论文信息 - Stochastic relaxations and em algorithms for markov random fields

Stochastic relaxations and em algorithms for markov random fields

This paper presents a simulation study of block (one line or two lines of pixels) updating for Markov random fields. Point and line relaxation methods are compared. Some pseudo-likelihoods, based on the conditional density of a block of pixels, are used together with modified EM algorithms to estimate parameters from noisy images.

W. Qian | D. Titterington

[1] J. Besag. Spatial Interaction and the Statistical Analysis of Lattice Systems , 1974 .

[2] J. Besag. Statistical Analysis of Non-Lattice Data , 1975 .

[3] D. K. Pickard. Asymptotic inference for an Ising lattice , 1976 .

[4] D. Rubin,et al. Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[5] Donald Geman,et al. Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6] D. M. Titterington,et al. Image labelling and the statistical analysis of incomplete data , 1986 .

[7] J. Besag. On the Statistical Analysis of Dirty Pictures , 1986 .

[8] Wang,et al. Nonuniversal critical dynamics in Monte Carlo simulations. , 1987, Physical review letters.

[9] D. K. Pickard. Inference for Discrete Markov Fields: The Simplest Nontrivial Case , 1987 .

[10] W. Qian,et al. On the use of Gibbs Markov chain models in the analysis of images based on second-order pairwise interactive distributions , 1989 .

[11] A note on asymptotic inference for an asymmetric Ising model around a torus , 1990 .

[12] B. D. Riley,et al. Iterative simulation methods , 1990 .

[13] W. Qian,et al. Parameter estimation for hidden Gibbs chains , 1990 .

[14] Gerhard Winkler. AN ERGODIC L2-THEOREM FOR SIMULATED ANNEALING IN BAYESIAN IMAGE RECONSTRUCTION , 1990 .