Emotion recognition from EEG signals allows the direct assessment of the “inner” state of a user, which is considered an important factor in human-machine-interaction. Many methods for feature extraction have been studied and the selection of both appropriate features and electrode locations is usually based on neuro-scientific findings. Their suitability for emotion recognition, however, has been tested using a small amount of distinct feature sets and on different, usually small data sets. A major limitation is that no systematic comparison of features exists. Therefore, we review feature extraction methods for emotion recognition from EEG based on 33 studies. An experiment is conducted comparing these features using machine learning techniques for feature selection on a self recorded data set. Results are presented with respect to performance of different feature selection methods, usage of selected feature types, and selection of electrode locations. Features selected by multivariate methods slightly outperform univariate methods. Advanced feature extraction techniques are found to have advantages over commonly used spectral power bands. Results also suggest preference to locations over parietal and centro-parietal lobes.

Feature Extraction and Selection for Emotion Recognition from EEG

In this paper, a multichannel EEG emotion recognition method based on a novel dynamical graph convolutional neural networks (DGCNN) is proposed. The basic idea of the proposed EEG emotion recognition method is to use a graph to model the multichannel EEG features and then perform EEG emotion classification based on this model. Different from the traditional graph convolutional neural networks (GCNN) methods, the proposed DGCNN method can dynamically learn the intrinsic relationship between different electroencephalogram (EEG) channels, represented by an adjacency matrix, via training a neural network so as to benefit for more discriminative EEG feature extraction. Then, the learned adjacency matrix is used to learn more discriminative features for improving the EEG emotion recognition. We conduct extensive experiments on the SJTU emotion EEG dataset (SEED) and DREAMER dataset. The experimental results demonstrate that the proposed method achieves better recognition performance than the state-of-the-art methods, in which the average recognition accuracy of 90.4 percent is achieved for subject dependent experiment while 79.95 percent for subject independent cross-validation one on the SEED database, and the average accuracies of 86.23, 84.54 and 85.02 percent are respectively obtained for valence, arousal and dominance classifications on the DREAMER database.

EEG Emotion Recognition Using Dynamical Graph Convolutional Neural Networks

Our project focused on recognizing emotion from human brain activity, measured by EEG signals. We have proposed a system to analyze EEG signals and classify them into 5 classes on two emotional dimensions, valence and arousal. This system was designed using prior knowledge from other research, and is meant to assess the quality of emotion recognition using EEG signals in practice. In order to perform this assessment, we have gathered a dataset with EEG signals. This was done by measuring EEG signals from people that were emotionally stimulated by pictures. This method enabled us to teach our system the relationship between the characteristics of the brain activity and the emotion. We found that the EEG signals contained enough information to separate five different classes on both the valence and arousal dimension. However, using a 3-fold cross validation method for training and testing, we reached classification rates of 32% for recognizing the valence dimension from EEG signals and 37% for the arousal dimension. Much better classification rates were achieved when using only the extreme values on both dimensions, the rates were 71% and 81%.

https://sce.uhcl.edu/boetticher/CSCI5931%20Computer%20Human%20Interaction/Emotion%20Recognition%20using%20Brain%20Activity.pdf

Emotion recognition using brain activity

In this paper, a novel group sparse canonical correlation analysis (GSCCA) method is proposed for simultaneous electroencephalogram (EEG) channel selection and emotion recognition. GSCCA is a group sparse extension of the conventional CCA method to model the linear correlationship between emotional EEG class label vectors and the corresponding EEG feature vectors. In contrast to conventional CCA method or previous GSCCA methods, a major advantage of our GSCCA method is the ability of handling the group feature selection problem from raw EEG features, which makes it very suitable for simultaneously coping with both EEG emotion recognition and automatic channel selection issues where each EEG channel is associated with a group of raw EEG features. To deal with EEG emotion recognition problem, we adopt the popularly used frequency feature to describe the EEG signal by dividing the full EEG frequency band into five parts, i.e., $\boldsymbol {\delta }$ , $\boldsymbol {\theta }$ , $\boldsymbol {\alpha }$ , $\boldsymbol {\beta }$ , and $\boldsymbol {\gamma }$ frequency bands, and then extract the frequency band features from each band for GSCCA model learning and emotion recognition. Finally, we conduct extensive experiments on EEG-based emotion recognition based on the SJTU emotion EEG dataset and experimental results demonstrate that the proposed GSCCA method would outperform the state-of-the-art EEG-based emotion recognition approaches.

Multichannel EEG-Based Emotion Recognition via Group Sparse Canonical Correlation Analysis

EEG emotion recognition using fusion model of graph convolutional neural networks and LSTM

The state of mind of a person is supported by the brain activity, and hence features of the state of mind appear in the scalp potentials, as seen on an electroencephalogram (EEG). The EEG features have been extracted into a set of 135 state variables of cross-correlation coefficients of EEGs collected with ten scalp electrodes in the θ, α, and β frequency bands corresponding toanger, sadness, joy, andrelaxation. An emotion matrix is defined which transforms the set of 135 state variables into a four-element emotion vector of which the components are indexes corresponding to the four elementary emotional states. The maximum time resolution of the emotion analysis is 0.64s and it is done in real time. This new technique has a wide variety of applications in both medical and non-medical areas, and the technology suggests the possibility of direct control of systems by the human emotional state.

Hasnine A. Haque

Papers

Feature extraction from EEGs associated with emotions