Edge adaptive hybrid regularization model for image deblurring

Abstract Parameter selection is crucial to regularization-based image restoration methods. Generally speaking, a spatially fixed parameter for the regularization term does not perform well both edge and smooth areas. A larger reduces noise better in areas but blurs regions, while small sharpens causes residual noise. In this paper, an automated adaptive model, which combines harmonic total variation (TV) terms, proposed reconstruction from noisy blurred observation. The model detects edges then adjusts parameters of Tikhonov TV terms each pixel according information. Accordingly, information matrix will also be dynamically updated during iterations. Computationally, newly-established convex, can solved by semi-proximal alternating direction method multipliers with linear convergence rate. Numerical simulation results demonstrate that effectively preserves eliminates blur at same time. comparison state-of-the-art algorithms, it outperforms other methods peak signal ratio, structural similarity index visual quality.