links-ads/spada-dataset

SPADA Dataset

This dataset contains images and sparse labels used in the paper Land Cover Segmentation with Sparse Annotations from Sentinel-2 Imagery , published at IGARSS 2023.

---

• Repository: https://github.com/links-ads/igarss-spada
• Paper: https://paperswithcode.com/paper/land-cover-segmentation-with-sparse

---

Dataset Preparation

The dataset has been compressed into segmented tarballs for ease of use within Git LFS (that is, tar > gzip > split). To revert the process into files and directories, follow these steps:

$ git clone https://huggingface.co/datasets/links-ads/spada-dataset
$ cd spada-dataset
# revert the multipart compression: merge first, then untar
$ cat data/train/train.tar.* | tar -xzvf - -i
$ cat data/test/test.tar.* | tar -xzvf - -i
$ cat data/val/val.tar.* | tar -xzvf - -i

It is very likely that the extracted files will retain the internal directory structure, making the train/val/test directories useless. Adapt the output structure as you see fit, the original structure is shown below.

Dataset Structure

The main dataset used in the paper comprises the following inputs:

Folder	Data Type	Description	Format
Img	Sentinel-2 Image	L2A data from 2018, with 12 channels in reflectance/10k format	GeoTIFF (.tif)
CLC	Scribble labels	2D labels derived from Corine Land Cover (CLC), obtained after the scribbling process	GeoTIFF (.tif)
Fuel	Pointwise labels	LUCAS points rasterized in a 5x5 grid to obtain point-wise annotations	GeoTIFF (.tif)

The training set also provides overlapping 512x512 tiles, while validation and testing are divided into non-overlapped sections with size 2048x2048 for ease of use.

Source Data

• Sentinel-2 L2A 2018 images were downloaded from Microsoft Planetary Computer, using predefined AoIs.
• CLC and LUCAS data from 2018 were downloaded from the official Copernicus Land platform, at https://land.copernicus.eu/en.
• Urban Atlas (UA) can also be downloaded from there.

Licensing Information

CC-BY-4.0 https://creativecommons.org/licenses/by/4.0/

Citation Information

@inproceedings{galatola2023land,
  title={Land Cover Segmentation with Sparse Annotations from Sentinel-2 Imagery},
  author={Galatola, Marco and Arnaudo, Edoardo and Barco, Luca and Rossi, Claudio and Dominici, Fabrizio},
  booktitle={IGARSS 2023-2023 IEEE International Geoscience and Remote Sensing Symposium},
  pages={6952--6955},
  year={2023},
  organization={IEEE}
}

Contributions

• Edoardo Arnaudo (edoardo.arnaudo@polito.it | linksfoundation.com)
• Marco Galatola (marco.galatola@linksfoundation.com)
• Luca Barco (luca.barco@linksfoundation.com)

Acknowledgements

This dataset was created in the context of the H2020 SAFERS project, under G.A. n. 869353, and HEU OVERWATCH project, under G.A. n.101082320, with the support of the EU Agency for the Space Programme (EUSPA). More information:

• https://safers-project.eu/
• https://overwatchproject.eu/