Morphology-based macro-scale finite-element timber models

This paper presents a non-invasive technique that can extract an accurate geometrical description of growth layer surfaces in wood. The method has been validated for sawn spruce elements (Picea Abies Karst.). The aim is to implement a procedure to model domain geometry in the numerical analysis of wooden elements, taking into account the intrinsic variability of the material. The approach presented by the authors avoids internal imaging and achieves a digital 3D model of growth layers, using, as input data, images of the ring pattern, which represents the growth surface boundary curves, visible on all the cut faces of the wooden element.

Research highlights► Morphological macro-scale FEM models account for wood anisotropy and heterogeneity ► Wood growth layers are modeled from the images of the ring texture ► Detection of ring curves is done using mathematical morphology ► Ring curves are modeled as Non rational non periodic cubic B-spline ► Growth surfaces are modeled as filled surfaces from their boundaries curves