Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/87023
Title: Authenticating personal identities using palmprint recognition
Authors: Li, Wenxin
Degree: Ph.D.
Issue Date: 2004
Abstract: Palmprint recognition technology is based on the identification of creases, wrinkles, ridges and hollows of the inner surface of the human palm. Palmprint recognition research, especially where it integrates fingerprint and hand geometry, is an area of great promise in the search for ever more reliable personal identity authentication and security systems. Our research over the last five years has made a number of major practical and theoretical contributions to the field of palmprint recognition. Our development of an online capture device has proven the feasibility of using palmprint recognition technology to identify people in real time. Our establishment of a palmprint database (ten thousand images collected from one thousand volunteers) has allowed us to thoroughly test the development of several novel palmprint positioning/segmentation, feature extraction/presentation and matching algorithms. We have proposed three palmprint positioning algorithms based on end points of the heart line and the outer boundary of the palm, on conjunction points between fingers, and on an inscribed circle. For the reference of researchers in the field, we also developed a number of feature extraction and matching methods: texture analysis, Fourier Transform and bi-directional matching of creases. This last proved itself 97.5% accurate and we are working on a new method to improve it. Finally, we also implemented a proof-of-concept palmprint verification system integrated with a smart card. Our publications include eight journal papers and five conference papers with one paper currently under review. In our future work we hope to improve the capture device, develop, test and compare other methods, enlarge our database, and find a way to locate and identify a palmprint in a free-of-scale image. Palmprint recognition shows great promise as a way to authenticate personal identities. With further research it will certainly become an established and accepted technology in the increasingly important fields of security and personal identity authentication.
Subjects: Hong Kong Polytechnic University -- Dissertations
Biometric identification
Palmprints
Pages: iv, 139 leaves : ill. ; 30 cm
Appears in Collections:Thesis

Show full item record

Page views

39
Last Week
0
Last month
Citations as of Apr 21, 2024

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.