Development of computerized wood veneer colour sorting system for wood industry / Liew Shaer Jin

• Main Author: Liew , Shaer Jin
• Format: Thesis
• Published: 2024
• Subjects: Q Science (General); TA Engineering (General). Civil engineering (General)
• Online Access: http://studentsrepo.um.edu.my/15472/2/Liew_Shaer_Jin.pdf; http://studentsrepo.um.edu.my/15472/1/Liew_Shaer_Jin.pdf; http://studentsrepo.um.edu.my/15472/
• Institution: Universiti Malaya

Wood colour sorting is essential in woodworking to maintain uniformity and consistency in the appearance of the final products, thus, improving consumer satisfaction. Majority of the wood manufacturing companies in Malaysia are depending heavily on manual colour sorting that solely relies on human visual inspection, which can be subjective, inconsistent, laborious, and subject to errors. Automation is a goal, however, the cost for implementation of established technologies is always extortionate especially for small and medium industries (SMI). Therefore, the aim of this research is to develop a computerized vision system to perform colour sorting for multi-scale woodworking facilities. To achieve the research goal, our objectives are set to determine a suitable algorithm for colour features classification, to select the best features which contribute the most in the classification and to compare the effect of different cameras in the performance of the colour sorting. We have compared camera of different genres, namely an industrial camera, a prosumer action camera, and a webcam. Three cameras used were: i) Hikrobot® MV-CE200-10UC (CE200), ii) Logitech® C920 HD Pro (C920), and iii) Sony® RX0 II (RX0 II). After setting up a veneer imaging prototype, a total of 1,289 distinct images of American red oak (Quercus rubra), yellow poplar (Liriodendron tulipifera), and maple (Acer spp.) were acquired from each camera, summing up to 3,867 images from all cameras. After performing image preparations and calibrations, 26 features were extracted from each image. The features were based on the average and standard deviation of the wood basal colour and wood grain colour. Salient features were obtained using Sequential Forward Selection (SFS), which were then used to train a Self-Organizing Map (SOM). The results affirmed that the colour of the basal colour is highly correlated with human sorted colour groups. As expected, CE200 performed the best being of industrial grade. Interestingly, C920 exhibited comparable performance to CE200. RX0 II performed the worst due to its interface software limitations. This proposed system achieved accuracies of 89.0% for red oak, 94.3% for yellow poplar and 96.4% for maple. This research will assist the SMI to develop affordable vision systems for colour sorting.