Effects of Time-Lapse and Contact Lenses on Iris Recognition

Master's Thesis

Abstract

Iris biometrics is a flourishing field with multiple companies offering commercial systems being implemented in several real-world scenarios. However, before iris recognition can be implemented on a world-wide scale, important questions regard- ing its accuracy must be addressed.

From its conception, iris biometrics has operated with the assumption that the iris texture is highly stable over an individual’s lifetime and the claim has been made (and widely accepted) that enrollment is “once for a lifetime.” Using four years time-lapse data for 46 “iris-subjects?,” we present results from three different iris recognition systems that indicate that the iris biometrics template does “age”

Another element of iris biometrics folklore is that performance is not affected by sub jects wearing contact lenses during acquisitions. Our study on 124 non contact lens wearing sub jects and 88 contact lens sub jects yields results opposing this claim.

We present results debunking both of these common myths in iris recognition and show a real effect of time-lapse and contact lenses on iris biometric performance.

Attributes

Attribute NameValues
URN
  • etd-12012009-222648

Author Sarah Elizabeth Baker
Advisor Dr. Patrick Flynn
Contributor Dr. Gregory Madey, Committee Member
Contributor Dr. Kevin Bowyer, Committee Co-Chair
Contributor Dr. Patrick Flynn, Committee Co-Chair
Contributor Dr. Nitesh Chawla, Committee Member
Degree Level Master's Thesis
Degree Discipline Computer Science and Engineering
Degree Name MSCSE
Defense Date
  • 2009-08-17

Submission Date 2009-12-01
Country
  • United States of America

Subject
  • computer vision

  • time-lapse

  • iris biometrics

  • contact lenses

Publisher
  • University of Notre Dame

Language
  • English

Record Visibility and Access Public
Content License
  • All rights reserved

Departments and Units

Files

Please Note: You may encounter a delay before a download begins. Large or infrequently accessed files can take several minutes to retrieve from our archival storage system.