IIUM Repository

Formulation of hybrid 3D image segmentation algorithm based partial differential equation

Aboaba, A. A. and Hameed, Shihab A. and Khalifa, Othman Omran and Hassan Abdalla Hashim, Aisha and Chowdhury, Md. Sazzad Hossien (2012) Formulation of hybrid 3D image segmentation algorithm based partial differential equation. In: International Conference on Mathematical Application in Engineering (ICMAE2012), 03-05 July 2012, Kuala Lumpur.

[img]
Preview
PDF (FORMULATION OF HYBRID 3D IMAGE SEGMENTATION ALGORITHM BASED PARTIAL DIFFERENTIAL EQUATION) - Published Version
Download (728kB) | Preview
[img]
Preview
PDF
Download (1MB) | Preview

Abstract

Segmentation is an important tool for analysis and understanding of most images encountered in science and engineering. One of the best segmentation methods that can perform 3D segmentation is the level-set method which has its mathematical foundation in partial differential equation (PDE). Owing to its complex nature, it exhibits a level of unacceptable sluggishness on implementation hence a need to hasten up the process by hybridizing it with a faster region-based segmentation method which is inherently a logical approach to segmentation pivoted on thresholding but not as good in segmentation as the former. This work presents a mathematical hybrid of the two methods that is hoped to produce a better segmentation result.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Additional Information: 4428/28456
Uncontrolled Keywords: 3D segmentation, level-set, region-growing, partial differential equation (PDE), and hybrid segmentation
Subjects: Q Science > QA Mathematics > QA76 Computer software
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button): Kulliyyah of Engineering > Department of Science
Depositing User: Dr. Md Sazzad Hossien Chowdhury
Date Deposited: 17 Jan 2013 14:46
Last Modified: 15 Jun 2017 16:12
URI: http://irep.iium.edu.my/id/eprint/28456

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year