UMP Institutional Repository

Data Fitting by G1 Rational Cubic Bézier Curves Using Harmony Search

Najihah, Mohamed and Ahmad, Abd Majid and Abd Rahni, Mat Piah (2015) Data Fitting by G1 Rational Cubic Bézier Curves Using Harmony Search. Egyptian Informatics Journal, 16 (2). pp. 175-185. ISSN 1110-8665

[img]
Preview
PDF
Data fitting by G1 rational cubic Bézier curves using harmony search.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (1MB) | Preview

Abstract

A metaheuristic algorithm, called Harmony Search (HS) is implemented for data fitting by rational cubic Bézier curves. HS is a derivative-free real parameter optimization algorithm, and draws an inspiration from the musical improvisation process of searching for a perfect state of harmony. HS is suitable for multivariate non-linear optimization problem. It is mainly achieved by data fitting using rational cubic Bézier curves with G1 continuity for every joint of segments of the whole data sets. This approach has significant contributions in making the technique automated. HS is used to optimize positions of middle points and values of the shape parameters. Test outline images and comparative experimental analysis are presented to show effectiveness and robustness of the proposed method. Statistical testing between HS and two other different metaheuristic algorithms is used in the analysis on several outline images. All of the algorithms improvised a near optimal solution but the result that is obtained by the HS is better than the results of the other two algorithms.

Item Type: Article
Uncontrolled Keywords: Rational cubic Bézier; Data approximation; Harmony search
Subjects: Q Science > Q Science (General)
Faculty/Division: Faculty of Industrial Sciences And Technology
Depositing User: Mrs. Neng Sury Sulaiman
Date Deposited: 26 Aug 2015 00:41
Last Modified: 26 Aug 2015 00:41
URI: http://umpir.ump.edu.my/id/eprint/10001
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item