A Double Fittings Based Total Variation Model for Image Denoising
Mengmeng Li, Bing‐Zhao Li · 2021 IEEE 6th International Conference on Signal and Image Processing (ICSIP) · 2021
In this paper, weighted image gradients based fitting is introduced into the proposed model for image denoising, which is called double fittings based total variation (DFTV) model. The weighted gradients are designed to maintain image structures during the denoising process. The definition of the weight function makes it better to enhance edges of images and maintain structures. For solving the proposed model, alternating direction method of multipliers (ADMM) is explored. In order to show the effectiveness and efficiency of the proposed model, both of the visual results and quantity evaluation are provided for compared experiments.