Name | inX
Systems
Jaakko Vuorilehto |
Contact
|
Lintulammenkatu
6,
FIN-04250 Kerava Finland |
Phone | +358 9 41 333 111 |
Fax | +358 9 41 333 151 |
inx@inx.fi, jaakko.vuorilehto@inx.fi | |
URL | www.inx.fi |
Application | machine vision and learning software to improve the efficiency of sawmills |
Product type | inspection system |
Deployed technology | |
Industrial sector | sawmill industry |
Cognitive vision aspect | self-training |
Description
Project in the EUTIST-IMV
(www.spt.fi/eutist/) cluster, EUTIST Wood, KNOT
Previous studies have
indicated that there is significant promise for automatic knot and wane
detection. Optical measurement of knots and wane with black-and-white and
colour CCD-technology has improved significantly during recent years.
KNOT integrates automation,
optical and other measurement technologies. It introduces a completely
new concept for a longitudinal control unit. The KNOT control system uses
a true-colour line-scan camera and matrix camera technologies. The system
is remotely operated and fully automatic.
The laboratory prototype
system with a 3-colour line-scan, 3-side measuring capability and a specially
designed 2-dimensional array camera has been tested and proven to be functional.
The objective of further work is to develop a real-time colour vision and control system to measure the features of a board from three board faces as it passes longitudinally through the measuring unit immediately before entering an edger saw. Data stored on different features would be used to determine the best sawing pattern for a rough board in order to optimise its sawing. After this determination, the system would send control signals to the saw. This work innovatively couples self-training grading and analysis software with three-sided measurement in the edger saw and the combined use of knot and wane data. An essential design goal is that the system can be integrated into new and existing saws.
Further Information
http://hui01.bh.spt.fi/Eutist/eutist.nsf/5915fe1028d76d13c2256a6f0024903f/991760d1ab47e559c2256a6f0025a3fc?OpenDocument