Implementation of a mechanics based system for estimating the strength of a board using mixed signals of MOE and x-ray images

TitleImplementation of a mechanics based system for estimating the strength of a board using mixed signals of MOE and x-ray images
Publication TypeConference Paper
Year of Publication2003
AuthorsSaravi, A., P. D. Lawrence, and F. Lam
Conference NameCommunications, Computers and signal Processing, 2003. PACRim. 2003 IEEE Pacific Rim Conference on
Pagination413 - 417 vol.1
Date Publishedaug.
Keywords4900 mm, 89 mm, board strength estimation, elastic moduli, feature extraction, feature-extracting-processor, fluid flow analogy, grip length, intelligent mechanics-based lumber grading system, knot size, low-pass filter, low-pass filters, lumber curvature, lumber grading systems accuracy, lumber strength, mechanical strength, mechanics based system, modulus of elasticity mixed signals, signal detection, threshold filter, vibration effect compensation, x-ray images, X-ray imaging

The most accurate way of identifying the strength of lumber requires destructive testing which is clearly not useful for production of lumber. An intelligent mechanics-based lumber grading system was developed to provide a better estimation of the strength of a board nondestructively. In this study a mechanics-based system was implemented to estimate the strength of a board, using only one combined feature extracted from MOE (modulus of elasticity) profiles and x-ray images. The x-ray image analysis involved extracting the useful parts of the image and compensating for the effect of vibration. After that, the image was passed through a directional low-pass filter to reduce the noise. Furthermore, the image was resized by interpolation in such a way that the size of the signal was the same as the real size of the board, which is 89[mm] 4900 [mm]. The image was passed through a threshold filter to separate the knots based on the fact that the denser knots produce "high hills" in the x-ray image. Finally, information on all the knots such as geometry and location were detected from the threshold image. The knot size and location were fed to an FEM processor to generate the physical model and the associated stress field. In this study, simulating grain direction by analogy to fluid flow and reorienting the element coordinate system along the flow line direction generated the slope of grain. The stress fields were then fed to a feature-extracting-processor which produced one strength predicting feature. A coefficient of determination of 0.4158 was reached using x-ray images alone. The MOE part of the system uses output of CLT machine which contains top and bottom profiles. Due to lumber curvature, one profile may be higher than the other one. By averaging the two profiles this effect will be compensated. Since the grip length for tension tests was 15% of beginning part and end part of each profile, these parts were discarded. The minimum value of the remaining part was the base for calculating the strength. A coefficient of determination of 0.5805 was achieved using MOE alone. Then, the two MOE and x-ray extracted features were combined to a single feature to estimate the strength of the boards. By applying the described algorithm to a database of more than 1000 b- oards to estimate the strength, a coefficient of determination of 0.6417 was achieved. The results show a way to improve the accuracy of lumber grading systems using combined signals.


a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2020 The University of British Columbia