EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Oil Spill Detection Performance in a Multitype Polarimetric-Feature Space Using a Polarimetric Synthetic Aperture Radar: A Comparative Analysis

Guannan Li, Gaohuan Lv, Xiang Wang iD, Fen Zhao, Xiluo Teng

Sensors · 26.07.2026 · 26 (15) · 4750

DOI: 10.3390/s26154750 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 26.07.2026. Verfasst von Guannan Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: Ulsan Ship and Ocean College, Ludong University, Yantai 264025, China. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Shandong Province, Shandong Province Youth Innovation Team for Higher Education Institutions.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Marine oil spills severely threaten marine ecosystems, the coastal economy, and marine engineering structures. Because it enables all-weather and all-time acquisition of rich target information, fully polarimetric synthetic aperture radar (FP SAR) is widely used for monitoring marine oil spills. However, the differences in the scattering characteristics among oil types can cause variability in the information contained in the features extracted using FP SAR. Herein, RADARSAT-2 images obtained from a rare oil-on-water experiment conducted in the Norwegian North Sea were used to compare the distribution differences in polarimetric features based on the oil slick type and incident angle. Results showed that the incident angle exerted some influence on polarimetric features and the detection performance for oil spills with a low oil–water contrast, particularly at large incident angles. The polarimetric features related to scattering mechanisms exhibited good robustness and effectiveness across various incident angles. The polarimetric feature that combines the scattering entropy H and modified anisotropy A12 exhibited strong overall performance and high suitability for extracting information on oil spills at different incident angles. This study demonstrates that incorporating appropriate polarimetric features according to the incident angle enables the identification of different oil slick types and facilitates oil spill detection and monitoring.

Schlagwörter

Stichwörter der Autor:innen

comparative analysismarine oil spillsmultitype feature spacepolarimetric synthetic aperture radar

Autor:innen und Institutionen

  • Ulsan Ship and Ocean College, Ludong University, Yantai 264025, China.
  • Ocean Remote Sensing Division, National Marine Environmental Monitoring Center, Dalian 116026, China.
  • School of Resources and Environmental Engineering, Ludong University, Yantai 264025, China.
  • Faculty of Engineering and Information Technology, The University of Melbourne, Parkville, VIC 3010, Australia.

ORCID-Profile: Xiang Wang

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Shandong Province
  • Shandong Province Youth Innovation Team for Higher Education Institutions
  • Technological Small and Medium-sized Enterprises Innovation Ability Enhancement Project of Shandong Province

Zitieren

APA

Li, G., Lv, G., Wang, X., Zhao, F., & Teng, X. (2026). Oil Spill Detection Performance in a Multitype Polarimetric-Feature Space Using a Polarimetric Synthetic Aperture Radar: A Comparative Analysis. Sensors, 26(15), 4750. https://doi.org/10.3390/s26154750

Vancouver

Li G, Lv G, Wang X, Zhao F, Teng X. Oil Spill Detection Performance in a Multitype Polarimetric-Feature Space Using a Polarimetric Synthetic Aperture Radar: A Comparative Analysis. Sensors. 2026;26(15):4750. doi:10.3390/s26154750

BibTeX
@article{Li2026,
  author = {Li, Guannan and Lv, Gaohuan and Wang, Xiang and Zhao, Fen and Teng, Xiluo},
  title = {Oil Spill Detection Performance in a Multitype Polarimetric-Feature Space Using a Polarimetric Synthetic Aperture Radar: A Comparative Analysis},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {15},
  pages = {4750},
  doi = {10.3390/s26154750}
}
Kostenlos starten →