Fusion of feature sets for facial expression recognition

Emotion recognition has been an important research topic in the area of human computer interaction (HCI) for different application, in the last decade for instance proper emotion recognition has a wide range of applications in security, entertainment, and training. Emotion is expressed via facial muscle movements, speech, body and hand gestures, and various biological signals such heart rate. This paper focuses on facial expression to identify two universal human emotions: happiness and sadness. This is carried out by trying to extract facial feature (geometric & texture). We use spatial information for extracting geometric features and texture features by Gabor filter respectively. We classify emotions using Support Vector Machine (SVM) algorithms. After classification of emotions we use support vector regression (SVR) for intensity estimation of facial expression. The use of a public database "cohn-kanade" is conducted. The experimental results demonstrate that the proposed approach is an effective method to recognize emotions through facial expression with an emotion recognition rate more than 95% that demonstrate the efficiency and validity of the method.