Segment Anything for Satellite Imagery: A Strong Baseline and a Regional Dataset for Automatic Field Delineation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908415642566656 |
|---|---|
| author | Scribano, Carmelo Govi, Elena Bertellini, Paolo Parisi, Simone Franchini, Giorgia Bertogna, Marko |
| author_facet | Scribano, Carmelo Govi, Elena Bertellini, Paolo Parisi, Simone Franchini, Giorgia Bertogna, Marko |
| contents | Accurate mapping of agricultural field boundaries is essential for the efficient operation of agriculture. Automatic extraction from high-resolution satellite imagery, supported by computer vision techniques, can avoid costly ground surveys. In this paper, we present a pipeline for field delineation based on the Segment Anything Model (SAM), introducing a fine-tuning strategy to adapt SAM to this task. In addition to using published datasets, we describe a method for acquiring a complementary regional dataset that covers areas beyond current sources. Extensive experiments assess segmentation accuracy and evaluate the generalization capabilities. Our approach provides a robust baseline for automated field delineation. The new regional dataset, known as ERAS, is now publicly available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_16318 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Segment Anything for Satellite Imagery: A Strong Baseline and a Regional Dataset for Automatic Field Delineation Scribano, Carmelo Govi, Elena Bertellini, Paolo Parisi, Simone Franchini, Giorgia Bertogna, Marko Computer Vision and Pattern Recognition Artificial Intelligence Accurate mapping of agricultural field boundaries is essential for the efficient operation of agriculture. Automatic extraction from high-resolution satellite imagery, supported by computer vision techniques, can avoid costly ground surveys. In this paper, we present a pipeline for field delineation based on the Segment Anything Model (SAM), introducing a fine-tuning strategy to adapt SAM to this task. In addition to using published datasets, we describe a method for acquiring a complementary regional dataset that covers areas beyond current sources. Extensive experiments assess segmentation accuracy and evaluate the generalization capabilities. Our approach provides a robust baseline for automated field delineation. The new regional dataset, known as ERAS, is now publicly available. |
| title | Segment Anything for Satellite Imagery: A Strong Baseline and a Regional Dataset for Automatic Field Delineation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2506.16318 |