Abstract
Previous fingerprint verification systems have achieved good results, but these have been affected by the quality of input data. Fingerprint verification systems that use many fingers or multiple impressions of the same finger are more efficient and reliable than systems that use a single finger. However, multiple impressions give inconvenience to the user and increase the overall verification time. Therefore, we use only two fingers (multi-unit fusion) to improve performance of the fingerprint verification system. We show that performance can be improved by selecting a better quality fingerprint image of two fingerprints. Also, we propose a new quality checking algorithm composed of three stages. Our experimental results show that when the quality checking algorithm is performed by selecting a better quality fingerprint image of two fingerprints, there is a significance improvement in performance of the fingerprint verification system.
Original language | English |
---|---|
Pages (from-to) | 141-150 |
Number of pages | 10 |
Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Volume | 3546 |
DOIs | |
State | Published - 2005 |
Event | 5th International Conference on Audio - and Video-Based Biometric Person Authentication, AVBPA 2005 - Hilton Rye Town, NY, United States Duration: 20 Jul 2005 → 22 Jul 2005 |