Recent advances in automated fingerprint recognition systems coupled with the growing need for reliable and efficient identification system have resulted in an increased deployment of fingerprint biometric technology in many broad applications. The system has been adjudged one of the most publicized biometrics across the globe as it offers reliable means of personal identification because of its uniqueness and consistency over time. Fingerprint recognition system has been successfully used in law enforcement and forensics to identify suspects and victims for over a century. It is very relevant in border and access control, employment background checks, students’ examinations and class attendance, user authentication on laptops and mobile devices et.c. Yet, there are number of challenges that recedes the effectiveness of the system. For instance, errors in matching process that particularly occur in many applications with single mode representation due to distortions and noisy data. Consequently this has led to a significant reduction in the accuracy of the system. Therefore, this research work is focused at considering fingerprint recognition system using multiple representations. Specifically, minutiae and texture based fingerprint representations are considered. Samples data collected for the research work were trained and tested using the classifiers system designed for the work. The approach was found to be more robust than the single mode representation in terms of accuracy and speed.
Key words: Fingerprint recognition, matching process, unimodal biometric, multiple representations, fusion techniques.
Copyright © 2019 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0