ML Model for Crop Pests & Diseases
Photo Credit: Feepik
Overview
Common beans and Irish potatoes are lifeline crops for millions of Tanzanian farming families, yet diseases like bean rust and early blight can devastate harvests before farmers even recognize the threat. With limited access to agricultural extension services, rural communities have long had no reliable, affordable way to identify diseases early enough to act.
Developed through close collaboration with farmer groups and extension officers, the mobile application analyzes leaf imagery to detect crop diseases in real time, delivering personalized management recommendations in Swahili with full offline functionality for communities with limited connectivity. Farmers receive smartphones and hands-on training, ensuring even the most resource-constrained households can participate. Strict consent protocols and privacy safeguards protect participants, while publicly deposited datasets allow other researchers to build on the work without duplicating costly data collection.
Early detection is already helping farmers implement timely interventions that minimize crop losses and protect household incomes. By combining responsible AI deployment with genuine community partnership, the project offers a scalable model for digital agricultural transformation across East Africa, one that keeps smallholder farmers, not just technology, at the center.