Abstract: In this Paper Emotion Detection System is implemented which runs the system applications like Browser, Media- Player, etc. Respectively based on the detected Emotion. The system detects seven universal emotions that are Sad, Disgust, Surprise, Angry, Fear, Happy and Neutral. This system captures real time Images using Web-cam. The captured images are further processed for emotion detection and facial expression is recognized from the captured image. HOG (Histogram Oriented Gradients) is used for feature extraction which allows us to extract facial components from the captured image. SVM is used as a classifier for the extracted features. Proposed method is experimented on JAFFE dataset and an extended on Cohn-Kanade dataset.

Keywords: Emotion Detection, HOG features, SVM, Context Management System and Real Time Image.