A research team has developed a wheat powdery mildew index (WPMI) that uses hyperspectral imagery from unmanned aerial vehicles (UAVs) to detect, quantify, and track the spread of a destructive fungal disease in winter wheat. The index, detailed in a study published in the Journal of Remote Sensing on April 10, 2026, combines leaf-level spectroscopy, ground canopy measurements, and UAV-based data with spatial hot-spot analysis to identify infected areas and monitor disease progression or recovery over time.
Wheat powdery mildew (WPM) is a major fungal disease that damages leaf tissues, weakens plant growth, and can cause severe yield losses. Current field diagnosis relies heavily on expert visual inspection, which is labor-intensive, subjective, and difficult to scale. While hyperspectral remote sensing has shown promise for crop disease detection, many existing vegetation indices were designed for general stress monitoring rather than specific pathogen-host responses. Machine-learning methods often require large, high-quality training datasets. The researchers aimed to develop a specific, stable, and scalable method for WPM detection from ground to UAV scales.
The study, led by scientists from the Key Lab of Smart Agriculture System at China Agricultural University, the Information Technology Research Center at the Beijing Academy of Agriculture and Forestry Sciences, the Institute of Plant Protection at the Chinese Academy of Agricultural Sciences, and the College of Land Science and Technology at China Agricultural University, focused on rapid, field-scale monitoring of WPM in smallholder farms where disease spread is spatially uneven. The team developed two forms of WPMI: WPMIG = (R760 − R554)/(R661 + R554) and WPMIR = (R760 − R661)/(R661 + R554). These indices use disease-sensitive bands in the green, red, and near-infrared regions. Compared with traditional vegetation indices, WPMI more consistently distinguished healthy from infected wheat and better quantified disease index across leaf, ground canopy, and UAV canopy scales.
Data were collected from greenhouse and field experiments from 2022 to 2024, including 1,260 leaf spectra and 804 canopy spectra under various infection conditions, wheat varieties, and spatial scales. At the leaf scale, WPMI achieved the highest overall classification accuracy, reaching 85% and 86% for WPMIG and WPMIR, respectively, in 2022. In field conditions, accuracy ranged from 80% to 81%. For disease severity estimation, WPMIG achieved R² values of 0.55 to 0.93 at the ground scale and 0.48 to 0.90 at the UAV scale. UAV-derived WPMIG maps combined with Getis–Ord Gᵢ* hot-spot analysis identified clusters of likely infection and tracked spatiotemporal changes over three growing seasons.
The researchers noted that a disease-specific spectral index can move crop disease monitoring beyond simple image comparison. By linking UAV hyperspectral imagery with spatial hot-spot analysis, the method reveals where WPM is emerging, expanding, or declining, offering a potential basis for earlier warning and more targeted disease management. Leaf spectra were collected using a handheld hyperspectral camera, while canopy spectra came from a ground spectrometer and a DJI M600 UAV equipped with a Pika L hyperspectral camera. Linear discriminant analysis selected sensitive bands and evaluated classification performance. Disease index was measured through field surveys, and linear regression assessed the relationship between WPMI and disease index.
With further validation across regions, wheat varieties, sensors, and disease conditions, WPMI-based UAV monitoring could support precision plant protection and early warning systems. The approach may help farmers identify disease hot spots before severe outbreaks, reduce unnecessary pesticide use, and improve decision-making. More broadly, this strategy provides a framework for developing disease-specific remote sensing indices for other crop–pathogen systems, contributing to smarter and more resilient agricultural monitoring. The study was supported by the National Key Research and Development Program and other grants. The Journal of Remote Sensing, an online-only Open Access journal published in association with AIR-CAS, promotes remote sensing theory and interdisciplinary research. For more information, visit Chuanlink Innovations.
