Regularization Parameter Selection for Gaussian Mixture Model Based Image Denoising Method

摘要

Regularization parameter selection for image denoising has always been a hot issue. In this paper, an adaptive regularization parameter selection method is exploited for the Gaussian Mixture Model (GMM) based image restoration by combining the gradient matching and the local entropy of the image, which varies with different regions of the image and has a good robustness to noise. Experiment results demonstrate that our proposed adaptive regularization parameter for GMM based image restoration method performs comparatively well, both in visual effects and quantitative evaluations.