OpenEarthMap-SAR: A Benchmark Synthetic Aperture Radar Dataset for Global High-Resolution Land Cover Mapping

Overview: 

OpenEarthMap-SAR is a benchmark synthetic aperture radar dataset, for global high-resolution land cover mapping. It consists of 5033 images covering 35 regions from Japan, France and USA, with partically manually annotated and fully pesudo 8-class land cover labels at a 0.15–0.5m ground sampling distance. 

IEEE GRSS Data Fusion Contest 2025

OpenEarthmap-SAR also serves as the official dataset of IEEE GRSS DFC 2025 Track I.

Please download dfc25_track1_trainval.zip and unzip it. It contains training images & labels and validation images. 

Please download dfc25_track1_test.zip and unzip it. It contains test images. 

Please download dfc25_track1_val_labels.zip and dfc25_track1_test_labels.zip and unzip them It contains the labels of val and test images. 

Benchmark code related to the DFC 2025 can be found at this Github repo. 

The official leaderboard is located on the Codalab-DFC2025-Track I page. 

Paper & Reference

Details of OpenEarthMap-SAR can be refer to our paper. 

If OpenEarthMap-SAR is useful to research, please kindly consider cite our paper
 

@article{xia2025openearthmapsar,
      title={OpenEarthMap-SAR: A Benchmark Synthetic Aperture Radar Dataset for Global High-Resolution Land Cover Mapping}, 
      author={Junshi Xia and Hongruixuan Chen and Clifford Broni-Bediako and Yimin Wei and Jian Song and Naoto Yokoya},
      year={2025},
      journal={IEEE Geoscience and Remote Sensing Magazine}, 
}

License of Track 1
Optical images are provided by the National Institute of Geographic and Forest Information (IGN), France, under the CC BY 2.0 license, with contributions from the Geospatial Information Authority of Japan (GSI) and the National Agriculture Imagery Program (NAIP), USA.

SAR images are supplied by the Umbra Open Data Program under the CC BY 4.0 license.

Label datasets are shared under the same license as the original optical images, with specific terms varying by source dataset.