TREE APPROXIMATIONS TO MARKOV RANDOM-FIELDS

被引:20
作者
WU, CH [1 ]
DOERSCHUK, PC [1 ]
机构
[1] PURDUE UNIV,SCH ELECT ENGN,W LAFAYETTE,IN 47907
基金
美国国家科学基金会;
关键词
MARKOV RANDOM FIELD; BAYESIAN ESTIMATION; SPATIAL PATTERN CLASSIFICATION; IMAGE SEGMENTATION; IMAGE RESTORATION;
D O I
10.1109/34.385979
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Methods for approximately computing the marginal probability mass functions and means of a Markov random field (MRF) by approximating the lattice by a tree are described. Applied to the a posteriori MRP these methods solve Bayesian spatial pattern classification and image restoration problems. The methods are described, several theoretical results concerning fixed-point problems are proven, and four numerical examples are presented, including comparison with optimal estimators and the Iterated Conditional Mode estimator and including two agricultural optical remote sensing problems.
引用
收藏
页码:391 / 402
页数:12
相关论文
共 25 条
[1]  
[Anonymous], 1980, MARKOV RANDOM FIELDS, DOI DOI 10.1090/CONM/001
[2]  
Baxter R. J., 2007, EXACTLY SOLVED MODEL
[3]  
BESAG J, 1974, J ROY STAT SOC B MET, V36, P192
[4]  
BESAG J, 1986, J R STAT SOC B, V48, P259
[5]   DESIGN CHALLENGES OF THE THEMATIC MAPPER [J].
BLANCHARD, LE ;
WEINSTEIN, O .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 1980, 18 (02) :146-160
[6]   BAYES SMOOTHING ALGORITHMS FOR SEGMENTATION OF BINARY IMAGES MODELED BY MARKOV RANDOM-FIELDS [J].
DERIN, H ;
ELLIOTT, H ;
CRISTI, R ;
GEMAN, D .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1984, 6 (06) :707-720
[7]  
Devijver P. A., 1986, Eighth International Conference on Pattern Recognition. Proceedings (Cat. No.86CH2342-4), P259
[8]  
DEVIJVER PA, 1987, PATTERN RECOGN, P141
[9]  
ELFADEL IM, 1993, RLE579 MIT RES LAB E
[10]   PARALLEL AND DETERMINISTIC ALGORITHMS FROM MRFS - SURFACE RECONSTRUCTION [J].
GEIGER, D ;
GIROSI, F .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1991, 13 (05) :401-412