Manuscript Number : IJSRSET151124
Efficiency Improvement in Recognition of Human Facial Expression
Authors(1) :-Uttam L. Patel
Face recognition has been very important issue in computer vision and pattern recognition over the last several decades. One difficulty in face recognition is how to handle the variations in the expression, pose and illumination when only a limited number of training samples are available. Here we used two databases, one is an Indian database which is not a standard database and second one is a JAFFE (Japanese Female Facial Expression). When we implemented facial expression recognition system using Indian database then we got accuracy of the algorithm is 68%. Then we implemented same system with JAFFE database then we got accuracy of the algorithm is about 70-71% which gives quite poor Efficiency of the system. Then we implemented facial expression recognition system with Gabor filter and principal component analysis. Here Gabor filter we have selected because of its good feature extraction property. Then the output of the Gabor filter we have used as an input for the PCA. Principal Component Analysis has a good feature of dimension reduction so we choose it for that purpose. In this system we used JAFFE database for training and testing purpose and we got good results. We got efficiency of the system is about 76-77% which higher than the previous system.
Uttam L. Patel
Human Facial Expression, JAFFE, fingerprint, face recognition, DNA, iris recognition, ICA, Principal Component Analysis, Linear Discriminant Analysis, PCA
Publication Details
Published in :
Volume 3 | Issue 5 | July-August 2017 Article Preview
Department of Computer Science, C. U. Shah Science College, Ashram Road, Ahmedabad, Gujarat, India
Date of Publication :
2017-07-30
License: This work is licensed under a Creative Commons Attribution 4.0 International License.
Page(s) :
668-678
Manuscript Number :
IJSRSET151124
Publisher : Technoscience Academy
Journal URL :
https://ijsrset.com/IJSRSET151124