Simultaneous monitoring of moisture and deformation during wood drying using integrated hyperspectral imaging and digital image correlation

The anatomical heterogeneity of wood, particularly the structural difference between earlywood and latewood, leads to non-uniform moisture transport and differential shrinkage, often resulting in drying-induced defects. To elucidate the complex coupling mechanism between moisture distribution and mechanical deformation, this study establishes an integrated monitoring framework combining Hyperspectral Imaging (HSI) and Digital Image Correlation (DIC). A critical technical challenge, the optical interference of conventional carbon-based DIC speckle patterns with NIR moisture analysis—was resolved by utilizing a screened NIR-transparent ink. This novel patterning approach enabled high-contrast strain tracking in the visible spectrum without perturbing the underlying spectral moisture signal. Experiments conducted on Japanese cypress ( Chamaecyparis obtusa ) demonstrated that the PLS-calibrated HSI model achieved robust predictive accuracy (R 2 > 0.92, RMSE = 0.072) on independent specimens. The co-registered spatiotemporal analysis revealed that macroscopic deformation is governed by non-linear coupling mechanisms: while latewood exhibited lower initial moisture content (MC) and higher shrinkage, earlywood regions experienced transient tensile strain during the early drying stage due to asynchronous shrinkage onset. Furthermore, a "strain inversion" phenomenon was observed in later drying stages, driven by the cumulative moisture loss history relative to the fiber saturation point. This framework is intended to serve as a methodological bridge between microscale wood anatomy and macroscale drying behavior, providing a mechanistic basis for understanding drying-induced defect formation at the growth-ring scale and offering a foundation for future work aimed at optimizing drying schedules for larger-dimension timber.