Identification and Disease Index Inversion of Wheat Stripe Rust and Wheat Leaf Rust Based on Hyperspectral Data at Canopy Level

Table 4

The results of disease index inversion SVR models of wheat stripe rust and wheat leaf rust.

Model	Wheat disease	Optimal parameters		Training set		Testing set
				R^2	MSE		MSE
Model 6	Wheat stripe rust	32	0.03125	0.8895	0.0068	0.6750	0.0193
Model 7	Wheat leaf rust	4	0.03125	0.8435	0.0104	0.8099	0.0130
Model 8	Wheat stripe rust	32	0.03125	0.8942	0.0062	0.6597	0.0205
Model 9	Wheat leaf rust	4	0.03125	0.8479	0.0101	0.8068	0.0131
Model 10	Wheat stripe rust	16	0.2500	0.8623	0.0085	0.6250	0.0229
Model 11	Wheat leaf rust	4	0.1250	0.8461	0.0095	0.7982	0.0125

Note. Model 6 was the DI inversion SVR model of wheat stripe rust built based on the spectra with the disease indices greater than 0 acquired in the diseased period of stripe rust in the first group of spectral features. Model 7 was the DI inversion SVR model of wheat leaf rust built based on the spectra with the disease indices greater than 0 acquired in the diseased period of leaf rust in the first group of spectral features. Model 8 was the DI inversion SVR model of wheat stripe rust built based on the spectra with the disease indices greater than 0 acquired in the diseased period of stripe rust in the second group of spectral features. Model 9 was the DI inversion SVR model of wheat leaf rust built based on the spectra with the disease indices greater than 0 acquired in the diseased period of leaf rust in the second group of spectral features. Model 10 was the DI inversion SVR model of wheat stripe rust built based on the spectra with the disease indices greater than 0 acquired in the diseased period of stripe rust in the third group of spectral features. And Model 11 was the DI inversion SVR model of wheat leaf rust built based on the spectra with the disease indices greater than 0 acquired in the diseased period of leaf rust in the third group of spectral features.