Table 2 Results of different models for each prediction of moisture.

From: Prediction of Moisture Content for Congou Black Tea Withering Leaves Using Image Features and Nonlinear Method

Methods

NPC

Calibration set

Prediction set

Rc

RMSEC

Bias

Rp

RMSEP

Bias

SEP

CV

RPD

PLS

5

0.8983

0.0821

0.0417

0.8349

0.0607

0.0262

0.0073

0.1086

0.9834

BN

5

0.9644

0.0352

0.0035

0.5968

0.0918

0.0005

0.0169

0.2026

1.1743

BP-ANN

4

0.9522

0.0364

0.0117

0.8949

0.0513

0.0198

0.0093

0.1381

1.6205

SVM

4

0.9838

0.0239

0.0011

0.9314

0.0411

0.0185

0.0091

0.1365

1.8004

RF

4

0.9858

0.0464

0.0003

0.9172

0.0472

0.0240

0.0121

0.1390

1.6379

  1. BN, Bayesian Network; SD, standard deviation; NPC, used latent variables; RMSEC, root mean square error of calibration; RMSEP: root mean square error of prediction; SEP, standard error of prediction; CV, coefficient of variation; RPD, residual predictive deviation value of prediction.
Back to article page