Home > News > How Artificial Intelligence enhances the potential of satellite data    
  • 18.05.2021
  • Data for development
  • Local innovation

How Artificial Intelligence enhances the potential of satellite data    

Sentinel-2 image shows wheat fields and other crops in Beauce (France). The different colours indicate different growth stages of the various crops. Remote sensing can be used to make statements about the type and condition of crops.

Combining the wealth of open Earth observation data with AI-based analysis tools can contribute to many development challenges including sustainable resource production, climate protection and food security. Already today, AI is used to analyze satellite and fields data to predict crop yield in small-holder farms in Uganda, calculate the impact of transportation programs in India, and map land use patterns in Vietnam.   

Allowing for a regular bird’s-eye view, satellite imagery can be a crucial monitoring tool capable of providing important insights. Yet, it would be difficult to analyze the amount of satellite data manually. The advantage of AI is that it can process a lot of information and data at the same time. However, according to Dr. Hanifa Nabuuma from Makerere University in Uganda, a key stumbling block for leveraging this technology in the African context is the lack of access to training datasets that fit to the local context. 

To address this issue and better leverage the potential of geospatial data, FAIR Forward in collaboration with the Radiant Earth Foundation and the Makerere University has organized a virtual “Geospatial Machine Learning” bootcamp for professionals from Uganda, Rwanda, Ghana and South Africa. 40 learners are trained on how to use machine learning techniques to leverage geospatial data for sustainable development. After the end of the training, they are equipped to pass on their newly acquired knowledge within their organizations.  

In doing so, this collaboration also strengthens the community of Earth observation and AI experts in Africa to more efficiently use the potential of geospatial data for local challenges such as climate change, urbanization and food security. 



Radiant Earth: https://www.mlhub.earth/  

FAIR Forward: https://toolkit-digitalisierung.de/en/fair-forward/