This is the dataset used for flood mapping as a change detection task in the IGARSS 22 conference paper "Attentive Dual Stream Siamese U-Net for Flood Detection on Multi-Temporal Sentinel-1 Data"

Paper Abstract:

Due to climate and land-use change, natural disasters such as flooding have been increasing in recent years. Timely and reliable flood detection and mapping can help emergency response and disaster management. In this work, we propose a flood detection network using bi-temporal SAR acquisitions. The proposed segmentation network has an encoder-decoder architecture with two Siamese encoders for pre and post-flood images. The network's feature maps are fused and enhanced using attention blocks to achieve more accurate detection of the flooded areas. Our proposed network is evaluated on publicly available Sen1Flood11 [1] benchmark dataset. The network outperformed the existing state-of-the-art (uni-temporal) flood detection method by 6% IOU. The experiments highlight that the combination of bi-temporal SAR data with an effective network architecture achieves more accurate flood detection than uni-temporal methods.

Full Paper : https://ieeexplore.ieee.org/document/9883132

Sen1Floods11 is a globally distributed, georeferenced dataset designed to train and evaluate deep learning models for flood detection using Sentinel-1 Synthetic Aperture Radar (SAR) imagery. It comprises 4,831 image chips, each measuring 512×512 pixels at 10-meter resolution, covering approximately 120,406 km² across 11 flood events spanning six continents, 14 biomes, and 357 ecoregions. The dataset includes Sentinel-1 SAR and Sentinel-2 MS image pairs. Annotations indicating areas of floodwater and permanent water bodies.

KozaMateusz/sen1floods11 dataset hosted on Hugging Face and contributed by the HF Datasets community

floodsmeta.csv: FloodsNet Metadata

FloodsNet_0.7z: WorldFloods and Sen1Floods11 flood training data as part of the FloodsNet data set.

FloodsNet_1.7z: USGS flood training data as part of the FloodsNet data set.

FloodsNet_2.7z: FL20190715MMR from the UNOSAT flood training data.

FloodsNet_3-0.7z: Half of the data for FL20180723LAO from the UNOSAT flood training data (1/2).

FloodsNet_3-1.7z: Half of the data for FL20180723LAO from the UNOSAT flood training data (2/2).

FloodsNet_4-0.7z: Half of the data for FL20150703MMR from the UNOSAT flood training data (1/2).

FloodsNet_4-1.7z: Half of the data for FL20150703MMR from the UNOSAT flood training data (2/2).