EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PMA-SAM2: Prior-Memory Adapted Framework for Automatic Remote Sensing Water Body Extraction

Hongcai Li, Wanyu Zhou, Chaohua Wang, Enqing Chen

Remote Sensing · 11.11.2026 · 18 (22) · 3880

DOI: 10.3390/rs18223880 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 11.11.2026. Verfasst von Hongcai Li und 3 weiteren Autor:innen. Die Arbeit nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Qinchuangyuan High-level Innovative and Entrepreneurial Talent Introduction Project, Shaanxi Association for Science and Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

SAM2 has great potential for remote sensing water body extraction due to its strong feature representation capability. However, when directly applied to the water body segmentation task, SAM2 still faces several challenges, including domain discrepancy, spectral confusion between water bodies and shadow-like objects, significant intra-class variations, and dependence on manual prompts. To address these challenges, this paper proposes the PMA-SAM2, which integrates domain prior injection, prototype memory calibration, and automatic prompt guidance for robust water body extraction. First, a Multi-level Prior Adaptation Module is designed to inject spatial geometric information and texture features into the SAM2 encoder, enabling effective alignment between general visual representations and remote sensing domain features. Then, a Prototype Memory Calibration Module is introduced to dynamically accumulate water feature prototypes under different geographical environments and imaging conditions, improving the model’s robustness to intra-class variations and inter-class spectral ambiguity. Furthermore, an Automatic Prompt Decoder is proposed to transform prototype-retrieved similarity maps into dense prompts, replacing manual prompts and enabling fully automatic water body segmentation. Experimental results on four datasets demonstrate that the proposed method achieves superior water body extraction accuracy compared with mainstream segmentation methods, verifying its strong generalization capability and scene adaptability.

Förderung

  • Qinchuangyuan High-level Innovative and Entrepreneurial Talent Introduction Project
  • Shaanxi Association for Science and Technology

Zitieren

APA

Li, H., Zhou, W., Wang, C., & Chen, E. (2026). PMA-SAM2: Prior-Memory Adapted Framework for Automatic Remote Sensing Water Body Extraction. Remote Sensing, 18(22), 3880. https://doi.org/10.3390/rs18223880

Vancouver

Li H, Zhou W, Wang C, Chen E. PMA-SAM2: Prior-Memory Adapted Framework for Automatic Remote Sensing Water Body Extraction. Remote Sensing. 2026;18(22):3880. doi:10.3390/rs18223880

BibTeX
@article{Li2026,
  author = {Li, Hongcai and Zhou, Wanyu and Wang, Chaohua and Chen, Enqing},
  title = {PMA-SAM2: Prior-Memory Adapted Framework for Automatic Remote Sensing Water Body Extraction},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {22},
  pages = {3880},
  doi = {10.3390/rs18223880}
}
Kostenlos starten →