Face expression recognition using Scaled - conjugate gradient Back - Propagation algorithm

Since decades, face recognition has become an active area of research in computer vision and psychology. The rapid developments of face recognition are being fueled by numerous advances in computer vision. An ongoing challenge in this field is to design an effective human-computer interaction (HCII). In this paper we will study the latest work done that has been done in the field of facial expression recognition and analysis. In our work we have recognized six different expressions using Cohn-kanade database and system is trained using scaled conjugate gradient back-propagation algorithm. In proposed methodology we have used MATLAB’s computer vision toolbox for face detection & cropping the images and neural network toolbox. In our work we have achieved 100% training accuracy and 87.2% overall testing accuracy of six different expressions.