1：Experimental Design and Method Selection:

An NIR hyperspectral imaging camera with a focused halogen light source was used to capture spatially resolved reflectance images. Methods included histogram plotting, variance calculation, PCA for dimension reduction, and QDA for classification with five-fold cross-validation.

2：Sample Selection and Data Sources:

Fifteen wood species (five softwoods and ten hardwoods) were commercially purchased, with three air-dried samples per species prepared to dimensions of 12 mm (T) × 30 mm (L) × 80 mm (R). Only earlywood areas were studied.

3：List of Experimental Equipment and Materials:

NIR-HSI camera (Compovision, Sumitomo Electric Industries, Ltd.), focused halogen light source (? 1 mm), white reference plate, dark image cap, microtome for slicing, safranin solution for dyeing, bright-field microscope (PrimoStar, Carl Zeiss Microscopy Co.), and software (Matlab, The Mathworks, Inc.).

4：Experimental Procedures and Operational Workflow:

Samples were scanned line-by-line in push-broom manner with the camera and light source set at specific angles. Reflectance images were converted using a standardization equation. Histogram plots and variance calculations were performed on the images, followed by PCA on variance data and QDA classification.

5：Data Analysis Methods:

Variance calculation from histogram frequencies, PCA for dimension reduction, and QDA with five-fold cross-validation for species identification accuracy assessment.