Back to search

University of Kent

Intelligent interface agents for biometric applications

Abstract

dc:description.abstract

This thesis investigates the benefits of applying the intelligent agent paradigm to biometric identity verification systems. Multimodal biometric systems, despite their additional complexity, hold the promise of providing a higher degree of accuracy and robustness. Multimodal biometric systems are examined in this work leading to the design and implementation of a novel distributed multi-modal identity verification system based on an intelligent agent framework. User interface design issues are also important in the domain of biometric systems and present an exceptional opportunity for employing adaptive interface agents. Through the use of such interface agents, system performance may be improved, leading to an increase in recognition rates over a non-adaptive system while producing a more robust and agreeable user experience. The investigation of such adaptive systems has been a focus of the work reported in this thesis. The research presented in this thesis is divided into two main parts. Firstly, the design, development and testing of a novel distributed multi-modal authentication system employing intelligent agents is presented. The second part details design and implementation of an adaptive interface layer based on interface agent technology and demonstrates its integration with a commercial fingerprint recognition system. The performance of these systems is then evaluated using databases of biometric samples gathered during the research. The results obtained from the experimental evaluation of the multi-modal system demonstrated a clear improvement in the accuracy of the system compared to a unimodal biometric approach. The adoption of the intelligent agent architecture at the interface level resulted in a system where false reject rates were reduced when compared to a system that did not employ an intelligent interface. The results obtained from both systems clearly express the benefits of combining an intelligent agent framework with a biometric system to provide a more robust and flexible application.

Degree

thesis:*
Name dc:type.qualificationname
phd
Level dc:type.qualificationlevel
doctoral
Grantor dc:publisher.institution
University of Kent
Year dc:date.issued
2005

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Mavity, Nick

Subjects

dc:subject × 1

Rights

Language dc:language
en

Chain of custody

source
Harvested from
University of Kent
Base URL
kar.kent.ac.uk/cgi/oai2
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Mavity, Nick. Intelligent interface agents for biometric applications. doctoral thesis, University of Kent, 2005.