Efficient edge detection using fuzzy heuristic particle swarm optimization / Noor Elaiza Abdul Khalid, Mazani Manaf and Mohd Ezane Aziz

Abdul Khalid, Noor Elaiza and Manaf, Mazani and Aziz, Mohd Ezane (2009) Efficient edge detection using fuzzy heuristic particle swarm optimization / Noor Elaiza Abdul Khalid, Mazani Manaf and Mohd Ezane Aziz. Scientific Research Journal, 6 (1). pp. 43-63. ISSN 1675-7009

[img]
Preview
Text
AJ_NOOR ELAIZA ABDUL KHALID SRJ 09 1.pdf

Download (1MB) | Preview

Abstract

This paper presents a hybridization of Particle Swarm Optimization (PSO) and Fuzzy edge detector. The edge detector is used as the initial population and as the objective function. The purpose of hybridizing the algorithm is to create an optimized edge detector. Classical Fuzzy Heuristics (CFH) detects thick edges. These thick edges need to be optimized to obtain a thin line. In this research the PSO is used to optimize the edge detection detected by the CFH algorithm and it is referred to as FHPSO. The test images are radiographs images of the metacarpal. These images have been used, because there is a need to detect strong and thin edges. Radiograph images are noisy in nature, which makes it difficult to measure the cortical thickness, the cortical outline of the inner cortical and outer cortical of the long tubular bone. The outer cortical edges are considered to be the strong edges due to high discontinuity values and the inner cortical edges are considered weak edges due to low their discontinuity values. The performance of FHPSO in detecting edges has been shown to be quite efficient.

Item Type: Article
Uncontrolled Keywords: Fuzzy heuristics, Particle swarm optimization, tubular bone, Radiographs
Subjects: Q Science > QA Mathematics > Fuzzy logic
R Medicine > RC Internal Medicine > Radiography
Divisions: Research Management Institute (RMI)
Depositing User: Staf Pendigitan 1
Date Deposited: 20 Oct 2015 08:57
Last Modified: 27 May 2016 11:13
URI: http://ir.uitm.edu.my/id/eprint/12916

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year