MRF parameter estimation by MCMC method

作者：

Highlights：

•

摘要

Markov random field (MRF) modeling is a popular pattern analysis method and MRF parameter estimation plays an important role in MRF modeling. In this paper, a method based on Markov Chain Monte Carlo (MCMC) is proposed to estimate MRF parameters. Pseudo-likelihood is used to represent likelihood function and it gives a good estimation result.

论文关键词：MRF,MCMC,Least-squares fit,Parameter estimation,Pseudo-likelihood

论文评审过程：Received 13 January 1999, Revised 28 July 1999, Accepted 28 July 1999, Available online 7 June 2001.

论文官网地址：https://doi.org/10.1016/S0031-3203(99)00178-8