great working ability. Cloud computing technology is used to accommodate vast amounts of data. Such medical applications also use the cloud computing platform to store and access the data protected. Parkinson's disease (PD) could also be a chronic long condition of the central systema nervosum that primarily affects the motor system of the patient. The side effects of Parkinsons’ sickness incorporate muscle rigidity, tremors and modifications in speech. The objective of the framework is to analyse the parkinson’s disease in voice detection. As the indications are worsen, the patients motor and non-motor system will get failure. The framework is used for diagnosing the parkinson’s disease. The proposed system concentrates on improving the Parkinson's disease diagnosis using voice signals from patient detected from different sensors and uploaded to the cloud for processing. This frame work concentrates on improving disease diagnosis with experimental results using Support Vector Machine, Logistic Regression, Random Forest and eXtreme Gradient Boosting.Machine learning algorithm mainly concentrates on improving the prediction of parkinson’s disease diagnosis. © 2020 Alpha Publishers. All rights reserved." "57217412518;","Beyond earthworm: Keeping the promise",2020,"10.1785/0220190198","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089527994&doi=10.1785%2f0220190198&partnerID=40&md5=1dfedaec6042d4d4e631301371124920","The founding of the Advanced National Seismic System (ANSS) vision was originally presented in U.S. Geological Survey Circular 1188 (U.S. Geological Survey [USGS], 1999), after many years of discussions and workshops, described in detail by Filson and Arabasz (2016). Much has been accomplished in the ensuing two decades. Disparate and sometimes divergent developments that had been previously explored at individual private and public universities were finally centralized with increased efficiency and coherency of effort. The stated mission of the ANSS is to ""... provide accurate and timely data and information products for seismic events, including their effects on buildings and structures, employing modern monitoring methods and technologies.""In this article, an approach (xQuake) is proposed that does not interfere in any way with the mission of the National Earthquake Information Center and ANSS but instead restores much of the community focus and international collaboration that has been lost over the past two decades. xQuake uses an executable graph framework in a pipeline architecture; this framework can be seamlessly integrated into current ANSS quake monitoring systems. This newapproach incorporatesmodern approaches to computer analytics, includingmultitopic Kafka exchange rings, cloud computing, a self-configuring phase associator, and machine learning. The xGraph system is free for noncommercial use, open source, hardware agnostic (Windows, Linux, Mac), with no requirement for commercial datastores. © Seismological Society of America." "55576700800;57205299839;57200596843;35098748100;54785899100;","All-sky longwave downward radiation from satellite measurements: General parameterizations based on LST, column water vapor and cloud top temperature",2020,"10.1016/j.isprsjprs.2020.01.011","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077787169&doi=10.1016%2fj.isprsjprs.2020.01.011&partnerID=40&md5=b6042f028815807da31f4dbcd4f4c7c2","Remotely sensed surface longwave downward radiation (LWDR) plays an essential role in studying the surface energy budget and greenhouse effect. Most existing satellite-based methods or products depend on variables that are not readily available from space such as, liquid water path, air temperature, vapor pressure and/or cloud-base temperature etc., which seriously restrict the wide applications of satellite data. In this paper, new nonlinear parameterizations and a machine learning-based model for deriving all-sky LWDR are proposed based only on land surface temperature (LST), column water vapor and cloud-top temperature (CTT), that are relatively readily available day and night for most satellite missions. It is the first time to incorporate the CTT in the parameterizations for estimating LWDR under the cloudy-sky conditions. The results reveal that the new models work well and can derive all-sky global LWDR with reasonable accuracies (RMSE <23 W/m2, bias <2.0 W/m2). The convenience of input data makes the new models easy to use, and thus will definitely expand the applicability of remotely sensed measurements in radiation budget fields and many land applications. © 2020" "56768656200;57214920823;55986546100;36663710700;57211922206;55934244200;","Discriminative feature learning constrained unsupervised network for cloud detection in remote sensing imagery",2020,"10.3390/rs12030456","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85080882325&doi=10.3390%2frs12030456&partnerID=40&md5=afa54c8f3017e4109c318b6d2596b840","Cloud detection is a significant preprocessing step for increasing the exploitability of remote sensing imagery that faces various levels of difficulty due to the complexity of underlying surfaces, insufficient training data, and redundant information in high-dimensional data. To solve these problems, we propose an unsupervised network for cloud detection (UNCD) on multispectral (MS) and hyperspectral (HS) remote sensing images. The UNCD method enforces discriminative feature learning to obtain the residual error between the original input and the background in deep latent space, which is based on the observation that clouds are sparse and modeled as sparse outliers in remote sensing imagery. The UNCD enforces discriminative feature learning to obtain the residual error between the original input and the background in deep latent space, which is based on the observation that clouds are sparse and modeled as sparse outliers in remote sensing imagery. First, a compact representation of the original imagery is obtained by a latent adversarial learning constrained encoder. Meanwhile, the majority class with sufficient samples (i.e., background pixels) is more accurately reconstructed than the clouds with limited samples by the decoder. An image discriminator is used to prevent the generalization of out-of-class features caused by latent adversarial learning. To further highlight the background information in the deep latent space, a multivariate Gaussian distribution is introduced. In particular, the residual error with clouds highlighted and background samples suppressed is applied in the cloud detection in deep latent space. To evaluate the performance of the proposed UNCD method, experiments were conducted on both MS and HS datasets that were captured by various sensors over various scenes, and the results demonstrate its state-of-the-art performance. The sensors that captured the datasets include Landsat 8, GaoFen-1 (GF-1), and GaoFen-5 (GF-5). Landsat 8 was launched at Vandenberg Air Force Base in California on 11 February 2013, in a mission that was initially known as the Landsat Data Continuity Mission (LDCM). China launched the GF-1 satellite. The GF-5 satellite captures hyperspectral observations in the Chinese Key Projects of High-Re solution Earth Observation System. The overall accuracy (OA) values for Images I and II from the Landsat 8 dataset were 0.9526 and 0.9536, respectively, and the OA values for Images III and IV from the GF-1 wide field of view (WFV) dataset were 0.9957 and 0.9934, respectively. Hence, the proposed method outperformed the other considered methods. © 2020 by the authors." "7003480828;13607413100;","Re-imagining the future of education in the era of the fourth industrial revolution",2020,"10.1108/WHATT-10-2019-0066","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079449284&doi=10.1108%2fWHATT-10-2019-0066&partnerID=40&md5=8c8c6c6207f57561896d01db0fd7f984","Purpose: Globally, higher education has been, over the years, a source of innovation, policy, new knowledge and a national asset. However, the advent of the Fourth Industrial Revolution (4IR) is having an impact on the principles of learning from primary to tertiary levels. The purpose of this paper is to consider how the 4IR has and will continue to impact education at the various levels of learning. Design/methodology/approach: The paper aims to bridge the perceived information gap and provide insights into the kinds of educational preparation and the skills and qualifications that 4IR jobs require. In response, the following are considered: the need to tweak the curriculum, adopt the right technology for in class and online delivery and the projection of other learning techniques and skills that are often not considered pertinent. Data gathering for the report was by discussion with experts and consultation of relevant articles and write-ups from related websites. Findings: The advent of smart communication systems involving artificial intelligence, internet, robotics, virtual reality and digital textbooks has opened a new vista in relation to how and what is learnt in schools. Just as technologies brought about smart communication systems, the 4IR model of higher education is rapidly evolving and as such, curriculum development and review must be dynamic, and it must keep pace with the technological advances and skills required in the twenty first century. Research limitations/implications: More purposeful research needs to be conducted in universities and industries with the intention of accelerating internal and external innovations so that markets can be expanded. Furthermore, efforts to reduce the cost and time of generating innovations will need to be intensified. Practical implications: The value and emphasis that are placed on the acquisition of degrees and paper qualifications are changing rapidly. Although it is traditional for students to compete for admission to the face-to-face classroom model, it is no longer unusual for a student to take courses online from any part of the world and still be accepted into positions usually reserved for traditional classroom education. Social implications: As at today, examples of 4IR services include Uber, Airbnb, Cloud services, Artificial intelligence, Cyber-security, three-dimensional printers, driverless cars and robotics. Machine learning and drone technology are also of growing significance. As yet, subjects dealing with such inventions and innovations are not part of the curriculum of many institutions and this is a cause for concern. Originality/value: The 4IR era will bring great changes to how students are taught and what students must learn as the tools for transformational learning are already overwhelming. Jobs will be scarce for those without the requisite skills, whereas those with the right skills will have to keep up with the pace of technological development, otherwise they too will be left behind. Schools will increasingly become centres for the generation of innovation and its incubation and in all this, quality learning, teaching and knowledge impartation can easily be carried out online. © 2020, Emerald Publishing Limited." "56789262700;48662599000;8719489000;56654564800;7006430057;23767612400;56421152600;","Forest signal detection for photon counting LiDAR using Random Forest",2020,"10.1080/2150704X.2019.1682708","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074241228&doi=10.1080%2f2150704X.2019.1682708&partnerID=40&md5=ef3686a58bba4b1b5693c679fd3e9a77","ICESat (The Ice, Cloud, and Land Elevation Satellite)-2, as the new generation of NASA (National Aeronautics and Space Administration)’s ICESat mission, had been successfully launched in September 2018. The sensor onboard the satellite is a newly designed photon counting LiDAR (Light Detection And Ranging) system for the first time used in space. From the currently released airborne simulation data, it can be seen that there exist numerous noise photons scattering from the atmosphere to even below the ground, especially for the vegetation areas. Therefore, relevant research on methods to distinguish the signal photons effectively is crucial for further forestry applications. In this paper, a machine learning based approach was proposed to detect the potential signal photons from 14 MATLAS datasets across 3 sites in the USA. We found that k-NN (k-Nearest Neighbour) distance and the reachability of the photon towards the nearby signal centre showed good stability and contributed to a robust model establishment. The relevant quantitative assessment demonstrated that the machine learning approach could achieve high detection accuracy over 85% based on a very limited number of samples even in rough terrain conditions. Further analysis proved the potential of model transferability across different sites. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group." "57196237174;57213029726;55939316400;","MODIS probabilistic cloud masking over the Amazonian evergreen tropical forests: a comparison of machine learning-based methods",2020,"10.1080/01431161.2019.1637963","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85068638824&doi=10.1080%2f01431161.2019.1637963&partnerID=40&md5=75faac6f917fac230870f6a65f3e0810","Amazonian tropical forests play a significant role in global water, carbon and energy cycles. Satellite remote sensing is presented as a feasible means in order to monitor these forests. In particular, the Moderate Resolution Imaging Spectroradiometer (MODIS) is amongst major tools for studying this region. Nevertheless, MODIS operative surface variable retrieval was reported to be impacted by cloud contamination effects. A proper cloud masking is a major consideration in order to ensure accuracy when analysing Amazonian tropical forests current and future status. In the present study, the potential of supervised machine learning algorithms in order to overcome this issue is evaluated. In front of global operative MODIS cloud masking algorithms (MYD35 and the Multi-Angle Implementation of Atmospheric Correction Algorithm (MAIAC)) these algorithms benefit from the fact that they can be optimized to properly represent the local cloud conditions of the region. Models considered were: Gaussian Naïve Bayes (GNB), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Random Forests (RF), Support Vector Machine (SVM) and Multilayer Perceptron (MLP). These algorithms are able to provide a continuous measure of cloud masking uncertainty (i.e. a probability estimate of each pixel belonging to clear and cloudy class) and therefore can be used for probabilistic cloud masking. Truth reference dataset (a priori knowledge) requirement was satisfied by considering the collocation of Cloud Profiling Radar (CPR) and Cloud Aerosol Lidar with Orthogonal Polarization (CALIOP) observations with MODIS sensor. Model performance was tested using three independent datasets: 1) collocated CPR/CALIOP and MODIS data, 2) MODIS manually classified images and 3) in-situ ground data. For satellite image and in-situ testing results were additionally compared to current operative MYD35 (version 6.1) and MAIAC cloud masking algorithms. Satellite image and in-situ testing results show that machine learning algorithms are able to improve MODIS operative cloud masking performance over the region. MYD35 and MAIAC tend to underestimate and overestimate the cloud cover over the study region, respectively. Amongst the models considered, probabilistic algorithms (LDA, GNB and in less extent QDA) provided better performance than RF, SVM and MLP machine learning algorithms as they were able to better deal with the viewing conditions limitation that resulted from collocating MODIS and CPR/CALIOP observations. In particular, best performance was obtained for LDA with a difference in Kappa coefficient (model minus MODIS operative algorithm) of 0.293/0.155 (MYD35/MAIAC, respectively) considering satellite image testing validation. Worst performance was obtained for MLP with a difference in Kappa coefficient of 0.175/0.037. For in-situ testing, models overall accuracy (OA) and Kappa coefficient values are higher than MYD35/MAIAC respective values. Models are computationally efficient (swath image calculation time between 0.37 and 9.49 s) and thus being able to be implanted for remote-sensing vegetation retrieval processing chains over the Amazonian tropical forests. LDA stands out as the best candidate because of its maximum accuracy and minimum computational associated. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group." "57212584870;57188717852;55277716700;56664552300;57219294173;57219298103;57211501464;56088674100;7404589013;","Improved Mapping Results of 10 m Resolution Land Cover Classification in Guangdong, China Using Multisource Remote Sensing Data with Google Earth Engine",2020,"10.1109/JSTARS.2020.3022210","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092112240&doi=10.1109%2fJSTARS.2020.3022210&partnerID=40&md5=34ab9cea870c0e6677aefb0bf0f5e73b","Land cover information depicting the complex interactions between human activities and surface change is critically essential for nature conservation, social management, and sustainable development. Recent advances have shown great potentials of remote sensing data in generating high-resolution land cover maps, but it remains unclear how different models, data sources, and inclusive features affect the classification results, which hinders its applications in regional studies requiring more accurate land cover data. Informing these issues, here we developed a robust framework to improve the mapping results of 10 m resolution land cover classification in Guangdong Province, China using thousands of manually collected samples, multisource remote sensing data (Sentinel-1, Sentinel-2, and Luojia-1), machine learning algorithms, and a free cloud-based platform of Google Earth Engine. Results showed that an overall accuracy of 86.12% and a Kappa coefficient of 0.84 could be achieved for land cover classification in Guangdong for 2019. We found that random forest models achieved better performance than classification and regression trees, minimum distance, and support vector machine models. We also found that features derived from Sentinel-1 data and Sentinel-2 spectral indices greatly contributed to the classification process, while the feature of Luojia-1 data was not as much important as other configurations. A comparison between our results and several existing land cover products in terms of classification accuracy and visual interpretation further validated the effectiveness and robustness of the proposed framework. Our experiments and findings not only systematically elucidate the role of classification methods and data sources in deriving more accurate and reliable land cover maps but also provide certain guidelines for future land cover mapping from regional to global scales. © 2008-2012 IEEE." "7401526171;57218396843;35975568000;","Improving PERSIANN-CCS Using Passive Microwave Rainfall Estimation",2020,"10.1007/978-3-030-24568-9_21","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089068676&doi=10.1007%2f978-3-030-24568-9_21&partnerID=40&md5=cd6573882f174612b92fc8659e48e681","Re-calibrated PERSIANN-CCS is one of the algorithms used in “Integrated Multi-satellitE Retrievals for GPM” (IMERG) to provide high-resolution precipitation estimations from the NASA Global Precipitation Measurement (GPM) program and retrospective data generation for the period covered by the Tropical Rainfall Measurement Mission (TRMM). This study presents the development of a re-calibrated PERSIANN-CCS algorithm for the next-generation GPM multi-sensor precipitation retrieval algorithm (IMERG). The activities include implementing the probability matching method to update PERSIANN-CCS using passive microwave (PMW) rainfall estimation from low earth orbit (LEO) satellites and validation of precipitation estimation using radar rainfall measurement. Further improvement by the addition of warm rain estimation to the PERSIANN-CCS algorithm using warmer temperature thresholds for cloud image segmentation is also presented. Additionally, developments using multispectral image analysis and machine learning approaches are discussed and proposed for future studies. © 2020, Springer Nature Switzerland AG." "22433897700;57195300026;","Data-driven agriculture for rural smallholdings",2020,"10.5311/JOSIS.2020.20.669","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088784194&doi=10.5311%2fJOSIS.2020.20.669&partnerID=40&md5=842ca6f59783a61aa7ef1fb6ccdab3e3","Spatial information science has a critical role to play in meeting the major challenges facing society in the coming decades, including feeding a population of 10 billion by 2050, addressing environmental degradation, and acting on climate change. Agriculture and agri-food value-chains, dependent on spatial information, are also central. Due to agriculture's dual role as not only a producer of food, fibre, and fuel but also as a major land, water, and energy consumer, agriculture is at the centre of both the food-water-energyenvironment nexus and resource security debates. The recent confluence of a number of advances in data analytics, cloud computing, remote sensing, computer vision, robotic and drone platforms, and Internet of Things (IoT) sensors and networks have lead to a significant reduction in the cost of acquiring and processing data for decision support in the agricultural sector. When combined with costeffective automation through development of swarm farming technologies, these technologies have the potential to decouple productivity and cost efficiency from economies of size, reducing the need to increase farm size to remain economically viable. We argue that these pressures and opportunities are driving agricultural value-chains towards high-resolution data-driven decision-making, where even decisions made by small rural landowners can be data-driven. We survey recent innovations in data, especially focusing on sensor, spatial, and data mining technologies with a view to their agricultural application; discuss economic feasibility for small farmers; and identify some technical challenges that need to be solved to reap the benefits. Flexibly composable information resources, coupled with sophisticated data sharing technologies, and machine learning with transparently embedded spatial and aspatial methods are all required. © by the author(s)." "57200377471;55811589800;57215038513;","Transitioning a legacy reservoir simulator to cloud native services",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085775736&partnerID=40&md5=7210a84b869ed41ebe9062ef222c19f3","The digital transformation journey provides new opportunities for running simulations through cloud computing, with flexibility in hardware resources and availability of a wide array of software tools that can enhance decision- making. This work reviews the benefits of running reservoir simulations in a cloud environment and demonstrates the efficiency and cost savings. Additionally, a workflow for uncertainty analysis and history matching that integrates data analysis and machine-learning tools is presented. First, the hardware architecture should be designed to meet parallel reservoir simulation needs: significant message passing occurs between computer nodes; for satisfactory performance, these nodes should be connected by a low-latency network rather than randomly located. Second, to help ensure portability and easy replication across multiple cloud sites and platforms, the software performing the simulations should be containerized. Third, to reduce the time required to start a new simulation run, the Kubernetes platform is used to optimize resource allocation. Finally, reservoir simulation in the cloud is no longer merely the running of the simulation model; it is integrated with data management and data analysis tools for decision-making. The cloud-based simulation services discussed herein exhibit good results during scale up when a simulation operation requires a larger number of central processing units and/or greater memory and during scale out when thousands of operation scenarios are necessary for history matching. The ""pay as you go"" pricing model reduces the time and capital costs of acquiring the new computing infrastructure to nearly zero, and the effectively unlimited scale-out capability can reduce the elapsed time for history matching by 80%. The availability of data centers in different regions is good for team collaborations. It serves the data management tool well to track history data, perform data mining, extract more information, and make decisions. Compared to traditional reservoir simulation, the cloud-based reservoir simulation software as a service model simplifies the process and reduces hardware acquisition and maintenance costs. Integrating intelligent data analysis with simulation helps quantify the uncertainty in the model and enables improved decisions. Copyright 2020, International Petroleum Technology Conference." "57215574864;7006738371;35414779200;","A multilayer perceptron for obtaining quick parameter estimations of cool exoplanets from geometric albedo spectra",2020,"10.1088/1538-3873/ab740d","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85081251876&doi=10.1088%2f1538-3873%2fab740d&partnerID=40&md5=01f0d39d6c778d84ef4a132f03e5052a","Future space telescopes now in the concept and design stage aim to observe reflected light spectra of extrasolar planets. Assessing whether given notional mission and instrument design parameters will provide data suitable for constraining quantities of interest typically requires time consuming retrieval studies in which tens to hundreds of thousands of models are compared to data with a given assumed signal to noise ratio, thereby limiting the rapidity of design iterations. Here we present a machine learning approach employing a Multilayer Perceptron (MLP) trained on model albedo spectra of extrasolar giant planets to estimate a planet’s atmospheric metallicity, gravity, effective temperature, and cloud properties given simulated observed spectra. The stand-alone C++ code we have developed can train new MLP’s on new training sets within minutes to hours, depending upon the dimensions of input spectra, size of the training set, desired output, and desired accuracy. After the MLP is trained, it can classify new input spectra within a second, potentially helping speed observation and mission design planning. Our MLP’s were trained using a grid of model spectra that varied in metallicity, gravity, temperature, and cloud properties. The results show that a trained MLP is an elegant means for reliable in situ estimations when applied to model spectra. We analyzed the effect of using models in a grid range known to have degeneracies. © 2020. The Astronomical Society of the Pacific. All rights reserved." "57189756538;57212864756;57212876123;57217192429;55844950200;","Combining Participatory Mapping, Cloud Computing and Machine Learning for Mapping Climate Induced Landslide Susceptibility in Lembeh Island, North Sulawesi",2019,"10.1088/1755-1315/363/1/012020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077471375&doi=10.1088%2f1755-1315%2f363%2f1%2f012020&partnerID=40&md5=c65dcd92e32d5bb97ff84c13ee4a2c5a","This study explored the use of participatory mapping and several machine learning algorithms (Naïve Bayes, GMO Maxent, SVM, CART, and Random Forest) to map climate induced landslide susceptibility in Lembeh Island, North Sulawesi, based on Earth Observation data available in Google Earth Engine. Participatory mapping on landslide incidents were conducted in three villages, i.e., Kareko, Pintu Kota, and Pasir Panjang. Data used include digital elevation model from SRTM, multispectral imageries from Sentinel 2, and precipitation from CHIRPS. Terrain modelling was done to DEM to come up with elevation, slope, curvature, and aspect. A cloud free mosaic of Sentinel Images was created using the median reducer and then NDVI was calculated. Precipitation data from CHIRPS was sampled and interpolated using kriging and reduced to maximum and mean. Each algorithm was trained using 70% participatory mapping data and then the prediction was tested for accuracy using the last 30%. Results showed that Random Forest, SVM, CART, and GMO Maxent gave 0.98 testing accuracy and Naïve Bayes only 0.90. The map showed that due to the terrain condition, Lembeh Island is prone to Landslide and even though previously BNPB already provide a landslide hazard risk map, there were many areas not included on that map. Therefore, the map could become an input for BNPB and the Bitung City for developing a mitigation and adaptation strategy. Machine learning and cloud computing along with participatory mapping could also complement mechanistic or multi-criteria analysis using GIS model for landslide susceptibility mapping. © Published under licence by IOP Publishing Ltd." "57209862484;57195480773;56193847400;57200567420;55949825400;","Cloud detection from FY-4A's geostationary interferometric infrared sounder using machine learning approaches",2019,"10.3390/rs11243035","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077887807&doi=10.3390%2frs11243035&partnerID=40&md5=fc48466a1b7e26d371b4c12e93d1a928","FengYun-4A (FY-4A)'s Geostationary Interferometric Infrared Sounder (GIIRS) is the first hyperspectral infrared sounder on board a geostationary satellite, enabling the collection of infrared detection data with high temporal and spectral resolution. As clouds have complex spectral characteristics, and the retrieval of atmospheric profiles incorporating clouds is a significant problem, it is often necessary to undertake cloud detection before further processing procedures for cloud pixels when infrared hyperspectral data is entered into assimilation system. In this study, we proposed machine-learning-based cloud detection models using two kinds of GIIRS channel observation sets (689 channels and 38 channels) as features. Due to differences in surface cover and meteorological elements between land and sea, we chose logistic regression (lr) model for the land and extremely randomized tree (et) model for the sea respectively. Six hundred and eighty-nine channels models produced slightly higher performance (Heidke skill score (HSS) of 0.780 and false alarm rate (FAR) of 16.6% on land, HSS of 0.945 and FAR of 4.7% at sea) than 38 channels models (HSSof 0.741 and FAR of 17.7% on land, HSS of 0.912 and FAR of 7.1% at sea). By comparing visualized cloud detection results with the Himawari-8 Advanced Himawari Imager (AHI) cloud images, the proposed method has a good ability to identify clouds under circumstances such as typhoons, snow covered land, and bright broken clouds. In addition, compared with the collocated Advanced Geosynchronous Radiation Imager (AGRI)-GIIRS cloud detection method, the machine learning cloud detection method has a significant advantage in time cost. This method is not effective for the detection of partially cloudy GIIRS's field of views, and there are limitations in the scope of spatial application. © 2019 by the authors." "57207565169;56680839900;57191839329;6506008310;6602842679;55812857100;","Combination of an automated 3D field phenotyping workflow and predictive modelling for high-throughput and non-invasive phenotyping of grape bunches",2019,"10.3390/rs11242953","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077850531&doi=10.3390%2frs11242953&partnerID=40&md5=360a0187fbffbdbe9917308b359fd5bd","In grapevine breeding, loose grape bunch architecture is one of the most important selection traits, contributing to an increased resilience towards Botrytis bunch rot. Grape bunch architecture is mainly influenced by the berry number, berry size, the total berry volume, and bunch width and length. For an objective, precise, and high-throughput assessment of these architectural traits, the 3D imaging sensor Artec® Spider was applied to gather dense point clouds of the visible side of grape bunches directly in the field. Data acquisition in the field is much faster and non-destructive in comparison to lab applications but results in incomplete point clouds and, thus, mostly incomplete phenotypic values. Therefore, lab scans of whole bunches (360°) were used as ground truth. We observed strong correlations between field and lab data but also shifts in mean and max values, especially for the berry number and total berry volume. For this reason, the present study is focused on the training and validation of different predictive regression models using 3D data from approximately 2000 different grape bunches in order to predict incomplete bunch traits from field data. Modeling concepts included simple linear regression and machine learning-based approaches. The support vector machine was the best and most robust regression model, predicting the phenotypic traits with an R2 of 0.70-0.91. As a breeding orientated proof-of-concept, we additionally performed a Quantitative Trait Loci (QTL)-analysis with both the field modeled and lab data. All types of data resulted in joint QTL regions, indicating that this innovative, fast, and non-destructive phenotyping method is also applicable for molecular marker development and grapevine breeding research. © 2019 by the authors." "9335419500;57202985540;57211938397;56425382100;","An azure aces early warning system for air quality index deteriorating",2019,"10.3390/ijerph16234679","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075465582&doi=10.3390%2fijerph16234679&partnerID=40&md5=787334dcc19eb38776e538b6eb62a074","With the development of industrialization and urbanization, air pollution in many countries has become more serious and has affected people’s health. The air quality has been continuously concerned by environmental managers and the public. Therefore, accurate air quality deterioration warning system can avoid health hazards. In this study, an air quality index (AQI) warning system based on Azure cloud computing platform is proposed. The prediction model is based on DFR (Decision Forest Regression), NNR (Neural Network Regression), and LR (Linear Regression) machine learning algorithms. The best algorithm was selected to calculate the 6 pollutants required for the AQI calculation of the air quality monitoring in real time. The experimental results show that the LR algorithm has the best performance, and the method of this study has a good prediction on the AQI index warning for the next one to three hours. Based on the ACES system proposed, it is hoped that it can prevent personal health hazards and help to reduce medical costs in public. © 2019 by the authors. Licensee MDPI, Basel, Switzerland." "57214713846;57193133682;","Sensor data collection and its architecture with internet of things",2019,"10.35940/ijeat.A9564.109119","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074558358&doi=10.35940%2fijeat.A9564.109119&partnerID=40&md5=34402b0aa7e063620e65afc3d72f7928","Sensors are gadgets, which can screen temperature, moistness, weight, commotion levels, setting mindfulness, lighting condition and identify speed, position, and size of an Object. Sensor information are getting accumulated in gigantic amount thus they are overseen utilizing NOSQL. The information will be gathered in an IOT cloud stage where it will be additionally prepared with machine learning methods for prescient examination. What's more, eventually with the required answer for the business structure will be created. This paper explain the proposed system for IoT data collection with AWS (Amazon Web Service) cloud platform. Various system components like Kinesis stream, M2M platform, Notification service and secured IoT service layout. The complete BMS system architecture is detailed in this paper. © BEIESP." "55575342100;35226698000;57208121325;7103389378;7801621724;7102811204;","Biomass Assessment of Agricultural Crops Using Multi-temporal Dual-Polarimetric TerraSAR-X Data",2019,"10.1007/s41064-019-00076-x","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074137730&doi=10.1007%2fs41064-019-00076-x&partnerID=40&md5=20fc2e59826d1939c53e1105d797fbe3","The biomass of three agricultural crops, winter wheat (Triticum aestivum L.), barley (Hordeum vulgare L.), and canola (Brassica napus L.), was studied using multi-temporal dual-polarimetric TerraSAR-X data. The radar backscattering coefficient sigma nought of the two polarization channels HH and VV was extracted from the satellite images. Subsequently, combinations of HH and VV polarizations were calculated (e.g. HH/VV, HH + VV, HH × VV) to establish relationships between SAR data and the fresh and dry biomass of each crop type using multiple stepwise regression. Additionally, the semi-empirical water cloud model (WCM) was used to account for the effect of crop biomass on radar backscatter data. The potential of the Random Forest (RF) machine learning approach was also explored. The split sampling approach (i.e. 70% training and 30% testing) was carried out to validate the stepwise models, WCM and RF. The multiple stepwise regression method using dual-polarimetric data was capable to retrieve the biomass of the three crops, particularly for dry biomass, with R2 > 0.7, without any external input variable, such as information on the (actual) soil moisture. A comparison of the random forest technique with the WCM reveals that the RF technique remarkably outperformed the WCM in biomass estimation, especially for the fresh biomass. For example, the R2 > 0.68 for the fresh biomass estimation of different crop types using RF whereas WCM show R2 < 0.35 only. However, for the dry biomass, the results of both approaches resembled each other. © 2019, Deutsche Gesellschaft für Photogrammetrie, Fernerkundung und Geoinformation (DGPF) e.V." "57207828620;55440959500;","Predictive ability of covariate-dependent Markov models and classification tree for analyzing rainfall data in Bangladesh",2019,"10.1007/s00704-019-02812-0","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85062964084&doi=10.1007%2fs00704-019-02812-0&partnerID=40&md5=6d9ad96bcaa31acff328cd0445d036f6","This study attempts to make comparison between different parametric regressive models for the bivariate binary data with a machine learning technique. The data on sequential occurrence of rainfall in consecutive days is considered. The outcomes are classified as rainfall in both days, rainfall in one of the consecutive days, and no rainfall in both days. The occurrence of rainfall in consecutive days is analyzed by using statistical models with covariate dependence and classification tree for the period from 1980 to 2014. We have used relative humidity, minimum temperature, maximum temperature, sea level pressure, sunshine hour, and cloud cover in the model as covariates. The binary outcome variable is defined as the occurrence or non-occurrence of rainfall. Five regions of Bangladesh are considered in this study and one station from each region is selected on the basis of two criteria: (i) contains fewer missing values and (ii) representative of the regional characteristics geographically. Several measures are used to compare the models based on Markov chain and classification tree. It is found that for yearly data, both the Markov model and classification tree performed satisfactorily. However, the seasonal data show variation of rainfall. In some seasons, both models perform equally good such as monsoon, pre-monsoon, and post-monsoon, but in the winter season, the Markov model works poorly whereas classification tree fails to work. Additionally, we also observe that the Markov model performed consistently for each season and performs better compared with the classification tree. It has been demonstrated that the covariate-dependent Markov models can be used as classifiers alternative to the classification tree. It is revealed that the predictive ability of the covariate-dependent Markov model based on Markovian assumption performs either better or equally good compared with the classification tree. The joint models also consistently showed better predictive performance compared with regressive model for whole year data as well as for several seasonal data. © 2019, Springer-Verlag GmbH Austria, part of Springer Nature." "57215487421;14053460100;57217204638;57217199418;","A weighted normalized likelihood procedure for empirical land change modeling",2019,"10.1007/s40808-019-00584-0","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086732005&doi=10.1007%2fs40808-019-00584-0&partnerID=40&md5=282d1e940129a74d77460aec681d7d66","A critical foundation for empirical land change modeling is the mapping of transition potentials—quantitative evaluations of the readiness or suitability of land to go through a transition. This paper presents a procedure based on empirically determined normalized likelihoods of transition. It shows that these normalized likelihoods equate to posterior probabilities if case–control sampling is carried out among historical instances of change and persistence. The posterior probabilities can then be aggregated at the pixel level across multiple covariates using linear opinion pooling where the pixel-specific weight for each covariate is determined locally by its ability to distinguish between the alternatives of change or persistence. Thus, covariates with spatially varying diagnostic ability can be productively incorporated. The resulting algorithm is shown to have a skill comparable to that of a multi-layer perceptron approach with the advantage of high efficiency and amenability to distributed processing in a cloud environment. © 2019, Springer Nature Switzerland AG." "8109934200;57215870896;57211137457;56992503700;57215489736;36069985300;57213216388;57214356147;57196711847;24783260400;6602980349;21935606200;","SATVAM: Toward an iot cyber-infrastructure for low-cost urban air quality monitoring",2019,"10.1109/eScience.2019.00014","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85081005328&doi=10.1109%2feScience.2019.00014&partnerID=40&md5=094bff23645d2ecd1e9a2d887eb44343","Air pollution is a public health emergency in large cities. The availability of commodity sensors and the advent of Internet of Things (IoT) enable the deployment of a city-wide network of 1000's of low-cost real-Time air quality monitors to help manage this challenge. This needs to be supported by an IoT cyber-infrastructure for reliable and scalable data acquisition from the edge to the Cloud. The low accuracy of such sensors also motivates the need for data-driven calibration models that can accurately predict the science variables from the raw sensor signals. Here, we offer our experiences with designing and deploying such an IoT software platform and calibration models, and validate it through a pilot field deployment at two mega-cities, Delhi and Mumbai. Our edge data service is able to even-out the differential bandwidths from the sensing devices and to the Cloud repository, and recover from transient failures. Our analytical models reduce the errors of the sensors from a best-case of 63% using the factory baseline to as low as 21%, and substantially advances the state-of-The-Art in this domain. © 2019 IEEE." "57193933577;7004479957;","Single-Column Emulation of Reanalysis of the Northeast Pacific Marine Boundary Layer",2019,"10.1029/2019GL083646","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071267445&doi=10.1029%2f2019GL083646&partnerID=40&md5=c7dce819b65137a844ceb16dcba60ea9","An artificial neural network is trained to reproduce thermodynamic tendencies and boundary layer properties from European Center for Medium-Range Weather Forecasts Reanalysis 5th Generation high resolution realization reanalysis data over the summertime northeast Pacific stratocumulus to trade cumulus transition region. The network is trained prognostically using 7-day forecasts rather than using diagnosed instantaneous tendencies alone. The resulting model, Machine-Assisted Reanalysis Boundary Layer Emulation, skillfully reproduces the boundary layer structure and cloud properties of the reanalysis data in 7-day single-column prognostic simulations over withheld testing periods. Radiative heating profiles are well simulated, and the mean climatology and variability of the stratocumulus to cumulus transition are accurately reproduced. Machine-Assisted Reanalysis Boundary Layer Emulation more closely tracks the reanalysis than does a comparable configuration of the underlying forecast model. ©2019. American Geophysical Union. All Rights Reserved." "56097323700;35182211000;37123014300;","Development of a high spatiotemporal resolution cloud-type classification approach using Himawari-8 and CloudSat",2019,"10.1080/01431161.2019.1594438","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85063149357&doi=10.1080%2f01431161.2019.1594438&partnerID=40&md5=7b074027ffec2bc1075691f27123bda8","With the cloud scenario products from CloudSat, we developed a high spatiotemporal resolution cloud-type classification procedure for Himawari-8 multispectral datasets using maximum-likelihood estimation (MLE) and random forests (RF) classification. The training and classification procedures were processed independently, and both algorithms provided cloud-type results with a good performance. Validation indicated that the use of the visible (VIS) channel significantly improved the cloud-type identification capabilities, while the use of three or more channels simultaneously resulted in considerable improvements over the use of bispectral combinations. The comparison among different classifiers also revealed that RF was more sensitive than MLE to the quality and distribution of the training data. After retraining the RF using MLE-based clustered samples, we produced two more-reasonable and efficient classifiers that can be used during the day and night. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group." "55715413100;36664709300;14028364400;57207459537;","Gobal forest cover mapping using landsat and google earth engine cloud computing",2019,"10.1109/Agro-Geoinformatics.2019.8820469","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85072927621&doi=10.1109%2fAgro-Geoinformatics.2019.8820469&partnerID=40&md5=333c08a02b5855495d6f2c1f78f660bc","Nowadays, the access of Landsat data-sets and the ever-lowering costs of computing make it feasible to monitor the Earth's land cover at Landsat resolutions of 30 meter. However, the rapid forest-covered products on a large scale, such as intercontinental or global, is still challenging. By utilizing the huge catalog of satellite imagery as well as the high-performance computing capacity of Google Earth Engine, we proposed an automated pipeline for generating 30-meter resolution global-scale forest map from time-series of Landsat images, and a novel 30-meter resolution global forest map of 2018 is released. In this paper, we describe the methods to create products of forest cover at Landsat resolutions. First, we partitioned the landscapes into sub-regions of similar forest type and spatial continuity, thus maximizing spectral differentiation, simplifying classifier model and improving classification accuracy. Then, with the existing forest cover, which come from a variety of sources, a multi-source forest/non-forest sample set was established for machine algorithm learning training. Finally, a machine learning algorithm was used to obtain samples automatically, extract the characteristics of satellite images and establish the forest / non-forest classifier models. Taking the Landsat8 images in 2018 as a case, selecting satellite image features based on the study of forest reflectance, including onboard reflectivity, the index of forest vegetation and the texture features of each band, using established forest eco-zoning and multi-source forest / non-forest sample points, we realized automated learning and classification of forest cover for three initial zones. The accuracy verification of forest cover products in the three region was carried on two aspects: collecting verification points on high resolution satellite imagery (e.g. google earth), and cross-validating the current globally disclosed forest cover products. These two methods will illustrate the accuracy of the forest cover product. © 2019 IEEE." "57205765622;57191616594;56050235900;57205541304;56915184300;36711526700;","Farming on the edge: Architectural goals",2019,"10.1109/Agro-Geoinformatics.2019.8820424","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85072921206&doi=10.1109%2fAgro-Geoinformatics.2019.8820424&partnerID=40&md5=c3fbe1274a9e36a04953f92a1765fecc","This research investigates how advances in Internet of Things (IoT) and availability of internet connection would enable Edge Solutions to promote smart utilization of existing machines at the edge. The presented results are based on experiments performed in real scenarios using the proposed solution. Whereas scenarios were cloned from real environments it is important to have in mind that experiments were performed with low load in terms of data and small number of devices in terms of distribution. As result of extensive architecture investigation for an optimal edge solution and its possible correlation to industrial applications, this paper will provide evidences supporting the use of edge solutions in challenging conditions which arise at the edge, including smart factories and smart agriculture. The present work assumes that the reader has some exposition to Edge computing, Cloud computing and software development. The paper will present some important findings on this area, compare main architectural aspects and will provide a broad view of how edge solutions might be built for this particular scenario. Having discussed how the ideal architecture works and having provided an overview about how it may be applied to industrial plants, the final section of this paper addresses how artificial intelligence will fit into edge solutions, forming a new source of 'smart capabilities' to existing environments. © 2019 IEEE." "57208372273;57205760598;57208597081;","Using Convolutional Neural Networks for Cloud Detection from Meteor-M No. 2 MSU-MR Data",2019,"10.3103/S1068373919070045","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85070222008&doi=10.3103%2fS1068373919070045&partnerID=40&md5=bceac4b26f4af32c7aa883edc850638b","A method for cloud detection using the machine-learning algorithm based on a convolutional neural network is presented. Input data are satellite images received from the MSU-MR multispectral low-resolution scanning unit onboard the Meteor-M No. 2 satellite. The developed method can be an alternative to the traditional algorithms of cloud detection based on the calculation of differential indices and thresholds. The algorithm is verified using the machine-learning metrics, comparing the resulting cloud mask with the reference one obtained by interpreting the satellite image by an experienced meteorologist. It was also compared (for verification) with a similar product based on VIIRS spectroradiometer data. The cloud mask computed using the algorithm allows the automatic thematic processing of satellite images. © 2019, Allerton Press, Inc." "57202977053;57209296376;57201667638;56544915700;26030052000;15923105200;","A high-speed particle phase discriminator (PPD-HS) for the classification of airborne particles, as tested in a continuous flow diffusion chamber",2019,"10.5194/amt-12-3183-2019","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85067259848&doi=10.5194%2famt-12-3183-2019&partnerID=40&md5=ff24ecdc6e7bba88bf2dde7df78ee37f","A new instrument, the High-speed Particle Phase Discriminator (PPD-HS), developed at the University of Hertfordshire, for sizing individual cloud hydrometeors and determining their phase is described herein. PPD-HS performs an in situ analysis of the spatial intensity distribution of near-forward scattered light for individual hydrometeors yielding shape properties. Discrimination of spherical and aspherical particles is based on an analysis of the symmetry of the recorded scattering patterns. Scattering patterns are collected onto two linear detector arrays, reducing the complete 2-D scattering pattern to scattered light intensities captured onto two linear, one-dimensional strips of light sensitive pixels. Using this reduced scattering information, we calculate symmetry indicators that are used for particle shape and ultimately phase analysis. This reduction of information allows for detection rates of a few hundred particles per second. Here, we present a comprehensive analysis of instrument performance using both spherical and aspherical particles generated in a well-controlled laboratory setting using a vibrating orifice aerosol generator (VOAG) and covering a size range of approximately 3-32 μm. We use supervised machine learning to train a random forest model on the VOAG data sets that can be used to classify any particles detected by PPD-HS. Classification results show that the PPD-HS can successfully discriminate between spherical and aspherical particles, with misclassification below 5% for diameters >3μm. This phase discrimination method is subsequently applied to classify simulated cloud particles produced in a continuous flow diffusion chamber setup. We report observations of small, near-spherical ice crystals at early stages of the ice nucleation experiments, where shape analysis fails to correctly determine the particle phase. Nevertheless, in the case of simultaneous presence of cloud droplets and ice crystals, the introduced particle shape indicators allow for a clear distinction between these two classes, independent of optical particle size. From our laboratory experiments we conclude that PPD-HS constitutes a powerful new instrument to size and discriminate the phase of cloud hydrometeors. The working principle of PPD-HS forms a basis for future instruments to study microphysical properties of atmospheric mixed-phase clouds that represent a major source of uncertainty in aerosol-indirect effect for future climate projections.. © Author(s) 2019." "57210343217;57210336943;57210340168;26639841900;","Implementation of machine learning techniques applied to the network intrusion detection system",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85070482626&partnerID=40&md5=d34b66f1d3f8adff05593710ed100ab2","An immense amount of data is being generated every second due to technological advancement and reforms. Social networking and cloud computing are generating a huge amount of data every second. Every minute data is being captured in the computing world from the click of the mouse to video people tend to watch generating an immediate recommendation. Everything a user is doing on the internet is being captured in different ways for multiple intents. Now it all ends up to monitor the system and network and, secure lines and servers. This mechanism is called Intrusion Detection System(IDS). Hacker uses multiple numbers of ways to attack the system which can be detected through a number of algorithm and techniques. A comprehensive survey of some major techniques of machine learning implemented on intrusion Detection was done where techniques based on kmeans, K-means with principal component analysis, Random Forest algorithm Extreme learning the ma-chine, techniques, classification algorithms such as Naive Bayes algorithm, Hoeffding Tree algorithm.Also, Accuracy Updated Ensemble algorithm, Accuracy Weighted Ensemble algorithm, Support Vector Machine, Genetic algorithm and Deep learning were studied. Now some of these algorithms are applied upon the NSL-KDD data set and compared on the basis of their accuracy. © BEIESP." "57210211142;57210204611;36467916800;","Video summarization using adaptive thresholding by machine learning for distributed cloud storage",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85069926529&partnerID=40&md5=857110a07a1bea9cd7d72df977cb602a","The growth in the video content-based communication and information management have motivated the applications dealing with the video contents to be placed on cloud-based data-storages, which are associated with datacentres. The applications ranging from social media to distance education and surveillance to business communication or any applications related to the e-governess are considering the video data as one of the most preferred mode of communication. The video data is enriched with audio and visual contents, which makes analysis or expressing information high convenient. Computerized video is an electronic portrayal of moving visual pictures as encoded advanced information. This is as opposed to simple video, which speaks to moving visual pictures with a simple sign. The advanced video contains a progression of computerized pictures showed in quick progression. The number of applications, as mentioned, dealing with video data is increasing and as a result a large amount of video content is generated every day. Hence, the complexity of retrieving the information from the video contents are also increasing. The bottleneck of the video retrieval process is for a lower sized segment of the video content can be retrieved in low time complexity and if the information to be preserved to a higher extend, then the retrieval time complexity increases. Thus, a good number of parallel researchers have introduced various methods for video content summarization and retrieval using summarization with efficient searching methods, but most of the parallel research outcomes are criticized for either higher time complexity or for higher information loss. This problem can be ideally solved by finding the highly accurate ratio of key information video frames from the total video content. Henceforth, this work, presents a novel machine learning method for identifying the key frames, not only based on the information available in the frame, also validating the key frames with the thresholds of the objects or changes in the frame. The work is again enhanced by considering the adaptive thresholding method for distributed and collaborative video information. The measures taken in the proposed algorithm produces a 98% accuracy for video information representation and nearly 99% reduction in the video frames, which results into nearly 99% reduction in the time complexity. © BEIESP." "57210207708;56623155300;57210206706;","Cloud based healthcare framework for criticality level analysis",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85069912205&partnerID=40&md5=13f695c05b4814864bc7c9041c952cf3","In a Cloud based development Amazon Web Service(AWS) is a platform which secures the cloud services, offers database storage, content delivery, computer power and also provides other functionality to help business scale and grow. This proposed work aims that medical healthcare inputs from various sensors have been automatically retrieved and directly loaded into the cloud. Once the data has been loaded creating the inference engine, setting up a big data cloud environment and store the data into a cloud based dataset. Medical data is calculated using machine learning algorithms such as K-Nearest Neighbor (KNN), Naïve Bayes and Support Vector Machine (SVM) through R shiny web application. The cloud system stores the health care data and transmitted to practitioners through the web service network. Based on these medical data the score value of a patient is calculated and displays criticality of patient. © BEIESP." "57209139058;55694076900;57191433477;57191247319;56029052900;8529553300;57071147200;7701309571;","Detecting Comma-Shaped Clouds for Severe Weather Forecasting Using Shape and Motion",2019,"10.1109/TGRS.2018.2887206","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85066611927&doi=10.1109%2fTGRS.2018.2887206&partnerID=40&md5=bf6cd6bd28b389265d5429ffea35a4d7","Meteorologists use shapes and movements of clouds in satellite images as indicators of several major types of severe storms. Yet, because satellite image data are in increasingly higher resolution, both spatially and temporally, meteorologists cannot fully leverage the data in their forecasts. Automatic satellite image analysis methods that can find storm-related cloud patterns are thus in demand. We propose a machine learning and pattern recognition-based approach to detect 'comma-shaped' clouds in satellite images, which are specific cloud distribution patterns strongly associated with cyclone formulation. In order to detect regions with the targeted movement patterns, we use manually annotated cloud examples represented by both shape and motion-sensitive features to train the computer to analyze satellite images. Sliding windows in different scales ensure the capture of dense clouds, and we implement effective selection rules to shrink the region of interest among these sliding windows. Finally, we evaluate the method on a hold-out annotated comma-shaped cloud data set and cross match the results with recorded storm events in the severe weather database. The validated utility and accuracy of our method suggest a high potential for assisting meteorologists in weather forecasting. © 2019 IEEE." "56979231400;55634326100;7201369119;","Semi-automated open water iceberg detection from Landsat applied to Disko Bay, West Greenland",2019,"10.1017/jog.2019.23","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065863688&doi=10.1017%2fjog.2019.23&partnerID=40&md5=335f25d70d68f10cf8a8eeef497226fb","Changes in Greenland's marine-terminating outlet glaciers have led to changes in the flux of icebergs into Greenland's coastal waters, yet icebergs remain a relatively understudied component of the ice-ocean system. We developed a simple iceberg delineation algorithm for Landsat imagery. A machine learning-based cloud mask incorporated into the algorithm enables us to extract iceberg size distributions from open water even in partially cloudy scenes. We applied the algorithm to the Landsat archive covering Disko Bay, West Greenland, to derive a time series of iceberg size distributions from 2000-02 and 2013-15. The time series captures a change in iceberg size distributions, which we interpret as a result of changes in the calving regime of the parent glacier, Sermeq Kujalleq (Jakobshavn Isbræ). The change in calving style associated with the disintegration and disappearance of Sermeq Kujalleq's floating ice tongue resulted in the production of more small icebergs. The increased number of small icebergs resulted in increasingly negative power law slopes fit to iceberg size distributions in Disko Bay, suggesting that iceberg size distribution time series provide useful insights into changes in calving dynamics. © 2019 The Author(s)." "7006413284;57214912047;7006940184;56928095700;57214910980;","Data driven smart monitoring for pipeline integrity assessment",2019,"10.2118/197327-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088200548&doi=10.2118%2f197327-ms&partnerID=40&md5=3901aba2ba55ea3a50eb9eb7d4c15b49","Efficiency and safety are primary requirements for oil & gas fluid filled transportation system. However, the complexity of the asset makes it challenging to derive a theoretical framework for managing the control parameters. The current frontier for a real time monitoring exploits the ""digital tansformation"", i.e. the acquisition and the analysis of large datasets recorded along the whole asset lifecycle, which are used to infer ""data driven"" relations and to predict the evolution of the asset integrity. This paper presents some results of a research project for the design, implementation and testing of a ""machine learning"" approach to vibroacoustic data recorded continuously by acquisition units installed every 10-20 km along a pipeline. In a fluid transportation system, vibroacoustic signals are generated by the flow regulation equipment (i.e. pumping, valves, metering), by the fluid flowing (i.e. turbulence, cavitation, bubbles), by third party interference (i.e. spillage, sabotage, illegal tapping), by internal inspection using PIGs operations), and by natural hazards (i.e. microseismic, subsidence, landslides). The basic principle of machine learning is to ""observe"", for an appropriate time interval, a series of descriptors, in this stage related to vibroacoustic signals but that can be integrated with other physical data (i.e. temperature, density, viscosity), in order to ""learn"" their safe range of variation or, when properly fed to a classification procedure, to obtain automatically a discrete set of operational status. The classification criteria are then applied to new data, highlighting the presence of system anomalies. The paper considers vibroacoustic signals collected at the flow stations of an oil trunkline in Nigeria. The vibroacoustic signals are the static pressure, the acceleration and the pressure transients recorded at the departure and at the arrival terminals. More than one year of data is available. Derived smart indicators are defined, which are directly linked to the asset parameters: for instance, the cross-correlation of the pressure transients at adjacent measuring locations permits to estimate the fluid channel continuity (correlation value), the sound velocity (time of correlation peak), and the sound attenuation (amplitude versus frequency amplitude decay). A portion of the data during normal operation is used for training and tuning a reference model. After that, new data are compared with the model, and anomalies are automatically detected. Two kind of errors are raised: i) sensors; ii) alerts. Sensor errors are referred to missing or corrupted sensors data. Alerts are raised when the measured physical quantities are not coherent with the functional and known service behaviors of the transport system. The system model is not static over time, and in fact it can be updated by the operators' feedback, that can tag false alarms and thus, automatically, re-define the set of operational scenarios of the upstream system. The medium-long term construction and update of data driven models is effective for predictive maintenance, automatic anomalies detection, optimization of operational procedures. Moreover, the new policy of data management and the opportunity of gaining awareness by interconnecting the monitoring experience of different assets leverages the introduction of new technologies (cloud, big data), new professional figures (smart data scientist), new operational and business models. © 2019, Society of Petroleum Engineers" "56338549300;57210363428;57216230666;57208208641;57200248379;57210364558;35617712500;","Enhanced reservoir geosteering and geomapping from refined models of ultra-deep LWD resistivity inversions using machine-learning algorithms",2019,"10.30632/T60ALS-2019_EE","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083952227&doi=10.30632%2fT60ALS-2019_EE&partnerID=40&md5=02ecbd0b7afc2ffe43398486ff3e437e","Ultradeep logging-while-drilling (LWD) resistivity tools have been widely used in the borehole resistivity applications. Previous field examples successfully demonstrated a detection range more than 200 ft away from a wellbore. As a result of the ultradeep detection capability, the complexity of the inversion process increases to accommodate a larger number of layered models, as compared to the conventional approaches. Cloud-based distributed solutions were implemented in the algorithms to efficiently provide in-time geological inverted models for real-time decisions. These available software platforms and successful field data from the ultradeep-reading tools initialized further studies and evaluations of the advanced, machine-learning algorithms applied into the existing inversion process. Based on a large database of