*Corresponding author information: E-mail: keisuke@fjt.info.gifu-u.ac.jp, Telephone: +81-58-230-6519 Personal Identification Based on Blood Vessels of Retinal Fundus Images Keisuke Fukuta* a , Toshiaki Nakagawa a , Yoshinori Hayashi b , Yuji Hatanaka c , Takeshi Hara a , Hiroshi Fujita a a Department of Intelligent Image Information, Graduate School of Medicine, Gifu University, 1-1 Yanagido, Gifu-shi, Gifu 501-1194, Japan b TAK Co., Ltd., 4-35-12 Kono, Ogaki-shi, Gifu 503-0803, Japan c Department of Electronic Control Engineering, Gifu National College of Technology, 2236-2 Kamimakuwa, Motosu-shi, Gifu 501-0495, Japan ABSTRACT Biometric technique has been implemented instead of conventional identification methods such as password in computer, automatic teller machine (ATM), and entrance and exit management system. We propose a personal identification (PI) system using color retinal fundus images which are unique to each individual. The proposed procedure for identification is based on comparison of an input fundus image with reference fundus images in the database. In the first step, registration between the input image and the reference image is performed. The step includes translational and rotational movement. The PI is based on the measure of similarity between blood vessel images generated from the input and reference images. The similarity measure is defined as the cross-correlation coefficient calculated from the pixel values. When the similarity is greater than a predetermined threshold, the input image is identified. This means both the input and the reference images are associated to the same person. Four hundred sixty-two fundus images including forty-one same-person’s image pairs were used for the estimation of the proposed technique. The false rejection rate and the false acceptance rate were 9.9×10 -5 % and 4.3×10 -5 %, respectively. The results indicate that the proposed method has a higher performance than other biometrics except for DNA. To be used for practical application in the public, the device which can take retinal fundus images easily is needed. The proposed method is applied to not only the PI but also the system which warns about misfiling of fundus images in medical facilities. Keywords: retinal fundus image, identification, biometrics, blood vessel extraction, similarity 1. INTRODUCTION Recently, several systems such as personal computers, cellular phones, automatic teller machines (ATM), and entrance and exit management systems have begun using automatic personal identification (PI). These systems are exposed to security attacks such as password and encryption key cracking and spoofing, and it is important to enhance security procedures to protect personal information. Therefore, biometric technologies, which identify individuals based on unique physiological and behavioral characteristics, have been implemented instead of conventional identification methods such as passwords and personal identification number (PIN) codes [1]-[5]. Biometric identification compare newly entered data (input data) with stored data (reference data) in database as well as the conventional identification methods. The ideal biometric system must be easy to use, fast in calculation, non-invasive for humans, convenient to the user, and socially acceptable. However, when certain equipment or a facility requires tight security, biometric solutions must prioritize the minimization of the risk of stolen personal identity information and high recognition accuracy over the convenience of implementation. Biometrics using retinal recognition is one of the most suitable methods for such situations. A relatively bulky device is required for stably obtaining biometric information from the retina. However, in Medical Imaging 2008: Image Processing, edited by Joseph M. Reinhardt, Josien P. W. Pluim, Proc. of SPIE Vol. 6914, 69141V, (2008) 1605-7422/08/$18 · doi: 10.1117/12.769330 Proc. of SPIE Vol. 6914 69141V-1