Article: A hybrid technique to secure fingerprint templates using Delaunay neighbour structures Journal: International Journal of Trust Management in Computing and Communications (IJTMCC) 2016 Vol.3 No.4 pp.319 - 335 Abstract: The rampant increase in the use of biometric authentication systems in a variety of applications made template protection for biometrics an important issue to be considered in the past decade. The leakage of biometric information causes privacy threats. Hence, biometric templates should be protected. In this paper, we propose a hybrid method for fingerprint template protection by combining cancellable biometrics and bio-cryptosystems. First, we construct Delaunay neighbour structure (DNS) for all minutia points in the fingerprint image. Second, we transform the DNS of each minutia point based on the distance and orientation difference between the minutia points in the DNS. Third, these transformed features are encrypted by using Bose, Chaudhuri, and Hocquenghem (BCH) error correction code in fuzzy commitment scheme. Hence, a secure template is obtained. An EER of 1.15%, 1.49%, 5.81%, 10.76%, 11.54% and 15.21% is obtained for the proposed method on FVC 2002 DB1, DB2, DB3, FVC 2004 DB1, DB2, and DB3, respectively. Experimental results and analysis prove the efficacy of the proposed method. Inderscience Publishers - linking academia, business and industry through research

Title: A hybrid technique to secure fingerprint templates using Delaunay neighbour structures

Authors: Mulagala Sandhya; Munaga V.N.K. Prasad

Addresses: Institute for Research and Development in Banking Technology, Road No. 1, Castle Hills, Masab Tank, Hyderabad, India 500057; School of Computer and Information Sciences, University of Hyderabad, Gachibowli, Hyderabad, 500046, India ' Institute for Research and Development in Banking Technology, Road No. 1, Castle Hills, Masab Tank, Hyderabad, 500057, India

Abstract: The rampant increase in the use of biometric authentication systems in a variety of applications made template protection for biometrics an important issue to be considered in the past decade. The leakage of biometric information causes privacy threats. Hence, biometric templates should be protected. In this paper, we propose a hybrid method for fingerprint template protection by combining cancellable biometrics and bio-cryptosystems. First, we construct Delaunay neighbour structure (DNS) for all minutia points in the fingerprint image. Second, we transform the DNS of each minutia point based on the distance and orientation difference between the minutia points in the DNS. Third, these transformed features are encrypted by using Bose, Chaudhuri, and Hocquenghem (BCH) error correction code in fuzzy commitment scheme. Hence, a secure template is obtained. An EER of 1.15%, 1.49%, 5.81%, 10.76%, 11.54% and 15.21% is obtained for the proposed method on FVC 2002 DB1, DB2, DB3, FVC 2004 DB1, DB2, and DB3, respectively. Experimental results and analysis prove the efficacy of the proposed method.

Keywords: cancellable template; biometric cryptosystem; fuzzy commitment scheme; BCH code; dimensionality reduction.

DOI: 10.1504/IJTMCC.2016.084560

International Journal of Trust Management in Computing and Communications, 2016 Vol.3 No.4, pp.319 - 335

Received: 15 Apr 2016
Accepted: 18 Jul 2016
Published online: 14 Jun 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: A hybrid technique to secure fingerprint templates using Delaunay neighbour structures

Keep up-to-date