ultradeep resistivity measurements from past successful field jobs and pre-well modeling, this paper presents several deep-learning algorithms to improve the existing inversion process for extracting more geological information (Payrazyan et al., 2017; Xu et al., 2019). The proposed methods identify similarities among numerous solutions attained by individual steps of the existing inversion process. Then, most likely distributions are detected within a detection range of the inversion to remove outlier signals and models, and to further produce more geologically reasonable representations. The proposed methods also enable automatic boundary-picking of layers with a major resistivity contrast between them. The determined connections learned from the previous sets of measurements are used to train any future processes based on a new set of measurements, enabling more efficient evaluations and calculations. Both modeling and field examples establish better geological interpolations acquired from the presented machine-learning algorithms than the original inversion approach. The use of the machine-learning concepts on the ultradeep resistivity measurements efficiently enhances the quality of the final geological interpretation over long detection distances into the formations. This enhancement benefits operations by optimizing reservoir development, maximizing assets, and reducing overall operational cost. Copyright 2019, held jointly by the Society of Petrophysicists and Well Log Analysts (SPWLA) and the submitting authors." "57208528190;55970576100;17434697300;57209911880;","The transparency of big data, data harvesting and digital Twins",2019,"10.1007/978-3-030-11289-9_6","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076088790&doi=10.1007%2f978-3-030-11289-9_6&partnerID=40&md5=336127021231df4d6177aa9b8d779ea5","Computer storage and cloud computing has become more powerful with multiple algorithms running complex data analysis looking at intelligence trends, user behaviour, profiling and ways to make use of these outputs. Added with the artificial intelligence (AI) interaction has meant a new and dynamic method to create models forging analysis to be more clinical, proficient and continually seeking more improvement with the self-learning and intelligent programming of machine learning (ML). In the healthcare sector there is deep interest in collecting, curating the data and making the best use of silo’d data through methods such as blockchain. This can then lead to a multitude of innovations such as precision based medicine, targeting individual variability in genes, their environment, etc. It also means that big data analytics in healthcare is evolving into providing these insights from very large data sets and improving outcomes while reducing costs and inefficiencies. However, there also are some ethical impacts in the process of Digital Twins which can lead to segmentation and discrimination. Or perhaps the data that is automatically collected from healthcare sensors in IoMT and what type of governance are they scrutinized to. It is clear that data is the most important asset of not just an organisation but also to the individual and why the General Data Protection Regulation (GDPR) has taken an important stance in data protection by design and default, that all organisations needs to follow. This chapter aims to highlight some of the concerns. © Springer Nature Switzerland AG 2019." "56292149100;57200597357;6602942477;","Development of an operational algorithm for automated deforestation mapping via the Bayesian integration of long-term optical and microwave satellite data",2019,"10.3390/rs11172038","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071993587&doi=10.3390%2frs11172038&partnerID=40&md5=d6d190f979b2cb6d1902ef9ee8bf3a47","The frequent fine-scale monitoring of deforestation using satellite sensors is important for the sustainable management of forests. Traditional optical satellite sensors suffer from cloud interruption, particularly in tropical regions, and recent active microwave sensors (i.e., synthetic aperture radar) demonstrate the difficulty in data interpretation owing to their inherent sensor noise and complicated backscatter features of forests. Although the sensor integration of optical and microwave sensors is of compelling research interest, particularly in the conduct of deforestation monitoring, this topic has not been widely studied. In this paper, we introduce an operational algorithm for automated deforestation mapping using long-term optical and L-band SAR data, including a simple time-series analysis of Landsat stacks and a multilayered neural network with Advanced Spaceborne Thermal Emission and Reflection Radiometer and Phased Array-type L-band Synthetic Aperture Radar-2, followed by sensor integration based on the Bayesian Updating of Land-Cover. We applied the algorithm over a deciduous tropical forest in Cambodia in 2003-2018 for validation, and the algorithm demonstrated better accuracy than existing approaches, which only depend on optical data or SAR data. Owing to the cloud penetration ability of SAR, observation gaps of optical data under cloudy conditions were filled, resulting in a prompter detection of deforestation even in the tropical rainy season. We also investigated the effect of posterior probability constraints in the Bayesian approach. The land-cover maps (forest/deforestation) created by the well-tuned Bayesian approach achieved 94.0% ± 4.5%, 80.0% ± 10.1%, and 96.4% ± 1.9% for the user's accuracy, producer's accuracy, and overall accuracy, respectively. In the future, small-scale commission errors in the resultant maps should be improved by using more sophisticated machine-learning approaches and considering the reforestation effects in the algorithm. The application of the algorithm to other landscapes with other sensor combinations is also desirable. © 2019 by the authors." "35173655800;55916474900;6506175565;7004175711;","Calibrating long-period variables as standard candles with machine learning",2019,"10.1093/mnras/sty3495","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85067187700&doi=10.1093%2fmnras%2fsty3495&partnerID=40&md5=391b362f078591f5e5fb08ee9fb34d5b","Variable stars with well-calibrated period-luminosity (PL) relationships provide accurate distance measurements to nearby galaxies and are therefore a vital tool for cosmology and astrophysics. While these measurements typically rely on samples of Cepheid and RR-Lyrae stars, abundant populations of luminous variable stars with longer periods of 10-1000 d remain largely unused. We apply machine learning to derive a mapping between light-curve features of these variable stars and their magnitude to extend the traditional PL relation commonly used for Cepheid samples. Using photometric data for long-period variable stars in the Large Magellanic Cloud (LMC), we demonstrate that our predictions produce residual errors comparable to those obtained on the corresponding Cepheid population. We show that our model generalizes well to other samples by performing a blind test on photometric data from the Small Magellanic Cloud (SMC). Our predictions on the SMC again show small residual errors and biases, comparable to results that employ PL relations fitted on Cepheid samples. The residual biases are complementary between the long-period variable and Cepheid fits, which provides exciting prospects to better control sources of systematic error in cosmological distance measurements. We finally show that the proposed methodology can be used to optimize samples of variable stars as standard candles independent of any prior variable star classification. © 2018 The Author(s)." "57205427051;57205427084;57205427187;57205427288;","An applied machine learning approach to subsea asset inspection",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85059991653&partnerID=40&md5=4c939405f4c6d9e1ad6b5cb5f590388f","Subsea infrastructure inspection campaigns are conducted on a fixed time schedule to ensure the integrity of offshore assets while guaranteeing compliance with government and other stakeholders' legislation. These thorough assessments of subsea assets are achieved by acquiring and processing survey results alongside the relevant engineering data and environmental conditions. The current approach is not ideal. The infrastructure operators would prefer to have more frequent, on-demand field visits and base the assessments on an up-to-date high resolution, precise asset model. Unfortunately, this has historically not been technically feasible. The underwater inspection vehicles moved relatively slow, the positioning accuracy was limited, and the underwater sensors' accuracy and resolution was not comparable with those used in air. This is not the case anymore. We now operate inspection vehicles traveling with speeds over six knots! We can control them remotely from the office-based command center. Next to the acoustic technology, the vehicles are additionally equipped with optic sensors, dynamically collecting point clouds, images and videos of unprecedented resolution. Gigabytes of data hits the vessel every few minutes. And that's where it used to end… Processing and, so called, ""Digital Twin"" modelling from this data has traditionally been taking weeks for every day of inspection. In many cases it was simply not technically possible since the data files were prohibitively big. Data required subsampling, more data processors had to go offshore, asset operators had to wait longer. Given the increasing capacity of satellite links, limitless potential of commoditized cloud computing and sophisticated Deep Learning algorithms, we are now able to move subsea asset management to the new era. Through one of their latest research & development programmes - Roames, Fugro has been able to revolutionize the traditional workflows using the state-of-the-art digital technology. The resultant product is a bespoke, web-based service that enables asset (initially pipelines) inspection data to be uploaded to the secure cloud environment, processed using Machine Learning, verified by experts on shore and visualized in an intuitive 4D web viewer. All delivered to the geographically spread stakeholders, operators and decision makers in near real time. This approach greatly reduces the cost of infrastructure management practices, lowers the risk exposure and contributes to extended life of the asset. The Roames, Machine Learning based methodology was validated using a vast archive of survey data prior to testing the workflow on a live project. Detailed method statement and field results are presented in this paper. © Copyright 2018, Society of Petroleum Engineers." "57205425235;57205425268;","Enabling the best by preparing for the worst: Lessons from disaster response for industrial IoT in oil and gas",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85059972627&partnerID=40&md5=eceb60791925ee46536223aa632fd684","Objectives/Scope: As more oil and gas companies develop Internet of Things (IoT) strategies and beginning their digital transformation to Industry 4.0 or Smart Manufacturing, they face challenges in adopting technologies due to regulatory restrictions for highly combustible atmospheres such as exist in some of the world's largest and most critical industries - oil & gas, chemical, pharmaceutical, energy and others. In Zone 1 classified hazardous areas worldwide, up to 15% of personnel do not have access to mobile computing devices unless they are certified ""intrinsically safe,"" or incapable of causing a spark that could ignite a combustible environment. Thus, the human ""sensor"" in hazardous area operations, who could conceivably detect perceived anomalies or problems in the maintenance, workflow, process or function of these operations, is relegated to recording observations with pencil and paper and then entering data manually into ERP systems hours or days later. Such lack of real-time communication and data management results in inefficiency, increased costs and elevated safety and asset risk, causing potential down-time and even loss of life in extreme cases. Methods, Procedures, Process: By deploying new IoT technologies that allow people to use technology inside Zone 1 hazardous areas, humans can actively interact with machines in real time to dramatically improve productivity, safety and the bottom line in hazardous operations. A new style of IoT platform built especially for oil & gas hazardous area operations, would need to include various and affordable types of sensors to cover vast spaces, real-time communications, cloud computing, machine learning, rights management, security, big data storage, analytics and user-friendly visualization, all functioning in highly explosive conditions. This paper considers the advantages for productivity and safety of an IoT Platform for Hazardous Locations, based on hands-on research conducted by AegexTechnologies, Verizon, Nokia and multiple technology partners that tested various edge technologies with first responders in realistic disaster scenarios during two annual events, Operation Convergent Response 2017 (#OCR2017) and Operation Convergent Response 2018 (#OCR2018 - to take place 5-8 November 2018)). The events provide unparalleled opportunities to test IoTunder extreme conditions with real people, such as a staged refinery collapse caused by an earthquake. Results, Observations, Conclusions: #OCR2017 and #OCR2018 showed how enabling real-time communications and data management via cutting-edge technologies, such as intrinsically safe tablets and IoT sensors, can strategically assist first responders to better handle emergencies. The studies' results give insight into the need for continued collaboration on IoT capabilities that can better manage not only emergency response, but everyday operations in hazardous industries such as oil and gas. Novel/Additive Information: Equipping oil and gas facilities with pervasive, smart IoT data-sensing capabilities, and equipping oil and gas personnel with real-time communications and data management tools, could result in dramatic improvements in productivity, safety, emergency response and disaster mitigation. © Copyright 2018, Society of Petroleum Engineers." "29867490900;8213128600;6603354695;8213128500;57197087688;6603888005;","Pattern Recognition Scheme for Large-Scale Cloud Detection over Landmarks",2018,"10.1109/JSTARS.2018.2863383","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85051767950&doi=10.1109%2fJSTARS.2018.2863383&partnerID=40&md5=12fc72c5a419b57417e2751d014371de","Landmark recognition and matching is a critical step in many image navigation and registration models for geostationary satellite services, as well as to maintain the geometric quality assessment in the instrument data processing chain of earth observation satellites. Matching the landmark accurately is of paramount relevance, and the process can be strongly impacted by the cloud contamination of a given landmark. This paper introduces a complete pattern recognition methodology able to detect the presence of clouds over landmarks using Meteosat second generation (MSG) data. The methodology is based on the ensemble combination of dedicated support vector machines dependent on the particular landmark and illumination conditions. This divide-and-conquer strategy is motivated by the data complexity and follows a physically based strategy that considers variability both in seasonality and illumination conditions along the day to split observations. In addition, it allows training the classification scheme with millions of samples at an affordable computational costs. The image archive was composed of 200 landmark test sites with near 7 million multispectral images that correspond to MSG acquisitions during 2010. Results are analyzed in terms of cloud detection accuracy and computational cost. We provide illustrative source code and a portion of the huge training data to the community. © 2018 IEEE." "6602574676;56227666500;56151374100;57061262300;36098286300;24166367300;7006393267;","Snow-covered area using machine learning techniques",2018,"10.1109/IGARSS.2018.8519443","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85064194187&doi=10.1109%2fIGARSS.2018.8519443&partnerID=40&md5=5e1fe13d5a1d90894f14a226f6e1de07","In this study, we used an artificial neural network method to estimate the fractional snow cover area (fSCA), which is fast and accurate, and that can be easily adapted to different remote sensing instruments. We tested our approach using SnowEx data from NASA's Cloud Absorption Radiometer (CAR) over Grand Mesa; one of the largest flat-topped mountains in the world, which features sufficient forested stands with a range of density and height (and a variety of other forest conditions); a spread of snow depth/snow water equivalent conditions over sufficiently flat snowcovered terrain. The retrieved fractional snowcovered area from CAR compares reasonably with a Sentinel-2 image over the same location and demonstrates CAR's unique capability to improve the retrieval of snow properties using machine learning. The retrieved snow fraction parameter from our method is expected to minimize the error associated with the traditional binary snow detection scheme, and improve the retrieval quality of key parameters such as surface albedo. © 2018 IEEE." "13409639400;57203231038;7405934557;","The potential of sentinel satellites for large area aboveground forest biomass mapping",2018,"10.1109/IGARSS.2018.8517597","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85063154821&doi=10.1109%2fIGARSS.2018.8517597&partnerID=40&md5=3f4f6582feadd31b318f5a0367b6ec07","Estimation of aboveground forest biomass is critical for regional carbon policies and sustainable forest management. Both passive optical remote sensing and active microwave remote sensing can play an important role in the monitoring of forest biomass. In this study, the recently launched Sentinel-2 Multi Spectral Instrument satellite and Sentinel-1 SAR satellite systems were evaluated and integrated to investigate the relative strengths of each sensor for mapping aboveground forest biomass at a regional scale. The Australian state of Victoria, with its wide range of forest vegetation was chosen as the study area to demonstrate the scalability and transferability of the approach. In this study aboveground forest biomass (AGB) was defined as the tons of carbon per hectare for the aboveground components (stem, branches, leaves) of all live large trees greater than 10 cm in diameter at breast height (DBHOB). Sentinel-2 and Sentinel-1 data were fused within a machine learning framework using a boosted regression tree model and high-quality ground survey data. Multi-criteria evaluations showed the use of the two independent and fundamentally different Sentinel satellite systems were able to provide robust estimates (R 2 of 0.62, RMSE of 32.2 t.C.ha -1 ) of aboveground forest biomass, with each sensor compensating for the weakness (cloud perturbations and spectral saturation for Sentinel 2, and sensitivity to ground moisture for Sentinel 1) of each other. As archives for Sentinel-2 and Sentinel-1 continue to grow, mapping aboveground forest biomass and dynamics at moderate resolution over large regions should become increasingly feasible. © 2018 IEEE" "55083178100;57202647645;57192690229;","Towards extraction of lianas from terrestrial lidar scans of tropical forests",2018,"10.1109/IGARSS.2018.8517634","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85063144970&doi=10.1109%2fIGARSS.2018.8517634&partnerID=40&md5=0aef4f09fd61fe99b51a9e277003cc99","Increased liana abundance results in reduced tree growth and increased tree mortality in tropical forest. The impact of lianas on forest-wide carbon storage has been a special interest for many researchers. The vertical and horizontal spatial distribution of lianas in tropical forest will determine the interaction with trees and the forest carbon cycle. In this study, we will introduce an algorithm to extract lianas from terrestrial laser scanning (TLS) data of a tropical forest. We developed a classification method for separating liana points from other points in a point cloud under canopy. We used a Random Forests machine learning algorithm for the classification of liana points from the other points. The leaf-wood and liana-tree classification accuracies are 90.69% and 94.42%, respectively. The results show the potential of TLS data for analysis the spatial distribution of lianas in forest stands and we explore the potential of extracting lianas from TLS point clouds. © 2018 IEEE" "57208219543;57189034361;","Study on Feature Selection and Feature Deep Learning Model for Big Data",2018,"10.1109/ICSCSE.2018.00171","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065861133&doi=10.1109%2fICSCSE.2018.00171&partnerID=40&md5=3b9d777ba6856795dfa1c2ba04095318","In the era of big data, agricultural big data effectively mined in the agricultural cloud service platform helps to provide intelligent services for agricultural production and management[1]. Two big data feature selection methods are proposed based on the presence or absence labels of big data and potential value of data is mined better by effectively utilizing big data feature learning technologies. Transformation of data from data with primitive rough extraction features to data with features with stronger separability and high-level semantic features is of great significance for target task learning. © 2018 IEEE." "56416697700;38863569800;24544853800;14618921800;57191157218;7003503895;7005618744;57203162576;","Artificial intelligence based directional mesh network design for spectrum efficiency",2018,"10.1109/AERO.2018.8396558","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85049869729&doi=10.1109%2fAERO.2018.8396558&partnerID=40&md5=dab355616fd2f3799805985d66794970","The paper presents a novel directional mesh network (DMN) design that can distribute the limited radio spectrum resources more efficiently for a DMN by applying artificial intelligence machine learning (ML) techniques. The proposed DMN framework analyzes time-sensitive signal data close to the signal source using fog computing with different types of ML techniques. Depending on the computational capabilities of the fog nodes, different feature extraction methods such as energy detection, match filter, and cyclostationary detection are selected to optimize spectrum allocation. The proposed system also takes the antenna power gain into consideration, which can further reduce probability of detection and interference of the DMN system. Also, the computing nodes send the periodic signal summary which is much smaller than the original signal to the cloud so that the overall system spectrum source allocation strategies are dynamically updated. Instead of just detecting the spectrum holes for secondary users to transmit the signal, the proposed system can optimize the signal transmission path from the cloud to the end user under the interference and relay constraints. The distributed nodes can further improve the strategy based on the sensing information from the fog. Applying fog computing, the system is more adaptive to the local environment and robust to spectrum changes. It will significantly improve the network reliability, resiliency, and flexibility. Designing the proposed system doesn't necessary need change much of the current communication network platform. As most of the signal data is processed at the fog level, it further strengthens the system security by reducing the communication burden of the communications network. © 2018 IEEE." "57200418714;55557536600;57037229400;","COMPARISON of SINGLE and MULTI-SCALE METHOD for LEAF and WOOD POINTS CLASSIFICATION from TERRESTRIAL LASER SCANNING DATA",2018,"10.5194/isprs-annals-IV-3-217-2018","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85046755955&doi=10.5194%2fisprs-annals-IV-3-217-2018&partnerID=40&md5=2550243857f10376fe9d5c2d908d3212","The classification of leaf and wood points is an essential preprocessing step for extracting inventory measurements and canopy characterization of trees from the terrestrial laser scanning (TLS) data. The geometry-based approach is one of the widely used classification method. In the geometry-based method, it is common practice to extract salient features at one single scale before the features are used for classification. It remains unclear how different scale(s) used affect the classification accuracy and efficiency. To assess the scale effect on the classification accuracy and efficiency, we extracted the single-scale and multi-scale salient features from the point clouds of two oak trees of different sizes and conducted the classification on leaf and wood. Our experimental results show that the balanced accuracy of the multi-scale method is higher than the average balanced accuracy of the single-scale method by about 10 % for both trees. The average speed-up ratio of single scale classifiers over multi-scale classifier for each tree is higher than 30. © Authors 2018." "57202990666;57200762357;57203899789;57203899549;57203898383;","Enabling autonomous well optimization via using IoT-enabled devices and machine learning in bakken horizontal wells",2018,"10.2118/190955-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088405164&doi=10.2118%2f190955-ms&partnerID=40&md5=5673b223eb4a7e17e6e0f9a4fbba14f3","During the mid-twentieth century, Dr. Sam Gibbs developed math that continues today to serve as the basis for the downhole monitoring and control of many wells. However, analytical methods based on this calculation, including wave equation techniques, assume vertical, single-well pads with little to no side loading. Modern E&P development has shifted to multi-well horizontal well pads with shallow kickoffs uphole and backbuilds downhole with high dog-leg severity, which causes high friction and dynamic conditions in the pumping system. The Gibbs wave equation and fillage calculation do not take into account key wellbore forces such as mechanical or Coulomb friction arising from rod on tubing wear due to deviation in the well. Friction due to deviation can result in downhole dynamometer card shapes that distort fillage calculations and downhole card analysis. Additionally, most rod pump control systems are based on Programmable Logic Control systems (PLC). PLC systems are simple to program but are limited in their computation capabilities and are unable to accommodate sophisticated mathematics. Increased computational capabilities are required to execute higher-order mathematics that accurately calculate downhole parameters and enable well autonomy. One approach to driving autonomous well classification and optimization of setpoints is the deployment of a system that is capable of real-time analysis and higher-order mathematics. An Internet of Things (IoT) device with high-performance computational capabilities and direct communication with a cloud-based analytics software platform was developed with the capabilities to execute higher-order mathematics, artificial intelligence and machine learning on high resolution data, sampled in real-time from the rod pump control system. Equinor deployed this technology on 50 wells in the Bakken. The 50 wells chosen are highly representative of ""typical"" Bakken horizontal wells. The device was connected into the legacy rod pump controller via Modbus connection. Immediate differences in key downhole parameters were observed when comparing the results from the traditional rod pump controller to the IoT device. The higher-accuracy physics-based inputs feed into machine learning algorithms, which dynamically classify wells into key operating states of under-pumping, over-pumping, and dialed in. Using improved downhole information, Equinor was able to automate well optimization setpoint decisions, resulting in reduced well volatility, better pump efficiency, and increased pump fillage. Equinor was able to achieve these improvements while maintaining production in all cases. By identifying wells that were over-pumping and under-pumping to optimize SPM setpoints, Equinor was able to achieve higher efficiency outcomes with equal or increased production. For wells that were under-pumping, Equinor was able to increase oil production by up to 33%. For wells that were over-pumping, Equinor was able to decrease the number of strokes by 11% and increase pump efficiency by 14%. Copyright 2018, Society of Petroleum Engineers." "7101806774;8308170700;22233735600;35776720900;36718621600;55448544300;","Towards automation of satellite-based radar imagery for iceberg surveillance - Machine learning of ship and iceberg discrimination",2018,"10.4043/29130-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088073432&doi=10.4043%2f29130-ms&partnerID=40&md5=782f76cc42bb033cea8ba25972b77aef","Drifting icebergs can threaten navigation and marine operations and are prevalent in a number of regions that have active oil and gas exploration and development. Satellite synthetic aperture radar (SAR) is naturally applicable to map and monitor icebergs and sea ice due its ability to capture images day or night, as well as through cloud, fog and various wind conditions. There are several notable examples of its use to support operations, including Grand Banks, Barents Sea, offshore Greenland and Kara Sea. New constellations of satellites and the increasing volume of satellite data becoming available present a new paradigm for ice surveillance, in terms of persistence, reliability and cost. To fully extract the value of the data from these constellations, automation and cloud-based processing must be implemented. This will allow more timely and efficient processing, lowering monitoring costs by at least an order of magnitude. The increase in data persistence and processing capability allows large regions to be monitored daily for ice incursions, thus increasing safety and efficiency during offshore operations in those regions. The process of automating SAR-based iceberg surveillance involves creating a process flow that is robust and requires limited human intervention. The process flow involves land-masking, target detection, target discrimination and product dissemination. Land masking involves the removal of high-clutter land from the imagery to eliminate false detection from these locations. Target detection usually involves an adaptive threshold to separate true targets from the background ocean clutter. A constant false alarm rate (CFAR) is a standard technique used in radar image processing for this purpose. Target discrimination involves an examination of the distinct features of a target to determine if they match the features of icebergs, vessels or other ‘false alarms’ (e.g., marine wildlife, clutter). The final stage is the production of an output surveillance product, which can be a standard iceberg chart (e.g., MANICE) or something that can be ingested into a GIS system (e.g., ESRI shapefile, Google KML). The target discrimination phase is one of the most important phases because it provides feedback to operations about the presence of targets of interest (icebergs and vessels). The authors have used computer vision techniques successfully to train target classifiers. Standard techniques usually result in classifier accuracies of between 85%-95%, depending on the resolution of the SAR (higher resolutions produce more accurate results) and the availability of multiple polarizations. To see if new machine learning techniques could be applied to increase classifier accuracy, a dataset of 5000 ship and iceberg targets were extracted from Sentinel-1 multi-channel data (HH,HV). The images were collected in several regions (Greenland, Grand Banks, and Strait of Gibraltar). Validation either came by way of supporting information from the offshore operations, or was inferred by location. An online machine learning competition was hosted by Kaggle, a company that conducts online competitions on behalf of their clients. The detection data were made available by Kaggle to the broad internet community. Kaggle has a loyal following of data scientists who regularly participate in Kaggle competitions. The competition was hosted over a three-month period; over 3300 teams participated in the competition. The competition produced an improved classifier over standard computer vision techniques; the top three competitors had 4-5 stage classifiers that increased classification accuracy by approximately 5%. Copyright 2018, Offshore Technology Conference." "57204047405;57201703592;57203344412;57204049934;","IOT based advanced medicine dispenser integrated with an interactive web application",2018,"10.14419/ijet.v7i4.10.20704","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85054278045&doi=10.14419%2fijet.v7i4.10.20704&partnerID=40&md5=2a76b2622d3185fc060dc2e8da1a8ff4","Internet of Things (IOT) is a development of the internet which plays a major role in integrating human-machine interaction by allowing everyday objects to send and receive data in a variety of applications. Using IOT in healthcare monitoring provides an avenue for doctors and patients to interact and to track the dosage of medication administered. The paper presents an interactive, user friendly network integrated with an automated medicine dispenser which uses IOT, cloud computing and machine learning. The network was built on a python tornado framework with a front end developed using materialise CSS. The feasibility of this approach was validated by building a prototype and conducting a survey. © 2018 Authors." "56427952900;57203866180;57209970192;","Machine learning for the communication optimization in distributed systems",2018,"10.14419/ijet.v7i4.1.19491","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85053241855&doi=10.14419%2fijet.v7i4.1.19491&partnerID=40&md5=5ddd4c4a69dc0310a755262f48808fd6","The objective of the work that is presented in this paper was the problem of the communication optimization and detection of the issues of computing resources performance degradation [1, 2] with the usage of machine learning techniques. Computer networks transmit payload data and the meta-data from numerous sources towards vast number of destinations, especially in multitenant environments [3, 4]. Meta data describes the payload data and could be analyzed for anomalies detection in the communication patterns. Communication patterns depend on the payload itself and technical protocol used. The technical patterns are the research target as their analysis could spotlight the vulnerable behavior, for example: unusual traffic, extra load transported and etc. There was a big data used to train model with a supervised machine learning. Dataset was collected from the network interfaces of the distributed application infrastructure. Machine Learning tools had been retained from the cloud services provider - Amazon Web Services. The stochastic gradient descent technique was utilized for the model training, so that it could represent the communication patterns in the system. The learning target parameter was a packet length, the regression was performed to understand the relationship between packet meta-data (timestamp, protocol, the source server) and its length. The root mean square error calculation was applied to evaluate the learning efficiency. After model was prepared using training dataset, the model was tested with the test dataset and then applied on the target dataset (dataset for prediction) to check whether it was capable to detect anomalies. The experimental part showed the applicability of machine learning for the communication optimization in the distributed application environment. By means of the trained artificial intelligence model, it was possible to predict target parameters of traffic and computing resources usage with purpose to avoid service degradation. Additionally, one could reveal anomalies in the transferred traffic between application components. The application of techniques is envisioned in information security field and in the field of efficient network resources planning. Further research could be in application machine learning techniques for more complicated distributed environments and enlarging the number of protocols to prepare communication patterns. © 2018 Authors." "57201990332;57201661007;54894167400;57201974960;","Cold Chain Management Using Model Based Design, Machine Learning Algorithms and Data Analytics",2018,"10.4271/2018-01-1201","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85046654459&doi=10.4271%2f2018-01-1201&partnerID=40&md5=8f676dd5b40e4c7bc79d193d92bafd75","In the food industry, there is an increased demand for generic pharmaceutical products and perishable food without compromising with the changes in texture and taste that occur in the transit. With this demand, there is a need for better visibility of products in the logistics network, to minimize wastage, to ensure product integrity, influence productivity, transparently track the fleet and to identify pathogens before a potential outbreak. In Cold Chain Management, information is power: with potentially billions of dollars' worth of cargo (such as food items, vaccines, serums, tests or chemicals) at stake worldwide. Hence, careful live monitoring, inspection, supervision, validation and documentation of business-critical information is essential. In this paper, we have proposed a framework for Cold Chain Management using Internet of Things (IoT) combined with other technological innovations such as: Cloud Computing, Machine Learning and Big Data Analytics to revolutionize the cold transport industry. By establishing such an architecture, we have tried to monitor, visualize, track and control various platform dependent parameters thereby providing a complete solution across the fleet cycle with assured freshness and palpability. © 2018 SAE International. All Rights Reserved." "22941140200;57200795580;55885604100;","Comparison of the filtering models for airborne LiDAR data by three classifiers with exploration on model transfer",2018,"10.1117/1.JRS.12.016021","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85042439463&doi=10.1117%2f1.JRS.12.016021&partnerID=40&md5=84b019e516d3fa422e055685320d277a","This paper discusses airborne light detection and ranging (LiDAR) point cloud filtering (a binary classification problem) from the machine learning point of view. We compared three supervised classifiers for point cloud filtering, namely, Adaptive Boosting, support vector machine, and random forest (RF). Nineteen features were generated from raw LiDAR point cloud based on height and other geometric information within a given neighborhood. The test datasets issued by the International Society for Photogrammetry and Remote Sensing (ISPRS) were used to evaluate the performance of the three filtering algorithms; RF showed the best results with an average total error of 5.50%. The paper also makes tentative exploration in the application of transfer learning theory to point cloud filtering, which has not been introduced into the LiDAR field to the authors' knowledge. We performed filtering of three datasets from real projects carried out in China with RF models constructed by learning from the 15 ISPRS datasets and then transferred with little to no change of the parameters. Reliable results were achieved, especially in rural area (overall accuracy achieved 95.64%), indicating the feasibility of model transfer in the context of point cloud filtering for both easy automation and acceptable accuracy. © 2018 Society of Photo-Optical Instrumentation Engineers (SPIE)." "57194471739;57200559318;57015781300;57202084374;","Resource allocation of cloud application through machine learning: A case study",2017,"10.1109/ICGI.2017.52","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85041695403&doi=10.1109%2fICGI.2017.52&partnerID=40&md5=17881c3c968b8dabf4ab555b41b238fc","With the rapid development of WEB applications, the demand for dynamically adjusting computing resources based on the load variation is increasing. However, most of the traditional WEB systems have limited ability to respond to load changes. In order to solve the problem, software self-Adaptation technology has been applied to the resource management of WEB systems. Many researchers have tried to propose various software self-Adaptation models, all of which contain the control loop 'Monitor-Analyze-Plan-Execute'. However, the 'Knowledge-base' of these models is based on predefined strategy or configuration, which increases the complexity of system development and maintenance. In this paper, machine learning is applied to software self-Adaptation through a case study, and the 'Knowledge-base' is given by machine learning, which greatly reduces the workload of system maintenance and rule configuration. The results show the case based on machine learning can construct the corresponding management strategy according to WEB system runtime status, and conduct software self-Adaptive management. © 2017 IEEE." "35094561600;57016429700;57192366005;13410467900;22952160500;","Demonstrating PlanetSense: Gathering geo-spatial intelligence from crowd-sourced and social-media data",2016,"10.1145/2996913.2996975","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85011030479&doi=10.1145%2f2996913.2996975&partnerID=40&md5=b8cb71c65f605744f3650d8259cefa4f","Crowd-sourced and volunteered information, social media, and participatory sensors are capable of providing real-time activity data. Monitoring these sources in time of relevance and then using them to gather operational knowledge is important during crisis management. Beyond that, it's important to curate this information for geo-spatial research purposes, including land use classification and population occupancy analysis. In this demonstration, we will showcase PlanetSense - a geo-spatial research platform built to harness the existing power of archived data and add to that, the dynamics of heterogeneous real-time streaming data from social media and volunteered sources, seamlessly integrated with sophisticated machine learning algorithms and visualization tools. A demonstration will focus on - 1) Recent initiative emphasizing the need to harness crowdsources, volunteered, and social media data at scale; 2) Anatomy and insight into data collection workflow. We will show the ability to harvest and process several terabytes of raw data in realtime; 3) A detailed discussion with insight into more than 20 sources of data will be given. These sources include text, sensors, as well as imagery data; 4) PlanetSense's end to end distributed architecture will be discussed with focus on collecting and processing high-volumes of streaming data in a Geo-Data Cloud. Data fusion methods and algorithms for integrating disparate data sources with existing legacy products. Data analytics and machine learning methods for generating operational intelligence on the fly; 5) In addition, PlanetSense ""App"" platform will be shown with hands-on application enabling interested audience to quickly develop and deploy solutions. 6) Several case studies will be discussed relevant to, land use classification, monitoring transient population, high-resolution occupancy analysis, mapping special events population, ability to uncover global breaking events and reactions in near-real time, ability to track protest, unrest, and monitor other societal turbulences as they happen, and real-time monitoring of infrastructure outages. © 2016 ACM." "57191839724;55648722300;55105432400;","Automated equipment recognition and classification from scattered point clouds for construction management applications",2016,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-84994323171&partnerID=40&md5=77dba3ba80adb2df840d73eb46f7b6c2","Recognizing construction assets from as-is point cloud data of construction environment provides essential information for engineering and management applications including progress monitoring, safety management, supply-chain management, and quality control. This study proposes a fast and automated processing pipeline for construction target assets recognition from scattered as-is point clouds. The recognition tasks can be subdivided into object detection, which involves computing the bounding box around each construction equipment, and object classification, which involves labelling point cloud clusters from discrete equipment categories, such as backhoe loader, bulldozer, dump truck, excavator and front loader. The object detection step consists of point cloud down-sampling, segmentation and clustering. For the object classification step, machine learning methods were employed to determine class membership probability using features derived from the ESF (Ensemble of Shape Functions) descriptor. The classifiers were trained on synthetic point clouds generated from CAD (Computer-aided Design) models. The method was validated using laser scanned point clouds from an equipment yard. The test results demonstrate promising advancements towards semantic labelling and scene understanding of point cloud data." "56330233100;","Predictive policing",2014,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-84906231906&partnerID=40&md5=66c4231e00cd8e96a480ae16c6031b2a","The article examines how machine learning can be combined with geospatial technologies to predict where crimes will happen next. Current law enforcement trends reflect an interest in predictive analysis, increasing adoption of cloud services, pervasive location information and the proliferation of mobile devices. These trends are emerging in extremely constrained budget environments for many agencies and communities. Azavea recently set out to take advantage of these trends by combining geographic analysis with contemporary machine learning algorithms and cloud computing to help police departments make more effective use of their limited resources. The team developing HunchLab sought not only to implement the yearly warning system represented by the original prototypes, but also to use crime patterns to predict areas at higher than normal risk. The goal was to help officers prevent spikes in crime from even occurring. The team reviewed the academic literature and decided to implement two analytic techniques: near-repeat pattern analysis and cyclic-load forecasting." "55932962700;6506085008;56516456400;8984213200;57217293480;","A new framework for classification of distributed denial of service (DDOS) attack in cloud computing by machine learning techniques",2014,"10.1166/asl.2014.5288","https://www.scopus.com/inward/record.uri?eid=2-s2.0-84887857071&doi=10.1166%2fasl.2014.5288&partnerID=40&md5=c37e2c58568fa5442cc93480d7645401","The growing nature of cloud computing, increases the urgency of appropriate solution for its relevant threats. Security of cloud is the top challenge among all challenges that cloud computing deals with. Indeed, security issues of cloud have a reverse relation with cloud users' desires to move their assets to cloud environment. Distributed Denial of Service (DDOS) attack is one of the most popular attacks for security experts, which is easy to be carried out; however, it has some catastrophic effects. On the other hand, if a kind of attack like DDOS happens in cloud environment, cloud providers are not eager to share this information with their users due to loss of reputation. However, cloud users demand a mechanism to find out whether their assets are secure or not. To address the aforementioned problem, in this paper a new framework for classification of DDOS attack by utilization of machine learning techniques has been proposed. For this purpose, we will collect a dataset from the resources that are dedicated to user's virtual machine. This file will be delivered to a machine learning tool for future activities. © 2014 American Scientific Publishers All rights reserved." "14831081900;57219272291;","A two-moment machine learning parameterization of the autoconversion process",2021,"10.1016/j.atmosres.2020.105269","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091987463&doi=10.1016%2fj.atmosres.2020.105269&partnerID=40&md5=1ef132ae04b9cd1b3f16472ed3806bf6","Autoconversion is the mass transfer from cloud to precipitation water in an early stage of cloud development, and is the dominant process in the formation of embryonic droplets that trigger precipitation formation. The accurate parameterization of this process is key, in order to improve the interaction between cloud microphysics and cloud dynamics for models from cloud scale to the global climate scale. For model based parameterizations of the auto-conversion process, the usual approach to develop an autoconversion parameterization is by curve fitting the autoconversion rates obtained from simulations or numerical solutions of the kinetic collection equation under a wide range of initial conditions. However, in this case, the autoconversion is modeled by a function that is a nonlinear product of liquid water content and droplet concentration and depends on a small number of parameters. As a result, a large amount of scatter around the actual values can be obtained, indicating a weak relationship between actual and fitted autoconversion rates. The purpose of this paper is to analyze whether neural networks are better than traditional curve fitting or regression to obtain parameterizations of autoconversion. Then, a deep neural network was trained from an autconversion rates dataset generated by solving the kinetic collection equation for a wide range of droplet concentrations and liquid water contents. The obtained machine learned parameterization shows a very good match with actual rates calculated from the kinetic collection equation. © 2020 Elsevier B.V." "57219955783;36620043400;57215559368;57213355177;57216965386;","Extraction of built-up area using multi-sensor data—A case study based on Google earth engine in Zhejiang Province, China",2021,"10.1080/01431161.2020.1809027","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096244274&doi=10.1080%2f01431161.2020.1809027&partnerID=40&md5=57ed9e48eaf1e8ff57f62890d363981c","Accurate and up-to-date built-up area mapping is of great importance to the science community, decision-makers, and society. Therefore, satellite-based, built-up area (BUA) extraction at medium resolution with supervised classification has been widely carried out. However, the spectral confusion between BUA and bare land (BL) is the primary hindering factor for accurate BUA mapping over large regions. Here we propose a new methodology for the efficient BUA extraction using multi-sensor data under Google Earth Engine cloud computing platform. The proposed method mainly employs intra-annual satellite imagery for water and vegetation masks, and a random-forest machine learning classifier combined with auxiliary data to discriminate between BUA and BL. First, a vegetation mask and water mask are generated using NDVI (normalized differenced vegetation index) max in vegetation growth periods and the annual water-occurrence frequency. Second, to accurately extract BUA from unmasked pixels, consisting of BUA and BL, random-forest-based classification is conducted using multi-sensor features, including temperature, night-time light, backscattering, topography, optical spectra, and NDVI time-series metrics. This approach is applied in Zhejiang Province, China, and an overall accuracy of 92.5% is obtained, which is 3.4% higher than classification with spectral data only. For large-scale BUA mapping, it is feasible to enhance the performance of BUA mapping with multi-temporal and multi-sensor data, which takes full advantage of datasets available in Google Earth Engine. © 2020 Informa UK Limited, trading as Taylor & Francis Group." "36117793500;16178105300;6504046934;","Toward making canopy hemispherical photography independent of illumination conditions: A deep-learning-based approach",2021,"10.1016/j.agrformet.2020.108234","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85095700535&doi=10.1016%2fj.agrformet.2020.108234&partnerID=40&md5=17ed4e609f91323e12beefc7f222ee16","Hemispherical photography produces the most accurate results when working with well-exposed photographs acquired under diffuse light conditions (diffuse-light images). Obtaining such data can be prohibitively expensive when surveying hundreds of plots is required. A relatively inexpensive alternative is using photographs acquired under direct sunlight (sunlight images). However, this practice leads to high errors since the standard processing algorithms expect diffuse-light imagery. Here, instead of using classification algorithms, which is the unique dominant practice, we approached the processing of sunlight images using deep learning (DL) regression. We implemented DL systems by using the general-purpose convolutional neural networks known as VGGNet 16, VGGNet 19, Res-Net, and SE-ResNet. We trained them with 608 samples acquired in a South American temperate forest populated by Nothofagus pumilio. For their evaluation, we used 113 independent samples. Each sample (X, Y) consisted of one or several sunlight images (X), and the plant area index (PAI) and effective PAI (PAIe) extracted from a diffuse-light image (Y). The sunlight images include clear sky and broken clouds with sun elevation from 15° to 47°. We obtained the best results with the SE-ResNet architecture. The system requires a low-resolution input reprojected to cylindrical, and it can make predictions with 10% root mean square error, even from pictures acquired with automatic exposure, which challenge previous findings. Furthermore, similar results (R2= 0.9, n = 104) can be obtained by feeding the system with photographs acquired with an inexpensive fisheye converter attached to a smartphone. Altogether, results indicate that our approach is a cost-efficient option for surveying hundreds of plots under direct sunlight. Therefore, combining our method with the traditional procedures provides processing solutions for virtually all kinds of illumination conditions. © 2020" "57214306009;55782832600;57189906017;","Roof bolt identification in underground coal mines from 3D point cloud data using local point descriptors and artificial neural network",2021,"10.1080/2150704X.2020.1809734","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096040035&doi=10.1080%2f2150704X.2020.1809734&partnerID=40&md5=82ebc9a4117da16f3e2c6c298cf7930d","Roof bolts are commonly used to provide structural support in underground mines. Frequent and automated assessment of roof bolt is critical to closely monitor any change in the roof conditions while preventing major hazards such as roof fall. However, due to challenging conditions at mine sites such as sub-optimal lighting and restrictive access, it is difficult to routinely assess roof bolts by visual inspection or traditional surveying. To overcome these challenges, this study presents an automated method of roof bolt identification from 3D point cloud data, to assist in spatio-temporal monitoring efforts at mine sites. An artificial neural network was used to classify roof bolts and extract them from 3D point cloud using local point descriptors such as the proportion of variance (POV) over multiple scales, radial surface descriptor (RSD) over multiple scales and fast point feature histogram (FPFH). Accuracy was evaluated in terms ofprecision, recall and quality metric generally used in classification studies. The generated results were compared against other machine learning algorithms such as weighted k-nearest neighbours (k-NN), ensemble subspace k-NN, support vector machine (SVM) and random forest (RF), and was found to be superior by up to 8% in terms of the achieved quality metric. © 2020 Informa UK Limited, trading as Taylor & Francis Group." "57210946877;56770984700;6701474321;","A machine learning approach for the detection of supporting rock bolts from laser scan data in an underground mine",2021,"10.1016/j.tust.2020.103656","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092391600&doi=10.1016%2fj.tust.2020.103656&partnerID=40&md5=9d04bdc4a5e8c3acecae449065671350","Rock bolts are a crucial part of underground infrastructure support; however, current methods to locate and record their positions are manual, time consuming and generally incomplete. This paper describes an effective method to automatically locate supporting rock bolts from a 3D laser scanned point cloud. The proposed method utilises a machine learning classifier combined with point descriptors based on neighbourhood properties to classify all data points as either ‘bolt’ or ‘not-bolt’ before using the Density Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to divide the results into candidate bolt objects. The centroids of these objects are then computed and output as simple georeferenced 3D coordinates to be used by surveyors, mine managers and automated machines. Two classifiers were tested, a random forest and a shallow neural network, with the neural network providing the more accurate results. Alongside the different classifiers, different input feature types were also examined, including the eigenvalue based geometric features popular in the remote sensing community and the point histogram based features more common in the mobile robotics community. It was found that a combination of both feature sets provided the strongest results. The obtained precision and recall scores were 0.59 and 0.70 for the individual laser points and 0.93 and 0.86 for the bolt objects. This demonstrates that the model is robust to noise and misclassifications, as the bolt is still detected even if edge points are misclassified, provided that there are enough correct points to form a cluster. In some cases, the model can detect bolts which are not visible to the human interpreter. © 2020" "57204794658;6603846580;57215661230;57214862416;54421097700;57163606900;57194165611;57193830300;56786235200;","Soil variability and quantification based on Sentinel-2 and Landsat-8 bare soil images: A comparison",2021,"10.1016/j.rse.2020.112117","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091914235&doi=10.1016%2fj.rse.2020.112117&partnerID=40&md5=55094a26d6493e88c53586c0e3024e55","There is a worldwide need for detailed spatial information to support soil mapping, mainly in the tropics, where main agricultural areas are concentrated. In this line, satellite images are useful tools that can assist in obtaining soil information from a synoptic point of view. This study aimed at evaluating how satellite images at different resolutions (spatial, spectral and temporal) can influence the representation of soil variability over time, the percentage of bare soil areas and spatial predictions of soil properties in southeastern Brazil. We used single-date and multi-temporal images (SYSI, Synthetic Soil Images) of bare soil pixels from the Sentinel2-MultiSpectral Instrument (S2-MSI) and the Landsat-8 Operational Land Imager (L8-OLI) to conduct this research. Two SYSIs were obtained from images acquired in four years (2016–2019) for each satellite (SYSI S2-MSI and SYSI L8-OLI) and a third SYSI, named SYSI Combined, was obtained by combining the images from both satellites. The single-date images for each satellite was acquired in September, when the influence of clouds was low and bare soil pixels was predominant. Single-date images and SYSIs were compared by means of their spectral patterns and ability to predict topsoil properties (clay, sand, silt, and organic matter contents and soil color) using the Cubist algorithm. We found that the SYSIs outperformed single-date images and that the SYSI Combined and SYSI L8-OLI provided the best prediction performances. The SYSIs also had the highest percentage of areas with bare soil pixels (~30–50%) when compared to the single-date images (~20%). Our results suggest that bare soil images obtained by combining Landsat-8 and Sentinel-2 images are more important for soil mapping than spatial or spectral resolutions. Soil maps obtained via satellite images are important tools for soil survey, land planning, management and precision agriculture. © 2020 Elsevier Inc." "25931139100;6701481007;6506992267;","Global monitoring of deep convection using passive microwave observations",2021,"10.1016/j.atmosres.2020.105244","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090568601&doi=10.1016%2fj.atmosres.2020.105244&partnerID=40&md5=ea1918a664ad2600dc803abe0b4b1d09","In this study, we present the DEEPSTORM (DEEP moiSt aTmospheric cOnvection from micRowave radioMeter) algorithm, able to retrieve ice water path (IWP) and to detect deep moist atmospheric convection (DC) from 80°S to 80°N using observations from four spaceborne passive microwave radiometers. DEEPSTORM is based on a machine learning approach and is fitted against observations from the CPR (Cloud Profiling Radar) spaceborne radar on-board CloudSat. IWP predictions show an average root mean square error of 0.27 kg/m2 and a correlation index of 0.87. DC occurrence is detected with a probability of 59% and a false alarm rate of 24%. The prediction accuracy of IWP and DC is significantly better when the IWP exceeds 0.5 kg/m2 showing that DEEPSTORM is well suited to detect and characterise the strongest DC events. Overall DC detection is more accurate in the tropics than in mid-latitudes while the IWP retrieval works better in the mid-latitudes. Two examples illustrating the potential of DEEPSTORM are presented: the IWP is retrieved during Hurricane Matthew in 2016, and a climatology of DC occurrences between September 2016 and December 2016 is presented. This work will allow building a quasi-worldwide and 20-year long database of DC occurrence and intensity. © 2020 Elsevier B.V." "57208864702;57201315852;7401683250;35337273500;36136336500;57190869353;12781228700;57204719415;57205301324;","Observations of single-stroke flashes from five isolated small thunderstorms in East China",2020,"10.1016/j.jastp.2020.105441","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092165196&doi=10.1016%2fj.jastp.2020.105441&partnerID=40&md5=a044bcc6bdb6384824f4406653f283fb","This paper presents the ground truth datasets of negative cloud-to-ground (NCG) flashes with one single stroke in five isolated small thunderstorms. By developing a machine-learning method based on the convolutional neural network, we identify the return strokes during the whole life of thunderstorms with the wideband electric field waveform of lightning discharges detected by Jianghuai Area Sferic Array. The distribution of flash multiplicity in these thunderstorms exhibits an exponential decrease pattern. The single-stroke flashes (with multiplicity = 1) account for more than 30% of all NCG flashes on a thunderstorm basis, and the proportion of single-stroke flashes tended to be most abundant. When observed at a 20-min interval, the percentage of single-stroke flashes tends to change dramatically during thunderstorm developments, which seems to show an opposite trend with time to the maximum flash multiplicity. Single-stroke flashes tended to be associated with a weaker peak current of initial stroke compared to that of multiple-stroke flashes. It is inferred that the horizontal scale of negative charge regions in thunderclouds might play an important role in enhancing the flash multiplicity. © 2020 Elsevier Ltd" "55939190800;57053470200;35518435600;57211533913;56245008400;22960711400;57150278600;16029719200;13905662200;7007022537;51863613500;","Ensemble-based deep learning for estimating PM2.5 over California with multisource big data including wildfire smoke",2020,"10.1016/j.envint.2020.106143","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091646921&doi=10.1016%2fj.envint.2020.106143&partnerID=40&md5=2e5b08366201ecce30c17488798fef97","Introduction: Estimating PM2.5 concentrations and their prediction uncertainties at a high spatiotemporal resolution is important for air pollution health effect studies. This is particularly challenging for California, which has high variability in natural (e.g, wildfires, dust) and anthropogenic emissions, meteorology, topography (e.g. desert surfaces, mountains, snow cover) and land use. Methods: Using ensemble-based deep learning with big data fused from multiple sources we developed a PM2.5 prediction model with uncertainty estimates at a high spatial (1 km × 1 km) and temporal (weekly) resolution for a 10-year time span (2008–2017). We leveraged autoencoder-based full residual deep networks to model complex nonlinear interrelationships among PM2.5 emission, transport and dispersion factors and other influential features. These included remote sensing data (MAIAC aerosol optical depth (AOD), normalized difference vegetation index, impervious surface), MERRA-2 GMI Replay Simulation (M2GMI) output, wildfire smoke plume dispersion, meteorology, land cover, traffic, elevation, and spatiotemporal trends (geo-coordinates, temporal basis functions, time index). As one of the primary predictors of interest with substantial missing data in California related to bright surfaces, cloud cover and other known interferences, missing MAIAC AOD observations were imputed and adjusted for relative humidity and vertical distribution. Wildfire smoke contribution to PM2.5 was also calculated through HYSPLIT dispersion modeling of smoke emissions derived from MODIS fire radiative power using the Fire Energetics and Emissions Research version 1.0 model. Results: Ensemble deep learning to predict PM2.5 achieved an overall mean training RMSE of 1.54 μg/m3 (R2: 0.94) and test RMSE of 2.29 μg/m3 (R2: 0.87). The top predictors included M2GMI carbon monoxide mixing ratio in the bottom layer, temporal basis functions, spatial location, air temperature, MAIAC AOD, and PM2.5 sea salt mass concentration. In an independent test using three long-term AQS sites and one short-term non-AQS site, our model achieved a high correlation (>0.8) and a low RMSE (<3 μg/m3). Statewide predictions indicated that our model can capture the spatial distribution and temporal peaks in wildfire-related PM2.5. The coefficient of variation indicated highest uncertainty over deciduous and mixed forests and open water land covers. Conclusion: Our method can be generalized to other regions, including those having a mix of major urban areas, deserts, intensive smoke events, snow cover and complex terrains, where PM2.5 has previously been challenging to predict. Prediction uncertainty estimates can also inform further model development and measurement error evaluations in exposure and health studies. © 2020 The Author(s)" "57218881802;56531367400;7005742190;55628589750;56203143700;","Determinants of fog and low stratus occurrence in continental central Europe – a quantitative satellite-based evaluation",2020,"10.1016/j.jhydrol.2020.125451","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090574057&doi=10.1016%2fj.jhydrol.2020.125451&partnerID=40&md5=c950c27b3e1f9a37b765a645a8fb51c8","The formation and development of fog and low stratus clouds (FLS) depend on meteorological and land surface conditions and their interactions with each other. While analyses of temporal and spatial patterns of FLS in Europe exist, the interactions between FLS determinants underlying them have not been studied explicitly and quantitatively at a continental scale yet. In this study, a state-of-the-art machine learning technique is applied to model FLS occurrence over continental Europe, using meteorological and land surface parameters from geostationary satellite and reanalysis data. Spatially explicit model units are created to test for spatial and seasonal differences in model performance and FLS sensitivities to changes in predictors, and effects of different data preprocessing procedures are evaluated. The statistical models show good performance in predicting FLS occurrence during validation, with R2>0.9 especially in winter high pressure situations.The predictive skill of the models seems to be dependent on data availability, data preprocessing, time period, and geographic characteristics. It is shown that atmospheric proxies are more important determinants of FLS presence than surface characteristics, in particular mean sea level pressure, near-surface wind speed and evapotranspiration are crucial, together with FLS occurrence on the previous day. Higher wind speeds, higher land surface temperatures and higher evapotranspiration tend to be negatively related to FLS. Spatial patterns of feature importance show the dominant influence of mean sea level pressure on FLS occurrence throughout the central European domain. When only high pressure situations are considered, wind speed (in the western study region) and evapotranspiration (in the eastern study region) gain importance, highlighting the influence of moisture advection on FLS occurrence in the western parts of the central European domain. This study shows that FLS occurrence can be accurately modeled using machine learning techniques in large spatial domains based on meteorological and land surface predictors. The statistical models used in this study provide a novel analysis tool for investigating empirical relationships in the FLS – land surface system and possibly infer processes. © 2020 Elsevier B.V." "36617930000;57204769007;35517567400;57212508254;57218645527;57218652245;57218652947;","PODPAC: open-source Python software for enabling harmonized, plug-and-play processing of disparate earth observation data sets and seamless transition onto the serverless cloud by earth scientists",2020,"10.1007/s12145-020-00506-0","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089885051&doi=10.1007%2fs12145-020-00506-0&partnerID=40&md5=8909c5c52e617c6c6133947b87cd7de3","In this paper, we present the Pipeline for Observational Data Processing, Analysis, and Collaboration (PODPAC) software. PODPAC is an open-source Python library designed to enable widespread exploitation of NASA earth science data by enabling multi-scale and multi-windowed access, exploration, and integration of available earth science datasets to support analysis and analytics; automatic accounting for geospatial data formats, projections, and resolutions; simplified implementation and parallelization of geospatial data processing routines; standardized sharing of data and algorithms; and seamless transition of algorithms and data products from local development to distributed, serverless processing on commercial cloud computing environments. We describe the key elements of PODPAC’s architecture, including Nodes for unified encapsulation of disparate scientific data sources; Algorithms for plug-and-play processing and harmonization of multiple data source Nodes; and Lambda functions for serverless execution and sharing of new data products via the cloud. We provide an overview of our open-source code implementation and testing process for development and deployment of PODPAC. We describe our interactive, JupyterLab-based end-user documentation including quick-start examples and detailed use case studies. We conclude with examples of PODPAC’s application to: encapsulate data sources available on Amazon Web Services (AWS) Open Data repository; harmonize processing of multiple earth science data sets for downscaling of NASA Soil Moisture Active Passive (SMAP) soil moisture data; and deploy a serverless SMAP-based drought monitoring application for use access from mobile devices. We postulate that PODPAC will also be an effective tool for wrangling and standardizing massive earth science data sets for use in model training for machine learning applications. