Wood species classification based on wood porosity distribution of macroscopic image.

Sugiarto, Bambang and Suaib, Norhaida and Mohd Rahim, Mohd Shafry and Prakasa, Esa and Fairus Ismail, Npr Anita and Mustafa Albakri, Ikmal Faiq Albakri (2024) Wood species classification based on wood porosity distribution of macroscopic image. In: Digest of Technical Papers - IEEE International Conference on Consumer Electronics.

Text Wood Species Classification Based on Wood Porosity Distribution of Macroscopic Image.pdf Download (973kB)

Abstract

In the context of illegal logging, it is crucial to identify the wood species that are being traded. Wood species classification plays a vital role in inspecting the movement of wood in and out of an area. Furthermore, wood classification is particularly significant to determine the species of the wood test sample to obtain high accuracy. Therefore, the development of automatic wood species classification based on computer vision is highly required. Each wood species possesses unique anatomical characteristics, such as its vessel or pore structure. This research aims to propose an automatic algorithm that can classify wood species based on the porosity distribution of the macroscopic image. This research utilized wood images from six different wood species. The Contour method is used to identify pores first and then many features will be extracted. It will detect the pore size and extract pore features such as the number of pores, size, standard deviation, skewness, kurtosis, and density. By using the macroscopic image of the six wood species, the result showed that 300 data could be extracted and forming a model based on pore features. The accuracy of the wood classification was evaluated using the Random Forest classifier, yielding an accuracy rate of 92.00%. This indicates that the wood porosity distribution features can be effectively utilized for wood species classification.

Item Type:	Conference or Workshop Item (Paper)
Divisions:	Faculty of Information and Communication Technology
Depositing User:	NURHASHIRAH BORHAN
Date Deposited:	31 Jul 2026 07:45
Last Modified:	31 Jul 2026 07:45
URI:	http://eprints.utem.edu.my/id/eprint/30128
Statistic Details:	View Download Statistic

Actions (login required)

View Item