Security enhancement of fingerprint authentication
Date of Issue2009
School of Electrical and Electronic Engineering
Biometric recognition deals with the use of distinctive physiological and behavioral characteristics of an individual such as iris, fingerprint, hand geometry, face, signature, gait etc which are called biometrics or biometric identifiers to automatically recognize a person. Fingerprint is currently the most widely used biometrics as it has good accuracy, easy to use and is cost effective. In spite of their numerous advantages, the fingerprint systems are vulnerable to some attacks, which can decrease their security level. Previous studies have shown that fingerprint scanners can be deceived very easily, using artificial fingerprints made using simple, inexpensive materials like gelatin or play-doh. A lot of research is also ongoing on how to cope with this problem and many of these researches have suggested liveness detection as one of the most apt solution. In this project, we propose a novel method of using surface texture details and perspiration details of fingertip to detect a fake finger attack. Although no formal definition of image texture exits, intuitively this provides a measure of properties such as surface smoothness, coarseness, and regularity. The variations in making a fake finger cause the creation of a much coarser surface in the case of a fake finger.
Final Year Project (FYP)
Nanyang Technological University