Emotion Recognition from EEG Using RASM and LSTM

In the field of human-computer interaction, automatic emotion recognition is an important and challenging task. As a physiological signal that directly reflects the brain activity, EEG has advantages in emotion recognition. However, previous studies seldom consider together the temporal, spatial, and frequency characteristics of EEG signals, and the reported emotion recognition accuracy is not adequate for applications. To address this issue, this study proposes a new approach which extracts RASM as the feature to describe the frequency-space domain characteristics of EEG signals and constructs a LSTM network as the classifier to explore the temporal correlations of EEG signals. It is implemented on the DEAP dataset for a trial-level emotion recognition task. In a comparison with a number of relevant studies on DEAP, its mean accuracy of 76.67% ranks the first, which approves the effectiveness of this new approach.

[1]  Mohammad Soleymani,et al.  Analysis of EEG Signals and Facial Expressions for Continuous Emotion Detection , 2016, IEEE Transactions on Affective Computing.

[2]  Yuan-Pin Lin,et al.  EEG-Based Emotion Recognition in Music Listening , 2010, IEEE Transactions on Biomedical Engineering.

[3]  P. Lang.  The emotion probe. Studies of motivation and attention. , 1995, The American psychologist.

[4]  Bao-Liang Lu,et al.  Identifying Stable Patterns over Time for Emotion Recognition from EEG , 2016, IEEE Transactions on Affective Computing.

[5]  Bin Hu,et al.  Feature-level fusion of multimodal physiological signals for emotion recognition , 2015, 2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[6]  Thierry Pun,et al.  DEAP: A Database for Emotion Analysis ;Using Physiological Signals , 2012, IEEE Transactions on Affective Computing.

[7]  Uma Shanker Tiwary,et al.  Affect representation and recognition in 3D continuous valence–arousal–dominance space , 2016, Multimedia Tools and Applications.

[8]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[9]  Varvara Kollia.  Personalization Effect on Emotion Recognition from Physiological Data: An Investigation of Performance on Different Setups and Classifiers , 2016, ArXiv.

[10]  Xiang Li,et al.  Emotion recognition from multi-channel EEG data through Convolutional Recurrent Neural Network , 2016, 2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).