Power-based PolSAR Ship Detector with the Sub-look Decomposition

Le résumé fourni par la source

PolSAR ship detection in complex maritime environments is a challenging task due to the influence of sea clutter, varying sea states, and different target characteristics. To address these challenges, this paper introduces ASPAN and RSPAN, two novel methods that integrate sub-look decomposition with polarimetric decomposition to improve detection performance. Sub-look decomposition extracts spatial features based on the azimuth or range direction, while polarimetric decomposition uses scattering characteristics to suppress clutter and improve target discrimination. The experimental results demonstrate that ASPAN achieves the highest detection rate by detecting 26 ships with 2 false alarms, while RSPAN eliminates false alarms entirely but detects 25 ships. Quantitative evaluations based on the Target-to-Clutter Ratio (TCR) further confirm the superiority of ASPAN and RSPAN over existing methods, highlighting their potential for robust and precise ship detection in complex maritime scenarios.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.