TreeSatAI Benchmark Archive for Deep Learning in Forest Applications
50,000+ aerial and Sentinel image triplets labeled for forest species and age
- Source
- Zenodo
- Published
- 2022
- Region
- Europe
- Format
- multiple formats
- Access
- Open access
- License
- CC-BY-4.0
- Themes
- Forests, Machine Learning & AI, Remote Sensing & Satellites
About this dataset
Contains over 50,000 image triplets (aerial, Sentinel-1, Sentinel-2) with labels for species, age, and forest stand types, suitable for benchmarking deep learning models.
How the data was collected
Derived from high-resolution aerial imagery and satellite data, annotated for various forest attributes.