Share Email Print

Proceedings Paper

Automation of wood mechanical grading: coupling of vision and microwave devices
Author(s): Denise Choffel
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Wood industry is confronted with problems induced from the raw material itself. Indeed, presenting a lot of growth and sawing up characteristics, with such a variability, wood pieces require unitary inspection. The grading rules recommended by the standards are based on aesthetic and/or physical criterions depending on the foreseen use. But they often have subjective applications. We are focusing here on the mechanical grading, that is the assignment of a class to a sawn timber according to the estimation of three physical and mechanical properties, and the presence of singularities such as knots or wane. A two-sensor approach has ben used for this in-line inspection: a vision device for surface analysis and a microwave device for depth measurements. One linear camera gives profiles so as to calculate timber dimensions and main singularities detection. A microwave sensor with sixteen receptors provides an image of the wood internal structure, from which it is possible to detect knots and to estimate the mechanical properties. Each system computes its own treatment then the result are merged. On the first hand, the fusion produces - the singularities localization by the comparison: redundancy and complementary - the estimation of strength, stiffness and density by the means of multivariate models. On the other hand, the grading rules are translated into computer form: each class is described by its boundaries so as to define which class the current timber belongs to. The paper present the methodology from the rough signals to the class decision.

Paper Details

Date Published: 27 August 1999
PDF: 8 pages
Proc. SPIE 3836, Machine Vision Systems for Inspection and Metrology VIII, (27 August 1999); doi: 10.1117/12.360263
Show Author Affiliations
Denise Choffel, CRAN and ENSTIB (France)

Published in SPIE Proceedings Vol. 3836:
Machine Vision Systems for Inspection and Metrology VIII
John W. V. Miller; Susan Snell Solomon; Bruce G. Batchelor, Editor(s)

© SPIE. Terms of Use
Back to Top