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature." "8575756400;57193212462;57217205306;57217202313;8510127900;","Enabling radiation tolerant heterogeneous GPU-based onboard data processing in space",2020,"10.1007/s12567-020-00321-9","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086737276&doi=10.1007%2fs12567-020-00321-9&partnerID=40&md5=5a642eef8e8bc1c608758f0a19bd9290","The last decade has seen a dramatic increase in small satellite missions for commercial, public, and government intelligence applications. Given the rapid commercialization of constellation-driven services in Earth Observation, situational domain awareness, communications including machine-to-machine interface, exploration etc., small satellites represent an enabling technology for a large growth market generating truly Big Data. Examples of modern sensors that can generate very large amounts of data are optical sensing, hyperspectral, Synthetic Aperture Radar (SAR), and Infrared imaging. Traditional handling and downloading of Big Data from space requires a large onboard mass storage and high bandwidth downlink with a trend towards optical links. Many missions and applications can benefit significantly from onboard cloud computing similarly to Earth-based cloud services. Hence, enabling space systems to provide near real-time data and enable low latency distribution of critical and time sensitive information to users. In addition, the downlink capability can be more effectively utilized by applying more onboard processing to reduce the data and create high value information products. This paper discusses current implementations and roadmap for leveraging high performance computing tools and methods on small satellites with radiation tolerant hardware. This includes runtime analysis with benchmarks of convolutional neural networks and matrix multiplications using industry standard tools (e.g., TensorFlow and PlaidML). In addition, a ½ CubeSat volume unit (0.5U) (10 × 10 × 5 cm3) cloud computing solution, called SpaceCloud™ iX5100 based on AMD 28 nm APU technology is presented as an example of heterogeneous computer solution. An evaluation of the AMD 14 nm Ryzen APU is presented as a candidate for future advanced onboard processing for space vehicles. © 2020, The Author(s)." "26422389600;57208228475;6602942477;35473149000;","Landslide detection in mountainous forest areas using polarimetry and interferometric coherence",2020,"10.1186/s40623-020-01191-5","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084662765&doi=10.1186%2fs40623-020-01191-5&partnerID=40&md5=588acb58605878691f49f9ccdb777f5e","The cloud-free, wide-swath, day-and-night observation capability of synthetic aperture radar (SAR) has an important role in rapid landslide monitoring to reduce economic and human losses. Although interferometric SAR (InSAR) analysis is widely used to monitor landslides, it is difficult to use that for rapid landslide detection in mountainous forest areas because of significant decorrelation. We combined polarimetric SAR (PolSAR), InSAR, and digital elevation model (DEM) analysis to detect landslides induced by the July 2017 Heavy Rain in Northern Kyushu and by the 2018 Hokkaido Eastern Iburi Earthquake. This study uses fully polarimetric L-band SAR data from the ALOS-2 PALSAR-2 satellite. The simple thresholding of polarimetric parameters (alpha angle and Pauli components) was found to be effective. The study also found that supervised classification using PolSAR, InSAR, and DEM parameters provided high accuracy, although this method should be used carefully because its accuracy depends on the geological characteristics of the training data. Regarding polarimetric configurations, at least dual-polarimetry (e.g., HH and HV) is required for landslide detection, and quad-polarimetry is recommended. These results demonstrate the feasibility of rapid landslide detection using L-band SAR images.[Figure not available: see fulltext.] © 2020, The Author(s)." "57219925946;57217865914;","Using machine learning to derive cloud condensation nuclei number concentrations from commonly available measurements",2020,"10.5194/acp-20-12853-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096103971&doi=10.5194%2facp-20-12853-2020&partnerID=40&md5=4a8649af36ec39f211d06d7e193b37d7","Cloud condensation nuclei (CCN) number concentrations are an important aspect of aerosol-cloud interactions and the subsequent climate effects; however, their measurements are very limited. We use a machine learning tool, random decision forests, to develop a random forest regression model (RFRM) to derive CCN at 0.4% supersaturation ([CCN0.4]) from commonly available measurements. The RFRM is trained on the long-Term simulations in a global size-resolved particle microphysics model. Using atmospheric state and composition variables as predictors, through associations of their variabilities, the RFRM is able to learn the underlying dependence of [CCN0.4] on these predictors, which are as follows: eight fractions of PM2:5 (NH4, SO4, NO3, secondary organic aerosol (SOA), black carbon (BC), primary organic carbon (POC), dust, and salt), seven gaseous species (NOx , NH3, O3, SO2, OH, isoprene, and monoterpene), and four meteorological variables (temperature (T), relative humidity (RH), precipitation, and solar radiation). The RFRM is highly robust: it has a median mean fractional bias (MFB) of 4:4% with 96:33% of the derived [CCN0.4] within a good agreement range of-60% MFB C60% and strong correlation of Kendall's coefficient 0:88. The RFRM demonstrates its robustness over 4 orders of magnitude of [CCN0.4] over varying spatial (such as continental to oceanic, clean to polluted, and nearsurface to upper troposphere) and temporal (from the hourly to the decadal) scales. At the Atmospheric Radiation Measurement Southern Great Plains observatory (ARM SGP) in Lamont, Oklahoma, United States, long-Term measurements for PM2:5 speciation (NH4, SO4, NO3, and organic carbon (OC)), NOx , O3, SO2, T, and RH, as well as [CCN0.4] are available. We modify, optimize, and retrain the developed RFRM to make predictions from 19 to 9 of these available predictors. This retrained RFRM (RFRM-ShortVars) shows a reduction in performance due to the unavailability and sparsity of measurements (predictors); it captures the [CCN0.4] variability and magnitude at SGP with 67:02% of the derived values in the good agreement range. This work shows the potential of using the more commonly available measurements of PM2:5 speciation to alleviate the sparsity of CCN number concentrations' measurements. © 2020 Copernicus GmbH. All rights reserved." "57219947521;57203656607;","A hybrid spatio-temporal prediction model for solar photovoltaic generation using numerical weather data and satellite images",2020,"10.3390/rs12223706","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096228596&doi=10.3390%2frs12223706&partnerID=40&md5=26b4d7f4a5274091fadb3739918a065c","Precise and accurate prediction of solar photovoltaic (PV) generation plays a major role in developing plans for the supply and demand of power grid systems. Most previous studies on the prediction of solar PV generation employed only weather data composed of numerical text data. The numerical text weather data can reflect temporal factors, however, they cannot consider the movement features related to the wind direction of the spatial characteristics, which include the amount of both clouds and particulate matter (PM) among other weather features. This study aims developing a hybrid spatio-temporal prediction model by combining general weather data and data extracted from satellite images having spatial characteristics. A model for hourly prediction of solar PV generation is proposed using data collected from a solar PV power plant in Incheon, South Korea. To evaluate the performance of the prediction model, we compared and performed ARIMAX analysis, which is a traditional statistical time-series analysis method, and SVR, ANN, and DNN, which are based on machine learning algorithms. The models that reflect the temporal and spatial characteristics exhibited better performance than those using only the general weather numerical data or the satellite image data. © MDPI AG. All rights reserved." "57219949429;23487156500;57219950276;56372626300;57219950011;","Combining phenological camera photos and modis reflectance data to predict gpp daily dynamics for alpine meadows on the tibetan plateau",2020,"10.3390/rs12223735","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096227333&doi=10.3390%2frs12223735&partnerID=40&md5=162222b642ee9f9a7108bee37dfb00e4","Gross primary production (GPP) is the overall photosynthetic fixation of carbon per unit space and time. Due to uncertainties resulting from clouds, snow, aerosol, and topography, it is a challenging task to accurately estimate daily GPP. Daily digital photos from a phenological camera record vegetation daily greenness dynamics with little cloud or aerosol disturbance. It can be fused with satellite remote sensing data to improve daily GPP prediction accuracy. In this study, we combine the two types of datasets to improve the estimation accuracy of GPP for alpine meadow on the Tibetan Plateau. To examine the performance of different methods and vegetation indices (VIs), three experiments were designed. First, GPP was estimated with the light use efficiency (LUE) model with the green chromatic coordinate (GCC) from the phenological camera and vegetation index from MODIS, respectively. Second, GPP was estimated with the Backpropagation neural network machine learning algorithm (BNNA) method with GCC from the phenological camera and vegetation index from MODIS, respectively. Finally, GPP was estimated with the BNNA method using GCC and vegetation index as inputs at the same time. Compared with eddy covariance GPP, GPP predicted by the BNNA method with GCC and vegetation indices as inputs at the same time showed the highest accuracy of all the experiments. The results indicated that GCC had a higher accuracy than NDVI and EVI when only one vegetation index data was used in the LUE model or the BNNA method. The R2 of GPP estimated by BNNA and GPP from eddy covariance increased by 0.12 on average, RMSE decreased by 1.13 g C·m−2·day −1 on average, and MAD decreased by 0.87 g C·m−2·day −1 on average compared with GPP estimated by the traditional LUE model and GPP from eddy covariance. This study puts forth a new way to improve the estimation accuracy of GPP on the Tibetan Plateau. With the emergence of a large number of phenological cameras, this method has great potential for use on the Tibetan Plateau, which is heavily affected by clouds and snow. © MDPI AG. All rights reserved." "57219937609;35763111200;","Object-oriented lulc classification in google earth engine combining snic, glcm, and machine learning algorithms",2020,"10.3390/rs12223776","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096173988&doi=10.3390%2frs12223776&partnerID=40&md5=e36dbd8f108e196677d7ce1e93936f40","Google Earth Engine (GEE) is a versatile cloud platform in which pixel-based (PB) and object-oriented (OO) Land Use–Land Cover (LULC) classification approaches can be implemented, thanks to the availability of the many state-of-art functions comprising various Machine Learning (ML) algorithms. OO approaches, including both object segmentation and object textural analysis, are still not common in the GEE environment, probably due to the difficulties existing in concatenating the proper functions, and in tuning the various parameters to overcome the GEE computational limits. In this context, this work is aimed at developing and testing an OO classification approach combining the Simple Non-Iterative Clustering (SNIC) algorithm to identify spatial clusters, the Gray-Level Co-occurrence Matrix (GLCM) to calculate cluster textural indices, and two ML algorithms (Random Forest (RF) or Support Vector Machine (SVM)) to perform the final classification. A Principal Components Analysis (PCA) is applied to the main seven GLCM indices to synthesize in one band the textural information used for the OO classification. The proposed approach is implemented in a user-friendly, freely available GEE code useful to perform the OO classification, tuning various parameters (e.g., choose the input bands, select the classification algorithm, test various segmentation scales) and compare it with a PB approach. The accuracy of OO and PB classifications can be assessed both visually and through two confusion matrices that can be used to calculate the relevant statistics (producer’s, user’s, overall accuracy (OA)). The proposed methodology was broadly tested in a 154 km2 study area, located in the Lake Trasimeno area (central Italy), using Landsat 8 (L8), Sentinel 2 (S2), and PlanetScope (PS) data. The area was selected considering its complex LULC mosaic mainly composed of artificial surfaces, annual and permanent crops, small lakes, and wooded areas. In the study area, the various tests produced interesting results on the different datasets (OA: PB RF (L8 = 72.7%, S2 = 82%, PS = 74.2), PB SVM (L8 = 79.1%, S2 = 80.2%, PS = 74.8%), OO RF (L8 = 64%, S2 = 89.3%, PS = 77.9), OO SVM (L8 = 70.4, S2 = 86.9%, PS = 73.9)). The broad code application demonstrated very good reliability of the whole process, even though the OO classification process resulted, sometimes, too demanding on higher resolution data, considering the available computational GEE resources. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "16485521200;7102795549;7004063850;","Exploring the Constraints on Simulated Aerosol Sources and Transport Across the North Atlantic With Island-Based Sun Photometers",2020,"10.1029/2020EA001392","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096467787&doi=10.1029%2f2020EA001392&partnerID=40&md5=95c6345f6b7f0a381c6fee9263b0aff7","Atmospheric aerosol over the North Atlantic Ocean impacts regional clouds and climate. In this work, we use a set of sun photometer observations of aerosol optical depth (AOD) located on the Graciosa and Cape Verde islands, along with the GEOS-Chem chemical transport model to investigate the sources of these aerosol and their transport over the North Atlantic Ocean. At both locations, the largest simulated contributor to aerosol extinction is the local source of sea-salt aerosol. In addition to this large source, we find that signatures consistent with long-range transport of anthropogenic, biomass burning, and dust emissions are apparent throughout the year at both locations. Model simulations suggest that this signal of long-range transport in AOD is more apparent at higher elevation locations; the influence of anthropogenic and biomass burning aerosol extinction is particularly pronounced at the height of Pico Mountain, near the Graciosa Island site. Using a machine learning approach, we further show that simulated observations at these three sites (near Graciosa, Pico Mountain, and Cape Verde) can be used to predict the simulated background aerosol imported into cities on the European mainland, particularly during the local winter months, highlighting the utility of background AOD monitoring for understanding downwind air quality. ©2020. The Authors." "56113469400;57203048222;","Modeling cloud-to-ground lightning probability in Alaskan tundra through the integration of Weather Research and Forecast (WRF) model and machine learning method",2020,"10.1088/1748-9326/abbc3b","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096331143&doi=10.1088%2f1748-9326%2fabbc3b&partnerID=40&md5=1dbb3fe7419d156e7d27ba454b1c498c","Wildland fires exert substantial impacts on tundra ecosystems of the high northern latitudes (HNL), ranging from biogeochemical impact on climate system to habitat suitability for various species. Cloud-to-ground (CG) lightning is the primary ignition source of wildfires. It is critical to understand mechanisms and factors driving lightning strikes in this cold, treeless environment to support operational modeling and forecasting of fire activity. Existing studies on lightning strikes primarily focus on Alaskan and Canadian boreal forests where land-atmospheric interactions are different and, thus, not likely to represent tundra conditions. In this study, we designed an empirical-dynamical method integrating Weather Research and Forecast (WRF) simulation and machine learning algorithm to model the probability of lightning strikes across Alaskan tundra between 2001 and 2017. We recommended using Thompson 2-moment and Mellor-Yamada-Janjic schemes as microphysics and planetary boundary layer parameterizations for WRF simulations in the tundra. Our modeling and forecasting test results have shown a strong capability to predict CG lightning probability in Alaskan tundra, with the values of area under the receiver operator characteristics curves above 0.9. We found that parcel lifted index and vertical profiles of atmospheric variables, including geopotential height, dew point temperature, relative humidity, and velocity speed, important in predicting lightning occurrence, suggesting the key role of convection in lightning formation in the tundra. Our method can be applied to data-scarce regions and support future studies of fire potential in the HNL. © 2020 The Author(s). Published by IOP Publishing Ltd." "57202739456;57203383511;57201106974;23089892300;","Discovery of new stellar groups in the Orion complex: Towards a robust unsupervised approach",2020,"10.1051/0004-6361/201935955","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096035966&doi=10.1051%2f0004-6361%2f201935955&partnerID=40&md5=c1e257b7038de2cbd1d76053c03b0ee5","We test the ability of two unsupervised machine learning algorithms, EnLink and Shared Nearest Neighbor (SNN), to identify stellar groupings in the Orion star-forming complex as an application to the 5D astrometric data from Gaia DR2. The algorithms represent two distinct approaches to limiting user bias when selecting parameter values and evaluating the relative weights among astrometric parameters. EnLink adopts a locally adaptive distance metric and eliminates the need for parameter tuning through automation. The original SNN relies only on human input for parameter tuning so we modified SNN to run in two stages. We first ran the original SNN 7000 times, each with a randomly generated sample according to within-source co-variance matrices provided in Gaia DR2 and random parameter values within reasonable ranges. During the second stage, we modified SNN to identify the most repeating stellar groups from the 25 798 we obtained in the first stage. We recovered 22 spatially and kinematically coherent groups in the Orion complex, 12 of which were previously unknown. The groups show a wide distribution of distances extending as far as about 150 pc in front of the star-forming Orion molecular clouds, to about 50 pc beyond them, where we, unexpectedly, find several groups. Our results reveal the wealth of sub-structure in the OB association, within and beyond the classical Blaauw Orion OBI sub-groups. A full characterization of the new groups is essential as it offers the potential to unveil how star formation proceeds globally in large complexes such as Orion. © ESO 2020." "57217524312;37117655800;36473337800;15842202000;","Automatic mapping of rice growth stages using the integration of sentinel-2, mod13q1, and sentinel-1",2020,"10.3390/rs12213613","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85095771467&doi=10.3390%2frs12213613&partnerID=40&md5=9ae21c4f0087dfa2015a893c35c1dae2","Rice (Oryza sativa L.) is a staple food crop for more than half of the world’s population. Rice production is facing a myriad of problems, including water shortage, climate, and land-use change. Accurate maps of rice growth stages are critical for monitoring rice production and assessing its impacts on national and global food security. Rice growth stages are typically monitored by coarse-resolution satellite imagery. However, it is difficult to accurately map due to the occurrence of mixed pixels in fragmented and patchy rice fields, as well as cloud cover, particularly in tropical countries. To solve these problems, we developed an automated mapping workflow to produce near real-time multi-temporal maps of rice growth stages at a 10-m spatial resolution using multisource remote sensing data (Sentinel-2, MOD13Q1, and Sentinel-1). This study was investigated between 1 June and 29 September 2018 in two (wet and dry) areas of Java Island in Indonesia. First, we built prediction models based on Sentinel-2, and fusion of MOD13Q1/Sentinel-1 using the ground truth information. Second, we applied the prediction models on all images in area and time and separation between the non-rice planting class and rice planting class over the cropping pattern. Moreover, the model’s consistency on the multitemporal map with a 5–30-day lag was investigated. The result indicates that the Sentinel-2 based model classification gives a high overall accuracy of 90.6% and the fusion model MOD13Q1/Sentinel-1 shows 78.3%. The performance of multitemporal maps was consistent between time lags with an accuracy of 83.27–90.39% for Sentinel-2 and 84.15% for the integration of Sentinel-2/MOD13Q1/Sentinel-1. The results from this study show that it is possible to integrate multisource remote sensing for regular monitoring of rice phenology, thereby generating spatial information to support local-, national-, and regional-scale food security applications. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57219842847;26664885100;35145722600;15220487300;6601942051;","A survey on the usage of blockchain technology for cyber-threats in the context of industry 4.0",2020,"10.3390/su12219179","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85095688402&doi=10.3390%2fsu12219179&partnerID=40&md5=16ddaf5169707b26c0dc07ef0296f325","A systematic review of the literature is presented related to the usage of blockchain technology (BCT) for cyber-threats in the context of Industry 4.0. BCT plays a crucial role in creating smart factories and it is recognized as a core technology that triggers a disruptive revolution in Industry 4.0. Beyond security, authentication, asset tracking and the exchange of smart contracts, BCTs allow terminals to exchange information according to mutually agreed rules within a secured manner. Consequently, BCT can play a crucial role in industrial sustainability by preserving the assets and the environment and by enhancing the quality of life of citizens. In this work, a classification of the most important cyber-attacks that occurred in the last decade in Industry 4.0 is proposed based on four classes. The latter classes cover scanning, local to remote, power of root and denial of service (DoS). BCT is also defined and various types belong to BCT are introduced and highlighted. Likewise, BCT protocols and implementations are discussed as well. BCT implementation includes linear structure and directed acyclic graph (DAG) technology. Then, a comparative study of the most relevant works based on BCT in Industry 4.0 is conducted in terms of confidentiality, integrity, availability, privacy and multifactor authentication features. Our review shows that the integration of BCT in industry can ensure data confidentiality and integrity and should be enforced to preserve data availability and privacy. Future research directions towards enforcing BCT in the industrial field by considering machine learning, 5G/6G mobile systems and new emergent technologies are presented. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "56816836100;7402706393;57208431792;6701726073;","Multi-temporal predictive modelling of sorghum biomass using uav-based hyperspectral and lidar data",2020,"10.3390/rs12213587","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85094980527&doi=10.3390%2frs12213587&partnerID=40&md5=44bac44a7a31c17d2fd6e11ed7a33821","High-throughput phenotyping using high spatial, spectral, and temporal resolution remote sensing (RS) data has become a critical part of the plant breeding chain focused on reducing the time and cost of the selection process for the “best” genotypes with respect to the trait(s) of interest. In this paper, the potential of accurate and reliable sorghum biomass prediction using visible and near infrared (VNIR) and short-wave infrared (SWIR) hyperspectral data as well as light detection and ranging (LiDAR) data acquired by sensors mounted on UAV platforms is investigated. Predictive models are developed using classical regression-based machine learning methods for nine experiments conducted during the 2017 and 2018 growing seasons at the Agronomy Center for Research and Education (ACRE) at Purdue University, Indiana, USA. The impact of the regression method, data source, timing of RS and field-based biomass reference data acquisition, and the number of samples on the prediction results are investigated. R2 values for end-of-season biomass ranged from 0.64 to 0.89 for different experiments when features from all the data sources were included. Geometry-based features derived from the LiDAR point cloud to characterize plant structure and chemistry-based features extracted from hyperspectral data provided the most accurate predictions. Evaluation of the impact of the time of data acquisition during the growing season on the prediction results indicated that although the most accurate and reliable predictions of final biomass were achieved using remotely sensed data from mid-season to end-of-season, predictions in mid-season provided adequate results to differentiate between promising varieties for selection. The analysis of variance (ANOVA) of the accuracies of the predictive models showed that both the data source and regression method are important factors for a reliable prediction; however, the data source was more important with 69% significance, versus 28% significance for the regression method. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "56677549900;9233163800;6507594945;7004181239;57206924573;57209398568;57193788172;6701412834;","Characterization of the far infrared properties and radiative forcing of antarctic ice and water clouds exploiting the spectrometer-lidar synergy",2020,"10.3390/rs12213574","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85094928144&doi=10.3390%2frs12213574&partnerID=40&md5=5345b21e4cc169ea21b4b6cd2e4259e6","Optical and microphysical cloud properties are retrieved from measurements acquired in 2013 and 2014 at the Concordia base station in the Antarctic Plateau. Two sensors are used synergistically: a Fourier transform spectroradiometer named REFIR-PAD (Radiation Explorer in Far Infrared-Prototype for Applications and Developments) and a backscattering-depolarization LiDAR. First, in order to identify the cloudy scenes and assess the cloud thermodynamic phase, the REFIR-PAD spectral radiances are ingested by a machine learning algorithm called Cloud Identification and Classification (CIC). For each of the identified cloudy scenes, the nearest (in time) LiDAR backscattering profile is processed by the Polar Threshold (PT) algorithm that allows derivation of the cloud top and bottom heights. Subsequently, using the CIC and PT results as external constraints, the Simultaneous Atmospheric and Clouds Retrieval (SACR) code is applied to the REFIR-PAD spectral radiances. SACR simultaneously retrieves cloud optical depth and effective dimensions and atmospheric vertical profiles of water vapor and temperature. The analysis determines an average effective diameter of 28 µm with an optical depth of 0.76 for the ice clouds. Water clouds are only detected during the austral Summer, and the retrieved properties provide an average droplet diameter of 9 µm and average optical depth equal to four. The estimated retrieval error is about 1% for the ice crystal/droplet size and 2% for the cloud optical depth. The sensitivity of the retrieved parameters to the assumed crystal shape is also assessed. New parametrizations of the optical depth and the longwave downwelling forcing for Antarctic ice and water clouds, as a function of the ice/liquid water path, are presented. The longwave downwelling flux, computed from the top of the atmosphere to the surface, ranges between 70 and 220 W/m2 . The estimated cloud longwave forcing at the surface is (31 ± 7) W/m2 and (29 ± 6) W/m2 for ice clouds and (64 ± 12) and (62 ± 11) W/m2 for water clouds, in 2013 and 2014, respectively. The total average cloud forcing for the two years investigated is (46 ± 9) W/m2 . © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57202643803;7004870145;6507811592;7102254283;57203178537;","Weather types affect rain microstructure: Implications for estimating rain rate",2020,"10.3390/rs12213572","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85094871679&doi=10.3390%2frs12213572&partnerID=40&md5=b2e2da0dd65cab4085fb69249fbeed89","Quantitative precipitation estimation (QPE) through remote sensing has to take rain microstructure into consideration, because it influences the relationship between radar reflectivity Z and rain intensity R. For this reason, separate equations are used to estimate rain intensity of convective and stratiform rain types. Here, we investigate whether incorporating synoptic scale meteorology could yield further QPE improvements. Depending on large-scale weather types, variability in cloud condensation nuclei and the humidity content may lead to variation in rain microstructure. In a case study for Bavaria, we measured rain microstructure at ten locations with laser-based disdrometers, covering a combined 18,600 h of rain in a period of 36 months. Rain was classified on a temporal scale of one minute into convective and stratiform based on a machine learning model. Large-scale wind direction classes were on a daily scale to represent the synoptic weather types. Significant variations in rain microstructure parameters were evident not only for rain types, but also for wind direction classes. The main contrast was observed between westerly and easterly circulations, with the latter characterized by smaller average size of drops and a higher average concentration. This led to substantial variation in the parameters of the radar rain intensity retrieval equation Z–R. The effect of wind direction on Z–R parameters was more pronounced for stratiform than convective rain types. We conclude that building separate Z–R retrieval equations for regional wind direction classes should improve radar-based QPE, especially for stratiform rain events. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "56684747900;56898969100;57194760236;57204481038;57218657147;57190089926;56528895600;10139058400;55946316700;7003505161;16070064500;","Application of google earth engine cloud computing platform, sentinel imagery, and neural networks for crop mapping in Canada",2020,"10.3390/rs12213561","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85094844639&doi=10.3390%2frs12213561&partnerID=40&md5=b59a405e6dfdc6c87553ff564e2257cb","The ability of the Canadian agriculture sector to make better decisions and manage its operations more competitively in the long term is only as good as the information available to inform decision-making. At all levels of Government, a reliable flow of information between scientists, practitioners, policy-makers, and commodity groups is critical for developing and supporting agricultural policies and programs. Given the vastness and complexity of Canada’s agricultural regions, space-based remote sensing is one of the most reliable approaches to get detailed information describing the evolving state of the country’s environment. Agriculture and Agri-Food Canada (AAFC)-the Canadian federal department responsible for agriculture-produces the Annual Space-Based Crop Inventory (ACI) maps for Canada. These maps are valuable operational space-based remote sensing products which cover the agricultural land use and non-agricultural land cover found within Canada’s agricultural extent. Developing and implementing novel methods for improving these products are an ongoing priority of AAFC. Consequently, it is beneficial to implement advanced machine learning and big data processing methods along with open-access satellite imagery to effectively produce accurate ACI maps. In this study, for the first time, the Google Earth Engine (GEE) cloud computing platform was used along with an Artificial Neural Networks (ANN) algorithm and Sentinel-1,-2 images to produce an object-based ACI map for 2018. Furthermore, different limitations of the proposed method were discussed, and several suggestions were provided for future studies. The Overall Accuracy (OA) and Kappa Coefficient (KC) of the final 2018 ACI map using the proposed GEE cloud method were 77% and 0.74, respectively. Moreover, the average Producer Accuracy (PA) and User Accuracy (UA) for the 17 cropland classes were 79% and 77%, respectively. Although these levels of accuracies were slightly lower than those of the AAFC’s ACI map, this study demonstrated that the proposed cloud computing method should be investigated further because it was more efficient in terms of cost, time, computation, and automation. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57203228090;57202987792;57213283167;57219670652;57216680871;22996553100;34769664200;15728256600;","Ffau—framework for fully autonomous uavs",2020,"10.3390/rs12213533","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85094681335&doi=10.3390%2frs12213533&partnerID=40&md5=3d2268b30d8d1bd9b81451ac6711fdfc","Unmanned Aerial Vehicles (UAVs), although hardly a new technology, have recently gained a prominent role in many industries being widely used not only among enthusiastic consumers, but also in high demanding professional situations, and will have a massive societal impact over the coming years. However, the operation of UAVs is fraught with serious safety risks, such as collisions with dynamic obstacles (birds, other UAVs, or randomly thrown objects). These collision scenarios are complex to analyze in real-time, sometimes being computationally impossible to solve with existing State of the Art (SoA) algorithms, making the use of UAVs an operational hazard and therefore significantly reducing their commercial applicability in urban environments. In this work, a conceptual framework for both stand-alone and swarm (networked) UAVs is introduced, with a focus on the architectural requirements of the collision avoidance subsystem to achieve acceptable levels of safety and reliability. The SoA principles for collision avoidance against stationary objects are reviewed and a novel approach is described, using deep learning techniques to solve the computational intensive problem of real-time collision avoidance with dynamic objects. The proposed framework includes a web-interface allowing the full control of UAVs as remote clients with a supervisor cloud-based platform. The feasibility of the proposed approach was demonstrated through experimental tests using a UAV, developed from scratch using the proposed framework. Test flight results are presented for an autonomous UAV monitored from multiple countries across the world. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57219671352;55927784300;","Forecasting spatio-temporal dynamics on the land surface using earth observation data—a review",2020,"10.3390/rs12213513","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85094636124&doi=10.3390%2frs12213513&partnerID=40&md5=e22d19be8815166c62ec3896abf35ab3","Reliable forecasts on the impacts of global change on the land surface are vital to inform the actions of policy and decision makers to mitigate consequences and secure livelihoods. Geospatial Earth Observation (EO) data from remote sensing satellites has been collected continuously for 40 years and has the potential to facilitate the spatio-temporal forecasting of land surface dynamics. In this review we compiled 143 papers on EO-based forecasting of all aspects of the land surface published in 16 high-ranking remote sensing journals within the past decade. We analyzed the literature regarding research focus, the spatial scope of the study, the forecasting method applied, as well as the temporal and technical properties of the input data. We categorized the identified forecasting methods according to their temporal forecasting mechanism and the type of input data. Time-lagged regressions which are predominantly used for crop yield forecasting and approaches based on Markov Chains for future land use and land cover simulation are the most established methods. The use of external climate projections allows the forecasting of numerical land surface parameters up to one hundred years into the future, while auto-regressive time series modeling can account for intra-annual variances. Machine learning methods have been increasingly used in all categories and multivariate modeling that integrates multiple data sources appears to be more popular than univariate auto-regressive modeling despite the availability of continuously expanding time series data. Regardless of the method, reliable EO-based forecasting requires high-level remote sensing data products and the resulting computational demand appears to be the main reason that most forecasts are conducted only on a local scale. In the upcoming years, however, we expect this to change with further advances in the field of machine learning, the publication of new global datasets, and the further establishment of cloud computing for data processing. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "56237086200;55628589750;56531367400;56735478500;57203925011;","A new satellite-based retrieval of low-cloud liquid-water path using machine learning and meteosat seviri data",2020,"10.3390/rs12213475","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093977443&doi=10.3390%2frs12213475&partnerID=40&md5=aee607a40b64b9aa3936a32d688c9574","Clouds are one of the major uncertainties of the climate system. The study of cloud processes requires information on cloud physical properties, in particular liquid water path (LWP). This parameter is commonly retrieved from satellite data using look-up table approaches. However, existing LWP retrievals come with uncertainties related to assumptions inherent in physical retrievals. Here, we present a new retrieval technique for cloud LWP based on a statistical machine learning model. The approach utilizes spectral information from geostationary satellite channels of Meteosat Spinning-Enhanced Visible and Infrared Imager (SEVIRI), as well as satellite viewing geometry. As ground truth, data from CloudNet stations were used to train the model. We found that LWP predicted by the machine-learning model agrees substantially better with CloudNet observations than a current physics-based product, the Climate Monitoring Satellite Application Facility (CM SAF) CLoud property dAtAset using SEVIRI, edition 2 (CLAAS-2), highlighting the potential of such approaches for future retrieval developments. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57219573318;54782254200;57211326042;57219570795;6603095001;","High-resolution soybean yield mapping across the us midwest using subfield harvester data",2020,"10.3390/rs12213471","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093972903&doi=10.3390%2frs12213471&partnerID=40&md5=d231007c25bc9aeebf57af335fbed26f","Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20-and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r2 = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r2 = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r2 = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r2 = 0.27 at the 30 m scale) and provide a new, parsimonious model. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57207771158;57218911059;57218911847;57218909762;35751750800;56108483100;57218910175;","Platform design for lifelog-based smart lighting control",2020,"10.1016/j.buildenv.2020.107267","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090725653&doi=10.1016%2fj.buildenv.2020.107267&partnerID=40&md5=afe07a24c9a15793c7fd063b1f6b6506","Lifelog refers to daily life records of individuals and includes their environmental, activity, emotional, and biometric information. The utilization of lifelogs makes it possible to provide a personalized lighting environment in line with a user's characteristics. However, no such customized lighting environment has been proposed thus far because methods for collecting and classifying the necessary data and a platform for synthesizing the data are not available. Therefore, in this study, the concept of the lifelog-based smart lighting control is introduced, and lifelog collection methods and a method to recommend the appropriate lighting environment for an individual by analyzing the collected information are proposed. Sensors, lighting controllers, and control interfaces required for lifelog-based smart lighting control were installed in a mock-up space, a machine learning server was set up, and a platform for implementing the optimal lighting environment was constructed by connecting these devices to the cloud. The platform constructed in this study creates a personalized lighting environment by utilizing lifelogs that are divided into emotional information collected via instant message (IM) text analysis, activity information collected using a location and activity tracker, and environmental information collected from weather data. © 2020 Elsevier Ltd" "57210706438;55893287300;57216237989;7004287554;","Vocalisation of the rare and flagship species Pharomachrus mocinno (Aves: Trogonidae): implications for its taxonomy, evolution and conservation",2020,"10.1080/09524622.2019.1647877","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071264664&doi=10.1080%2f09524622.2019.1647877&partnerID=40&md5=21995e7597855aa6c544e3b8fd6b8230","The Resplendent Quetzal Pharomachrus mocinno is a rare Neotropical bird included in the IUCN red list as Near Threatened. Fragmentation of its habitat, the cloud forest, is considered as the principal threat. Two subspecies are currently recognised but genetic and morphometric studies suggested they could be considered as full species. We assessed whether male vocalisation would support a species delimitation hypothesis. We recorded in the field and downloaded from sound archives vocalisation of 57 individuals from 30 different localities distributed in 11 countries. We estimated the acoustic differences of all the Pharomachrus taxa with multivariate analyses and machine learning techniques. Our results show vocal differences between P. m. mocinno and P. m. costaricensis that could have a molecular basis, potentially due to genetic drift developed during the more than three million years of separation of P. m. mocinno (from Mexico to Nicaragua) and P. m. costaricensis (Costa Rica and Panama). We therefore suggest that P. mocinno could potentially be divided into two species. A possible separation of these taxa into two species could have important consequences for the conservation status of the Resplendent Quetzals, and redirect conservation efforts for these taxa. © 2019 Informa UK Limited, trading as Taylor & Francis Group." "36718754900;7501959001;53663285500;7404403569;57218555441;57218546431;8939135600;57218546374;","Machine learning approaches for rice crop yield predictions using time-series satellite data in Taiwan",2020,"10.1080/01431161.2020.1766148","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089490090&doi=10.1080%2f01431161.2020.1766148&partnerID=40&md5=1725c43d04cf2d1b2f285b2016fe9cd1","Rice is the most important food crop in Taiwan, directly feeding more than 23 million people in the country. Information on rice production is thus crucial for crop management and food policymaking. This study aims to develop a machine learning (ML) approach for predicting rice crop yields in Taiwan using time-series Moderate Resolution Imaging Spectroradiometer (MODIS) data. We processed the data for the period from 2000 to 2018, following three main steps: (1) data pre-processing to generate smooth time-series Normalized Difference Vegetation Index (NDVI) data, (2) establishment of models for yield predictions using the heading date (HD) NDVI value, and the accumulated NDVI value of the dates from heading to maturity (DHM). The data from 2000 to 2017 were used for building predictive models using the random forests (RF) and support vector machines (SVM), leaving the 2018 data for model assessment, and (3) evaluation of model performance. The results compared with the government’s yield statistics indicated good predictions, with the root mean square error (RMSE) and mean absolute error (MAE) values between 7.1% and 11.8%, and Willmott’s index of agreement (d) values between 0.81 and 0.84 for the first crop, and 5.6% and 11.3% and d values between 0.91 and 0.95 for the second crop, respectively. A slight underestimation of yield predictions was observed for both crops, with the relative error (RE) values of −6.5% to −8.2% and −3.8% to −6% for the first and second crops, respectively. The results of regression analysis also confirmed a close agreement between these two datasets, with the correlation coefficient (r) higher than 0.84 (p-value <0.05), in both cases. Although some factors, including mixed-pixel issues, boundary effects, and cloud cover potentially affected the modelling results, our study demonstrated the effectiveness of ML methods for regional rice yield predictions from MODIS NDVI data in Taiwan. © 2020 Informa UK Limited, trading as Taylor & Francis Group." "36657489600;7101984634;7404548584;7006423243;","Leveraging spatial textures, through machine learning, to identify aerosols and distinct cloud types from multispectral observations",2020,"10.5194/amt-13-5459-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093500613&doi=10.5194%2famt-13-5459-2020&partnerID=40&md5=4d56ed1362f65276c7a11bb1f418ea57","Current cloud and aerosol identification methods for multispectral radiometers, such as the Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS), employ multichannel spectral tests on individual pixels (i.e., fields of view). The use of the spatial information in cloud and aerosol algorithms has been primarily through statistical parameters such as nonuniformity tests of surrounding pixels with cloud classification provided by the multispectral microphysical retrievals such as phase and cloud top height. With these methodologies there is uncertainty in identifying optically thick aerosols, since aerosols and clouds have similar spectral properties in coarse-spectral-resolution measurements. Furthermore, identifying clouds regimes (e.g., stratiform, cumuliform) from just spectral measurements is difficult, since low-altitude cloud regimes have similar spectral properties. Recent advances in computer vision using deep neural networks provide a new opportunity to better leverage the coherent spatial information in multispectral imagery. Using a combination of machine learning techniques combined with a new methodology to create the necessary training data, we demonstrate improvements in the discrimination between cloud and severe aerosols and an expanded capability to classify cloud types. The labeled training dataset was created from an adapted NASA Worldview platform that provides an efficient user interface to assemble a human-labeled database of cloud and aerosol types. The convolutional neural network (CNN) labeling accuracy of aerosols and cloud types was quantified using independent Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) and MODIS cloud and aerosol products. By harnessing CNNs with a unique labeled dataset, we demonstrate the improvement of the identification of aerosols and distinct cloud types from MODIS and VIIRS images compared to a per-pixel spectral and standard deviation thresholding method. The paper concludes with case studies that compare the CNN methodology results with the MODIS cloud and aerosol products. © Author(s) 2020. This work is distributed under the Creative Commons Attribution 4.0 License." "55479265100;","A study on the transformation of accounting based on new technologies: Evidence from korea",2020,"10.3390/su12208669","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092911020&doi=10.3390%2fsu12208669&partnerID=40&md5=7d5754fe6afc2514064f0e3e8d6f75ec","This study identifies the new accounting technologies into Cloud, Artificial Intelligence, Big Data, and Blockchain, and introduces the case of Korean companies applying new technologies to their accounting process. The purpose of this study is to help understand accounting technologies and provide examples of the adoption of these technologies in actual practice. To achieve this aim of the study, a systematic review of the literature of the major academic publications and professional reports and websites was used as a research methodology. In order to select the cases, it performed the analytical process of reviewing Korean major business and economic newspaper articles. This study provides evidence from Korea to companies contemplating the transformation of their accounting process using technology. Such companies can consider the cases presented in this study as a benchmark. It also offers guidance for the application of technologies to accounting practices for businesses and related researchers. The technology transformation is expected to be accelerated, especially after COVID-19. Therefore, it is necessary to understand and explore ways to effectively apply them. Further, while new technologies offer many opportunities, associated risks and threats should be addressed. © 2020 by the author. Licensee MDPI, Basel, Switzerland." "57203415030;57193802017;7003272359;36544749400;7006307181;6505462735;6507765711;","National scale land cover classification for ecosystem services mapping and assessment, using multitemporal copernicus EO data and google earth engine",2020,"10.3390/rs12203303","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092889651&doi=10.3390%2frs12203303&partnerID=40&md5=5863f528b289eed99bbd6826bb99b12d","Land-Use/Land-Cover (LULC) products are a common source of information and a key input for spatially explicit models of ecosystem service (ES) supply and demand. Global, continental, and regional, readily available, and free land-cover products generated through Earth Observation (EO) data, can be potentially used as relevant to ES mapping and assessment processes from regional to national scales. However, several limitations exist in these products, highlighting the need for timely land-cover extraction on demand, that could replace or complement existing products. This study focuses on the development of a classification workflow for fine-scale, object-based land cover mapping, employed on terrestrial ES mapping, within the Greek terrestrial territory. The processing was implemented in the Google Earth Engine cloud computing environment using 10 m spatial resolution Sentinel-1 and Sentinel-2 data. Furthermore, the relevance of different training data extraction strategies and temporal EO information for increasing the classification accuracy was also evaluated. The different classification schemes demonstrated differences in overall accuracy ranging from 0.88% to 4.94% with the most accurate classification scheme being the manual sampling/monthly feature classification achieving a 79.55% overall accuracy. The classification results suggest that existing LULC data must be cautiously considered for automated extraction of training samples, in the case of new supervised land cover classifications aiming also to discern complex vegetation classes. The code used in this study is available on GitHub and runs on the Google Earth Engine web platform. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57219452272;37088138800;36816017700;7005865248;","Lowland rice mapping in Sédhiou region (Senegal) using sentinel 1 and sentinel 2 data and random forest",2020,"10.3390/rs12203403","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092765171&doi=10.3390%2frs12203403&partnerID=40&md5=45f2cc442d698ab383bb50dd55796c7a","In developing countries, information on the area and spatial distribution of paddy rice fields is an essential requirement for ensuring food security and facilitating targeted actions of both technical assistance and restoration of degraded production areas. In this study, Sentinel 1 (S1) and Sentinel 2 (S2) imagery was used to map lowland rice crop areas in the Sédhiou region (Senegal) for the 2017, 2018, and 2019 growing seasons using the Random Forest (RF) algorithm. Ground sample datasets were annually collected (416, 455, and 400 samples) for training and testing yearly RF classification. A procedure was preliminarily applied to process S2 scenes and yield a normalized difference vegetation index (NDVI) time series less affected by clouds. A total of 93 predictors were calculated from S2 NDVI time series and S1 vertical transmit–horizontal receive (VH) and vertical transmit–vertical receive (VV) backscatters. Guided regularized random forest (GRRF) was used to deal with the arising multicollinearity and identify the most important predictors. The RF classifier was then applied to the selected predictors. The algorithm predicted the five land cover types present in the test areas, with a maximum accuracy of 87% and kappa coefficient of 0.8 in 2019. The broad land cover maps identified around 12,500 (2017), 13,800 (2018), and 12,800 (2019) ha of lowland rice crops. The study highlighted a partial difficulty of the classifier to distinguish rice from natural herbaceous vegetation (NHV) due to similar temporal patterns and high intra-class variability. Moreover, the results of this investigation indicated that S2-derived predictors provided more valuable information compared to VV and VH backscatter-derived predictors, but a combination of radar and optical imagery always outperformed a classification based on single-sensor inputs. An example is finally provided that illustrates how the maps obtained can be combined with ground observations through a ratio estimator in order to yield a statistically sound prediction of rice area all over the study region. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57219293912;7405490236;57219054933;","Evaluation and enhancement of unmanned aircraft system photogrammetric data quality for coastal wetlands",2020,"10.1080/15481603.2020.1819720","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092107074&doi=10.1080%2f15481603.2020.1819720&partnerID=40&md5=7f1642edf786da2c4c621aad90075446","Understanding the impacts of flight configuration and post-mission data processing techniques on unmanned aircraft system (UAS) photogrammetric data quality is essential for employing this popular technique in coastal wetland ecosystems. In this study, we systematically evaluated the effects of flight configuration (flying altitude, image overlap, and lighting conditions) on UAS photogrammetric level 1 products: orthoimagery and point clouds, and level 2 products: digital terrain models (DTM) and canopy height models (CHM). We also developed an object-based machine learning approach to correct UAS DTMs to mitigate data uncertainties caused by flight configuration and dense vegetation. Flying altitude was identified as the leading parameter in the quality of level 1 products, while image overlap was the most influential determinant for the quality of level 2 products. The correction approach effectively reduced the vertical error of DTMs for two study sites. This study informs UAS photogrammetric survey design and data enhancement for applications in coastal wetlands. © 2020 Informa UK Limited, trading as Taylor & Francis Group." "57191287722;22433611700;57198063860;","Applications of remote sensing in precision agriculture: A review",2020,"10.3390/rs12193136","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093863787&doi=10.3390%2frs12193136&partnerID=40&md5=cc2c416926af0e18f7ecb09cb46b208a","Agriculture provides for the most basic needs of humankind: food and fiber. The introduction of new farming techniques in the past century (e.g., during the Green Revolution) has helped agriculture keep pace with growing demands for food and other agricultural products. However, further increases in food demand, a growing population, and rising income levels are likely to put additional strain on natural resources. With growing recognition of the negative impacts of agriculture on the environment, new techniques and approaches should be able to meet future food demands while maintaining or reducing the environmental footprint of agriculture. Emerging technologies, such as geospatial technologies, Internet of Things (IoT), Big Data analysis, and artificial intelligence (AI), could be utilized to make informed management decisions aimed to increase crop production. Precision agriculture (PA) entails the application of a suite of such technologies to optimize agricultural inputs to increase agricultural production and reduce input losses. Use of remote sensing technologies for PA has increased rapidly during the past few decades. The unprecedented availability of high resolution (spatial, spectral and temporal) satellite images has promoted the use of remote sensing in many PA applications, including crop monitoring, irrigation management, nutrient application, disease and pest management, and yield prediction. In this paper, we provide an overview of remote sensing systems, techniques, and vegetation indices along with their recent (2015–2020) applications in PA. Remote-sensing-based PA technologies such as variable fertilizer rate application technology in Green Seeker and Crop Circle have already been incorporated in commercial agriculture. Use of unmanned aerial vehicles (UAVs) has increased tremendously during the last decade due to their cost-effectiveness and flexibility in obtaining the high-resolution (cm-scale) images needed for PA applications. At the same time, the availability of a large amount of satellite data has prompted researchers to explore advanced data storage and processing techniques such as cloud computing and machine learning. Given the complexity of image processing and the amount of technical knowledge and expertise needed, it is critical to explore and develop a simple yet reliable workflow for the real-time application of remote sensing in PA. Development of accurate yet easy to use, user-friendly systems is likely to result in broader adoption of remote sensing technologies in commercial and non-commercial PA applications. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57202286610;36697084100;35762297500;57212672124;57219552485;","Retrieving the national main commodity maps in indonesia based on high-resolution remotely sensed data using cloud computing platform",2020,"10.3390/land9100377","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093847415&doi=10.3390%2fland9100377&partnerID=40&md5=a764bd528c14fc15187f9dfa5ab0da81","Indonesia has the most favorable climates for agriculture because of its location in the tropical climatic zones. The country has several commodities to support economics growth that are driven by key export commodities—e.g., oil palm, rubber, paddy, cacao, and coffee. Thus, identifying the main commodities in Indonesia using spatially-explicit tools is essential to understand the precise productivity derived from the agricultural sectors. Many previous studies have used predictions developed using binary maps of general crop cover. Here, we present national commodity maps for Indonesia based on remote sensing data using Google Earth Engine. We evaluated a machine learning algorithm—i.e., Random Forest to parameterize how the area in commodity varied in Indonesia. We used various predictors to estimate the productivity of various commodities based on multispectral satellite imageries (36 predictors) at 30-meters spatial resolution. The national commodity map has a relatively high accuracy, with an overall accuracy of about 95% and Kappa coefficient of about 0.90. The results suggest that the oil palm plantation was the highest commodity product that occupied the largest land of Indonesia. However, this study also showed that the land area in rubber, rice paddies, and cacao commodities was underestimated due to its lack of training samples. Improvement in training data collection for each commodity should be done to increase the accuracy of the commodity maps. The commodity data can be viewed online (website can be found in the end of conclusions). This data can further provide significant information related to the agricultural sectors to investigate food provisioning, particularly in Indonesia. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57184370500;57204759617;7201908774;8434627900;","Auroral Image Classification With Deep Neural Networks",2020,"10.1029/2020JA027808","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093817691&doi=10.1029%2f2020JA027808&partnerID=40&md5=98f502eafb7bb33fb8f6ae59ebb9f065","Results from a study of automatic aurora classification using machine learning techniques are presented. The aurora is the manifestation of physical phenomena in the ionosphere-magnetosphere environment. Automatic classification of millions of auroral images from the Arctic and Antarctic is therefore an attractive tool for developing auroral statistics and for supporting scientists to study auroral images in an objective, organized, and repeatable manner. Although previous studies have presented tools for detecting aurora, there has been a lack of tools for classifying aurora into subclasses with a high precision (>90%). This work considers seven auroral subclasses: breakup, colored, arcs, discrete, patchy, edge, and faint. Six different deep neural network architectures have been tested along with the well-known classification algorithms: k-nearest neighbor (KNN) and a support vector machine (SVM). A set of clean nighttime color auroral images, without clearly ambiguous auroral forms, moonlight, twilight, clouds, and so forth, were used for training and testing the classifiers. The deep neural networks generally outperformed the KNN and SVM methods, and the ResNet-50 architecture achieved the highest performance with an average classification precision of 92%. © 2020. The Authors." "57200570867;6602610108;","Development of a machine learning-based radiometric bias correction for noaa’s microwave integrated retrieval system (Mirs)",2020,"10.3390/rs12193160","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092904232&doi=10.3390%2frs12193160&partnerID=40&md5=0f38227dfb09f6353432189f9baaefab","We present the development of a dynamic over-ocean radiometric bias correction for the Microwave Integrated Retrieval System (MiRS) which accounts for spatial, temporal, spectral, and angular dependence of the systematic differences between observed and forward model-simulated radiances. The dynamic bias correction, which utilizes a deep neural network approach, is designed to incorporate dependence on the atmospheric and surface conditions that impact forward model biases. The approach utilizes collocations of observed Suomi National Polar-orbiting Partnership/Advanced Technology Microwave Sounder (SNPP/ATMS) radiances and European Centre for Medium-Range Weather Forecasts (ECMWF) model analyses which are used as input to the Community Radiative Transfer Model (CRTM) forward model to develop training data of radiometric biases. Analysis of the neural network performance indicates that in many channels, the dynamic bias is able to reproduce realistically both the spatial patterns of the original bias and its probability distribution function. Furthermore, retrieval impact experiments on independent data show that, compared with the baseline static bias correction, using the dynamic bias correction can improve temperature and water vapor profile retrievals, particularly in regions with higher Cloud Liquid Water (CLW) amounts. Ocean surface emissivity retrievals are also improved, for example at 23.8 GHz, showing an increase in correlation from 0.59 to 0.67 and a reduction of standard deviation from 0.035 to 0.026. