Back to search

University of Alabama Birmingham

Multimodality Image Registration

Abstract

dc:description.abstract

Image registration has been a major area of research in the last few decades. It is a complex problem that involves handling large data sets. The relation between the real object and the image is very intricate and encompasses the object’s shape, surface properties, position, and illumination. In this thesis, a general solution to this problem was devised that can be applied to multimodality and within-modality image registration. The approach followed here is such that various factors such as illumination do not affect the algorithm. The algorithm considers the intensity of the images as raw data for the registration. The method presented in this research applies mutual information (MI) to measure the information redundancy. The redundancy is calculated between the image intensities of corresponding voxels in both images. If the images are aligned with each other closely, then the MI is assumed to be maximal. Maximization of MI is a powerful criterion for image registration. The result of the image registration algorithm is validated for computed tomography and magnetic resonance imaging by comparing with a sub-pixel registration based on intensity. Image registration is obtained automatically, as well as manually, without following any preprocessing steps such as segmentation and feature selection.

Degree

thesis:*
Level thesis:degree_level
Thesis

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Prasai, Persis
Contributors dc:contributor
  • Murat M Tanik
  • Gary J Grimes
  • John L Hartman IV

Subjects

dc:subject × 1

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
University of Alabama Birmingham
Base URL
digitalcommons.library.uab.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Prasai, Persis. Multimodality Image Registration. Thesis thesis, https://digitalcommons.library.uab.edu/etd-collection/3622