Efficient Fingerprint Matching Algorithm for Reduced False Similarity Contribution In Forensics and Partial Finger Prints  

Waqas Ellahi1, Kashif Sardar1, Mohsin Raza2, Nauman Aslam3, Wajih Ullah Baig1,

1 Center for Advanced Studies in Engineering (CASE), Islamabad, Pakistan

2 Department of Math, Physics and Electrical Engineering, Northumbria University, Newcastle, UK

3 Department of Computer Science and Digital Technologies Northumbria University, UK

Abstract

Minutiae based matching algorithms use both local and global matching techniques for a better discrimination of scores. However, these algorithms inherently produce high false matching scores for partial fingerprint identification. This paper introduces a novel mechanism to reduce false matching scores without degrading true matching scores by evaluating minutiae points that reside in over-lapping fingerprint area. The main contributions of the paper are 1) introduction of new and faster algorithm that replaces the GrahamScan algorithm to find a point inside a convex hull 2) reduction of false match scores. The performance of the proposed algorithm is evaluated using FVC-2002 and other databases. FingerjetFx is used for feature extraction. The evaluated results show that the proposed algorithm offers improved accuracy along with over 300 percent improvement in computational efficiency.

Keywords: Fingerprint Recognition; Minutiae; Convex Hull; Point slope formula; Equal Error Rate (EER), optimization.

Citation: Ellahi W, Sardar K, Raza M, Aslam N, Baig WU (2017). Efficient Fingerprint Matching Algorithm for Reduced False Similarity Contribution In Forensics and Partial Finger PrintsWorkshop on Sustainability and Green Technology 2017, eds Le L (Ho Chi Minh City, Viet Nam). 
Full-text Download: PDF