Research Of The Adaptive HDTV Regularization To Image Restoration And Segmentation Problem

Image processing is widely applied in various fields of modern life. Image restoration is the basis of image processing steps, and image segmentation is the key steps from image processing to image analysis. Therefore, to improve the method of image restoration and segmentation, and propose the corresponding models and algorithms have realistic significance. Main work of this article is based on the existing total variation restoration model and Mumford Shah(M-S) segmentation model, in view of the total variational model in the restored image is easy to have a ladder effect and not well in segmenting image with complex boundary problem, put forward with edge detection function of High Degree Total Variation(HDTV) regularization restoration model and segmentation model. Since the models are non-smooth optimization problems, therefore, using the operator splitting method convert into easy optimization problems, and analyzing convergence for the algorithm. Finally numerical experiments demonstrate the effectiveness of the two algorithms.