International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 64 - Number 18 |
Year of Publication: 2013 |
Authors: Nazil Perveen, Darshan Kumar, Ishan Bhardwaj |
10.5120/10733-5573 |
Nazil Perveen, Darshan Kumar, Ishan Bhardwaj . Facial Expression Recognition by Statistical, Spatial Features and using Decision Tree. International Journal of Computer Applications. 64, 18 ( February 2013), 15-21. DOI=10.5120/10733-5573
Facial Expression reflects the emotional stage of the person. Sensing and responding appropriately to the user's emotional state is one of the most powerful, natural and abrupt means, which have the capability to enrich man-machine interaction and to regulate inter-personal behavior. In this paper we apply a novel technique to recognize different expression effectively using classification and regression trees (CART). Firstly, we compute spatial features of the face which provide 73. 66% correct classification rate. Secondly, we compute statistical features of the face which provide 79. 4%. In last we merge both features in order to increase accuracy and classification rate increases to 83. 4%. The proposed technique is tested using JAFFE database and implemented in MATLAB environment 7. 0.