This technology is a machine learning system that can use images taken by an unmanned vehicle to screen crops for disease.
Unmet Need: Efficient and accurate screening for disease in crops
Detection of crop-destroying diseases currently requires time-consuming direct observation and can vary in accuracy and precision, depending on the level of expertise of the observer. The development of inexpensive drones may provide a method to survey crops for sign of disease; however, there is still a need for a method that efficiently processes the collected data.
The Technology: Automated plant disease screening system using low-cost drones
This technology is a system for mass screening of plant disease in a crop that requires minimal operator involvement. In this method, an unmanned aerial or ground vehicle travels around a field, periodically taking images of individual plants in a crop. These images are then analyzed by software, which has been trained to identify lesions associated with certain plant diseases, providing information about the presence and location of disease in the crop. As such, this technology could streamline plant disease identification and management for large commercial farming operations, reducing costs and enhancing food security.
This technology has been validated on drone-acquired test images screening for northern leaf blight in a maize crop.
Applications:
- Screen for the presence of disease in plants
- Isolation of location and origin of disease in a crop
- Automated and targeted treatment of disease
- Surveying of phenotypes for crop selection and breeding
Advantages:
- Large-scale plant disease screening
- Automated analysis algorithm
- Does not require a trained plant diagnostics expert
- High precision and accuracy
- Time efficient
- Can be further trained to user’s discretion
- Customizable to the detection of user-specified phenotype
Lead Inventor:
Hod Lipson, Ph.D.
Patent Information:
Patent Status
Related Publications:
Wiesner-Hanks T, Wu H, Stewart E, DeChant C, Kaczmar N, Lipson H, Gore MA, Nelson RJ. “Millimeter-level plant disease detection from aerial photographs via deep learning and crowdsourced data” Front Plant Sci. 2019 Dec; 10: 1550.
Wu H, Wiesner-Hanks T, Stewart EL, DeChant C, Kaczmar N, Gore MA, Nelson RJ, Lipson H. “Autonomous detection of plant disease symptoms directly from aerial imagery” Plant Phenome J. 2019 Nov; 2(1): 1-9.
Stewart EL, Wiesner-Hanks T, Kaczmar N, DeChant C, Wu H, Lipson H, Nelson RJ, Gore MA. “Quantitative phenotyping of Northern Leaf Blight in UAV images using deep learning” Remote Sens. 2019 Jul; 11(19): 2209.
Wiesner-Hanks T, Stewart EL, Kaczmar N, DeChant C, Wu H, Nelson RJ, Lipson H, Gore MA. “Image set for deep learning: field images of maize annotated with disease symptoms” BMC Res Notes. 2018 Jul; 11: 440.
DeChant C, Wiesner-Hanks T, Chen S, Stewart EL, Yosinski J, Gore MA, Nelson RJ, Lipson H. “Automated identification of Northern Leaf Blight-infected maize plants from field imagery using deep learning” Phytopathology. 2017 Nov; 107(11): 1426-1432.
Tech Ventures Reference: