Application of near-infrared spectroscopy for moisture-based sorting of green hem-fir timber

A rapid, non-destructive, in-line method suitable for sorting green hem-fir timbers (115-mm square) based on moisture content was established by near-infrared (NIR) spectroscopy. The accuracy of NIR sorting was compared with a commercial capacitance-type moisture meter. Mixedspecies samples consisting of three moisture classes were assessed in this study. The NIR-based prediction model showed positive correlation with the actual calculated values as determined by oven-drying, regardless of knots, surface roughness, and the mix of two wood species. NIR proved to be capable of detecting the moisture content between all pairs of the three moisture groups, whereas the capacitance-type moisture meter failed to establish a significant difference between middle- and high-moisture groups. These findings clearly demonstrate that NIR spectroscopy has a potential to estimate average moisture of green timber indirectly, although it inherently gives only surface moisture content values, as it is limited by scan depth.