isp-uv-es/CloudSEN12Plus

CloudSEN12+ is a significant extension of the CloudSEN12 dataset, which doubles the number of expert-reviewed labels, making it, by a large margin, the largest cloud detection dataset to date for Sentinel-2. All labels from the previous version have been curated and refined, enhancing the dataset's trustworthiness. This new release is licensed under CC0, which puts it in the public domain and allows anyone to use, modify, and distribute it without permission or attribution.

Data Folder order

The CloudSEN12+ dataset is organized into train, val, and test splits. The images have been padded from 509x509 to 512x512 and 2000x2000 to 2048x2048 to ensure that the patches are divisible by 32. The padding is filled with zeros in the left and bottom sides of the image. For those who prefer traditional storage formats, GeoTIFF files are available in our ScienceDataBank repository.

CloudSEN12+ spatial coverage. The terms p509 and p2000 denote the patch size 509 × 509 and 2000 × 2000, respectively. ‘high’, ‘scribble’, and ‘nolabel’ refer to the types of expert-labeled annotations

ML-STAC Snippet

import mlstac
dataset = mlstac.load('isp-uv-es/CloudSEN12Plus')

Sensor: Sentinel2 - MSI

ML-STAC Task: image-segmentation

ML-STAC Dataset Version: 1.0.0

Data raw repository: https://cloudsen12.github.io/

Dataset discussion: https://huggingface.co/datasets/isp-uv-es/CloudSEN12Plus/discussions

Split_strategy: stratified

Paper: https://www.sciencedirect.com/science/article/pii/S2352340924008163

Data Providers

Name	Role	URL
Image & Signal Processing	['host']	https://isp.uv.es/
ESA	['producer']	https://www.esa.int/

Curators

Name	Organization	URL
Cesar Aybar	Image & Signal Processing	http://csaybar.github.io/

Labels

For human high-quality labels (also UnetMobV2_V2 & UnetMobV2_V1 predictions).

Name	Value
clear	0
thick-cloud	1
thin-cloud	2
cloud-shadow	3

For human scribble labels.

Name	Value
clear	0
thick-cloud border	1
thick-cloud center	2
thin-cloud border	3
thin-cloud center	4
cloud-shadow border	5
cloud-shadow center	6

Dimensions

Axis	Name	Description
0	C	Spectral bands
1	H	Height
2	W	Width

Spectral Bands

Name	Common Name	Description	Center Wavelength	Full Width Half Max	Index
B01	coastal aerosol	Band 1 - Coastal aerosol - 60m	443.5	17.0	0
B02	blue	Band 2 - Blue - 10m	496.5	53.0	1
B03	green	Band 3 - Green - 10m	560.0	34.0	2
B04	red	Band 4 - Red - 10m	664.5	29.0	3
B05	red edge 1	Band 5 - Vegetation red edge 1 - 20m	704.5	13.0	4
B06	red edge 2	Band 6 - Vegetation red edge 2 - 20m	740.5	13.0	5
B07	red edge 3	Band 7 - Vegetation red edge 3 - 20m	783.0	18.0	6
B08	NIR	Band 8 - Near infrared - 10m	840.0	114.0	7
B8A	red edge 4	Band 8A - Vegetation red edge 4 - 20m	864.5	19.0	8
B09	water vapor	Band 9 - Water vapor - 60m	945.0	18.0	9
B10	cirrus	Band 10 - Cirrus - 60m	1375.5	31.0	10
B11	SWIR 1	Band 11 - Shortwave infrared 1 - 20m	1613.5	89.0	11
B12	SWIR 2	Band 12 - Shortwave infrared 2 - 20m	2199.5	173.0	12
CM1	Cloud Mask 1	Expert-labeled image.	-	-	13
CM2	Cloud Mask 2	UnetMobV2-V1 labeled image.	-	-	14

Data Structure

We use .mls format to store the data in HugginFace and GeoTIFF for ScienceDataBank.

Folder Structure

The fixed/ folder contains high and scribble labels, which have been improved in this new version. These changes have already been integrated.

The demo/ folder contains examples illustrating how to utilize the models trained with CLoudSEN12 to estimate the hardness and trustworthiness indices.

The images/ folder contains the CloudSEN12+ imagery

Download

The code below can be used to download the dataset using the mlstac library. For a more detailed example, please refer to the examples section in our website https://cloudsen12.github.io/.

import mlstac
import matplotlib.pyplot as plt
import numpy as np

ds = mlstac.load(snippet="isp-uv-es/CloudSEN12Plus")
subset = ds.metadata[(ds.metadata["split"] == "test") & (ds.metadata["label_type"] == "high") & (ds.metadata["proj_shape"] == 509)][10:14]
datacube = mlstac.get_data(dataset=subset)

Make a plot of the data point downloaded

datapoint = datacube[2]
datapoint_rgb = np.moveaxis(datapoint[[3, 2, 1]], 0, -1) / 5_000
fig, ax = plt.subplots(1, 3, figsize=(10, 5))
ax[0].imshow(datapoint_rgb)
ax[0].set_title("RGB")
ax[1].imshow(datapoint[13], cmap="gray")
ax[1].set_title("Human label")
ax[2].imshow(datapoint[14], cmap="gray")
ax[2].set_title("UnetMobV2 v1.0")

Citation

Cite the dataset as:

@article{aybar2024cloudsen12+,
  title={CloudSEN12+: The largest dataset of expert-labeled pixels for cloud and cloud shadow detection in Sentinel-2},
  author={Aybar, Cesar and Bautista, Lesly and Montero, David and Contreras, Julio and Ayala, Daryl and Prudencio, Fernando and Loja, Jhomira and Ysuhuaylas, Luis and Herrera, Fernando and Gonzales, Karen and others},
  journal={Data in Brief},
  pages={110852},
  year={2024},
  DOI={10.1016/j.dib.2024.110852},
  publisher={Elsevier}
}