Electroencephalography and Physiological Signals for Emotion Analysis

A novel method for Electroencephalography (EEG) based emotion analysis using Gray Level Co-occurrence Matrix1 (GLCM) features contrast, correlation, energy, and homogeneity has been discussed with peripheral physiological signals. Emotions are classified using Linear Discriminant Analysis (LDA) and obtained an accuracy of 93.8. The proposed novel method discussed the effect of distances, and direction on GLCM features for different emotions. This paper concluded that GLCM features are an effective measure to discriminate the emotions and give significant knowledge for each emotion. The proposed novel methodology can be used as a tool for emotion analysis and it can also be useful for observing brain lobe variation globally.