Emily Biller

Date of Graduation

2017

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Jeremy M Dawson

Committee Co-Chair

Donald Adjeroh

Committee Member

Thirimachos Bourlai

Abstract

Fingerprints are an essential biometric identifier that have been used for centuries due to the high diversity among individuals. Contactless fingerprint systems possess advantages over conventional contact-based systems, such as increased throughput in check point screening scenarios. Increased acceptance of contactless fingerprinting technologies requires careful operational validation due to the inherent differences in their capture techniques. To leverage the maturation of contactless technologies, reliability and data interoperability with legacy fingerprint datasets must be considered. The goal of this study is to evaluate the interoperability of contactless fingerprinting devices with contact-based fingerprinting devices. To achieve this goal, the fingerprint dataset of the ManTech Non-Contact Multi-Sensor Fingerprint Collection Phase II collected by West Virginia University, has been employed. The main objectives this study aims to achieve are:;Assessing the match performance of the contactless versus contact-based fingerprint devices. Statistical match performance evaluation using match score statistics. Assessing the match performance of demographic strata in comparison to the total dataset.;Following the generation of genuine and imposter match score distributions, Receiver Operating Characteristic Curves were plotted to compare the match performance of each device. The Kullback Leibler Divergence measurements have been calculated which signify the amount of variation between match score distributions as well as the Equal Error Rates for the genuine and imposter distributions. With the help of these procedures, the interoperability of contactless devices and contact-based devices has been examined and analyzed. Results obtained from the devices, representing a 'snapshot' of capability during their development cycle, indicate that Rank1 accuracy is between 74.33 and 98.68 percent with contact based methods. The Rank 1 accuracy is between 74.66 and 97.27 percent with other contactless devices.

Download

Share

COinS