By Topic

Estimating the Strength of Boards Using Mixed Signals of MOE and X-Ray Images

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Saravi, A. ; Dept. of Wood Sci., Univ. of British Columbia, Vancouver, BC ; Lawrence, P.D. ; Lam, F.

The most accurate way of identifying the strength of lumber requires destructive testing, which is clearly not useful for the production of lumber. An intelligent mechanics-based lumber grading system was developed to nondestructively provide better estimation of the strength of a board. This system processed X-ray-extracted geometric features (of 1080 boards that eventually underwent destructive strength testing) by using a physical model of the lumber based on finite-element methods (FEMs) to generate associated stress fields. The stress fields were then fed to a feature-extracting processor, which produced one strength-predicting feature. The modulus of elasticity (MOE) profiles were separately processed, and another feature was extracted based on the minimum point in the MOE averaged profile, with 15% of the data cut from each end. Then, the two MOE and X-ray extracted features were combined (with four different algorithms) into a single feature to estimate the strength of the boards. By applying four different algorithms to a database of more than 1000 boards, to estimate the strength of the boards, coefficient of determination r 2 values of 0.64, 0.65, 0.65, and 0.65 were achieved for the different algorithms, respectively. The results were improved by dividing the database into two sets (based on the dates that the two batches were delivered), and r 2 values of 0.69, 0.71, 0.71, and 0.71 were achieved for the different algorithms, respectively.

Published in:

Instrumentation and Measurement, IEEE Transactions on  (Volume:57 ,  Issue: 11 )