MAXIMUM LIKELIHOOD ESTIMATION OF GAUSSIAN MARKOV RANDOM FIELD PARAMETERS.

Chee Sun Won, Haluk Derin

Research output: Contribution to journalConference articlepeer-review

5 Scopus citations

Abstract

The problem is discussed of fitting Gaussian Markov random field (GMRF) models to natural textures through maximum likelihood (ML) estimation of the model parameters. In particular, in two implementation issues on ML estimation are emphasized. First, the log-likelihood function to be maximized is not a concave function with respect to the parameters. The second concern comes from the fact that the suppose-to-be covariance matrix expressed in terms of the estimated parameters must be non-negative definite. To resolve these difficulties one uses a version of the so-called Multi-Start algorithm, which is a variation of the deterministic relaxation algorithm. Some experimental results with natural textures show that often the best fit of natural images to GMRF models via ML estimation occurs at the boundary of the allowable parameter set.

Original languageEnglish
Pages (from-to)1040-1043
Number of pages4
JournalProceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing
StatePublished - 1988

Fingerprint

Dive into the research topics of 'MAXIMUM LIKELIHOOD ESTIMATION OF GAUSSIAN MARKOV RANDOM FIELD PARAMETERS.'. Together they form a unique fingerprint.

Cite this