Edge structure preserving image denoising using OAGSM/NC statistical model

Abstract

It is a challenging work to design an edge structure preserving image denoising. In recent years, Bayes least squares-Gaussian scale mixtures (BLS-GSM) has emerged as one of the most powerful methods for image denoising. Its strength relies on providing a simple and, yet, very effective local statistical description of oriented pyramid coefficient neighborhoods via a GSM vector. This can be viewed as a fine adaptation of the model to the signal variance at each scale, orientation, and spatial location. Combining with Bayes least squares estimator, we describe a method for removing noise from digital images, based on orientation-adapted GSM with nonoriented component (OAGSM/NC) in shiftable complex directional pyramid (PDTDFB) domain in this paper, which can be seen a modified version of the BLS-GSM. By introducing a coarser adaptation level, we model the distribution of PDTDFB coefficients with OAGSM/NC. The statistical model is then used to obtain the denoised coefficients from the noisy image decomposition by Bayes least squares estimator. Extensive experimental results demonstrate that our method can obtain better performances in terms of both subjective and objective evaluations than those state-of-the-art denoising techniques. Especially, the proposed method can preserve edges very well while removing noise.