EEG signal processing and emotion recognition using Convolutional Neural Network

As an important task in the advanced stage of artificial intelligence, the research of emotional EEG has received more and more attention in recent years. In order to improve the accuracy of EEG signal emotion recognition, in this paper, Fast Fourier Transform (FFT) and Continuous Wavelet Transform (CWT) are used to extract the features of EEG signals on the DEAP data set and build two CNN models for emotion recognition. The results show that the proposed algorithm is effective for EEG signal emotion recognition. The average recognition accuracy of emotion valence can reach 75.9%; the arousal can reach 79.3%; the like/dislike can reach 80.7%. This research can provide practical application reference for continuous dimension emotion automatic analysis and machine recognition.