Authors: Abdallah Meraoumia; Djamel Samai; Salim Chitroub
Addresses: Laboratory of Mathematics, Informatics and Systems (LAMIS), University of Tebessa, Tebessa, 12002, Algeria ' Lab. de Génie Électrique, Fac. des nouvelles technologies de l'information et de la communication, Univ. Ouargla, Ouargla, 30 000, Algeria ' Laboratory of Intelligent and Communication Systems Engineering (LISIC), University of Houari Boumediene, Algiers, 16111, Algeria
Abstract: Communication via the internet has become vital for any kind of information exchange private, public, commercial, or military. Banks are the first that have used the internet for financial transactions (e-banking). However, the safe use of e-banking implies that all precautions have been considered to identify legitimate users and thus avoid economic and social damage that may be caused by any possible fraud. In this context, we propose in this paper a secure biometric system dedicated to e-banking for reducing the fraud risk and strengthening the customer confidence. The fuzzy commitment concept associated with the finger-knuckle-print (FKP) is the core of our proposed system. However, such a system will only be efficient if the FKP features are accurately extracted. For this, we have developed a new method of feature extraction called adaptive extended binary pattern (AELBP). The obtained experimental results have been judged promising for a high security of e-banking with guaranteed trust from costumers.
Keywords: information security; cryptography; fuzzy commitment; biometrics; feature extraction; finger-knuckle-print; FKP; local binary pattern; LBP; data fusion.
International Journal of Embedded Systems, 2019 Vol.11 No.3, pp.325 - 339
Received: 18 Jul 2016
Accepted: 09 Mar 2017
Published online: 23 Apr 2019 *