Skip to Main Content
Testing and analysis of internal hardwood log defect prediction modelsAuthor(s): R. Edward Thomas
Source: In: Proceedings of the 3rd international scientific conference on hardwood processing; 2011 Oct. 16-18; Blackburg, VA. Blacksburg, VA: Virginia Polytechnic Institute and State University: 85-94.
Publication Series: Scientific Journal (JRNL)
Station: Northern Research Station
PDF: Download Publication (467.85 KB)
DescriptionThe severity and location of internal defects determine the quality and value of lumber sawn from hardwood logs. Models have been developed to predict the size and position of internal defects based on external defect indicator measurements. These models were shown to predict approximately 80% of all internal knots based on external knot indicators. However, the size of the actual knot differed from the size of its prediction on average by 0.32 inch in length and 1.49 inches in width. Depending on the defect type, the mean absolute error of the prediction model varies from 0.4 to 1.8 inches in length and 0.3 to 0.8 inch in width at the defect cross-sectional size at the midpoint depth. Given the current models and their associated known prediction errors, this paper seeks to identify the effect, if any, these errors would have on the quality and value of lumber sawn from logs whose internal defects are generated using the prediction models. Twenty-six high-resolution laser-scanned logs were digitally sawn and the resulting lumber graded and analyzed to test the models.
- Check the Northern Research Station web site to request a printed copy of this publication.
- Our on-line publications are scanned and captured using Adobe Acrobat.
- During the capture process some typographical errors may occur.
CitationThomas, R. Edward. 2011. Testing and analysis of internal hardwood log defect prediction models. In: Proceedings of the 3rd international scientific conference on hardwood processing; 2011 Oct. 16-18; Blackburg, VA. Blacksburg, VA: Virginia Polytechnic Institute and State University: 85-94.
- Predicting internal red oak (Quercus rubra) log defect features using surface defect defect measurements
- Validation of an internal hardwood log defect prediction model
- Predicting internal yellow-poplar log defect features using surface indicators
XML: View XML