Bayesian Estimation for Homogeneous and Inhomogeneous Gaussian Random Fields

Authors:
Robert G. Aykroyd
Affiliations:
Univ. of Leeds, Leeds, UK
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
1998

Citing 3
Cited 4

Constrained Restoration and the Recovery of Discontinuities

IEEE Transactions on Pattern Analysis and Machine Intelligence
Exact sampling with coupled Markov chains and applications to statistical mechanics

Proceedings of the seventh international conference on Random structures and algorithms
Unexpected spatial patterns in exponential family auto models

Graphical Models and Image Processing

Approximations for Gibbs Distribution Normalising Constants

Statistics and Computing
Maximum entropy random fields for texture analysis

Pattern Recognition Letters
A statistical approach to the problem of restoring damaged and contaminated images

Pattern Recognition
Data-driven neighborhood selection of a Gaussian field

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.17

Visualization

Abstract

This paper investigates Bayesian estimation for Gaussian Markov random fields. In particular, a new class of compound model is proposed which describes the observed intensities using an inhomogeneous model and the degree of spatial variation described using a second random field. The coupled Markov random fields are used as prior distributions, and combined with Gaussian noise models to produce posterior distributions on which estimation is based. All model parameters are estimated, in a fully Bayesian setting, using the Metropolis-Hastings algorithm. The full posterior estimation procedures are illustrated and compared using various artificial examples. For these examples the inhomogeneous model performs very favorably when compared to the homogeneous model, allowing differential degrees of smoothing and varying local textures.