EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LRA-UNet: A Lightweight Residual Attention Network for SAR Marine Oil Spill Detection

Yu Cai, Jingjing Su, Jun Song, Dekai Xu, Liankang Zhang, Gaoyuan Shen

Journal of Marine Science and Engineering · 12.06.2025 · 13 (6) · 1161

DOI: 10.3390/jmse13061161 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Marine Science and Engineering, online erschienen am 12.06.2025. Verfasst von Yu Cai und 5 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Plan of Liaoning Province, Dalian Science and Technology Innovation Fund, Dalian Science and Technology Program for Innovation Talents of Dalian.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Oil spills represent a serious threat to marine ecosystems. Remote sensing monitoring, especially based on synthetic aperture radar (SAR), have been extensively employed in marine environments due to its unique advantages. However, SAR images of marine oil spills exhibit challenges of weak boundaries, confusion with look-alike phenomena, and the difficulty of detecting small-scale targets. To address these issues, we propose LRA-UNet, a Lightweight Residual Attention UNet for semantic segmentation in SAR images. Our model integrates depthwise separable convolutions to reduce feature redundancy and computational cost, while adopting a residual encoder enhanced with the Simple Attention Module (SimAM) to improve the precise extraction of target features. Additionally, we design a joint loss function that incorporates Sobel-based edge information, emphasizing boundary features during training to enhance edge sharpness. Experimental results show that LRA-UNet achieves superior segmentation results, with a mIoU of 67.36%, surpassing the original UNet by 4.41%, and a 5.17% improvement in IoU for the oil spill category. These results confirm the effectiveness of our approach in accurately extracting oil spill regions from complex SAR imagery.

Förderung

  • Science and Technology Plan of Liaoning Province
  • Dalian Science and Technology Innovation Fund
  • Dalian Science and Technology Program for Innovation Talents of Dalian
  • Liaoning Province Education Department Scientific research platform construction project
  • Basic scientific research funds of Dalian Ocean University

Zitieren

APA

Cai, Y., Su, J., Song, J., Xu, D., Zhang, L., & Shen, G. (2025). LRA-UNet: A Lightweight Residual Attention Network for SAR Marine Oil Spill Detection. Journal of Marine Science and Engineering, 13(6), 1161. https://doi.org/10.3390/jmse13061161

Vancouver

Cai Y, Su J, Song J, Xu D, Zhang L, Shen G. LRA-UNet: A Lightweight Residual Attention Network for SAR Marine Oil Spill Detection. Journal of Marine Science and Engineering. 2025;13(6):1161. doi:10.3390/jmse13061161

BibTeX
@article{Cai2025,
  author = {Cai, Yu and Su, Jingjing and Song, Jun and Xu, Dekai and Zhang, Liankang and Shen, Gaoyuan},
  title = {LRA-UNet: A Lightweight Residual Attention Network for SAR Marine Oil Spill Detection},
  journal = {Journal of Marine Science and Engineering},
  year = {2025},
  volume = {13},
  number = {6},
  pages = {1161},
  doi = {10.3390/jmse13061161}
}
Kostenlos starten →