Probing the complexity of wood with computer vision: from pixels to properties

Mirko Lukovic, Laure Ciernik, Gauthier Müller, Dan Beat Kluser, Tuan Pham Huu, Ingo Burgert, Mark Schubert

Abstract: We use data produced by industrial wood grading machines to train a machine learning model for predicting strength-related properties of wood lamellae from colour images of their surfaces. The focus was on samples of Norway spruce (Picea abies) wood, which display visible fibre pattern for- mations on their surfaces. We used a pre-trained machine learning model based on the residual network ResNet50 that we trained with over $15000$ high-definition images labelled with the indicating properties measured by the grading machine. With the help of augmentation techniques, we were able to achieve a coefficient of determination ($R^2$) value of just over $0.9$. Considering the ever-increasing demand for construction-grade wood, we argue that computer vision should be considered a viable option for the automatic sorting and grading of wood lamellae in the future.