Finger Vein Authentication System

Finger vein authentication system (FVAS) is more secure than other forms of authentication, such as signatures and fingerprint authentication. This is because of the fact that veins are present beneath the human skin, which makes them nearly impossible to replicate. Furthermore, the vein pattern remains persistent during the span of a person's life, so it eliminates the risk of alterations in contrast to fingerprint and iris biometrics. In fingerprints, the ridges begin wearing out and start getting spaced out as a person ages. By using Iris authentication, if a person gets a cataract surgery, the iris pattern is altered, which causes a significant problem for security systems. In this paper, the authors plan to develop an authentication system using a finger vein spattern as a unique biometric. An image of the target finger is captured with the help of a Near-Infrared (NIR) camera under the Infrared (IR) light transmission. This NIR camera produces images of the finger containing vein patterns. The image additionally contains various shades produced by the thickness of finger muscles, bones, and tissue networks surrounding the vein. This paper focuses on developing a system to identify finger veins and the procedure to make transactions secure. The proposed method provides an accuracy of over 90% using filtering methods such as Segmentation, Gabor filtering, Masking, and Skeletonization of the image to extract ROI, which will be used for comparison and assist in lowering the occurrences of forgery instances that are caused by other biometrics. The identity of a user is registered through their finger vein image and stored in the database which will be used to test against future authentication thus making sure that the finger vein cannot be replaced.

I. Introduction

Biometric security implies that an individual's body turns into a “key”, which can help that individual gain access to various resources. Biometric technology comes with its personal set of problems. Biometric models work with physical traits such as fingerprint, iris, face, finger veins, palm veins, out of which face, iris patterns, and fingerprint can be altered by various methods. These biometrics are currently in use over a wide range of applications, including credit card authentication, automobile security, employee time and attendance tracking, computer and network authentication, endpoint security automated teller machine (ATM).