Skip to main content

Table 6 The prediction results of kiwifruit SSC 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

0.9564

0.2645

0.7800

0.5903

CARS

0.9644

0.2364

0.9317

0.3290

SPA

0.6383

0.7581

0.6007

0.7864

UVE

0.7981

0.5702

0.7456

0.6308

SVR

None

0.9152

0.3649

0.7548

0.6257

CARS

0.9514

0.2709

0.9170

0.3651

SPA

0.6378

0.7558

0.5641

0.8309

UVE

0.8539

0.4848

0.7137

0.6582

  1. Bold text indicates the optimal model structure