Emotion Recognition using EEG based Topographic Images

Abstract

Emotional recognition play a vital role in developing affective computing applications. Brain electrical activity bears the emotional cues needed for emotion detection, but very modestresearch has been done to extract those cues. Most of the work are either classification of emotion on arousal-valance scale or recognizing basic emotions but no work has been done so far for predicting the actual response of user which is self-assessment feedback after each stimuli presentation. Only user knows what he felt after stimuli presentation, so self-assessment feedback is the actual state of mind of user. In this thesis we proposed a method to predict the response of user after stimuli presentation. For this purpose we used event related topographic images extracted from EEG. We extracted pattern related features from this topographs and this features are used by artificial neural network for emotion recognition.