Finger Nail Plate: A New Biometric Identifier

Abstract

The key objective of this paper is to investigate a new biometric identifier for forensic and civilian applications. We propose a novel biometric authentication system based on low resolution finger nail plate images. Despite the uniqueness and high stability of the nail plate, its usage as a biometric identifier has not been extensively investigated for personal authentication. This paper implements a convenient and computationally efficient approach on low resolution nail plate images acquired with a contactless and unconstrained imaging setup. It exploits the local shape and texture of the nail plate by employing appearance based and texture based feature descriptors. Scorelevel fusion rules are utilized for integrating the nail plates from mainly three fingers of the hand. Extensive experimentation in both the biometric verification and recognition scenarios is carried out and the results validate the finger nail plates as a potential biometric identifier.