Abstract

Abstract- Many applications are available on mobile phone, for providing services to the user.so for providing the emotional care to the user, it is required to develop real time application which recognize the emotion of user on mobile phone. So the emotion recognition algorithms implemented for recognizing the emotion of user, should take smaller processing power, and less time requirement. In the proposed system, the image of the user is captured using the camera of android phone. The phone has inbuilt capacity of face detection. On the detected face area bandlet transform is applied. After applying bandlet transform the face area is divided into non overlapping blocks. Then Local Binary Pattern histograms are calculated for each block and concatenated and it will give feature set for image. Kruskal-Wallis feature selection is applied to select the subset of a features from feature set. Then Convolutional Neural Network based classifier is used to classify the emotion.