A Fast C-V Model for Vector-valued Image Segmentation

To achieve fast and effective vector-valued image segmentation,an improved active contour model without re-initialization for object segmentation is proposed on the basis of analysis on the image colors,spatial information and characteristics of the level set function.Nonlinear heat equation with balanced diffusion rate is added to the C-V model to maintain the signed distance function property.Therefore the costly re-initialization procedure is completely eliminated.The proposed method employs the two-dimensional spatial rotation-invariance gradient and divergence operator instead of the traditional discretization approach.Experimental results show that the proposed method is fast,efficient and robust with respect to noise and weak object boundaries.