Discriminant Analyzing System for Wood Wastes Using a Visible—Near-Infrared Chemometric Imaging Technique

Abstract

A new optical system was developed and applied to automated separation of wood wastes, using a combined technique of visible–near-infrared (Vis-NIR) imaging analysis and chemometrics. Three kinds of typical wood wastes were used, i.e., non-treated, impregnated, and plastic-film overlaid wood. The classification model based on soft independent modeling of class analogy (SIMCA) was examined using the difference luminance brightness of a sample. Our newly developed system showed a good/promising performance in separation of wood wastes, with an average rate of correct separation of 89%. Hence, it is concluded that the system is efficiently feasible for online monitoring and separation of wood wastes in recycling mills.