Computerized system for fingerprint identification for biometric security

Abstract

AFTS (automated fingerprint identification system) is used for identification of fingerprint images. AFIS is a type of biometric system which uses digital imaging methodology to capture, analyze and store fingerprint data. This fingerprint data can be compared with the database to determine the identity of individual. The fingerprint images of poor quality need to be enhanced before the feature extraction and feature matching. In this paper, we present a novel automated approach for fingerprint matching, which involve three steps. In the first step, we enhance image to remove noise. In the second step, we extract required feature that is necessary for matching and remove unwanted feature. In the third step, we do matching by using adjacent feature vector technique. The advantage of adjacent feature vector technique is that, it shows a good performance to differentiate matching and non- matching minutiae pairs. AFV (Adjacent Feature Vector) also reduce the computational cost and matching error. We have used standard fingerprint image database FVC2004 (Fingerprint Verification Competition). Experimental result shows that our approach is effective and gets high quality matching.