Synthetic Aperture Radar Ship Detection Using Modified Gamma Fisher Metric

This article proposes a novel ship detection method for high-resolution SAR images. Our goal is to look at this question from a information geometry point of view. The method consists of two steps: construction of revised metric and Riemann structure, and extraction of targets. For the first step of the process, a revised metric is introduced on Gamma 2-manifold. We construct a special Riemannian structure by using the proposed metric. For the second step, the regions of interest (ROIs) are extracted out based on the Riemann structure. Experimental results of the detection method on SAR images show that the algorithm presented is effective.