Prediction of Soil Available Boron Content in Visible-Near-Infrared Hyperspectral Based on Different Preprocessing Transformations and Characteristic Wavelengths Modeling

Table 1

Pretreatment methods utilized for the visible-near-infrared ray spectroscopy of collected soil samples.

Pretreatment methods	Abbreviations
Reflection spectrum without pretreatment method	RS
Dislodge tendency	DT
First derivative	FD
Second derivative	SD
Mean center	MC
Logarithmic transformation	LG
First derivative with logarithmic transformation	LG + FD
Second derivative with logarithmic transformation	LG + SD
Multiplicative scatter correction	MSC
First derivative with multiplicative scatter correction	MSC + FD
Second derivative with multiplicative scatter correction	MSC + SD
Standard normal variate	SNV
Dislodge tendency with standard normal variate	SNV + DT
First derivative with standard normal variate	SNV + FD
Second derivative with standard normal variate	SNV + SD
Savitzky–Golay	SG
Dislodge tendency with Savitzky–Golay	SG + DT
First derivative with Savitzky–Golay	SG + FD
Second derivative with Savitzky–Golay	SG + SD
Mean center with Savitzky–Golay	SG + MC
Logarithmic transformation with Savitzky–Golay	SG + LG
First derivative with logarithmic transformation and Savitzky–Golay	SG + LG + FD
Second derivative with logarithmic transformation and Savitzky–Golay	SG + LG + SD
Multiplicative scatter correction with Savitzky–Golay	SG + MSC
First derivative with multiplicative scatter correction and Savitzky–Golay	SG + MSC + FD
Second derivative with multiplicative scatter correction and Savitzky–Golay	SG + MSC + SD
Standard normal variate with Savitzky–Golay	SG + SNV
Dislodge tendency with standard normal variate and Savitzky–Golay	SG + SNV + DT
First derivative with standard normal variate and Savitzky–Golay	SG + SNV + FD
Second derivative with standard normal variate and Savitzky–Golay	SG + SNV + SD