A Fingerprint Verification System based on Minutiae Matching
| Autor: | Chao Gwo-Cheng, 趙國成 |
|---|---|
| Rok vydání: | 2004 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 93 Fingerprint based identification has been known and used for a long time. Owing to the uniqueness and immutability, fingerprints are today the most widely used biometric features. Fingerprint verification is based on the minutiae of the fingerprints. Extracting minutiae from fingerprint patterns is one of the most important steps in automatic system for fingerprint identification. This model, adopted in most automatic systems, is based on a two-class minutiae classification: endpoint and bifurcation .The accuracy of the minutiae will affect the correctness of the fingerprint matching; therefore the fingerprint enhancement will be done to reduce the false minutiae. In this thesis, an effective fingerprint verification system is implemented by passing the image pre/post-processing, feature extraction and matching processes. We describe an improved method on fingerprint enhancement and post-processing. We directly enhance the fingerprint on gray-scale image and reduce the most false minutiae in the post-processing step. For minutiae matching, an alignment-based elastic matching algorithm has been developed. This algorithm is capable of finding the corresponding point pairs between input and template minutiae. It also has the ability to adaptively compensate for the nonlinear deformations between fingerprints. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|