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "6602533891;57219363922;6506175855;","Monitoring forest change in the amazon using multi-temporal remote sensing data and machine learning classification on Google Earth Engine",2020,"10.3390/ijgi9100580","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092420606&doi=10.3390%2fijgi9100580&partnerID=40&md5=2c791a7ef23271ffc999badad2441d34","Deforestation causes diverse and profound consequences for the environment and species. Direct or indirect effects can be related to climate change, biodiversity loss, soil erosion, floods, landslides, etc. As such a significant process, timely and continuous monitoring of forest dynamics is important, to constantly follow existing policies and develop new mitigation measures. The present work had the aim of mapping and monitoring the forest change from 2000 to 2019 and of simulating the future forest development of a rainforest region located in the Pará state, Brazil. The land cover dynamics were mapped at five-year intervals based on a supervised classification model deployed on the cloud processing platform Google Earth Engine. Besides the benefits of reduced computational time, the service is coupled with a vast data catalogue providing useful access to global products, such as multispectral images of the missions Landsat five, seven, eight and Sentinel-2. The validation procedures were done through photointerpretation of highresolution panchromatic images obtained from CBERS (China-Brazil Earth Resources Satellite). The more than satisfactory results allowed an estimation of peak deforestation rates for the period 2000- 2006; for the period 2006-2015, a significant decrease and stabilization, followed by a slight increase till 2019. Based on the derived trends a forest dynamics was simulated for the period 2019-2028, estimating a decrease in the deforestation rate. These results demonstrate that such a fusion of satellite observations, machine learning, and cloud processing, benefits the analysis of the forest dynamics and can provide useful information for the development of forest policies. © 2020 by the authors." "57197738532;26649033300;57192933085;23014409400;","Anomaly detection based on machine learning in IoT-based vertical plant wall for indoor climate control",2020,"10.1016/j.buildenv.2020.107212","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089804800&doi=10.1016%2fj.buildenv.2020.107212&partnerID=40&md5=6a773d9322021c8c563e894d00da1d10","Indoor climate is closely related to human health, comfort and productivity. Vertical plant wall systems, embedded with sensors and actuators, have become a promising application for indoor climate control. In this study, we explore the possibility of applying machine learning based anomaly detection methods to vertical plant wall systems so as to enhance the automation and improve the intelligence to realize predictive maintenance of the indoor climate. Two categories of anomalies, namely point anomalies and contextual anomalies are researched. Prediction-based and pattern recognition-based methods are investigated and applied to indoor climate anomaly detection. The results show that neural network-based models, specifically the autoencoder (AE) and the long short-term memory encoder decoder (LSTM-ED) model surpass the others in terms of detecting point anomalies and contextual anomalies, respectively, therefore can be deployed into vertical plant walls systems in industrial practice. Based on the results, a new data cleaning method is proposed and a prediction-based method is deployed to the cloud in practice as a proof-of-concept. This study showcases the advancements in machine learning and Internet of things can be fully utilized by researches on building environment to accelerate the solution development. © 2020 The Authors" "15840467900;57206259084;57218294888;57218293495;57218295171;57218290463;","Evaluation and comparison of a machine learning cloud identification algorithm for the SLSTR in polar regions",2020,"10.1016/j.rse.2020.111999","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088642244&doi=10.1016%2fj.rse.2020.111999&partnerID=40&md5=7fefd97fa53d3bd6c858104b9f535b8a","A Feed Forward Neural Net (NN) approach to distinguish between clouds and the surface has been applied to the Sea and Land Surface Temperature Radiometer in polar regions. The masking algorithm covers the Arctic, Antarctic and regions typically classified as the cryosphere such as northern hemisphere permafrost. The mask has been trained using collocations with the CALIOP active lidar, which in narrow strips provide more accurate detection of cloud, and was subsequently evaluated as a function of cloud type and surface type. The mask was compared with the existing operational Bayesian and Empirical cloud masks by eye and also statistically using CALIOP data. It was found to perform exceptionally well in the polar regions. The Kuiper skill score improved from 0.28, for the operational Bayesian and 0.17 for the Empirical masks to 0.77 for the NN. The NN algorithm also has a much more homogeneous performance over all surface types. The key improvement came from better identification of clear scenes; for the NN mask, the same performance in terms of contamination of cloudy pixels in the sample of identified clear pixels can be achieved while retaining 40% of the clear pixels compared with 10% for the operational cloud identification. The algorithm performed with almost the same skill over sea and land. The best performance was achieved for opaque clouds while transparent and broken clouds showed slightly reduced accuracy. © 2020 Elsevier Inc." "57219097296;55332214100;56893485300;6701640417;","GLM and ABI Characteristics of Severe and Convective Storms",2020,"10.1029/2020JD032858","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091204819&doi=10.1029%2f2020JD032858&partnerID=40&md5=27753df453ecfa35998ba2a1eb36953a","The recent deployment of the Geostationary Lightning Mapper (GLM) on board GOES-16 and GOES-17 provides a new perspective of total lightning production for the severe convective storms research and operational communities. While the GLM has met its performance targets, further understanding flash characteristics and the physical limitations of the GLM are required to increase the applicability of the data. Derived cloud-top height and infrared (IR) brightness temperature products from the Advanced Baseline Imager (ABI) are used to assess data quality and characteristics from gridded GLM imagery across 7 weeks of active severe weather: 13 April through 31 May 2019. Areas with cloud tops colder than 240 K typically produced lightning, though this becomes less certain near the edge of the field of view due to algorithm limitations. Increasing flash rates were observed to correlate with decreasing flash areas, increasing cloud-top heights, and colder cloud-top temperatures. However, flash rates and size were more strongly tied to convective intensity and proximity to convective hazards at the surface due to the ability to delineate between convective and stratiform precipitation. Results show that merging ABI and GLM data sets could add value to both machine learning and statistical-based algorithms and also forecast applications with each providing unique details, although parameters such as GOES-16 viewing angle should be considered. Lastly, two case studies (24 and 27 May 2019) are used to help interpret the results from the 7-week sampling period and identify GLM and ABI trends related to thunderstorm evolution. ©2020. American Geophysical Union. All Rights Reserved." "57200857362;7004563395;55962154500;56583139400;8284949000;56504563100;57094306300;","DSCOVR/EPIC-derived global hourly and daily downward shortwave and photosynthetically active radiation data at 0.1° × 0.1° resolution",2020,"10.5194/essd-12-2209-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093534894&doi=10.5194%2fessd-12-2209-2020&partnerID=40&md5=993605e47eededdce75d699fa19e18ea","Downward shortwave radiation (SW) and photosynthetically active radiation (PAR) play crucial roles in Earth system dynamics. Spaceborne remote sensing techniques provide a unique means for mapping accurate spatiotemporally continuous SW-PAR, globally. However, any individual polar-orbiting or geostationary satellite cannot satisfy the desired high temporal resolution (sub-daily) and global coverage simultaneously, while integrating and fusing multisource data from complementary satellites/sensors is challenging because of coregistration, intercalibration, near real-time data delivery and the effects of discrepancies in orbital geometry. The Earth Polychromatic Imaging Camera (EPIC) on board the Deep Space Climate Observatory (DSCOVR), launched in February 2015, offers an unprecedented possibility to bridge the gap between high temporal resolution and global coverage and characterize the diurnal cycles of SW-PAR globally. In this study, we adopted a suite of well-validated data-driven machine-learning models to generate the first global land products of SW-PAR, from June 2015 to June 2019, based on DSCOVR/EPIC data. The derived products have high temporal resolution (hourly) and medium spatial resolution (0.1° × 0.1°), and they include estimates of the direct and diffuse components of SW-PAR. We used independently widely distributed ground station data from the Baseline Surface Radiation Network (BSRN), the Surface Radiation Budget Network (SURFRAD), NOAA's Global Monitoring Division and the U.S. Department of Energy's Atmospheric System Research (ASR) program to evaluate the performance of our products, and we further analyzed and compared the spatiotemporal characteristics of the derived products with the benchmarking Clouds and the Earth's Radiant Energy System Synoptic (CERES) data. We found both the hourly and daily products to be consistent with ground-based observations (e.g., hourly and daily total SWs have low biases of-3.96 and-0.71 W m-2 and root-mean-square errors (RMSEs) of 103.50 and 35.40 W m-2, respectively). The developed products capture the complex spatiotemporal patterns well and accurately track substantial diurnal, monthly, and seasonal variations in SW-PAR when compared to CERES data. They provide a reliable and valuable alternative for solar photovoltaic applications worldwide and can be used to improve our understanding of the diurnal and seasonal variabilities of the terrestrial water, carbon and energy fluxes at various spatial scales. The products are freely available at https.//doi.org/10.25584/1595069 (Hao et al., 2020). © 2020 Author(s)." "57213686348;16642370700;23394040900;","Automatic extraction of high-voltage bundle subconductors using airborne LiDAR data",2020,"10.3390/RS12183078","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092271541&doi=10.3390%2fRS12183078&partnerID=40&md5=505baf5474bd509de08223e8b38085c3","Overhead high-voltage conductors are the chief components of power lines and their safety has a strong influence on social and daily life. In the recent decade, the airborne laser scanning (ALS) technique has been widely used to capture the three-dimensional (3D) information of power lines and surrounding objects. Most of the existing methods focused on extraction of single conductors or extracted all conductors as one object class by applying machine learning techniques. Nevertheless, power line corridors (PLCs) are built with multi-loop, multi-phase structures (bundle conductors) and exist in intricate environments (e.g., mountains and forests), and thus raise challenges to process ALS data for extraction of individual conductors. This paper proposes an automated method to extract individual subconductors in bundles from complex structure of PLCs using a combined image- and point-based approach. First, the input point cloud data are grouped into 3D voxel grid and PL points and separated from pylon and tree points using the fact that pylons and trees are vertical objects while power lines are non-vertical objects. These pylons are further separated from trees by employing a statistical analysis technique and used to extract span points between two consecutive pylons; then, by using the distribution properties of power lines in each individual span, the bundles located at different height levels are extracted using image-based processing; finally, subconductors in each bundle are detected and extracted by introducing a window that slides over the individual bundle. The orthogonal plane transformation and recursive clustering procedures are exploited in each window position and a point-based processing is conducted iteratively for extraction of complete individual subconductors in each bundle. The feasibility and validity of the proposed method are verified on two Australian sites having bundle conductors in high-voltage transmission lines. Our experiments show that the proposed method achieves a reliable result by extracting the real structure of bundle conductors in power lines with correctness of 100% and 90% in the two test sites, respectively. © 2020 by the authors." "57205522425;57212792894;57219311534;57219314338;57219312305;57189580728;6603095001;","Mapping crop types in southeast india with smartphone crowdsourcing and deep learning",2020,"10.3390/RS12182957","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092200075&doi=10.3390%2fRS12182957&partnerID=40&md5=a7146aa44ed68ae6098bf875aa811231","High resolution satellite imagery and modern machine learning methods hold the potential to fill existing data gaps in where crops are grown around the world at a sub-field level. However, high resolution crop type maps have remained challenging to create in developing regions due to a lack of ground truth labels for model development. In this work, we explore the use of crowdsourced data, Sentinel-2 and DigitalGlobe imagery, and convolutional neural networks (CNNs) for crop type mapping in India. Plantix, a free app that uses image recognition to help farmers diagnose crop diseases, logged 9 million geolocated photos from 2017-2019 in India, 2 million of which are in the states of Andhra Pradesh and Telangana in India. Crop type labels based on farmer-submitted images were added by domain experts and deep CNNs. The resulting dataset of crop type at coordinates is high in volume, but also high in noise due to location inaccuracies, submissions from out-of-field, and labeling errors. We employed a number of steps to clean the dataset, which included training a CNN on very high resolution DigitalGlobe imagery to filter for points that are within a crop field. With this cleaned dataset, we extracted Sentinel time series at each point and trained another CNN to predict the crop type at each pixel. When evaluated on the highest quality subset of crowdsourced data, the CNN distinguishes rice, cotton, and ""other"" crops with 74% accuracy in a 3-way classification and outperforms a random forest trained on harmonic regression features. Furthermore, model performance remains stable when low quality points are introduced into the training set. Our results illustrate the potential of non-traditional, high-volume/high-noise datasets for crop type mapping, some improvements that neural networks can achieve over random forests, and the robustness of such methods against moderate levels of training set noise. Lastly, we caution that obstacles like the lack of good Sentinel-2 cloud mask, imperfect mobile device location accuracy, and preservation of privacy while improving data access will need to be addressed before crowdsourcing can widely and reliably be used to map crops in smallholder systems. © 2020 by the authors." "57201284843;57203927097;47461334700;57186194700;57203928635;6506182755;57211096480;16551253800;57211096034;57211351934;55949126400;","Integrated geological and geophysical mapping of a carbonatite-hosting outcrop in siilinjärvi, finland, using unmanned aerial systems",2020,"10.3390/RS12182998","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092116770&doi=10.3390%2fRS12182998&partnerID=40&md5=53718447b7bf877f11556978c04f208e","Mapping geological outcrops is a crucial part of mineral exploration, mine planning and ore extraction. With the advent of unmanned aerial systems (UASs) for rapid spatial and spectral mapping, opportunities arise in fields where traditional ground-based approaches are established and trusted, but fail to cover sufficient area or compromise personal safety. Multi-sensor UAS are a technology that change geoscientific research, but they are still not routinely used for geological mapping in exploration and mining due to lack of trust in their added value and missing expertise and guidance in the selection and combination of drones and sensors. To address these limitations and highlight the potential of using UAS in exploration settings, we present an UAS multi-sensor mapping approach based on the integration of drone-borne photography, multi- and hyperspectral imaging and magnetics. Data are processed with conventional methods as well as innovative machine learning algorithms and validated by geological field mapping, yielding a comprehensive and geologically interpretable product. As a case study, we chose the northern extension of the Siilinjärvi apatite mine in Finland, in a brownfield exploration setting with plenty of ground truth data available and a survey area that is partly covered by vegetation. We conducted rapid UAS surveys from which we created a multi-layered data set to investigate properties of the ore-bearing carbonatite-glimmerite body. Our resulting geologic map discriminates between the principal lithologic units and distinguishes ore-bearing from waste rocks. Structural orientations and lithological units are deduced based on high-resolution, hyperspectral image-enhanced point clouds. UAS-based magnetic data allow an insight into their subsurface geometry through modeling based on magnetic interpretation. We validate our results via ground survey including rock specimen sampling, geochemical and mineralogical analysis and spectroscopic point measurements. We are convinced that the presented non-invasive, data-driven mapping approach can complement traditional workflows in mineral exploration as a flexible tool. Mapping products based on UAS data increase efficiency and maximize safety of the resource extraction process, and reduce expenses and incidental wastes. © 2020 by the authors." "7401911971;57219293007;55971004500;26322732800;56011074300;57206895864;57203474131;","Mapping paddy rice fields by combining multi-temporal vegetation index and synthetic aperture radar remote sensing data using Google Earth Engine machine learning platform",2020,"10.3390/RS12182992","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092076309&doi=10.3390%2fRS12182992&partnerID=40&md5=b56a53bd4f900888e392dfd019d29174","The knowledge of the area and spatial distribution of paddy rice fields is important for water resource management. However, accurate map of paddy rice is a long-term challenge because of its spatiotemporal discontinuity and short duration. To solve this problem, this study proposed a paddy rice area extraction approach by using the combination of optical vegetation indices and synthetic aperture radar (SAR) data. This method is designed to overcome the data-missing problem due to cloud contamination and spatiotemporal discontinuities of the traditional optical remote sensing method. More specifically, the Sentinel-1A SAR and the Sentinel-2 multispectral imager (MSI) Level-2A imagery are used to identify paddy rice with a high temporal and spatial resolution. Three vegetation indices, namely normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and land surface water index (LSWI), are estimated from optical bands. Two polarization bands (VH (vertical-horizontal) and VV (vertical-vertical)) are used to overcome the cloud contamination problem. This approach was applied with the random forest machine learning algorithm on the Google Earth Engine platform for the Jianghan Plain in China as an experimental area. The results of 39 experiments uncovered the effect of different factors. The results indicated that the combination of VV and VH band showed a better performance compared with other polarization bands; the average producer's accuracy of paddy rice (PA) is 72.79%, 1.58% higher than the second one VH. Secondly, the combination of three indices also showed a better result than others, with average PA 73.82%, 1.42% higher than using NDVI alone. The classification result presented the best combination is EVI, VV, and VH polarization band. The producer's accuracy of paddy rice was 76.67%, with the overall accuracy (OA) of 66.07%, and Kappa statistics of 0.45. However, NDVI, EVI, and VH showed better performance in mapping the morphology. The results demonstrated the method developed in this study can be successfully applied to the cloud-prone area for mapping paddy rice to overcome the data missing caused by cloud and rain during the paddy growing season. © 2020 by the authors." "16318937600;57196872978;7004385955;35477147000;7006056290;","Atmospheric parameters of Cepheids from flux ratios with ATHOS: I. The temperature scale",2020,"10.1051/0004-6361/202038277","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091840760&doi=10.1051%2f0004-6361%2f202038277&partnerID=40&md5=cb6ca943073030015930a37b04025a03","Context. The effective temperature is a key parameter governing the properties of a star. For stellar chemistry, it has the strongest impact on the accuracy of the abundances derived. Since Cepheids are pulsating stars, determining their effective temperature is more complicated than in the case of nonvariable stars. Aims. We want to provide a new temperature scale for classical Cepheids, with a high precision and full control of the systematics. Methods. Using a data-driven machine learning technique employing observed spectra, and in taking great care to accurately phase single-epoch observations, we tied flux ratios to (label) temperatures derived using the infrared surface brightness method. Results. We identified 143 flux ratios, which allow us to determine the effective temperature with a precision of a few Kelvin and an accuracy better than 150 K, which is in line with the most accurate temperature measures available to date. The method does not require a normalization of the input spectra and provides homogeneous temperatures for low-And high-resolution spectra, even at the lowest signal-To-noise ratios. Due to the lack of a dataset with a sufficient sample size for Small Magellanic Cloud Cepheids, the temperature scale does not extend to Cepheids with [Fe/H] <-0.6 dex. However, it nevertheless provides an exquisite, homogeneous means of characterizing Galactic and Large Magellanic Cloud Cepheids. Conclusions. The temperature scale will be extremely useful in the context of spectroscopic surveys for Milky Way archaeology with the WEAVE and 4MOST spectrographs. It paves the way for highly accurate and precise metallicity estimates, which will allow us to assess the possible metallicity dependence of Cepheids' period-luminosity relations and, in turn, to improve our measurement of the Hubble constant H0. © ESO 2020." "57219197101;55738957800;8905764300;56119479900;","A Moist Physics Parameterization Based on Deep Learning",2020,"10.1029/2020MS002076","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091654322&doi=10.1029%2f2020MS002076&partnerID=40&md5=9f3000e915b558d172c344c9d6ec9af9","Current moist physics parameterization schemes in general circulation models (GCMs) are the main source of biases in simulated precipitation and atmospheric circulation. Recent advances in machine learning make it possible to explore data-driven approaches to developing parameterization for moist physics processes such as convection and clouds. This study aims to develop a new moist physics parameterization scheme based on deep learning. We use a residual convolutional neural network (ResNet) for this purpose. It is trained with 1-year simulation from a superparameterized GCM, SPCAM. An independent year of SPCAM simulation is used for evaluation. In the design of the neural network, referred to as ResCu, the moist static energy conservation during moist processes is considered. In addition, the past history of the atmospheric states, convection, and clouds is also considered. The predicted variables from the neural network are GCM grid-scale heating and drying rates by convection and clouds, and cloud liquid and ice water contents. Precipitation is derived from predicted moisture tendency. In the independent data test, ResCu can accurately reproduce the SPCAM simulation in both time mean and temporal variance. Comparison with other neural networks demonstrates the superior performance of ResNet architecture. ResCu is further tested in a single-column model for both continental midlatitude warm season convection and tropical monsoonal convection. In both cases, it simulates the timing and intensity of convective events well. In the prognostic test of tropical convection case, the simulated temperature and moisture biases with ResCu are smaller than those using conventional convection and cloud parameterizations. © 2020. The Authors." "57219177671;57219172532;57219019454;57219180968;57212948551;","Ultra-short-term prediction method of photovoltaic electric field power based on ground-based cloud image segmentation",2020,"10.1051/e3sconf/202018501052","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091590959&doi=10.1051%2fe3sconf%2f202018501052&partnerID=40&md5=7b95ffe5105196bd92f2cba45cc18a15","As a large number of photovoltaic power stations are built and put into operation, the total amount of photovoltaic power generation accounts for an increasing proportion of the total electricity. The inability to accurately predict solar energy output has brought great uncertainty to the grid. Therefore, predicting the future power of photovoltaic fields is of great significance. According to different time scales, predictions are divided into long-term, medium-term and ultra-short-term predictions. The main difficulty of ultra-short-term forecasting lies in the power fluctuations caused by sudden and drastic changes in environmental factors. The shading of clouds is directly related to the irradiance received on the surface of the photovoltaic panel, which has become the main factor affecting the fluctuation of photovoltaic power generation. Therefore, sky images captured by conventional cameras installed near solar panels can be used to analyze cloud characteristics and improve the accuracy of ultra-short-term predictions. This paper uses historical power information of photovoltaic power plants and cloud image data, combined with machine learning methods, to provide ultra-short-term predictions of the power generation of photovoltaic power plants. First, the random forest method is used to use historical power generation data to establish a single time series prediction model to predict ultra-short-term power generation. Compared with the continuous model, the root mean square (RMSE) error of prediction is reduced by 28.38%. Secondly, the Unet network is used to segment the cloud image, and the cloud amount information is analyzed and input into the random forest prediction model to obtain the bivariate prediction model. The experimental results prove that, based on the cloud amount information contained in the cloud chart, the bivariate prediction model has an 11.56% increase in prediction accuracy compared with the single time series prediction model, and an increase of 36.66% compared with the continuous model. © The Authors, published by EDP Sciences, 2020." "53264602200;57219054187;57203388944;25625412100;17345922900;","Different Fates of Young Star Clusters after Gas Expulsion",2020,"10.3847/2041-8213/abad28","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091169604&doi=10.3847%2f2041-8213%2fabad28&partnerID=40&md5=cccd0c26927454cc45c36d5028efec4c","We identify structures of the young star cluster NGC 2232 in the solar neighborhood (323.0 pc) and a newly discovered star cluster, LP 2439 (289.1 pc). Member candidates are identified using the Gaia DR2 sky position, parallax, and proper-motion data by an unsupervised machine-learning method, StarGO. Member contamination from the Galactic disk is further removed using the color-magnitude diagram. The four identified groups (NGC 2232, LP 2439, and two filamentary structures) of stars are coeval with an age of 25 Myr and were likely formed in the same giant molecular cloud. We correct the distance asymmetry from the parallax error with a Bayesian method. The 3D morphology shows the two spherical distributions of clusters NGC 2232 and LP 2439. Two filamentary structures are spatially and kinematically connected to NGC 2232. Both NGC 2232 and LP 2439 are expanding. The expansion is more significant in LP 2439, generating a loose spatial distribution with shallow volume number and mass density profiles. The expansion is suggested to be mainly driven by gas expulsion. With 73% of the cluster mass bound, NGC 2232 is currently experiencing a process of revirialization, However, LP 2439, with 52% of the cluster mass unbound, may fully dissolve in the near future. The different survivability traces the different dynamical states of NGC 2232 and LP 2439 prior to the onset of gas expulsion. While NGC 2232 may have been substructured and subvirial, LP 2439 may have either been virial/supervirial or experienced a much faster rate of gas removal. © 2020. The American Astronomical Society. All rights reserved." "35239293700;8630380500;57217779740;56402112700;55939843000;","Spatial mapping of short-term solar radiation prediction incorporating geostationary satellite images coupled with deep convolutional LSTM networks for South Korea",2020,"10.1088/1748-9326/ab9467","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090892326&doi=10.1088%2f1748-9326%2fab9467&partnerID=40&md5=4373093ddef43c1cc22dceb27eb1df54","A practical approach to continuously monitor and provide real-time solar energy prediction can help support reliable renewable energy supply and relevant energy security systems. In this study on the Korean Peninsula, contemporaneous solar radiation images obtained from the Communication, Ocean and Meteorological Satellite (COMS) Meteorological Imager (MI) system, were used to design a convolutional neural network and a long short-term memory network predictive model, ConvLSTM. This model was applied to predict one-hour ahead solar radiation and spatially map solar energy potential. The newly designed ConvLSTM model enabled reliable prediction of solar radiation, incorporating spatial changes in atmospheric conditions and capturing the temporal sequence-to-sequence variations that are likely to influence solar driven power supply and its overall stability. Results showed that the proposed ConvLSTM model successfully captured cloud-induced variations in ground level solar radiation when compared with reference images from a physical model. A comparison with ground pyranometer measurements indicated that the short-term prediction of global solar radiation by the proposed ConvLSTM had the highest accuracy [root mean square error (RMSE) = 83.458 W • m-2, mean bias error (MBE) = 4.466 W • m-2, coefficient of determination (R2) = 0.874] when compared with results of conventional artificial neural network (ANN) [RMSE = 94.085 W • m-2, MBE =-6.039 W • m-2, R2 = 0.821] and random forest (RF) [RMSE = 95.262 W • m-2, MBE =-11.576 W • m-2, R2 = 0.839] models. In addition, ConvLSTM better captured the temporal variations in predicted solar radiation, mainly due to cloud attenuation effects when compared with two selected ground stations. The study showed that contemporaneous satellite images over short-term or near real-time intervals can successfully support solar energy exploration in areas without continuous environmental monitoring systems, where satellite footprints are available to model and monitor solar energy management systems supporting real-life power grid systems. © 2020 The Author(s). Published by IOP Publishing Ltd." "57195330111;57194694881;57209009649;57188734162;56460209100;6507122674;","Comparing machine and deep learning methods for large 3D heritage semantic segmentation",2020,"10.3390/ijgi9090535","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090876880&doi=10.3390%2fijgi9090535&partnerID=40&md5=c3a4164be648f13fce9dacf04fefec63","In recent years semantic segmentation of 3D point clouds has been an argument that involves different fields of application. Cultural heritage scenarios have become the subject of this study mainly thanks to the development of photogrammetry and laser scanning techniques. Classification algorithms based on machine and deep learning methods allow to process huge amounts of data as 3D point clouds. In this context, the aim of this paper is to make a comparison between machine and deep learning methods for large 3D cultural heritage classification. Then, considering the best performances of both techniques, it proposes an architecture named DGCNN-Mod+3Dfeat that combines the positive aspects and advantages of these two methodologies for semantic segmentation of cultural heritage point clouds. To demonstrate the validity of our idea, several experiments from the ArCH benchmark are reported and commented. © 2020 by the authors." "57218877342;57202676994;57195343832;57218880958;6603232744;","IM2ELEVATION: Building height estimation from single-view aerial imagery",2020,"10.3390/RS12172719","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090549355&doi=10.3390%2fRS12172719&partnerID=40&md5=9abed4b44067c3aa027cad63c4a307c1","Estimation of the Digital Surface Model (DSM) and building heights from single-view aerial imagery is a challenging inherently ill-posed problem that we address in this paper by resorting to machine learning. We propose an end-to-end trainable convolutional-deconvolutional deep neural network architecture that enables learning mapping from a single aerial imagery to a DSM for analysis of urban scenes. We perform multisensor fusion of aerial optical and aerial light detection and ranging (Lidar) data to prepare the training data for our pipeline. The dataset quality is key to successful estimation performance. Typically, a substantial amount of misregistration artifacts are present due to georeferencing/projection errors, sensor calibration inaccuracies, and scene changes between acquisitions. To overcome these issues, we propose a registration procedure to improve Lidar and optical data alignment that relies on Mutual Information, followed by Hough transform-based validation step to adjust misregistered image patches. We validate our building height estimation model on a high-resolution dataset captured over central Dublin, Ireland: Lidar point cloud of 2015 and optical aerial images from 2017. These data allow us to validate the proposed registration procedure and perform 3D model reconstruction from single-view aerial imagery. We also report state-of-the-art performance of our proposed architecture on several popular DSM estimation datasets. © 2020 by the authors." "57218363740;56226977100;54684821900;57210823248;55963151700;57200700310;6603888005;7401925341;57208121325;","Gaussian processes retrieval of LAI from Sentinel-2 top-of-atmosphere radiance data",2020,"10.1016/j.isprsjprs.2020.07.004","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088914980&doi=10.1016%2fj.isprsjprs.2020.07.004&partnerID=40&md5=e4fbed46a2e8b16a80a0efbdb8428a12","Retrieval of vegetation properties from satellite and airborne optical data usually takes place after atmospheric correction, yet it is also possible to develop retrieval algorithms directly from top-of-atmosphere (TOA) radiance data. One of the key vegetation variables that can be retrieved from at-sensor TOA radiance data is leaf area index (LAI) if algorithms account for variability in atmosphere. We demonstrate the feasibility of LAI retrieval from Sentinel-2 (S2) TOA radiance data (L1C product) in a hybrid machine learning framework. To achieve this, the coupled leaf-canopy-atmosphere radiative transfer models PROSAIL-6SV were used to simulate a look-up table (LUT) of TOA radiance data and associated input variables. This LUT was then used to train the Bayesian machine learning algorithms Gaussian processes regression (GPR) and variational heteroscedastic GPR (VHGPR). PROSAIL simulations were also used to train GPR and VHGPR models for LAI retrieval from S2 images at bottom-of-atmosphere (BOA) level (L2A product) for comparison purposes. The BOA and TOA LAI products were consistently validated against a field dataset with GPR (R2 of 0.78) and with VHGPR (R2 of 0.80) and for both cases a slightly lower RMSE for the TOA LAI product (about 10% reduction). Because of delivering superior accuracies and lower uncertainties, the VHGPR models were further applied for LAI mapping using S2 acquisitions over the agricultural sites Marchfeld (Austria) and Barrax (Spain). The models led to consistent LAI maps at BOA and TOA scale. The LAI maps were also compared against LAI maps as generated by the SNAP toolbox, which is based on a neural network (NN). Maps were again consistent, however the SNAP NN model tends to overestimate over dense vegetation cover. Overall, this study demonstrated that hybrid LAI retrieval algorithms can be developed from TOA radiance data given a cloud-free sky, thus without the need of atmospheric correction. To the benefit of the community, the development of such hybrid models for the retrieval vegetation properties from BOA or TOA images has been streamlined in the freely downloadable ALG-ARTMO software framework. © 2020 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)" "57202079855;55494663500;","Volcano video data characterized and classified using computer vision and machine learning algorithms",2020,"10.1016/j.gsf.2020.01.016","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85081897711&doi=10.1016%2fj.gsf.2020.01.016&partnerID=40&md5=7ad1f0ad3d3a661782aedaa7a7550f10","Video cameras are common at volcano observatories, but their utility is often limited during periods of crisis due to the large data volume from continuous acquisition and time requirements for manual analysis. For cameras to serve as effective monitoring tools, video frames must be synthesized into relevant time series signals and further analyzed to classify and characterize observable activity. In this study, we use computer vision and machine learning algorithms to identify periods of volcanic activity and quantify plume rise velocities from video observations. Data were collected at Villarrica Volcano, Chile from two visible band cameras located ~17 km from the vent that recorded at 0.1 and 30 frames per second between February and April 2015. Over these two months, Villarrica exhibited a diverse range of eruptive activity, including a paroxysmal eruption on 3 March. Prior to and after the eruption, activity included nighttime incandescence, dark and light emissions, inactivity, and periods of cloud cover. We quantify the color and spatial extent of plume emissions using a blob detection algorithm, whose outputs are fed into a trained artificial neural network that categorizes the observable activity into five classes. Activity shifts from primarily nighttime incandescence to ash emissions following the 3 March paroxysm, which likely relates to the reemergence of the buried lava lake. Time periods exhibiting plume emissions are further analyzed using a row and column projection algorithm that identifies plume onsets and calculates apparent plume horizontal and vertical rise velocities. Plume onsets are episodic, occurring with an average period of ~50 s and suggests a puffing style of degassing, which is commonly observed at Villarrica. However, the lack of clear acoustic transients in the accompanying infrasound record suggests puffing may be controlled by atmospheric effects rather than a degassing regime at the vent. Methods presented here offer a generalized toolset for volcano monitors to classify and track emission statistics at a variety of volcanoes to better monitor periods of unrest and ultimately forecast major eruptions. © 2020 China University of Geosciences (Beijing) and Peking University" "56203601500;24072139200;","Utilization of Google Earth Engine (GEE) for land cover monitoring over Klang Valley, Malaysia",2020,"10.1088/1755-1315/540/1/012003","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090156914&doi=10.1088%2f1755-1315%2f540%2f1%2f012003&partnerID=40&md5=1e957d60bb508a88a751c952da0d5bf2","Geospatial Big Data is currently received overwhelming attention and are on highlight globally and Google Earth Engine (GEE) is currently the hot pot platform to cater big data processing for Remote Sensing and GIS. Currently few or no study regarding the usage of this platform to study land use/cover changes over years in Malaysia. The objective is to evaluate the feasibility of GEE as a free cloud-based platform by performing classification of Klang Valley area from Landsat composites of three different years (1988-2003-2018) using multiple Machine Learning Algorithms (MLA). The best classification results were then imported and further processed to quantify the changes over the years using commercial software. Although, the classification results are of high accuracy but CART shows the best accuracy with 94.71%, 97.72% and 96.57% in 1988, 2003 and 2018 in comparison with RF and SVM. Some misclassified pixels were encountered because the annual composited images were compiled without taken into considerations of crops phenological stages (paddy) which resulted to the misclassified agricultural land into urban and bare land. Hence, the selection and composition of data initially had to be structured and strategized prior to processing as they can affect the classification result and further analysis. Regardless, GEE has performed quite well and fast in term of time and processing complexity of multiple datasets with minimal human interaction and intervention. Generally, GEE has proven to be reliable in fulfilling the objectives of this study to evaluate the GEE feasibility by performing classification and quantifying the land use/cover of studied area and provide good base for further analysis using different platform. © Published under licence by IOP Publishing Ltd." "57219308076;57190759179;55053263400;","3D city models for urban mining: Point cloud based semantic enrichment for spectral variation identification in hyperspectral imagery",2020,"10.5194/isprs-Annals-V-4-2020-223-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092182432&doi=10.5194%2fisprs-Annals-V-4-2020-223-2020&partnerID=40&md5=1bebddcb9b0c91887a2c89db06843300","Urban mining aims at reusing building materials enclosed in our cities. Therefore, it requires accurate information on the availability of these materials for each separate building. While recent publications have demonstrated that such information can be obtained using machine learning and data fusion techniques applied to hyperspectral imagery, challenges still persist. One of these is the so-called 'salt-And-pepper noise', i.e.The oversensitivity to the presence of several materials within one pixel (e.g. chimneys, roof windows). For the specific case of identifying roof materials, this research demonstrates the potential of 3D city models to identify and filter out such unreliable pixels beforehand. As, from a geometrical point of view, most available 3D city models are too generalized for this purpose (e.g. in CityGML Level of Detail 2), semantic enrichment using a point cloud is proposed to compensate missing details. So-called deviations are mapped onto a 3D building model by comparing it with a point cloud. Seeded region growing approach based on distance and orientation features is used for the comparison. Further, the results of a validation carried out for parts of Rotterdam and resulting in KHAT values as high as 0.7 are discussed. © Copyright:" "57195695111;23005893600;57216759668;","Supervised Classification and Its Repeatability for Point Clouds from Dense Vhr Tri-Stereo Satellite Image Matching Using Machine Learning",2020,"10.5194/isprs-annals-V-2-2020-525-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091102393&doi=10.5194%2fisprs-annals-V-2-2020-525-2020&partnerID=40&md5=21c722c8bc08a99d4e0b266991fa510b","Image matching of aerial or satellite images and Airborne Laser Scanning (ALS) are the two main techniques for the acquisition of geospatial information (3D point clouds), used for mapping and 3D modelling of large surface areas. While ALS point cloud classification is a widely investigated topic, there are fewer studies related to the image-derived point clouds, even less for point clouds derived from stereo satellite imagery. Therefore, the main focus of this contribution is a comparative analysis and evaluation of a supervised machine learning classification method that exploits the full 3D content of point clouds generated by dense image matching of tri-stereo Very High Resolution (VHR) satellite imagery. The images were collected with two different sensors (Pleíades and WorldView-3) at different timestamps for a study area covering a surface of 24 km2, located in Waldviertel, Lower Austria. In particular, we evaluate the performance and precision of the classifier by analysing the variation of the results obtained after multiple scenarios using different training and test data sets. The temporal difference of the two Pleíades acquisitions (7 days) allowed us to calculate the repeatability of the adopted machine learning algorithm for the classification. Additionally, we investigate how the different acquisition geometries (ground sample distance, viewing and convergence angles) influence the performance of classifying the satellite image-derived point clouds into five object classes: ground, trees, roads, buildings, and vehicles. Our experimental results indicate that, in overall the classifier performs very similar in all situations, with values for the F1-score between 0.63 and 0.65 and overall accuracies beyond 93%. As a measure of repeatability, stable classes such as buildings and roads show a variation below 3% for the F1-score between the two Pleíades acquisitions, proving the stability of the model. © 2020 Copernicus GmbH. All rights reserved." "57202445503;23767232700;6602209960;","Classification of Tree Species and Standing Dead Trees by Fusing Uav-Based Lidar Data and Multispectral Imagery in the 3D Deep Neural Network Pointnet++",2020,"10.5194/isprs-annals-V-2-2020-203-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091088085&doi=10.5194%2fisprs-annals-V-2-2020-203-2020&partnerID=40&md5=f57a7ce628cb5ba3acfbe2d7a609ebe0","Knowledge of tree species mapping and of dead wood in particular is fundamental to managing our forests. Although individual tree-based approaches using lidar can successfully distinguish between deciduous and coniferous trees, the classification of multiple tree species is still limited in accuracy. Moreover, the combined mapping of standing dead trees after pest infestation is becoming increasingly important. New deep learning methods outperform baseline machine learning approaches and promise a significant accuracy gain for tree mapping. In this study, we performed a classification of multiple tree species (pine, birch, alder) and standing dead trees with crowns using the 3D deep neural network (DNN) PointNet++ along with UAV-based lidar data and multispectral (MS) imagery. Aside from 3D geometry, we also integrated laser echo pulse width values and MS features into the classification process. In a preprocessing step, we generated the 3D segments of single trees using a 3D detection method. Our approach achieved an overall accuracy (OA) of 90.2% and was clearly superior to a baseline method using a random forest classifier and handcrafted features (OA Combining double low line 85.3%). All in all, we demonstrate that the performance of the 3D DNN is highly promising for the classification of multiple tree species and standing dead trees in practice. © 2020 Copernicus GmbH. All rights reserved." "26221613300;57218827452;55586607900;24536694300;","Automated updating of forest cover maps from cloud-free sentinel-2 mosaic images using object-based image analysis and machine learning methods",2020,"10.5194/isprs-Annals-V-3-2020-803-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090343402&doi=10.5194%2fisprs-Annals-V-3-2020-803-2020&partnerID=40&md5=dad63df872c4d4a0b24f30bfd4bcd860","Planning sustainable use of land resources and environmental monitoring benefit from accurate and detailed forest information. The basis of accurate forest information is data on the spatial extent of forests. In Norway land resource maps have been carefully created by field visits and aerial image interpretation for over four decades with periodic updating. However, due to prioritization of agricultural and built-up areas, and high requirements with respect to the map accuracy, forest areas and outfields have not been frequently updated. Consequently, in some part of the country, the map has not been updated since its first creation in the 1960s. The Sentinel-2 satellite acquires images with high spatial and temporal resolution which provides opportunities for creating cloud-free mosaic images over areas that are often covered with clouds. Here, we combine object-based image analysis with machine learning methods in an automated framework to map forest area in Sentinel-2 mosaic images. The images are segmented using the eCogntion™ software. Training data are collected automatically from the existing land resource map and filtered using height and greenness information so that the training samples certainly represent their respective classes. Two machine learning algorithms, namely Random Forest (RF) and the Multilayer Perceptron Neural Network (MLP), are then trained and validated before mapping forest area. The effects of including and excluding some features on the classification accuracy is investigated. The results show that the method produces forest cover map at very high accuracy (up to 97%). The MLP performs better than the RF algorithm both in classification accuracy and in robustness against inclusion and exclusion of features. © Authors 2020. All rights reserved." "57218827879;57218822575;6507146779;57218829464;57210999953;35203565800;57218822623;57192830024;57198996571;","Machine learning for classification of an eroding scarp surface using terrestrial photogrammetry with nir and rgb imagery",2020,"10.5194/isprs-Annals-V-3-2020-431-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090337515&doi=10.5194%2fisprs-Annals-V-3-2020-431-2020&partnerID=40&md5=d0043d3d0246cf9f82296bf8f7bdb9a0","Increasingly advanced and affordable close-range sensing techniques are employed by an ever-broadening range of users, with varying competence and experience. In this context a method was tested that uses photogrammetry and classification by machine learning to divide a point cloud into different surface type classes. The study site is a peat scarp 20 metres long in the actively eroding river bank of the Rotmoos valley near Obergurgl, Austria. Imagery from near-infra red (NIR) and conventional (RGB) sensors, georeferenced with coordinates of targets surveyed with a total station, was used to create a point cloud using structure from motion and dense image matching. NIR and RGB information were merged into a single point cloud and 18 geometric features were extracted using three different radii (0.02 m, 0.05 m and 0.1 m) totalling 58 variables on which to apply the machine learning classification. Segments representing six classes, dry grass, green grass, peat, rock, snow and target, were extracted from the point cloud and split into a training set and a testing set. A Random Forest machine learning model was trained using machine learning packages in the R-CRAN environment. The overall classification accuracy and Kappa Index were 98% and 97% respectively. Rock, snow and target classes had the highest producer and user accuracies. Dry and green grass had the highest omission (1.9% and 5.6% respectively) and commission errors (3.3% and 3.4% respectively). Analysis of feature importance revealed that the spectral descriptors (NIR, R, G, B) were by far the most important determinants followed by verticality at 0.1 m radius. © Authors 2020. All rights reserved." "57188763259;57196055256;","Virtual infrastructure provisioning virtual machine with machine learning prediction in green cloud computing",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091597411&partnerID=40&md5=49996417f1b5d648a644662201da215f","Cloud computing is one of the emerging technology in the world it refer different technologies, services concepts and architectures. Cloud has provided everything as a service to end users, medium and large scale enterprises across globally. The datacenters contribute major role in the cloud, it has different architecture, service provided by cloud providers. The datacenters level required lots of optimization, it will helps to gain the less energy consumption, less pollution and carbon emission, provide more powerful performance to consumer with less cost, QoS within SLA, innovation. To achieve these datacenter levels required continuously monitoring, analysis and take decisions on the log level, monitoring metrics, thresholds. The datacenters resource objects using training and prediction on usage during provisioning and server consolidation. The application workflow request and response experiment analyses. There selections of energy selection on datacenter. The agent based monitoring, agent less script based monitoring, automation, artificial intelligence, neural network. This paper mainly concentrated analysis metrics, thresholds on cloud, capacity analysis, application workflow and Resource such as CPU, RAM utilization and prediction using Machine Learning techniques such as LR, RantomTree, RandomForest and 10 folds cross validation in cloud. The data center power source, architecture selection and usage of resource in usage optimum will helps us to reduce carbon emission and green cloud computing. © 2020 Alpha Publishers. All rights reserved." "55764719600;57217651295;55415880000;6506685277;","Optimized energy cost and carbon emission-aware virtual machine allocation in sustainable data centers",2020,"10.3390/SU12166383","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090083510&doi=10.3390%2fSU12166383&partnerID=40&md5=f66ddaee4b3bf7b872c8e80cd15eb3d1","Cloud data center's total operating cost is conquered by electricity cost and carbon tax incurred due to energy consumption from the grid and its associated carbon emission. In this work, we consider geo-distributed sustainable datacenter's with varying on-site green energy generation, electricity prices, carbon intensity and carbon tax. The objective function is devised to reduce the operating cost including electricity cost and carbon cost incurred on the power consumption of servers and cooling devices. We propose renewable-aware algorithms to schedule the workload to the data centers with an aim to maximize the green energy usage. Due to the uncertainty and time variant nature of renewable energy availability, an investigation is performed to identify the impact of carbon footprint, carbon tax and electricity cost in data center selection on total operating cost reduction. In addition, on-demand dynamic optimal frequency-based load distribution within the cluster nodes is performed to eliminate hot spots due to high processor utilization. The work suggests optimal virtual machine placement decision to maximize green energy usage with reduced operating cost and carbon emission. © 2020 by the authors." "57190034210;57194694881;55444846100;54787529900;6507122674;","A hierarchical machine learning approach for multi-level and multi-resolution 3d point cloud classification",2020,"10.3390/RS12162598","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090017553&doi=10.3390%2fRS12162598&partnerID=40&md5=1d252318efd2db65b905fc1cde7c5ed9","The recent years saw an extensive use of 3D point cloud data for heritage documentation, valorisation and visualisation. Although rich in metric quality, these 3D data lack structured information such as semantics and hierarchy between parts. In this context, the introduction of point cloud classification methods can play an essential role for better data usage, model definition, analysis and conservation. The paper aims to extend a machine learning (ML) classification method with a multi-level and multi-resolution (MLMR) approach. The proposed MLMR approach improves the learning process and optimises 3D classification results through a hierarchical concept. The MLMR procedure is tested and evaluated on two large-scale and complex datasets: the Pomposa Abbey (Italy) and the Milan Cathedral (Italy). Classification results show the reliability and replicability of the developed method, allowing the identification of the necessary architectural classes at each geometric resolution. © 2020 by the authors." "57211519015;8600097300;57213346004;57192512150;57205863098;","A novel classification extension-based cloud detection method for medium-resolution optical images",2020,"10.3390/RS12152365","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089689487&doi=10.3390%2fRS12152365&partnerID=40&md5=147cee8296c6f2bf6a93c639729a2d6d","Accurate cloud detection using medium-resolution multispectral satellite imagery (such as Landsat and Sentinel data) is always difficult due to the complex land surfaces, diverse cloud types, and limited number of available spectral bands, especially in the case of images without thermal bands. In this paper, a novel classification extension-based cloud detection (CECD) method was proposed for masking clouds in the medium-resolution images. The new method does not rely on thermal bands and can be used for masking clouds in different types of medium-resolution satellite imagery. First, with the support of low-resolution satellite imagery with short revisit periods, cloud and non-cloud pixels were identified in the resampled low-resolution version of the medium-resolution cloudy image. Then, based on the identified cloud and non-cloud pixels and the resampled cloudy image, training samples were automatically collected to develop a random forest (RF) classifier. Finally, the developed RF classifier was extended to the corresponding medium-resolution cloudy image to generate an accurate cloud mask. The CECD method was applied to Landsat-8 and Sentinel-2 imagery to test the performance for different satellite images, and the well-known function of mask (FMASK) method was employed for comparison with our method. The results indicate that CECD is more accurate at detecting clouds in Landsat-8 and Sentinel-2 imagery, giving an average F-measure value of 97.65% and 97.11% for Landsat-8 and Sentinel-2 imagery, respectively, as against corresponding results of 90.80% and 88.47% for FMASK. It is concluded, therefore, that the proposed CECD algorithm is an effective cloud-classification algorithm that can be applied to the medium-resolution optical satellite imagery. © 2020 by the authors." "57218454517;24822286200;","Lightning Distance Estimation Using LF Lightning Radio Signals via Analytical and Machine-Learned Models",2020,"10.1109/TGRS.2020.2972153","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089239621&doi=10.1109%2fTGRS.2020.2972153&partnerID=40&md5=9f5269d976911f2ce5480140d766b9f5","Lightning geolocation is useful in a variety of applications, ranging from weather nowcasting to a better understanding of thunderstorm evolution processes and remote sensing of the ionosphere. Lightning-generated radio signals can be used in range estimation of lightning return strokes, for which the most commonly employed technique is the time difference of arrival in lightning detection networks. Though these instrument networks provide the most reliability and best accuracy, users without access to them can instead benefit from lightning geolocation using a standalone instrument. In this article, we present the framework for training fast models capable of estimating negative cloud-to-ground lightning location from single-instrument observations of very low frequency/low frequency (VLF/LF, 3-300 kHz) radio pulses or 'sferics,' without knowledge of the ionosphere's D-region state. The models are generated using an analytical method, based on the delay between ground and skywave, and a machine learning method. The training framework is applied to three different data sets to assess model accuracy. Validation of the machine-learned models for these data sets, which include both simulated and observed sferics, confirms this technique as a promising solution for lightning distance estimation using a single receiver. Distance estimates using a machine-learned model for observed sferics in Kansas yield an RMSE of 53 km with 68% of them being within 9.8 km. Estimates using the analytical method are found to have an RMSE of 54 km with 68% of them being within 32 km. Limitations of our methodology and potential improvements to be investigated are also discussed. © 1980-2012 IEEE." "57218474008;15023850900;55630272400;57217680171;57217687864;57217675347;22234853700;","Thin cloud removal in optical remote sensing images based on generative adversarial networks and physical model of cloud distortion",2020,"10.1016/j.isprsjprs.2020.06.021","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087526039&doi=10.1016%2fj.isprsjprs.2020.06.021&partnerID=40&md5=94e3dce17405ccf1cf7703204fe5079b","Cloud contamination is an inevitable problem in optical remote sensing images. Unlike thick clouds, thin clouds do not completely block out background which makes it possible to restore background information. In this paper, we propose a semi-supervised method based on generative adversarial networks (GANs) and a physical model of cloud distortion (CR-GAN-PM) for thin cloud removal with unpaired images from different regions. A physical model of cloud distortion which takes the absorption of cloud into consideration was also defined in this paper. It is worth noting that many state-of-the-art methods based on deep learning require paired cloud and cloud-free images from the same region, which is often unavailable or time-consuming to collect. CR-GAN-PM has two main steps: first, the cloud-free background and cloud distortion layers were decomposed from an input cloudy image based on GANs and the principles of image decomposition; then, the input cloudy image was reconstructed by putting those layers into the redefined physical model of cloud distortion. The decomposition process ensured that the decomposed background layer was cloud-free and the reconstruction process ensured that generated background layer was correlated with the input cloudy image. Experiments were conducted on Sentinel-2A imagery to validate the proposed CR-GAN-PM. Averaged over all testing images, the SSIMs values (structural similarity index measurement) of CR-GAN-PM were 0.72, 0.77, 0.81 and 0.83 for visible and NIR bands respectively. Those results were similar to the end-to-end deep learning-based methods and better than traditional methods. The number of input bands and values of hyper-parameters affected little on the performance of CR-GAN-PM. Experimental results show that CR-GAN-PM is effective and robust for thin cloud removal in different bands. © 2020 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)" "57194328647;55807752200;7004635295;","Machine learning techniques for regional scale estimation of high-resolution cloud-free daily sea surface temperatures from MODIS data",2020,"10.1016/j.isprsjprs.2020.06.008","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086632052&doi=10.1016%2fj.isprsjprs.2020.06.008&partnerID=40&md5=4cebe15a5dbe418a6b996de59fe95052","High-resolution sea surface temperature (SST) estimates are dependent on satellite-based infrared radiometers, which are proven to be highly accurate in the past decades. However, the presence of clouds is a big stumbling block when physical approaches are used to derive SST. This problem is more prominent across tropical regions such as Arabian Sea(AS) and Bay of Bengal(BoB), restricting the availability of high-resolution SST data for ocean applications. The previous studies for developing daily high-resolution cloud-free SST products mainly focus on fusion of multiple satellites and in-situ data products that are computationally expensive and often time consuming. At the same time, it was observed that the capabilities of data-driven approaches are not yet fully explored in the estimation of cloud-free high-resolution SST data. Hence, in this study an attempt has been made for the first time to estimate daily cloud free SST from a single sensor (MODIS Aqua) dataset using advanced machine learning techniques. Here, three distinct machine learning techniques such as Artificial Neural Networks (ANN), Support Vector Regression (SVR) and Random Forest (RF)-based algorithms were developed and evaluated over two different study areas within the AS and BoB using 10 years of MODIS data and in-situ reference data. Among the developed algorithms, the SVR-based algorithm performs consistently better. In AS region, while testing, the SVR-based SST estimates was able to achieve an adjusted coefficient of determination (Radj2) of 0.82 and root mean square error (RMSE) of 0.71 °C with respect to the in situ data. Similarly, in BoB too, the SVR algorithm outperforms the other algorithms with Radj2 of 0.78 with RMSE of 0.88 °C. Further, a spatio-temporal and visual analysis of the results as well as an inter-comparision with NOAA AVHRR daily optimally interpolated global SST (a standard SST product available in practice) the suggest that the proposed SVR-based algorithm has huge potential to produce operational high-resolution cloud-free SST estimates, even if there is cloud cover in the image. © 2020 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)" "57218292971;7409462943;57190281074;57215817129;57196043593;57196040405;57218393669;56809128600;37041512600;","Upward Expansion of Supra-Glacial Debris Cover in the Hunza Valley, Karakoram, During 1990 ∼ 2019",2020,"10.3389/feart.2020.00308","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089099686&doi=10.3389%2ffeart.2020.00308&partnerID=40&md5=4c2f8c9b7afcb4de435632ddc3fd0e5c","Supra-glacial debris cover is key to glacier ablation through increasing (thin debris layer) or decreasing (thick debris layer) melt rates, thereby regulating the mass balance of a glacier and its meltwater runoff. The thickening or lateral expansion of supra-glacial debris cover correlates with a reduction of glacier ablation and, consequently, runoff generation, which is also considered to be an influential factor on the rheology and dynamics of a glacierized system. Studies on supra-glacial debris cover have recently attracted wide attention especially