Open Machine Learning for Earth Observation (ML4EO)
The availability of satellite image data enables organizations to monitor various aspects of the Earth’s surface, including activities on the ground, such as agricultural practices. However, business leaders and decision-makers are also facing challenges, such as how to analyze and use massive amounts of data (e.g., for land cover mapping of crops) and building the skills needed to work with Earth Observation (EO) data. The advent of machine learning has been a significant development, providing an increasingly useful method for making sense of the vast and diverse amount of EO data.
The intersection of Machine Learning and Earth Observation is an emerging but rapidly growing field. An important first step for Rwanda to become more competitive in the Machine Learning for Earth Observation (ML4EO) space is technical training and capacity building for young professionals. In cooperation with the University of Rwanda Center for Geographical Information Systems (CGIS), GFA developed a curriculum to train junior data scientists in ML4EO methods. The overall objective was to strengthen Rwandan skills in the ML4EO field. In addition, the program enables participants to directly apply their newly-acquired skills by working on relevant ML4EO applications in the agricultural sector.