Wood Artificial Intelligence - Knot modelling by Computed Tomography (WAI-KnotCT)

Purpose and goal:

Knots are the most important feature for the quality of structural and appearance sorting of sawn timber. Within the project, we aim to increase the production efficiency of both structural and appearance sawn wood by developing innovative and unique algorithms for detecting knots based on artificial intelligence (AI) and X-ray computed tomography (CT) methods.

Expected results and effects:

Industrial CT scanning of logs is becoming an increasingly widespread tool in sawmill production and has been shown to significantly increase the volume of high-quality sawn timber. Nevertheless, there is great potential for improving its accuracy. The new AI methods will enable more accurate data on the internal knot structure of logs that can be used to assess and optimize forest management and processing processes in the wood industry.

Approach and implementation:

The project is divided into 9 WPs covering different levels of the processing chain; central is the knot structure and linking the tree growth characteristics with the mechanical wood properties (strength and stiffness). Six institutions and companies are involved in the three-year-long project, where collaborative data collections will take place. Knowledge and AI models will be shared between the parties and will be made available after the project.