EEG-based Person Identification Research Articles

Fusing the spatial structure of electroencephalogram channels can increase the individualization of the functional connectivity network.

An electroencephalogram (EEG) functional connectivity (FC) network is individualized and plays a significant role in EEG-based person identification. Traditional FC networks are constructed by statistical dependence and correlation between EEG channels, without considering the spatial relationships between the channels. The individual identification algorithm based on traditional FC networks is sensitive to the integrity of channels and crucially relies on signal preprocessing; therefore, finding a new presentation for FC networks may help increase the performance of the identification algorithms. EEG signals are smooth across space owing to the volume conduction effect. Considering such spatial relationships among channels can provide a more accurate representation of FC networks. In this study, we propose an EEG FC network with virtual nodes that combines the spatial relationships and functional connectivity of channels. The comparison results for individual identification show that the novel EEG network is more individualized and achieves an accuracy of 98.64% for data without preprocessing. Furthermore, our algorithm is more robust in reducing the number of channels and can perform well even when a large area of channels is removed.

AITST—Affective EEG-based person identification via interrelated temporal–spatial transformer

Compared with other biometrics, such as fingerprints and irises, EEG signals contain richer identity information due to their high subject dependence. In this study, we proposed an affective EEG-based person identification by using an affective interrelated temporal–spatial transformer (AITST) to overcome the influence of different emotional states in affective EEG-based person identification. The AITST contains an interrelated temporal–spatial attention that can preserve the interrelationships of temporal and spatial domains while extracting temporal and spatial features of the EEG signal. We evaluated the proposed AITST on the public emotional database DEAP. The experimental results show that our AITST can reach an average recognition accuracy of 98.94±0.04% in the low valence, low arousal state, 99.06±0.02% in low valence, high arousal state, 98.97±0.02% in high valence, low arousal state, and 99.21±0.03% in the high valence, high arousal state, significantly outperforming the other compared models and overcoming the influence of different emotional states. Furthermore, we explored the contribution of different frequency bands, different emotional states and different features of EEG for EEG-based person identification. The results showed that gamma band of EEG, power spectrum density (PSD), and high arousal state have the greatest contribution to affective EEG-based person identification. Source code is available at https://github.com/JerryKingQAQ/AITST.

EEG-Based Person Identification during Escalating Cognitive Load.

With the development of human society, there is an increasing importance for reliable person identification and authentication to protect a person’s material and intellectual property. Person identification based on brain signals has captured substantial attention in recent years. These signals are characterized by original patterns for a specific person and are capable of providing security and privacy of an individual in biometric identification. This study presents a biometric identification method based on a novel paradigm with accrual cognitive brain load from relaxing with eyes closed to the end of a serious game, which includes three levels with increasing difficulty. The used database contains EEG data from 21 different subjects. Specific patterns of EEG signals are recognized in the time domain and classified using a 1D Convolutional Neural Network proposed in the MATLAB environment. The ability of person identification based on individual tasks corresponding to a given degree of load and their fusion are examined by 5-fold cross-validation. Final accuracies of more than 99% and 98% were achieved for individual tasks and task fusion, respectively. The reduction of EEG channels is also investigated. The results imply that this approach is suitable to real applications.

A Novel Tensorial Scheme for EEG-Based Person Identification

Biometrics have attracted growing research interests as the information security and safety gain increasing attention to date. As a kind of important biomedical signal, Electroen-cephalogram (EEG) contains valuable information about identity, emotionality, personality, etc. Thus, automatically distinguishing the identities based on EEG is beneficial to the development of biometrics, forensics and informatics. Although deep learning has absorbed much research attention for the issue of EEG-based person identification, the performance enhancement of this methodology seems to have hit a bottleneck recently. Hence, by rethinking the problems haunting this issue, we plan to reinvigorate the conventional method pipeline, and put forward a novel and effective tensorial scheme away from the deep learning mainstream. Specifically, the proposed tensorial scheme extracts the effective tensorial representation from multi-channel EEG at first; then, the scheme performs the designed tensorial learning to improve the discriminability of the feature space; finally, the scheme carries out the devised tensorial measurement in the learned metric space for classification. Experimental results have demonstrated the superiority of proposed scheme over the related advanced approaches by means of the challenging benchmark databases DEAP, SEED and DREAMER.

CNN-Based Personal Identification System Using Resting State Electroencephalography.

As a biometric characteristic, electroencephalography (EEG) signals have the advantages of being hard to steal and easy to detect liveness, which attract researchers to study EEG-based personal identification technique. Among different EEG protocols, resting state signals are the most practical option since it is more convenient to operate than the other protocols. In this paper, a personal identification system based on resting state EEG is proposed, in which data augmentation and convolutional neural network are combined. The cross-validation is performed on a public database of 109 subjects. The experimental results show that when only 14 EEG channels and 0.5 seconds data are employed, the average accuracy and average equal error rate of the system can reach 99.32% and 0.18%, respectively. Compared with some existing representative works, the proposed system has the advantages of short acquisition time, low computational complexity, and rapid deployment using market available low-cost EEG sensors, which further advances the implementation of practical EEG-based identification systems.

