A Level Set Image Segmentation Method Combined with GVF and CV Model

Owning to the non-convex functional of with the Chan-Vese(CV) model, one can only obtain a local optimal solution. It is difficult to achieve an ideal result for image segmentation in the global range. Therefore a new level set based image segmentation method that combining CV model and gradient vector flow(GVF) was proposed. The edge gradient information is spreaded to the entire image with GVF, which guides the evolution of CV model to the correct target edge in the global range and retains the basic advantages of CV model. The experimental results indicate that the present method are obviously better than the traditional CV model.