Predicting moisture content and density distribution of Scots pine by microwave scanning of sawn timber II: evaluation of models generated on a pixel level
Lundgren, N. L. C., Hagman, O. and Johansson, J.
2006 Journal of Wood Science, 52(1): 1611-4663
Lundgren, N. L. C., Hagman, O. and Johansson, J., (2006), "Predicting moisture content and density distribution of Scots pine by microwave scanning of sawn timber II: evaluation of models generated on a pixel level", Journal of Wood Science, 52(1): 1611-4663.
Abstract The purpose of this study was to use images from a microwave sensor on a pixel level for simultaneous prediction of moisture content and density of wood. The microwave sensor functions as a line-scan camera with a pixel size of 8mm. Boards of Scots pine (Pinus sylvestris), 25 and 50mm thick, were scanned at three different moisture contents. Dry density and moisture content for each pixel were calculated from measurements with a computed tomography scanner. It was possible to create models for prediction of density on a pixel level. Models for prediction of moisture content had to be based on average values over homogeneous regions. Accuracy will be improved if it is possible to make a classification of knots, heartwood, sapwood, etc., and calibrate different models for different types of wood. The limitations of the sensor used are high noise in amplitude measurements and the restriction to one period for phase measurements.
LUNDGREN NILS(Lulea Univ. Technol., Skelleftea, Swe) HAGMAN OLLE(Lulea Univ. Technol., Skelleftea, Swe) JOHANSSON JAN
