A Preliminary Study of Wood Species Classifacation System Based on Wood Knot Texture Using K-Nearest Neighbour With Optimized Features From Binary Magnetic Optimization Algorithm Selection

Osman, Khairuddin and Mohamad, Syahrul Hisham and Jaafar, Hazriq Izzuan (2013) A Preliminary Study of Wood Species Classifacation System Based on Wood Knot Texture Using K-Nearest Neighbour With Optimized Features From Binary Magnetic Optimization Algorithm Selection. In: Science & Engineering Technology National Conference 2013, 3-4 July 2013, Kuala Lumpur, Malaysia..

PDF 2013_Conf._3July_SETNC2013_(2).pdf Restricted to Registered users only Download (1MB)

Abstract

The Classification of wood type can be done by studying the texture of the wood knot. This paper proposed a classification of wood species using k-Nearest Neighbour with optimized features. The images of wood knots are taken from Universiti Teknologi Malaysia’s CAIRO Wood Database. The database consists of 25 species of tropical woods. The features of the wood knot images are extracted using Gray Level Co-Occurrence Matrix. Binary Magnetic Optimization Algorithm is use to optimize the feature selection process. Binary Magnetic Optimization Alogirthm also use to optimize parameters of k-Nearest Neighbour and Gray Level Co-Occurrence Matrix. The result indicates that the proposed approach can perform as good as previous literature with fewer features used as input for the classifier.

Item Type:	Conference or Workshop Item (Paper)
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Faculty of Electronics and Computer Engineering > Department of Industrial Electronics
Depositing User:	HAZRIQ IZZUAN JAAFAR
Date Deposited:	13 Jan 2014 04:25
Last Modified:	28 May 2015 04:12
URI:	http://eprints.utem.edu.my/id/eprint/10656
Statistic Details:	View Download Statistic

Actions (login required)

View Item