Table 4 Prediction modeling results of nicotine content in cigars.

From: Retrieval of nicotine content in cigar leaves by remote analysis of aerial hyperspectral combining machine learning methods

Target

Characteristic band selection

Model

Training sets

Test sets

Rc2

RMSEC

RMSECV

Rp2

RMSEP

RPD

Nicotine

UVE

PLSR

0.749

0.095

0.080

0.673

0.101

1.690

BP

0.762

0.081

0.079

0.744

0.088

1.930

RF

0.882

0.086

0.067

0.681

0.097

1.648

SVM

0.766

0.093

0.102

0.664

0.090

1.705

CARS

PLSR

0.876

0.068

0.067

0.643

0.127

1.673

BP

0.721

0.044

0.063

0.797

0.078

2.182

RF

0.909

0.059

0.059

0.757

0.064

1.945

SVM

0.797

0.087

0.087

0.784

0.078

1.977

SPA

PLSR

0.723

0.090

0.100

0.671

0.097

1.636

BP

0.767

0.094

0.088

0.546

0.117

1.459

RF

0.888

0.065

0.057

0.749

0.085

1.895

SVM

0.758

0.095

0.095

0.733

0.098

1.562