An Initial Contour Method for Active Contour Model

Image segmentation which is the basis of image analysis and image understanding is a very important step in the area of image processing. The result of image segmentation determines whether it is suitable to go on the following processing. Considering the defect of the existing image segmentation method such as the slow speed of segmentation and acquiring more manual operations in the area of medical image segmentation, a new method of setting initial contour is proposed in this paper. Firstly the image will be processed simply in which a rough and continuous contour will be obtained, then 8-adjacent segmentation method will be adopted to detect the edge, finally the initial contour will converge to the boundary needed by using gradient vector flow field. Experimental results show that, this method can get a better result of convergence of the target boundary, and is an ideal method for medical image segmentation with fast speed and high degree of automation.