NASA-Funded Scientist Uses EO Imagery and AI to Improve Agriculture in Uganda

Republished from Devex

NASA Harvest

Catherine Nakalembe, the Africa program director for NASA Harvest, leads efforts to map crop conditions and build early warning systems for weather events by developing tools like maps, dashboards, apps and radio to make satellite insights accessible and useful for local farmers and policy-makers across Eastern and Southern Africa.

The programme is executed with local partners, policymakers and researchers to develop tools best suited to the local farmers and increase agricultural production.

One of NASA Harvest’s projects is Helmets Labeling Crops, a ground data collection effort underway in Kenya, Mali, Rwanda, Tanzania, and Uganda, which involves taking pictures of fields from cameras mounted on motorcycle helmets or cars. The ground data is then used to analyse satellite data to accurately assess food insecurity and climate change. A related NASA Harvest effort called Street2Sat transforms these images into large datasets of georeferenced labels, with information on location and crop type. This data trains algorithms to recognise specific crops like maize or sugarcane, parse the photos to predict which crops are shown, and then turn that data into crop type maps and other tools for individual farmers or national crop monitoring initiatives.

According to Catherine Nakalembe, Assistant Professor at the University of Maryland, more investment is needed to ensure partners across Africa can leverage earth observation, ground data, and artificial intelligence to improve food security, and despite growing interest in satellite imagery as a tool for addressing food security, there isn’t sufficient donor funding to ensure regions like East and southern Africa can benefit.

The NASA Harvest programme is a food security and agriculture project that monitors crops from space and uses a combination of satellite imagery and data from the ground to help farmers and policymakers on the continent make more informed decisions.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.