Table 4 Presents the percentage change in performance metrics between the indices prediction model and the combined indices and canopy temperature model.

From: Novel indices and multi-source data fusion for monitoring plant moisture stress in winter wheat fields

Moisture types

ML models

RSI

NDSI

Published indices

R2

RMSE

MAE

R2

RMSE

MAE

R2

RMSE

MAE

LMC

RF

0.190

-0.296

2.547

16.426

-15.808

-7.812

5.384

-8.520

-2.669

PLSR

-44.678

30.654

17.975

2.209

-1.649

-10.349

5.761

2.156

-0.752

SVM

3.338

-8.143

-1.677

5.149

-7.313

-4.634

5.307

-7.314

-11.430

ANN

-27.653

31.825

18.491

6.816

-3.301

-3.215

-8.116

5.409

14.163

PMC

RF

-0.559

1.239

-0.507

1.055

-2.267

-6.836

2.258

-4.886

-5.181

PLSR

0.805

-1.692

2.289

0.319

-0.754

1.842

-2.500

5.981

-3.702

SVM

13.820

-15.890

-18.327

0.948

-2.302

1.880

4.736

-6.289

-10.683

ANN

0.132

-0.372

-2.498

-0.248

0.669

0.402

4.632

-11.848

-25.015

  1. Note: positive values indicate a percentage increase, while negative values indicate a decrease.