Impact of Image Quality on Performance: Comparison of Young and Elderly Fingerprints

Abstract

Performance of fingerprint recognition systems is heavily influenced by the quality of fingerprints provided by the user. Image quality analysis is traditionally performed using local and global structures of fingerprint images like ridge flow, analysis of ridge-valley structures, contrast ratios etc. With large scale deployment of fingerprint recognition in systems like US VISIT program, image quality issues of fingerprint images from extreme age groups becomes even a more important issue. The impact of image quality on performance of fingerprint recognition systems should be a positive one i.e. higher image quality should lead to better overall performance of the system, and removal of lower quality images should improve performance of the system. This research study studied the impact of fingerprint image quality of two different age groups: 18-25, and 62 and above on overall performance using two different matchers. The difference in image quality between the two age groups was analyzed, and then the impact of image quality on performance of fingerprint matchers between the two groups was analyzed. Image quality analysis was performed using NFIQ which is part of NIST Fingerprint Image Software (NFIS). Neurotechnologija VeriFinger and bozorth3 (NFIS) matchers were used to assess overall performance. For the purposes of the research study, overall performance was measured using False Non Matches.