An algorithm for comparing density in CT-images taken before and after compression of Pinus sylvestris

Ten plain-sawn boards of Scots pine (Pinus sylvestris L.) were compressed semi-isostatically according to the CaLignum process in a Quintus press at 25 C and 140 MPa. X-ray computerised tomography images of the same cross-sections were captured before and after compression and compared using a transformation algorithm. In the algorithm the shape of compressed cross-sections were converted to the same shape as non-compressed using the density variation between pixels. Density after compression and the increase in density were analysed using PLS regression. The regressor variables described position in the cross-section and wood properties of each pixel. Heartwood and other resinous wood were less compressed than sapwood, particularly in boards with high resin content. Density also increased little close to the press table, especially in whorls with large knots. Higher original density gives lower degree of compression but still higher density after compression.