Texture segmentation based on MRMRF modeling

被引：17

作者：

Wang, L ^{[1
]}

Liu, J ^{[1
]}

机构：

[1] Nanyang Technol Univ, Div Informat Engn, Sch Elect & Elect Engn, Singapore 639798, Singapore

来源：

PATTERN RECOGNITION LETTERS | 2000年 / 21卷 / 02期

关键词：

multiresolution MRF (MRMRF); Markov chain Monte Carlo (MCMC); texture segmentation; parameter estimation;

D O I：

10.1016/S0167-8655(99)00146-4

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Texture segmentation remains a fundamental issue in low-level image analysis, pattern recognition and computer vision. Texture segmentation problem can be solved in two directions: model fitting and non-parametric classification. In this paper, we propose to use multiresolution MRF (MRMRF) modeling in texture segmentation. A novel MRMRF parameter estimation method based on MCMC approach is presented. The experimental result shows that the method is suitable to segment textured images. (C) 2000 Elsevier Science B.V. All rights reserved.

引用

页码：189 / 200

页数：12

共 22 条

[1] IMAGE SEGMENTATION USING SIMPLE MARKOV FIELD MODELS [J].

HANSEN, FR ;

ELLIOTT, H .

COMPUTER GRAPHICS AND IMAGE PROCESSING, 1982, 20 (02) :101-132

[2]

HO WJ, 1991, IEEE IJCNN, P981, DOI 10.1109/IJCNN.1991.170527

[3] SEGMENTATION AND CLASSIFICATION OF RANGE IMAGES [J].

HOFFMAN, R ;

JAIN, AK .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1987, 9 (05) :608-620

[4] ITERATIVE FUZZY IMAGE SEGMENTATION [J].

HUNTSBERGER, TL ;

JACOBS, CL ;

CANNON, RL .

PATTERN RECOGNITION, 1985, 18 (02) :131-138

[5] UNSUPERVISED TEXTURE SEGMENTATION USING GABOR FILTERS [J].

JAIN, AK ;

FARROKHNIA, F .

PATTERN RECOGNITION, 1991, 24 (12) :1167-1186

[6]

Klir G, 1995, Fuzzy Sets and Fuzzy Logic: Theory and Applications, V4

[7] Multiresolution Gauss-Markov random field models for texture segmentation [J].

Krishnamachari, S ;

Chellappa, R .

IEEE TRANSACTIONS ON IMAGE PROCESSING, 1997, 6 (02) :251-267

[8]

LAINE A, 1993, TR93025

[9]

Li SZ., 1995, Markov random field modeling in computer vision

[10] FUZZY THRESHOLDING - MATHEMATICAL FRAMEWORK, BOUND FUNCTIONS AND WEIGHTED MOVING AVERAGE TECHNIQUE [J].

MURTHY, CA ;

PAL, SK .

PATTERN RECOGNITION LETTERS, 1990, 11 (03) :197-206

← 1 2 3 →