Learning optimal filter representation for texture classification

Zhang Peng, Jing Peng, Bill Buckles

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

2 Citations (Scopus)

Abstract

Crucial to texture classification are texture features and classifiers that operate on the features. There are several approaches to computing texture features. Of particular interest is multichannel filtering because of its simplicity. Multichannel filtering works by decomposing the frequency domain of an image, resulting in a bank of filtered feature images. Many techniques have been proposed to optimize multichannel filtering. However, the optimization is with respect to image representation, thus giving no guarantee for texture classification. This paper proposes a novel technique for learning optimal filters for texture classification. We use regularization techniques such as support vector machines (SVMs) to learn multichannel filters. Since filter training in our approach is naturally tied to classifier training, the resulting filters are optimized for classification. Experimental results validate the efficacy of our proposed technique.

Original languageEnglish
Title of host publicationProceedings - 18th International Conference on Pattern Recognition, ICPR 2006
Pages1138-1141
Number of pages4
DOIs
StatePublished - 1 Dec 2006
Event18th International Conference on Pattern Recognition, ICPR 2006 - Hong Kong, China
Duration: 20 Aug 200624 Aug 2006

Publication series

NameProceedings - International Conference on Pattern Recognition
Volume2
ISSN (Print)1051-4651

Other

Other18th International Conference on Pattern Recognition, ICPR 2006
CountryChina
CityHong Kong
Period20/08/0624/08/06

Fingerprint

Textures
Classifiers
Support vector machines

Cite this

Peng, Z., Peng, J., & Buckles, B. (2006). Learning optimal filter representation for texture classification. In Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006 (pp. 1138-1141). [1699410] (Proceedings - International Conference on Pattern Recognition; Vol. 2). https://doi.org/10.1109/ICPR.2006.753
Peng, Zhang ; Peng, Jing ; Buckles, Bill. / Learning optimal filter representation for texture classification. Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006. 2006. pp. 1138-1141 (Proceedings - International Conference on Pattern Recognition).
@inproceedings{d01f22eb20bf4d5b8ea5919862f2c1e8,
title = "Learning optimal filter representation for texture classification",
abstract = "Crucial to texture classification are texture features and classifiers that operate on the features. There are several approaches to computing texture features. Of particular interest is multichannel filtering because of its simplicity. Multichannel filtering works by decomposing the frequency domain of an image, resulting in a bank of filtered feature images. Many techniques have been proposed to optimize multichannel filtering. However, the optimization is with respect to image representation, thus giving no guarantee for texture classification. This paper proposes a novel technique for learning optimal filters for texture classification. We use regularization techniques such as support vector machines (SVMs) to learn multichannel filters. Since filter training in our approach is naturally tied to classifier training, the resulting filters are optimized for classification. Experimental results validate the efficacy of our proposed technique.",
author = "Zhang Peng and Jing Peng and Bill Buckles",
year = "2006",
month = "12",
day = "1",
doi = "10.1109/ICPR.2006.753",
language = "English",
isbn = "0769525210",
series = "Proceedings - International Conference on Pattern Recognition",
pages = "1138--1141",
booktitle = "Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006",

}

Peng, Z, Peng, J & Buckles, B 2006, Learning optimal filter representation for texture classification. in Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006., 1699410, Proceedings - International Conference on Pattern Recognition, vol. 2, pp. 1138-1141, 18th International Conference on Pattern Recognition, ICPR 2006, Hong Kong, China, 20/08/06. https://doi.org/10.1109/ICPR.2006.753

Learning optimal filter representation for texture classification. / Peng, Zhang; Peng, Jing; Buckles, Bill.

Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006. 2006. p. 1138-1141 1699410 (Proceedings - International Conference on Pattern Recognition; Vol. 2).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Learning optimal filter representation for texture classification

AU - Peng, Zhang

AU - Peng, Jing

AU - Buckles, Bill

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Crucial to texture classification are texture features and classifiers that operate on the features. There are several approaches to computing texture features. Of particular interest is multichannel filtering because of its simplicity. Multichannel filtering works by decomposing the frequency domain of an image, resulting in a bank of filtered feature images. Many techniques have been proposed to optimize multichannel filtering. However, the optimization is with respect to image representation, thus giving no guarantee for texture classification. This paper proposes a novel technique for learning optimal filters for texture classification. We use regularization techniques such as support vector machines (SVMs) to learn multichannel filters. Since filter training in our approach is naturally tied to classifier training, the resulting filters are optimized for classification. Experimental results validate the efficacy of our proposed technique.

AB - Crucial to texture classification are texture features and classifiers that operate on the features. There are several approaches to computing texture features. Of particular interest is multichannel filtering because of its simplicity. Multichannel filtering works by decomposing the frequency domain of an image, resulting in a bank of filtered feature images. Many techniques have been proposed to optimize multichannel filtering. However, the optimization is with respect to image representation, thus giving no guarantee for texture classification. This paper proposes a novel technique for learning optimal filters for texture classification. We use regularization techniques such as support vector machines (SVMs) to learn multichannel filters. Since filter training in our approach is naturally tied to classifier training, the resulting filters are optimized for classification. Experimental results validate the efficacy of our proposed technique.

UR - http://www.scopus.com/inward/record.url?scp=34047192530&partnerID=8YFLogxK

U2 - 10.1109/ICPR.2006.753

DO - 10.1109/ICPR.2006.753

M3 - Conference contribution

SN - 0769525210

SN - 9780769525211

T3 - Proceedings - International Conference on Pattern Recognition

SP - 1138

EP - 1141

BT - Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006

ER -

Peng Z, Peng J, Buckles B. Learning optimal filter representation for texture classification. In Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006. 2006. p. 1138-1141. 1699410. (Proceedings - International Conference on Pattern Recognition). https://doi.org/10.1109/ICPR.2006.753