Semester

Fall

Date of Graduation

2007

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

Matthew Valenti.

Abstract

Current methods for comparing fingerprints have weaknesses that have opened them to criticism. Current methods concentrate on the comparison of minutia in the print either manually or with the assistance of a computer algorithm. This causes these methods to depend highly on the presence of minutia and their relationship to one another. Absence or rotations of minutia can prevent current methods form making accurate comparisons. The goal of this process is to develop a new method for analyzing fingerprints that addresses many of the concerns with current methods.;The developed process uses an iterated function sequence (IFS) to convert the image of a fingerprint into a fractal pattern. The input for the IFS is constructed by a random walk through the image. Once a fingerprint is converted into a fractal pattern, the fractals can be used to make comparisons. Fractals are well defined mathematical objects that make them far easier to compare than fingerprints themselves. This process addresses many of the issues with current methods. This method is global in nature and thus it is not dependent on a set number of minutiae. Moreover, the rules for the random walk are constructed so as to make the fractal produced invariant of orientation of the print.;This method offers a new fast way to compare images. This method can be used to increase confidence, both in court and public opinion, in the use of fingerprints as identification. It can offer both an independent and/or supplemental method to the current ones used.

Download

Share

COinS