HOME › BLOGS

Meet the 10 Sub Grantees Leading the Way to Securing Africa’s Food Systems

April 11, 2023

Agriculture and food systems are critical to the livelihoods of millions of people in Africa, playing a vital role in ensuring food security and economic development. To harness the power of artificial intelligence (AI) for sustainable agricultural practices in Africa, the AI4D Africa Hub for Agriculture and Food Systems has selected 10 outstanding sub grantees to join their initiative. These sub grantees, chosen from across the continent, represent a diverse range of innovative projects that are leveraging AI to address challenges in agriculture and food systems, with the ultimate goal of improving the lives of farmers and communities in Africa.

In this article, we are excited to introduce you to these 10 sub grantees and showcase the groundbreaking work they are doing in the field of agricultural AI. These projects are at the forefront of cutting-edge technologies that have the potential to revolutionise agriculture and food systems in Africa. With a strong focus on sustainability, inclusivity, and responsible AI practices, these sub grantees are paving the way for a more resilient and prosperous future for African farmers and communities.

Project 1: Monitoring and Artificial Intelligence Tools for Smart Agriculture in Cape Verde

Location: Cape Verde

Project Lead: Dr. Sonia Semedo, University of Cabo Verde

Grant Amount: US$50,000

Agriculture has always been a vital aspect of Cape Verde’s economy and livelihood, but recent cycles of drought have had a direct impact on agricultural production. As a solution to these challenges, this project has been initiated to explore the feasibility of integrating emerging technologies such as IoT and AI into agriculture, with a focus on small scale farmers who practice subsistence farming. It is hoped that this project will help them improve their yields and water management practices. The project’s main goal is to create an approach for agriculture that is suitable for Cape Verde and can increase production levels while creating mechanisms for managing soil health and water resources. By working closely with small farmers, the project aims to develop a more sustainable and resilient agricultural sector that can withstand the effects of climate change while improving food security.

Project 2: Development of Machine Learning Model for Crop Pests and Diseases Diagnosis Based on Crop Imagery Data

Location: Tanzania

Project Lead: Dr. Hudson Laizer, Mbeya University of Science and Technology

Grant Amount: US$51,000

This project aims to develop a Machine Learning model to diagnose crop pests and diseases using crop imagery data. Common beans and Irish potatoes are vital food and cash crops for smallholder farmers in Tanzania, but yields are often low due to diseases such as Bean rust, Bean anthracnose, Early blight, and Late blight. The project will collect over 120,000 crop leaf images to develop a mobile application that can detect these diseases early and help farmers manage their spread. The Machine Learning model will be trained on localized datasets to accurately detect crop diseases and will be used by farmers and agricultural extension officers. The project aims to alleviate food security problems in Africa and contribute to responsible AI practices and carbon footprint reporting.

Project 3: Enhancing Farm-scale Crop Yield Prediction using Machine Learning for Internet of Agro –Things in Tanzania

Location: Tanzania

Project Lead: Dr. Barakabitze Alcardo Alex, Sokoine University of Agriculture

Grant Amount: US$51,000

By analyzing satellite vegetation indices, climate data, and other relevant data sources, the project aims to develop farm-scale crop yield prediction models that will provide recommendations and predictions to local farmers early in the season and at the end of the season. This project is unique in Tanzania as it considers different data sources for different crops and regions, ensuring tailored and accurate information for smallholder farmers. The agricultural sector in Tanzania, which contributes significantly to the country’s GDP and employs a large portion of the workforce, stands to benefit from the implementation of this project, helping to modernize the industry and improve crop yields for smallholder farmers

Project 4: Using Artificial Intelligence to enhance the Production, Management and Marketing of Nsukka Yellow Pepper(Capsicum Chinense Nsukkadrilus)

Location: Nigeria

Project Lead: Engr Prof. Chinenye Anyadike

Grant Amount: US$59,000

The Nsukka yellow pepper is a popular variety of pepper in Nigeria, but its production faces challenges such as depleted soil nutrients, water demand due to climate change, pest attacks, and limited market access. To address these issues the team will embark on a project that will leverage AI for early pest detection, soil nutrient loss detection, smart irrigation, and e-extension services to support farmers and improve market access. The AI system will be developed using Python programming language and FastAI library, deployed on Raspberry PI with calibrated sensors. The project will also prioritize gender equality and inclusion, empowering women, youths, and vulnerable groups in using modern technology in farming practices. Expected outputs include training manuals, crop datasets for machine learning, soil nutrient and smart irrigation system, and an e-extension service application for farmers’ support. The project aims to promote synergy between government, private sector, and academia, utilize AI for water conservation, generate datasets for local machine learning, and support farmers in marketing their products.

Project 5:  Scaling Smartphone-Based Tools for Early Crop Diseases Detection & Monitoring

Location: Uganda

Project Lead: Dr. Owomugisha Godliver of Busitema University

Grant Amount: US$55,000

This research aims to leverage low-cost smartphones with embedded assisted technologies for early crop disease detection and monitoring. The project will focus on cassava, a staple food crop in Africa, and aims to scale existing diagnostic tools to create a system that can be run by non-experts using smartphones. The project will utilise a novel approach of identifying diseases in plants using non-symptomatic disease data acquired with a spectrometer device. The outcomes of this research could pave the way for similar tools for other crops, benefiting food security in Sub-Saharan Africa.

