In day to day interactions, human being possesses an ability of communication through facial emotions with others. Some study has fascinated the human computer interaction environments in recognizing facial emotions. The recent literature on facial emotion recognition specifies the requirement of two desired directions i.e. one on the image processing techniques which are highly relevant for identifying facial features under uneven lighting and the other is on interpreting the face emotion through the processed facial features. In this paper, both of these problems are under taken. In proposed work Scale invariant feature transform(SIFT) is used for feature extraction as it is invariant to rotation and scaling and partially invariant to illumination changes and is hybridized with genetic algorithm and neural network for recognizing six emotions i.e. HAPPY, SAD, ANGRY, NEUTRAL, SURPRISE and FEAR. The whole simulation has been taken place in MATLAB environment.
展开▼