EEG-based Emotion Recognition using Transfer Learning Based Feature Extraction and Convolutional Neural Network

In this paper, a novel method for EEG(Electroencephalography) based emotion recognition is introduced. This uses transfer learning to extract features from multichannel EEG signals, these are then arranged in an 8×9 map represent their spatial location on scalp and we introduce CNN model which takes the feature extracts relations between channel finally classify emotions. First, signals converted spectrogram passed through pre-trained image classification get vector of EEG. Then, vectors different channels rearranged presented as input or dependencies part training. Finally, outputs flattened dense layer classes. study, SEED, SEED-IV SEED-V data-sets used our achieves best accuracy 97.09% 89.81% 88.23% data-set with fivefold cross validation.