Search

This study aimed to identify and assess vegetation indices (VIs) and their optimal band combinations using unmanned aerial vehicle (UAV) hyperspectral imagery for the quantitative inversion of wheat stripe rust. This would offer guidance for selecting rust-resistant phenotypes and facilitate large-scale disease monitoring using aerial and spaceborne remote sensing images. The experimental design encompassed 960 wheat varieties (strains) in agricultural fields. Hyperspectral imagery was acquired at 100-m altitude during different disease stages, and disease index (DI) was investigated per plot. A custom program explored VIs with two, three, and four bands using 30 calculation methods and 3463790 band combinations. Regression models employed threefold cross-validation and multilayer perceptron (MLP) algorithms, with the mean $R^{2}$ value indicating optimal index and band combinations. The results revealed that the chosen VIs were effective in inverting the DI. Selected two-band VIs included MGRVI (531, 571), with $R^{2}$ = 0.746 ± 0.01618; the optimal three-band VIs was ARI2 (531, 550 ± 10, 640 ± 25), with $R^{2}$ = 0.755 ± 0.00896; and the best four-band VIs were DBSI (531, 551, 750, 799), with $R^{2}$ = 0.778 ± 0.01300, which was comparable to full-band modeling ( $R^{2}$ = 0.775 ± 0.01508). The models’ performance improved with an increasing number of bands in the VIs. This study demonstrated that appropriate multi-VIs modeling enhances performance compared to single-VIs modeling, e.g., six combinations of VIs achieved $R^{2}$ = 0.790 ± 0.01141. These findings underscore the potential of integrating machine learning algorithms and VIs for quantifying wheat rust diseases, laying the foundation for developing airborne and spaceborne imaging sensors for large-scale wheat rust monitoring.