A model of two-factor authentication using facial recognition in automated teller machines

Abstract

The general idea of this research work is to use facial recognition technique to reinforce security on Automatic Teller Machines. The use of face recognition is to authenticate an individual by identifying and verifying him/her in an existing and up – to - d ate database of peoples face images. The level of security in ATM‘s has changed tremendously since their introduction in the late 70‘s. This has caused them to be very vulnerable because technology has brought in a new kind of attackers who have been using the advancement of technology to compromise their security. Something has to be done to guarantee security when it comes to people‘s funds and other confidential information. Face recognition is a technique that uses application software to determine the identity of an individual and is within the category of biometrics which is defined as ―the automatic recognition of a person using distinguishing traits‖ (RAND, 2003). Face recognition is among other biometrics including iris recognition, fingerprint, h and print and retina scans. This research provides a review of face - based authentication in accessing critical and confidential information in computer systems, the underlying motivation being to provide the latest review of the existing literature on faci al recognition and to bring to the limelight, the studies of computer vision in recognition of human faces. In order to provide a comprehensive review, we have provided detailed descriptions of representative methods and related topics such as issues of (l ighting) illumination and pose variation.