Plywood is composed of several thin layers of wood bonded together by glue. The adhesive integrity of the glue formulation employed must surpass the structural integrity of the wood species within a given panel of plywood. The American Plywood Association (APA) regularly tests the plywood produced at various plywood manufacturing plants to ensure that this particular performance requirement is consistently met. One of the procedures used by the APA to test this requirement consists of 1) milling a plywood panel to be tested into small rectangular blocks called samples, 2) conditioning these samples with various treatments to simulate natural aging, 3) shearing each sample into two halves, and 4) estimating the percent wood failure (as opposed to glue failure) produced by the shear by visually inspecting these sample halves. A region of solid wood or a region of wood fibers embedded in glue on the shear of a sample half is a region of wood failure while a region of glue is a region of glue failure. If the wood failure of samples from a significant number of panels is too low, the right to use APA trademarks is withdrawn from the plant where the sampling occurred. Since measurements obtained by human visual inspection can contain inaccuracies due to fatigue, boredom, state of mind, etc., an automatic vision system to determine percent wood failure is proposed. The method presented is a refinement of the method developed by McMillin and is divided into three tasks. The first task is to locate the area of shear on a given sample half. The second task is to distinguish the areas of wood from the areas of glue on the shear of a sample half. Solid wood is distinguished from glue based on the difference in gray level intensity that exists between solid wood and glue. Wood fiber is distinguished from glue based on the difference in texture, i.e., edge patterns, that exists between fiber and glue. The third task is to compare the areas of shear on the two sample halves comprising a sample to determine the percent wood failure of the sample. / Master of Science
| Identifer | oai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/46460 |
| Date | 30 December 2008 |
| Creators | Avent, R. Richard |
| Contributors | Electrical Engineering, Conners, Richard W., Abbott, A. Lynn, Kline, D. Earl |
| Publisher | Virginia Tech |
| Source Sets | Virginia Tech Theses and Dissertation |
| Language | English |
| Detected Language | English |
| Type | Thesis, Text |
| Format | xiv, 115 leaves, BTD, application/pdf, application/pdf |
| Rights | In Copyright, http://rightsstatements.org/vocab/InC/1.0/ |
| Relation | OCLC# 23612885, LD5655.V855_1990.A987.pdf |
Page generated in 0.0033 seconds