Faint Ship Wake Detection in PolSAR Images

Focusing on the faint turbulent wake detection in polarimetric synthetic aperture radar (PolSAR) images, this letter introduces a novel two-step (coarse and fine processing) detector. Based on the polarization decomposition theory, a new parameter that enhances the contrast between wake and sea, which is called surface scattering randomness (SSR), is proposed. The coarse detection process extracts the regions of potential ship wakes by digital axoids transform of SSR. During the fine detection process, the regions detected by the coarse detection are segmented and processed independently in the polarization feature domain. Finally, by introducing the regularity least-squares method, the line parameters of ship wake are obtained, while the a priori knowledge is obtained from Radon transform. Experiments are performed on Radarsat-2 data, and the results demonstrate that the proposed algorithm has strong ability to detect faint turbulent wake in PolSAR images compared with the current detectors, which are based on the Radon or Hough transform.