Summary
International Technical Conference on Circuits/Systems, Computers and Communications
2008
Session Number:G4
Session:
Number:G4-5
Personal Authentication Using New Feature Vector of Brain Wave
Sadanao Baba, Chisei Miyamoto, Isao Nakanishi,
pp.-
Publication Date:2008/7/7
Online ISSN:2188-5079
DOI:10.34385/proc.39.G4-5
PDF download (102.8KB)
Summary:
Conventional person authentication methods using the brain wave: electroencephalogram (EEG) have a problem that they require heavy computational road. In this paper, we propose a new feature vector with light computational load, which consists of maximum spectral powers, their frequency values, and accumulated spectral powers which are greater than a threshold. The verification performance is examined in authentication experiments using 23 subjects and as a result the verification performance of 72% is obtained.