Facial expression recognition using log-Gabor filters and local binary pattern operators

Facial expression recognition using log-Gabor filters and local binary pattern operators

Lajevardi, S and Hussain, Z 2009, 'Facial expression recognition using log-Gabor filters and local binary pattern operators', in H.E. Dr. Ali Al-Bimani (ed.) Proceedings of the International Conference on Communication, Computer and Power (ICCCP'09), Muscat, Oman, 15 - 18 Feb 2009, pp. 349-353.

Document type:	Conference Paper
Collection:	Conference Papers

Title	Facial expression recognition using log-Gabor filters and local binary pattern operators
Author(s)	Lajevardi, S Hussain, Z
Year	2009
Conference name	2009 International Conference on Communication, Computer and Power (ICCCP'09)
Conference location	Muscat, Oman
Conference dates	15 - 18 Feb 2009
Proceedings title	Proceedings of the International Conference on Communication, Computer and Power (ICCCP'09)
Editor(s)	H.E. Dr. Ali Al-Bimani
Publisher	SQU
Place of publication	Muscat, Oman
Start page	349
End page	353
Total pages	5
Abstract	This study investigates two different methods of feature extraction for person-independent facial expression recognition from images. The logarithmic Gabor filters and the local binary pattern operator (LBP) were used for feature extraction. Then, the optimum features were selected based on minimum redundancy and maximum relevance algorithm (MRMR). Six different facial expressions were considered. The selected features were classified using the naive bayesian (NB) classifier. The percentage of correct classifications varied across different expressions from 62.8% to 90.5% for the log-Gabor filter approach,and from 71.8% to 94.0% for the LBP approach. Experiments carried out on Cohn-Kanade database showed comparable performance between Log-Gabor filters and LBP operator, with a classification accuracy of around 82.3% and 81.7% respectively. This was achieved on low-resolution images, without the need to precisely locate facial points on each face image.
Subjects	Signal Processing
Keyword(s)	Feature selection minimum redundancy maximum dependency mutual information quotient log-Gabor filters LBP operator Facial Expression Recognition.
Copyright notice	© SQU 2009
ISSN	1813-419X

Access Statistics:	29 Abstract Views - Detailed Statistics
Created:	Fri, 12 Aug 2011, 10:33:00 EST by Catalyst Administrator