Texture classification using spectral histograms

被引:131
作者
Liu, XW [1 ]
Wang, DL
机构
[1] Florida State Univ, Dept Comp Sci, Tallahassee, FL 32306 USA
[2] Ohio State Univ, Dept Comp & Informat Sci, Ctr Cognit Sci, Columbus, OH 43210 USA
基金
美国国家科学基金会;
关键词
filtering; spectral histogram; texture analysis; texture classification;
D O I
10.1109/TIP.2003.812327
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Based on a local spatial/frequency representation,we employ a spectral histogram as a feature statistic for texture classification. The spectral histogram consists of marginal distributions of responses of a bank of filters and encodes implicitly the local structure of images through the filtering stage and the global appearance through the histogram stage. The distance between two spectral histograms is measured using chi(2)-statistic. The spectral histogram with the associated distance measure exhibits several properties that are necessary for texture classification. A filter selection algorithm is proposed to maximize classification performance of a given dataset. Our classification experiments using natural texture images reveal that the spectral histogram representation provides a robust feature statistic for textures and generalizes well. Comparisons show that our method produces a marked improvement in classification performance. Finally we point out the relationships between existing texture features and the spectral histogram, suggesting that the latter may provide a unified texture feature.
引用
收藏
页码:661 / 670
页数:10
相关论文
共 44 条
[1]  
[Anonymous], 1995, DIALOGUES PERCEPTION
[2]   Texture classification using windowed Fourier filters [J].
Azencott, R ;
Wang, JP ;
Younes, L .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1997, 19 (02) :148-153
[3]   The ''independent components'' of natural scenes are edge filters [J].
Bell, AJ ;
Sejnowski, TJ .
VISION RESEARCH, 1997, 37 (23) :3327-3338
[4]   EARLY VISION AND TEXTURE-PERCEPTION [J].
BERGEN, JR ;
ADELSON, EH .
NATURE, 1988, 333 (6171) :363-364
[5]   APPLICATION OF FOURIER ANALYSIS TO VISIBILITY OF GRATINGS [J].
CAMPBELL, FW ;
ROBSON, JG .
JOURNAL OF PHYSIOLOGY-LONDON, 1968, 197 (03) :551-&
[7]   Texture segmentation using Gaussian-Markov random fields and neural oscillator networks [J].
Çesmeli, E ;
Wang, DL .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2001, 12 (02) :394-404
[8]   MARKOV RANDOM FIELD TEXTURE MODELS [J].
CROSS, GR ;
JAIN, AK .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1983, 5 (01) :25-39
[9]  
DAUGMAN J, 1985, J OPT SOC AM A, V2, P23
[10]   Texture synthesis and pattern recognition for partially ordered Markov models [J].
Davidson, JL ;
Cressie, N ;
Hua, X .
PATTERN RECOGNITION, 1999, 32 (09) :1475-1505