for glaciers in the Himalayas and Karakoram, where the glaciers have heterogeneously responded to climate change. In this study, we used 32 images from the Landsat Thematic Mapper, Enhanced Thematic Mapper Plus, and Operational Land Imager archive, going back to 1990, which are available on the Google Earth Engine cloud-computing platform, to map the supra-glacial debris cover in the Hunza Valley, Karakoram, Pakistan, based on a band ratio segmentation method (normalized difference snow index [NDSI] < 0.4), Otsu thresholding, and machine learning algorithms. Compared with manual digitization, the random forest (RF) model was found to have the greatest accuracy in identifying supra-glacial debris, with a Kappa coefficient of 0.94 ± 0.01 and an overall accuracy of 95.5 ± 0.9%. Overall, the supra-glacial debris cover in the study area showed an increasing trend, and the total area expanded by 8.1–21.3% for various glaciers from 1990 to 2019. The other two methods (Otsu thresholding and NDSI < 0.4) generally overestimated the supra-glacial debris covered area, by 36.3 and 18.8%, respectively, compared to that of the RF model. The supra-glacial debris cover has migrated upward on the glaciers, with intensive variation near the equilibrium-line altitude zone (4,500–5,500 m a.s.l.). The increase in ice or snow avalanche activity at high altitudes may be responsible for this upward expansion of supra-glacial debris cover in the Hunza Valley, which is attributed to the combined effect of temperature decrease and precipitation increase in the study area. © Copyright © 2020 Xie, Liu, Wu, Zhu, Gao, Qi, Duan, Saifullah and Tahir." "57213598861;36921601500;7003341789;55636444100;55479763800;8791306500;7003817690;15770244400;","Exploration of machine learning methods for the classification of infrared limb spectra of polar stratospheric clouds",2020,"10.5194/amt-13-3661-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088032027&doi=10.5194%2famt-13-3661-2020&partnerID=40&md5=48014d96d020cc98e0b2b7432f66887b","Polar stratospheric clouds (PSCs) play a key role in polar ozone depletion in the stratosphere. Improved observations and continuous monitoring of PSCs can help to validate and improve chemistry-climate models that are used to predict the evolution of the polar ozone hole. In this paper, we explore the potential of applying machine learning (ML) methods to classify PSC observations of infrared limb sounders. Two datasets were considered in this study. The first dataset is a collection of infrared spectra captured in Northern Hemisphere winter 2006/2007 and Southern Hemisphere winter 2009 by the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) instrument on board the European Space Agency's (ESA) Envisat satellite. The second dataset is the cloud scenario database (CSDB) of simulated MIPAS spectra. We first performed an initial analysis to assess the basic characteristics of the CSDB and to decide which features to extract from it. Here, we focused on an approach using brightness temperature differences (BTDs). From both the measured and the simulated infrared spectra, more than 10 000 BTD features were generated. Next, we assessed the use of ML methods for the reduction of the dimensionality of this large feature space using principal component analysis (PCA) and kernel principal component analysis (KPCA) followed by a classification with the support vector machine (SVM). The random forest (RF) technique, which embeds the feature selection step, has also been used as a classifier. All methods were found to be suitable to retrieve information on the composition of PSCs. Of these, RF seems to be the most promising method, being less prone to overfitting and producing results that agree well with established results based on conventional classification methods. © Author(s) 2020. This work is distributed under the Creative Commons Attribution 4.0 License." "55266523600;57188869324;57215429741;57215414947;6603393238;35762297500;6508239454;","A preliminary study on machine learning and google earth engine for mangrove mapping",2020,"10.1088/1755-1315/500/1/012038","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087755156&doi=10.1088%2f1755-1315%2f500%2f1%2f012038&partnerID=40&md5=2c43b7e9b8788ad2f2a69f47d87ae9d8","The alarming rate of global mangrove forest degradation corroborates the need for providing fast, up-to-date and accurate mangrove maps. Conventional scene by scene image classification approach is inefficient and time consuming. The development of Google Earth Engine (GEE) provides a cloud platform to access and seamlessly process large amount of freely available satellite imagery. The GEE also provides a set of the state-of-the-art classifiers for pixel-based classification that can be used for mangrove mapping. This study is an initial effort which is aimed to combine machine learning and GEE for mapping mangrove extent. We used two Landsat 8 scenes over Agats and Timika Papua area as pilot images for this study; path 102 row 64 (2014/10/19) and path 103 row 63 (2013/05/16). The first image was used to develop local training areas for the machine learning classification, while the second one was used as a test image for GEE on the cloud. A total of 838 points samples were collected representing mangroves (244), non-mangroves (161), water bodies (311), and cloud (122) class. These training areas were used by support vector machine classifier in GEE to classify the first image. The classification result show mangrove objects could be efficiently delineated by this algorithm as confirmed by visual checking. This algorithm was then applied to the second image in GEE to check the consistency of the result. A simultaneous view of both classified images shows a corresponding pattern of mangrove forest, which mean the mangrove object has been consistently delineated by the algorithm. © 2020 IOP Publishing Ltd. All rights reserved." "57209638348;57217094757;57208372273;57188746436;","Snow and cloud detection using a convolutional neural network and low-resolution data from the Electro-L No. 2 Satellite",2020,"10.1117/1.JRS.14.034506","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092645221&doi=10.1117%2f1.JRS.14.034506&partnerID=40&md5=6c3b3d056a9a651aeb99b4efec2b2893","We describe an algorithm based on a convolutional neural network that detects cloud formations and snow cover in satellite images using textures. Herein, multispectral satellite images, received from a multizone scanning instrument used for hydrometeorological support and installed on the Russian satellite Electro-L No. 2, are used as input data. The problem of snow and cloud classification in the absence of a spectral channel in the range of 1.4 to 1.8 μm, which is necessary for their accurate separation, is considered. The developed algorithm can produce cloud and snow cover masks for an area limited by the values of the solar zenith angle in the range of 0 deg to 80 deg for daytime. Algorithm accuracy was evaluated using machine learning metrics and comparing its results with ground truth masks segmented manually by an experienced interpreter. In addition, we compared the resulting masks with a similar cloud mask product from the European Organisation for the Exploitation of Meteorological Satellites based on the data of the Spinning Enhanced Visible and Infrared Imager (SEVIRI) instrument installed on the Meteosat-8 satellite. According to the results of this comparison, we conclude that the cloud masks produced by the proposed convolutional neural network-based algorithm have a lower probability of false detection than products based on the SEVIRI data. The proposed algorithm is fully automatic, and it works in any season of the year during the daytime. © 2020 Society of Photo-Optical Instrumentation Engineers (SPIE)." "57211970265;57211855484;57210936702;57218799260;","A novel intelligent diagnosis and disease prediction algorithm in green cloud using machine learning approach",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090686536&partnerID=40&md5=e8b122ffbd1395c2d96bc43218ac29ba","Public health threats and epidemics are affecting the human life day-to-day. These include obesity, diabetes, cardiovascular diseases, cancer, osteoporosis and dental diseases. The trouble of chronic diseases may be cured during initial stages if it is properly predicted and it requires a lot of training using the medical data. Existing models can support disease diagnosis to certain extent based on the training. This can predict unique diseases and separate system is required for different diagnosis. A generic disease prediction model can reduce the burden of physician while making clinical decisions and has not been evolved yet. Machine learning and artificial intelligence offer one such a generic model with principled approach for intelligent disease diagnosis. A novel hybrid algorithm and a prediction model are proposed in this paper based on disease symptoms of the patient Collected from hospital and are stored in cloud.The features of KNN and CNN are combined to provide high speed prediction analysis. The enormous amount of data growth in the field of medical diagnosis helps the proposed system to find hidden patterns with respect to individual diseases. © 2020 Alpha Publishers. All rights reserved." "57201727508;55740373500;35323728000;","Combining forecasts of day-ahead solar power",2020,"10.1016/j.energy.2020.117743","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089285975&doi=10.1016%2fj.energy.2020.117743&partnerID=40&md5=e7edef1d4b83edb6e19fa50afdc5bcc0","Solar power forecasting is important for the reliable and economic operation of power systems with high penetration of solar energy. The solar power forecasts for the day-ahead time horizon are more erroneous than the hour-ahead time horizon. Numerical weather prediction (NWP) variables such as irradiance, cloud cover, precipitation etc. are used as input to day-ahead forecasting models. The uncertainty in NWP varies with weather conditions. Different forecasting algorithms based on a single method are available in the literature. Combination of individual forecasting algorithms increases the accuracy of the forecasts. However, the combined-forecast has yet not been analysed much in the area of day-ahead solar power forecasting. This paper thus explores different combined-forecast methods such as mean, median, linear regression and non-linear regressions using supervised machine learning algorithms. The number of models required for day-ahead solar power forecasts is studied. One for all hour (same) or separate models for each hour of the day are possible. The effects of retraining frequency on the performance of the forecasting models, which is important for the computational burden of the system, are also studied. Forecasting algorithms are applied to three solar plants in Australia. © 2020 Elsevier Ltd" "57216154309;35332559300;57217729762;21933438300;","Hydrometeor identification using multiple-frequency microwave links: A numerical simulation",2020,"10.3390/rs12132158","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087547745&doi=10.3390%2frs12132158&partnerID=40&md5=e88a1272b2a0ca67620ab317fefbdcd0","A method for identifying hydrometeor types (rain, graupel, and wet snow) based on a microwave link is proposed in this paper. The measured hydrometeor size distribution (HSD) data from the winters of 2014 to 2019 in Nanjing, China, were used to carry out simulation experiments to verify the performance of the model. Single-, dual-, and tri-frequency models (combinations of 15 GHz, 18 GHz, 25 GHz, 38 GHz, 50 GHz, 60 GHz, 70 GHz, and 80 GHz) were established with the extreme learning machine (ELM) algorithm. The results showed that the performance of the tri-frequency models was overall better than that of the dual-frequency models, for which the performance was better than that of the single-frequency models. The mean (maximum) test set accuracies of the single-frequency, dual-frequency, and tri-frequency models reached 75.8%, 80.7%, and 83.2% (83.0%, 84.4%, and 85.6%), respectively. For the dual-frequency and tri-frequency models, it was found that the accuracy increased with the overall frequency or the frequency difference. In addition, the influences of different noise levels on the model performance were also analyzed. Finally, the effects of position and length of link relative to precipitation cell were analyzed and are also discussed. © 2020 by the authors." "57193788172;57209398568;57206924573;6506051565;7004171611;36622868000;55914904100;55466977400;7102689523;57217728673;8525147900;","Cirrus cloud identification from airborne far-infrared and mid-infrared spectra",2020,"10.3390/rs12132097","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087543275&doi=10.3390%2frs12132097&partnerID=40&md5=676968cfa40c35c7e83f55bda62a3c94","Airborne interferometric data, obtained from the Cirrus Coupled Cloud-Radiation Experiment (CIRCCREX) and from the PiknMix-F field campaign, are used to test the ability of a machine learning cloud identification and classification algorithm (CIC). Data comprise a set of spectral radiances measured by the Tropospheric Airborne Fourier Transform Spectrometer (TAFTS) and the Airborne Research Interferometer Evaluation System (ARIES). Co-located measurements of the two sensors allow observations of the upwelling radiance for clear and cloudy conditions across the far-and mid-infrared part of the spectrum. Theoretical sensitivity studies show that the performance of the CIC algorithm improves with cloud altitude. These tests also suggest that, for conditions encompassing those sampled by the flight campaigns, the additional information contained within the far-infrared improves the algorithm's performance compared to using mid-infrared data only. When the CIC is applied to the airborne radiance measurements, the classification performance of the algorithm is very high. However, in this case, the limited temporal and spatial variability in the measured spectra results in a less obvious advantage being apparent when using both mid-and far-infrared radiances compared to using mid-infrared information only. These results suggest that the CIC algorithm will be a useful addition to existing cloud classification tools but that further analyses of nadir radiance observations spanning the infrared and sampling a wider range of atmospheric and cloud conditions are required to fully probe its capabilities. This will be realised with the launch of the Far-infrared Outgoing Radiation Understanding and Monitoring (FORUM) mission, ESA's 9th Earth Explorer. © 2020 by the authors. Licensee MDPI, Basel, Switzerland." "57204396783;57190372479;57188719498;57216602841;12785706100;57214208469;","GRNet: Geometric relation network for 3D object detection from point clouds",2020,"10.1016/j.isprsjprs.2020.05.008","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085217744&doi=10.1016%2fj.isprsjprs.2020.05.008&partnerID=40&md5=cbfa85069ee2beaa1de323270b6b1b68","Rapid detection of 3D objects in indoor environments is essential for indoor mapping and modeling, robotic perception and localization, and building reconstruction. 3D point clouds acquired by a low-cost RGB-D camera have become one of the most commonly used data sources for 3D indoor mapping. However, due to the sparse surface, empty object center, and various scales of point cloud objects, 3D bounding boxes are challenging to be estimated and located accurately. To address this, geometric shape, topological structure, and object relation are commonly employed to extract box reasoning information. In this paper, we describe the geometric feature among object points as an intra-object feature and the relation feature between different objects as an inter-object feature. Based on these two features, we propose an end-to-end point cloud geometric relation network focusing on 3D object detection, which is termed as geometric relation network (GRNet). GRNet first extracts intra-object and inter-object features for each representative point using our proposed backbone network. Then, a centralization module with a scalable loss function is proposed to centralize each representative object point to its center. Next, proposal points are sampled from these shifted points, following a proposal feature pooling operation. Finally, an object-relation learning module is applied to predict bounding box parameters. Such parameters are the additive sum of prediction results from the relation-based inter-object feature and the aggregated intra-object feature. Our model achieves state-of-the-art 3D detection results with 59.1% mAP@0.25 and 39.1% mAP@0.5 on ScanNetV2 dataset, 58.4% mAP@0.25 and 34.9% mAP@0.5 on SUN RGB-D dataset. © 2020" "57197867114;24450860900;","Linking large-scale circulation patterns to low-cloud properties",2020,"10.5194/acp-20-7125-2020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086723013&doi=10.5194%2facp-20-7125-2020&partnerID=40&md5=fb995243ca9423be7244438cf5e1f936","The North Pacific High (NPH) is a fundamental meteorological feature present during the boreal warm season. Marine boundary layer (MBL) clouds, which are persistent in this oceanic region, are influenced directly by the NPH. In this study, we combine 11 years of reanalysis and an unsupervised machine learning technique to examine the gamut of 850 hPa synoptic-scale circulation patterns. This approach reveals two distinguishable regimes-a dominant NPH setup and a land-falling cyclone-and in between a spectrum of large-scale patterns. We then use satellite retrievals to elucidate for the first time the explicit dependence of MBL cloud properties (namely cloud droplet number concentration, liquid water path, and shortwave cloud radiative effect-CRESW) on 850 hPa circulation patterns over the northeast Pacific Ocean. We find that CRESW spans from-146.8 to-115.5Wm-2, indicating that the range of observed MBL cloud properties must be accounted for in global and regional climate models. Our results demonstrate the value of combining reanalysis and satellite retrievals to help clarify the relationship between synoptic-scale dynamics and cloud physics. © 2020 Author(s)." "14035135900;57216397984;57219482763;","A digital twin model of a pasteurization system for food beverages: Tools and architecture",2020,"10.1109/ICE/ITMC49519.2020.9198625","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093120907&doi=10.1109%2fICE%2fITMC49519.2020.9198625&partnerID=40&md5=ea9fe5677b66bb4e2467f46e8abe2278","Many enabling technologies of Industry 4.0 (Internet of Things 'IoT', Cloud systems, Big Data Analytics) contribute to the creation of what is the Digital Twin or virtual twin of a physical process, that is a mathematical model capable of describing the process, product or service in a precise way in order to carry out analyses and apply strategies. Digital Twin models integrate artificial intelligence, machine learning and analytics software with the data collected from the production plants to create digital simulation models that update when the parameters of the production processes or the working conditions change. This is a self-learning mechanism, which makes use of data collected from various sources (sensors that transmit operating conditions; experts, such as engineers with deep knowledge of the industrial domain; other similar machines or fleets of similar machines) and integrates also historical data relating to the past use of the machine. Starting from the virtual twin vision, simulation plays a key role within the Industry 4.0 transformation. Creating a virtual prototype has become necessary and strategic to raise the safety levels of the operators engaged in the maintenance phases, but above all the integration of the digital model with the IoT has become particularly effective, as the advent of software platforms offers the possibility of integrating real-time data with all the digital information that a company owns on a given process, ensuring the realization of the Digital Twin. In this context, this work aims at developing optimized solutions for application in a beverage pasteurization system using the Digital Twin approach, capable of creating a virtual modelling of the process and preventing high-risk events for operators. © 2020 IEEE." "57213701812;14027120600;","Green IoT: Advancements and Sustainability with Environment by 2050",2020,"10.1109/ICRITO48877.2020.9197796","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093115224&doi=10.1109%2fICRITO48877.2020.9197796&partnerID=40&md5=25f144593f3fd246251cf0ccc3e1c9fc","Green IoT is eco-friendly technology evolves from IoT, in today's scenario each and every device connected over internet and store on cloud and these devices are known as 'Smart Devices'. At present 31 billion devices are connected as IoT devices and by 2050 it surge pass 170 billion limit so on average Iot devices are increases 12% annually as per the reports in proportionally the carbon footprint percentage and GHG(green-house gas) emission percentage increases and enhance the overall pollution percentage on earth, so to look after this major environmental issues and for proper initiatives against these issues technically as well as environmental prospective first objective of this paper is analyze the carbon footprints of smart devices with IoT for sustainability with environment and provide major steps to minimize it and emphasize the use of 'Green IoT' instead of IoT for environment safety perspectives and second objective is improve the LCA assessment model with Deep Learning and Data Mining techniques with various impact factors for better and efficient results.The proposed method cut down the production and manufacturing cycle rate and promotes the recyclability and usability of devices from environmental point of view and from technical point of view it cut down the overall cost and save energy and time during installing phase of smart devices with updating service packs and in overall estimation of GHG(green-house gas)emission rate and carbon footprint rate. © 2020 IEEE." "57219486486;7004522901;15730011900;","Potential Bottleneck and Measuring Performance of Serverless Computing: A Literature Study",2020,"10.1109/ICRITO48877.2020.9197837","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093103522&doi=10.1109%2fICRITO48877.2020.9197837&partnerID=40&md5=48c6321f04c6fcb41cb3ef2aa12106cd","Trending form of cloud computing is Serverless computing, where developer just needs to focus on his code rather than worrying about server management. In serverless computing, application is nothing but collection of one or more functions, written for specific business functionality, which triggers on an event. There are various cloud service providers, i.e. Amazon, Microsoft, Google, IBM, etc. who provide serverless services, on pay as you use and auto scalable solution to execute the application code as a function. The developer just needs to upload the code for execution. The performance of the serverless computing may vary due to dynamic configuration of the solution, technologies and different technology used by the service provider.This paper reviews various past and recent work in the serverless computing to identify possible bottlenecks and the scope of measuring performance of serverless computing. It will also put some light to leverage machine learning in various possible ways to do performance engineering for future research. © 2020 IEEE." "57193812309;","Towards developing a classification model for water potability in philippine rural areas",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090490193&partnerID=40&md5=e2966654b973e6810a5402f32e6c59ff","In the Philippines, access to safe and sustainable water source is a major problem especially in rural areas. Thus, water monitoring in different water resources has been practiced to ensure safe drinking water. However, manual monitoring of safe drinking water is known to be inconvenient since it requires high operational and transportation costs, and time consuming. This study develops a data-driven water classification model for rural household areas using sensor nodes and machine learning algorithm. Sensor nodes are installed in several water sources in different rural areas to collect water parameters such as pH, turbidity, total dissolved solids, and temperature which are wirelessly transmitted to a base station. The collected sensor data is used to build and train the model to classify water potability using a hard-voting method in ensemble learning. The ensemble learning combined three machine learning algorithms namely k-nearest Neighbor, Naive Bayes, and Classification and Regression Tree. Finally, data are sent to a cloud for data storage and remote monitoring. Results show that the voting classifier model achieves an accuracy of 97% compared with other stand-alone classification algorithms. Furthermore, the model achieves 90% match with conventional industrial laboratory test. © 2020 ASEAN University Network/Southeast Asia Engineering Education Development Network. All right reserved." "36675430500;57164754800;","Enhanced intrusion detection system for cloud environment using machine learning techniques",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088987770&partnerID=40&md5=ff147155d3fcb7f0accc61de20dfcc07","Tremendous growth in Cloud computing environment brings great happiness to the IT sector. Easy access and pay per use are the attractive glimpse towards the cloud environment. Users can’t put out the cloud though there may be many threatening security problems. This is because they tasted the cloud easy deployment anywhere. But the security compliance is not yet attained its goal. Cloud has endless boundaries in the field so there is much crucial vulnerability arising every time.IDS being a very good solution to switch off the vulnerabilities. This paper is going to discuss about the cloud IDS which is tested using the feature selection method and eliminates unwanted attributes to gain the effective ids. The feature selection takes only the essential attributes from the total 42 features to bring enhanced model to suit the Cloud computing. The effective ids is found using NSL KDD dataset applying on the methods that is MLP and J48 algorithm to bring higher accuracy rate to improve the intrusion detection in the Cloud environment. © 2020 Alpha Publishers. All rights reserved." "57212030072;55772930800;","Sentiment analysis of green product using cloud system",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088973077&partnerID=40&md5=aca7e32eca427e242650f17221432c54","A typical manner in which valuable information can be obtained by extracting the sentiment or opinion from any message is called sentiment analysis. The sentiment classification exploits the technologies in machine learning owing to their ability to learn from training data set to predict and support decision making with high accuracy level. Some algorithms do not maintain proper scalability for large datasets. Today, there is a need to deal with some big datasets for involving features in high numbers. The methods of feature selection have been aiming at the elimination of the noisy, the irrelevant or the redundant features that can bring down the performance of classification. Most of the traditional methods lack the capability to be able to cope with the results within a given time. In this work, Term Frequency (TF) is used for feature extraction. The focus is on the Green product opinion mining done using the Information Gain (IG) in feature selection and it is compared with the Group Search Optimization (GSO). This method of feature selection has reduced the original feature sets by means of removing the irrelevant features that enhance the accuracy of classification and bring down the run time of the learning algorithms. The proposed method has been evaluated using the Support Vector Machine (SVM) classifier. The experimental results have proved that the proposed method had achieved better performance. © 2020 Alpha Publishers. All rights reserved." "57204919704;55636320055;56509029800;57216081524;57214837520;","Time-series prediction of sea level change in the east coast of Peninsular Malaysia from the supervised learning approach",2020,"10.18280/ijdne.150314","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087835571&doi=10.18280%2fijdne.150314&partnerID=40&md5=ac5524177d180a9214fb459c723884ef","Analyzing and predicting the rises in sea level are vital elements in oceanography and marine management especially in managing low-lying coastal areas. The present study aims to analyze the ability of machine learning algorithm viz. regression support vector machine (RSVM) in predicting the changes in the sea level on the east coast of Peninsular Malaysia. The selected inputs for the proposed model are monthly mean sea level (MMSL), monthly sea surface temperature (SST), rainfall and mean cloud cover (MCC) for the period from January 2007 to December 2017. A total of 132 data points for each meteorological parameter were used, where 92 (70%) data points from January 2007 to December 2015 were used for training and 40 (30%) data points from January 2016 to December 2017 were used for validating and testing. Results showed based on the correlation coefficient that the model predicts the sea level rises accurately (R= 0.861, 0.825 and 0.857) for Kerteh, Tanjung Sedili, and Tioman Island, respectively. Moreover, the predicted values were similar to the historical tide-gauge data with very low error, which indicates that the proposed RSVM model can be a promising tool for decision-makers and can be reliable to predict monthly mean sea level rises in Malaysia. © 2020 WITPress. All rights reserved." "57211816997;6602927210;","Fine-tuning self-organizing maps for Sentinel-2 imagery: Separating clouds from bright surfaces",2020,"10.3390/rs12121923","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086987137&doi=10.3390%2frs12121923&partnerID=40&md5=0c00f87308837b0a83a14de070f2b572","Removal of cloud interference is a crucial step for the exploitation of the spectral information stored in optical satellite images. Several cloud masking approaches have been developed through time, based on direct interpretation of the spectral and temporal properties of clouds through thresholds. The problem has also been tackled by machine learning methods with artificial neural networks being among the most recent ones. Detection of bright non-cloud objects is one of the most difficult tasks in cloud masking applications since spectral information alone often proves inadequate for their separation from clouds. Scientific attention has recently been redrawn on self-organizing maps (SOMs) because of their unique ability to preserve topologic relations, added to the advantage of faster training time and more interpretative behavior compared to other types of artificial neural networks. This study evaluated a SOM for cloud masking Sentinel-2 images and proposed a fine-tuning methodology to separate clouds from bright land areas. The fine-tuning process which is based on the output of the non-fine-tuned network, at first directly locates the neurons that correspond to the misclassified pixels. Then, the incorrect labels of the neurons are altered without applying further training. The fine-tuning method follows a general procedure, thus its applicability is broad and not confined only in the field of cloud-masking. The network was trained on the largest publicly available spectral database for Sentinel-2 cloud masking applications and was tested on a truly independent database of Sentinel-2 cloud masks. It was evaluated both qualitatively and quantitatively with the interpretation of its behavior through multiple visualization techniques being a main part of the evaluation. It was shown that the fine-tuned SOM successfully recognized the bright non-cloud areas and outperformed the state-of-the-art algorithms: Sen2Cor and Fmask, as well as the version that was not fine-tuned. © 2020 by the authors." "56326408200;55739545700;57209326451;7102410570;","Data reconstruction for remotely sensed Chlorophyll-a concentration in the ross sea using ensemble-based machine learning",2020,"10.3390/rs12111898","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086444937&doi=10.3390%2frs12111898&partnerID=40&md5=7d084e43b8f9402c226913c1b6962746","Polar regions are too harsh to be continuously observed using ocean color (OC) sensors because of various limitations due to low solar elevations, ice effects, peculiar phytoplankton photosynthetic parameters, optical complexity of seawater and persistence of clouds and fog. Therefore, the OC data undergo a quality-control process, eventually accompanied by considerable data loss. We attempted to reconstruct these missing values for chlorophyll-a concentration (CHL) data using a machine-learning technique based on multiple datasets (satellite and reanalysis datasets) in the Ross Sea, Antarctica. This technique-based on an ensemble tree called random forest (RF)-was used for the reconstruction. The performance of the RF model was robust, and the reconstructed CHL data were consistent with satellite measurements. The reconstructed CHL data allowed a high intrinsic resolution of OC to be used without specific techniques (e.g., spatial average). Therefore, we believe that it is possible to study multiple characteristics of phytoplankton dynamics more quantitatively, such as bloom initiation/termination timings and peaks, as well as the variability in time scales of phytoplankton growth. In addition, because the reconstructed CHL showed relatively higher accuracy than satellite observations compared with the in situ data, our product may enable more accurate planktonic research. © 2020 by the authors." "26032307400;57199127093;6603002503;","Identifying corporate sustainability issues by analyzing shareholder resolutions: A machine-learning text analytics approach",2020,"10.3390/su12114753","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086368657&doi=10.3390%2fsu12114753&partnerID=40&md5=bc75d52091851fe8011c32d5ab9f9fa8","Corporations have embraced the idea of corporate environmental, social, and governance (ESG) under the general framework of sustainability. Studies have measured and analyzed the impact of internal sustainability efforts on the performance of individual companies, policies, and projects. This exploratory study attempts to extract useful insight from shareholder sustainability resolutions using machine learning-based text analytics. Prior research has studied corporate sustainability disclosures from public reports. By studying shareholder resolutions, we gain insight into the shareholders' perspectives and objectives. The primary source for this study is the Ceres sustainability shareholder resolution database, with 1737 records spanning 2009-2019. The study utilizes a combination of text analytic approaches (i.e., word cloud, co-occurrence, row-similarities, clustering, classification, etc.) to extract insights. These are novel methods of transforming textual data into useful knowledge about corporate sustainability endeavors. This study demonstrates that stakeholders, such as shareholders, can influence corporate sustainability via resolutions. The incorporation of text analytic techniques offers insight to researchers who study vast collections of unstructured bodies of text, improving the understanding of shareholder resolutions and reaching a wider audience. © 2020 by the authors." "35098748100;57205299839;57216524741;55576700800;56095856700;51461664100;36815873200;8284949000;57214786637;8437626600;","A review of the estimation of downward surface shortwave radiation based on satellite data: Methods, progress and problems",2020,"10.1007/s11430-019-9589-0","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083806719&doi=10.1007%2fs11430-019-9589-0&partnerID=40&md5=922d29aa22cdaad3079ccfa7a6122e9d","The estimation of downward surface shortwave radiation (DSSR) is important for the Earth’s energy budget and climate change studies. This review was organised from the perspectives of satellite sensors, algorithms and future trends, retrospects and summaries of the satellite-based retrieval methods of DSSR that have been developed over the past 10 years. The shortwave radiation reaching the Earth’s surface is affected by both atmospheric and land surface parameters. In recent years, studies have given detailed considerations to the factors which affect DSSR. It is important to improve the retrieval accuracy of cloud microphysical parameters and aerosols and to reduce the uncertainties caused by complex topographies and high-albedo surfaces (such as snow-covered areas) on DSSR estimation. This review classified DSSR retrieval methods into four categories: empirical, parameterisation, look-up table and machine-learning methods, and evaluated their advantages, disadvantages and accuracy. Further efforts are needed to improve the calculation accuracy of atmospheric parameters such as cloud, haze, water vapor and other land surface parameters such as albedo of complex terrain and bright surface, organically combine machine learning and other methods, use the new-generation geostationary satellite and polar orbit satellite data to produce high-resolution DSSR products, and promote the application of radiation products in hydrological and climate models. © 2020, Science China Press and Springer-Verlag GmbH Germany, part of Springer Nature." "56313481400;57146989100;7102653996;36799026400;","Hyper-local, efficient extreme heat projection and analysis using machine learning to augment a hybrid dynamical-statistical downscaling technique",2020,"10.1016/j.uclim.2020.100606","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079842964&doi=10.1016%2fj.uclim.2020.100606&partnerID=40&md5=00c814a5ce2b487119121e119c9ca6f6","This paper describes a scalable system for quantifying hyper-local heat stress in urban environments and its expected response within the changing climate. A hybrid dynamical-statistical downscaling approach links Global Climate Models (GCMs) with dynamically downscaled extreme heat events using the Weather Research and Forecasting model (WRF). Downscaled historical simulations in WRF incorporate urban canopy physics to better describe localized land surface details in the urban environment relevant to extreme heat. This downscaled library is then used in an analog-based approach. This contribution reports a series of enhancements to existing analog-based methods which can efficiently produce more detailed results. The system here uses advanced statistical methods and simple machine learning (ML) techniques to optimize analog selection, perform spatially-consistent bias correction, and decompose patterns of extreme heat into dynamic components such as the land-sea contrast and inland sea-breeze penetration. Hindcast projections are validated against observational data from in-situ weather observing stations. The results demonstrate the scalability and efficiency of this system as it is deployed in cloud-based architectures with parallelized code. Downscaled predictions are equally applicable to heat stress at weather and climate time scales, supporting infrastructure resilience and adaptation, and emergency response. © 2020 Elsevier B.V." "55174951200;36537144100;57210959173;57216249671;57219418376;","Development of an Assistive Device via Smart Glasses",2020,"10.1109/ECBIOS50299.2020.9203629","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85092632906&doi=10.1109%2fECBIOS50299.2020.9203629&partnerID=40&md5=a5f71bbe1bba1a16e01b6c46760d7b7b","This study used smart glasses as the carrier to develop three applications aimed at the elderly. Taiwan entered an elderly society in 2018, with the proportion of the elderly population exceeding 14%. It is estimated that the elderly population will exceed 20% in 2026, and Taiwan will enter an ultra-elderly society. To improve the quality of life of the elderly and attach importance to the health care of the elderly, more and more scientific and technological products are used to improve the life and consumption patterns of the elderly, thereby promoting the popularization of smart wearable devices and sensing technologies. In view of the purpose, this research develops related functional applications with smart glasses. First, the memory recall mechanism is carried out based on face recognition. When the user uses the function, an embedded camera in glasses is turned on to capture the target personnel. Afterwards, a KNN (K Nearest Neighbor) machine learning approach is applied to identify the target, and the personal data is performed in the glasses through searching on the developed cloud database. Besides, the user can add, delete and modify the personal data through the computer, which the entrance is carried out by scanning the two-dimension code shown in the computer program with glasses for security purpose. Secondly, the image enhancement function shows that the captured texts can be expanded and enlarged in the projection monitor of glasses. Finally, the construction of voice commands is performed through the microphone within the suite of smart glasses. The results show that the accuracy rate of KNN face recognition is 93.3%, which can be applied to general life situations. © 2020 IEEE." "57218116828;57218283904;57218120506;","Mobile mapping, machine learning and digital twin for road infrastructure monitoring and maintenance: Case study of mohammed VI bridge in Morocco",2020,"10.1109/Morgeo49228.2020.9121882","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087967317&doi=10.1109%2fMorgeo49228.2020.9121882&partnerID=40&md5=f64c9686dd6e9bcc47bedf0562162dde","The concepts of Digital Twin has been recently introduced, it refers to functional connections between a complex physical system and its high-fidelity digital replica. Digital Twin process workflow is proposed in case of Mohammed VI Bridge modeling in Morocco. The current maintenance of a road infrastructure is based on a manual inspection and a system based on traditional tools. Aging infrastructures require a new approach to maintenance in terms of inspection, bridge maintenance system, simulation and systematic evaluation. This system now exists and is called the Digital Twin. Digital Twin can be thought of as a virtual prototype in service that changes dynamically in near real time as its physical twin changes. An urban infrastructure digital twin is a virtual instance of his physical twin that is continuously updated with multisource, multisensor and multitemporal data that can be used for monitoring, simulating and forecasting any potential problem that may appear in the structure and proposing planning for repair and maintenance of health status throughout the life cycle of this infrastructure. This work presents a general vision and a justification for integrating DT technology with geospatial data. The paper examines the benefits of integrating 3D GIS data acquired by automated mobile mapping (MMS) workflows for modeling the reality of a major bridge infrastructure in Morocco. This allowed to study the future performance of this bridge structure on virtual twin structures under different environmental conditions. Cloud point data are acquired by a Mobile Mapping System on Mohammed VI Bridge and converted in BIM model by a scan to BIM process and is integrated in a GIS and BIM virtual environment and shows the efficiency of volumetric auscultation in terms of surface flatness and distortion inspection. This project provides a new bridge maintenance system using the concept of a Digital Twin. This digital model is a platform that allows to collect, organize and share the maintenance history of this important road infrastructure in Morocco. © 2020 IEEE." "36615943200;57218539795;35180845800;15031611500;37046365700;56477626600;56372294100;57196301839;14060679800;","Searching for Molecular Outflows with Support Vector Machines: The Dark Cloud Complex in Cygnus",2020,"10.3847/1538-4365/ab879a","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087208421&doi=10.3847%2f1538-4365%2fab879a&partnerID=40&md5=2afa08c0212e32f3e08e2e163b0697d7","We present a survey of molecular outflows across the dark cloud complex in the Cygnus region, based on a 46.75 deg2 field of CO isotopologue data from the Milky Way Imaging Scroll Painting survey. A supervised machine-learning algorithm, the support vector machine, is introduced to accelerate our visual assessment of outflow features in the data cube of 12CO and 13CO J = 1-0 emission. A total of 130 outflow candidates are identified, 77 of which show bipolar structures and 118 are new detections. Spatially, these outflows are located inside dense molecular clouds, and some of them are found in clusters or in elongated linear structures tracing the underlying gas filament morphology. Along the line of sight, 97, 31, and 2 candidates reside in the Local, Perseus, and Outer Arms, respectively. Young stellar objects as outflow drivers are found near most outflows, while 36 candidates show no associated source. The clusters of outflows that we detect are inhomogeneous in their properties; nevertheless, we show that the outflows cannot inject turbulent energy on cloud scales. Instead, at best, they are restricted to affecting the so-called ""clump""and ""core""scales, and only on short (∼0.3 Myr) estimated timescales. Combined with outflow samples in the literature, our work shows a tight outflow mass-size correlation. © 2020. The American Astronomical Society. All rights reserved." "56094876500;14821958500;35581315600;57208478400;57216672036;7006392180;","Maximum fraction images derived from year-based project for on-board autonomy-vegetation (PROBA-V) data for the rapid assessment of land use and land cover areas in Mato Grosso state, Brazil",2020,"10.3390/LAND9050139","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085590974&doi=10.3390%2fLAND9050139&partnerID=40&md5=a298842ae08e04076da51a961ace5866","This paper presents a new approach for rapidly assessing the extent of land use and land cover (LULC) areas in Mato Grosso state, Brazil. The novel idea is the use of an annual time series of fraction images derived from the linear spectral mixing model (LSMM) instead of original bands. The LSMM was applied to the Project for On-Board Autonomy-Vegetation (PROBA-V) 100-m data composites from 2015 (~73 scenes/year, cloud-free images, in theory), generating vegetation, soil, and shade fraction images. These fraction images highlight the LULC components inside the pixels. The other new idea is to reduce these time series to only six single bands representing the maximum and standard deviation values of these fraction images in an annual composite, reducing the volume of data to classify the main LULC classes. The whole image classification process was conducted in the Google Earth Engine platform using the pixel-based random forest algorithm. A set of 622 samples of each LULC class was collected by visual inspection of PROBA-V and Landsat-8 Operational Land Imager (OLI) images and divided into training and validation datasets. The performance of the method was evaluated by the overall accuracy and confusion matrix. The overall accuracy was 92.4%, with the lowest misclassification found for cropland and forestland (<9% error). The same validation data set showed 88% agreement with the LULC map made available by the Landsat-based MapBiomas project. This proposed method has the potential to be used operationally to accurately map the main LULC areas and to rapidly use the PROBA-V dataset at regional or national levels. © 2020 by the authors." "57006249500;36803116800;57211112377;57201257574;8593424100;57200294885;","Integration of multi-sensor data to estimate plot-level stem volume using machine learning algorithms-case study of evergreen conifer planted forests in Japan",2020,"10.3390/rs12101649","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085577942&doi=10.3390%2frs12101649&partnerID=40&md5=b6e6c0ada259f8e19c630b9cd63def67","The development of new methods for estimating precise forest structure parameters is essential for the quantitative evaluation of forest resources. Conventional use of satellite image data, increasing use of terrestrial laser scanning (TLS), and emerging trends in the use of unmanned aerial systems (UASs) highlight the importance of modern technologies in the realm of forest observation. Each technology has different advantages, and this work seeks to incorporate multiple satellite, TLS-and UAS-based remote sensing data sets to improve the ability to estimate forest structure parameters. In this paper, two regression analysis approaches are considered for the estimation: random forest regression (RFR) and support vector regression (SVR). To collect the dependent variable, in situ measurements of individual tree parameters (tree height and diameter at breast height (DBH) were taken in a Japanese cypress forest using the nondestructive TLS method, which scans the forest to obtain dense and accurate point clouds under the tree canopy. Based on the TLS data, the stem volume was then computed and treated as ground truth information. Topographic and UAS information was then used to calculate various remotely sensed explanatory variables, such as canopy size, canopy cover, and tree height. Canopy cover and canopy shapes were computed via the orthoimages derived from the UAS and watershed segmentation method, respectively. Tree height was computed by combining the digital surface model (DSM) from the UAS and the digital terrain model (DTM) from the TLS data. Topographic variables were computed from the DTM. The backscattering intensity in the satellite imagery was obtained based on L-band (Advanced Land Observing Satellite-2 (ALOS-2) Phased Array type L-band Synthetic Aperture Radar-2 (PALSAR-2) and C-band (Sentinel-1) synthetic aperture radar (SAR). All satellite (10-25 m resolution), TLS (3.4 mm resolution) and UAS (2.3-4.6 cm resolution) data were then combined, and RFR and SVR were trained; the resulting predictive powers were then compared. The RFR method yielded fitting R2 up to 0.665 and RMSE up to 66.87 m3/ha (rRMSE = 11.95%) depending on the input variables (best result with canopy height, canopy size, canopy cover, and Sentinel-1 data), and the SVR method showed fitting R2 up to 0.519 and RMSE up to 80.12 m3/ha (rRMSE = 12.67%). The RFR outperformed the SVR method, which could delineate the relationship between the variables for better model accuracy. This work has demonstrated that incorporating various remote sensing data to satellite data, especially adding finer resolution data, can provide good estimates of forest parameters at a plot level (10 by 10 m), potentially allowing advancements in precision forestry. © 2020 by the authors." "57194068533;7006499360;6603916527;7202643131;","Sentinel-1 observation frequency significantly increases burnt area detectability in tropical SE Asia",2020,"10.1088/1748-9326/ab7765","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085275425&doi=10.1088%2f1748-9326%2fab7765&partnerID=40&md5=78c6cfc1ee28121ba469faa080162086","Frequent cloud cover in the tropics significantly affects the observation of the surface by satellites. This has enormous implications for current approaches that estimate greenhouse gas (GHG) emissions from fires or map fire scars. These mainly employ data acquired in the visible to middle infrared bands to map fire scars or thermal data to estimate fire radiative power and consequently derive emissions. The analysis here instead explores the use of microwave data from the operational Sentinel-1A (S-1A) in dual-polarisation mode (VV and VH) acquired over Central Kalimantan during the 2015 fire season. Burnt areas were mapped in three consecutive periods between August and October 2015 using the random forests machine learning algorithm. In each mapping period, the omission and commission errors of the unburnt class were always below 3%, while the omission and commission errors of the burnt class were below 20% and 5% respectively. Summing the detections from the three periods gave a total burnt area of ∼1.6 million ha, but this dropped to ∼1.2 million ha if using only a pair of pre- and post-fire season S-1A images. Hence the ability of Sentinel-1 to make frequent observations significantly increases fire scar detection. Comparison with burnt area estimates from the Moderate Resolution Imaging Spectroradiometer (MODIS) burnt area product at 5 km scale showed poor agreement, with consistently much lower estimates produced by the MODIS data-on average 14%-51% of those obtained in this study. The method presented in this study offers a way to reduce the substantial errors likely to occur in optical-based estimates of GHG emissions from fires in tropical areas affected by substantial cloud cover. © 2020 The Author(s). Published by IOP Publishing Ltd." "57215057713;37089364100;57215056447;47962102500;37088003400;55626446400;","Improvement of the Rapid-Development Thunderstorm (RDT) Algorithm for Use with the GK2A Satellite",2020,"10.1007/s13143-020-00182-6","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079799174&doi=10.1007%2fs13143-020-00182-6&partnerID=40&md5=b6dabc24aafc3aaef3f706875d78a457","New technologies for the classification of convective cloud lifecycles and the prediction of their movements are needed to detect severe convective weather and to support objective cloud guidance. Satellites enable earlier detection of severe weather over larger coverage areas than ground-based observations or radar. The use of satellite observations for nowcasting is thus likely. In this study, convective initiation (CI) data are paired with a modified rapid-development thunderstorm (RDT) algorithm for the analysis of new data from the Geostationary Korea Multi-Purpose Satellite-2A (GEO-KOMPSAT-2A, GK2A). The RDT algorithm is further modified to accommodate the additional GK2A satellite channels, and new satellite data are used to continuously analyze thunderstorms associated with severe weather in Korea. The logistic regression (LR) machine learning approach is used to optimize the criteria of interest fields and weighting coefficients of the RDT algorithm for convective detection. In addition, auxiliary data (cloud type, convective rainfall rate, and cloud top temperature/height) calculated from RDT sub-module is replaced with GK2A derived products. The fully modified RDT algorithm (K-RDT) is quantitatively verified using lightning data from summer convection cases. The probability of detection (POD) for convective clouds is increased by 30–40%, and the threat score (TS) for average lightning activity is improved by 10–30%. The channel properties of Japan Himawari-8 satellite are similar to those of the GK2A satellite. Due to the lack of GK2A satellite data during the development period, CI data from the Himawari-8 satellite are used as proxies. © 2020, Korean Meteorological Society and Springer Nature B.V." "57216206627;","Reaction: Can We Grow a Quantum Processor?",2020,"10.1016/j.chempr.2020.03.018","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082831288&doi=10.1016%2fj.chempr.2020.03.018&partnerID=40&md5=9b352f6a55423ef6b2a37d05df410c13","Alexei Marchenkov is an experimental physicist who has devoted most of his research career to studying quantum phenomena in materials and structures at cryogenic temperatures, including graphene, atomic-sized wires and contacts, and superconducting quantum bits and circuits. He is a former associate professor at the Georgia Institute of Technology and founder and chief executive officer of the Bleximo Corp., a startup developing application-specific superconducting quantum computers. © 2020 Elsevier Inc.Today's quantum computers contain tens of qubits and are readily available for experimentation in the cloud. However, building quantum computers capable of outperforming digital supercomputers in practical simulation, optimization, and machine-learning tasks requires more coherent qubits assembled in larger processors. Although molecular electron-spin qubits have shown early promise, validating whether they can provide a viable platform for practical quantum processors and compete with the leading superconducting and trapped-ion modalities requires a significant experimental effort. © 2020 Elsevier Inc." "57117012400;57216413349;57193791654;57211840394;","The potential of integrating blockchain technology into smart sustainable city development",2020,"10.1088/1755-1315/463/1/012020","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083428106&doi=10.1088%2f1755-1315%2f463%2f1%2f012020&partnerID=40&md5=51493834ae51217e05553b3da09f1314","The rise of global urbanisation has led to massive pressures on resources such as food, water, infrastructure, and energy demand to support growing populations. It brings adverse impacts on the liveable condition and economic growth of a country if this problem remains unsolved. Smart city is a potential solution to address the challenges of urbanisation by leveraging the technological breakthrough such as internet of things (IoT), Artificial Intelligence (AI), machine learning, big data, and cloud computing to facilitate scarce resources planning and management. With numerous connected devices and vast communication networks, it poses a challenges of security threat which cannot be addressed by the conventional cybersecurity solutions. Blockchain offers a solution in securing the huge numbers of connected devices in smart city network. The application of blockchain technology is leading in the banking and financial industry. However, the uses and implementations in smart city have emerged in recent years. The combination of blockchain technology and smart city has offered a great potential for sustainable development. Thus, it is imperative to discuss the potential of these two elements in making the city safer and sustainable. This paper explores how the blockchain technology application can help in managing smart city and achieve sustainability. The findings revealed that there are five key areas of blockchain application in smart city which are smart governance, smart mobility, smart asset, smart utility and smart logistic. A framework for smart sustainable city with blockchain technology is presented as an outcome of this study. It gives a clear overview for the policy makers and regulators of how blockchain supports within smart city framework. It facilitates the transition towards smart and sustainable cities through the use of blockchain. © 2020 Institute of Physics Publishing. All rights reserved." "55811505000;56589932000;57148539400;57211159673;","An optimal machine learning classifier for cloud based web URL phishing detection",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088855154&partnerID=40&md5=201849c9b90764b0fba7965940c306d6","Cloud based Web URL phishing is a crucial threat in the cyber world a decade before and even till now. An enormous web server instance is created in the cloud platform and hosted online. Phishing is an act of impersonating to achieve sensitive information from the intended recipient by deceiving them. The users access information through various mediums such as websites, emails, mobile applications, SMS, and social media. Phishing is demonstrated using a malicious URL that the attacker replicates from the original website and sends it to the target or by using a mail that injects malware in the target’s computer to obtain sensitive information. Malicious URL is detected using machine learning techniques that efficiently identifies a malicious website or a new website. Random Forest's anti-phishing classifier has outperformed to accomplish a genuine positive rate as 98.42% and false positive rate as 1.58% compared to all other ML classifiers. © 2020 Alpha Publishers." "57203397955;57216490159;57218081287;","An Efficient Solution for Semantic Segmentation of Three Ground-based Cloud Datasets",2020,"10.1029/2019EA001040","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083636404&doi=10.1029%2f2019EA001040&partnerID=40&md5=96ab039a3967ec6569507b3fe7d657e5","The machine learning approach has shown its state-of-the-art ability to handle segmentation and detection tasks. It is increasingly employed to extract patterns and spatiotemporal features from the ever-increasing stream of Earth system data. However, there is still a significant challenge, which is the generalization capability of the model on cloud images in different types and weather conditions. After studying several popular methods, we propose a semantic segmentation neural network for cloud segmentation. It extracts features learned by source and target domains in an end-to-end behavior, which can address the problem of significant lack of labels in the observed cloud image data. It is further evaluated on the Singapore Whole Sky Image Segmentation (SWIMSEG) dataset by using Mean Intersection-over-Union, recall, F-score, and accuracy matrices. The scores of these matrices are 86%, 97%, 92%, and 96%, which prove that it has excellent efficiency and robustness. Most importantly, a new benchmark based on the SWIMSEG dataset for the task of cloud segmentation is introduced. The others, BENCHMARK, Cirrus Cumulus Stratus Nimbus are evaluated through the model trained from the SWIMSEG dataset by way of visualization. ©2020. The Authors." "55941476900;57014496500;57197784493;55612256200;","Snow cover estimation from MODIS and Sentinel-1 SAR data using machine learning algorithms in the western part of the Tianshan Mountains",2020,"10.1007/s11629-019-5723-1","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083324480&doi=10.1007%2fs11629-019-5723-1&partnerID=40&md5=de77ff3430d69bad6df3d3b4017216cc","Obtaining the spatial distribution of snow cover in mountainous areas using the optical image of remote sensing technology is difficult because of cloud and fog. In this study, the object-based principle component analysis-support vector machine (PCA-SVM) method is proposed for snow cover mapping through the integration of moderateresolution imaging spectroradiometer (MODIS) snow cover products and the Sentinel-1 synthetic aperture radar (SAR) scattering characteristics. First, derived from the Sentinel-1A SAR images, the feature parameters, including VV/VH backscatter, scattering entropy, and scattering alpha, were used to describe the variations of snow and non-snow covers. Second, the optimum feature combinations of snow cover were formed from the feature parameters using the principle component analysis (PCA) algorithm. Finally, using the optimum feature combinations, a snow cover map with a 20 m spatial resolution was extracted by means of an object-based SVM classifier. This method was applied in the study area of the Xinyuan County, which is located in the western part of the Tianshan Mountains in Xinjiang, China. The accuracies in this method were analyzed according to the data observed at different experimental sites. Results showed that the snow cover pixels of the extraction were less than those in the actual situation (FB1=93.86, FB2=59.78). The evaluation of the threat score (TS), probability of detection (POD), and false alarm ratio (FAR) for the snow-covered pixels obtained from the two-stage SAR images were different (TS1=86.84, POD1=90.10, FAR1=4.01; TS2=56.40, POD2=57.62, FAR2=3.62). False and misclassifications of the snow cover and non-snow cover pixels were found. Although the classifications were not highly accurate, the approach showed potential for integrating different sources to retrieve the spatial distribution of snow covers during a stable period. © 2020, Science Press, Institute of Mountain Hazards and Environment, CAS and Springer-Verlag GmbH Germany, part of Springer Nature." "57202642788;7202994044;56333692000;57192701006;57202917893;","A Three-Dimensional Deep Learning Framework for Human Behavior Analysis Using Range-Doppler Time Points",2020,"10.1109/LGRS.2019.2930636","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082883574&doi=10.1109%2fLGRS.2019.2930636&partnerID=40&md5=d598b8eec8031fdc8f2c1d3119c3ec99","Deep neural networks have shown promise in the radar-based human activity analysis application. Different from existing deep learning models that take either micro-Doppler spectrograms or range profiles as their input, the proposed method can process micromotion signatures in a 3-D way. In this letter, we first transform radar echoes into range-Doppler (RD) time points and then directly process the point sets via a designed 3-D network called the RD PointNet. In fact, our point model is a discrete representation of the motion trajectory. Through this quantitative model, we can use the 3-D network to simultaneously capture human motion profiles and temporal variations. The motion capture simulations and ultrawideband radar measurements show that the proposed framework can achieve superior classification accuracy and noise robustness when compared with image-based methods. © 2004-2012 IEEE." "57215037375;36443824600;49661020300;57212631853;55738672900;56026541800;56168282700;","Climate-based approach for modeling the distribution of montane forest vegetation in Taiwan",2020,"10.1111/avsc.12485","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85081245150&doi=10.1111%2favsc.12485&partnerID=40&md5=091dcef2a352bcb71f80ce5eabdd3024","Aims: Climate shapes forest types on our planet and also drives the differentiation of zonal vegetation at regional scale. A climate-based ecological model may provide an effective alternative to the traditional approach for assessing limitations, thresholds, and the potential distribution of forests. The main objective of this study is to develop such a model, with a machine-learning approach based on scale-free climate variable estimates and classified vegetation plots, to generate a fine-scale predicted vegetation map of Taiwan, a subtropical mountainous island. Location: Taiwan. Methods: A total of 3,824 plots from 13 climate-related forest types and 57 climatic variable estimates for each plot were used to build an individual ecological niche model for each forest type with random forest (RF). A predicted vegetation map was developed through the assemblage of RF predictions for each forest type at the spatial resolution of 100 m. The accuracy of the ensemble RF model was evaluated by comparing the predicted forest type with its original classification by plot. Results: The climate environment of regions higher than 100 m above sea level in Taiwan was classified into potential habitats of 13 forest types by using model predictions. The predicted vegetation map displays a distinct altitudinal zonation from subalpine to montane cloud forests, followed by the latitudinal differentiation of subtropical mountain forests in the north and tropical montane forests in the south, with an average mismatch rate of 6.59%. An elevational profile and 3D visualization demonstrate the excellence of the model in estimating a fine, precise, and topographically corresponding potential distribution of forests. Conclusions: The machine-learning approach is effective for handling a large number of variables and to provide accurate predictions. This study provides a statistical procedure integrating two sources of training data: (a) the locations of field sampling plots; and (b) their corresponding climate variable estimates, to predict the potential distribution of climate-related forests. © 2020 International Association for Vegetation Science" "57216035656;57189929871;","Impact of Hurricane Maria on Beach Erosion in Puerto Rico: Remote Sensing and Causal Inference",2020,"10.1029/2020GL087306","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082532550&doi=10.1029%2f2020GL087306&partnerID=40&md5=398ac8ebf167235c5d24785bf37fbbdb","Beach erosion due to large storms critically affects coastal vulnerability, but is challenging to monitor and quantify. Attributing erosion to a specific storm requires a reliable counterfactual scenario: hypothetical beach conditions, absent the storm. Calibrating models to construct counterfactuals requires numerous observations that are rarely available. Storm paths are unpredictable, making long-term instrumentation of specific beaches costly. Optical remote sensing is hampered by persistent cloud cover. We use Sentinel-1 satellite radar imagery to monitor shoreline changes through clouds and propose regression discontinuity as a strategy to estimate the causal effect of large storms on beach erosion. Applied to 75 beaches across Puerto Rico, the approach detects shoreline changes with a root-mean-square error comparable to the resolution of the imagery. Hurricane Maria caused an erosion of 3 to 5 m along its path, up to 40 m at particular beaches. Results reveal strong local disparities that are consistent with simulated nearshore hydrodynamic conditions. ©2020. American Geophysical Union. All Rights Reserved." "57216129885;","Research on the Application of Artificial Intelligence in Energy Science and Engineering Monitoring Software Engineering Technology under the Background of Big Data",2020,"10.1088/1755-1315/440/3/032058","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082627416&doi=10.1088%2f1755-1315%2f440%2f3%2f032058&partnerID=40&md5=a1f26bac185a5867810e292641832fa4","Artificial intelligence belongs to the world's leading technology and technology, which can make people's work and life become more intelligent and faster. The arrival of the era of big data accelerates the development of artificial intelligence technology, and the artificial intelligence big data platform begins to appear. Through the analysis of the relationship between big data and artificial intelligence, the AI big data platform and its application are comprehensively discussed, aiming to promote the development of artificial intelligence big data platform and improve the overall level of domestic artificial intelligence. The smart grid is artificial. One of the important application areas of intelligence, the advancement and breakthrough of the new generation of artificial intelligence technology, which is mainly represented by advanced machine learning theory, big data and cloud computing, will promote the development of smart grid. Therefore, an energy science and engineering data monitoring software is needed to realize the monitoring and analysis and unified scheduling of energy science. The article details the design and implementation of data monitoring software for power and energy intelligence. © 2020 Published under licence by IOP Publishing Ltd." "57212588620;57005349200;15846943000;57212587521;57212585560;","Point-cloud detection of buildings based on a latent Dirichlet allocation model with waveform data",2020,"10.1080/2150704X.2019.1706006","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077157679&doi=10.1080%2f2150704X.2019.1706006&partnerID=40&md5=0ede6552b6aa57c4b5daf7c97aa1ecd6","Light detection and ranging are important methods for acquiring digital surface models and can be used to extract building data. Point-cloud detection of buildings is a prerequisite for the model-based expression of buildings. Existing methods are insufficient because of their abstractness of feature extraction and poor accuracy of the detection results. This paper proposes a method for the point-cloud detection of buildings based on a latent Dirichlet allocation (LDA) model with waveform data. This method can extract waveform data via the global convergence Levenberg Marquard algorithm, convert discrete point clouds into point-cluster objects via super voxel segmentation, and detect the point clouds of buildings via the LDA model. Moreover, it supports vector machine classification. Experimental results demonstrate that waveform features and the LDA model both improve the accuracy of building detection. In addition, this method is less susceptible to variations in feature dimensions and is robust in terms of the number of topics and words. