Kennesaw State student develops crop monitoring system to detect plant disease

Kennesaw State University computer science major Brendan Miller is developing a low-cost crop monitoring system designed to help farmers detect plant disease before it spreads. The research, conducted through KSU’s Summer Undergraduate Research Program with information technology assistant professor Xu Tao, addresses a global agricultural challenge: plant pests and diseases destroy up to 40% of crops worldwide each year, according to the U.N. Food and Agriculture Organization.

The system uses edge AI, allowing models to process data directly on local hardware instead of relying on remote servers or data centers. A microcontroller gathers readings for a small computer, including pH levels, water temperature, electrical conductivity, humidity, and light intensity. The device can compare those signals, identify unusual conditions, and alert farmers even in areas with limited power or internet access.

Early testing has shown progress. Miller trained an edge AI vision model running entirely on a Raspberry Pi computer to distinguish healthy lettuce leaves from leaves affected by bacterial or fungal diseases, achieving about 87% accuracy during testing. Future work will focus on detecting risk factors before visible symptoms appear, while improving performance on hardware with limited memory and processing power.