Gaussian MRF rotation-invariant features for image classification

被引：116

作者：

Deng, HW ^{[1
]}

Clausi, DA ^{[1
]}

机构：

[1] Univ Waterloo, Dept Syst Design Engn, Waterloo, ON N2L 3G1, Canada

来源：

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 2004年 / 26卷 / 07期

关键词：

Markov random field (MRF); Gaussian MRF (GMRF) model; isotropic; anisotropic; least squares estimate (LSE); discrete Fourier transform (DFT); rotational invariance; texture analysis; classification;

D O I：

10.1109/TPAMI.2004.30

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Features based on Markov random field (MRF) models are sensitive to texture rotation. This paper develops an anisotropic circular Gaussian MRF (ACGMRF) model for retrieving rotation-invariant texture features. To overcome the singularity problem of the least squares estimate method, an approximate least squares estimate method is designed and implemented. Rotation-invariant features are obtained from the ACGMRF model parameters using the discrete Fourier transform. The ACGMRF model is demonstrated to be a statistical improvement over three published methods. The three methods include a Laplacian pyramid, an isotropic circular GMRF (ICGMRF), and gray level cooccurrence probability features.

引用

页码：951 / 955

页数：5

共 15 条

[1]

[Anonymous], TEXTURE PHOTOGRAPHIC

[2] Circular neighbourhood and 1-D DFT features for texture classification and segmentation [J].

Arof, H ;

Deravi, F .

IEE PROCEEDINGS-VISION IMAGE AND SIGNAL PROCESSING, 1998, 145 (03) :167-172

[3]

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

[4]

BISHOP T, 1975, DISCRETE MULTIVARIAT

[5] An analysis of co-occurrence texture statistics as a function of grey level quantization [J].

Clausi, DA .

CANADIAN JOURNAL OF REMOTE SENSING, 2002, 28 (01) :45-62

[6] CLASSIFICATION OF ROTATED AND SCALED TEXTURED IMAGES USING GAUSSIAN MARKOV RANDOM FIELD MODELS [J].

COHEN, FS ;

FAN, ZG ;

PATEL, MA .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1991, 13 (02) :192-202

[7]

GREENSPAN H, 1994, INT C PATT RECOG, P162, DOI 10.1109/ICPR.1994.576896

[8] Rotation-invariant texture classification using a complete space-frequency model [J].

Haley, GM ;

Manjunath, BS .

IEEE TRANSACTIONS ON IMAGE PROCESSING, 1999, 8 (02) :255-269

[9] TEXTURAL FEATURES FOR IMAGE CLASSIFICATION [J].

HARALICK, RM ;

SHANMUGAM, K ;

DINSTEIN, I .

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1973, SMC3 (06) :610-621

[10] A MODEL-BASED METHOD FOR ROTATION INVARIANT TEXTURE CLASSIFICATION [J].

KASHYAP, RL ;

KHOTANZAD, A .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1986, 8 (04) :472-481

← 1 2 →