Back to search

West Virginia University

Segmentation of images with low-contrast edges

Abstract

dc:description.abstract

A vast amount of the current research in medical image analysis has aimed to develop improved techniques of image segmentation. Of the existing approaches, active contour methods have proven effective by incorporating edge or region information from the image into a level set formulation. However, complications arise in images containing regions of low-contrast due to noise, occlusions, or partial volume effects, which are often unavoidable in practical applications. Incorporating prior shape information into the segmentation framework provides a more accurate and robust solution by constraining the evolving contour to resemble a target shape. Two methods are presented to incorporate a shape prior into existing active contour segmentation methods, including the edge-based geodesic active contours model and a fast update implementation of the region-based Chan-Vese model. Applying these methods to synthetic and real images demonstrates that an improved result can be obtained for images containing low-contrast edge regions.

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
  • Madden, Matthew J.

Subjects

dc:subject × 2

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

Madden, Matthew J.. Segmentation of images with low-contrast edges. Thesis thesis, https://doi.org/10.33915/etd.1833