Skip to main content

Table 5 The prediction results of kiwifruit firmness values by PLSR and SVR models using Savitzky–Golay 1st derivative spectral data and different wavelength selections

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

Prediction model

Wavelength selection

\({R}_{Cal}^{2}\)

\({RMSE}_{Cal}\)

\({R}_{CV}^{2}\)

\({RMSE}_{CV}\)

PLSR

None

0.9691

0.3739

0.9363

0.6940

CARS

0.9879

0.2761

0.9723

0.4211

SPA

0.9164

0.7266

0.8595

0.9416

UVE

0.8985

0.8038

0.8709

0.9109

SVR

None

0.9532

0.4318

0.9105

0.7687

CARS

0.9698

0.3723

0.9616

0.4868

SPA

0.9171

0.7245

0.8753

0.8852

UVE

0.9218

0.7009

0.8583

0.8947

  1. Bold text indicates the optimal model structure