Abstract:

Feature extraction is an important problem of polarimetric synthetic aperture radar (SAR) image pressing and the key of maritime target detection. Similarity parameter and polarization entropy can represent the electromagnetic scattering characteristics of the targets. In order to enhance the contrast of targets and clutter, a new parameter is proposed based on feature parameter fusion. This parameter could express the regional differences in electromagnetic scattering characteristics. After processing, the contrast between targets and backgrounds becomes more apparent. The clutter model of this new parameter is studied, and a new method of ship detection is proposed. By using this parameter, ships are detected for the case of multi-look polarimetric SAR data. Finally the effectiveness of the proposed method is validated by airborne SAR (AirSAR) data.