Regularized constrained restoration of wavelet-compressed image

Abstract

Wavelet-compressed images suffer from coding artifacts, such as ringing and blurring, resulted from the quantization of transform coefficients. In this paper we propose a new algorithm that reduces such coding artifacts in wavelet-compressed images by using regularized iterative image restoration. We, first, propose an appropriate model for the image degradation system which represents the wavelet-based image compression system. Then the model is used to formulate the regularized iterative restoration algorithm. The proposed algorithm adopts a couple of constraints, and adaptivity is imposed to the general regularization process on both spatial and frequency domain. Experimental results show that the solution of the proposed iteration converges to the image in which both ringing and blurring are significantly reduced.