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group." "57219195975;57193749143;","Estimating Crop Yields with Remote Sensing and Deep Learning",2020,"10.1109/LAGIRS48042.2020.9165608","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091676099&doi=10.1109%2fLAGIRS48042.2020.9165608&partnerID=40&md5=bc587f96d7c10752b3d83087d268f4ba","Increasing the accuracy of crop yield estimates may allow improvements in the whole crop production chain, allowing farmers to better plan for harvest, and for insurers to better understand risks of production, to name a few advantages. To perform their predictions, most current machine learning models use NDVI data, which can be hard to use, due to the presence of clouds and their shadows in acquired images, and due to the absence of reliable crop masks for large areas, especially in developing countries. In this paper, we present a deep learning model able to perform pre-season and in-season predictions for five different crops. Our model uses crop calendars, easy-to-obtain remote sensing data and weather forecast information to provide accurate yield estimates. © 2020 IEEE." "56494144900;57219200521;6602269604;36623278200;36136174500;","Brazildam: A Benchmark Dataset for Tailings Dam Detection",2020,"10.1109/LAGIRS48042.2020.9165620","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091634482&doi=10.1109%2fLAGIRS48042.2020.9165620&partnerID=40&md5=97f530e672ba28f5f148d1939522ae9d","In this work we present BrazilDAM, a novel public dataset based on Sentinel-2 and Landsat-8 satellite images covering all tailings dams cataloged by the Brazilian National Mining Agency (ANM). The dataset was built using georeferenced images from 769 dams, recorded between 2016 and 2019. The time series were processed in order to produce cloud free images. The dams contain mining waste from different ore categories and have highly varying shapes, areas and volumes, making BrazilDAM particularly interesting and challenging to be used in machine learning benchmarks. The original catalog contains, besides the dam coordinates, information about: the main ore, constructive method, risk category, and associated potential damage. To evaluate BrazilDAM's predictive potential we performed classification essays using state-of-the-art deep Convolutional Neural Network (CNNs). In the experiments, we achieved an average classification accuracy of 94.11% in tailing dam binary classification task. In addition, others four setups of experiments were made using the complementary information from the original catalog, exhaustively exploiting the capacity of the proposed dataset. © 2020 IEEE." "57198861721;7006011851;37023334200;","Characterizing lognormal fractional-Brownian-motion density fields with a convolutional neural network",2020,"10.1093/mnras/staa122","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091418077&doi=10.1093%2fmnras%2fstaa122&partnerID=40&md5=8e2e9b830de56de95eda55fca52d2044","In attempting to quantify statistically the density structure of the interstellar medium, astronomers have considered a variety of fractal models. Here, we argue that, to properly characterize a fractal model, one needs to define precisely the algorithm used to generate the density field, and to specify – at least – three parameters: one parameter constrains the spatial structure of the field, one parameter constrains the density contrast between structures on different scales, and one parameter constrains the dynamic range of spatial scales over which self-similarity is expected (either due to physical considerations, or due to the limitations of the observational or numerical technique generating the input data). A realistic fractal field must also be noisy and non-periodic. We illustrate this with the exponentiated fractional Brownian motion (xfBm) algorithm, which is popular because it delivers an approximately lognormal density field, and for which the three parameters are, respectively, the power spectrum exponent, β, the exponentiating factor, S, and the dynamic range, R. We then explore and compare two approaches that might be used to estimate these parameters: machine learning and the established ∆-Variance procedure. We show that for 2 ≤ β ≤ 4 and 0 ≤ S ≤ 3, a suitably trained Convolutional Neural Network is able to estimate objectively both β (with root-mean-square error ∊β ∼ 0.12) and S (with ∊S ∼ 0.29). ∆-variance is also able to estimate β, albeit with a somewhat larger error (∊β ∼ 0.17) and with some human intervention, but is not able to estimate S. © 2020 The Author(s)" "6506328135;34881780600;7005446873;6701754792;","A Machine Learning Assisted Development of a Model for the Populations of Convective and Stratiform Clouds",2020,"10.1029/2019MS001798","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083242186&doi=10.1029%2f2019MS001798&partnerID=40&md5=29d46bf7c828c63bc3cf2fea84e4afd2","Traditional parameterizations of the interaction between convection and the environment have relied on an assumption that the slowly varying large-scale environment is in statistical equilibrium with a large number of small and short-lived convective clouds. They fail to capture nonequilibrium transitions such as the diurnal cycle and the formation of mesoscale convective systems as well as observed precipitation statistics and extremes. Informed by analysis of radar observations, cloud-permitting model simulation, theory, and machine learning, this work presents a new stochastic cloud population dynamics model for characterizing the interactions between convective and stratiform clouds, with the goal of informing the representation of these interactions in global climate models. Fifteen wet seasons of precipitating cloud observations by a C-band radar at Darwin, Australia are fed into a machine learning algorithm to obtain transition functions that close a set of coupled equations relating large-scale forcing, mass flux, the convective cell size distribution, and the stratiform area. Under realistic large-scale forcing, the derived transition functions show that, on the one hand, interactions with stratiform clouds act to dampen the variability in the size and number of convective cells and therefore in the convective mass flux. On the other, for a given convective area fraction, a larger number of smaller cells is more favorable for the growth of stratiform area than a smaller number of larger cells. The combination of these two factors gives rise to solutions with a few convective cells embedded in a large stratiform area, reminiscent of mesoscale convective systems. ©2020. The Authors." "57200602927;57191057256;15034793900;56233739200;56579557400;55640064500;57214086717;57203541311;","Learning sequential slice representation with an attention-embedding network for 3D shape recognition and retrieval in MLS point clouds",2020,"10.1016/j.isprsjprs.2020.01.003","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85078134393&doi=10.1016%2fj.isprsjprs.2020.01.003&partnerID=40&md5=5f832be3a7282de12eccc8d6b31782c2","The representation of 3D data is the key issue for shape analysis. However, most of the existing representations suffer from high computational cost and structure information loss. This paper presents a novel sequential slice representation with an attention-embedding network, named RSSNet, for 3D point cloud recognition and retrieval in road environments. RSSNet has two main branches. Firstly, a sequential slice module is designed to map disordered 3D point clouds to ordered sequence of shallow feature vectors. A gated recurrent unit (GRU) module is applied to encode the spatial and content information of these sequential vectors. The second branch consists of a key-point based graph convolution network (GCN) with an embedding attention strategy to fuse the sequential and global features to refine the structure discriminability. Three datasets were used to evaluate the proposed method, one acquired by our mobile laser scanning (MLS) system and two public datasets (KITTI and Sydney Urban Objects). Experimental results indicated that the proposed method achieved better performance than recognition and retrieval state-of-the-art methods. RSSNet provided recognition rates of 98.08%, 95.77% and 70.83% for the above three datasets, respectively. For the retrieval task, RSSNet obtained excellent mAP values of 95.56%, 87.16% and 69.99% on three datasets, respectively. © 2020 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)" "6602981892;","Geospatial Artificial Intelligence: Potentials of Machine Learning for 3D Point Clouds and Geospatial Digital Twins",2020,"10.1007/s41064-020-00102-3","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85080980476&doi=10.1007%2fs41064-020-00102-3&partnerID=40&md5=423f399f054e2449ed47dcf3b3c110c1","Artificial intelligence (AI) is changing fundamentally the way how IT solutions are implemented and operated across all application domains, including the geospatial domain. This contribution outlines AI-based techniques for 3D point clouds and geospatial digital twins as generic components of geospatial AI. First, we briefly reflect on the term “AI” and outline technology developments needed to apply AI to IT solutions, seen from a software engineering perspective. Next, we characterize 3D point clouds as key category of geodata and their role for creating the basis for geospatial digital twins; we explain the feasibility of machine learning (ML) and deep learning (DL) approaches for 3D point clouds. In particular, we argue that 3D point clouds can be seen as a corpus with similar properties as natural language corpora and formulate a “Naturalness Hypothesis” for 3D point clouds. In the main part, we introduce a workflow for interpreting 3D point clouds based on ML/DL approaches that derive domain-specific and application-specific semantics for 3D point clouds without having to create explicit spatial 3D models or explicit rule sets. Finally, examples are shown how ML/DL enables us to efficiently build and maintain base data for geospatial digital twins such as virtual 3D city models, indoor models, or building information models. © 2020, The Author(s)." "56099899900;56943040200;57219173203;57219185242;7404338756;","Data-driven fault diagnosis of industrial robots with a cloud computing framework",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091572258&partnerID=40&md5=6d54aee9433a5ea1e1e2d844806ea760","In modern manufacturing industry, industrial robots have been widely used. Health status of the robots should be always monitored to prevent a sudden shutdown of manufacturing lines. Supervising the signals measured from the industrial robot and diagnosing the status of machines in real time are essential tasks for us to manage the manufacturing lines. In this work, we developed a system for data-driven fault diagnosis of industrial robots, which includes a data acquisition and mining process, a machining learning process, and a cloud computing framework. The signals gathered from attached sensors on the robot are stored in a database within the framework. Structured data are extracted from the stored raw signals. The most important features are selected from the structured data for preventing the overfitting problems in the machine learning process. The fault diagnosis models are trained based on several machine learning algorithms and selected features. Finally, the fault diagnosis results are monitored by operators using mobile devices in real time. All monitoring and diagnosing processes including signal processing, feature extraction, feature selection, and diagnosis operate in the server of our cloud computing framework. Copyright © Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology, EUSPEN 2020. All rights reserved." "57211095441;57217271715;57213186458;57218919599;57193959504;55817600100;","Impact of Satellite Sounding Data on Virtual Visible Imagery Generation Using Conditional Generative Adversarial Network",2020,"10.1109/JSTARS.2020.3013598","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090798696&doi=10.1109%2fJSTARS.2020.3013598&partnerID=40&md5=bea9a3e9113e0d6aa18712e58991c4ab","The visible band of satellite sensors is of limited use during the night due to a lack of solar reflection. This study presents an improved conditional generative adversarial networks (CGANs) model to generate virtual nighttime visible imagery using infrared (IR) multiband satellite observations and the brightness temperature difference between the two IR bands in the communication, ocean, and meteorological satellite. For the summer daytime case study with visible band imagery, our multiband CGAN model showed better statistical results [correlation coefficient (CC) = 0.952, bias = -1.752 (in a digital number (DN) unit from 0 to 255, converted from reflectance from 0 to 1), and root-mean-square-error (RMSE) = 26.851 DN] than the single-band CGAN model using a pair of visible and IR bands (CC = 0.916, bias = -4.073 DN, and RMSE = 35.349 DN). The proposed multiband CGAN model performed better than the single-band CGAN model, particularly, in convective clouds and typhoons, because of the sounding effects from the water vapor band. In addition, our multiband CGAN model provided detailed patterns for clouds and typhoons at twilight. Therefore, our results could be used for visible-based nighttime weather analysis of convective clouds and typhoons, using data from next-generation geostationary meteorological satellites. © 2008-2012 IEEE." "57211499439;35849722200;55914518400;57198885698;","CDL: A Cloud Detection Algorithm over Land for MWHS-2 Based on the Gradient Boosting Decision Tree",2020,"10.1109/JSTARS.2020.3014136","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090762033&doi=10.1109%2fJSTARS.2020.3014136&partnerID=40&md5=e7a77ccbba5ff99cee7251d3e72a49d5","This article presents a standalone cloud detection algorithm over the land (CDL) for microwave humidity sounder-2 (MWHS-2), which is characterized by the first operational satellite sensor measuring 118.75 GHz. The CDL is based on the advanced machine learning algorithm gradient boosting decision tree, which achieves the state-of-the-art performance on tabular data with high accuracy, fast training speed, great generalization ability, and weight factor ranking of predictors (or features). Given that the new-generation weather radar of China (CINRAD) provides improved cloud information with extensive temporal-spatial coverage, the observations from CINRAD are used to train the algorithm in this study. There are four groups of radiometric information employed to evaluate the CDL: all frequency ranges from MWHS-2 (all-algorithm), the humidity channels near 183.31 GHz (hum-algorithm), the temperature channels near 118.75 GHz (tem-algorithm), and the window channels at 89 and 150 GHz (win-algorithm). It is revealed that the tem-algorithm (around 118.75 GHz) has a superior performance for CDL along with the optimal values of most evaluation metrics. Although the all-algorithm uses all available frequencies, it shows inferior ability for CDL. Followed are the win-algorithm and hum-algorithm, and the win-algorithm performs better. The analysis also indicates that the latitude, zenith angle, and the azimuth are the top-ranking features for all four algorithms. The presented algorithm CDL can be applied in the quality control processes of assimilating microwave radiances or in the retrieval of atmospheric and surface parameters for cloud filtering. © 2008-2012 IEEE." "56119224700;57208747547;57202362638;24502843200;6602345158;57191895403;","Road infrastructure heritage: From scan to infrabim",2020,"10.37394/232015.2020.16.65","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090685608&doi=10.37394%2f232015.2020.16.65&partnerID=40&md5=dfe11c30022e85640748b806687a30cb","-BIM is a methodology applied to the realization of design models applied to new buildings. To date, however, most of the building interventions as it happens in the field of cultural heritage are developed in the existing. For this reason, scan to BIM procedures are improved and improved every day to make the use of BIM easier. This document will describe the combination of different geomatics techniques used by the Geomatics Laboratory of the University of the Mediterranean in Reggio Calabria to create a Building Information Model of a highway viaduct (infraBIM). In particular, we paid more attention to the scan to BIM phase through the segmentation of the point cloud using machine learning techniques that allow to obtain the constitutive parametric elements of the 3D model. The model containing the geometric and physical data made available by the ANAS management body in order to use the potential of infraBIM. This methodology today is of particular importance for the control, monitoring, intervention, and maintenance of road infrastructures, optimizing the procedures existing up to now. The advantages would be even more evident considering that we are living in a particular historical moment, in which a large number of bridges and viaducts in our nation are subject to advanced forms of degradation. © 2020, World Scientific and Engineering Academy and Society. All rights reserved." "57214044022;","An Optimized Approach-Based Machine Learning to Mitigate DDoS Attack in Cloud Computing",2020,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090649698&partnerID=40&md5=e4cccfbbfa4c2a779dbb7fe78b120ca4","With the emerging growth of cloud computing technology and on-demand services, users can access cloud services and software freely and applications based on the ""pay-as-you go"" concept. This innovation reduced service costs and made them cheaper with high reliability. One of the most significant characteristics of the cloud concept is on-demand services. One can access the applications of cloud computing at any time at a much lower cost. In addition to providing cloud users with much-needed services, the cloud also gets rid of security concerns which are not tolerated by the cloud. One of the most security problems in the cloud environment is Distributed Denial of Service (DDoS) attack that are responsible for overloading the cloud servers. This paper highlights a prevention technique (CS-ANN) which detect the DDoS attack and makes the server side more sensitive by integrating a Cuckoo Search (CS) approach with the Artificial Neural Network (ANN) approach. The cloud user features, along with the attacker features, are optimized using CS as a nature-inspired approach. Later on, these optimized features are passed to the ANN structure. The trained features are stored in the database and used during testing process to match the test features with the trained features and hence provide results in terms of attacker and normal cloud users. The test results of CS-ANN show a True Positive Rate (TPR), False Positive Rate (FPR) and detection accuracy of 0.99, 0.0105 and 0.9865% respectively. The proposed approach outperforms in contrast to the other two state-of-the-art techniques. © International Research Publication House." "57192543260;57204349215;57191667621;56249968800;","Tree detection and health monitoring in multispectral aerial imagery and photogrammetric pointclouds using machine learning",2020,"10.1109/JSTARS.2020.2995391","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090504711&doi=10.1109%2fJSTARS.2020.2995391&partnerID=40&md5=ca5486ecadcff129dc5cc2bbb6bf3b02","A machine learning methodology is developed for the detection of individual trees, classification of health, and detection of dead/dying trees in 125 mm resolution aerial multispectral or-thoimagery and photogrammetric pointcloud data. The novelty of the proposed method lies in its flexible utilization of features from different georegistered data sources. The methodology is evaluated on a commercial Pinus radiata plantation with a known outbreak of Sirex noctilio. An analysis was carried out to determine the value of the different sensors/data sources and the influence of the spatial resolution of the data on performance. Using the proposed methodology, trees were detected with approximate commission and omission errors of 5% and 22%, respectively, whilst dead or dying trees could be detected with commission and omission errors of approximately 26% and 9%. It was found that the multispectral imagery was the most informative sourceofdata for the given tasks. Tree detection in general was found to be sensitive to the spatial resolution of the data, whilst diseased tree detection was found to be more robust. © 2020 Institute of Electrical and Electronics Engineers. All rights reserved." "57218501642;55904509900;","Research on a Single-Tree Point Cloud Segmentation Method Based on UAV Tilt Photography and Deep Learning Algorithm",2020,"10.1109/JSTARS.2020.3008918","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85089341117&doi=10.1109%2fJSTARS.2020.3008918&partnerID=40&md5=e7089d787531c317c0a1ab8650df4465","Developing a robust point cloud segmentation algorithm for individual trees from an amount of point cloud data has great significance for tracking tree changes. This method can measure the size, growth, and mortality of individual trees to track and understand forest carbon storage and variation. Traditional measurement methods are not only slow but also tardy. In order to obtain forest information better and faster, this article focuses on two aspects: The first is using UAVs to obtain multiview remote sensing images of the forest, and then using the structure from motion algorithm to construct the forest sparse point cloud and patch-based MVS algorithm to construct the dense point cloud. The second is that a targeted point cloud deep learning method is proposed to extract the point cloud of a single tree. The research results show that the accuracy of single-tree point cloud segmentation of deep learning methods is more than 90%, and the accuracy is far better than traditional planar image segmentation and point cloud segmentation. The combination of point cloud data acquisition with UAV remote sensing and point cloud deep learning algorithms can meet the needs of forestry surveys. It is undeniable that this method, as a forestry survey tool, has a large space for promotion and possible future development. © 2008-2012 IEEE." "57212311690;16023225600;55339456900;57208524615;36052693100;","Deep convolutional neural networks for forensic age estimation: A review",2020,"10.1007/978-3-030-35746-7_17","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085219163&doi=10.1007%2f978-3-030-35746-7_17&partnerID=40&md5=ad7f1f3d077118448629a22db4548bc8","Forensic age estimation is usually requested by courts, but applications can go beyond the legal requirement to enforce policies or offer age-sensitive services. Various biological features such as the face, bones, skeletal and dental structures can be utilised to estimate age. This article will cover how modern technology has developed to provide new methods and algorithms to digitalise this process for the medical community and beyond. The scientific study of Machine Learning (ML) have introduced statistical models without relying on explicit instructions, instead, these models rely on patterns and inference. Furthermore, the large-scale availability of relevant data (medical images) and computational power facilitated by the availability of powerful Graphics Processing Units (GPUs) and Cloud Computing services have accelerated this transformation in age estimation. Magnetic Resonant Imaging (MRI) and X-ray are examples of imaging techniques used to document bones and dental structures with attention to detail making them suitable for age estimation. We discuss how Convolutional Neural Network (CNN) can be used for this purpose and the advantage of using deep CNNs over traditional methods. The article also aims to evaluate various databases and algorithms used for age estimation using facial images and dental images. © 2020, Springer Nature Switzerland AG." "57208754607;6701824774;12753037900;6506881892;57209202081;57214155278;55905144900;57215408871;","Assessing the changes in the moisture/dryness of water cavity surfaces in imlili sebkha in southwestern morocco by using machine learning classification in google earth engine",2020,"10.3390/rs12010131","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85078350488&doi=10.3390%2frs12010131&partnerID=40&md5=cfe4979da56272e8541c6ff05961c792","Imlili Sebkha is a stable and flat depression in southern Morocco that is more than 10 km long and almost 3 km wide. This region is mainly sandy, but its northern part holds permanent water pockets that contain fauna and flora despite their hypersaline water. Google Earth Engine (GEE) has revolutionized land monitoring analysis by allowing the use of satellite imagery and other datasets via cloud computing technology and server-side JavaScript programming. This work highlights the potential application of GEE in processing large amounts of satellite Earth Observation (EO) Big Data for the free, long-term, and wide spatio-temporal wet/dry permanent salt water cavities and moisture monitoring of Imlili Sebkha. Optical and radar images were used to understand the functions of Imlili Sebkha in discovering underground hydrological networks. The main objective of this work was to investigate and evaluate the complementarity of optical Landsat, Sentinel-2 data, and Sentinel-1 radar data in such a desert environment. Results show that radar images are not only well suited in studying desertic areas but also in mapping the water cavities in desert wetland zones. The sensitivity of these images to the variations in the slope of the topographic surface facilitated the geological and geomorphological analyses of desert zones and helped reveal the hydrological functions of Imlili Sebkha in discovering buried underground networks. © 2020 by the authors." "7004111660;56606212400;6602954401;7004167798;","Improvement of the 24 hr forecast of surface UV radiation using an ensemble approach",2020,"10.1002/met.1865","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076739442&doi=10.1002%2fmet.1865&partnerID=40&md5=5d5025679b0f9f5e9130e2a9cd3a91d8","A methodology is proposed to improve the 24 hr forecast of the ultraviolet (UV) index and the duration of exposure to obtain the minimal erythemal dose (MED). A forecast ensemble consisting of 10 members (differing in initial and boundary conditions) is examined to search for the best performed ensemble member. Routine UV measurements are used for the forecast validation. These are carried out at Belsk (20.8 ° E, 51.8 ° N) and in Racibórz (18.2 ° E, 50.1 ° N) representing a rural and an urban site in Poland, respectively. Each ensemble member is built using the clear-sky simulations by a radiative transfer model. The clear-sky irradiance is attenuated using the cloud modification factor (CMF) depending on the cloud cover by low- and mid-level clouds. The 24 hr forecast of cloudiness is obtained by the Weather Research and Forecasting (WRF) model. Every day, for each ensemble member, the optimal CMF values are built by the offline bootstrapping of the original CMF matrix. The performance of all ensemble members is evaluated for the day preceding the forecast. The best one is subsequently used for the next-day forecast. This procedure provides a more accurate forecast than that based on a single member of the ensemble. For both sites, the root mean square percentage error for the duration of the MED exposure changes from about 30% to about 15%, and mean absolute percentage error from about 20–25% to about 10%. © 2019 The Authors. Meteorological Applications published by John Wiley & Sons Ltd on behalf of the Royal Meteorological Society." "13204812100;57217112643;","Information technology for sustainable agriculture",2019,"10.1007/978-3-030-23169-9_19","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086296686&doi=10.1007%2f978-3-030-23169-9_19&partnerID=40&md5=3f0dd5e12c09862fd7e165f5d856f1a9","Feeding global population by 2050 requires 60% increase in the agricultural production. Agricultural transformation has a role to play for food security, poverty reduction and economic growth. However, sustainability and climate risk management are the challenges. The recent advancements in information technology (I. T) delivered smart devices, computing and sensor technologies. Application of those smart technologies have the potential to enable agricultural industry meets its productivity and sustainability challenge as well as solving indigenous agricultural problems of the developing nations. The geospatial data archives and real-time data from satellites, UAVs, RFIDs in combination with weather data, digitized soil data, and other real-time data streams coming from in-situ smart sensors can now give us a better understanding of the interaction between crops, weather and soils than ever before. Further the analytics of that big data assisted by machine learning can provide decision support in this regard. The customized I. T packages are required where, e-farm production system based on precision agriculture techniques, crops and livestock management, precision irrigation applications, crop water and pest/disease management, wireless moisture sensing networks, wireless communication in UAVs used for vegetation health detection, rainfall monitoring system based on mobile communication data, cloud services for knowledgebase on soils, nutrients, yields by making soil, nutrient and yield maps and disseminating through mobile networks and variable rate application based on GPS and GIS systems. Every passing day, the use of internet and smartphones is enhancing rapidly. The cloud-based services for big data analytics in agriculture and data sharing apps with linkages to integrated platforms and models are the future of farming in both modern and developing world. © Springer Nature Switzerland AG 2019." "57216256526;7404843459;57216256773;","Super-Resolution Algorithm of Satellite Cloud Image Based on WGAN-GP",2019,"10.1109/ICMO49322.2019.9026112","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082963009&doi=10.1109%2fICMO49322.2019.9026112&partnerID=40&md5=50fe0e227441937edf12f5ac4aa39e44","The resolution of an image is an important indicator for measuring image quality. The higher the resolution, the more detailed information is contained in the image, which is more conducive to subsequent image analysis and other tasks. Improving the resolution of images has always been the unremitting pursuit of industry and academia. In the past, people used hardware devices to increase the resolution, which is a practical solution. However, there are many limitations in the method of improving the image resolution by hardware devices. We use software-based image super-resolution technology, which transforms low-resolution images into high-resolution images through a series of machine learning algorithms. The classic GAN algorithm is difficult to train a model, and the improved Wasserstein GAN algorithm can make the model training more stable. Based on SRGAN model, this algorithm replaces the classical GAN algorithm with the improved WGAN algorithm. We will use the FY-3D satellite's Medium Resolution Spectral Imager Type II (MERSI-II) data, using super-resolution algorithms to make the reconstructed image significantly better visually. We conducted four sets of controlled experiments using four different improved methods. We will evaluate the image from three aspects: peak signal to noise ratio value, structural similarity value and visual effect. We applied the WGAN-GP algorithm to super-resolution tasks and achieved the desired results. © 2019 IEEE." "57216260024;57214130629;57216259010;55688898000;57213519323;","The Cloud Images Classification Based on Convolutional Neural Network",2019,"10.1109/ICMO49322.2019.9026121","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082932298&doi=10.1109%2fICMO49322.2019.9026121&partnerID=40&md5=17324f5a5ed39740b08a4e67012cd8cd","In order to achieve the goal of clouds images classification, a deep-learning method based on Convolutional Neural Network (CNN) is adopted.2300 raw true-color images of eight types of clouds are collected as the datasets from the website www.baidu.com, along with the corresponding labels. Before feeding to machine learning, some preprocessing steps are conducted. Firstly, every photo is preprocessed using the method of the sliding window that can crop the initial photo into a series of sub-images with 200∗200 pixels and RGB channels. Not only can this step increase the size of datasets, but it also can make more features reserved from inputs. Next, the image histogram equalization is adopted. Besides, In the CNN network construction, the ReLU activation function is placed directly after the convolutional layer when there isn't a BN layer behind it, while the ReLU function is placed after the BN layer if the BN layer exists. This kind of structure is designed to avoid the overfitting problem. The result obviously performs better when adopts more large and comprehensive datasets than that when datasets are unhandled. Meanwhile, owing to the MBSD gradient algorithm, the adjustment of arguments and layers can influence the final results as well. The classification accuracy has reached a comparatively high value, which can nearly meet our requirements. © 2019 IEEE." "57212174713;","Large-scale Ship Fault Data Retrieval Algorithm Supporting Complex Query in Cloud Computing",2019,"10.2112/SI97-034.1","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076112150&doi=10.2112%2fSI97-034.1&partnerID=40&md5=a24e6b408b5b717495918fc637c758cf","In the cloud computing environment, the mass ship fault data retrieval is easy to be interfered by the association rule items, the fuzzy clustering of the data retrieval is not good, the fault diagnosis efficiency of the ship is reduced, and in order to improve the fault diagnosis capability of the ship, The invention provides a mass ship fault data retrieval technology based on complex query support in a cloud computing environment. The distributed storage structure analysis of mass ship fault data is carried out by adopting a vector quantization characteristic coding technology, the spectral characteristic analysis of the mass ship fault data is carried out by adopting a subsection adaptive regression analysis method, the quantitative recursive analysis model is used for extracting the mass ship fault data, the method comprises the following steps of: extracting an association rule feature set reflecting the attribute of a mass ship fault data category, carrying out data classification retrieval on the extracted mass ship fault data feature quantity by using a BP neural network classifier, introducing a machine learning factor to perform convergence control on a support vector machine, And the global stability of the mass ship fault data retrieval is improved. The simulation results show that the accuracy of the data retrieval is high, the error rate is small, and the fault diagnosis ability of the ship is improved. © Coastal Education and Research Foundation, Inc. 2019." "57216288692;57216289852;57204629821;15060531500;","Non-invasive blood glucose monitoring using near-infrared spectroscopy based on internet of things using machine learning",2019,"10.1109/R10-HTC47129.2019.9042479","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083028663&doi=10.1109%2fR10-HTC47129.2019.9042479&partnerID=40&md5=d73eb77ca11d08344de0b59b254038c8","According to International Diabetes Federation, Indonesia ranked 6th in country with highest number of diabetes patients in the world. Now, the measurement of blood glucose in Indonesia mostly done in invasive manner, which is expensive, painful, and impractical. Moreover, according to Indonesia's Ministry of Health, there are more diabetes patient in rural areas, who have economic shortages and limited access to health care. So, a cheap and easy to use non-invasive blood glucose measurement system is needed to solve this problem. One of the current trend non-invasive blood glucose is the use of Near-Infrared Spectroscopy (NIR). In this paper, we discuss the development of a NIR based blood glucose monitor. The device's sensor consists of a pair of LED and photodiode which transmit and receive light with wavelength of 940 nm. The light intensity reading from the sensor will be amplified and filtered to reduce noise, then transmitted to the smart-phone. In the smart-phone application, the reading result will be converted to blood glucose level using a machine learning model embedded in the application. The model used in this final project is a sequential, layer-based neural network model provided by Keras. The model was built and trained on top of TensorFlow, and then converted for mobile use with the help of TensorFlow Lite, The model achieved an acceptable result with Mean Absolute Error (MAE) of 5.855 mg/dL. The result then could be stored in the online database using Cloud Firestore. © 2019 IEEE." "57203835698;57203835732;34969310900;","Short-Term Prediction of Electricity Outages Caused by Convective Storms",2019,"10.1109/TGRS.2019.2921809","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074498250&doi=10.1109%2fTGRS.2019.2921809&partnerID=40&md5=7bfda0ba98fbb093ee6fe885c45832d8","Prediction of power outages caused by convective storms, which are highly localized in space and time, is of crucial importance to power grid operators. We propose a new machine learning approach to predict the damage caused by storms. This approach hinges identifying and tracking of storm cells using weather radar images on the application of machine learning techniques. Overall prediction process consists of identifying storm cells from CAPPI weather radar images by contouring them with a solid 35-dBZ threshold, predicting a track of storm cells, and classifying them based on their damage potential to power grid operators. Tracked storm cells are then classified by combining data obtained from weather radar, ground weather observations, and lightning detectors. We compare random forest classifiers and deep neural networks as alternative methods to classify storm cells. The main challenge is that the training data are heavily imbalanced, as extreme weather events are rare. © 1980-2012 IEEE." "57197462965;56825747200;54784958200;55646959300;","Design and Implementation of Digital Twin for Predicting Failures in Automobiles Using Machine Learning Algorithms",2019,"10.4271/2019-28-0159","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074758484&doi=10.4271%2f2019-28-0159&partnerID=40&md5=328fbd7149521ccfb1133a2ed3345c64","The drastic technological advancements in the field of autonomous vehicles and connected cars lead to substantial progression in the commercial values of automobile industries. However, these advancements force the Original Equipment Manufacturers (OEMs) to shift from feedback-based reactive business analysis to operational-data based predictive analysis thereby enhancing both the customer satisfaction as well as business opportunities. The operational data is nothing but the parameters obtained from several parts of an automobile during its operation such as, temperature in radiator, viscosity of the engine oil and force applied over the brake disk. These operational data are gathered using several sensors implanted in different parts of an automobile and are continuously transmitted to backend computers to develop Digital Twin, which is a virtual model of the physical automobile. Later, gathered operational data are analyzed using data mining algorithms to predict the failures of an automobile well in advance, better insights into performance of an automobile thereby recommending alternative design choices and remote service management of failures by a professional technician. Firstly, this research work illustrates the platform for the creation of digital twin using Eclipse Hono, Eclipse Kura and Eclipse Ditto. Secondly, it explains about the operational data gathering and processing at the nearby edge devices as well as the remote cloud. Finally, the prediction of failures is demonstrated using Turbofan Engine Degradation Simulation Dataset by means of several machine learning regression algorithms and compare their accuracies. Finally, it is concluded that Gradient Boost Regressor provides better accuracy in predicting future failures. © 2019 SAE International. All Rights Reserved." "57217665158;56041758500;","Machine learning strategies for enhancing bathymetry extraction from imbalanced lidar point clouds",2019,"10.23919/OCEANS40490.2019.8962400","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079077294&doi=10.23919%2fOCEANS40490.2019.8962400&partnerID=40&md5=d6271c8cca635611b3559ba5b13d9f98","Density-based approaches to extract bathymetry from airborne lidar point clouds generally rely on histogram/frequency-based disambiguation rules to separate noise from signal. The present work targets the improvement of such disambiguation rules by enhancing each pulse with a machine learning-based estimate of its p(Bathy) - i.e., its probability of truly being bathymetry. Extreme gradient boosting (XGB) is used to assess the strength of bathymetric signal in pulse return metadata. Because lidar point clouds can be highly imbalanced between Bathymetry and NotBathymetry, three strategies for mitigating the effects of imbalanced samples were examined. Impacts of an imbalanced lidar point cloud were successfully mitigated by: •Applying an 'optimal' decision threshold (ODT) that equalizes accuracy for Bathymetry and NotBathymetry to p(Bathy) rather than using a conventional probability decision threshold (PDT) of 0.50, and •Using proportional class weighting to fit the XGB model. However, decomposing a confusion matrix by iteratively discarding misclassified points and re-fitting an XGB model was not successful in improving the strength or detectability of the bathymetric signal in the metadata. The same was true for iteratively discarding correctly classified points. The bathymetric signal in the metadata was found to be sufficiently strong to explore the operational incorporation of results into the disambiguation rules of density-based bathymertric extraction methods. © 2019 Marine Technology Society." "57211910620;57211908810;","AI & CoE",2019,"10.35940/ijeat.A9846.109119","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075295961&doi=10.35940%2fijeat.A9846.109119&partnerID=40&md5=314d5856d171ad0701babe1765f8b3fc","Artificial Intelligence or AI is being positioned as the panacea for all organizational problems; while Centre of Excellence or CoE, which is distinctly different from Research and Development activities helps organizations in their pursuit of higher revenue and profit. In this paper, the researchers have analysed the growth and importance of each of these two concepts–AI and the CoE; and have worked towards putting them together for creating a unique combination which shall benefit the organizations and hence the economy at large. In the process, a framework is provided for companies to improve, innovate, optimize, and eventually automate their management systems while making the core-competencies of their business AI proof. It is hoped that through this framework, organizations will be able to create a substantial impact by improving existing capabilities and actively creating new strategic resources in the interest of all the stakeholders. © BEIESP." "57211621229;56910124600;","PM2.5 prediction using machine learning hybrid model for smart health",2019,"10.35940/ijeat.A1187.109119","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074658732&doi=10.35940%2fijeat.A1187.109119&partnerID=40&md5=175dcd98b3f99d08082557214863bda2","Air Pollution is one of the current serious issue attributable to people's health causing cardiopulmonary deaths, lung cancer and several respiratory problems. Air is polluted by numerous air pollutants, among which Particulate Matter (PM2.5) is considered harmful consists of suspended particles with a diameter less than 2.5 micrometers.This paper aims to acquire PM2.5 data through IoT devices,store it in Cloud and propose an improved hybrid model that predicts the PM2.5 concentration in the air. Finally through forecasting system we alert the public in case of an undesired condition. The experimental result shows that our proposed hybrid model achieve better performance than other regression models. © BEIESP." "57215597635;57203397955;57216355489;56722821200;57201399639;7101627008;","A hybrid algorithm for mineral dust detection using satellite data",2019,"10.1109/eScience.2019.00012","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083233958&doi=10.1109%2feScience.2019.00012&partnerID=40&md5=bbf101231a6b3d1365a70fcf2a134609","Mineral dust, defined as aerosol originating from the soil, can have various harmful effects to the environment and human health. The detection of dust, and particularly incoming dust storms, may help prevent some of these negative impacts. In this paper, using satellite observations from Moderate Resolution Imaging Spectroradiometer (MODIS) and the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation Observation (CALIPSO), we compared several machine learning algorithms to traditional physical models and evaluated their performance regarding mineral dust detection. Based on the comparison results, we proposed a hybrid algorithm to integrate physical model with the data mining model, which achieved the best accuracy result among all the methods. Further, we identified the ranking of different channels of MODIS data based on the importance of the band wavelengths in dust detection. Our model also showed the quantitative relationships between the dust and the different band wavelengths. © 2019 IEEE." "57188690965;57203079997;57194625045;56046460900;","Estimation of -CG lightning distances using single-station E-field measurements and machine learning techniques",2019,"10.1109/SIPDA47030.2019.9004484","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85080936194&doi=10.1109%2fSIPDA47030.2019.9004484&partnerID=40&md5=f763a6eb28b3220890ac10c75c7b34eb","Machine learning (ML) techniques have been used around the world to solve different problems. In this work, we applied ML techniques to estimate lightning distance for return strokes (RS) in negative cloud-to-ground (-CG) flashes. The approach uses E-field records from a single-station system. A lightning electric field waveform dataset containing more than 1500 waveforms of negative RS recorded at LOG, Florida, US was used to train and validate the ML classifiers. The dataset was split into day time and night time records. For day-time records, the quadratic Support Vector Machine (SVM) classifier was the one with the best accuracy (80%) and for night-time, the best one was the linear SVM with an accuracy of 88%. The ML classifiers were applied to estimate lightning distance in thunderstorms in the Amazon region of Brazil, and the results were compared against GOES-16 images and STARNET lightning location data. The main application of such methodology is for regions with no lightning location systems or no communication links to obtain lightning location data. © 2019 IEEE." "57195197915;56271680000;","Machine learning algorithm for early detection of heart diseases using 3-tier IOT architecture",2019,"10.35940/ijeat.F1360.0986S319","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075458510&doi=10.35940%2fijeat.F1360.0986S319&partnerID=40&md5=83d384750db497fdc187418d9cf3c375","Among the applications empowered by the Internet of Things (IoT), regular health monitoring framework is an important one. Wearable sensor gadgets utilized in IoT health monitoring framework have been producing huge amount of data on regular basis. The speed of data generation by IoT sensor gadgets is very high. Henceforth, the volume of data generated from the IoT-based health monitoring framework is also very high. So as to overcome this problem, this paper proposes adaptable three-tier architecture to store and process such immense volume of wearable sensor data. Tier 1 focuses on gathering of data from IoT wearable sensor gadgets. Tier 2 employs Apache HBase for storing substantial volume of wearable IoT sensor data in cloud computing. Likewise, Tier-3 utilizes Apache Mahout for building up logistic regression-based prediction model for heart related issues. At long last, ROC examination is performed to identify the most significant clinical parameters to get heart diseases. © BEIESP." "57202077645;55663177800;37012907500;36988104100;","Cloud-based non-intrusive load monitoring system (NILM)",2019,"10.35940/ijeat.F1021.0986S319","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075453130&doi=10.35940%2fijeat.F1021.0986S319&partnerID=40&md5=d0c920bccedb92ef7ad74ca5c73cdbc0","Design and development of a cloud-based non-intrusive load monitoring System (NILM) is presented. It serves for monitoring and disaggregating the aggregated data such as smart metering into appliance-level load information by using cloud computing and machine learning algorithms implemented in cloud. The existing NILM systems are lack of scalability and limited in computing resources (computation and data storage) due to dedicated, closed and proprietary-based characteristics. They are inaccessible to variety of heterogeneity data (electrical and non-electrical data) openly for improving NILM performance. Therefore, this paper proposed a novel cloud-based NILM system to enable collection of these open data for load monitoring and other energy-related services. The collected data such as smart meter or data acquisition unit (DAQ), is pre-processed and uploaded to the cloud platform. A classifier algorithm based on Artificial Neural Network (ANN) is implemented in Azure ML Studio (AzureML), followed by the classifier testing with different combinations of feature set for the performance comparison. Furthermore, a web service is deployed for web APIs (Application Programming Interfaces) of applications such as smart grid and smart cities. The results shows that the ANN classifier for multiclass classification has improved performance with additional features of harmonics, apart from active and reactive powers used. It also demonstrates the feasibility of proposed cloud-based classifier model for load monitoring. Therefore, the proposed solution offers a convenient and cost-effective way of load monitoring via cloud computing technology for smart grid and smart home applications. Further work includes the use of other ML algorithms for classifier, performance analysis, development of cloud-based universal appliance data and use cases. © BEIESP." "57211938438;","5G-smart diabetes: Personalized diabetes diagnosis with human services big data clouds",2019,"10.35940/ijeat.F1335.0986S319","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85075445463&doi=10.35940%2fijeat.F1335.0986S319&partnerID=40&md5=5f900f2f6c5db1acecc2965507dbc6aa","Ongoing advances in remote systems administration and huge information innovations, for example, 5G systems, medicinal huge information investigation, and the Internet of Things, alongside ongoing improvements in wearable figuring and man-made brainpower, are empowering the advancement and usage of imaginative diabetes checking frameworks and applications. Because of the deep rooted and efficient damage endured by diabetes patients, it is basic to plan powerful strategies for the determination and treatment of diabetes. In light of our far reaching examination, this paper characterizes those strategies into Diabetes 1.0 and Diabetes 2.0, which show insufficiencies as far as systems administration and insight. Consequently, our objective is to structure a manageable, financially savvy, and insightful diabetes finding arrangement with customized treatment. In this paper, we initially propose the 5G-Smart Diabetes framework, which joins the best in class advancements, for example, wearable 2.0, machine learning, and huge information to create complete detecting and investigation for patients experiencing diabetes. At that point we present the information sharing system and customized information examination display for 5G-Smart Diabetes. At last, we construct a 5G-Smart Diabetes testbed that incorporates savvy dress, cell phone, and huge information mists. The trial results demonstrate that our framework can successfully give customized analysis and treatment proposals to patients. © BEIESP." "55982189400;55639123200;","Research on inefficiency analysis method of building energy utilizing time series data",2019,"10.1088/1755-1315/294/1/012052","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071868044&doi=10.1088%2f1755-1315%2f294%2f1%2f012052&partnerID=40&md5=c347553e2fd7a298028bccc18c067908","Purpose; Our research purpose is to utilize time series data related to building energy and perform an inefficiency analysis of building energy efficiency. Development; We developed a cloud building energy and environment monitoring system (cloud Building Management System (BEMS)) and an optimized control system (an artificial intelligence [AI] control system). Methodology; First, we designed an inefficiency analysis model and the analysis steps. Next, this model was tested in several actual buildings, and the method was confirmed to be adaptable. Finally, through iterative adaptations based on the results, we significantly improved our method. Findings; (1) The developed cloud BEMS and AI control system were confirmed to be useful for a) improvement of building management, b) improvement of equipment and system, and c) renovation. (2) Our analytical method and its steps can quantify inefficiency. (3) When using this methodology, we can quantitatively predict before implementation the effect of a) building improvement, b) equipment and system improvement, and c) renovation. Originality/value; Cloud BEMS collects time series data related to building energy consumption from sensors and building automation systems and accumulates data in a cloud server. The AI control system and analytical method finds energy inefficiencies and improves the operation of the building. Moreover, this improvement in operation is unmanned and carried out automatically. Keywords; Sustainable, Internet of things (IoT), machine learning, artificial intelligence (AI), energy management system (EMS), energy conservation, environment efficiency, optimized control, CO2 reduction. Paper type; Academic paper. © Published under licence by IOP Publishing Ltd." "57211289630;57203689843;","A systematic access through machine learning methods for expectation in malady related qualities",2019,"10.35940/ijeat.F1193.0886S19","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85073348095&doi=10.35940%2fijeat.F1193.0886S19&partnerID=40&md5=7177f48073f74d611d246edbf05383ae","There are Many learning strategies that are been identified with distinguish infection based related qualities. At the early, they as a rule moved toward this issue as a parallel arrangement, where preparing set is involved examples. Examples developed sickness qualities, whereas negative examples are there mining which are not known to be connected with contaminations. This is the essential of the twofold deals based diagrams, since the negative arranging set ought to be true non-infection qualities; regardless, advancement of this set is on a very basic level unfeasible in biomedical inspects. Therefore, to reduce this delicacy, insightfully sensible social gathering based techniques have been proposed. For example, unary outline strategy subject to one-class SVM framework was proposed by grabbing from fundamentally positive models. Also, there mining set may contain cloud torment qualities; as such, semi-formed methodologies, for example, twofold semi-controlled & positive & unlabeled (PU) learning blueprints have been proposed. Specifically, PU learning frameworks, which increase from both known suffering qualities & there mining attributes, were appeared to beat others. In these examinations, information sources are commonly tended to by vectorial plan for cemented classifiers, while they are in bit frameworks for unary & PU learning ones. The bit based information blend is reasonable for information with various sorts & it has the majority of the stores of being uncalled for or the relationship subject to various information diagrams. In like manner, in this examination, we looked accumulating structures for the ailment quality figure dependent on vectorial delineation of tests. The spread outcome demonstrated that the unary strategy structure, which joins both thickness & class likelihood estimation approachs, accomplished the best execution, where as it is most recognizably stunning for the one-class SVM-based technique. fascinatingly, execution of a best twofold outline framework is in each rational sense misty with that of uneven SVM-based PU learning & twofold semidirected hoarding strategy they are altogether improved. © BEIESP." "57205578378;57211300508;","IOT and data research in industrial power management",2019,"10.35940/ijeat.F1011.0886S19","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85073334195&doi=10.35940%2fijeat.F1011.0886S19&partnerID=40&md5=27fc7500d246f55dc33e1006c34f4f30","IOT plays an important role in collecting data and machine learning for prediction in variety of applications like homecare, healthcare and energy management. In energy management there are various variables such as future power demands, generation status weather conditions and current battery status hard to expect high efficiency. Here, in this proposed idea, for higher efficiency of renewable energy, an IOT system is needed to monitor and collect these Statuses and provide energy management services. Energy will be consumed of passive operation according to hourly variation in price and battery status will be predicted by using machine learning algorithms like Logistic regression, SVM, and k-NN. We trained the system by considering five random variables in datasheet such as Current time, Current cost, predicted time, predicted cost and Solar battery status from the device. This integrated system is used for uploading power related details of Grid and Solar to IBM cloud. Depending on previous datasheet, analytics will be done by resulting which source has to be triggered to drive the load either Solar or Grid. APIs and Node-Red Tool were used for wiring sensor data and Model predicted output. In future power demands, this design will help to predict the price according to hourly variation based on the units and to trigger the source. © BEIESP." "57210971634;56771498800;57210972983;","Detection of intrusion using hybrid feature selection and flexible rule based machine learning",2019,"10.35940/ijeat.F8783.088619","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85072062467&doi=10.35940%2fijeat.F8783.088619&partnerID=40&md5=11c21ef35407df6c7b020615abdab1d6","With the rapid growth in the data processing and data sharing, the application owners and the consumers of the applications are more influenced to use the remote storage on cloud-based data centre and the application generated data is also growing ups and bounds. Nevertheless, the adaptation of the data sharing, and data processing applications were not easy for the consumers. The application owners and the service providers have struggled with the sensitive data of the consumers and the consumers were also faced trust issues with the complete framework. The standard legacy applications were designed for the traditional centralized scenarios, where the intrusion detection can be performed only using the network status analysis and the application characteristics analysis. Moreover, most of the parallel calculations initially enhance the hybrid likelihood and change likelihood of GA as indicated by the populace advancement variable-based math and wellness esteem. Nevertheless, the population of data and the attacks on the data is high and the correct population size is highly difficult to determine. Regardless to mention, that the use of fitness functions will restrict the attack detection to certain types and these algorithms are bound to fail in case of a newer attack. However, with the migration of application to the data processing framework, the consumers have started demanding more security against the intrusions. A good number of research attempts were made to map the traditional security algorithms into the data processing space, nonetheless, the attempts were highly criticized due to the lack of proper analysis of security attacks on data processing applications. Hence, this work proposes a novel framework to detect the intrusions on data processing framework with justifying attack characteristics. This work proposes a novel algorithm to reduce the features of attack characteristics to justify the gaps on data processing frameworks with significant reduction in time for processing and further, proposes an algorithm to derive a strong rule engine to analyse the attack characteristics for detecting newer attacks. The complete proposed framework demonstrates nearly 93% and higher accuracy, which is much higher than the existing parallel research outcomes with least time complexity. © BEIESP." "57190491796;56252036100;","Decision tree: A predictive modeling tool used in cloud trust prediction",2019,"10.35940/ijeat.F8763.088619","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85072045101&doi=10.35940%2fijeat.F8763.088619&partnerID=40&md5=5fe4f597066479a6aee3adf32caa475f","Trust is one of the important challenges faced by the cloud industry. Ever increasing data theft cases are contributing in worsening this issue. Regarding trust, author has a perception that this challenge can be handled to some extend if consumer can evaluate “Trust Value “ of the provider or can predict the same on some reliable basis. Current research is using predictive modeling for predicting trustworthiness of cloud provider. This paper is an attempt to utilize the data mining algorithm for predictive modeling. Decision Tree, a supervised data mining algorithm has been used in the current work for making predictions. Certification attainment criteria as prime basis for trust evaluation. In current scenario, data mining algorithm will classify providers in category of low, medium and high category of trust on the basis of information displayed on the public domain. ©BEIESP." "57210957409;56040721700;57210959142;","Upgrade-data security in cloud by machine learning and cryptography techniques",2019,"10.35940/ijeat.F8395.088619","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85072019154&doi=10.35940%2fijeat.F8395.088619&partnerID=40&md5=8cd96d23d16ce5f9c3a71b563cff0441","The term cloud computing is referred as the shared pool of customizable computer resources and high quantity services which can easily be provisioned with less management endeavours via internet. It transfigured the mode associations reach IT, which enables them to be more perceptive, launch new business models, and minimise the IT costs. These technologies are to be administrated in an interdisciplinary collection of architectures, characterized into various deployment and service models, and can synchronize with other related