Skip to main content

Table 4 The classification results of kiwifruit samples by LDA and SVM models using SNV spectral data and different wavelength selections

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

Classifier

Wavelength selection

\({accuracy}_{CV}\)

\({RMSE}_{CV}\)

\({accuracy}_{EV}\)

\({RMSE}_{EV}\)

LDA

None

93.33

0.1286

93.33

0.1289

CARS

82.00

0.2065

80.00

0.2076

SPA

84.67

0.2038

82.67

0.2061

UVE

90.00

0.1532

89.33

0.1528

SVM

None

80.67

0.2057

77.33

0.2204

CARS

80.00

0.2074

69.33

0.2487

SPA

74.67

0.2297

64.00

0.2631

UVE

76.00

0.2281

64.00

0.2633

  1. Bold text indicates the optimal model structure