Study on an multi-modal personal authentication using visual information and behavior characteristics

2016 Fiscal Year Final Research Report

• Project/Area Number: 25330226
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Human interface and interaction
• Research Institution: Utsunomiya University
• Principal Investigator: Hasegawa Madoka 宇都宮大学, 工学(系)研究科(研究院), 教授 (80322014)
• Co-Investigator(Renkei-kenkyūsha): SHINODA Kazuma 宇都宮大学, 大学院工学研究科, 助教 (50639200)
• Project Period (FY): 2013-04-01 – 2017-03-31
• Keywords: 画像選択型認証 / ユーザブルセキュリティ / ハイブリッド画像 / 加速度センサ

Outline of Final Research Achievements

We studied on a usable authentication method for handy mobile terminals. Our system consists of a graphical password using hybrid images and a personal authentication system based on arm swing operation.
For the graphical password, we proposed an image synthesis method based on the wavelet transform. We also proposed an objective evaluation score considering the visibility of the foreground image in the hybrid image for both the legitimate user and the shoulder-surfer. Experimental results show that our score can choose suitable hybrid images for user authentication.
For the authentication system based on arm swing operation, we developed a method to calculate the difference between the pre-registered motion data and the user's current motion data by using the Euclidean distance and the error angle. This method is effective to reduce the False Rejection Rate. Simulation results show that this method can reduce Equal Error Rate.

Free Research Field

画像処理