technologies. The widespread issues with cloud computing are security, reliability, data privacy and anonymity. Cloud computing provides a way to share distributed sources and services that are owned by different organizations or sites. Since it shares distributed resources via network in open environment that results in security issues. In this paper, our aim is to upgrade the security of data in the cloud and also to annihilate the difficulties related to the data security with encipher algorithm. In our proposed plan, some key services of security like authentication and cryptographic techniques are assigned in cloud computing environment. © BEIESP." "55430538200;8529553300;55908790400;36170340400;36806957800;57210434487;57210436461;","Microservices and machine learning algorithms for adaptive green buildings",2019,"10.3390/su11164320","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85070712595&doi=10.3390%2fsu11164320&partnerID=40&md5=547ebabe27f2f4e288d0a8b65e03a6b6","In recent years, the use of services for Open Systems development has consolidated and strengthened. Advances in the Service Science and Engineering (SSE) community, promoted by the reinforcement ofWeb Services and SemanticWeb technologies and the presence of new Cloud computing techniques, such as the proliferation of microservices solutions, have allowed software architects to experiment and develop new ways of building open and adaptable computer systems at runtime. Home automation, intelligent buildings, robotics, graphical user interfaces are some of the social atmosphere environments suitable in which to apply certain innovative trends. This paper presents a schema for the adaptation of Dynamic Computer Systems (DCS) using interdisciplinary techniques on model-driven engineering, service engineering and soft computing. The proposal manages an orchestrated microservices schema for adapting component-based software architectural systems at runtime. This schema has been developed as a three-layer adaptive transformation process that is supported on a rule-based decision-making service implemented by means of Machine Learning (ML) algorithms. The experimental development was implemented in the Solar Energy Research Center (CIESOL) applying the proposed microservices schema for adapting home architectural atmosphere systems on Green Buildings. © 2019 by the authors." "57205651985;57193516410;8502218200;57195735997;6507631512;55802422900;35595209900;","Evaluation of Cloud Type Classification Based on Split Window Algorithm Using Himawari-8 Satellite Data",2019,"10.1109/IGARSS.2019.8898451","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077721443&doi=10.1109%2fIGARSS.2019.8898451&partnerID=40&md5=f04c62e5f9884e5bdc96b3080796574f","Precise evaluation of cloud types is indispensable for the detailed analysis of the Earth's radiation budget. The split window algorithm (SWA) is an algorithm that has been widely employed for cloud type classification from meteorological satellite imagery. In this study, we apply the SWA to analyze the clouds that appear in the Japan area using the imagery of Himawari-8 meteorological satellite. The brightness temperature (BT) information from band 13 (BT13, 10 μm) and band 15 (BT15, 12 μm) are employed with the BT difference (BTD) between these two bands (BTD13-15). For daytime analysis, the albedo of band 1 (0.47 μm) is also used to discriminate the cloudy and cloud-free areas. The validation of the resulting cloud type (SWA13-15), which includes ten classes including cloud-free condition, is carried out using the space-borne lidar data concurrent with the satellite observations. In addition, two different classifiers, namely, the sequential minimal optimization (SMO) and Naïve Bayes (NB) classifiers are tested with the results of SWA. When about 10% of 2 million data points are used for training the classifiers, the test results reveal that the correctly classified points are 97.0% and 89.5% for the first dataset (observed in July 2015) and 97.4%, and 92.1% for the second dataset (July 2016) for SMO and NB, respectively. © 2019 IEEE." "55918225100;57189026667;55706555700;","Cloud Detection and Classification for S-NPP FSR CRIS Data Using Supervised Machine Learning",2019,"10.1109/IGARSS.2019.8898876","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077694896&doi=10.1109%2fIGARSS.2019.8898876&partnerID=40&md5=d85a9995d1c16c9c0fb6f069d14350b3","Cloud detection and classification are important and challenging in satellite infrared remote sensing. This study proposes an end-to-end cloud detection and classification method for the Cross-track Infrared Sounder (CrIS) based on big data analysis and relevant machine learning schemes. Using the full spectral resolution (FSR) CrIS data, a set of FSR cloud detection indexes (FCDIs) is derived from the brightness temperatures of the selected CrIS LWIR-SWIR channel pairs. Linear and cubic regressions are compared and discussed when deriving FCDI. It's shown that FCDIs can capture clouds and structures well by comparing with the Visible Infrared Imaging Radiometer Suite (VIIRS) cloud cover/layer product. Selected from several supervised machine learning schemes, the extreme learning machine (ELM) is applied to train FCDI for cloud detection. Overall the ELM classification accuracy is about 80%. Results and discussion are provided. © 2019 IEEE." "57213189957;56049862900;57191270639;57213199230;57207365988;57213192204;","KORE Application: Potatoes Yield Assessment",2019,"10.1109/IGARSS.2019.8898996","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077673883&doi=10.1109%2fIGARSS.2019.8898996&partnerID=40&md5=6fab62b0e05f347ddab595b7c4adf558","KORE is a cloud-based precision agriculture platform, based on subscriptions, that combines satellites, drones and ground sensor data. This paper focuses on one of its applications: yield assessment in potato fields using UAV data and deep learning. © 2019 IEEE." "35176903900;57211186337;8976516300;7006684098;56913686000;55946457200;","Study on temporal and spatial adaptability of crop classification models",2019,"10.1109/Agro-Geoinformatics.2019.8820233","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85072912373&doi=10.1109%2fAgro-Geoinformatics.2019.8820233&partnerID=40&md5=b1383142e1c93c480ed146be76746c97","Crop classification is an important part of national agricultural management, and accurate crop classification is conducive to crop growth monitoring and yield assessment. However, due to the different growing years and regions, even the same crop has different growth processes and different phenological characteristics. Therefore, improving the spatial and temporal adaptability of the classification model is an important research content for large-scale crop classification. In this paper, several adjacent agricultural production areas are studied. Based on the stable time-series remote sensing image dataset, the adaptive changes of several machine learning classification methods with higher classification accuracy in spatial and temporal are studied. The paper selected Sentinel 1 satellite data with good anti-cloud interference and a short return visit cycle for experiments. Firstly, the training of each classification model in the same area is completed, and then the spatial adaptability of the model is studied in different adjacent ranges. Finally, the adaptability of different classification models to the change of the growth cycle of the same type of crop is also compared. The paper finds that the models such as CNN+LSTM and BinConvLSTM perform better in temporal and spatial. © 2019 IEEE." "57211353806;57206774247;56669065400;57211351208;35750657400;57211347521;57211353913;","Sharing learnings: The methodology, optimisation and benefits of moving subsurface data to the public cloud",2019,"10.3997/2214-4609.201900773","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088067716&doi=10.3997%2f2214-4609.201900773&partnerID=40&md5=1c5035aa0abfba12108487d076eeda22","We detail the application of cloud technology to Chevron's seismic data repository, share the applicable learnings and highlight areas of workflow evolution that delivered value. The aim of the project, run as a piloted field trial, was to couple the transformative capability of public cloud services with subsurface data. The learnings shared in this paper are designed to inform and assist other subsurface data custodians whether they are energy companies, national data repositories, service companies or academia. Cloud technology enables us to reduce seismic data duplication to a single data version which can be accessed securely by all appropriate stakeholders. We discuss a highly automated, scalable data migration process which included seismic ingestion, machine learning and serverless architecture. This automated the data management process, progressing data from loading to global analysis in minutes. The project has provided access to subsurface data anytime, anywhere and on any device, delivering a more accessible data environment at lower costs and connecting via an API to traditional workflows. By approaching the subsurface data challenge in an innovative way, this project has provided multiple learnings to share and built a greater understanding of the value case for faster adoption of public cloud infrastructure. © 81st EAGE Conference and Exhibition 2019. All rights reserved." "57215345521;24766116400;10044906200;57215343685;57219734114;","EarthNET a native cloud web based solution for next generation subsurface workflows",2019,"10.3997/2214-4609.201901974","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084021587&doi=10.3997%2f2214-4609.201901974&partnerID=40&md5=80dcde21cc5cd666339b46ec447515fd","Faster and better data driven decision-making and shorter times to first oil and gas top the list of expected benefits that digital technologies can drive for upstream oil and gas companies. In the oil industry, Artificial Intelligence (AI) and Machine Learning (ML) tools have already moved from R&D projects into G&G tool boxes, slowly transforming the subsurface workflow. We will discuss about cloud platforms and demonstrate how such an integrated platforms provides both the data access and applications required to apply ML at scale with examples that include integration of multi regional datasets. We will show that such platforms are not only enhancing further creativity and enabling data driven decisions but in addition will shorten time to oil which seems to be the next challenge of the industry. © 81st EAGE Conference and Exhibition 2019 Workshop Programme. All rights reserved." "56610006900;57202441572;57210218564;14052416300;","RoofN3D: A database for 3d building reconstruction with deep learning",2019,"10.14358/PERS.85.6.435","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85069962419&doi=10.14358%2fPERS.85.6.435&partnerID=40&md5=1232a616325793faf535573bd423c446","Machine learning methods, in particular those based on deep learning, have gained in importance through the latest development of artificial intelligence and computer hardware. However, the direct application of deep learning methods to improve the results of 3D building reconstruction is often not possible due, for example, to the lack of suitable training data. To address this issue, we present RoofN3D which provides a three-dimensional (3D) point cloud training dataset that can be used to train machine learning models for different tasks in the context of 3D building reconstruction. The details about RoofN3D and the developed framework to automatically derive such training data are described in this paper. Furthermore, we provide an overview of other available 3D point cloud training data and approaches from current literature in which solutions for the application of deep learning to 3D point cloud data are presented. Finally, we exemplarily demonstrate how the provided data can be used to classify building roofs with the PointNet framework. © 2019 American Society for Photogrammetry and Remote Sensing." "57210202528;57210205399;","Continuous authentication using smart health monitoring system",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85069929615&partnerID=40&md5=a6bb79c399bf3e9dd808737bab68d1d3","The main security issue in most of the computer is user identification and authentication. In traditional authentication schemes, the user is only validated during initial login which provides low security for the system. Hence continuous authentication has to be done to resolve the security issue. With increase in the number of smartphone users, continuous validation of the authenticated user is important. Continuous authentication mechanism can be made using two behavioral traits: app usage and touch based. There is a worldwide increase in the usage of apps that works on touchscreen, hence both can be used for authentication. Hence, the continuous authentication will be based on app usage and touch screen based which provide high security. In this paper, smart health monitoring system is used for continuous authentication. The data which is collected from wearable biomedical sensors for continuous health monitoring can also be used for continuous authentication. Although the biomedical signals are not highly discriminative a robust machine learning to obtain high accuracy levels is used. An android app is developed to gather data and send to cloud for data storage. The user is validated based on the decisions from the classifiers. The proposed work does not need any extra model for data collection as it uses the data gathered for health monitoring purpose, it can be used for low cost applications. © BEIESP." [No author id available],"2018 IEEE Region 10 Symposium, Tensymp 2018",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065105334&partnerID=40&md5=00c5515dd02d1e2a858fd90f4374dfe8","The proceedings contain 58 papers. The topics discussed include: optimal overcurrent relay coordination of a multi-loop distribution network with distributed generation using dual simplex algorithm; a customized system to assess foot plantar pressure: a case study on calloused and normal feet; distinguishing between cyber injection and faults using machine learning algorithms; computing the relations among three views based on artificial neural network; on-chip transient detection circuit for microelectronic systems against electrical transient disturbances due to ESD events; sparsity representation of beat signal in weather radar for compressive sampling; development of a compact-sized biodigester for pig manure and organic wastes with raspberry pi-based temperature, pressure, and pH level monitoring; an effective hexapod robot control design based on a fuzzy neural network and a kalman filter; potential of IoT system and cloud services for predicting agricultural pests and diseases; and performance evaluation of ideal nearest replica routing (NRR) against several forwarding strategies on named data networking(NDN)." "57204728675;57206315647;","Overcoming challenges in connected autonomous vehicles development: Open source vehicular data analytics platform",2019,"10.4271/2019-01-1083","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85064682024&doi=10.4271%2f2019-01-1083&partnerID=40&md5=32d2d20489e2058beb09f81fe00dcf5f","Data Science and Machine Learning are buzzwords in our everyday lives, as is evident from its applications, such as voice recognition features in vehicles and on cell phones, automatic facial and traffic sign recognition, etc. Analyzing big data on the basis of internet searches, pattern recognition, and learning algorithms, provides deep understanding of the behaviour of processes, systems, nature, and ultimately the people. The already implementable idea of autonomous driving is nearly a reality for many drivers today with the aid of ""lane keeping assistance"" and ""adaptive cruise control systems"" in the vehicle. The drift towards connected, autonomous, and artificially intelligent systems that constantly learns from big data and is able to make best-suited decisions, is progressing in ways that is fundamental to the growth of automotive industry. The paper envisages the future of connected-and-autonomous-vehicles (CAVs) as computers-on-wheels. These are pictured as sophisticated systems with sensors on board as data sources and a lot of other functions and running services to support autonomous-driving. These services are considered to be computationally expensive. The unit performing on-board has limited computing resources while on the other hand, the cloud-based architecture has unconstrained resources, but it suffers from extended unexpected latency that requires large-scale internet data transfer. To deal with this conflicting scenario, Open-Vehicular-Data-Analytics-Platform (Open VDAP) for CAVs may be used. This allows CAVs to detect dynamically status of each service, computation-overhead and the optimal-offloading-destination such that each service could be finished within an acceptable-latency. Open VDAP is an open-source platform that offers free APIs and real-field vehicle data to researchers and developers in the community, allowing them to deploy and evaluate applications in real environment. © 2019 SAE International. All Rights Reserved." "56461352400;57210411252;57210414069;57210418335;","Prediction and detection of heart attack using machine learning and internet of things",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85070610136&partnerID=40&md5=c328e679e2016e0c987aa1dc9e36fe39","In today’s modern world Cardiovascular infections are the basic reason for death worldwide in the course of the most recent couple of years in the developed as well as developing countries. Early recognition of heart diseases and persistent supervision of clinicians can lessen the death rate. Be that as it may, the exact discovery of heart diseases and meeting of a patient for 24 hours by a specialist isn’t accessible since it requires more insightfulness, time and aptitude. In this examination, a speculative plan of a cloud-based coronary illness identification and forecast the framework had been proposed to identify coronary illness utilizing Machine learning strategies. For the exact recognition of the coronary illness, a productive AI procedure ought to be utilized and we utilize numerous relapses for the expectation of heart attack illness. We use this algorithm to predict heart attack by taking different independent variables and we take pulse beat time to time as it varies from time to time. We use multiple regression to predict heart attack and we use IOT to communicate to the person and we use IOT devices and cloud platform in order to remind the person about his health condition of a heart attack. Besides, to screen the coronary illness tolerant a constant patient checking framework was created and displayed utilizing Arduino, able to do taking a parameter like a heartbeat rate utilizing beat sensor IOT device. The created framework can transmit the recorded information to a server which is updated at regular intervals. In this paper, I clarified the engineering for pulse or heartbeat rate and other information observing system and I likewise disclosed how to utilize an AI calculation like MULTIPLE REGRESSION calculations to foresee the heart attack by utilizing the gathered pulse information and another wellbeing related edge and how we use IOT devices for the location of a heart attack. © BEIESP." "57209471021;57209463826;57209472902;57209474569;","A machine learning algorithm for jitter reduction and video quality enhancement in IoT environment",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85067872532&partnerID=40&md5=fbf10404bb62e7f9b12ebc5387ea4c1a","Multimedia traffic has been abnormally increasing nowadays due to its greater usage and necessity. CCTV cameras (closed circuit television camera) are widely used in these days as it a matter of security concern. From shopping malls to home the usage CCTV camera plays a vital role. But the challenging part arises when the media data captured by the camera is to be transmitted to the display monitors of the owner. There are scenarios where more than one CCTV camera covers a particular region. The 70% of network traffic is caused due to CCTV surveillance. It is important to reduce the traffic and delay packets to deliver the data to the user on time. We have used wireless SDN network (software defined network) to transfer the multimedia data to the display monitor. The SDN control switch is integrated with AI module where machine learning algorithm is implemented (BAT algorithm) to prioritise the data packets. For more efficient storage data is uploaded to the IOT cloud. © BEIESP." "55665406600;","Advanced IT-based Future Sustainable Computing (2017-2018)",2019,"10.3390/su11082264","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85066877682&doi=10.3390%2fsu11082264&partnerID=40&md5=df013124a3524621672aaf1c1caa9799","Future Sustainability Computing (FSC) is an emerging concept that holds various types of paradigms, rules, procedures, and policies to support breadth and length of the deployment of Information Technology (IT) for abundant life. However, advanced IT-based FCS is facing several sustainability problems in different information processing and computing environments. Solutions to these problems can call upon various computational and algorithmic frameworks that employ optimization, integration, generation, and utilization technique within cloud, mobile, and cluster computing, such as meta-heuristics, decision support systems, prediction and control, dynamical systems, machine learning, and so on. Therefore, this special issue deals with various software and hardware design, novel architectures and frameworks, specific mathematical models, and efficient modeling-simulation for advance IT-based FCS. We accepted eighteen articles in the six different IT dimensions: machine learning, blockchain, optimized resource provision, communication network, IT governance, and information security. All accepted articles contribute to the applications and research in the FCS, such as software and information processing, cloud storage organization, smart devices, efficient algorithmic information processing and distribution. © 2019 by the authors." "57207992885;57207981392;57207982328;57193866840;57193770709;","Leading Edge Assembly Real Time Process Monitoring Using Industrial Internet of Things (IIoT)",2019,"10.4271/2019-01-1367","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85063433788&doi=10.4271%2f2019-01-1367&partnerID=40&md5=00e610f6f0409f45874cd5092949940c","The increasing global demand for commercial aircraft creates many new challenges in manufacturing including an increased need to maximize the automation of manufacturing processes. The purpose of this research is to develop the understanding of leading edge assembly processes using robot mounted tooling and an automated fixture with advanced process monitoring. Within this research real-time process monitoring data is acquired from an assembly operation and processed into an open cloud environment enabling advanced data analytics. Implementation of advanced analytics utilising process data could be developed for the use of machine learning algorithms which can lead to superior fault finding. The aim of this research is to improve product quality, reduce cost and increase process knowledge, enabling the potential for maximized online and offline process feedback. This paper details how an open Industrial Internet of Things (IIoT) environment can be used to interface with a number of proprietary devices to enable real-time process monitoring in aerospace manufacturing. The implementation of the software and hardware is detailed and followed by an initial evaluation of the system architecture by performing leading edge assembly operations. In addition, a baseline of current IIoT systems is discussed, with the comparison to the specific architecture used here. The system is based on a wireless integration of proprietary devices and sensors feeding real-time data to an open cloud environment. Data is analysed and visualised in real time with online access and report generation. Data is supplied from the automated fixture and Restricted Access Drilling Unit (RADU) end-effector prototype developed by the Manufacturing Technology Centre. Supplied process data includes hardware and system health, and environmental measurements during the assembly operation. Access to analysed data grants the ability to identify occurring abnormalities during the assembly process which, ultimately, will allow for advancements in increased component quality and reduce manufacturing costs. © 2019 SAE International. All Rights Reserved." "57204728675;","Autonomous Connected Aerial Vehicles: Overview of Application of Open Source Vehicular Data Analytics Platform to Urban Air Taxi",2019,"10.4271/2019-01-1351","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85063426508&doi=10.4271%2f2019-01-1351&partnerID=40&md5=9130b4c2d13e91c9cda41df96eb3a12d","Data Science and Machine Learning are playing a key role in our everyday lives, as is evident from its applications, such as voice recognition features in vehicles and on cell phones, automatic facial and traffic sign recognition etc. Analysing big data on the basis of searches, pattern recognition and learning algorithms provides deep understanding of the behaviour of processes, systems, nature, and ultimately the people. The already implementable idea of autonomous driving is nearly a reality for many drivers today with the help of lane keeping assistance and adaptive cruise control systems in the vehicle. The drift towards connected, autonomous, and artificially intelligent systems that constantly learns from the big data and is able to make best suited decisions is advancing in ways that are fundamentally important to many automotive industries. The paper envisages the future of connected-autonomous-urban-airtaxi (CUTs) as computers on wheels. These are pictured as sophisticated systems with on-board sensors as data sources and a lot of other functions and services running to support autonomous aerial driving/flying. These services are considered to be computationally expensive. The on-board computation unit has limited computing resources. On the other hand, the cloud-based architecture has unconstrained resources but it suffers from unexpected extended latency that leads to the large-scale Internet data transmission. To deal with this dilemma, Open Vehicular Data Analytics Platform (OpenVDAP) for CUTs may be used. This allows CUTs to dynamically detect status of each service, computation overhead and the optimal offloading destination so that each service could be finished within an acceptable latency. OpenVDAP is an open-source platform that offers free APIs and real-field aerial vehicle data to the researchers and developers in the community, allowing them to deploy and evaluate applications on the real environment. © 2019 SAE International. All Rights Reserved." "57200918295;24072139200;57207450851;56545781200;8632957700;","The utilisation of cloud computing and remote sensing approach to assess environmental sustainability in Malaysia",2019,"10.1088/1755-1315/230/1/012109","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85062512275&doi=10.1088%2f1755-1315%2f230%2f1%2f012109&partnerID=40&md5=d41eaa12c6a1016de50e502fce2b0f30","Monitoring of the environment over a large area will require huge amount of data. The implementation of conventional methods will be time consuming and very costly. Furthermore, one way to assess the environmental sustainability is by analysing the changes of land cover over two different periods. Therefore, this study implemented a cloud computing approach utilising an open source Remote Ecosystem Monitoring Assessment Pipeline (REMAP) to map the changes of Land use land cover (LULC) over Peninsular Malaysia utilising Landsat data obtained from two different periods (2003 and 2017). This approach has utilised a powerful inbuilt machine learning algorithm, Random Forest (RF) to test the performance of cloud computing using REMAP to produce LULC maps over Peninsular Malaysia. The results showed an acceptable LULC maps and the changes between two periods were analysed. Therefore, the utilisation of machine learning algorithm with the integration of cloud computing using REMAP can reduce the cost, lessen the processing time, produce LULC maps, and perform change analysis over large area. © 2019 Institute of Physics Publishing. All rights reserved." "57209835052;57073683700;56586072300;","An analysis of user oriented behaviour based malware variants detection system",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85068856096&partnerID=40&md5=79edbcdf8bfbca7093791a8bb0d64339","A virtualized infrastructure (VI) is implemented through one or more virtual machines that depend on built-in software defined multiple instances of hosting hardware. This infrastructure model added an advantage of gathering different computing resources and on-demand resource scaling that facilitated extensive deployment of VI to cloud computing services. BDSA finds the potential attack and protect in VI opposing vulnerabilities. HDFS is used to store the backend information. Security analytical algorithm applied on logs captured at various points within network to identify the attach existence. Ref[1]Graph based event correlation and Map Reduce Parser methodologies are used to identify the attack paths through the network logs obtained.Ref[1] Two step machine learning is used to determine the attack presence, attack’s conditional probability based on attributes is calculated through logistic regression and existence of attack on network is calculated through belief propagation. This has steered way for cyber attackers to launch attacks for illegal access on virtualized infrastructures. © BEIESP." "57204927880;57205443317;57205445803;","Methodology to Recognize Vehicle Loading Condition - An Indirect Method Using Telematics and Machine Learning",2019,"10.4271/2019-26-0019","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85060051881&doi=10.4271%2f2019-26-0019&partnerID=40&md5=f2b0edb50cf4b2952b03c9420d71b8a2","Connected vehicles technology is experiencing a boom across the globe. Vehicle manufacturers have started using telematics devices which leverage mobile connectivity to pool the data. Though the primary purpose of the telematics devices is location tracking, the additional vehicle information gathered through the devices can bring in much more insights about the vehicles and its working condition. Cloud computing is one of the major enabled for connected vehicles and its data-driven solutions. On the other hand, machine learning and data analytics enable a rich customer experience understanding different inferences from the available data. From a fleet owner perspective, the revenue and the maintenance costs are directly related to the usage conditions of the vehicle. Usage information like load condition could help in efficient vehicle planning, drive mode selection and proactive maintenance [1]. A common approach to vehicle load condition detection is by using exclusive load sensing sensors. This paper explores a possibility of detecting vehicle load conditions without making use of any sensors. Instead, a supervised machine learning model is developed to recognize real-time loading condition, by analyzing vehicle driving behavior. This paper covers a machine learning based approach for load detection of small commercial vehicles, which are less then 1Ton of loading capacity. In this study, the focus is given to differentiating the vehicle behavior at different loading conditions and to select the accurate parameters for machine learning model development. These selected features are based on the domain expertise in vehicle dynamics and statistics of the data. The output of this novel method can be used for optimizing different ADAS functionalities [2] at very low-cost, leveraging telematics units. © 2019 SAE International. All Rights Reserved." "36158741000;57203041761;57215132765;","Drainage strategy optimization - Making better decisions under uncertainty",2019,"10.2118/196683-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088773612&doi=10.2118%2f196683-ms&partnerID=40&md5=4c332c53cbffda633bbd17287434e192","Improved reservoir knowledge is key to extracting additional value from existing oil and gas assets. However, given the uncertainty in the subsurface, it is always a question if our current development strategy is the most robust choice, or if there are alternatives that can further increase the value of our field. This paper presents a novel solution that enables the asset team to answer these questions in a new way. Furthermore, the solution helps teams quickly identify and screen new opportunities that ultimately increase both subsurface understanding and the value of the field. The solution combines a quasi-Newton gradient based numerical optimization scheme with a stochastic simplex approximate gradient (StoSAG) algorithm. Because the algorithm is non-intrusive with respect to the fluid flow simulator, we can directly apply the solution on any flow optimization problem without the need to access the simulator source code. The solution is implemented using a microservice architecture that allows for efficient scaling and deployment either on cloud-based or internal systems. We demonstrate the proposed solution on a field containing 11 oil producers and 7 water injectors by optimizing the water injection and oil production rates. The machine learning algorithm allows us to quickly explore different drainage strategies, given the current understanding and associated uncertainties of the reservoir. Specifically, the software solution suggests that 6 of the 18 pre-defined well targets are high risk and/or of little value. Running a second development scenario where we do not drill these six wells reduces the investment cost of this field by 163 MUSD and increases the expected net present value per well of the field by 48 percent. Compared with the reactive control drainage strategy approach, we increase the expected net present value of the field by 9.0 %, while simultaneously lowering the associated risk. Copyright 2019, Society of Petroleum Engineers." "57215883934;55356184400;","Pipeline corrosion failure probability assessment from qualitative inspection leveraging artificial intelligence",2019,"10.2118/197484-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088771496&doi=10.2118%2f197484-ms&partnerID=40&md5=0e35767a5ab9ec0da127560e483cc55d","Assessing asset corrosion related failure risks without involving expensive testing, field trials, and intensive data mining followed by complex computational analyses is a significant challenge in Oil & Gas industry. More reliably, such assessment is still governed by practical experience and fundamental conception of the main corrosion mechanisms, though often with the significant advancement of corrosion failure risk assessment and computer technology. To that end the practicality of artificial intelligence and machine learning (AIML) techniques leveraging the characteristics inspection data to predict corrosion failure mimicking human practical experience were investigated. This paper attempts to merge the two complex areas of corrosion engineering and computer aided analyses for the direct beneficiary of the fast-growing discipline of corrosion and integrity management (C&IM). Over the past few years, the authors from their respective areas of expertise have combined forces to learn about each other's world and help make tangible progress and resolve common problems in pipeline engineering. In this paper, qualitative risks due to external and internal corrosion of pipeline were studied and assessed using six different artificially intelligent enabled machine learning algorithms leveraging the existing PHMSA Hazardous Liquid Accident's report. Relative importance insights or governing mechanisms that results in corrosion related failure were identified utilizing several AIMC algorithms. Where relevant and possible semi quantitative risk assessments have been appraised to perform a 'calibration check' on the findings determined. This is best done by both verification and validation (V&V). The former is usually a desktop or laboratory exercise, and the latter often a field trial. Based on a comparative study using various learning algorithms like Gradient boosting, Deep Learning, Support Vectors, Random Forest, Decision Trees and Linear regression, it is observed that a specific corrosion mechanistic pipe failure can be predicted within 97% accuracy. The process is automated in a cloud-based web application form for prediction from raw characteristics data of accident report, along with a Corrosion SME review of logic and reasonability. High accuracy and reliability demonstrate that use of artificial intelligent would enable decision making engineering and project managers to give a faster and more confident assessment of corrosion risks. © 2019, Society of Petroleum Engineers" "57200247152;57203115730;57203121765;56593035700;7004709202;","Automated pressure transient analysis: A cloud-based approach",2019,"10.2523/iptc-19060-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85088069110&doi=10.2523%2fiptc-19060-ms&partnerID=40&md5=0d29cee62fab00b56bd60b06e9f02037","Pressure transient analysis provides useful information to evaluate injection induced fracture geometry, permeability damage near wellbore and pressure elevation in injection zone. Manual analysis of pressure data after each injection cycle could be subjective and time-consuming. In this study a cloud-based approach to automatically analyze pressure data will be presented, which is aimed to improve the reliability and efficiency of pressure transient analysis. There are two fundamental requirements for the automated pressure transient analysis: 1) Pressure data needs to be automatically retrieved from field sites and fed to analyzer; 2) Analyzer can automatically select instantaneous shut-in pressure (ISIP), identify flow regimes, and determine fracture closure point. To meet these requirements and also take the advantages of cloud storage and computing technologies, a web based application has been developed to pull real time injection data from any field sites and push it to a cloud database. Besides analyzing any existing pressure data in the cloud database, a built-in pressure transient analyzer can also detect any real-time pressure data and perform pressure analysis automatically when required data is available. The automated, cloud-based pressure transient analysis has been applied to multiple injection projects. In general, the analysis results including permeability, fracture half length, skin factor, and fracture closure pressure are comparable to these yielded from manual analysis. The discrepancy is mainly caused by poor data quality. The inconsistent selections of ISIP and different slopes defined for G-function and flow regime analyses also contribute to the divergence. Overall, the automated pressure transient analysis provides consistent results as the exact same criteria are applied to the pressure data, and analysis results are independent on analyzer's experience and knowledge. In addition, machine learning algorithms are applied to continuously refine the criteria and improve the quality of analysis results. As data from oil/gas industry increases exponentially over time, automated data transmission, storage, analysis and access are essential to maximize the value of the data and reduce operation cost. The automated pressure transient analysis presented here demonstrates that cloud storage and computing combined with automated analysis tools is an optimal way to overcome big data challenges facing by oil/gas industry. © 2019, International Petroleum Technology Conference" "57214910519;57201316372;57214915718;","Seismic data management for big data era",2019,"10.2118/197369-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086058520&doi=10.2118%2f197369-ms&partnerID=40&md5=2ac570dfb21123763485da58b17ca338","Seismic data is one of earliest data acquired in a prospect evaluation and the data are utilized throughout the exploration and production stages of a prospect. With recent advances in the handling of big data, it is essential to re-evaluate the best practices in the seismic data ecosystem. This paper presents the idea to leveragingthe technology advancement in big data and cloud computing for Seismic data ecosystem with the aim to providing an improve user experience. This new seismic platform would be capable of handling, managing and delivering the full spectrum of seismic data varieties starting from acquired field data to interpretation ready processed data. The system to have the following capabilities: • Capability to entitle the right portion of data to every user as per interest • Organization of seismic data as per the business units • Data security by sharing data only with legitimate users/groups. • Direct or indirect integration with all the data sources and applications who are consuming and/ or generating data • Sharing of and collaboration on data within company and/or across organization for shareholding partner, perspective seismic buyer for trading and relinquishment, regulatory agency resource certifying agencies and service providers etc. over limited network connectivity. • Provide intergration/data deliverivey to End Users applications where this seismic data will be utilizaed Implementation of Seismic ecosystem will enable: • Sharing of seismic data by the acquisition, quality control, data processing and interpretation with user communities from one centralized storage • Collaboration of stake holders in real time over an encrypted network • Leveraging cloud and mobility technology advancement for agility and interaction. The system will be connected and interactive yet has the power of complex high-performance computing infrastructure on the background. • Data delivery and auditing to wider and more diverse user community that consumes data from different platforms. • Secure data access based on organizational business units to make sure data does not fall into unauthorized hand. • Reduction in seismic data turnaround time by reading and ingesting large volume of data through parallel input/output operation. • Improved data delivery and map interface with contextual information out of the centralized data store. • Augment traditional workflows with machine learning and artificial intelligence for example automated fault detection, etc., The proposed best practice aims to bring all of the different disciplines working with seismic data to one centralized seismic data repository and enable them to consume and share seismic data from big data lake. This is live and interactive when compared to traditional technologies of using the archive and restore system in standalone application. © 2019, Society of Petroleum Engineers" "57215884104;57215883959;57205425645;57205427894;36696718600;26026419000;57215884723;6602822699;36053538900;","Embracing the digital and artificial intelligence revolution for reservoir management - Intelligent integrated subsurface modelling IISM",2019,"10.2118/197388-ms","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084780776&doi=10.2118%2f197388-ms&partnerID=40&md5=ff7e24e91e17f48a296b22c93f594d6e","Challenges associated with volatile oil and gas prices and an enhanced emphasis on a cleaner energy world are pushing the oil and gas industry to re-consider its fundamental existing business-models and establish a long-term, more sustainable vision for the future. That vision needs to be more competitive, innovative, sustainable and profitable. To move along that path the oil and gas industry must proactively embrace the 4th Industrial Revolution (oil and gas 4.0) across every part of its business. This will help to overcome time constraints in the understanding and utilization of the terabytes of data that have been and are continuously being produced. There is a clear need to streamline and enhance the critical decision-making processes to deliver on key value drivers, reducing the cost per barrel, enabling greater efficiencies, enhanced sustainability and more predictable production. Latest advances in software and hardware technologies enabled by virtually unlimited cloud compute and artificial intelligence (AI) capabilities are used to integrate the different petro-technical disciplines that feed into massive reservoir management programs. The presented work in this paper is the foundation of a future ADNOC digital reservoir management system that can power the business for the next several decades. In order to achieve that goal, we are integrating next generation data management systems, reservoir modeling workflows and AI assisted interpretation systems across all domains through the Intelligent Integrated Subsurface Modelling (IISM) program. The IISM is a multi-stage program, aimed at establishing a synergy between all domains including drilling, petrophysics, geology, geophysics, fluid modeling and reservoir engineering. A continuous feedback loop helps identify and deliver optimum solutions across the entire reservoir characterization and management workflow. The intent is to dramatically reduce the turnaround time, improve accuracy and understanding of the reservoir for better and more timely reservoir management decisions. This would ultimately make the management of the resources more efficient, agile and sustainable. Data-driven machine learning (ML) workflows are currently being built across numerous petro-technical domains to enable quicker data processing, interpretation and insights from both structured and unstructured data. Automated quality controls and cross domain integration are integral to the system. This would ensure a better performance and deliver improvements in safety, efficiency and economics. This paper highlights how applying artificial intelligence, automation and cloud computing to complex reservoir management processes can transform a traditionally slow and disconnected set of processes into a near real time, fully integrated, workflow that can optimize efficiency, safety, performance and drive long term sustainability of the resource. © 2019, Society of Petroleum Engineers" "57205425143;57214913847;57214915820;","Using machine learning-based predictive models to enable preventative maintenance and prevent ESP downtime",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084166530&partnerID=40&md5=4ef62b9d3311b8f58ed52aeabb45c0de","This paper focuses on the use of artificial intelligence (AI) and machine learning (ML) algorithms to implement anomaly detection and shows how this concept can be extended to implement autonomous well surveillance. Today, critical equipment is monitored by implementing automation and control systems with built-in protection logic for safe operation of equipment and for shutdown of the system in the event that operation deviates outside of valid process conditions. These automation and control systems require constant surveillance by a human operator to verify that all processes are running normally. It is the human operator's responsibility to react to any alarm conditions that occur during operation. Often, the alarm trigger event occurs without early notification and the operator has a short period of time to react. This way of controlling operations requires skilled operators with a great deal of experience to monitor and control the system in an effective manner. It also limits the amount of time the operator can allocate toward optimization. Autonomous surveillance is the concept of training an AI system to provide early detection of abnormal behavior. In this way, the system can take over the task of constant surveillance of process operation, leaving the experienced human operator with additional capacity to focus his time on more productive actions. For example, the human operator can use a combination of process data and decision support information from the AI system to consider current operating conditions and implement a more optimized setpoint, which could increase production or extend equipment life expectancy. In this paper, an example is presented, which is based on monitoring electric submersible pumps (ESPs) using a deep learning neural network that was trained on historical data from the process control system historian. In addition to outlining the benefits that AI-assisted surveillance provides when compared to conventional methods, the paper provides a system architecture blueprint for implementing an autonomous monitoring application across different types of ESP fleets by connecting sensor data directly to a cloud-based monitoring system. The paper builds on the work of previous SPE papers by providing up-to-date results of a pilot project, where a predictive maintenance model has been running for 10 months. On the project, 30 ESPs ranging in power from as low as 200-kW to as high as 500-kW were deployed and monitored using an AI-supported predictive maintenance model. To date, the results have been extremely positive. In one case, an ESP interruption was predicted by the application 12 days before the actual failure occurred. Following several months of testing/use of the ML-based predictive maintenance solution during the pilot deployment, the ESP fleet operator concluded that the system can detect multiple kinds of anomalies in advance, even previously unknown ones. This capability is a significant distinction between the AI-based model and conventional models employed by other ESP diagnostic tools. Although these new, unknown types of complex ESP operational anomalies were difficult to interpret as to their root causes, they could still have led to ESP performance degradation and possible failure nonetheless, if not mitigated or remediated. © 2019, Society of Petroleum Engineers" "56135789500;57194177429;10340225800;56136748300;","Traffic-related air pollution monitoring laboratory with computer vision based traffic count and cloud based data visualization techniques",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084019676&partnerID=40&md5=f3a4a474335ff15727f4ccee64edd0bf","Motor vehicle exhaust is a significant source of urban air pollution. The most widely reported pollutants in vehicular exhaust include carbon monoxide, nitrogen oxides, unburned hydrocarbons, particulate matter, polycyclic aromatic hydrocarbons, and other organic compounds. People living or otherwise spending substantial time near busy highways are exposed to these pollutants and it is becoming increasingly important to understand the emission and health effects of road traffic. Many roadside areas have no or limited sustained monitoring capabilities, making it difficult to quantify the major pollution sources. Real-time mobile air monitoring has been demonstrated to have an advantage of identifying spatial and temporal differences of on-road traffic pollutants from different road types, traffic intensities, and road features. The Houston Advanced Research Center (HARC) has developed the Mobile Acquisition of Real-time Concentrations (MARC) laboratory, which consists of a Ford F-350 passenger van outfitted with a Proton Transfer Reaction Mass Spectrometer (PTR-MS), NOx analyzer, CO analyzer, GPS unit, camera, and metrological equipment. Front view camera mounted on the front of the van is used to get the traffic intensities using a machine learning based computer vision algorithm by tracking and counting the number of the vehicle in the vicinity. This allows for the fast measurement of geotagged gaseous pollutant concentrations correlated with traffic intensities. The data recorded from the van is uploaded to an off-site database via an adaptive interface that supports 3G/4G infrastructure and the information is broadcasted to a website in real-time. MARC is used during 2018 the Health Impact Assessment (HIA) of the North Houston Highway Improvement Project (NHHIP) in Houston, TX. The main goal of the study was to quantify and explain the potential co-benefits and co-harms to health associated with the proposed changes to the I-45 corridor from Beltway 8 North to the I-69 spur south of downtown Houston. © 2019 Air and Waste Management Association. All rights reserved." "57211205900;","Remote sensing for leak detection and quantification",2019,"10.2118/195794-MS","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084017163&doi=10.2118%2f195794-MS&partnerID=40&md5=a0c858aa0e2d60ae27a8516f547250e1","Offshore oil and gas installations are (by their nature) located in remote locations that are both difficult and costly to access. While such challenges exist, the operate & maintain requirements associated with such assets are consistent and must be addressed, requiring operators to identify the most efficient form of service to reduce staffing levels, risk and cost. Offshore hydrocarbon production assets commonly incorporate equipment and processes that can lead to significant (fugitive) gas emissions. The consequences are both economic and social (environmental) in nature, requiring operators to perform emissions surveys with the objective of leak identification and remediation within the shortest possible timeframe. The frequency of this activity is naturally limited and must be balanced with the staffing and operating needs of the broader facility, which in-turn can lead to sub-optimal leak detection to fix timing and reliability. Addressing the three key challenges of access productivity, detection reliability and results quantification, Worley has developed a remote sensing platform that incorporates the use of productive remote access equipment such as unmanned aerial vehicles (UAV) and in-situ monitoring, with machine based emissions detection and algorithmic quantification to provide a solution that allows the operator to increase survey frequency, obtain more reliable results at lower cost, and perform the work in a manner consistent with safe and low-risk operations. In both testing and field deployments, the results have provided for significant reductions in both false positive and negatives and have produced datasets that allow for accurate indications of greenhouse gas reduction via comparison of volumetric emissions before and after leak repair activity has taken place. The technology is largely mathematical, utilizing coded routines for machine learning to perform gas detection under (initially) supervised modeling conditions, and algorithmic gas dispersion models for further emission quantification. The performance of the survey is typically carried out through the integration of existing, proven manufactured sensing equipment across several types of UAV or in-situ monitors which collect field data for transmission to a cloud-based portal which further processes the results. The approach has been shown effective in accessing hard or costly to reach areas, improving survey productivities, while the data processing and quantification allows the operator to benefit from improved measurability and prioritize leak repair accordingly. Copyright 2019, Society of Petroleum Engineers" "57211206961;57211206666;","Machine learning image analysis for asset inspection",2019,"10.2118/195773-MS","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084013834&doi=10.2118%2f195773-MS&partnerID=40&md5=ce3a7d7b71241f6488d0bf3b9067903e","Total E&P UK Ltd and Merkle have undertaken a proof of concept project to investigate machine learning image analysis applied to image and video data from onshore and offshore sites, in preparation for the deployment of an autonomous asset inspection ground robot in 2019. The aim is to better understand the feasibility of these methods, and to demonstrate the benefits of robotic inspection with regard to improving safety and efficiency, enhancing data capture and reducing operational costs. An object detection model was developed based on high-performance open source algorithms. Transfer learning was applied using a custom-built image library and the result is a model able to detect a range of different types of items in the industrial environment including mobile equipment, process equipment, infrastructure and personnel (with and without PPE). The object detection model is used to feed into object classification anomaly detection models to look at the state of selected pieces of identified equipment, such as whether a valve is open or closed, which can be placed in the context of the expected state of the process equipment by relating it to the digital twin for the asset. An additional object detection model was developed to operate as a gas leak detection system for infrared cameras. The object detection model achieved good results and model performance was driven by the number and quality of images used for the training. An anomaly detection model designed to detect whether ball valves were open or closed delivered good results, with high accuracy and balanced false positive and false negative detection rates. The overall performance of the infrared gas detection model was restricted by the limited volume and variability of the training data, although the false positive detection rate was very low. A significant part of the machine learning was devoted to the development of a consistent labelled image library for oil and gas equipment, infrastructure and gas leaks. Image transformations were tested but boosting the number of images using transforms gave variable results. Additional training and testing data is needed to ensure that the models are as robust as possible, especially for the gas leak detection model. Once the models are productionised and in use, additional data can be used to periodically retrain the models for improved performance. In addition to the machine learning algorithms, a fundamental aspect of the project is the development of the overall technical architecture, supporting the data science. This includes enabling the transfer of data from inspection robots or other connected camera devices into a data store in a cloud computing environment and returning the results to dashboard systems with different levels of detail depending on user requirements. Copyright 2019, Society of Petroleum Engineers" "57191583173;57214793809;57214793588;57214800500;57214797515;","Blocksat: On-demand access to shared-use satellite constellations",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079138023&partnerID=40&md5=94823c1be896751d15b26f031db23e92","As space launch costs drop and the feasibility of “small-satellite” distributed sensing and imaging improves, we note a growing interest in satellite swarms across academic, government, and industry labs. Many corporate proposals already explore satellite constellations as the backbone of a global, space-based Internet service [1], [2]. However, these constellation proposals are usually owned by a single entity or conglomerate, and operated to serve a particular business model. Building on the theme of crowd-sharing models and innovative research in distributed system-control algorithms, we propose a constellation-management system and multi-functional CubeSat hardware platform that would allow communal use of satellite functionality, including opportunities for non-conflicting concurrent processes and re-purposing of satellite hardware. By coupling communication and data ledger protocols (e.g. modified blockchains) with machine learning algorithms for smart task distribution and execution management, this research describes a way to orchestrate peer-to-peer network collaborations where satellite constellation functionality is dynamically rented, shared or reused between many applications. We envision multi-purpose, shared- use satellite constellations, bringing this category of space hardware into the “on-demand” services market (along the lines of Amazon Web Services, but for broad space applications) and moving beyond more narrow applications of Distributed Ledger Technologies (DLTs) currently deployed in orbit [3]. This model for allocation of a shared research resource, where a ledger is kept to distribute access fairly, builds on years of scientific collaboration processes [4] created around expensive, limited-use hardware resources (e.g., multinational collaborations that split time on large telescopes or the CERN particle accelerator [5]). This research proposes to automate and enhance this process, exploring remote-activation and operations of space-based resources. This paper will discuss the BlockSat mission architecture and ConOps (Concept of Operations), showing how the integration of a multi-functional hardware platform and cloud-computing participation model can create a new class of small satellite operations. We aim to free satellite deployments from the limitations of single-use applications and propose a new “open space” market that can democratize access to LEO space technology. Copyright © 2019 by the International Astronautical Federation (IAF). All rights reserved." "57208753024;56448070900;","Implementation of machine learning methods on FPGA for On-board spacecraft operation",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079124867&partnerID=40&md5=1fec763b3e230fa456042fe41e114983","Machine Learning (ML) techniques are increasingly being used in terrestrial applications. They have also been proposed for various space applications such as on-board data processing, planetary explorations, autonomous operations and various mission-specific applications. Terrestrial application of ML is facilitated by shared resources such as cloud services, powerful desktop computers and framework APIs. However, these enablers are limited in space environment. To take advantage of the progress in ML, space applications need to incorporate on-board inference. The first part of this paper presents a review on some space applications in which such algorithms are suitable or have been used. The review scope is limited to on-board and in-flight applications that have already been realized and doesn't factor in majority of those that are at a proposal stage. ML techniques at ground stations for analysing and processing data is not covered. The review also presents some hardware that has been or can be utilized in these implementations. Though there has been a plethora of hardware platforms geared towards ML inference on the edge, this review notes that Field Programmable Gate Arrays (FPGAs) and associated Systems-on-Chip (SoCs) are more suitable for ML inference in space applications. They also have flight-heritage, having been applied in mission-specific and satellite subsystem operations. They are also reconfigurable and can therefore be adopted for different tasks on the fly. Available models and networks optimized for edge ML inference are presented in this study as well. Such models are more appropriate for space applications since a custom implementation is time-consuming and prone to failure due to tight FPGA fabric timing constraints. Nevertheless, the workflow for a custom Artificial Neural Network (ANN) implementation on an FPGA has been presented. Xilinx Kintex-7 FPGA has been used as the target FPGA in the implementation and evaluation of the network. Vivado IDE and Verilog Hardware Description Language have been used in the demonstration. Copyright © 2019 by the International Astronautical Federation (IAF). All rights reserved." [No author id available],"14th 3D GeoInfo Conference 2019",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85078019390&partnerID=40&md5=23b92da0f5e61589ef44364fe466e3ce","The proceedings contain 19 papers. The topics discussed include: generating 3D city models based on the semantic segmentation of lidar data using convolutional neural networks; voxel-based visibility analysis for safety assessment of urban environments; cityreconstruction from airborne lidar: a computational geometry approach; raise the roof: towards generating LOD2 models without aerial surveys using machine learning; a big data approach for comprehensive urban shadow analysis from airborne laser scanning point clouds; and an improved automatic pointwise semantic segmentation of a 3D urban scene from mobile terrestrial and airborne lidar point clouds: a machine learning approach." "36614966600;54930662100;56055664200;55542665400;55893847400;","Active sampling and model based prediction for fast and robust detection and reconstruction of complex roofs in 3D point clouds",2019,"10.5194/isprs-annals-IV-4-W8-43-2019","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077959601&doi=10.5194%2fisprs-annals-IV-4-W8-43-2019&partnerID=40&md5=293352e189996bf8d36129daedcc3405","3D city models in Level-of-Detail 2 (LoD2) are nowadays inevitable for many applications such as solar radiation calculation and energy demand estimation. City-wide models are required which can solely be acquired by fully automatic approaches. In this paper we propose a novel method for the 3D-reconstruction of LoD2 buildings with structured roofs and dormers from LIDAR data. We apply a hybrid strategy which combines the strengths of top-down and bottom-up methods. The main contribution is the introduction of an active sampling strategy which applies a cascade of filters focusing on promising samples in an early stage and avoiding the pitfalls of RANSAC based approaches. Such filters are based on prior knowledge represented by (non-parametric) density distributions. Samples are pairs of surflets, i.e. 3D points together with normal vectors derived from a plane approximation of their neighborhood. Surflet pairs imply immediately important roof parameters such as azimuth, inclination and ridge height, as well as parameters for internal precision and consistency, giving a good base for assessment and ranking. Ranking of samples leads to a small number of promising hypotheses. Model selection is based on predictions for example of ridge positions which can easily be falsified based on the given observations. Our approach does not require building footprints as prerequisite. They are derived in a preprocessing step using machine learning methods, in particular Support Vector Machines (SVM). © Authors 2019. CC BY 4.0 License." "57212882647;54893965400;57212866934;57212877790;","Potential distribution model of leontochir ovallei using remote sensing data [Modelo de distribución potencial de leontochir ovallei con datos de sensores remotos]",2019,"10.4995/raet.2019.12792","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85077460375&doi=10.4995%2fraet.2019.12792&partnerID=40&md5=2457565e13138944a884c1da5e9a4187","Predicting the potential distribution of short-lived species with a narrow natural distribution range is a difficult task, especially when there is limited field data. The possible distribution of L. ovallei was modeled using the maximum entropy approach. This species has a very restricted distribution along the hyperarid coastal desert in northern Chile. Our results showed that local and regional environmental factors define its distribution. Changes in altitude and microhabitat related to the landforms are of critical importance at the local scale, whereas cloud cover variations associated with coastal fog was the principal factor determining the presence of L. ovallei at the regional level. This study verified the value of the maximum entropy in understanding the factors that influence the distribution of plant species with restricted distribution ranges. © 2019, Universitat Politecnica de Valencia. All rights