Image Recovery By Peaceman-Rachford Operator Splitting Method

In the solution of image recovery inverse problem, the objective functional can often be modeled as the minimization of the sum of two (not necessarily differential) lower semicontinuous convex functional in a real Hilbert space. A decomposition method derived from Peaceman-Rachford splitting algorithm for monotone operator-splitting is proposed to solve this generic problem. Applications to image deblurring and non-Gaussian denoising in a tight frame are also presented. Finally numerical computations comparing with other classical methods are demonstrated.