Pests and diseases destroy 20 to 40% of crops worldwide. Farmers in developing countries are often not sure what exactly is hurting their fields. Decreased crop yields and loss of income often contribute to famine.
Helping smallholder farmers detect and manage crop pests and diseases can increase yields and incomes, but given the variety and intensity of pest and disease threats, this is a major undertaking.
To help overcome this problem, scientists are using digital technologies, recently adding potato and sweet potato diseases to the PlantVillage Nuru diagnostic application.
Nuru (meaning “light” in Swahili) is a free smartphone app that provides real-time plant disease or pest diagnosis in the field, even in areas off mobile networks. It was developed by scientists from the University Pennsylvania and several CGIAR centers and is part of PlantVillage – a farmer support platform that combines artificial intelligence and satellite technology.
Nuru was launched in Africa in 2018 with the ability to diagnose two cassava diseases (CMD and CBSD). It was subsequently expanded in cooperation with the Food and Agriculture Organization of the United Nations (FAO).
Scientists from International Potato Center (CIP) and the State of Pennsylvania have collaborated to develop artificial intelligence (AI) for Nuru to diagnose yam and potato late blight virus infections. Sweet Potato Virus Diagnosis has been added to the Google Play version of the app since March 2022, while Late Blight Diagnosis was added at the end of 2021.
Potatoes and sweet potatoes are becoming increasingly important food crops in Africa, where they are grown by small farmers who usually suffer crop losses due to disease.
“By giving farmers access to rapid diagnosis of the most serious diseases affecting some of Africa's most important crops, this innovation contributes to food security and livelihoods,” says Jan Kreize, Head of CIP's Crop Production and Systems Science Division.
About 30 types of viruses can infect sweet potato plants. They are spread by whiteflies, aphids and with planting material. Late blight, a rapidly spreading wind-borne disease, can destroy 60% or more of a potato crop in a few rainy weeks if left untreated.
Yield Tools
The AI that powers Nuru's diagnostic capabilities was developed using tens of thousands of photographs of diseased and healthy plants and improved through repeated field assessments.
In sweet potatoes, viral infections are often asymptomatic, and noticeable symptoms differ from one variety to another. To allow Nuru to diagnose viral infections in plants with few or no symptoms, the researchers used diagnostic tests to identify asymptomatic infected plants in hopes of teaching Nuru's AI to detect symptoms that even experts might not notice.
When Nuru diagnoses a disease, it connects farmers to information and management guidance on the PlantVillage website.
Since smartphones are still relatively rare in rural areas, the leader of each group of 20-100 farmers is given the opportunity to use the app. In 2021, CIP distributed 240 smartphones to seed producers and farm group leaders in Kenya and Tanzania.
“Nuru has helped farmers a lot,” Kreiser says. “They are learning more about late blight, how to prevent it and how to deal with the disease, and they are improving potato production.”