EEG-based emotion recognition using empirical wavelet transform

Emotion recognition has a prominent status in the applications of brain-machine interface. An approach on recognizing Electroencephalography (EEG) emotion using empirical wavelet transform (EWT) and autoregressive (AR) model is given in this paper. The proposed method chooses two channels in a certain time segment to perform feature extraction. The EWT is first used to decompose EEG-based emotion data into several empirical modes, and then AR coefficients are calculated based on the selected empirical modes. Furthermore, these features constitute a feature vector and are then input into a classifier to perform emotion recognition. This paper implemented multiple experiments to verify the performance of our proposed approach on DEAP dataset. The best recognition rate of our approach achieves 67.3% for arousal dimension and 64.3% for valence dimension. The obtained results show that our proposed approach is superior to some exist methods.