Abstract
Feature Matching is the last, deciding stage in a fingerprint-based verification system. In this paper, a Fingerprint Feature-Matching algorithm which is tailor-made for CNNs is presented. Together with the previously developed CNN Image-Preprocessing and Feature-Extraction algorithms, it realizes a CNN-based fingerprint verification system. The effectiveness of the proposed algorithm, more precisely, the performance of the entire system is evaluated by using a real fingerprint database.
Original language | English |
---|---|
Pages (from-to) | III73-III76 |
Journal | Proceedings - IEEE International Symposium on Circuits and Systems |
Volume | 3 |
State | Published - 2004 |
Externally published | Yes |
Event | 2004 IEEE International Symposium on Cirquits and Systems - Proceedings - Vancouver, BC, Canada Duration: 23 May 2004 → 26 May 2004 |