CloudSEN12+: Expert-Labeled Pixels for Cloud Detection in Sentinel-2

CloudSEN12+ version 1.1.2 is the largest dataset of expert-labeled pixels for cloud and cloud shadow detection in Sentinel-2 imagery. It is a significant extension of the CloudSEN12 dataset, doubling the number of expert-reviewed labels. The dataset is maintained by the tacofoundation and was last updated on the Hugging Face platform in January 2025.

Use Cases

• Training cloud detection models based on expert-labeled pixel data.
• Benchmarking remote sensing algorithms using the largest labeled dataset for Sentinel-2 cloud detection.
• Improving atmospheric correction in satellite image processing pipelines based on accurate cloud and shadow masks.

Strengths

• Described as the largest cloud detection dataset for Sentinel-2 imagery to date.
• Contains a significant extension of labels, doubling the number of expert-reviewed labels from a previous version.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown, which may limit suitability assessment.

Provenance

Source

tacofoundation on Hugging Face

Collection Method

Expert-reviewed labeling of Sentinel-2 satellite imagery pixels.

Freshness

Last updated 2025-01-05 14:47:21; freshness should be verified.