An efficient method to reduce grain angle influence on NIR spectra for predicting extractives content from heartwood stem cores of Toona. sinensis

Table 2 Analysis of two PLS regression models (EC and grain angle) using sMC selected spectra variables with SNV+1st derivative preprocessing method on calibration and validation set

	Pre-treatment	Number of variables	Calibration			validation
	Pre-treatment	Number of variables	R²_Cal	RMSE_Cal (%)	LVs	R²_V	RMSE_V (%)
EC	SNV+1st derivative	19	0.84	1.21	5	0.80	1.42
Grain angle	SNV+1st derivative	19	0.36	39	5	0.30	45

R²_Cal The coefficient of determination on calibration, RMSE_Cal root-mean-square error on calibration, R²v The coefficient of determination on validation, RMSE_V root-mean-square error on validation, LVs latent variables

ISSN: 1746-4811