Automatic Extraction Of Linear Arranged Targets From Polarimetric Sar Imagery

In this paper we propose a new extraction scheme for linear arranged targets in polarimetric SAR images based on a contrario theory. In this scheme, to reduce the influence of speckle, firstly a polarimetric whitening filter is applied to combine four images from different channels into a single channel image. Then a Cell-Averaging Constant False Alarm Rate detector with Weibull clutter background and a post-processing operator is used to detect point-like targets in the image. Finally, a searching approach based on contrario theory is applied to extract the linear arranged targets from other point-like targets. The experimental results varify the effectiveness in linear arranged targets extraction.