The complex structure of timber has traditionally been difficult to model as it is a highly heterogeneous material. The density and material properties for structural species such as Pinus radiata (radiata pine) can vary greatly across the growth rings. Numerical simulation methods are becoming more prevalent as a method of predicting moisture migration, stress and strain distributions, and fungal/rot intrusion in engineered wood products (EWPs). All these applications require a computational mesh that captures the growth ring structure to facilitate an accurate assessment of the performance of EWPs. In this work, a low-cost image-based algorithm is developed for generating a virtual representation of a small cross laminated timber panel sample. Specifically, the proposed method results in a virtual description of an EWP sample comprised of a triangular prismatic mesh where the nodes are aligned on the growth rings of each EWP component, with specific wood material properties allocated to each mesh element. Each small sample is treated individually and we assume there is no longitudinal variation in the density, pith location, and pith angle within the mesh structure. The initial step involves analysing an image of the end grain pattern of a single clear wood sample to identify the growth rings using a spectral clustering algorithm. Next, the centre of the tree (pith) is located through an iterative constrained least-squares algorithm to determine the pith angle. Image analysis of an anatomical image combined with the pith location allows for a constant density value to be assigned to each mesh element. The capability of this framework is then demonstrated by simulating the moisture migration and heat transfer throughout a CLT sample under atmospheric and saturating boundary conditions.