Hierarchical classification of surface defects on dusty wood boards

Abstract

Classification of surface defects on wood boards is one of the important steps towards a completely automated wood processing plant. This paper presents a hierarchical approach to classify sample boards of red oak into the nine classes, eight types of surface defects and clear wood. It utilizes a priori knowledge about surface defects and their texture properties. Since the boards in a wood processing plant are usually covered by dust, the feasibility of using the proposed approach in such an environment is investigated. Three types of image data including those of dusty board surfaces are generated and utilized in the classification experiments.