Application of Particle Swarm Optimization in Optimizing Stereo Matching Algorithm’s Parameters for Star Fruit Inspection System

Nasroddin, Saidatul Nizan and Mohd Mokji, Musa and Tan, Kok and Zainal Abidin, Amar Faiz and Amirulah, Rahman and Nordin, Nur Anis and Hasim, Saipol Hadi and Zakaria, Hamzah and Hassan, Jefery and Jaafar, Hazriq Izzuan and Khairuddin, Osman (2014) Application of Particle Swarm Optimization in Optimizing Stereo Matching Algorithm’s Parameters for Star Fruit Inspection System. In: International Conference Recent treads in Engineering & Technology (ICRET’2014), 13-14 Feb 2014, Batam (Indonesia).

[img] PDF
2014_Conf._13Feb_ICRET2014_BATAM_(3).pdf

Download (698kB)

Abstract

This paper reports the finding of the experimentation of the Particle Swarm Optimization in optimizing the stereo matching algorithm’s parameters for the star fruit inspection system. The star fruit inspection system is built by CvviP Universiti Teknologi Malaysia. While the stereo matching algorithm used in the experiment is taken from the Matlab library. Each particle of Particle Swarm Optimization in the search pace repsents a set of candidate numerical value of the stereo matching’s parameters. The fitness function for this application is the sum of absolute error of the gray scale value of both images. Based on this information, the particles will improve its position in the search space by moving towards its best record and the swarm best record. The process repeated until the maximum iteration met. The result indicates that there is potential application of Particle Swarm Optimization in stereo matching’s parameters tuning.

Item Type: Conference or Workshop Item (Paper)
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Electrical Engineering > Department of Control, Instrumentation & Automation
Depositing User: HAZRIQ IZZUAN JAAFAR
Date Deposited: 05 Mar 2014 04:52
Last Modified: 28 May 2015 04:19
URI: http://eprints.utem.edu.my/id/eprint/11672
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item