The satellite image deconvolution problem is ill-posed and must be regularized. Herein, we use an edge-preserving regularization model using a phi function, involving two hyperparameters. Our goal is to estimate the optimal parameters in order to automatically reconstruct images. We propose to use the maximum-likelihood estimator (MLE), applied to the observed image. We need sampling from prior and posterior distributions. Since the convolution prevents use of standard samplers, we have developed a modified Geman-Yang algorithm, using an auxiliary variable and a cosine transform. We present a Markov chain Monte Carlo maximum-likelihood (MCMCML) technique which is able to simultaneously achieve the estimation and the reconstruction.
Description
ScienceDirect - Pattern Recognition : Hyperparameter estimation for satellite image restoration using a MCMC maximum-likelihood method
%0 Journal Article
%1 Jalobeanu2002341
%A Jalobeanu, André
%A Blanc-Féraud, Laure
%A Zerubia, Josiane
%D 2002
%J Pattern Recognition
%K bayesian imageprocessing inference inversions markovrandomfields models regularization remotesensing statistics uncertainty
%N 2
%P 341--352
%R DOI: 10.1016/S0031-3203(00)00178-3
%T Hyperparameter estimation for satellite image restoration using a MCMC maximum-likelihood method
%U http://www.sciencedirect.com/science/article/B6V14-44HT45G-4/2/ebb6155d9b582e44ff8944d34283d91f
%V 35
%X The satellite image deconvolution problem is ill-posed and must be regularized. Herein, we use an edge-preserving regularization model using a phi function, involving two hyperparameters. Our goal is to estimate the optimal parameters in order to automatically reconstruct images. We propose to use the maximum-likelihood estimator (MLE), applied to the observed image. We need sampling from prior and posterior distributions. Since the convolution prevents use of standard samplers, we have developed a modified Geman-Yang algorithm, using an auxiliary variable and a cosine transform. We present a Markov chain Monte Carlo maximum-likelihood (MCMCML) technique which is able to simultaneously achieve the estimation and the reconstruction.
@article{Jalobeanu2002341,
abstract = {The satellite image deconvolution problem is ill-posed and must be regularized. Herein, we use an edge-preserving regularization model using a [phi] function, involving two hyperparameters. Our goal is to estimate the optimal parameters in order to automatically reconstruct images. We propose to use the maximum-likelihood estimator (MLE), applied to the observed image. We need sampling from prior and posterior distributions. Since the convolution prevents use of standard samplers, we have developed a modified Geman-Yang algorithm, using an auxiliary variable and a cosine transform. We present a Markov chain Monte Carlo maximum-likelihood (MCMCML) technique which is able to simultaneously achieve the estimation and the reconstruction.},
added-at = {2009-08-24T16:22:14.000+0200},
author = {Jalobeanu, André and Blanc-Féraud, Laure and Zerubia, Josiane},
biburl = {https://www.bibsonomy.org/bibtex/2f9d4cc271ecc61799e0308065fbacc16/jgomezdans},
description = {ScienceDirect - Pattern Recognition : Hyperparameter estimation for satellite image restoration using a MCMC maximum-likelihood method},
doi = {DOI: 10.1016/S0031-3203(00)00178-3},
interhash = {a09fc2d1823691949177ee71e18ad9e7},
intrahash = {f9d4cc271ecc61799e0308065fbacc16},
issn = {0031-3203},
journal = {Pattern Recognition},
keywords = {bayesian imageprocessing inference inversions markovrandomfields models regularization remotesensing statistics uncertainty},
number = 2,
pages = {341--352},
timestamp = {2009-08-24T16:22:14.000+0200},
title = {Hyperparameter estimation for satellite image restoration using a MCMC maximum-likelihood method},
url = {http://www.sciencedirect.com/science/article/B6V14-44HT45G-4/2/ebb6155d9b582e44ff8944d34283d91f},
volume = 35,
year = 2002
}