Affective EEG-Based Person Identification Using Channel Attention Convolutional Neural Dense Connection Network

Abstract

In the biometric recognition mode, the use of electroencephalogram (EEG) for biometric recognition has many advantages such as anticounterfeiting and nonsteal ability. Compared with traditional biometrics, EEG biometric recognition is safer and more concealed. Generally, EEG-based biometric recognition is to perform person identification (PI) through EEG signals collected by performing motor imagination and visual evoked tasks. The aim of this paper is to improve the performance of different affective EEG-based PI using a channel attention mechanism of convolutional neural dense connection network (CADCNN net) approach. Channel attention mechanism (CA) is used to handle the channel information from the EEG, while convolutional neural dense connection network (DCNN net) extracts the unique biological characteristics information for PI. The proposed method is evaluated on the state-of-the-art affective data set HEADIT. The results indicate that CADCNN net can perform PI from different affective states and reach up to 95%-96% mean correct recognition rate. This significantly outperformed a random forest (RF) and multilayer perceptron (MLP). We compared our method with the state-of-the-art EEG classifiers and models of EEG biometrics. The results show that the further extraction of the feature matrix is more robust than the direct use of the feature matrix. Moreover, the CADCNN net can effectively and efficiently capture discriminative traits, thus generalizing better over diverse human states.

Electroencephalographic (EEG) based Deep Learning (DL): A Comparative Review

“…In [9] datasets from four different experiments measuring endogenous brain functions (driving fatigue and emotion) in addition to time-locked artificially created brain responses from 157 subjects, [5] datasets including emotion and combined data. In [31] [35] (DREAMER, HEADIT) datasets performed emotion recognition tasks were used in them experiments. In [29] a dataset created to evaluate EEG-based biometric methods, named (BED) created to test the effectiveness of consumer-grade hardware.…”

“…After then, respondents were instructed to imagine making left and right-handed movements every 25 times in time with the notification that appeared on the screen for 4 seconds. In [35], combine two protocols VEP and MI. Tasks Evoked (TE) by external stimulation, in ref.…”

“…These models can learn to identify patterns and relationships in the data and generate predictions or classifications based on these features. In ISSN: 2710-2165 http://doi.org/10.58564/IJSER.1.2.2023.70 https://ijser.aliraqia.edu.iq [1], [4], [6], [8], [14], [18], [20], [22], [25], [28], [32], [35][36][37][38][39] used the (CNNs) models or changing in the structure of the (CNN) [8], to extract features, while [10], [13] apply a model called Siamese CNNs, another researchers use the graph convolutional neural network (GCNN) to detain deep essential structural representations from EEG graphs immediately [5], and in [19] propose an adversarial inference learning with CNN to extend DL models. In [24], [33], used RNN and Long Short-Term Memory (LSTM) which is a special form of RNN architecture with FaceNet to create a new LSTM-RNN model that aided in the extraction of the feature vector for a given 100s EEG signal which leads with a support vector machine (SVM) used as a clustering tool.…”

“…A DL model integrating functional connectivity was proposed for biometric identification in a recent work [16]. In recent studies CNNs have been proposed to EEG-based person identification and authentication, achieving promising results [2], [8][9][10][11][12][13][14][15][16][17][18][19][20], [22], [25], [28], [29], [31], [32] and [34][35][36][37][38][39]. These studies use CNNs model directly on the EEG amplitude fluctuations.…”

“…There were more than (35) identified studies. According to the kind of different tasks, input data type, EEG preprocessing techniques, and DL architecture, those studies were analyzed, to provide the main results.…”