An improved level set evolution without re-initialization for vector-valued image segmentation

This paper presents an improved variational formulation for active contours model that forces level set function to be fast and stably close to signed distance function. The improvement can completely eliminates the need of the costly Re-initialization procedure. A restriction item that is nonlinear heat equation with balanced diffusion rate is added to the traditional Chan-Vese vector-valued model. The proposed variational level set formulation is implemented by finite difference scheme with spatial rotation-invariance gradient and divergence operator. Consequently it computes more efficiently. The proposed algorithm has been applied to both synthetic and real images with promising results.