FACIAL EMOTION RECOGNITION USING GABOR (HAPPY, SAD, ANGER, NEUTRAL MOOD)

The best way to communicate the emotions and intentions is facial expressions. This is powerful because as per the psychological research 55% of the total communicated message is the human facial expression. Therefore deriving an effective facial representation from the original face image is a vital step and very tough task in the field of computer science for successful facial expression recognition. Human facial recognition usually uses image processing, gesture signal proces sing and physiological signal processing. Most use of this system is in the area of security, psychological studies, and social interaction [Ekman et al., 1997] [5]. Facial expression reflects not only emotions but other mental activities, social interacti on and physiological signals. Gabor filters are used to extract the features of facial expressions. GF represents the behavior of receptive fields in human visual systems (HVS) very effectively even in case of slight object rotation, distortion and variati on in illumination. Here we have considered face database in which the different expressions of facial images are stored. Different facial expressions will be recognized as neutral, disgust, happy, sad, and anger. In this first we extracted the features of face by using Gabor filter and then applied SVM to classify into different mood. Finally mood is recognized and verified by using confusion matrix.