Ship Detection with Sar Data Using a Notch Filter Based on Perturbation Analysis

The aim of the work described in this paper has been the development of an innovative ship detector, based on synthetic aperture radar (SAR) polarimetry and the methodology pioneered in [1-3], namely perturbation analysis. Ship detection is a key topic for the surveillance of coastal areas. In particular, SAR presents an advantageous technology, since the observations are independent on weather condition and solar illumination. Specifically, we achieve the detection by exploiting the difference between the polarimetric characteristics of sea clutter and ships. In the literature, several papers have described ship detection using radar polarimetry [4-6], however the technique proposed here is entirely innovative since it makes use of a methodology introduced by the authors in [1-3]. The proposed detector will be referred to as a notch filter since the detector rejects the target selected (in our case the sea) and detects anything different from sea. It acts in a similar manner to that of a notch filter, with the null corresponding to the target selected, however, the space where the filter is applied is now the polarization space and not that of the frequency). The algorithm is based on a physical rather than a statistical technique. In the following a brief introduction to polarimetry is presented. A single target can be characterized using a unique scattering matrix or equivalently a scattering vector: