Non-destructive pre-symptomatic detection of gray mold infection in kiwifruit using hyperspectral data and chemometrics

Table 7 The prediction results of kiwifruit TA values by PLSR and SVR models using Savitzky–Golay 1^st derivative spectral data and different wavelength selections

Prediction model	Wavelength selection	\({R}_{Cal}^{2}\)	\({RMSE}_{Cal}\)	\({R}_{CV}^{2}\)	\({RMSE}_{CV}\)
PLSR	None	0.9619	0.0225	0.8183	0.0515
	CARS	0.9797	0.0164	0.9500	0.0259
	SPA	0.5800	0.0726	0.5714	0.0720
	UVE	0.8072	0.0497	0.7761	0.0537
SVR	None	0.9608	0.0231	0.7982	0.0561
	CARS	0.9642	0.0218	0.9771	0.0317
	SPA	0.5847	0.0745	0.5750	0.0742
	UVE	0.8570	0.0437	0.7812	0.0587

ISSN: 1746-4811