Research On Image Denoising Algorithm Based On Total Variation Model

Image restoration is one of the most widely studied topics in the field of digital image processing.In order to get faster and better image restoration effect,image denoising is an essential step.The main goal of image denoising is to recover the corresponding clear image from the noisy image.Image noise mainly comes from shooting and the process of image transmission,reproduction and scanning.So far,many image dcnoising algorithms have been proposed by related scholars.In this paper,some classit image denoising algorithms are briefly introduced,and the advantages and disadvantages of different algorithms are analyzed.However,these algorithms are easy to blur image edge information when image denoising.Therefore,the total variation denoising model has attracted much attention because of its excellent edge protection ability.The research shows that the improved algorithm based on the total variation model is mainly divided into two aspects,one is the numerical solution algorithm of the optimization model,the other is to improve the model.In view of these two aspects,this paper proposes the new improved algorithm based on the total variation model.The main contents of this paper include the following two aspects:(1)For optimizing the numerical calculation algorithm of model,a numerical algorithm of adaptive total variation model is proposed.It is the most commonly used numerical method to solve the partial differential equation based on the total variation model.Therefore,by optimizing the solution process of PDE,this paper proposes a numerical solution algorithm of adaptive total variation model.First,the corresponding partial differential equation is obtained by constructing the model functional;then,the diffusion function of the partial differential equation is solved;finally,the diffusion function is disseretized numerically.The experiment results show that the algorithm can effectively minimize the adaptive total variation model and get the effective denoising effect.Besides,this paper also analyzes the value of weight ? in the adaptive total variadon model,and obtains the most suitable value ? through a large number of experiments.(2)For improving the model,a region fusion based split Bregman(RFSB)denoising algorithm is proposed.First of all,the algorithm introduces the segmentation algorithm of image region;then;Then,the split Bregman algorithm is used to solve the total variation model and the isotropic diffusion model,and the region segmentation algorithm is used to combine the two models;finally,it uses the fast non local mean filtering as the post-processing,improving the denoising effect.Experiment results show that compared with other denoising algorithms based on the total variation model,the new algorithm proposed in this paper not only has a very good improvement in the denoising effect,but also the time complexity has been optimized.