reserved." "57211500142;57201118368;","IoT platforms for the Mining Industry: An overview",2019,"10.29227/IM-2019-01-47","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074188305&doi=10.29227%2fIM-2019-01-47&partnerID=40&md5=9ff1d7c6c5fa4f5731543fbd9816b59e","Industry 4.0 and the Internet of Things are now very common concepts as solutions that can revolutionize the industry. Constant technological progress increases the possibilities of using computer tools and solutions to support processes in industry and production optimization. The use of the Internet of Things is particularly important in complex processes in mining, enabling the extraction of valuable information from data. The integration of physical facilities in the enterprise enables the digitization of production processes and the increase of efficiency and security. This article presents an overview of the selected internet of things platforms and analytical tools that can be used in industry, with particular emphasis on the mining sector. It is pointed out, that the number of suppliers of IoT technologies and analytical tools offering advanced data analytics services for industry is significant and constantly evolving. The aim of the article is to evaluate selected IoT solutions based on the following criteria: offering predictive analytics, implemented artificial intelligence (AI) or machine learning (ML) algorithms, a mining-oriented process approach, advanced data visualization, interoperability, real-time data capture, remote device management and cloud-based technology. The review was prepared to provide knowledge about IoT vendors operating on the market, as well as to indicate the functionalities that are the most popular among solutions. © 2019 Polish Mineral Engineering Society. All rights reserved." "55927897800;6507214354;35488207000;44061832600;55257235200;6506805816;","Editorial for the Special Issue ""Frontiers in spectral imaging and 3D technologies for geospatial solutions""",2019,"10.3390/rs11141714","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071529952&doi=10.3390%2frs11141714&partnerID=40&md5=eefaea7ac07c55cb355ae2539b1d3e06","This Special Issue hosts papers on the integrated use of spectral imaging and 3D technologies in remote sensing, including novel sensors, evolving machine learning technologies for data analysis, and the utilization of these technologies in a variety of geospatial applications. The presented results showed improved results when multimodal data was used in object analysis. © 2019 by the authors." "36082483100;56132443500;57210283137;","How Cost Effective Is Machine Learning/AI Applied to Leak Detection and Pipe Replacement Prioritization?",2019,"10.1061/9780784482506.029","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85070217696&doi=10.1061%2f9780784482506.029&partnerID=40&md5=41a73bd907f914899b006a883d66c0aa","Artificial intelligence (AI) and machine learning (ML) offer a cost-effective way of evaluating the condition of buried water mains taking in hundreds of variables which is more than normal management consulting practices. Machine learning as a cloud-based solution is able to leverage existing water main pipe data and produce a more accurate model than typical age-based models. Detecting more high-risk water main pipes using ML likelihood of failure (LoF) probability analysis provides improved decision making for directing leak detection, direct pipe inspection, and renewal and replacement activities. This paper explores the applications of AI/ML in underground water pipe distribution systems using example data sets from for a large water utility and a medium sized water utility analyzing the accuracy of an aged-based methodology versus applied ML. © 2019 American Society of Civil Engineers." "44061451800;57206996179;57205673606;57197936167;","Advancing In-line Inspection technology and pipeline risk management through advanced analytics of big data",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85064745247&partnerID=40&md5=15d27070dc69bc1db6fd04d7d5ee75ed","Every year we hear more and more about Big Data, how it's going to be the future, how machine learning is going to change what we are capable of and what we do. However, after all this time, the same questions remain as it applies to our industry: How do I get Big Data? How will it make a difference to integrity management program efficiency and ultimately pipeline safety? And, when will it be here? This paper focuses on answering these simple questions by providing a context for existing, active systems that are successfully employed today in the in-line inspection world and, particularly, their subsequent role in pipeline integrity decision making. Additionally, we'll examine how such computing technologies as Big Data, ""machine learning"", AI, and cloud computing are evolving with the increasing expectation of getting more from the significant volumes of information and performance data available to the pipeline operator and Inline Inspection (ILI) vendors. We'll discuss key considerations necessary for establishing a robust and viable framework needed to manage and analyse the vast quantities of pipeline inspection data being collected every day as a means for performance and reliability improvement, including current and historical digital technologies of both inline and NDE dig excavation systems. Furthermore, how, with an ability to analyse pipeline defect characteristics over a wide spectrum of pipelines and conditions, development processes for advanced characterization and accuracy improvements can be dramatically accelerated relative to historic methods. Insight will be provided as to how advanced analytics are now being further enabled with the massive processing power readily available through these cloud-based systems. Lastly, regarding higher accuracy and more consistent field calibration data, when we combined these Big Data environments with the ongoing improvements in ILI, we will outline how that can open up the possibilities for operators to better understand their pipeline risks and how to prioritize most efficiently when it comes to managing repair and maintenance. © 2019 Clarion Technical Conferences." "57205610424;55538601800;","Mobile observatory uses sight and sound to quanify drainage conditions for O&M and real time control",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85060818314&partnerID=40&md5=30611c27d9063ec55d8edd8cb295753c","Sight&Sound patterns produced by sewer flows can be modeled with Machine Learning to affordable quantify flow parameters at a vast number of drainage sites. Such quantifications produce low cost data to fill in gaps of understanding about very expensive invisible infrastructure lying beneath cities. As a more complete operational understanding of infrastructure becomes affordable, a more cognitive ML system can use that performance data to creation and fine tune a predictive mathematical model for system wide drainage fine tuned by on going flow parameters supplied from affordable flow monitors. With such modeling real time or near real time flow control holds much promise to mitigate overflows into our basements, streets, streams, lakes, rivers and oceans. A final note-the altaStation design also comprehends the need for actuators in sewers. An over-the-air software update available in 2019 will provide closed loop control functions that facilitate cloud based, supervisory control system. Copyright © 2018 Water Environment Federation" "57205427916;57205427465;57205428926;","From insight to foresight: Knowing how to apply artificial intelligence in the oil & gas industry",2019,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85059978832&partnerID=40&md5=d3bc1967b1cbd4b60dc12d1eb62ca26a","We are in an era where digital technologies are developing at exponential rates and transforming industries wholesale. The confluence of machine learning advances, accelerated growth in acquired data, on-demand CPU and GPU driven computing such as cloud infrastructure, and other advances in automation and robotics are causing an industrial revolution that some term as the ""Fourth Industrial Revolution"". Given that all these transformative technologies are now available and rapidly reinventing other industries, why is the rate of adoption in the oil and gas industry so slow? How can we best utilize these advances to stop drowning in data and instead transform this data into information and knowledge in order to enable secure and intelligent automation in oilfield operations? The oil and gas industry has attempted, at times successfully, a multitude of big data and analytical techniques to further describe and analyze the systems' or system of systems' subsurface interactions. While the proofs of concepts have shown promise, structural difficulties embedded in the design of 20th century systems hinder the implementation of the methods and procedures now part and parcel of the 21st century, driven forth because of the Fourth Industrial Revolution. Unfortunately, 20th century procedures are not able to incorporate 21st century driven processes and methods of conducting business. We outline some of the structural challenges facing the oil and gas industry and describe a few of the solutions that have been developed to help companies in the industry. These include applications from the subsurface in geophysics, completions design, and production. Overcoming data silos in traditional data infrastructure requires a novel approach to cloud infrastructure that respects user access, data privacy, and data residency requirements of companies. Assessing data for quality and for reasonable diversity and variation in order to answer questions posed by oil & gas companies can be quite profound. This critical step prevents companies from spending lots of non-productive time and money trying to develop and tune machine learning algorithms to produce answers that are simply not available in the data. Further, getting data to be in a form suitable to apply artificial intelligence can be quite involved. We illustrate the above challenges by several subsurface examples and then describe the implementation of novel solutions. What we will show is that the oil and gas digital highway presently has data traffic jams preventing it from moving at the speed of light. Removing these traffic jams offers decision-makers the Revolution on the previous one while showcasing the concept of [3] Creative Destruction that supplants one technology with another. According to Joseph Schumpeter, the ""gale of creative destruction"" describes the ""process of industrial mutation that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one"". The various Industrial Revolutions are described in Figure 1 as: 1. IR 1.0 was steam power and the advent of industrialization 2. IR 2.0 was the introduction of factories, mass manufacturing and other techniques of mass production 3. IR 3.0 was the semiconductor revolution with advances in computers, electronics and robotics/ automation 4. IR4.0 is about cyberphysical systems and the marriage of hardware and software to produce autonomous systems [4] The Cyber-Physical Systems (CPS) in Figure 2 are integrations of computation, networking, and physical processes. Embedded computers and networks monitor and control the physical processes, with feedback loops where physical processes affect computations and vice versa. Independent systems have existed and are continuously moving towards becoming autonomous in nature, for example, the above surface rotating equipment in the oilfield whose health is monitored in real-time and event triggers manage and alleviate catastrophic failures. [5] Artificial Intelligence is realizing the intelligent interoperations of system of systems defined as a collection of task-oriented or dedicated systems that pool their resources and capabilities together to create a new, more complex system which offers more functionality and performance than simply the sum of the constituent systems. Upstream oil and gas has specialized over the last 100 years into highly developed scientific and commercial niches. While the outcome has been engineering and academic disciplines driving forth accuracy in decisions within the disciplines, the integration of the various niches, which can be stated as the intelligent communication of information between and among systems, has remained a struggle. We discuss this further under the section entitled ""Data Management"". It is easy to recognize that where there is high quality communication of information among systems, artificial intelligence has made significant strides. For example, facial recognition is now used routinely for expediting in airports for expediting immigration lines, online retail and companies like Netflix use artificial intelligence to understand customer preferences and refine and target their offerings, personal assistants like Microsoft Cortana, Amazon Alexa, Google Now, and Apple Siri use voice recognition for many applications such as home automation. opportunity to move from insight to foresight - looking out in front instead of the rearview mirror to drive change. © Copyright 2018, Society of Petroleum Engineers." "57192386104;57205735747;57217749096;57205747497;57205742396;7006093015;","Remote cloud-based automated stroke rehabilitation assessment using wearables",2018,"10.1109/eScience.2018.00063","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85061384336&doi=10.1109%2feScience.2018.00063&partnerID=40&md5=321330a162eaa20badba3d9ace5b91a7","We outline a system enabling accurate remote assessment of stroke rehabilitation levels using wrist worn accelerometer time series data. The system is built based on features generated from clustering models across sliding windows in the data and makes use of computation in the cloud. Predictive models are built using advanced machine learning techniques. © 2018 IEEE." "57205500439;56366080900;22634069200;57037171600;55967354600;","Study of enteromorpha idendification based on machine learning technology",2018,"10.1109/OCEANSKOBE.2018.8559374","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85060296421&doi=10.1109%2fOCEANSKOBE.2018.8559374&partnerID=40&md5=89dddcb3a49d6a55c05ab796718845e0","In this paper, we construct an enteromorpha remote sensing dataset. We take the eight channels data, from Band 1 to Band 8, in Landset 8 as spectral features of the dataset and consider the features extracted by Local Binary Patterns algorithm(LBP) and Gray-level Co-occurrence Matrix algorithm (GLCM) as texture features. Combining these two types of features, we create a novel machine learning dataset including enteromorpha, land, seawater, and cloud samples. Then we use the single-channel threshold method and the multi-channel ratio method to label the entire data set pixel-by-pixel based on the feature of spectral characteristics and ground reflection spectra. Four classifiers are trained via Support Vector Machine (SVM), K Nearest Neighbor (KNN), Expectation Maximization (EM) and Stack Autoencoders (SAE) on our dataset respectively. A new classifier with a set of weight coefficients, which is called weighted classifier, can be obtained by combining above classifiers according to the accuracy of the four classifiers on all kinds of samples. Experiments show that the classification accuracy of these classifiers in our dataset are higher than 90% and weighted classifier reaches 93.59%. If we only focus on enteromorpha data, the best classifier is KNN classifier which has the correct rate of 95.8%. © 2018 IEEE." "57189042712;57203813157;57203820620;57203820084;","Mobile Mechanic - An innovative step towards Digital Automobile Service",2018,"10.1109/ICSCET.2018.8537243","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85059410451&doi=10.1109%2fICSCET.2018.8537243&partnerID=40&md5=bbf6dacc30e06a18b74ad80ea40654a3","In the 21 st century's work engrossed environment, everyone is in a rush to reach their destination on time, nothing should be a hindrance to their way of achieving their goal. Considering a real-life problem where a person faces troubles when their car breaks down in the middle of the road, can really become a major problem especially when this occurs on highways or distant places. This can be solved by using our application which can help the users to find nearby location of service or a repairing center. The application would be subdivided into two main sections namely the Mechanic's portal and the User's portal. Both these portals would require the respective user to either login or register themselves up before using the application. The location can be provided or obtained using the GPS available in mobile phones. The User's portal would display a map with the current location of themselves along with all the available repair shops around them, which shall be consistently updated using the plethora of location services available on Microsoft Azure Cloud. User can get detailed information about these shops along with a route to reach the desired mechanic shop, using Google Map API. Further, the same data shall be utilized for providing interest/ choice based mechanic shops to future users, by implementing Data Analytics and Machine learning available as API's on Azure Cloud Services. This same data can be thereafter be cleansed and provided as employment opportunities for locations which have lesser mechanic reach and for various other assistance, by using Azure Big Data. The application will also be providing a SOS feature for women's safety and a DIY section consisting of videos and documents within the application would let the users try their hands at fixing their own vehicles on occasion of no network coverage or poor internet connectivity. © 2018 IEEE." "56019512200;7102383497;","Data mining for classification of high volume dense lidar data in an Urban Area",2018,"10.5194/isprs-annals-IV-5-391-2018","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85057600992&doi=10.5194%2fisprs-annals-IV-5-391-2018&partnerID=40&md5=2e79e30f940d35407ec0f51fa8852d02","3D LiDAR point cloud obtained from the laser scanner is too dense and contains millions of points with information. For such huge volume of data to be sorted, identified, validated and be used for prediction, data mining provides immense scope and has been used to achieve the same. Certain unique attributes were selected as an input for creating models through machine learning. Supervised models were thus built for prediction of classes through the available LiDAR data using random forest algorithm. The algorithm was chosen owing to its efficiency and accuracy over other data mining algorithms. The models created using random forest were then tested on an unclassified point cloud data of an urban area. The method shows promising results in terms of classification accuracy as overall accuracy of 91.71% was achieved for pixel-based classification. The method also displays enhanced efficiency over common classification algorithms as the time taken to make predictions about the data is reduced considerably for a set of dense LiDAR data. This shows positive foresight of making use of data mining and machine learning to handle large volume of LiDAR data and can go a long way in augmenting efficient processing of LiDAR data. © 2018 Authors." [No author id available],"1st International Conference on Data Science and Analytics, PuneCon 2018 - Proceedings",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85070317866&partnerID=40&md5=327aa6d9ece80221278f9401385fc965","The proceedings contain 71 papers. The topics discussed include: a working paper on use of social media by selected Indian public sector banks; a comprehensive analysis between popular symmetric encryption algorithms; CReP: a packet forwarding protocol in mobile social networks; real-time disease forecasting using climatic factors: supervised analytical methodology; standards and techniques to remove data remanence in cloud storage; using machine learning to find out when to use box-cox transformation on time series data; embedded control system for flash lamp pumped solid-state Nd:glass laser power supply; diabetes readmission prediction using distributed and collaborative paradigms; and text detection from scene videos having blurriness and text of different sizes." "55574821400;6701728686;21741875100;55346329500;15046810600;55587759100;22943323900;57196442844;55407788800;57208247654;","Bridging climate and earth observation data analytics in a federated cloud infrastructure using interoperable multidisciplinary workflows",2018,"10.1109/IGARSS.2018.8519076","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85064210594&doi=10.1109%2fIGARSS.2018.8519076&partnerID=40&md5=0cde5233d7b9caff4a848db4739c1afc","Managing large data sets is a challenge that is compounded when data is distributed across multiple sites. In the case of the combined use of Earth Observation (EO) and climate data, traditional big data approaches are not adequate. This has lead to the development of a new generation of federated cyberinfrastructures, such as the Platform for the Analysis and Visualization of Climate Science (PAVICS), developed by the Computer Research Institute of Montreal (CRIM) and Ouranos, a Montreal-based consortium on regional climatology and adaptation to climate change. PAVICS also reuses key components of Birdhouse, a collaborative project led by the German Climate Computing Center (DKRZ). PAVICS uses standards, such as the Web Processing Service (WPS) developed by the Open Geospatial Consortium (OGC) as a standard interface for providing access to pre-defined geospatial processing services to facilitate integration and foster reuse. This paper presents some of the significant advancements that have been made in recent projects conducted in Canada, in the United States, and in Europe. © 2018 IEEE." "57189391340;6602942477;57200597357;6701615368;","Temporal difference and density-based learning method applied for deforestation detection using ALOS-2/PALSAR-2",2018,"10.1109/IGARSS.2018.8518412","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85063129681&doi=10.1109%2fIGARSS.2018.8518412&partnerID=40&md5=4e9e8a462135c6cd3891427d8cd22096","Remote sensing has established as key technology for monitoring of environmental degradation such as forest clearing. One of the state-of-the-art microwave EO systems for forest monitoring is Japan's L-band ALOS-2/PALSAR-2 which provides outstanding means for observing tropical forests due its cloud and canopy penetration capability. However, the complexity of the physical backscattering properties of forests and the associated spatial and temporal variabilities, render straightforward change detection methods based on simple thresholding rather inaccurate with high false alarm rates. In this paper, we develop a framework to alleviate problems caused by forest backscatter variability. We define three essential elements, namely ""structures of density"", ""speed of change"", and ""expansion patterns"" which are obtained by differential computing between two repeat-pass PALSAR-2 images. To improve both the detection and assessing of deforestation, a ""deforestation behavior pattern"" is sought through temporal machine learning mechanism of the three proposed elements. Our results indicate that the use of ""structure of density"" can introduce a more robust performance for detecting deforestation. Meanwhile, ""speed of change"" and ""expansion pattern"" are capable to provide additional information with respect to the drivers of deforestation and the land-use change. © 2018 IEEE." "57209097561;37015389000;","Augmented Map Based Traffic Density Estimation for Robot Navigation",2018,"10.1109/TENCONSpring.2018.8692049","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065071563&doi=10.1109%2fTENCONSpring.2018.8692049&partnerID=40&md5=14254f2497037f853ce236f0fc3ec826","Most of the work done on robot navigation is focused on map building, localization and obstacle avoidance. In this paper, we particularly focus on intelligent robot navigation based on scene traffic. In this course we initially generate augmented maps using scene traffic flow captured at discrete times of same days, different days and various weather conditions. For facilitating the same, we manually navigate a mobile robot(X80SV) for capturing data(image and sonar map data) across a campus; segmented into zones. Zone specific point cloud maps are then created by merging scene object detection results with sonar map. A Gaussian model is further used for estimating trend for traffic density from the point clouds. This information is later utilized for navigational purposes, to detect the minimum crowded path from a source to destination. Object detection and classification is done using a fine tuned AlexNet. For labelling the training set used for fine tuning, we perform foreground extraction, transfer learning and SVM classification of the scenes. 16 classes of objects including background is used. During realtime navigation, we also perform object detection and classification so that the current traffic trend can be matched with the augmented maps. © 2018 IEEE." "57208553138;55556221300;","Potential of IoT System and Cloud Services for Predicting Agricultural Pests and Diseases",2018,"10.1109/TENCONSpring.2018.8691951","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065055826&doi=10.1109%2fTENCONSpring.2018.8691951&partnerID=40&md5=8e78e66899bbd0081f47cec8372d4f83","Controlling the outbreaks of pests and diseases in agricultural environment, it is still a big challenge to the farmers due to the changing climatic conditions. In this paper we are proposing the alternative method of predicting occurrences of pest and diseases in the plantation, by combining the advantage of IoT farmland monitoring system and Amazon Machine Learning cloud-based services to find hidden patterns into data. Logistic regression algorithm used to train our IoT collected dataset and classify the data with acceptable model quality score, to estimate the diseases forecasting based on sensing technology. © 2018 IEEE." "57217699055;57205580374;57202620886;57188878642;57205578338;","Elasticdocs as an automated information retrieval platform for unstructured reservoir data utilizing a sequence of smart machine learning methods within a hybrid cloud container",2018,"10.3997/2214-4609.201803242","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85093689436&doi=10.3997%2f2214-4609.201803242&partnerID=40&md5=119c8cebc117deec49ef0cfb8202b351","There is a tremendous amount of information available and stored in digital geoscientific documents and published reports in the energy industry. These documents contain a distillation of reservoir information from diverse discipline of geologists, geophysicists, petrophysicists and drillers, that are stored in unstructured format, which find further use in succeeding reservoir modeling stages. In particular, national data management repositories and oil companies hosts these huge amounts of historical well reports containing information such as lithology, hydrocarbon shows, and other reservoir data. Due to the large volume, vintage variety, and non-standardized formats, extraction of valuable information that are used as inputs for interpretation, is an arduous, very time-consuming task. Our solution is to develop ElasticDocs a machine learning-enabled platform in a hybrid cloud container that automatically reads and understand hundreds or thousand of technical documents with little human supervision through a smart combination of machine learning algorithms including optical character recognition (OCR), elatic search, natural language processing (NLP), clustering and deep convolutional neural network. The platform uses a hybrid, 2-tier data service architecture leveraging on the strength of both the strength of local servers and cloud to enhance data security, integrity, and accessibility. © EAGE Reservoir Geoscience Conference, ResGeo 2018. All rights reserved." "57209498999;57194870778;26424302500;56428234100;","Cloud-based borehole image interpretation workflow using machine learning",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85083938338&partnerID=40&md5=b9113ad487a010bfb70ef0fffce3c447","We propose a new workflow for borehole image interpretation incorporating machine learning methods and recent cloud technologies. In order to predict facies from borehole image, clean labels, i.e. the borehole image with annotated facies intervals, must be collected. We propose a solution to accelerate the labelling process by introducing an automatic segmentation and unsupervised learning algorithm based on image texture features. We then present a web based application and pipeline from the user interface to the cloud database that allows the user to easily construct and update the digital facies library and we show first results of our automated image recognition. © 2018 Society of Petroleum Engineers. All rights reserved." "57215911633;57215913812;57217951270;36781870300;","Image processing and restriction of video downloads using cloud",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082353982&partnerID=40&md5=cf8cd87311f13e313084c5958f2ba4fc","Flower image classification using deep learning and convolutional neural network (CNN) based on machine learning in Tensor flow. Tensor flow IDE is used to implement machine learning algorithms. Flower image processing is based on supervised learning which detects the parameters of image. Parameters of the image were compared by decision algorithms. These images are classified by neurons in convolutional neural network. Video processing based on machine learning is used in restriction of downloading the videos by preventing the second response from the server and enabling the debugging of the video by removing the request from the user. © 2018 Authors." "57203576845;57208855690;57201914273;55445979600;56401125500;55446100900;","A Novel approach for prediction of heart disease: Machine learning techniques",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082344458&partnerID=40&md5=ebd40c790915a462e6489d18ba36b2b8","Heart disease and machine learning are the two different words where one is related to medical field and another one to artificial intelligence. In medical filed most of them are facing the problems with the heart disease and machine learning is developing area in computer science. Heart disease is general called cardiac disease where it gives the more data or information, it is to be collected to give the reports for the patients and the machine learning also requires the data for predicting and to solve the problems. Machine learning techniques are used in prediction of heart diseases where it gives the faster prediction with less computation time and better accuracy to progress their health. Heart disease prediction requires lot of data for predicting and in cloud computing also we have more data and the data available in cloud it is difficult to analyze. So we use machine learning algorithms or techniques to predict the heart disease and the in the similar way we can apply these algorithms or techniques to predict or analyze the data that is available in cloud. In this paper we are going to use machine learning algorithms called Backpropagation Algorithm and later we use optimization algorithm later. Backpropagation algorithm deals with the artificial neural networks. Backpropagation is a method used to calculate the error contribution of each neuron after a batch of data (in image recognition, multiple images) is processed. This is used by an enveloping optimization algorithm to adjust the weight of each neuron, completing the learning process for that case. Machine learning algorithms and techniques are used for recognize the intensity of risk issues in humans and it helps the patients to take safety measures in well advances to save the patient's life. © 2018 Authors." "57215914751;55150308500;","An understanding of machine learning techniques in big data analytics: A survey",2018,"10.14419/ijet.v7i3.12.16450","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082343877&doi=10.14419%2fijet.v7i3.12.16450&partnerID=40&md5=25abd611f3e4e2b900eea96b03be76d1","Big data is a Firing Term in the recent era of the modern world, due to the information exploita-tion; there is an enormous amount of data produced. Big data is a powerful momentum of infor-mation and communication technology field due to the effect of growing data in healthcare, IOT, cloud computing, online education, online businesses, and public management. The produced data is not only large but also complex. Big data has a large amount of unstructured data so that there is a need to develop advanced tools and techniques for han-dling big data. Machine Learning is a prominent area of Artificial Intelligence. It makes the system to make intelligent resolutions by giving the knowledge to achieve the goals. This study reviews the various challenges and innovative ideas for big data analytics with machine learning in different fields over the past ten years. This paper mainly organized to identify the research projects based on the discussions over machine learning techniques for big data analytics and provide suggestions to develop the new projects. © 2018 Authors." "57210929194;24402703700;57204647920;","The guidance seeker onboard romote sensing satellite to enhance mission effectiveness",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071949027&partnerID=40&md5=2b8e463546fb362d128812403f9957dd","Nowadays, the swath has trend to get smaller with the development of high-resolution satellite for fixed number of pixels. So it is getting more important to capture clear target image efficiently. Otherwise, it might take several times to have a clear image if there is cloud between the remote sensing instrument (RSI) and target image. The objective of this paper is to develop an on-board guidance seeker on RSI. The on-board guidance seeker is used to get wideangle images and detect target before the task target is captured. Based on wide-Angle images and detected-Target, the path of satellite can be fine turned to direct ground target without cloud blocking ground target to get clear image or to find target ships and so on. The on-board guidance seeker includes a wide-Angle camera, an on-board machine learning unit and an on-board real time path fine-Tuning unit. This kind of on-board AI-driven design can enhance the mission effectiveness. It also can be applied to rescue in the open sea. © 2018 Asian Association on Remote Sensing. All Rights Reserved." "55357841300;57208632282;47161085800;57208628181;57205390249;57207195307;","Research and application of machine-learning-oriented spacecraft health management platform",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065341564&partnerID=40&md5=d234c8a94632d492afa4177a9f536cd9","The complex environment, conditions, aging failure and other comprehensive factors, making the spacecraft fault detection, diagnosis, prediction exceptionally difficult. The capability of traditional expert knowledge systems in spacecraft's system-level fault handling is limited, and still require designers and domain experts to spend a lot of time on mechanism analysis, formula derivation, and experimental verification. The traditional manual-analysis-based work model obviously cannot meet the development requirements of the spacecraft's high-reliability and quantity growth. In recent years, the new machine learning platforms (e.g., Microsoft's Azure, Google's Cloud Machine Learning, Alibaba's PAI), which have friendly process analysis framework, rich plugs and play machine learning tools and distributed services, can provide new ideas of complex problem handling in the fields of spacecraft. It proposed a machine-learning-oriented spacecraft health management platform design based on the analysis of the difficulties in spacecraft fault diagnosis and fault prediction, including modeling, health management platform architecture, massive data preprocessing methods, TensorFlow and other typical machine learning tools integration method, diagnosis and prediction of distributed service design and the results display and evaluation design, etc. Finally, the actual application effect is verified with the solar array power forecasting and other cases. The experiment results show that the research can provide technical reference for the research and application of spacecraft health management technology based on machine learning, and ultimately improve the safety of the spacecraft. © 2018 by the International Astronautical Federation (IAF). All rights reserved." [No author id available],"Delivering solutions at the intersection satellite big data, cloud computing, machine learning and IoT technology - The case of SatSure",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065339959&partnerID=40&md5=96a9e219274f676fcd9618ce4a60bcae","Doubling farm income by 2022-23 translates into enabling farmers to grow at more than 1.5 times the rate of Gross Domestic Product (GDP) growth in India over the next 5 years. This is no mean feat in leapfrogging to be achieved in a community where 67% of the 193.76 million hectares cropped area in the country is held by marginal farmers with holdings below one hectare. The three pillars that support farmers in the entire lifecycle of agriculture from a monetary perspective are credit, insurance and trade institutions. While much of the discussion and debate today centres around institutions that are involved in producing better seeds, fertilizer and equipment for better farm outcomes, it is important to acknowledge that discussing the future of monetary institutions and trade aspects in both technology and policy is as necessary as every other aspect of the farm and farmers. Utilizing satellite Big Data for driving efficiencies in financing agriculture, as global digitization drive is set to help both sides (farmers and monetary institutions). Gaining access to credit and insurance based on farm/farmer level analytics will also mean fixing problems in fraudulent utilization of loans as well as claims. For example, a bank providing loans for a particular crop can also have an automated flagging system for the entire lifecycle of the farm starting from sowing to harvest. This shall make sure that the credit is utilized effectively and will also give a perspective to the banker on the estimations of yield based, the health of the crop, etc. Similarly, an underwriter can get a perspective on the risks of monsoon, availability of irrigation, land fertility over time, weather uncertainties over cropping period, etc., providing complete transparency in sizing of the claim in the underwriting process. Copyright © 2018 by the International Astronautical Federation." "57202998689;36730895700;57205433409;57191746914;55002247800;57204106697;57205433321;57203025148;","It's raining barrels: Cloud computing in the O&G industry",2018,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85060023883&partnerID=40&md5=bd92f4a7803a068792b57f7655b82fc6","Cloud computing has become the buzz word in the last few years. All the service industries from different fields are using cloud computing based data analytics to optimize their operations in order to improve the customer user experiences and the overall efficiency. It is mainly due to the use of high volume computing backed by significant development in advanced hardware capabilities. This paper describes the 'what', 'why' and 'how' based questions on cloud computing. To start with, a brief introduction of cloud computing has been discussed along with the history of computation usage in oil industry. Followed by that a brief introduction highlighting the significance of Artificial intelligence and machine learning in the current computing environment has been explained. As the industry moves towards more and more usage of digital oilfield techniques, the dominance of high end computing and data analytics in the oil and gas industry is also showcased. Once cloud computing is established as a standard, the paper further discusses different modes of cloud delivery service models- Infrastructure as a service (IaaS), Platform as a service (PaaS) and Software as a Service (SaaS). The paper also identifies the challenges and concerns/issues that comes as a part of cloud computing methodology and then describes how these all challenges are being addresses. This paper elucidates the comprehensive view of the cloud computing landscape in oil and gas arena through a review of available noteworthy open source literature. © 2018 Society of Petroleum Engineers. All rights reserved." "57205244267;55758943200;","Machine learning in cloud: Sentiment analyzing system",2018,"10.14419/ijet.v7i4.40.24387","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85059261090&doi=10.14419%2fijet.v7i4.40.24387&partnerID=40&md5=1b2a83bde69e4f5193b10f0b2ed7b935","As the number of computer users increases, numerous content has been generated by them. Machine learning as one of the main direction of natural language processing, allows computer systems to extract various information from the generated content. Processing results determine the sentiments of the text to extract the author's emotional evaluation that is expressed in the text. The aim of the project was to develop the Sentiment Analyzing system by using Machine Learning algorithms on cloud-based system. The paper describes the development process of Sentiment Analyzing System in English language. Two Machine Learning algorithms, SVM and Naïve Bayes classifier, have been inspected and Cloud computing used to develop and publish web application. The testing results demonstrate the accuracy of the work in proposed method. © 2018 Authors." "57195102154;57204491819;56217320100;57202647606;","Automatic LOD0 classification of airborne LiDAR data in urban and non-urban areas",2018,"10.1080/22797254.2018.1522934","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85055260551&doi=10.1080%2f22797254.2018.1522934&partnerID=40&md5=8177a266b6cce93946555996c72c9c27","Point clouds are a very detailed and accurate vector data model of 3D geographic information. In contrast to other data models, no standard has been defined for visualization and management of point clouds based on levels of detail (LOD). This paper proposes the application of the concept of LODs to point clouds and defines the LOD0 for point cloud classification (the lowest possible level of detail) as urban and non-urban. A methodology based on the use of machine learning techniques is developed to perform LOD0 classification to airborne LiDAR data. Point clouds acquired with airborne laser scanner (ALS) are structured in grid maps and geometric features related with Z distribution and roughness are extracted from each cell. Six machine learning classifiers have been trained with datasets including urban (cities) and non-urban samples (farmlands and forests). The influence of grid size, point density, number of features and classifier type are analysed in detail. The classifiers have been tested in three case studies. The best results correspond to a grid size of 100 m and the use of 12 geometric features. The accuracy is around 90% in all tests and Cohen’s Kappa index reaches 81% in the best of cases. © 2018, © 2018 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group." "57218593297;57190957993;","Securing cloud by mitigating insider data theft attacks with decoy technology using Hadoop",2018,"10.14419/ijet.v7i2.31.13407","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85047856614&doi=10.14419%2fijet.v7i2.31.13407&partnerID=40&md5=1882bcc94adc67c572c7440c40d06ce9","Cloud Computing has been intrinsically changing the way we utilize computers to keep and retrieve our personal & business data. With the advent of this emerging paradigm of computing, it arises the new security challenges. Existent cryptographic data security techniques i.e., encryption deteriorated in preventing data theft attacks once the key is compromised, especially those perpetrated by insiders. Cloud Security Alliance reckoned this threat as a significant danger of Cloud Computing. Although the majority of Cloud users are very much known of this risk, they are leftover with the only choice of trusting the cloud service provider, regards to their data protection. In this paper, we propose an alternate way to secure data on the cloud which is more efficient and secure by the concoction of user profile mapping using Hadoop framework and offensive decoy technology. © 2018 Authors." "57194041416;57201646954;55312192000;57194253460;","LiDAR Based Sensor Verification",2018,"10.4271/2018-01-0043","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85045507351&doi=10.4271%2f2018-01-0043&partnerID=40&md5=43971302da8f9b3dfd98732008e0ab2a","In the world of automated driving, sensing accuracy is of the utmost importance, and proving that your sensors can do the job is serious business. This is where ground-truth labeling has an important role in Autoliv's validation process. Currently, annotating ground-truth data is a tedious and manual effort, involving finding the important events of interest and using the human eye to determine objects from LiDAR point cloud images. We present a workflow we developed in MATLAB to alleviate some of the pains associated with labeling point cloud data from a LiDAR sensor and the advantages that the workflow provides to the labeler. We discuss the capabilities of a tool we developed to assist users in visualizing, navigating, and annotating objects in point cloud data, tracking these objects through time over multiple frames, and then using the labeled data for developing machine learning based classifiers. We describe how the output of the labeling process is used to train deep neural nets to provide a fully automated way to produce vehicle objects of interest which can be used to find false-negative events. To do this with a human analyst takes as much time as to play back the entire data set. However, with a fully automated approach it can be run on many computers to reduce the analysis time. We present this time savings as well as the accuracy of the labels achieved and show how this approach provides substantial benefit to Autoliv's validation process. © 2018 SAE International. All Rights Reserved." "57201469360;57201362771;57201368459;57201475091;57201475351;","A proposals of convolution neural network system for malicious code analysis based on cloud systems",2018,"10.14419/ijet.v7i2.12.11040","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85045007763&doi=10.14419%2fijet.v7i2.12.11040&partnerID=40&md5=d0fcebe8fcba09ea4c5db00b90d76cf0","Background/Objectives: In the information security field, artificial intelligence must be applied first. This is because the frequency of malicious code is too high and the processing method is too difficult, which is very difficult for human to handle. Methods/Statistical analysis: In this paper, we developed a program to classify malicious codes into images and a Tensorflow system to classify malicious codes. The malware used as input was the computer virus code used in the BIG 2015 Challenge. This dataset, called a Kaggle dataset, consists of 10,868 bytes of train set. Findings: We used the Tensorflow SLIM library to develop this machine learning malware learning machine. This resulted in more than 80% accuracy. Especially, when the CRIS-Ensemble algorithm was added, the accuracy was 97%. The study of malicious code analysis using machine learning consists of two major parts. First, the process of making the virus into images is important. To classify 10,868 Kaggle malware datasets that the BIG 2015 winner showed 99.6% accuracy, Tensorflow's accuracy and parameter tuning are important, but finding the way to make good images is the most important technique Improvements/Applications: The results show that the malicious code classification system using machine learning can be an effective method to classify malicious code of malicious code by the accuracy of the result and ease of use. © 2018 Yong-kyu Park et. al." [No author id available],"Proceedings - 2017 International Conference on Green Informatics, ICGI 2017",2017,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85041684643&partnerID=40&md5=3d2ecb82b6f6b0568554d30e975c2014","The proceedings contain 47 papers. The topics discussed include: efficient GPU-based parallel Kriging algorithm for predicting the air quality index; hybridization of PMF and LSTM for recommendation of intelligent resource; research and application of clustering algorithm based on shared nearest neighbor; a low complexity extended Kalman filter algorithm for neural network digital predistortion of power amplifier; a monocular ranging algorithm for detecting illegal vehicle jumping; abstract super points from core network by unique candidate list; character-based convolutional grid neural network for breast cancer classification; an independent forwarding algorithm based on multidimensional spatial superposition model in SDN; machine learning based LncRNA function prediction; resource reconfiguration module for cross-cloud system; the implementation of a GPU-accelerated virtual desktop infrastructure platform; implement a virtual development platform based on QEMU; improving tasks scheduling performance in cloud computing environment by using analytic hierarchy process model; cloud bank liquidity risk prediction and identification, liquidity creation, and resource fragility; a task scheduling scheme for preventing temperature hotspot on GPU heterogeneous cluster; exploring critical success factors of building green logistics business in Fuzhou; cooperative self-organized energy-saving mechanism of cellular network based on hybrid energy supplies; encouraging knowledge sharing among green fashion communities; the key success factors of developing intelligent logistics within pharmaceutical industry in Fujian free trade area; and to assess the core competitiveness of Taiwanese design departments' students." "6507534695;","Machine-learning point cloud classification",2017,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85035035463&partnerID=40&md5=5ea0eb317d6bcf3d03056dd22d6feb8e",[No abstract available] "48161127100;7006517690;","Geocomputation: Data, Methods, and Applications in a New Era",2017,"10.1016/B978-0-12-409548-9.09600-7","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082357776&doi=10.1016%2fB978-0-12-409548-9.09600-7&partnerID=40&md5=a409bf27164db92e9e4bf2fd18a3487e","Geocomputation is a research field where computational technology and methods are applied to geographic data. We are in the midst of a fundamental change that affects how computers are used in handling geographic data. The Web has provided access to much new data as well as hardware agnostic software platforms for mapping and analysis. These new data and capabilities offer both opportunities and challenges to researchers. Traditional methods of geocomputation face new challenges to cope with and take advantage of these new opportunities. In this article we survey these changes and discuss how they are influencing new directions of spatial data applications. © 2018 Elsevier Inc. All rights reserved." "56783909900;55918988300;23013616900;6701381246;","Erratum: A machine learning method for co-registration and individual tree matching of forest inventory and airborne laser scanning data [Remote Sens., 9, 505 (2017)] DOI: 10.3390/rs9050505",2017,"10.3390/rs9070692","https://www.scopus.com/inward/record.uri?eid=2-s2.0-85022337303&doi=10.3390%2frs9070692&partnerID=40&md5=04f9ba7ee1450f546c2cbafdb8d008ab","Since Equation (2) has been rearranged incorrectly during preparation for this article [1], the authors would like to correct the relevant text of Section 3.4.3 as follows: 3.4.3. Synthetic Tree Detections A tree crown is defined by the tree species-specific parameters b and c (see Table 2), the tree height h, crown base height cbh, and its crown radius r (all provided by the Waldplaner 2.0 software, [2]). The height above ground ei of a point with distance di to the tree location less than or equal to r is calculated by Equation (2). Please note that since the parameters of the Douglas fir were not given by Pretzsch [3], these were assumed to be similar to the parameters of the Norway spruce. ei = h - c · (h - cbh) · (di/r)1/b (2) For each plot, we generated uniquely-distributed xy-coordinates with the same pulse density as the original point cloud. For each point, the z-coordinate was calculated by applying Equation (2) for all trees and adding the value of the DTM (of the original ALS point cloud) at the tree location. Finally, the point is linked to the tree which resulted in the maximum z-coordinate and-for a subsequent evaluation-it is labeled with the ID of that tree. We apologize for any inconvenience caused to the readers by these changes. The changes do not affect the scientific results. The manuscript will be updated and the original will remain online on the article webpage. © 2017 by the authors." "6507194541;57219176330;7005583247;57219173568;57219173295;57219176366;57219174081;7401595700;7004902724;","Online advanced uncertain reasoning architecture with binomial event discriminator system for novelty detection in smart water networks",2017,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091596172&partnerID=40&md5=322e650678bf3047dbb0b68a96fc1079","Minimising the loss of treated water from water supply systems due to bursts is an ongoing issue for water service providers around the world. Sensor technology and the 'big data' they generate combined with machine learning based analytics are providing an opportunity for automated event detection. AURA-Alert has been developed as an online (Software as a Service) system which automates the training data selection (by selecting data with acceptable Match Strength and with regular retraining). The addition of a Binomial Event Discriminator service can produce alerts based on windows of thresholded match distances. A pilot deployment on over 200 live data streams in the cloud has been deployed as part of the SmartWater4Europe project. Examples of analysis for real events are presented. For a historic subset of eight data streams over a three month period up to 58% of bursts were detected (depending on window used for evaluation). It is concluded that the system is an effective and viable tool for novelty detection for water network time series data with potential for wider applicability. Key strengths include lack of per site configuration, data-driven self-learning (from periods of normality), real-time, high scalability and full automation of model retraining. © 2020 CCWI 2017 - 15th International Conference on Computing and Control for the Water Industry. All rights reserved." "57196831038;57203371490;","Cloud based processing of free and commercial earth observation data with PCI GXL, populating and analysing data with the Australian Geoscience Data Cube software",2017,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85051431120&partnerID=40&md5=49b90a935d7c2f77de4a29ca2fa75293","Over recent years, corporate and public satellite operators have provided API’s to their image archives and distributed image processing has moved into the cloud. With this in place, the development of multi-sensor cloud enabled automated image pre-processing and analysis tools, feeding data cubes such as released by Geoscience Australia, have become necessary. Such a system is PCI GXL. The system can run on an Amazon cloud and extract time series of data from the archive of Planet Labs or ESA through their respective API’s. A range of image preprocessing techniques can be undertaken with a set of software libraries and executables. GXL powered automated workflows would typically include image calibration, atmospheric corrections, image to image registration to 1/10th of the image pixel, image compositing and image fusion. Pre-processing is optimized in a distributed way to feed into a data cube released as open source by Geoscience Australia. Once imagery has been deposited in the data cube, quantitative time series analysis can be performed by GXL to enable rapid identification of changes. A new approach to the analysis of multi resolution data supports the combined use of high spatial resolution commercial data with lower spatial resolution imagery collected at a high temporal frequency. Image segmentation and feature calculation based on pure pixels only allow the abstraction of data to units of homogeneous surfaces reducing computational efforts and complexity for a multi resolution spatial and temporal image analysis. A set of software libraries dedicated to SAR image processing for polarimetry and interferometry enable the use of SAR data with all its attributes in the data stack. Near real time change detection and mapping applications for disaster management can be addressed in this way. Machine learning classifiers are able to convert data in the data cube to thematic geo-spatial information efficiently and accurately. The system can be integrated with cloud enabled multi-mission satellite ground segments such as FarEarth to complete the loop from satellite tasking to geo-spatial information delivery to end users. © 2018 International Astronautical Federation IAF. All rights reserved." "57189347919;6603684955;57218448132;","Using Cloud-Based Analytics to Save Lives",2016,"10.1016/B978-0-12-803192-6.00012-8","https://www.scopus.com/inward/record.uri?eid=2-s2.0-84969730169&doi=10.1016%2fB978-0-12-803192-6.00012-8&partnerID=40&md5=9d7a1f5095a607d9f2dc84a6856caa2d","The National Flood Interoperability Experiment (NFIE) is research initiative among government, academia, and industry to help demonstrate the next generation of national flood hydrology modeling to enable early warning systems and emergency response. The goal of NFIE is to answer the questions-What if it were easier to predict more accurately where floods will occur? What if more flood information could be shared in real time to aid in response that is more effective for planning and prevent deaths and property damage? This paper presents an approach to these problems based on cloud computing and machine learning. The paper also addresses the characteristics of cloud computing that can make it an attractive alternative to on-premises infrastructure for much scientific collaboration. © 2016 Elsevier Inc. All rights reserved." "55331191600;26655335900;6603227150;24765165900;22735922800;36944449700;","Large scale thematic mapping by supervised machine learning on 'big data' distributed cluster computing frameworks",2015,"10.1109/IGARSS.2015.7326065","https://www.scopus.com/inward/record.uri?eid=2-s2.0-84962510511&doi=10.1109%2fIGARSS.2015.7326065&partnerID=40&md5=635b34400d2f45a070d8f3337791942c","The Petabyte-scale data volumes in Earth Observation (EO) archives are not efficiently manageable with serial processes running on large isolated servers. Distributed storage and processing based on 'big data' cloud computing frameworks needs to be considered as a part of the solution. © 2015 IEEE." "55308134300;6701525647;57195650018;","Do open clusters have distinguishable chemical signatures?",2014,"10.1051/eas/1567023","https://www.scopus.com/inward/record.uri?eid=2-s2.0-84937806199&doi=10.1051%2feas%2f1567023&partnerID=40&md5=638109ea41b3cfb63dba9b17ebfc40ed","Past studies have already shown that stars in open clusters are chemically homogeneous (e.g. De Silva et al. 2006, 2007 and 2009). These results support the idea that stars born from the same giant molecular cloud should have the same chemical composition. In this context, the chemical tagging technique was proposed by Freeman et al. (2002). The principle is to recover disrupted stellar clusters by looking only to the stellar chemical composition. In order to evaluate the feasibility of this approach, it is necessary to test if we can distinguish between stars born from different molecular clouds. For this purpose, we studied the chemical composition of stars in 32 old and intermediate-age open clusters, and we applied machine learning algorithms to recover the original cluster by only considering the chemical signatures. © EAS, EDP Sciences, 2015." [No author id available],"Proceedings - 7th International Congress on Environmental Modelling and Software: Bold Visions for Environmental Modeling, iEMSs 2014",2014,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-84912141184&partnerID=40&md5=2d9f8f03df6a5bb62de6895101d8825d","The proceedings contain 310 papers. The topics discussed include: use of near real time earth observation data infrastructures and open source tools for water resources monitoring and assessment; the virtual machine (VM) scaler: an infrastructure manager supporting environmental modeling on infrastructure-as-a-service clouds; physically based landslide susceptibility models with different degree of complexity: integration in OMS, calibration and verification; a spatial planning tool for the evaluation of the effect of hydrological and land-use changes on ecosystem services; modelling land-use changes in Godavari river basin : a comparison of two districts in Andhra Pradesh; abstractions from sensor data with complex event processing and machine learning; graph clustering based on social network community detection algorithms; linking governance storylines with the D-EXPANSE model to explore the power system transition pathways; and object-based analysis of multispectral RS data and GIS for detection of climate change impact on the Karakoram range northern Pakistan." [No author id available],"Proceedings - 7th International Congress on Environmental Modelling and Software: Bold Visions for Environmental Modeling, iEMSs 2014",2014,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-84912126716&partnerID=40&md5=80a36610703d119c81553cdf6b93ae43","The proceedings contain 310 papers. The topics discussed include: use of near real time earth observation data infrastructures and open source tools for water resources monitoring and assessment; the virtual machine (VM) scaler: an infrastructure manager supporting environmental modeling on infrastructure-as-a-service clouds; physically based landslide susceptibility models with different degree of complexity: integration in OMS, calibration and verification; a spatial planning tool for the evaluation of the effect of hydrological and land-use changes on ecosystem services; modelling land-use changes in Godavari river basin : a comparison of two districts in Andhra Pradesh; abstractions from sensor data with complex event processing and machine learning; graph clustering based on social network community detection algorithms; linking governance storylines with the D-EXPANSE model to explore the power system transition pathways; and object-based analysis of multispectral RS data and GIS for detection of climate change impact on the Karakoram range northern Pakistan." [No author id available],"Proceedings - 7th International Congress on Environmental Modelling and Software: Bold Visions for Environmental Modeling, iEMSs 2014",2014,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-84912102891&partnerID=40&md5=47e05dae7b53af68795265e83ba42fcd","The proceedings contain 310 papers. The topics discussed include: use of near real time earth observation data infrastructures and open source tools for water resources monitoring and assessment; the virtual machine (VM) scaler: an infrastructure manager supporting environmental modeling on infrastructure-as-a-service clouds; physically based landslide susceptibility models with different degree of complexity: integration in OMS, calibration and verification; a spatial planning tool for the evaluation of the effect of hydrological and land-use changes on ecosystem services; modelling land-use changes in Godavari river basin : a comparison of two districts in Andhra Pradesh; abstractions from sensor data with complex event processing and machine learning; graph clustering based on social network community detection algorithms; linking governance storylines with the D-EXPANSE model to explore the power system transition pathways; and object-based analysis of multispectral RS data and GIS for detection of climate change impact on the Karakoram range northern Pakistan." [No author id available],"Proceedings - 7th International Congress on Environmental Modelling and Software: Bold Visions for Environmental Modeling, iEMSs 2014",2014,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-84912073084&partnerID=40&md5=7621fca50733c52856391bc79137f46c","The proceedings contain 310 papers. The topics discussed include: use of near real time earth observation data infrastructures and open source tools for water resources monitoring and assessment; the virtual machine (VM) scaler: an infrastructure manager supporting environmental modeling on infrastructure-as-a-service clouds; physically based landslide susceptibility models with different degree of complexity: integration in OMS, calibration and verification; a spatial planning tool for the evaluation of the effect of hydrological and land-use changes on ecosystem services; modelling land-use changes in Godavari river basin : a comparison of two districts in Andhra Pradesh; abstractions from sensor data with complex event processing and machine learning; graph clustering based on social network community detection algorithms; linking governance storylines with the D-EXPANSE model to explore the power system transition pathways; and object-based analysis of multispectral RS data and GIS for detection of climate change impact on the Karakoram range northern Pakistan." [No author id available],"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences",2013,,"https://www.scopus.com/inward/record.uri?eid=2-s2.0-85060427922&partnerID=40&md5=b36c702b10aaa44ace64caa1e6400cc8","The proceedings contain 31 papers. The topics discussed include: canopy surface reconstruction and tropical forest parameters prediction from airborne laser scanner for large forest area; comparison of spatiotemporal mapping techniques for enormous ETL and exploitation patterns; a spatiotemporal prediction framework for air pollution based on deep RNN; applying Thiessen polygon catchment areas and gridded population weights to estimate conflict-driven population changes in south Sudan; detection of behavior patterns of interest using big data which have spatial and temporal attributes; a simple spatially weighted measure of temporal stability for data with limited temporal observations; detecting vessels carrying migrants using machine learning; spatio-temporal data model for integrating evolving nation-level datasets; spatio-temporal visualization of time-series satellite-derived CO2 flux data using volume rendering and gpu-based interpolation on a cloud-driven digital earth; extracting spatiotemporal objects from raster data to represent physical features and analyze related processes; spatiotemporal patterns and socioeconomic dimensions of shared accommodations: the case of AirBNB in Los Angeles, California; elastic cloud computing architecture and system for heterogeneous spatiotemporal computing; efficient lidar point cloud data managing and processing in a hadoop-based distributed framework; a novel approach of indexing and retrieving spatial polygons for efficient spatial region queries; and normalization strategies for enhancing spatio-temporal analysis of social media responses during extreme events: a case study based on analysis of four extreme events using socioenvironmental data explorer (SEDE)."