Multi Channel EEG Based Biometric System with a Custom Designed Convolutional Neural Network

Bakircioğlu, Kaan; Tanimu, Musa Bindawa; Özkurt, Nalan; Seçmen, Mustafa; Güzeliş, Cüneyt; Yıldız, Osman

doi:10.1007/978-3-031-01984-5_10

Cited by 5 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, other studies also using non‐expensive devices reported lower precision ratio results, such as in Ref. [20, 62], or slightly lower precision in Ref. [24, 27] compared to the proposed method here.…”

Section: Experiments Results and Discussioncontrasting

confidence: 57%

“…Nevertheless, motor imagery as a signal acquisition strategy [21,33,34,58,63] seems to offer slightly better results than cognitive mental tasks [20,24,27,28]. Regarding classification algorithms, SVM has been the most widely used and has usually provided the best performances [20,21,23,26,28] compared to others commonly used in the literature, such as DA [33,58], NN [28,33] or Convolutional Neural Networks (CNN) [34,62,63], and Random Forest (RnF) [24,27]. However, there are less conventional techniques such as the Pearson correlation coefficient (PCC) [61] or the application of long short-term memory (LSTM) [60] that report competent results.…”

Section: Comparison Discussion With Recent Related Studiesmentioning

confidence: 99%

See 1 more Smart Citation

Brainprint based on functional connectivity and asymmetry indices of brain regions: A case study of biometric person identification with non‐expensive electroencephalogram headsets

et al. 2023

View full text Add to dashboard Cite

Brain-computer interface applications for biometric person identification have increased their interest in recent years since they are potentially more secure and more difficult to counterfeit than traditional biometric techniques. However, it is necessary to consider how brain waves are acquired for this purpose, not only in terms of efficiency but also of practical comfort for the user and the affordability degree of the biosignal acquisition device so that their everyday application can become a realistic possibility. In this context, this paper presents the capabilities of using a non-expensive wireless electroencephalogram (EEG) device to extract spectral-related and functional connectivity information of brain activity. The proposed method achieved a sufficient biometric identification with two datasets of 13 and 109 subjects when comparing the performance of a sizeable classification algorithm set. In addition, a novel feature in EEG biometric identification, called asymmetry index, is introduced here. Furthermore, this is the first study in this field to consider the effect of the time-lapse between different recording sessions on the system's behaviour when using a low-cost EEG device with identification accuracy rates of up to 100%. K E Y W O R D S biometrics brain-computer interface (BCI), classification, electroencephalogram (EEG) | INTRODUCTIONThe extraction of information from human brain electrical biosignals has been the subject of much attention in recent years. The electroencephalogram, encephalogram, or EEG is a non-invasive measure used to study the central nervous system's functioning, specifically the cerebral cortex's electrical activity [1]. EEG analysis has been widely used during the last century in medicine and as a basis for interfaces between brain and machine (BMI) [2,3]. In this context, the application of feature extraction techniques and automatic classification algorithms to the EEG signals has served as support tools in various utilities such as diagnosing neurological disorder diseases [4,5], since BMIs provide their users with an alternativeThis is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

show abstract

Section: Experiments Results and Discussioncontrasting

confidence: 57%

Section: Comparison Discussion With Recent Related Studiesmentioning

confidence: 99%

Brainprint based on functional connectivity and asymmetry indices of brain regions: A case study of biometric person identification with non‐expensive electroencephalogram headsets

et al. 2023

View full text Add to dashboard Cite

show abstract

“…This result demonstrates that high accuracy does not entail using all available EEG channels. The authors in Reference 20 designed a customized CNN model to identify multi‐channel raw EEG signals along with sliding window and cross‐validation approach increasing the generalization ability of the network. Their proposed method showed that the performance and training time vary with the number of specific channels used to classify the signals.…”

Section: Related Workmentioning

confidence: 99%

EEG‐based biometric authentication system using convolutional neural network for military applications

Vadher,

Patel,

Nair

et al. 2023

Security and Privacy

View full text Add to dashboard Cite

In this technological era, as the need for security arises, the use of biometrics is increasing in authentication systems as a secure and convenient method of human identification and verification. Electroencephalogram (EEG) signals have gained significant attention among the various biometric modalities available because of their unique and unforgeable characteristics. In this study, we have proposed an EEG‐based multi‐subject and multi‐task biometric authentication system for the military applications that address the challenges associated with multi‐task variation in EEG signals. The proposed work considers the use of respective EEG signals for the access of artillery, entrance to highly confidential places for the military and so forth by authenticated personnel only. We have used a multi‐subject, multi‐session, and multi‐task () dataset. The dataset was partially preprocessed with basic signal processing techniques such as bad channel repairing, independent component analysis for artifact removal, downsampling to 250 Hz, and an audio filter of 0.01–200 Hz for signal improvisation. This partially preprocessed dataset was further processed and was used in our deep learning model (DL) architectures. For EEG‐based biometric authentication, convolutional neural network (CNN) outperforms many of the state‐of‐the‐art DL architectures with a validation accuracy of approximately 99.86%, training accuracy of 98.49% and precision, recall and F1‐score with values of 99.91% that makes this EEG‐based approach for authentication more reliable. The DL models were also compared based on training and inference time, where CNN used the most training time but took the least time to predict the output. We compared the performance of the CNN model for three preprocessing techniques by feeding mel spectrograms, chromagrams and mel frequency cepstral coefficients, out of which mel spectrograms provided better results. This proposed architecture proves to be robust and efficient for military applications.

show abstract

A Short-Length Single Channel EEG Based Personal Identification System

Hendrawan,

Ratsanjani,

Noprianto

et al. 2023

2023 IEEE International Conference on Communication, Networks and Satellite (COMNETSAT)

View full text Add to dashboard Cite

Multi Channel EEG Based Biometric System with a Custom Designed Convolutional Neural Network

Cited by 5 publications

References 20 publications

Brainprint based on functional connectivity and asymmetry indices of brain regions: A case study of biometric person identification with non‐expensive electroencephalogram headsets

Brainprint based on functional connectivity and asymmetry indices of brain regions: A case study of biometric person identification with non‐expensive electroencephalogram headsets

EEG‐based biometric authentication system using convolutional neural network for military applications

A Short-Length Single Channel EEG Based Personal Identification System

Contact Info

Product

Resources

About