International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 59 - Number 3 |
Year of Publication: 2012 |
Authors: Zaenal Abidin, Agus Harjoko |
10.5120/9531-3956 |
Zaenal Abidin, Agus Harjoko . A Neural Network based Facial Expression Recognition using Fisherface. International Journal of Computer Applications. 59, 3 ( December 2012), 30-34. DOI=10.5120/9531-3956
Facial expression plays significant role for human beings to communicate their emotions. Automatic facial expression analysis is a flourishing area of research in computer science, and it is also still a challenge. This paper discusses the application of a natural network based facial expression recognition using fisherface. Back propagation neural network is used as a classifier for classifying the expressions. For face portion segmentation and localization, integral projection method is used. The accuracy of system performance have evaluated on a public database "Japanese Female Facial Expression (JAFFE)". The experimental results show the effectiveness of our scheme. The best average recognition rate achieves 89. 20%.