An improved Chan-Vese model based on local information for image segmentation

It is a difficult task to accurately segment images with intensity inhomogeneity, because most of representative algorithms assume that the image intensity is homogeneous. In this paper, an improved Chan-Vese model based on local information is proposed, which utilizes both global and local image information. The proposed method has been defined by the intensity fitting term and the regularization term. Firstly, the data evolution equation of the level set function is the gradient descent flow that minimizes the global binary fitting energy functional. The local intensity fitting value based on the Generalized Gaussian kernel function is then incorporated into the global intensity fitting value to form the weighted intensity fitting value on the two sides of the contour. Finally, the regularization term is used to control the smoothness of level set function and avoid complicated re-initialization. Experimental results and comparisons with other models of inhomogeneous images, synthetic and infrared images have shown the advantages of the proposed method in terms of accuracy and robustness of initial contour.