Development of biometric recognition system using electroencephalogram (EEG) signals

• Main Author: Ericsen
• Other Authors: Vinod Achutavarrier Prasad
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering
• Online Access: http://hdl.handle.net/10356/70239
• Institution: Nanyang Technological University

As the technology advancing at phenomenal rate, security had become an integral part of life. Regardless of whether it is a personal, business security or government security, security becomes an important issue which requiring constant upgrading. Biometric recognition system had been deployed widely; however, the security risk for the biometric system had grown significantly. Hence, there is a need to find more biological traits which are more secure against potential security attack. As of recent, there had been much research on the electroencephalogram signals as a new modality that can be used to develop a more robust biometric system due to its uniqueness and its robustness against the threat. This report will present the use of EEG as a potential biometric system with the use of self and relatives’ image. In the offline experiment, the signal will be preprocessed and extracted to form a template vector. The template vector for each subject will be correlated with test EEG features. The significance of correlation determines the acceptance/rejection of a person during online authentication. P value determines how significantly correlated the two signals are. A predefined p-value is employed in the authentication procedure, chosen from a number of validation sessions. During the online experiment, the P value of self-face in theta band and relative face in the beta band will be compared with the predefined value. If both conditions are less or equal to the predefined threshold, then the person will be accepted into the system and vice versa. Based on the result from three subjects, the overall accuracy of the proposed solution is 70.83%, the average False Acceptance Rate (FAR) is 33.33%, and the average False Rejection Rate (FRR) is 25%.