Fusion of Hidden Markov Random Field Models and Its Bayesian Estimation

Author

Destrempes, Francois ; Angers, Jean-Francois ; Mignotte, Max

Author_Institution

DIRO, Univ. de Montreal, Que.

Volume

15

Issue

10

fYear

2006

Firstpage

2920

Lastpage

2935

Abstract

In this paper, we present a Hidden Markov Random Field (HMRF) data-fusion model. The proposed model is applied to the segmentation of natural images based on the fusion of colors and textons into Julesz ensembles. The corresponding Exploration/Selection/Estimation (ESE) procedure for the estimation of the parameters is presented. This method achieves the estimation of the parameters of the Gaussian kernels, the mixture proportions, the region labels, the number of regions, and the Markov hyper-parameter. Meanwhile, we present a new proof of the asymptotic convergence of the ESE procedure, based on original finite time bounds for the rate of convergence

Keywords

Bayes methods; Gaussian processes; hidden Markov models; image colour analysis; image segmentation; parameter estimation; sensor fusion; Bayesian estimation; Gaussian kernels; Julesz ensemble textons; Markov hyper-parameter; asymptotic convergence; color fusion; exploration-selection-estimation procedure; hidden Markov random field data-fusion model; mixture proportions; natural image segmentation; parameter estimation; region labels; Bayesian methods; Biomedical imaging; Convergence; Entropy; Hidden Markov models; Image processing; Image segmentation; Kernel; Monte Carlo methods; Parameter estimation; Bayesian estimation; Exploration/Selection algorithm; Exploration/Selection/Estimation procedure; Julesz ensembles; Markov Chain Monte Carlo (MCMC) algorithm; color and texture segmentation; fusion of hidden Markov random field models;

fLanguage

English

Journal_Title

Image Processing, IEEE Transactions on

Publisher

ieee

ISSN

1057-7149

Type

jour

DOI

10.1109/TIP.2006.877522

Filename

1703583