Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Computing | en_US |
| dc.creator | Xu, Yan | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/6417 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | - |
| dc.rights | All rights reserved | en_US |
| dc.title | Study of pore extraction algorithms for high resolution fingerprint recognition | en_US |
| dcterms.abstract | Fingerprint recognition is widely used in many domains and has matured as a field of study. Although, there are three levels of fingerprint features. Most of the existing automatic fingerprint recognition systems (AFRS) are based on level-2 features (minutiae). In some situations, however, using only level-2 features cannot accurately recognize the fingerprint and level-3 features are combined with level-2 features (pores) to improve recognition. However, pore-based fingerprint recognition has a major limitation in that it requires high resolution and good quality fingerprint images. Thanks to the development of high resolution technique pore-based fingerprint recognition can now be implemented with greater success. This thesis introduces adaptive fingerprint pore extraction algorithm. It first partitions blocks as well-defined, ill-defined and background blocks based on ridge orientation and ridge frequency. Well-defined blocks are applied in the adaptive fingerprint pore module and ill-defined blocks are used in the DoG module to generate a matched filter through which pores can be detected. Then the direct pore matching algorithm is used to finish pore matching. There are two key steps in this process: first, descriptors are used to calculate the initial pore correspondence based on the similarity of two pores; second, it refines and fixes the errors of matched pores by using the RANSAC (RANdom SAmple Consensus) algorithm according to the initial pore correspondence. After finishing the mentioned two steps, the match score can be calculated. The last part depicts feature fusion. Experimental results indicate that combining pores and minutiae to recognize fingerprints is more accurate than only using pores or minutiae. | en_US |
| dcterms.extent | vii, 55 leaves : ill. ; 30 cm. | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2012 | en_US |
| dcterms.educationalLevel | All Master | en_US |
| dcterms.educationalLevel | M.Sc. | en_US |
| dcterms.LCSH | Fingerprints -- Identification -- Data processing. | en_US |
| dcterms.LCSH | Pattern recognition systems. | en_US |
| dcterms.LCSH | High resolution imaging. | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | restricted access | en_US |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b24736612.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 2.78 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/6417