A contextual Segmentation of Sea SAR Images to Detect Dark Spots in Mediterranean Sea

In this paper, we investigate the performance of partition features derived from histogram analysis to isolate dark spots which are candidates to be oil spills in SAR images. First partition is carried out to obtain preliminary clusters of the pixels on the basis of their grey level intensities and the two threshold values deduced from the histogram. The detection process is achieved by a contextual partition where the conflict pixels are attributed to their region involving local information about a pre-etiqueted pixels neighboring the pixel in question. For pixel's assignment, we proposed two decision criteria: first based on local probability maximization and the second use the khi2 test. We considered variable context in order to characterize the sea texture and a dark spots. This method is tested on ERS-2 SAR precision image containing several dark spots with different shapes and gave a promising results which are useful for the identification process