Crop Disease Detector
- March 28, 2024
- Godfrey Naman
- 0
Research Problem
- Crop diseases pose major threats to income and food security for smallholder farmers in Africa.
- Traditional diagnosis relies on visual symptoms on leaves, stems, and other plant parts, leading to challenges such as: Late diagnosis beyond recovery, High management and control costs, Significant or total crop loss.
Findings
- The project successfully developed a mobile application that utilizes machine learning models to detect diseases in maize and banana crops based on leaf image data.
- The project created the publicly available datasets for maize and banana crops.
- The machine learning models demonstrated high accuracy in detecting specific diseases affecting maize and banana crops, allowing for early diagnosis and intervention.
- The mobile application was designed with a user-friendly interface, making it accessible and easy to use for smallholder farmers and other stakeholders in agriculture.
Impact
- The application enables farmers to make timely and informed decisions regarding disease management, potentially reducing the spread of diseases and minimizing crop losses.
- The project facilitated knowledge sharing among farmers, agricultural extension officers, and other stakeholders, promoting best practices in disease detection and management.
- The application has the potential for scalability and adaptability to other crops and regions, indicating its usefulness in broader agricultural contexts.
Research Credits
- Team: Dr Neema Mduma, Mr Christian Elinisa & Ms Flavia Mayo
- Partners: Makerere University AI Lab – Uganda, Namibia University of Science and Technology – Namibia & KaraAgro AI Foundation – Ghana
- Funders: Rockefeller Foundation, Google.org, and Canada’s International Development Research Centre (IDRC) through Lacuna Fund in Agriculture; Swedish International Development Cooperation Agency (SIDA) and International Development Research Centre (IDRC) through the African Centre for Technology Studies (ACTS), Data Science Africa (DSA) & GrowFurther
- African Centre for Technology Studies (ACTS) – Dr. Mduma
Announcing Our Inaugural Round of Funding for Agricultural Datasets for AI
Two New Available Agriculture & Language Datasets, 2023 Grantee Convening