Project 6: Pest Occurrence Early Warning System and Diagnostic Tool for Tomato Leaf Miner and Whiteflies in Kenya

Location: Kenya

Project Leader: Dr. Hilda Manzi, Geospatial Research International

Grant Amount: US$ 52,000.00

The Kenya Horticulture sub-sector, which contributes significantly to the agricultural GDP, has been impacted by climate change, leading to increased pest pressure on crops like tomato. Tomato production is hampered by pests such as tomato leaf miner and whiteflies, causing significant losses. This research project aims to develop an AI-based spatial tool that utilizes remote sensing and artificial intelligence to monitor and control these pests in Machakos County, Kenya. The tool will provide real-time monitoring of pest occurrence and support early identification and control measures. It will be implemented among smallholder farmers through a business model involving youth and women for e-extension services, with a focus on providing solutions to users with limited pest control knowledge.

Project 7: Empowering Smallholder Farmers (SHF) in Busia County using Low-Cost IoT and AI Tools

Location: Kenya

Project Leader: Dr. Betsy Muriithi from Strathmore University

Grant Amount: US$ 53,000.

This project aims to empower smallholder farmers in Busia County, Kenya, by utilizing low-cost Internet of Things (IoT)and Artificial Intelligence (AI) tools. Smallholder farmers, who produce more than 70% of the food consumed in Kenya, lack access to localized productivity maximizing data and struggle to mitigate the impacts of climate change. The project will deploy affordable mini-weather stations in Busia county to collect and aggregate weather data, which will be made accessible to farmers through a platform synced with existing information from delegated bodies. The goal is to enhance crop yield and support sustainable agriculture, ultimately ensuring food security in a county with high food insecurity.

Project 8: Building the artificial intelligence (AI) for soil moisture and nutrient monitoring under irrigated agriculture among smallholder farmers, academic and agriculture experts in Malawi

Location: Malawi

Project Leader: Dr. Isaac Fandika, Department of Agricultural Research Services (DARS)

Grant Amount: US$ 54,000.

The project aims to build artificial intelligence (AI) tools for soil moisture and nutrient monitoring in irrigated agriculture among smallholder farmers in Malawi. The project, seeks to institutionalize the deployment and scaling of innovative digital solutions for improved agriculture and food availability. It will involve the development and deployment of sensor technology for soil water and nutrient monitoring in 30 irrigation schemes, covering 660 farmers. Public-private partnerships will be established to develop and distribute the AI tools, and farmers, extension workers, and data collectors will be trained to use the tools. The project aims to benefit a total of 2,660 farmers, with a focus on women and youth, by promoting smart irrigation practices, increasing productivity, and creating job opportunities. The long-term goals include promoting sustainable agriculture, achieving food availability and stability, and contributing to the Sustainable Development Goal 2 of zero hunger.

Project 9: TOLBI AI, an AI-based digital tool for smart, sustainable, and efficient agriculture

Location: Senegal

Project Leader: Mouhamadou Kebe, Corniche Ouest, Dakar, Senegal

Grant Amount: US$ 48,000

TOLBI AI is an AI-based digital tool aimed at promoting smart, sustainable, and efficient agriculture in Senegal. The project will use a combination of Artificial Intelligence (AI), satellite images, and local languages to provide real-time information on yield forecasts and field management practices to small-scale agriculture producers and inform national agricultural policies. The expected results of the project include reducing post-harvest losses by 60-80% through an AI-based decision-making system, increasing production and income of farmers by 30% through optimized agricultural inputs and better plant health management, and impacting 80,000 farmers (50.43% women) by 2023 in Senegal and 1 Million farmers by 2025 across Africa. The goal of TOLBI AI is to contribute significantly to food security in Africa, with a focus on Senegal, by leveraging the power of Artificial Intelligence in agriculture.

Project 10: Detection of Crop Pests and Diseases on Web and Mobile Devices using Deep Learning.

Location: Ghana

Project Leader: Dr. Patrick Mensah, Department of Computer Science, University of Energy and Natural Resources 

Grant Amount: US$ 49,000

This project aims to develop a deep learning-based mobile and web app for detecting crop pests and diseases in Ghana. The University of Energy and Natural Resources, in collaboration with Ghana Developing Communities Association and DIGILECT SYSTEM, will train deep learning models on Google Collab using images of healthy and sick plants of cassava, maize, tomatoes, and cashew. The trained models will be embedded in a mobile app for Android and iOS devices using the Tensor Flow lite framework. The app will allow users to capture or scan plant images using their phone’s camera, and provide instantaneous results with a probability value for detection certainty. For uncertain outputs, users will be alerted to seek expert clarification. The app will be user-friendly with text-to-voice facility in English and the local language “Twi” for accessibility by visually impaired and illiterate farmers. The models will prioritize privacy and security, with frequent updates. The app will not require internet connectivity for detecting plant pests and diseases to address low internet penetration. To ensure sustainability and scale-up, e-kiosks will be set up in five communities to provide services to illiterate farmers, visually impaired individuals, and those without phones and internet connectivity at a subsidized fee. The e-kiosks will be managed by women and disabled persons to promote gender equality and inclusion. The Ministry of Agriculture, district assemblies, partner NGO, and SME will be involved in disseminating information about the app to farmers.

We are excited to see the lasting and positive impact from these project in the coming years as they make significant strides in harnessing the power of responsible AI for the betterment of agriculture and food systems in Africa.