期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2014
卷号：63
期号：3
出版社：Journal of Theoretical and Applied
摘要：Knots are common wood defects. A knot is a specific imperfection in timber, reducing its strength which can be exploited for artistic effect, resulting in knot selection being an important matter in the wood industry. The value of wood is related to its quality, and this in turn is determined by defect numbers and distribution. This is challenging as in some instances, selection/classification is manual. In this paper, it is proposed to detect and classify the knots in timber boards. Hilbert transforms and Gabor filters are used for pre-processing the image of knots. The features obtained from preprocessing were classified using Multi Layer Perceptron (MLP) and Neural Network (NN) with Particle Swarm Optimization (PSO) and Invasive Weed Optimization (IWO) for momentum and learning rate.
