A comprehensive survey on fingerprint recognition systems

Abstract

Matching finger prints is the most popular biometric technique used for providing authentication. Fingerprint recognition systems scans for raw image, performs little preprocessing, features are extracted as vectors and stored in fingerprint databases. A review on various aspects of fingerprint recognition systems is presented in this paper. The paper briefs various types of fingerprint patterns, followed by minutiae based approach. Fingerprint ridges called minutiae are able to capture the invariant and discriminatory information present in the fingerprint images. Pattern recognition based approach is also studied followed by wavelet based approaches. The challenges and issues relating to fingerprint recognition system are critically reviewed in this paper. It is important for fingerprint recognition system to use good quality, noise free fingerprint image as input to achieve high accuracy. Various fingerprint image enhancement techniques were also analyzed and discussed in this paper.