A Study On Improvements for Fingerprint Identification
| Autor: | Yu-ren Lai, 賴郁仁 |
|---|---|
| Rok vydání: | 2008 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 96 In this paper, we present a fast geometric fingerprint matching based on graph matching principles. We use a representation called K-plets to match local neighborhood minutia, and we use the other presentation called CBFS that is a graph traversal algorithm for consolidating the local neighborhood matches. The proposed method is making effort to decrease the computational complexity. Then we use convex-hull algorithm to narrow down fingerprint minutiae for matching and reducing the matching time. In the other hand, fingerprint database classification method is our concerned. We use fingerprint’s ridges and estimate which fingerprint cluster is. We hope automated fingerprint identification system will be used in our daily life. In our experiment, our proposed method shows not only the same recognition ratio but also less matching time. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|