Scanning timber for strength predictions
Abstract
This paper describes the use of neural networks to predict the strength of structural timber on the basis of data obtained from six scanners. These networks were found to be simple to apply, but there are doubts on the robustness of the algorithms for coping with changes in the characteristics of the timber resource. One method suggested for coping with this difficulty is to use a combination of several algorithms, each one set up to cope with a different sub-group of timber.
Fields of Study & Subjects
Wood Treatment and Properties
Neural Networks and Applications
Industrial Vision Systems and Defect Detection
Publication details
- Journal / source
- CRC Press eBooks
- Pages
- 683–688
- Publication date
- 26 February 2026
- Language
- EN
- Document type
- Book chapter