Optimizing Gabor Filter and Local Binary Patterns for multi-texture classification

Alsadegh Mohamed, Zhongyu Lou, Qiang Xu, Jinwen Ma

Research output: Contribution to journalArticle

Abstract

In image classification by texture, it is important to maximize the discrimination
between different classes by using an effective descriptor. The objective of this research is a new hybrid approach using state of the art feature extraction methods and improving the classification percentage of optimum filter by combining it with optimized LBP and find low dimensional size of features. The Gabor filter (GF) parameters are processed by the Artificial Bee Colony (ABC) algorithm to select the optimum filter, whereas pertinent features from LBP histogram are obtained using Rough Set Theory (RST) without impacting
its classification rate. The classification implemented on texture classes is obtained from the Brodatz database. The results from the proposed approach show an improvement in the classification accuracy and processing time of k-folded cross- validation Neural Network classifier over the method of LBP with single filter and a reduced processing time of the classifier.
LanguageEnglish
Article number5
Pages55-67
Number of pages13
JournalEgyptian Computer Science Journal
Volume42
Issue number2
Publication statusPublished - May 2018

Fingerprint

Gabor filters
Textures
Classifiers
Rough set theory
Image classification
Processing
Feature extraction
Neural networks

Cite this

@article{7624fe555f5d4fe99edba75441f5c78f,
title = "Optimizing Gabor Filter and Local Binary Patterns for multi-texture classification",
abstract = "In image classification by texture, it is important to maximize the discriminationbetween different classes by using an effective descriptor. The objective of this research is a new hybrid approach using state of the art feature extraction methods and improving the classification percentage of optimum filter by combining it with optimized LBP and find low dimensional size of features. The Gabor filter (GF) parameters are processed by the Artificial Bee Colony (ABC) algorithm to select the optimum filter, whereas pertinent features from LBP histogram are obtained using Rough Set Theory (RST) without impactingits classification rate. The classification implemented on texture classes is obtained from the Brodatz database. The results from the proposed approach show an improvement in the classification accuracy and processing time of k-folded cross- validation Neural Network classifier over the method of LBP with single filter and a reduced processing time of the classifier.",
keywords = "Feature selection, ABC algorithm optimization, Rough Set Method,, GF feature extraction, LBP, Brodatz",
author = "Alsadegh Mohamed and Zhongyu Lou and Qiang Xu and Jinwen Ma",
year = "2018",
month = "5",
language = "English",
volume = "42",
pages = "55--67",
journal = "Egyptian Computer Science Journal",
issn = "1110-2586",
number = "2",

}

Optimizing Gabor Filter and Local Binary Patterns for multi-texture classification. / Mohamed, Alsadegh; Lou, Zhongyu; Xu, Qiang; Ma, Jinwen.

In: Egyptian Computer Science Journal, Vol. 42, No. 2, 5, 05.2018, p. 55-67.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Optimizing Gabor Filter and Local Binary Patterns for multi-texture classification

AU - Mohamed, Alsadegh

AU - Lou, Zhongyu

AU - Xu, Qiang

AU - Ma, Jinwen

PY - 2018/5

Y1 - 2018/5

N2 - In image classification by texture, it is important to maximize the discriminationbetween different classes by using an effective descriptor. The objective of this research is a new hybrid approach using state of the art feature extraction methods and improving the classification percentage of optimum filter by combining it with optimized LBP and find low dimensional size of features. The Gabor filter (GF) parameters are processed by the Artificial Bee Colony (ABC) algorithm to select the optimum filter, whereas pertinent features from LBP histogram are obtained using Rough Set Theory (RST) without impactingits classification rate. The classification implemented on texture classes is obtained from the Brodatz database. The results from the proposed approach show an improvement in the classification accuracy and processing time of k-folded cross- validation Neural Network classifier over the method of LBP with single filter and a reduced processing time of the classifier.

AB - In image classification by texture, it is important to maximize the discriminationbetween different classes by using an effective descriptor. The objective of this research is a new hybrid approach using state of the art feature extraction methods and improving the classification percentage of optimum filter by combining it with optimized LBP and find low dimensional size of features. The Gabor filter (GF) parameters are processed by the Artificial Bee Colony (ABC) algorithm to select the optimum filter, whereas pertinent features from LBP histogram are obtained using Rough Set Theory (RST) without impactingits classification rate. The classification implemented on texture classes is obtained from the Brodatz database. The results from the proposed approach show an improvement in the classification accuracy and processing time of k-folded cross- validation Neural Network classifier over the method of LBP with single filter and a reduced processing time of the classifier.

KW - Feature selection

KW - ABC algorithm optimization

KW - Rough Set Method,

KW - GF feature extraction

KW - LBP

KW - Brodatz

UR - http://ecsjournal.org/Archive/Volume42_Issue2.aspx

M3 - Article

VL - 42

SP - 55

EP - 67

JO - Egyptian Computer Science Journal

T2 - Egyptian Computer Science Journal

JF - Egyptian Computer Science Journal

SN - 1110-2586

IS - 2

M1 - 5

ER -