Towards online applications of EEG biometrics using visual evoked potentials

Electroencephalogram (EEG)-based biometrics have attracted increasing attention in recent years. A few studies have used visual evoked potentials (VEPs) in EEG biometrics due to their high signal-to-noise ratio (SNR) and good stability. However, a systematic comparison of different types of VEPs is still lacking. Therefore, this study proposes a system framework for VEP-based biometrics. We quantitatively compared the performance of three types of VEP signals in person identification. Flash VEPs (f-VEPs), steady-state VEPs (ss-VEPs), and code-modulated VEPs (c-VEPs) measured from a group of 21 subjects on two different days were used to estimate the correct recognition rate (CRR). We adopted a template-matching-based identification algorithm that was developed for VEP detection in brain-computer interfaces (BCIs) for person identification. Furthermore, this study demonstrates an online person identification system using c-VEPs with a group of 15 subjects. Among the three methods, c-VEPs achieved the highest CRRs of 100% using 3.15-s VEP data (a 5.25-s duration including 2.1-s intervals) in the intra-session condition and 99.48% using 10.5-s VEP data (a 17.5-s duration including 7-s intervals) in the cross-session condition. The online system achieved a cross-session CRR of 98.93% using 10.5-s VEP data (a 14-s duration including 3.5-s intervals). A systematic comparison of the performance of the three types of VEP signals in EEG-based person identification revealed that the c-VEP paradigm achieved the highest CRRs. The online system further demonstrated high performance in practical applications. The proposed VEP-based biometric system obtained promising identification performance, showing great potential for online person identification applications in real life.

Biometric recognition system performance measures for lossy compression on EEG signals

Abstract Electroencephalogram (EEG) plays an essential role in analysing and recognizing brain-related diseases. EEG has been increasingly used as a new type of biometrics in person identification and verification systems. These EEG-based systems are important components in applications for both police and civilian works, and both areas process a huge amount of EEG data. Storing and transmitting these huge amounts of data are significant challenges for data compression techniques. Lossy compression is used for EEG data as it provides a higher compression ratio (CR) than lossless compression techniques. However, lossy compression can negatively influence the performance of EEG-based person identification and verification systems via the loss of information in the reconstructed data. To address this, we propose introducing performance measures as additional features in evaluating lossy compression techniques for EEG data. Our research explores if a common value of CR exists for different systems using datasets with lossy compression that could provide almost the same system performance with those using datasets without lossy compression. We performed experiments on EEG-based person identification and verification systems using two large EEG datasets, CHB MIT Scalp and Alcoholism, to investigate the relationship between standard lossy compression measures and our proposed system performance measures with the two lossy compression techniques, discrete wavelet transform—adaptive arithmetic coding and discrete wavelet transform—set partitioning in hierarchical trees. Our experimental results showed a common value of CR exists for different systems, specifically, 70 for person identification systems and 50 for person verification systems.

EEG-based personal identification method using unsupervised feature extraction and its robustness against intra-subject variability

Objective. Brain activity signals are possible biomarkers for personal authentication. However, they are inherently variable due to measurement-environment factors and subject-dependent factors; electroencephalography (EEG) signals could be different in days even for the same task, subject, and experimental settings. This variability could cause loss of consistency of the signals across multiple measurements of a single subject, and hence decrease the performance of EEG-based personal identification. In this study, we evaluated the influence of the variability on personal EEG features by using our original EEG dataset. Approach. We collected EEG signals in twenty subjects across four rounds (morning and afternoon daily for two days). At each round, we reinstalled an EEG cap on the subjects’ scalps. To extract personal EEG features that were invariant across the sessions, we proposed unsupervised learning methods; common dictionary learning and t-distributed stochastic neighbor embedding. To assess the performance of personal identification, we compared two different experimental settings; test data recorded in the same round as the training data (Setting SR) and test data recorded in different rounds (Setting DR). Main results. The performance in SR was better than that in DR, suggesting that features dependent on the rounds were dominant. However, the 40% accuracy rate in DR, which is significantly higher than the chance level, suggests that our proposed method robustly extracted the personal features against the variability, in most cases. Furthermore, we also evaluated the performance of a problem, which involved detecting individuals who were not registered in the authentication system. In this problem, we obtained a similar result that the variability for the rounds influenced the performance. However, we obtained a good performance in the detection of some unknown subjects even in DR. Significance. We found the variability in EEG data actually affected the personal features that were used for personal identification. Even considering the variability in EEG data, however, we found our proposed method is applicable in personal authentication scenarios, i.e. personal identification and unknown detection.

