International Journal of Computer Applications (0975 – 8887) Volume 66– No.21, March 2013 1 Fingerprint Matching using Neighbourhood Distinctiveness Iwasokun Gabriel Babatunde Federal University of Technology, PMB 704, Akure, Ondo State, Nigeria, Akinyokun Oluwole Charles Federal University of Technology, PMB 704, Akure, Ondo State, Nigeria Angaye Cleopas Officer National Information Technology Development Agency, Abuja, Nigeria ABSTRACT The issue of identity management has continued to pose serious security challenge to different organizations. To cub this challenge, emphasis is now been shifted from what you know or have to what you are leading to increasing use of fingerprint, iris voice, face image and other physical biometrics for human verification and identification. Among these, fingerprint has proved most reliable and dependable. This has precipitated the emergence of a good number of Automated Fingerprint Identification Systems (AFIS) with different forms of matching algorithms. This paper presents the formulation and implementation of a minutiae based fingerprint pattern matching algorithm. The algorithm relies on the spatial characteristics defined over the 11 x 11 neighbourhood of the fingerprints core points to determine the matching scores, which exhibit the degree of resemblance for any two images. Results obtained from the implementation of the proposed algorithm show its good performance. Comparative analysis of the obtained FNMR, FMR and computation time values with values obtained from some other research works shows a superior performance of the proposed system. Keyword: Fingerprint, Pattern Matching, Core Point, Minutiae, FNMR, FMR 1. INTRODUCTION The issue of identity management of individuals, organizations and other public and private institutions poses a great challenge to government worldwide today. Biometric identification has featured prominently for individuals with fingerprint emerging as the dominant one. The dominance of fingerprint has been buttressed by the continuous emergence of different forms of Automated Fingerprint Identification Systems (AFIS). Fingerprints are the results of minute ridges and valleys in the fingers of every person. The ridges are the dark and raised portions while the valleys are the white and lowered portions as shown in Figure 1. The ridges do not change at any time, from birth until death. No matter what happens, they reappear within a short period. In most cases, they appear in any of the five major patterns; namely left loop, right loop, whorl, arch and tented arch as shown in Figure 2 [1-6]. In the loop pattern, the ridges enter from either side, re-curve round the core point and pass out (or tend to pass out) the same side they entered. In the right loop pattern, the ridges enter from the right side while they enter from the left side in the left loop pattern. In a whorl pattern, the ridges are usually circular round the core point while the ridges enter from one side, make a rise round the core point and exit generally on the opposite side in the arch pattern. A fingerprint may be described as captured or latent print [7]. A captured print is obtained for different purposes. When a person is arrested in connection with a crime, as part of the booking process, the police or other security agent rolled the arrestee’s fingertip in ink and then impressed it on a card. The card is subsequently stored in libraries (of such cards) maintained by local, state or national agencies. Captured prints may also be obtained using modern day finger scan system [8-9]. The dawning of electronically scanning fingers to obtain fingerprint images and intelligent computer processes and algorithms that enhance, enroll and match prints to an extremely high accuracy level has provided an efficient means of satisfying the need for a full automation of fingerprint identification process. The basic components of a fully automated finger-scan system shown in Figure 3 comprise of components for image acquisition, processing and template generation, matching and storage. These components can be located within a single peripheral or standalone device, or may be spread among a peripheral device, a local PC, and a central server [8]. Latent print in contrast, is typically produced at a crime scene and is usually not readily visible. It occurs when the natural secretions of the skin are deposited on a surface through fingertip contact. The best way to render latent fingerprints visible for photographing is complex and depends, for example, Ridges Valleys Figure 1: Fingerprint Ridges and Valleys Figure 2: Basic types of fingerprint pattern Left loop Right loop Whorl Arch Tented Arch