Back to results
US 11,442,006 B2Granted
Method and system for detecting moisture levels in wood products using near infrared imaging and machine learning
AssigneeBoise Cascade Corporation
InventorsDavid Bolton, Jude Richard Peek, Curtis Fennell
Life of this patent
Estimating the term…
Abstract
Near InfraRed NIR technology, including NIR cameras and detectors, and machine learning methods and systems, including one or more Machine Learning (ML) based moisture level detection models, are used to accurately identify moisture content and the specific locations of the moisture on an entire surface of a veneer sheet or other wood product and provide moisture level prediction data for the veneer sheet or other wood product. Based on the moisture level prediction data for a given wood product, one or more actions are taken with respect to wood product to ensure the wood product is put to the most efficient, effective, and valuable use.
Claims20
A method for detecting moisture levels in wood products comprising:
- training one or more machine learning based moisture level detection models using Near InfraRed (NIR) image data for one or more wood products and determined corresponding moisture levels for the one or more wood products;
- providing an NIR analysis station, the NIR analysis station including one or more sources of illumination positioned to illuminate a surface of a wood product, the NIR analysis station including one or more NIR cameras positioned to capture one or more NIR images of the illuminated surface of the wood product;
- positioning a wood product to be analyzed in the NIR analysis station such that a first surface of the wood product to be analyzed is illuminated by the one or more illumination sources;
- capturing, using the one or more NIR cameras, one or more NIR images of the illuminated first surface of the wood product;
- processing the one or more NIR images of the illuminated first surface of the wood product to generate NIR image data for the illuminated first surface of the wood product;
- providing the NIR image data for the illuminated first surface of the wood product to the one or more trained machine learning based moisture level detection models;
- obtaining moisture level prediction data for the wood product from the one or more trained machine learning based moisture level detection models;
- assigning a grade to the wood product based on the moisture level prediction data for the wood product;
- and
- based, at least in part, on the grade assigned to the wood product, taking one or more actions with respect to the wood product.
7 more elements
A system for detecting moisture levels in wood products comprising:
- one or more trained machine learning based moisture level detection models, the one or more trained machine learning based moisture level detection models having been trained using Near InfraRed (NIR) image data for one or more wood products and corresponding determined moisture levels for the one or more wood products;
- an NIR analysis station, the NIR analysis station including one or more sources of illumination positioned to illuminate a surface of a wood product, the NIR analysis station including one or more NIR cameras positioned to capture one or more NIR images of the illuminated surface of the wood product;
- a wood product to be analyzed in the NIR analysis station, the wood product to be analyzed being positioned such that a first surface of the wood product to be analyzed is illuminated by the one or more illumination sources;
- a physical memory, the physical memory including data representing one or more NIR images of the illuminated first surface of the wood product captured using the one or more NIR cameras;
- one or more processors for processing the data representing one or more NIR images of the illuminated first surface of the wood product to generate NIR image data for the illuminated first surface of the wood product;
- moisture level prediction data for the wood product obtained from the one or more trained machine learning based moisture level detection models;
- a grade assignment module for assigning a grade to the wood product based on the moisture level prediction data for the wood product;
- and
- an action activation module for initializing one or more actions with respect to the wood product based, at least in part, on the grade assigned to the wood product.
6 more elements
A method for detecting moisture levels in veneer sheets comprising:
- training one or more machine learning based moisture level detection models using Near InfraRed (NIR) image data for one or more veneer sheets and determined corresponding moisture levels for the one or more veneer sheets;
- providing an NIR analysis station, the NIR analysis station including one or more sources of illumination positioned to illuminate a surface of a veneer sheet, the NIR analysis station including one or more NIR cameras positioned to capture one or more NIR images of the illuminated surface of the veneer sheet;
- positioning a veneer sheet to be analyzed in the NIR analysis station such that a first surface of the veneer sheet to be analyzed is illuminated by the one or more illumination sources;
- capturing, using the one or more NIR cameras, one or more NIR images of the illuminated first surface of the veneer sheet;
- processing the one or more NIR images of the illuminated first surface of the veneer sheet to generate NIR image data for the illuminated first surface of the veneer sheet;
- providing the NIR image data for the illuminated first surface of the veneer sheet to the one or more trained machine learning based moisture level detection models;
- obtaining moisture level prediction data for the veneer sheet from the one or more trained machine learning based moisture level detection models;
- assigning a grade to the veneer sheet based on the moisture level prediction data for the veneer sheet;
- and
- based, at least in part, on the grade assigned to the veneer sheet, taking one or more actions with respect to the veneer sheet.
7 more elements
Dependencies are inferred from the claim text.