“This photo was taken in August, in the Sing’isi village in Arusha, northern Tanzania, where my colleagues and I were conducting a field visit to farmers. I was demonstrating how to use a mobile app — named KilimoAI — to examine crop leaves. The app, which we’ve developed in-house, works by analysing a photograph of the plant to detect possible disease symptoms.
This is part of my role in the Artificial Intelligence and Complexity Systems group at the Nelson Mandela African Institution of Science and Technology in Arusha. I guide projects that aim to apply artificial intelligence to real-world challenges in agriculture, conservation and development.
To create the app, my colleagues and I took thousands of photos of plant leaves at farms, of both healthy and diseased crops. So far, we’ve focused on diseases affecting maize (corn), beans (Phaseolus spp.), bananas and potatoes.
After collection, images go through a verification process with the help of plant pathologists at the Tanzania Agricultural Research Institute in Arusha. We use this data set to train machine-learning models to distinguish between healthy and diseased plants, and even to classify specific disease types. A portion of the data is set aside for testing, so that we can evaluate the model’s accuracy.
