WorldFloodsv2 dataset

This repository contains the WorldFloodsv2 dataset released with the publication:

E. Portalés-Julià, G. Mateo-García, C. Purcell, and L. Gómez-Chova Global flood extent segmentation in optical satellite images. Scientific Reports 13, 20316 (2023). DOI: 10.1038/s41598-023-47595-7.

The WorldFloodsv2 database contains 509 pairs of Sentinel-2 images and flood segmentation masks. Splitted in train, val and test sets. It requires approximately 76GB of hard-disk storage.

Download the dataset

huggingface-cli download --cache-dir /path/to/cachedir --local-dir /path/to/localdir/WorldFloodsv2 --repo-type dataset isp-uv-es/WorldFloodsv2

Explore the dataset

The following ml4floods tutorials may help you explore and use this dataset:

1. The exploring WorldFloodsv2 tutorial shows how to process the dataset and plot the images and masks.
2. The WorldFloodsV2 torch dataset tutorial shows how to load the dataset with pytorch data loaders.

Licence

The WorldFloods database and all pre-trained models are released under a Creative Commons non-commercial licence

Cite

If you find this work useful, please cite:

@article{portales-julia_global_2023,
    title = {Global flood extent segmentation in optical satellite images},
    volume = {13},
    issn = {2045-2322},
    doi = {10.1038/s41598-023-47595-7},
    number = {1},
    urldate = {2023-11-30},
    journal = {Scientific Reports},
    author = {Portalés-Julià, Enrique and Mateo-García, Gonzalo and Purcell, Cormac and Gómez-Chova, Luis},
    month = nov,
    year = {2023},
    pages = {20316},
}