The overall goal of this research project is to develop a prototype scanning system that can automatically identify important edging/trimming defects (knots, wane, decay, and voids) on rough hardwood lumber and can recommend optimal edge and trim cuts to achieve maximum lumber value for each board. To accomplish this goal we will complete 3 major tasks: (1) install and integrate scanning hardware, including camera, lasers, and a materials handling system, (2) develop computer-based machine vision software to automatically identify rough lumber defects from scanned images, and (3) develop optimal edging and trimming software that incorporates results from machine vision defect analysis. We will collect rough lumber images that contain gray-scale, grain, and profile information and we will develop defect recognition software to fuse information from those different images.

A commercial version of our prototype could be readily incorporated into sawmill operations by adding a marking station that physically places edge and trim lines on each board. Then, existing edging and trimming stations could line up their saws with those markings. This type of vision application would be relatively inexpensive for sawmills and would allow these small, rural business operations to realize lumber value gains of up to 30% while using existing log resources.

Daniel L. Schmoldt, A. Lynn Abbott (VA Tech), Matthew F. Winn, and Philip A. Araman

Accomplishments

• A materials handling system, camera, lasers, and computer hardware/software have been coupled to allow us to capture images of rough lumber.
• Software has been written to capture profile, gray-scale, and wood-pore information simultaneously.

Current and Future Work

• Finalize scanning parameters to ensure that high-quality and consistent images are captured.
• Develop defect recognition software and optimal edging/trimming software.
• Integrate imaging, recognition, and processing decisions.

Technology Transfer Efforts

• Demonstrate completed prototype to industry representatives in cooperation with Virginia Center for Innovative Technology.