Considering the low accuracy of Synthetic Aperture Radar （SAR） image segmentation in the marine spill oil detection

a segmentation method of marine spill oil images based on Tsallis entropy multilevel thresholding and improved Chan Vese （CV）model is proposed in this paper. First

the multi-threshold selection algorithm based on Tsallis entropy is used to make a coarse segmentation for marine spill oil images. The obtained spill oil region and its coarse contour provide local region and initial contour for CV model

respectively

which are used to reduce the scene complexity of CV model and its sensitivity to initial situation.The traditional CV model only considers the mean value of each region of image instead of the local information of image. Though it can get non-gradient defined image boundary

there are errors in the segmented results. We use an improved CV model with the motion factor

thus the segmentation errors are reduced and the convergence speed is increased. Experimental results show that the our method not only dispenses with initial condition

but also ensures accurate segmentation boundary and efficient operation.