Abstract | ||
---|---|---|
Security of biometric templates in a fingerprint authentication system is highly critical because the number of fingers of a person is limited and the raw fingerprints cannot be reset or replaced. A cancelable template is an efficient and powerful means to provide template protection. However, in most cancelable fingerprint templates, the many-to-one based non-invertible transformation acts on each single minutia, which may greatly decrease the discriminative capability of its feature representation, and more importantly, single minutia is more sensitive than a local structure to the non-linear distortion present in a fingerprint image. In this paper, we propose a cancelable template by utilizing some nice properties of Delaunay triangle-based local structures, e.g., excellent local structural stability, to achieve satisfactory performance. The non-invertible transformation is applied to each Delaunay triangle-based local structure rather than to each individual minutia so as to mitigate the influence of non-linear distortion and retain the discriminative power of the feature data after the many-to-one mapping. Experimental results on the publicly available databases demonstrate the validity of the proposed scheme. © Springer International Publishing Switzerland 2013. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1007/978-3-319-03584-0_7 | CSS |
Keywords | Field | DocType |
biometrics,cancellable template,delaunay triangles,fingerprint,local structure,security | Pattern recognition,Fingerprint recognition,Minutiae,Algorithm,Fingerprint,Artificial intelligence,Engineering,Biometrics,Template,Distortion,Delaunay triangulation,Feature data | Conference |
Volume | Issue | ISSN |
8300 LNCS | null | 16113349 |
Citations | PageRank | References |
17 | 0.65 | 21 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wencheng Yang | 1 | 85 | 10.34 |
Jiankun Hu | 2 | 1976 | 150.35 |
Song Wang | 3 | 321 | 16.09 |
JuCheng Yang | 4 | 72 | 7.49 |