Research On Mixed Poisson-Gaussian Noise Image Recovery Models And Algorithms

During the process of image capturing,storing,transmitting,and receiving,due to the influence of internal or external factors,it is easy to generate noise,which leads to the degradation of image quality.Traditional noise models often only consider a single type of noise,such as Gaussian noise,Poisson noise and impulse noise.However,considering the complexity of the actual imaging system,the image is often disturbed by mixed noise.Therefore,the study of mixed noise image restoration has important theoretical significance and application value.In this paper,we primarily focus on the total variation infimal convolution(TV-IC)image restoration model for eliminating mixed Poisson-Gaussian noise.The TV-IC model combining Kullback-Leibler and 2-norm data fidelity term works well for the inverse problem of mixed Poisson–Gaussian noise.The existing algorithms for solving the TV-IC model often rely on the Newton method to solve a nonlinear optimization subproblem,which inevitably increases the computation burden.Therefore,we propose two more efficient algorithms to solve the TV-IC model.This paper is divided into four chapters,the details are as follows.In the first chapter,We introduce the research background and the current research status at home and abroad of mixed Poisson-Gaussian noise,give the related preparatory knowledge,and describe the main research content of this paper.In the second chapter,we propose to apply the primal-dual Chambolle-Pock algorithm(PDCP)to solve the TV-IC model and give an effective algorithm to solve the subproblem of the joint proximal operator with the Kullback-Leibler divergence and 2-norm,which is based on the bilinear constraint alternating direction multiplier method.The proposed algorithm can avoid using the Newton iteration method to solve subproblems and is applicable to both mixed Poisson-Gaussian noise denoising and deblurring.The numerical experimental results demonstrate the effectiveness and superiority of the proposed algorithm.In the third chapter,we propose a complete splitting proximal bilinear constraint alternating direction method of multipliers(PBCA)to solve the TV-IC model,the proposed algorithm does not involve any inner iterations,and each subproblem has a closed-form solution.At the same time,we prove the convergence of the proposed algorithm under mild conditions.The numerical experimental results show that the proposed algorithm has better recovery effect and shorter computation time than the state-of-the-art methods.In the fourth chapter,we summarize the paper and give an outlook for future work.