Open AccessMulti-scale Image Denoising via a Regularization Method
Author(s) -
Yi Tan,
Jintu Fan,
Dong Sun,
Qingwei Gao,
Yue Lu
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2253/1/012030
Subject(s) - deblurring , image restoration , regularization (linguistics) , artificial intelligence , computer science , noise reduction , total variation denoising , image denoising , image processing , non local means , image (mathematics) , computer vision , pattern recognition (psychology) , mathematics
Image restoration is a widely studied problem in the field of image processing. Although the existing image restoration methods based on denoising regularization have shown relatively well performance, image restoration methods for different features of unknown images have not been proposed. Since images have different features, it seems necessary to adopt different priori regular terms for different features. In this paper, we propose a multiscale image regularization denoising framework that can simultaneously perform two or more denoising prior regularization terms to better obtain the overall image restoration results. We use the alternating direction multiplier method (ADMM) to optimize the model and combine multiple denoising algorithms for extensive image deblurring and image super-resolution experiments, and our algorithm shows better performance compared to the existing state-of-the-art image restoration methods.