Edge detection of oil spill using SAR image

The ocean pollution, which is mainly caused by marine oil spill accidents, is a huge environmental problem. The satellite especially the synthetic aperture radar (SAR) is a well-known effective tool in figuring out this problem. Because speckle noise is a large problem in SAR images since even a homogeneous area has a statistical distribution with large standard deviation, an improved edge detection algorithm, based on Canny algorithm and thresholding algorithm, for SAR image is proposed. By the consideration of the applicability and practicability of the proposed algorithm, the SAR images for the experiment include different satellites in different modes, such as ENVISAT and RADARSAT-1 and it makes the result more robust. Experimental results demonstrate that the method proposed in this paper is robust and effective to extract edges of SAR image accurately.