Development of continuous personal authentication method combined with eye blink characteristics and eye shape feature

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K12517
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Biomedical engineering/Biomaterial science and engineering
• Research Institution: Toyama Prefectural University
• Principal Investigator: Takano Hironobu 富山県立大学, 工学部, 准教授 (40363874)
• Co-Investigator(Kenkyū-buntansha): 中村 清実 富山県立大学, 工学部, 名誉教授 (20143860)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: まばたき / バイオメトリクス / 継続認証 / 行動的特徴 / セキュリティ

Outline of Final Research Achievements

In this study, we developed the continuous personal authentication method using eye blink. The evaluation experiment of blink discrimination indicates that the correct discrimination rate of SVM learned with own blink data is higher than that of SVM learned with blink data of other people. In addition, the suitable feature vector which provides the highest discrimination rate is different in each person. In the evaluation of personal authentication with blink data, the performance of authentication with voluntary blink data is better than that with spontaneous blink data.

Free Research Field

生体情報計測