This dataset was proposed by a research team from the Technical University of Munich in 2024. The relevant paper results are “UrbanSARFloods: Sentinel-1 SLC-Based Benchmark Dataset for Urban and Open-Area Flood Mapping". UrbanSARFloods is a dataset dedicated to urban and open area flood mapping, using preprocessed Sentinel-1 intensity data and interferometric coherence imagery. This addresses the lack of attention paid to urban floods in existing large-scale SAR-derived flood mapping studies. The dataset contains 8,879 512×512 image patches covering 807,500 square kilometers, spanning 20 land cover types and five continents, and covering 18 flood events. The dataset evaluates the performance of existing state-of-the-art convolutional neural networks on this dataset and points out the current challenges in urban flood detection, especially the problems of data imbalance and small training datasets.

Citation

@INPROCEEDINGS{10678367,
author={Zhao, Jie and Xiong, Zhitong and Zhu, Xiao Xiang},
booktitle={2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
title={UrbanSARFloods: Sentinel-1 SLC-Based Benchmark Dataset for Urban and Open-Area Flood Mapping},
year={2024},
pages={419-429},
keywords={Training;Satellites;Urban areas;Transfer learning;Sentinel-1;Land surface;Benchmark testing;Sentinel-1;flood mapping;benchmark dataset;urban flood},
doi={10.1109/CVPRW63382.2024.00047}
}