Cluster expansions for the deterministic computation of Bayesian estimators based on Markov random fields
- 1 March 1995
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 17 (3) , 275-293
- https://doi.org/10.1109/34.368192
Abstract
We describe a family of approximations, denoted by “cluster approximations”, for the computation of the mean of a Markov random field (MRF). This is a key computation in image processing when applied to the a posteriori MRF. The approximation is to account exactly for only spatially local interactions. Application of the approximation requires the solution of a nonlinear multivariable fixed-point equation for which we prove several existence, uniqueness, and convergence-of-algorithm results. Four numerical examples are presented, including comparison with Monte Carlo calculationsKeywords
This publication has 30 references indexed in Scilit:
- The mean field theory in EM procedures for blind Markov random field image restorationIEEE Transactions on Image Processing, 1993
- Compound Gauss-Markov random fields for image estimationIEEE Transactions on Signal Processing, 1991
- Numerical Continuation MethodsPublished by Springer Nature ,1990
- Simulated annealing in compound Gaussian random fields (image processing)IEEE Transactions on Information Theory, 1990
- Learning the Parameters of a Hidden Markov Random Field Image Model: A Simple ExamplePublished by Springer Nature ,1987
- Estimation and choice of neighbors in spatial-interaction models of imagesIEEE Transactions on Information Theory, 1983
- Digital image restoration using spatial interaction modelsIEEE Transactions on Acoustics, Speech, and Signal Processing, 1982
- Advances in mathematical models for image processingProceedings of the IEEE, 1981
- Markov image modelingIEEE Transactions on Automatic Control, 1978
- On Gaussian Fields with Given Conditional DistributionsTheory of Probability and Its Applications, 1967