Datasets:
The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
Sentinel-2 with Merged MultiSenGE Groundtruth
This dataset contains Sentinel-2 satellite imagery with corresponding ground truth labels derived from the MultiSenGE benchmark dataset. It is designed for remote sensing and machine learning applications, including land-cover classification, segmentation, and environmental monitoring. More about the original dataset, please visit the MultiSenGE page.
Dataset Description
Source Imagery: Sentinel-2 multispectral satellite images year 2020
Ground Truth: Merged patches derived from MultiSenGE (31TFN, 31UEP, 31UEQ)
Data Format:
- Sentinel-2 + Ground truth: multi-band
.zarrfiles (stacked bands and mask) - Sentinel-2 + Ground truth: multi-band
.tiffiles (stacked bands and mask)
Size: 4.16GB compressed in zip
License: CC-BY-4.0 (same as MultiSenGE)
This dataset was generated by:
- Downloading the original MultiSenGE ground truth patches.
- Merging the patches to form a complete ground truth map.
- Convert the labels from level 3 topology to level 1 topology. Ref: Wenger et al., 2022
- Downloading corresponding Sentinel-2 imagery.
- Stacking the ground truth labels with the Sentinel-2 images for aligned dataset creation.
- Downloads last month
- 13