Researchers in Northeast China have demonstrated a new approach using drone-mounted multispectral imaging to monitor and predict soybean bacterial blight disease, offering a promising tool for early detection and yield protection.
Aerial Spectroscopy Offers New Hope for Battling Soybean Blight
In an ambitious effort to counter soybean bacterial blight disease—a growing threat to crop yields in China—a team of scientists has successfully combined drone-mounted multispectral imaging with ground-based spectroscopy to monitor and predict disease severity in soybean fields. The study, published in Agronomy, provides compelling evidence that this cutting-edge technique could advance disease management in agriculture (1).
Conducted in 2022, the research focused on soybean fields in Northeast China, where rising temperatures and humidity have increased the incidence and severity of Pseudomonas savastanoi pv. glycinea infection. This bacterium causes characteristic leaf spots that progress from green and water-soaked to reddish-brown necrosis, ultimately reducing photosynthetic activity and soybean yield (1).
Drone with spectroscopy reveals hidden threats to soybean crops in China
© Та -chronicles-stock.adobe.com
The study was led by Weishi Meng, Xiaoshuang Li, Jing Zhang, Tianhao Pei, and Jiahuan Zhang, from the College of Plant Protection at Jilin Agricultural University, Changbaishan Customs, and the Key Laboratory of Soybean Disease and Pest Control, Ministry of Agriculture and Rural Affairs in China (1).
Multispectral Monitoring: A High-Tech Perspective
The core of the team's approach was to integrate aerial drone surveys using multispectral imaging (450 nm to 850 nm) with ground truth data, such as chlorophyll content index (CCI) and direct observations of disease grade. The researchers discovered a strong negative correlation between CCI and disease severity, confirming that chlorophyll levels drop as disease progresses. This relationship allowed them to indirectly estimate disease grade from spectroscopic data (1).
The green normalized difference vegetation index (GNDVI), derived from drone-mounted multispectral sensors, emerged as a particularly effective metric. As soybean plants became more diseased, their GNDVI values declined, reflecting reduced chlorophyll activity and plant vigor. This allowed researchers to classify disease stages more accurately than conventional visual inspection (1).
“The GNDVI replaces the red band in traditional NDVI with the green band,” the authors noted, “enhancing sensitivity to chlorophyll and making it more effective in monitoring canopy density and late-stage disease progression” (1).
Data Science Meets Agronomy
Beyond disease classification, the study leveraged random forest regression to model soybean yield loss based on GNDVI data. The model consistently outperformed traditional regression approaches, especially on key observation dates such as September 7, when yield estimation was most accurate. The greatest recorded yield loss was 8.6%, occurring when disease severity reached grade 4 on the researchers’ scale (1).
The team’s integration of field-based inoculation, drone imaging at 30 meters, and regression modeling marks a significant advancement in real-time, precision agriculture. Despite the high cost of low-altitude imaging, the researchers proposed a two-tiered approach for future applications: broad scans from high altitudes to locate potential outbreak zones, followed by detailed imaging closer to the crop canopy (1).
Challenges and Future Directions
While the study showed promise, it wasn’t without limitations. Due to environmental variability and the difficulty of capturing fully healthy (grade 0) and severely infected (grades 7–9) samples, the dataset was somewhat skewed. The authors addressed this by artificially inoculating plants to simulate a wide range of disease grades (1).
Color-rendered multispectral images provided reasonable discrimination between disease grades 1 to 6, although detection at lower severity levels was more difficult due to subtle color differences. Calibration methods improved classification accuracy, but the researchers acknowledged the need for further refinement in image processing and disease grading algorithms (1).
A Blueprint for Precision Plant Protection
This study establishes a foundation for broader implementation of drone-based, spectroscopic health and disease monitoring in crops (1–3). “This method provides a certain theoretical foundation and reference value for predicting and managing soybean bacterial blight disease,” the authors wrote. With further refinements in spectral calibration and sampling design, the approach could be scaled for use across China's soybean-producing regions—and potentially around the world (1).
References
(1) Meng, W.; Li, X.; Zhang, J.; Pei, T.; Zhang, J. Monitoring of Soybean Bacterial Blight Disease Using Drone-Mounted Multispectral Imaging: A Case Study in Northeast China. Agronomy 2025, 15 (4), 921. DOI: 10.3390/agronomy15040921
(2) Tanaka, T. S. T.; Wang, S.; Jørgensen, J. R.; Gentili, M.; Vidal, A. Z.; Mortensen, A. K.; Acharya, B. S.; Beck, B. D.; Gislum, R. Review of Crop Phenotyping in Field Plot Experiments Using UAV-Mounted Sensors and Algorithms. Drones 2024, 8 (6), 212. DOI: 10.3390/drones8060212
(3) Wang, Y.; An, J.; Shao, M.; Wu, J.; Zhou, D.; Yao, X.; Zhang, X.; Cao, W.; Jiang, C.; Zhu, Y. A Comprehensive Review of Proximal Spectral Sensing Devices and Diagnostic Equipment for Field Crop Growth Monitoring. Precis. Agric. 2025, 26 (3), 1–38. DOI: 10.1007/s11119-025-10251-3
Advancing Deep Soil Moisture Monitoring with AI-Powered Spectroscopy Drones
July 7th 2025A Virginia Tech study has combined drone-mounted NIR hyperspectral imaging (400 nm to 1100 nm) and AI to estimate soil moisture at root depths with remarkable accuracy, paving the way for smarter irrigation and resilient farming.
AI Boosts SERS for Next Generation Biomedical Breakthroughs
July 2nd 2025Researchers from Shanghai Jiao Tong University are harnessing artificial intelligence to elevate surface-enhanced Raman spectroscopy (SERS) for highly sensitive, multiplexed biomedical analysis, enabling faster diagnostics, imaging, and personalized treatments.
Artificial Intelligence Accelerates Molecular Vibration Analysis, Study Finds
July 1st 2025A new review led by researchers from MIT and Oak Ridge National Laboratory outlines how artificial intelligence (AI) is transforming the study of molecular vibrations and phonons, making spectroscopic analysis faster, more accurate, and more accessible.
Advancing Deep Soil Moisture Monitoring with AI-Powered Spectroscopy Drones
July 7th 2025A Virginia Tech study has combined drone-mounted NIR hyperspectral imaging (400 nm to 1100 nm) and AI to estimate soil moisture at root depths with remarkable accuracy, paving the way for smarter irrigation and resilient farming.
AI Boosts SERS for Next Generation Biomedical Breakthroughs
July 2nd 2025Researchers from Shanghai Jiao Tong University are harnessing artificial intelligence to elevate surface-enhanced Raman spectroscopy (SERS) for highly sensitive, multiplexed biomedical analysis, enabling faster diagnostics, imaging, and personalized treatments.
Artificial Intelligence Accelerates Molecular Vibration Analysis, Study Finds
July 1st 2025A new review led by researchers from MIT and Oak Ridge National Laboratory outlines how artificial intelligence (AI) is transforming the study of molecular vibrations and phonons, making spectroscopic analysis faster, more accurate, and more accessible.
2 Commerce Drive
Cranbury, NJ 08512
