University Links: Home Page | Site Map
Covenant University Repository

Development of an Improved Fingerprint Feature Extraction Algorithm for Personal Verification

Atuegwu, Charity and Daramola , S.A and Okokpujie, Kennedy O. and Noma-Osaghae, Etinosa (2018) Development of an Improved Fingerprint Feature Extraction Algorithm for Personal Verification. International Journal of Applied Engineering Research , 13 (9). pp. 6608-6612. ISSN 0973-4562

[img] PDF
Download (395Kb)

Abstract

New and sophisticated technologies are regularly developed to counter every new wave of breaches in data security. At the heart of some of these technologies is the personal verification system that rests on the oars of biometrics. Biometric systems use unique physical and behavioral traits for identification or verification. In this paper, an improved fingerprint feature extraction algorithm for personal verification is proposed. The improved fingerprint feature extraction algorithm is capable of recognizing authorized individuals and differentiating them from fraudulent imposters. The input images were preprocessed before extracting robust features for matching. Euclidean distance was used for classification. The proposed system was tested using the fingerprint images of fifty registered individuals and thirty imposters. The results obtained were a False Acceptance Rate and False Rejection Rate of 16% and 24% respectively. It is also faster than other feature extraction algorithms by forty (40) seconds Keywords: Fingerprint, biometrics, robust features, division into blocks, ridge pattern, euclidean distance, personal verification, feature extraction, classification.

Item Type: Article
Subjects: A General Works > AC Collections. Series. Collected works
T Technology > T Technology (General)
Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User: Engr. Kennedy O. Okokpujie
Date Deposited: 25 Apr 2018 13:44
Last Modified: 25 Apr 2018 13:44
URI: http://eprints.covenantuniversity.edu.ng/id/eprint/10694

Actions (login required)

View Item View Item