ML4EO – Machine Learning for Earth Observation
The use of the machine learning (ML) method has become widespread in recent years, but its accessibility varies globally. To address this disparity, the German development agency GIZ initiated the ML4EO project. This project aimed to impart knowledge and educate young, educated individuals in Rwanda on geoinformatics, ML, and earth observation. The majority of the project activities took place remotely. Before its commencement, we, along with our partners, traveled to Kigali, Rwanda’s capital, to familiarize ourselves with the project’s stakeholders, including partners, colleagues, and participants. GeoCodis contributed to the project by leveraging its expertise in Earth observation and ML. We developed, reviewed, and organized theoretical instructional materials and practical exercises, as well as consultations for the participants.
The project’s primary goals were to:
- develop an understanding of machine learning and Earth observation,
- explain of the theoretical principles of GIS, remote sensing, algorithms for processing products for agriculture,
- presentation of tools and methods for exploiting EO satellite data,
- promotion of exploitation and use of ML and EO in the field of agriculture,
- supporting business models/minimum viable projects (eg crop type, crop quality and health, land cover classification, etc.) applicable to the Rwandan agricultural sector.
Our partners included the German Agency for International Cooperation (GIZ), the Rwanda Space Agency (RSA), the University of Rwanda, the GFA Group, and the German Space Centre (DLR).
