On CFAR detection of oil slicks on the ocean surface by a multifrequency and/or multipolarization SAR

This paper addresses detection of oil spills on the sea surface as viewed by multifrequency and/or multipolarimetric synthetic aperture radars. We propose a two-stage processor: the first stage is a conventional detector designed to operate in a Gaussian environment; its detections are fed to a second stage designed to discriminate between candidate oil spills and templates of dark areas due to different phenomena, based upon different multifrequency and/or polarimetric signatures. Remarkably, the second stage possesses the constant false alarm rate property with respect to all of the unknown statistics of slick-free returns