A Novel De-Pccm Feature For Eeg-Based Emotion Recognition

Emotion recognition is a key work of research in Brain Computer Interactions. With the increasing concerns on affective computing, emotion recognition has attracted more and more attention in the past decades. Using electroencephalography(EEG) is a common way to distinguish emotions although it is also a challenging task. In this paper, we proposed a novel feature called DE-PCCM to improve the accuracy. The basic idea of DE-PCCM is to reveal the relationship between channels after extracting the differential entropy (DE) feature. The DE-PCCM feature can transform the DE features into 2D images so that it could be used as input of Convolutional neural network(CNN). In addition, we constructed a deep learning model for the DE-PCCM feature. Experiments are carried out on the SEED dataset, and our results demonstrate the superiority of the proposed method.