Genetic Feature Selection for Texture Classification Using 2-D Non-Separable Wavelet Bases

Jing-Wein WANG; Chin-Hsing CHEN; Jeng-Shyang PAN

Genetic Feature Selection for Texture Classification Using 2-D Non-Separable Wavelet Bases

Jing-Wein WANG, Chin-Hsing CHEN, Jeng-Shyang PAN

Full Text Views

0

Share
Cite this

Summary :

In this paper, the performances of texture classification based on pyramidal and uniform decomposition are comparatively studied with and without feature selection. This comparison using the subband variance as feature explores the dependence among features. It is shown that the main problem when employing 2-D non-separable wavelet transforms for texture classification is the determination of the suitable features that yields the best classification results. A Max-Max algorithm which is a novel evaluation function based on genetic algorithms is presented to evaluate the classification performance of each subset of selected features. It is shown that the performance with feature selection in which only about half of features are selected is comparable to that without feature selection. Moreover, the discriminatory characteristics of texture spread more in low-pass bands and the features extracted from the pyramidal decomposition are more representative than those from the uniform decomposition. Experimental results have verified the selectivity of the proposed approach and its texture capturing characteristics.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E81-A No.8 pp.1635-1644

Publication Date: 1998/08/25

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Section on Digital Signal Processing)

Category

Cite this

Copy

Jing-Wein WANG, Chin-Hsing CHEN, Jeng-Shyang PAN, "Genetic Feature Selection for Texture Classification Using 2-D Non-Separable Wavelet Bases" in IEICE TRANSACTIONS on Fundamentals, vol. E81-A, no. 8, pp. 1635-1644, August 1998, doi: .
Abstract: In this paper, the performances of texture classification based on pyramidal and uniform decomposition are comparatively studied with and without feature selection. This comparison using the subband variance as feature explores the dependence among features. It is shown that the main problem when employing 2-D non-separable wavelet transforms for texture classification is the determination of the suitable features that yields the best classification results. A Max-Max algorithm which is a novel evaluation function based on genetic algorithms is presented to evaluate the classification performance of each subset of selected features. It is shown that the performance with feature selection in which only about half of features are selected is comparable to that without feature selection. Moreover, the discriminatory characteristics of texture spread more in low-pass bands and the features extracted from the pyramidal decomposition are more representative than those from the uniform decomposition. Experimental results have verified the selectivity of the proposed approach and its texture capturing characteristics.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/e81-a_8_1635/_p

Copy

@ARTICLE{e81-a_8_1635,
author={Jing-Wein WANG, Chin-Hsing CHEN, Jeng-Shyang PAN, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Genetic Feature Selection for Texture Classification Using 2-D Non-Separable Wavelet Bases},
year={1998},
volume={E81-A},
number={8},
pages={1635-1644},
abstract={In this paper, the performances of texture classification based on pyramidal and uniform decomposition are comparatively studied with and without feature selection. This comparison using the subband variance as feature explores the dependence among features. It is shown that the main problem when employing 2-D non-separable wavelet transforms for texture classification is the determination of the suitable features that yields the best classification results. A Max-Max algorithm which is a novel evaluation function based on genetic algorithms is presented to evaluate the classification performance of each subset of selected features. It is shown that the performance with feature selection in which only about half of features are selected is comparable to that without feature selection. Moreover, the discriminatory characteristics of texture spread more in low-pass bands and the features extracted from the pyramidal decomposition are more representative than those from the uniform decomposition. Experimental results have verified the selectivity of the proposed approach and its texture capturing characteristics.},
keywords={},
doi={},
ISSN={},
month={August},}

Copy

TY - JOUR
TI - Genetic Feature Selection for Texture Classification Using 2-D Non-Separable Wavelet Bases
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1635
EP - 1644
AU - Jing-Wein WANG
AU - Chin-Hsing CHEN
AU - Jeng-Shyang PAN
PY - 1998
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E81-A
IS - 8
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - August 1998
AB - In this paper, the performances of texture classification based on pyramidal and uniform decomposition are comparatively studied with and without feature selection. This comparison using the subband variance as feature explores the dependence among features. It is shown that the main problem when employing 2-D non-separable wavelet transforms for texture classification is the determination of the suitable features that yields the best classification results. A Max-Max algorithm which is a novel evaluation function based on genetic algorithms is presented to evaluate the classification performance of each subset of selected features. It is shown that the performance with feature selection in which only about half of features are selected is comparable to that without feature selection. Moreover, the discriminatory characteristics of texture spread more in low-pass bands and the features extracted from the pyramidal decomposition are more representative than those from the uniform decomposition. Experimental results have verified the selectivity of the proposed approach and its texture capturing characteristics.
ER -