Polarimetric Feature-Based Weak Sea-Surface Target Detection with Machine Learning

To address the challenge of weak target detection in complex sea clutter, this paper proposes a polarimetric feature-based detection framework assisted by machine learning. Cloude polarization decomposition is applied to the scattering matrix along the slow-time dimension to extract sequences of polarimetric parameters. These parameters are combined with echo characteristics to construct a discriminative feature vector, which is then classified using XGBoost. Experimental results based on the IPIX radar dataset demonstrate that the proposed method significantly improves detection performance, exhibiting strong robustness and practical applicability.