In the study, we propose a region-based method for the automatic detection of oil spills in SAR images. The proposed method combines the image segmentation techniques and conventional detection theory to improve the accuracy of oil spills detection. From the image statistical characteristics, we first segment the image into proper regions by using one-dimensional moment preserving method. Their to get it more integrated segmentation result, we adopt N nearest neighbor rule to merge the image regions according to their spatial correlation. Performing, the split and merge procedure recursively we can partition the image into proper regions, oil-polluted and sea areas. Based on the segmentation results, then, we propose an oil spills detection algorithm, which involves data modeling of oil-polluted image data and the development of air automatic decision rule. Employing the built oil spills model and the generalized likelihood ratio (GLRE) detection theory. we derive it region-based decision rule for oil spills detection. Under the criterion of Constant False Alarm Ration (CFAR), we may determine the threshold automatically. Simulation results performed on ERS2-SAR images have demonstrated the efficiency of the proposed approach.
