NONCAUSAL GAUSS-MARKOV RANDOM-FIELDS - PARAMETER STRUCTURE AND ESTIMATION

被引：56

作者：

BALRAM, N ^{[1
]}

MOURA, JMF ^{[1
]}

机构：

[1] CARNEGIE MELLON UNIV,DEPT ELECT & COMP ENGN,PITTSBURGH,PA 15213

来源：

IEEE TRANSACTIONS ON INFORMATION THEORY | 1993年 / 39卷 / 04期

关键词：

RANDOM FIELDS; NONCAUSAL; GAUSS MARKOV RANDOM FIELDS; MAXIMUM LIKELIHOOD PARAMETER ESTIMATION; PARAMETER SPACE IN RANDOM FIELDS;

D O I：

10.1109/18.243450

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The parameter structure of noncausal homogeneous Gauss Markov random fields (GMRF) defined on finite lattices is studied. For first-order (nearest neighbor) and a special class of second-order fields, we provide a complete characterization of the parameter space and a fast implementation of the maximum likelihood (ML) estimator of the field parameters. For general higher order fields, tight bounds for the parameter space are presented and an efficient procedure for ML estimation is described. Experimental results illustrate the application of the approach presented and the viability of the present method in fitting noncausal models to 2-D data.

引用

页码：1333 / 1355

页数：23

共 38 条

[1] CLASSIFICATION OF BINARY RANDOM PATTERNS [J].

ABEND, K ;

HARLEY, TJ ;

KANAL, LN .

IEEE TRANSACTIONS ON INFORMATION THEORY, 1965, 11 (04) :538-544

[2]

Andrews H, 1977, DIGITAL IMAGE RESTOR

[3]

[Anonymous], 2019, MATRIX DIFFERENTIAL, DOI DOI 10.1002/9781119541219.CH5

[4]

BALRAM N, 1992, THESIS CARNEGIE MELL

[5] STATISTICAL-ANALYSIS OF NON-LATTICE DATA [J].

BESAG, J .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES D-THE STATISTICIAN, 1975, 24 (03) :179-195

[6]

BESAG J, 1974, J ROY STAT SOC B MET, V36, P192

[7]

BESAG JE, 1975, BIOMETRIKA, V62, P555

[8]

BESAG JE, 1977, J R STATIST SOC B, P73

[9] DIGITAL IMAGE-RESTORATION USING SPATIAL INTERACTION MODELS [J].

CHELLAPPA, R ;

KASHYAP, RL .

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1982, 30 (03) :461-472

[10]

Chellappa R., 1985, PATTERN RECOGNITION, V2, P79

← 1 2 3 4 →