Modeling Knot Geometry in Norway Spruce from Industrial CT Images

For spruce (Picea abies (L.) Karst.), as with most other species, the value of a log without major defects like decay or compression wood is to great extent determined by its knot structure. That is the reason why sawyers, aiming at the optimal utilization of a log, were naturally interested in the knots, their number, position, diameter and length. X-ray computer tomography (CT) scanning is able to detect internal density variations like knots. To use this information for a computer based optimization process, it is necessary to model different defects by geometric descriptions which are easy to handle. The purpose of this study is to show how to model knots in spruce logs from non-destructive measurements using industrial CT and automatic image analysis processes. A comparison between the results of these methods and the results of a destructive method gives an experimental evaluation of our approach.