Back to search

West Virginia University

Quality of iris segmentation as a predictor of verification performance

Abstract

dc:description.abstract

In this thesis, the accuracy of iris segmentation algorithms will be compared using quality measures based on both homogeneity and heterogeneity of the resulting image regions. After calculation of the quality measures, Principal Component Analysis (PCA) is performed. The resulting pairwise distances between the genuine pairs are compared to the hamming distance scores of the iris templates. The relationship between the two samples will be examined, and curve fitting by least squares regression is shown to be an adequate method of prediction for genuine Hamming distance score. The work proposes a segmentation quality metric that is highly correlated with the distance score.;By determining the relationship between the segmentation quality values and hamming distances, it is possible to apply the same quality measures for future segmented iris images to determine the approximate genuine Hamming distance score without having to first encode the iris template and then calculate the actual hamming distance. This would be of the most practical use in a verification biometric system, where the hamming distance score is used to confirm the identity of a user. The raw data obtained from the user can be examined and determined whether it should be used for verification after segmentation, without needing to have the hamming score calculated.

Degree

thesis:*
Name thesis:degree_name
MS
Level thesis:degree_level
Thesis
Discipline thesis:degree_discipline
Lane Department of Computer Science and Electrical Engineering

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Lee, Syvale

Subjects

dc:subject × 1

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
West Virginia University
Base URL
researchrepository.wvu.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Lee, Syvale. Quality of iris segmentation as a predictor of verification performance. Thesis thesis, https://doi.org/10.33915/etd.1857