EuroSAT: Satellite Imagery for Land Use and Land Cover Classification

EuroSAT is a dataset of satellite images for land use and land cover classification. It contains images labeled into 10 classes, including annual crop land, forest, and industrial buildings. The dataset was uploaded to Hugging Face by user 'tanganke' and was last updated on May 16, 2024.

Use Cases

• Land use classification based on satellite imagery.
• Training semantic segmentation models for identifying land cover types like forests and crop lands.
• Benchmarking computer vision models on geospatial image recognition tasks.
• Analyzing the distribution of natural and man-made features from aerial imagery.

Strengths

• Dataset contains images labeled into 10 distinct land use and land cover classes.
• Images are in RGB format, which is a standard for many computer vision models.
• The dataset is hosted on Hugging Face, facilitating easy access and integration with common ML tools.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last updated 2024-05-16 02:40:58; freshness should be verified.

Provenance

Source

huggingface

Freshness

2024-05-16 02:40:58