TABLE 4.

The accuracy rates of classification models based on fusion data.

Cultivar	Features	Variables	Model	Tr (%)	Val (%)	Te (%)	Model	Tr (%)	Val (%)	Te (%)	Model	Tr (%)	Val (%)	Te (%)
Zhefujing83	HSI-MIR	7,858	SVM	100	95.00	84.62	LR	100	100	86.54	CNN	100	95.00	84.62
	HSI-LIBS	22,426		100	100	76.92		100	100	69.23		100	100	90.38
	MIR-LIBS	29,504		100	100	86.54		100	100	80.77		100	100	86.54
	HSI-MIR-LIBS	29,894		100	100	90.38		100	100	82.69		100	100	92.31
AD516	HSI-MIR	7,858		100	100	92.19		100	100	100		100	95.00	85.94
	HSI-LIBS	22,426		100	100	81.25		100	100	90.63		100	100	84.38
	MIR-LIBS	29,504		100	100	84.38		100	100	92.19		100	100	84.38
	HSI-MIR-LIBS	29,894		100	100	85.94		100	100	92.19		100	100	81.25

Other abbreviations are similar to this fashion. Tr, training set; Val, validation set; Te, test set; HSI-MIR, the concatenation of full HSI and MIR spectra; HSI-LIBS, the concatenation of full HSI and LIBS spectra; MIR-LIBS, the concatenation of full MIR and LIBS spectra; HSI-MIR-LIBS, the concatenation of full HSI, MIR, and LIBS spectra.