Ocean Clutter Modeling for Ship Detection
Tao Ding, Stian Normann Anfinsen, Camilla Brekke · 2013
This work addresses the problem of covariance matrix estimation for ocean clutter modeling. For ship detec-tion based on polarimetric synthetic aperture radar (Pol-SAR) imagery and constant false alarm rate (CFAR) de-tectors, accurate ocean clutter modeling is essential. The covariance matrix provides all the polarimetric informa-tion of the ocean clutter and its estimate is always in-volved in PolSAR detection [1]. The aim of this work is to investigate and compare the behavior of different co-variance matrix estimators, i.e., the sample mean, fixed-point, and maximum likelihood estimators. An approx-imate maximum likelihood covariance matrix estimator is also proposed and discussed for better computational efficiency. Their performances are evaluated in terms of the Kullback-Leibler (KL) matrix distance, and compu-tational efficiency. Various textured ocean clutter con-ditions are considered, ranging from high texture to the non-textured case with Gaussian clutter. Experiments are performed on simulated ocean clutter data.