Intellectualization of Moving Transport Objects Recognition
Creators
- 1. State University of Infrastructure and Technology, Ukraine
Description
The purpose of the article is to research, analyze, and consider current problems and prospects for the development of software for the recognition of transport objects based on the use of pattern recognition theory, methods, and tools of artificial intelligence, and different types of neural networks.
The research methodology is basic methods and algorithms of pattern recognition, methods and means of artificial intelligence, and different types of neural networks. The article considers the main problems of intellectualization of processes occurring in transport. The main attention is paid to the intellectualization of the processes of transport objects’ recognition. The article analyzes the most common recognition methods. A study of these methods and approaches to the recognition of moving vehicles is conducted. It is determined which methods have high and which have low computational complexity. Among the considered methods are those that recognize static transport objects (primarily transport infrastructure objects) using intelligent technologies, statistical, probabilistic, and other methods. The main attention is paid to methods that recognize dynamic transport objects. The basis of intellectualization of the processes of recognition of this group of objects is the use of neural networks, in particular convolutional, recurrent, neural networks with long short-term memory (LSTM), etc.
The scientific novelty of the research is the analysis of modern methods of moving transport objects’ recognition, the results of which can be used in the development of their software product. The article emphasizes that the proposed modern approach to the recognition of transport objects (moving vehicles and transport infrastructure) involves solving a wide range of problems based on the use of intelligent technologies, in particular neural networks.
Conclusions. The most common methods for solving current problems of recognition of transport objects (vehicles and transport infrastructure) have been investigated and analyzed. Based on the analysis of methods for recognizing moving vehicles, it has been determined which methods have high and low computational complexity. The basis of the intellectualization of the recognition processes for this group of transport objects is the use of various neural networks.
Files
Intellectualization_of_Moving_Transport_Objects_Recognition.pdf
Files
(893.3 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:5f0701b5c1aa8d0575a65fccf9da862e
|
893.3 kB | Preview Download |
Additional details
References
- Ammar, A., Koubaa, A., Ahmed, M. and Saad, A., 2019. Aerial Images Processing for Car Detection using Convolutional Neural Networks: Comparison between Faster R-CNN and YoloV3. Preprints, [e-journal] 17 October. DOI: 10.20944/preprints201910.0195.v1.
- Borysov, H.O., Humen, T.F. and Trapezon, K.O., 2020. Doslidzhennia prohramnykh osoblyvostei obiednannia Android things na osnovi kontseptsii Internetu rechei [Study of software features of Android things integration based on the Internet of Things concept]. Scientific notes of Taurida National V.I. Vernadsky University. Series: Technical Sciences, [e-journal] 31(70/1), pp.29–35. https://doi.org/10.32838/2663-5941/2020.1-1/0.
- Introduction to computer vision: what it is and how it works, 2018. DataRobot, [online] 2 April. Available at: <https://www.datarobot.com/blog/introduction-to-computer-vision-what-it-is-and-how-it-works/> [Accessed 28 August 2022].
- Kartashov, V.M., Oleynikov, V.N., Sheyko, S.A., Babkin, S.I., Koryttsev, I.V. and Zubkov, O.V., 2019. Peculiarities of Small Unmanned Aerial Vehicles Detection and Recognition. Telecommunications and Radio Engineering, [e-journal] 78 (9), pp.771–781. https://doi.org/10.1615/TelecomRadEng.v78.i9.30.
- Markov, E., 2016. Fractal methods for extracting artificial objects from the unmanned aerial vehicle images. Journal of Applied Remote Sensing, [e-journal] 10 (2), art. 25020. https://doi.org/10.1117/1.JRS.10.025020.
- Nechiporenko, A.S., Gubarenko, E.V. and Gubarenko, M.S., 2019. Authentication of users of mobile devices by their motor reactions. Telecommunications and Radio Engineering, [e-journal] 78 (11), pp.987–1003. https://doi.org/10.1615/TelecomRadEng.v78.i11.60.
- Nidhi, G., 2017. The Incredible Future Of Public Transport With This Gyroscopic Vehicle Design. Industry Tap, [online] 26 August. Available at: <https://www.industrytap.com/incredible-future-public-transport-gyroscopic-vehicle-design/43587> [Accessed 31 August 2022].
- Prakash, J., 2018. The intuition behind RetinaNet. Medium, [online] 23 March. Available at: <https://medium.com/@14prakash/the-intuition-behind-retinanet-eb636755607d> [Accessed 28 August 2022].
- Redmon, J., Divvala, S., Girshick, R. and Farhadi, A., 2016. You Only Look Once: Unified, Real-Time Object Detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), [e-journal] 27-30 June 2016. Institute of Electrical and Electronics Engineers. pp.779– 788. https://doi.org/10.1109/CVPR.2016.91.
- Samaras, S., Diamantidou, E., Ataloglou, D., Sakellariou, N., Vafeiadis, A., Magoulianitis, V., Lalas, A., Dimou, A., Zarpalas, D., Votis, K., Daras, P. and Tzovaras, D., 2019. Deep Learning on Multi Sensor Data for Counter UAV Applications – A Systematic Review. Sensors, [e-journal] 19 (22), 4837. https://doi.org/10.3390/s19224837.
- Svatiuk, D.R., Svatiuk, O.R. and Belei, O.I., 2020. Zastosuvannia zghortkovykh neironnykh merezh dlia bezpeky rozpiznavannia obiektiv u videopototsi [Application of convolutional neural networks for object recognition security in video stream]. Cybersecurity: Education, Science, Technique, [e-journal] 4(8), pp.97–112. https://doi.org/10.28925/26634023.2020.8.97112.
- Tymoshyn, Yu.A. and Orlenko, S.P., 2018. Alhorytm rozpiznavannia oblychchia liudei na bazi zghortkovoi neironnoi merezhi [Human face recognition algorithm based on convolutional neural network]. Adaptyvni systemy avtomatychnoho upravlinnia, 1 (32), pp.166–173.
- Yaacoub, J.-P., Noura, H., Salman, O. and Chehab, A., 2020. Security analysis of drones systems: Attacks, limitations, and recommendations. Internet of Things, [e-journal] 11, pp.1–39. https://doi.org/10.1016/j.iot.2020.100218.