A Level Set Method for Image Segmentation by Integrating Channel Anisotropic Diffusion Information

A novel algorithm for image segmentation is proposed, which combines anisotropic diffusion and Chan-Vese (CV) model to improve contours´ performance on multichannel (color, multispectral, hyperspectral, etc) image segmentation. Firstly, the multichannel images are divided into each single channel. Secondly, a novel anisotropic diffusion method is used to smooth the image of each channel. Finally, we use an improved multichannel CV model which integrates the segmentation information of each channel. Simulation and real experiments on color images have shown that our algorithm can obtain better performance than the traditional CV method.