Multi-Counterpropagation Network Model For Colour Recognition

Article Sidebar

FULL TEXT
Published: Jun 1, 1999
Keywords:
CPN model competitive layer unsupervised learning supervised learning Minolta Chroma Meters

Main Article Content

Razali Yaakob
Department of Computer Science Faculty of Comp. Sc. And Inform. Tech., Universiti Putra Malaysia
Md. Nasir Sulaiman
Department of Computer Science Faculty of Comp. Sc. And Inform. Tech., Universiti Putra Malaysia
Ramlan Mahmod
Department of Computer Science Faculty of Comp. Sc. And Inform. Tech., Universiti Putra Malaysia
Mahmud Tengku Muda Mohamed
Department of Agronomy and Horticulture, Faculty of Agriculture, Universiti Putra Malaysia
Abd. Rahman Ramli
Department of Computer and Communications, Faculty of Engineering, Universiti Putra Malaysia

Abstract

Minolta Chroma Meters was used to convert colours into numbers. It offers five different colour systems for measuring absolute chromaticity, that is, CIE Yxy, L*a*b*, L*C*Ho, Hunter Lab and XYZ. In this study, only L*a*b* is used, and combinations of two counterpropagation network (CPN) are required to recognise 808 colours produced by The Royal Horticultural Society, based on RHS Colour Chart [1]. Our proposed neural network model is tested; the result shows that 99% of trained data are recognised, against 98% for untrained data.

Downloads

Download data is not yet available.

Article Details

How to Cite
Yaakob, R., Sulaiman, M. N., Mahmod, R., Mohamed, M. T. M., & Ramli, A. R. (1999). Multi-Counterpropagation Network Model For Colour Recognition. Malaysian Journal of Computer Science, 12(1), 38–46. Retrieved from https://mjcs.um.edu.my/index.php/MJCS/article/view/5746
Issue
Vol. 12 No. 1 (1999): Malaysian Journal of Computer Science
Section
Articles