Resin Pocket Enhancement through Anisotropic Diffusion

The increasing demand for wood products with better defined quality parameters has led the wood industry to find improved methods for converting logs into final products. The number and the type of defects (e.g. knots, cracks, resin pockets, wane, pith, and decay) define the quality of lumber. In particular, the presence or absence of these defects determines the visual appearance, constructive strength and, in combination, the quality and price of the product. Computer Tomography (CT) can be used to inspect the internal structure of logs and to determine their interior defects before breakdown. In this paper, we present a method for the localization of resin pockets in CT images, based on anisotropic diffusion filtering. We use this filter to enhance the resin pockets edges, afterwards detected by an edge detector.