Spectral image database for observing the quality of Nordic sawn timbers

In this study, a comprehensive spectral image database of Nordic sawn timbers for public use was measured. Economically significant Finnish wood species birch (Betula sp.), Norway spruce (Picea abies) and Scots pine (Pinus sylvestris) were chosen for inclusion in the database. The total of samples was 107 containing heartwood, sapwood, decayed wood, blue stain, mold, resin, early wood, late wood, knots, cracks, pith, reaction wood and bark. Board and crosscut samples were measured in frozen, melted and room-dried conditions. The reflectance of samples was measured over a 300- to 2,500-nm wavelength range. Additionally, the photoluminescence of samples excited by an ultraviolet B light source was measured. The spot size used was 250 μm with an 80 mm \(\times \) 200 mm imaging area, and produced all in all 44 million spectra. In this paper, examples of the possibilities of this spectral image database as a means of detection of the spatial distribution of aromatic lignin and the moisture content (MC) of nonfrozen timber were introduced and provided. From the results, it was found that it was possible to detect the lignin distribution from spectral images, and simple and robust methods for wood MC estimations were also introduced.