Mapping schools in Sudan using artificial intelligence
Thanks to a challenge run by Omdena, we now have a model that can identify schools in Sudan from satellite imagery with 95% accuracy.
Thanks to a challenge run by Omdena, we now have a model that can identify schools in Sudan from satellite imagery with 95% accuracy.
To accelerate our school mapping efforts, we collaborated with Mapbox to create a crowdsourced application to identify schools using satellite imagery.
Development Seed and UNICEF use AI to identify unmapped schools across Asia, Africa, and South America, building on our success mapping 23,100 unmapped schools in 8 countries.
Training a machine learning model to identify schools from satellite imagery of developing countries is a difficult task. UNICEF and Tryolabs partnered to understand how UNICEF’s Machine Learning model was actually performing by employing explainable artificial intelligence (XAI) methods.
SIMET data is shared with Giga, which aims to map and increase the number of connected schools around the world.
The project aims to support the universalization of high-speed Internet access and encourage the pedagogical use of digital technologies in basic education policy.
Three-year initiative to identify connectivity gaps in 35 countries is a critical first step in connecting every school to the internet