Image Restoration and Reconstruction using Variable Splitting and\n Class-adapted Image Priors

2016/02/12 by Afonso M. Teodoro, Teodoro, Afonso M., José M. Bioucas‐Dias +3 · 2 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Engineering · #47N10 #68U10 #94A08 #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #I.4.4 #I.4.5 #Image and Signal Denoising Methods #Photoacoustic and Ultrasonic Imaging #Sparse and Compressive Sensing Techniques #Spectroscopy Techniques in Biomedical and Chemical Research

paper · pdf · doi:10.48550/arxiv.1602.04052

openalex publication_date 2016/02/12 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28

Abstract

This paper proposes using a Gaussian mixture model as a prior, for solving\ntwo image inverse problems, namely image deblurring and compressive imaging. We\ncapitalize on the fact that variable splitting algorithms, like ADMM, are able\nto decouple the handling of the observation operator from that of the\nregularizer, and plug a state-of-the-art algorithm into the pure denoising\nstep. Furthermore, we show that, when applied to a specific type of image, a\nGaussian mixture model trained from an database of images of the same type is\nable to outperform current state-of-the-art methods.\n

Citations

• Multiplier and gradient methods
• Fast Image Recovery Using Variable Splitting and Constrained Optimization
• Nonlinear total variation based noise removal algorithms
• rm K-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation

Cited by

• Scene-Adapted Plug-and-Play Algorithm with Guaranteed Convergence: Applications to Data Fusion in Imaging
• Class-specific Poisson denoising by patch-based importance sampling

Related

• Image Restoration with Locally Selected Class-Adapted Models
• Blind image deblurring using class-adapted image priors
• Analysis Plug-and-Play Methods for Imaging Inverse Problems
• Deep Mean-Shift Priors for Image Restoration
• Image denoising with generalized Gaussian mixture model patch priors