Program Overview
Our digital agriculture research program integrates advanced remote sensing, molecular diagnostics, and artificial intelligence to develop scalable solutions for early crop disease detection, weed identification, and precision crop management. By combining drone-based hyperspectral and red-green-blue (RGB) imaging with molecular pathogen validation and machine learning, the program aims to provide accurate, timely, and field-ready decision support for producers.
Research Areas
Predictive Plant Health Analytics
Machine learning models that forecast disease emergence, pest pressure, and yield outcomes, enabling proactive management strategies across prairie agricultural systems.
Precision Agriculture Systems
Data integration platforms that optimize input applications, enhance resource efficiency, and support sustainable intensification of crop production.
Climate-Adaptive Decision Tools
Digital solutions that help producers adapt to climate variability through early warning systems, risk assessment, and management recommendations tailored to local conditions.
Knowledge Translation Platforms
User-centered technologies that translate complex agricultural data into actionable insights for producers, agronomists, and industry stakeholders.
Low-Altitude Drone Imagery Collection
High-quality imagery forms the foundation of our digital agriculture tools. We employ lightweight consumer-grade quadcopters equipped with RGB and hyperspectral sensors to collect ultra-high-resolution canopy imagery from research and producer fields. These compact drones can be transported easily, launched directly from field margins, and operated efficiently without specialized infrastructure, making them practical for both research and on-farm applications.
The resulting low-altitude imagery captures detailed crop canopy characteristics, including subtle spectral and structural changes associated with disease stress. These datasets support the development of computer vision algorithms for disease detection, crop health assessment, phenotyping, and precision management. By integrating aerial imagery with molecular pathogen confirmation and environmental data, the program develops reliable, scalable analytics that can be deployed across diverse cropping systems.
An example of the program in practice: this on-device model names the most likely species among nine common prairie weeds — running entirely in the producer's browser, with no photo ever leaving the device.
Impact
Our research supports Canada's digital agriculture strategy by developing practical solutions that enhance productivity, sustainability, and resilience in prairie farming systems. Through industry partnerships and producer engagement, we ensure our digital innovations address real-world challenges and drive measurable outcomes.