An effective alternating direction method of multipliers for color image restoration

Color image restoration is an ill-posed problem, and regularization necessary. In this paper, we first formulate the color problem into a constrained minimization that minimizes quadratic functional subject to constraint total variation of less than given parameter δ. The advantages over traditional unconstrained one δ has obvious physical meaning easy select. However solving generally more difficult corresponding form. We propose effective alternating direction method multipliers for by using structure problem. prove convergence in detail. Experimental results demonstrate proposed feasible much restoration.