Monitoring of drought stress and transpiration rate using proximal thermal and hyperspectral imaging in an indoor automated plant phenotyping platform

Table 4 Prediction accuracy of transpiration rate prediction models that use one thermal infrared index

RMSE	R²	MAPE	Selected Variables	Method
Linear: single TIR index
0.56	0.55	41	T_dry-T_p,T_BS, VPD_out, interaction_{Tdry-Tp_VPDout}	Linear: index + env
0.57	0.52	39	CWSI_{Ta_out}, VPD_out, PAR_gz, T_BS	Linear: index + env
0.60	0.57	37	CTD_out, VPD_out, PAR_gz	Linear: index + env
0.53	0.59	35	CWSI_{Ta_out},T_BS, WBI	Linear: Index + env + hyperspectral index
0.51	0.62	33	CTD_out, PAR_gz,, RMP_1483/1430, WBI	Linear: Index + env + hyperspectral index

Median test RMSE, R₂ and MAPE were calculated from 100 bootstrap samples. The data used to create these models consisted of TIR indices (index), environmental data (env), and hyperspectral indices (Table 1)

ISSN: 1746-4811