Fingerprint Authentication System based on Coalesce Portrayal Approach

Abstract

Fingerprint identification system is one of the most popularly used biometrics solutions for human identification and authentication as a result of its uniqueness and universality. It has been used in law enforcement and forensics, to identify suspects and victims for over a century. However, in spite of its applications in diverse identification systems, a number of challenges still recedes its effectiveness and efficiency, especially in applications with single mode representation. Study have shown that biometric system that uses a single representation for identification has this predisposition to contend with problems related to distortion, noisy data and error in matching, which often leads to reduction in the accuracy of the system. Therefore, this study considered fingerprint authentication system using coalesce portrayal approach in a single application to improve the efficiency of the system. A biometric system platform was developed and successfully used to capture fingerprint images from 50 persons and extracted salient features of both minutiae and texture of the fingerprint using standard enrolment devices with 500 pixels per inch (ppi). Each minutia was determined by calculating the Euclidean distance between the ridge ending and bifurcation, while the texture-based extraction was done using spatial grey level dependence matrix (SGLDM). Genuine and impostors’ identity were tested during the enrolment and verification processes. The performance of the system was measured in terms of false acceptance rate (FAR), false rejection rate (FRR), failure to enroll rate (FER). It took the system 58seconds to process a request with performance accuracy of 98% in 3 trials. The results from the implementation when compared with that of unimodal system, showed a better performance in terms of efficiency, accuracy and reliability.