Uloženo v:
| Hlavní autoři: | , |
|---|---|
| Médium: | Recurso digital |
| Jazyk: | angličtina |
| Vydáno: |
Zenodo
2025
|
| Témata: | |
| On-line přístup: | https://doi.org/10.5281/zenodo.14630397 |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
Obsah:
- <p>Reference field boundaries dataset generated in the paper<a href="https://www.sciencedirect.com/science/article/pii/S0168169925001929"> "FieldSeg: A scalable agricultural field extraction framework based on the Segment Anything Model and 10-m Sentinel-2 imagery</a>".</p> <p>A hand-annotated field boundary dataset (2022) covering 8 10x10 km areas across the world is made available. The study areas are located in Argentina, Australia, Brazil, China, South Africa, Spain, USA-California, and USA-Iowa.</p> <p>This dataset contains two files:</p> <ul> <li>reference_field_boundaries.gpkg: hand-annotated dataset, with polygons defining the field boundaries.</li> <li>study_areas.gpkg: polygons defining the limits of the study areas and additional metadata about each area.</li> </ul> <p>More information on how this dataset was prepared is available in the paper <a href="https://www.sciencedirect.com/science/article/pii/S0168169925001929">"FieldSeg: A scalable agricultural field extraction framework based on the Segment Anything Model and 10-m Sentinel-2 imagery"</a>.</p>