Satellite Segmentation: Sentinel-2 RGB Tiles with ESA Land-Cover Masks

Satellitesegmentation contains paired Sentinel-2 RGB image tiles and corresponding integer-labeled land-cover masks derived from ESA WorldCover data. The dataset is prepared for semantic segmentation tasks, with source imagery from Sentinel-2 L2A via Microsoft Planetary Computer. It was authored by nikolkoo and last updated on 2025-11-26.

Use Cases

• Training semantic segmentation models based on paired Sentinel-2 RGB images and land-cover masks.
• Benchmarking land-use classification algorithms based on ESA WorldCover-derived labels.
• Developing remote sensing applications for environmental monitoring based on 10-meter resolution satellite imagery.

Strengths

• Data pairs are explicitly aligned to the Sentinel-2 grid at a 10-meter spatial resolution.
• Land-cover masks are derived from the authoritative ESA WorldCover product.
• Source imagery is from Sentinel-2 L2A, accessed via Microsoft Planetary Computer's STAC API.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset's geographic coverage is suggested to be 'Regionus', potentially limiting global applicability.

Provenance

Source

Sentinel-2 L2A via Microsoft Planetary Computer; ESA WorldCover.

Collection Method

Paired RGB tiles and derived land-cover masks prepared for semantic segmentation.

Freshness

Last updated 2025-11-26 13:23:42; freshness should be verified.

Geography

Regionus (suggested by platform tags).