Linear Feature Detection in SAR Images.

摘要

The problem treated in this paper is the one of detection and restoration of ship wakes in Synthetic Aperture Radar (SAR) images. This cannot be easily made, because SAR images are corrupted by a granular noise, called speckle and because there is no information about the direction and the level of the wake. For these reasons, most detection algorithms use the Radon transform, which makes square a straight with a point We propose here a new method, based on the manage between the Radon transform and a filtering method used to interpolate the image in a rotating reference system, introduced by the Radon transform theory. This filtering technic is the stochastic matched filtering technic, which allows to maximize the signal to noise ratio after processing. Experimental results on SIR-C/X-SAR images are presented and compared to those obtained using the classical approach.