Curvelet-Based Iterative Regularization and Inverse Scale Space Methods

For regularization theory of inverse problem in image processing, a challenge is to find a proper space in which the image is well characterized and hence restorable faithfully. Therefore, one contribution of this paper is to propose a novel variational regularization model with the help of decomposition space theory. Furthermore, as an development of it two models for image denoising are given, namely, Curve let-based Iterative regularization method (C-IRM) and Inverse scale spaces (C-ISS) method. Finally, experimental results on some standard test images as well as comparisons with some available methods show that the proposed methods work well in image edge preservation while achieving pleasant performance in terms of Signal-to-noise ratio (SNR).