Skip to main content

An Investigation of Iris Recognition in Unconstrained Environments

Bonner, Richard (2014) An Investigation of Iris Recognition in Unconstrained Environments. Doctor of Philosophy (PhD) thesis, University of Kent,. (KAR id:50463)

PDF
Language: English
Download this file
(PDF/80MB)
[thumbnail of 201Thesis_Richard_Bonner.pdf]
Preview

Abstract

Iris biometrics is widely regarded as a reliable and accurate method for personal identification and the continuing advancements in the field have resulted in the technology being widely adopted in recent years and implemented in many different scenarios. Current typical iris biometric deployments, while generally expected to perform well, require a considerable level of co-operation from the system user. Specifically, the physical positioning of the human eye in relation to the iris capture device is a critical factor, which can substantially affect the performance of the overall iris biometric system. The work reported in this study will explore some of the important issues relating to the capture and identification of iris images at varying positions with respect to the capture device, and in particular presents an investigation into the analysis of iris images captured when the gaze angle of a subject is not aligned with the axis of the camera lens. A reliable method of acquiring off-angle iris images will be implemented, together with a study of a database thereby compiled of such images captured methodically. A detailed analysis of these so-called “off-angle” characteristics will be presented, making possible the implementation of new methods whereby significant enhancement of system performance can be achieved. The research carried out in this study suggests that implementing carefully new training methodologies to improve the classification performance can compensate effectively for the problem of off-angle iris images. The research also suggests that acquiring off-angle iris samples during the enrolment process for an iris biometric system and the implementation of the developed training configurations provides an increase in classification performance.

Item Type: Thesis (Doctor of Philosophy (PhD))
Thesis advisor: Fairhurst, Michael
Uncontrolled keywords: Iris recognition, Unconstrained environments, Non-ideal conditions, Off-axis iris images, Gaze angle
Subjects: Q Science > QC Physics
T Technology > TK Electrical engineering. Electronics. Nuclear engineering
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Engineering and Digital Arts
Funders: Engineering and Physical Sciences Research Council (https://ror.org/0439y7842)
Depositing User: Users 1 not found.
Date Deposited: 10 Sep 2015 17:00 UTC
Last Modified: 09 Dec 2022 17:24 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/50463 (The current URI for this page, for reference purposes)

University of Kent Author Information

Bonner, Richard.

Creator's ORCID:
CReDIT Contributor Roles:
  • Depositors only (login required):

Total unique views for this document in KAR since July 2020. For more details click on the image.