A Novel Technique for Fingerprint Based Secure User Authentication

Fingerprint biometrics are extremely popular among digital identity systems for authentication. Most fingerprint based authentication systems store unique minutiae points information of a user directly in the database as their template. An adversary can potentially attack the user template that could compromise information of minutiae points. It has been proven that a fingerprint can be reconstructed from minutiae points and hence there is an important requirement in fingerprint based biometric systems to provide protection and privacy. This article introduces a novel technique to provide enhanced protection for fingerprint based biometrics, taking into account the attributes of minutiae point, the ridge count between the neighbouring minutiae points and the local minutiae point structure. The proposed technique has been implemented and tested on a total of nine fingerprint databases of varying quality. In comparison to the other existing techniques under different scenarios, the proposed technique exhibits superior recognition performance. The proposed technique is a singular point independent and has shown high stability, robustness, and security.