Convolutional Neural Networks Using Dynamic Functional Connectivity for EEG-Based Person Identification in Diverse Human States

Highly secure access control requires Swiss-cheese-type multi-layer security protocols. The use of electroencephalogram (EEG) to provide cognitive indicators for human workload and fatigue has created environments where the EEG data are well-integrated into systems, making it readily available for more forms of innovative uses including biometrics. However, most of the existing studies on EEG biometrics rely on resting state signals or require specific and repetitive sensory stimulation, limiting their uses in naturalistic settings. Moreover, the limited discriminatory power of uni-variate measures denies an opportunity to use dependences information inherent in brain regions to design more robust biometric identifiers. In this paper, we proposed a novel model for ongoing EEG biometric identification using EEG collected during a diverse set of tasks. The novelty lies in representing EEG signals as a graph based on within-frequency and cross-frequency functional connectivity estimates, and the use of graph convolutional neural network (GCNN) to automatically capture deep intrinsic structural representations from the EEG graphs for person identification. An extensive investigation was carried out to assess the robustness of the method against diverse human states, including resting states under eye-open and eye-closed conditions and active states drawn during the performance of four different tasks. We compared our method with the state-of-the-art EEG features, classifiers, and models of EEG biometrics. Results show that the representation drawn from EEG functional connectivity graphs demonstrates more robust biometric traits than direct use of uni-variate features. Moreover, the GCNN can effectively and efficiently capture discriminative traits, thus generalizing better over diverse human states.

A spatio-temporal model for EEG-based person identification

Biometric representation of humans often requires the tasks such as identification and verification. This can be done using various modalities such as fingerprint, face, retina, voice, etc. However, existing biometric systems are vulnerable to various security attacks. Recently, Electroencephalography (EEG) is considered to be one of the alternatives to develop a robust biometric system. Brain activities represented by EEG signals are more sensitive, secure, and difficult to copy and steal. In this paper, we propose a spatio-temporal dense architecture for EEG-based person identification. Firstly, raw EEG are processed to extract robust and informative spatial features using Convolutional Neural Networks (CNN) as they are known for automatic feature extraction from the raw data. Then, a Long short-term memory (LSTM) network is utilized to process temporal data and person identification is carried out. The experiment has been carried out on a publicly available dataset consisting of EEG of 109 subjects. The architecture is tested on two baseline situations, i.e., eye closed (EC) and the eye opened (EO). Person identification rates of 99.95% and 98% have been recorded for EC and EO states using the proposed scheme. Experimental results demonstrate the robustness of the proposed scheme in terms of person identification and outstrip existing works.

Robustness of functional connectivity metrics for EEG-based personal identification over task-induced intra-class and inter-class variations

Growing interest is devoted to understanding in which situations and with what accuracy brain signals recorded from scalp electroencephalography (EEG) may represent unique fingerprints of individual neural activity. In this context, the present paper aims to investigate the impact of some of the most commonly used metrics to estimate functional connectivity on the ability to unveil personal distinctive patterns of inter-channel interactions. Different metrics were compared in terms of equal error rate. It is widely accepted that each connectivity metric carries specific information in respect to the underlying interactions. Experimental results on publicly available EEG recordings show that different connectivity metrics define peculiar subjective profile of connectivity and show different mechanisms to detect subject-specific patterns of inter-channel interactions. Moreover, these findings highlight that some measures are more accurate and more robust than others, regardless of the task performed by the user. Finally, it is important to consider that frequency content and spurious connectivity may still play a relevant role in determining subject-specific characteristics.

Feature Extraction with GMDH-Type Neural Networks for EEG-Based Person Identification

The brain activity observed on EEG electrodes is influenced by volume conduction and functional connectivity of a person performing a task. When the task is a biometric test the EEG signals represent the unique "brain print", which is defined by the functional connectivity that is represented by the interactions between electrodes, whilst the conduction components cause trivial correlations. Orthogonalization using autoregressive modeling minimizes the conduction components, and then the residuals are related to features correlated with the functional connectivity. However, the orthogonalization can be unreliable for high-dimensional EEG data. We have found that the dimensionality can be significantly reduced if the baselines required for estimating the residuals can be modeled by using relevant electrodes. In our approach, the required models are learnt by a Group Method of Data Handling (GMDH) algorithm which we have made capable of discovering reliable models from multidimensional EEG data. In our experiments on the EEG-MMI benchmark data which include 109 participants, the proposed method has correctly identified all the subjects and provided a statistically significant ([Formula: see text]) improvement of the identification accuracy. The experiments have shown that the proposed GMDH method can learn new features from multi-electrode EEG data, which are capable to improve the accuracy of biometric identification.

EEG-based person identification through Binary Flower Pollination Algorithm

Electroencephalogram (EEG) signal presents a great potential for highly secure biometric systems due to its characteristics of universality, uniqueness, and natural robustness to spoofing attacks. EEG signals are measured by sensors placed in various positions of a person’s head (channels). In this work, we address the problem of reducing the number of required sensors while maintaining a comparable performance. We evaluated a binary version of the Flower Pollination Algorithm under different transfer functions to select the best subset of channels that maximizes the accuracy, which is measured by means of the Optimum-Path Forest classifier. The experimental results show the proposed approach can make use of less than a half of the number of sensors while maintaining recognition rates up to 87%, which is crucial towards the effective use of EEG in biometric applications.