A modified genetic model based on the queen bee algorithm for facial expression classification

Amir Jamshidnezhad, Md. Jan Nordin

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Genetic Algorithm (GA) as an intelligent technique simulates human learning process. In this paper, the bee mating process is modeled as a modified GA which is combined with the fuzzy system to recognize the facial expression from the images in the Human Computer Interaction (HCI) machines. In the proposed hybrid model the core of expression recognition system is a Mamdani-type fuzzy rule based system to classify the emotions, also, a proposed Genetic Algorithm called Queen Bee Algorithm (QBA) is used with the purpose of making better performance and parameter optimization to improve the accuracy and robustness of the system. Therefore, QBA as a training technique sets the fuzzy membership functions under the adverse conditions. To evaluate the system performance, images from FG-Net database (FEED) were used to obtain the best functions parameters. Results showed that the hybrid model under the training process will be matched with every database automatically. Also, QBA showed higher accuracy rate of classification compare with using classic GA for facial expression recognition.

Original languageEnglish
Pages (from-to)1109-1114
Number of pages6
JournalJournal of Computational and Theoretical Nanoscience
Volume9
Issue number8
DOIs
Publication statusPublished - 2012

Fingerprint

bees
Facial Expression
genetic algorithms
Genetic algorithms
Genetic Algorithm
Model-based
Hybrid Model
education
emotions
fuzzy systems
membership functions
Facial Expression Recognition
Fuzzy Rule-based Systems
Fuzzy Membership Function
Performance Optimization
Knowledge based systems
Parameter Optimization
Fuzzy rules
Fuzzy systems
Membership functions

Keywords

  • Classifier
  • Facial Expression Recognition
  • Fuzzy System
  • Genetic Algorithm
  • Queen Bee Algorithm

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Electrical and Electronic Engineering
  • Materials Science(all)
  • Computational Mathematics
  • Chemistry(all)

Cite this

A modified genetic model based on the queen bee algorithm for facial expression classification. / Jamshidnezhad, Amir; Nordin, Md. Jan.

In: Journal of Computational and Theoretical Nanoscience, Vol. 9, No. 8, 2012, p. 1109-1114.

Research output: Contribution to journalArticle

@article{3541912928bc4491afe315e34f778a9c,
title = "A modified genetic model based on the queen bee algorithm for facial expression classification",
abstract = "Genetic Algorithm (GA) as an intelligent technique simulates human learning process. In this paper, the bee mating process is modeled as a modified GA which is combined with the fuzzy system to recognize the facial expression from the images in the Human Computer Interaction (HCI) machines. In the proposed hybrid model the core of expression recognition system is a Mamdani-type fuzzy rule based system to classify the emotions, also, a proposed Genetic Algorithm called Queen Bee Algorithm (QBA) is used with the purpose of making better performance and parameter optimization to improve the accuracy and robustness of the system. Therefore, QBA as a training technique sets the fuzzy membership functions under the adverse conditions. To evaluate the system performance, images from FG-Net database (FEED) were used to obtain the best functions parameters. Results showed that the hybrid model under the training process will be matched with every database automatically. Also, QBA showed higher accuracy rate of classification compare with using classic GA for facial expression recognition.",
keywords = "Classifier, Facial Expression Recognition, Fuzzy System, Genetic Algorithm, Queen Bee Algorithm",
author = "Amir Jamshidnezhad and Nordin, {Md. Jan}",
year = "2012",
doi = "10.1166/jctn.2012.2152",
language = "English",
volume = "9",
pages = "1109--1114",
journal = "Journal of Computational and Theoretical Nanoscience",
issn = "1546-1955",
publisher = "American Scientific Publishers",
number = "8",

}

TY - JOUR

T1 - A modified genetic model based on the queen bee algorithm for facial expression classification

AU - Jamshidnezhad, Amir

AU - Nordin, Md. Jan

PY - 2012

Y1 - 2012

N2 - Genetic Algorithm (GA) as an intelligent technique simulates human learning process. In this paper, the bee mating process is modeled as a modified GA which is combined with the fuzzy system to recognize the facial expression from the images in the Human Computer Interaction (HCI) machines. In the proposed hybrid model the core of expression recognition system is a Mamdani-type fuzzy rule based system to classify the emotions, also, a proposed Genetic Algorithm called Queen Bee Algorithm (QBA) is used with the purpose of making better performance and parameter optimization to improve the accuracy and robustness of the system. Therefore, QBA as a training technique sets the fuzzy membership functions under the adverse conditions. To evaluate the system performance, images from FG-Net database (FEED) were used to obtain the best functions parameters. Results showed that the hybrid model under the training process will be matched with every database automatically. Also, QBA showed higher accuracy rate of classification compare with using classic GA for facial expression recognition.

AB - Genetic Algorithm (GA) as an intelligent technique simulates human learning process. In this paper, the bee mating process is modeled as a modified GA which is combined with the fuzzy system to recognize the facial expression from the images in the Human Computer Interaction (HCI) machines. In the proposed hybrid model the core of expression recognition system is a Mamdani-type fuzzy rule based system to classify the emotions, also, a proposed Genetic Algorithm called Queen Bee Algorithm (QBA) is used with the purpose of making better performance and parameter optimization to improve the accuracy and robustness of the system. Therefore, QBA as a training technique sets the fuzzy membership functions under the adverse conditions. To evaluate the system performance, images from FG-Net database (FEED) were used to obtain the best functions parameters. Results showed that the hybrid model under the training process will be matched with every database automatically. Also, QBA showed higher accuracy rate of classification compare with using classic GA for facial expression recognition.

KW - Classifier

KW - Facial Expression Recognition

KW - Fuzzy System

KW - Genetic Algorithm

KW - Queen Bee Algorithm

UR - http://www.scopus.com/inward/record.url?scp=84863911389&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863911389&partnerID=8YFLogxK

U2 - 10.1166/jctn.2012.2152

DO - 10.1166/jctn.2012.2152

M3 - Article

AN - SCOPUS:84863911389

VL - 9

SP - 1109

EP - 1114

JO - Journal of Computational and Theoretical Nanoscience

JF - Journal of Computational and Theoretical Nanoscience

SN - 1546-1955

IS - 8

ER -