6810792
doi
10.5281/zenodo.6810792
oai:zenodo.org:6810792
user-remote-sensing
Ivan Oršolić
Why How Ltd
Freddie Kalaitzis
Oxford University, Why How Ltd
The WorldStrat Dataset: Open High-Resolution Satellite Imagery With Paired Multi-Temporal Low-Resolution
Julien Cornebise
University College London, WhyHow Ltd
info:eu-repo/semantics/openAccess
Other (Non-Commercial)
remote sensing
machine learning
image dataset
satellite imagery
<p><strong>What is this dataset?</strong></p>
<p><strong>Nearly 10,000 km² of free high-resolution and matched low-resolution satellite imagery</strong> of unique locations which ensure stratified representation of all types of land-use across the world: from agriculture to ice caps, from forests to multiple urbanization densities.</p>
<p>Those locations are also enriched with typically under-represented locations in ML datasets: sites of humanitarian interest, illegal mining sites, and settlements of persons at risk.</p>
<p>Each high-resolution image (1.5 m/pixel) comes with multiple temporally-matched low-resolution images from the freely accessible lower-resolution Sentinel-2 satellites (10 m/pixel).</p>
<p>We accompany this dataset with a paper, datasheet for datasets and an open-source Python package to: rebuild or extend the WorldStrat dataset, train and infer baseline algorithms, and learn with abundant tutorials, all compatible with the popular EO-learn toolbox.</p>
<p><strong>Why make this?</strong></p>
<p>We hope to foster broad-spectrum applications of ML to satellite imagery, and possibly develop the same power of analysis allowed by costly private high-resolution imagery from free public low-resolution Sentinel2 imagery. We illustrate this specific point by training and releasing several highly compute-efficient baselines on the task of Multi-Frame Super-Resolution.</p>
<p><strong>Licences</strong></p>
<ul>
<li>The high-resolution Airbus imagery is distributed, with authorization from Airbus, under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0).</li>
<li>The labels, Sentinel2 imagery, and trained weights are released under Creative Commons with Attribution 4.0 International (CC BY 4.0).</li>
<li>The source code (will be shortly released on GitHub) under 3-Clause BSD license.</li>
</ul>
Zenodo
2022-07-13
info:eu-repo/semantics/other
6810791
user-remote-sensing
1674412541.385991
14562136
md5:dfeb3348e79b719bf03c230d5d258839
https://zenodo.org/records/6810792/files/metadata.csv
2617946
md5:58d8e87c52bfaec962ab1ca5c3bf48b1
https://zenodo.org/records/6810792/files/WorldStrat_article_and_datasheet.pdf
282609
md5:745035835d835280aa0298a9dc1996d1
https://zenodo.org/records/6810792/files/stratified_train_val_test_split.csv
41542117051
md5:ca7167334006f3c17f9071f14c435335
https://zenodo.org/records/6810792/files/hr_dataset.tar.gz
39298
md5:d97b8d86da83f7e51f2d3205509e4a7b
https://zenodo.org/records/6810792/files/LICENSE.txt
11323245225
md5:40a4fd5241e91bffeddee0ed109778da
https://zenodo.org/records/6810792/files/hr_dataset_raw.tar.gz
26764158267
md5:8cfc6a477cee9e9cd8b20ea27227de65
https://zenodo.org/records/6810792/files/lr_dataset_l2a.tar.gz
27389228116
md5:d2dcafa207b1e1bc6c754607f15e9ed6
https://zenodo.org/records/6810792/files/lr_dataset_l1c.tar.gz
public
10.5281/zenodo.6810791
isVersionOf
doi