A novel approach to Real Time Human Emotion Recognition based on displacement of facial feature points

Abstract: Human emotion recognition is one of the vital areas of research in the broad field of image processing and human-computer interaction. In this paper, an implementation methodology is proposed for designing a real-time emotion recognition system through facial expression detection. In implementing this system, a webcam is used to capture the frontal face of the person and the YCbCr color model is used to detect the face part in the captured image and then various facial features are extracted from the face image and based on the distances measured between the various feature points of the face, the facial expression is detected and eventually classified into six basic emotions such as happiness, surprise, anger, fear, sorrow and neutral. The classification results varies in a time function thus reducing the latency between the man and machine making the system behave in a real-time manner.

Keywords: Real-time, YCbCr color model, Facial Expression, Emotion Recognition, Human-Computer Interaction.