Abstract

In this research study, we describe an enhanced automated vision-based system for the classification of facial expressions. The face within an image is firstly localized using a simplified method then it will be characterized in three different ways; by compacting its geometric characteristics using Zernike moments feature vector then by obtaining its spectral source model through AR Burg representation. Finally, a statistical distribution analysis of the luminance information is performed through the LBP method. The three characterization feature vectors are used separately to train three back propagation neural networks to perform facial expression recognition. Combined feature vectors are used to train neural networks. Finally, a fuzzy linear discriminant analysis is applied to these combined feature vectors in order to enhance facial expression recognition process. Experiments performed on the JAFFE database along with comparisons to other methods have affirmed the potency of the proposed approach attaining promising results compared to those reported in the literature.

Information

Item Type:	Conference
Divisions:	» Laboratory of Electrical Engineering and Renewable Energy » Faculty of Science and Technology
ePrint ID:	329
Date Deposited:	2015-03-30
Further Information:	Google Scholar
URI:	https://www.univ-soukahras.dz/en/publication/article/329

BibTex