Multispectral optical satellite data has become an essential source for digital agriculture in recent years. It is a reliable and accessible data source for crop monitoring and variable rate applications globally. However, cloud cover can render services based on optical satellite monitoring inaccessible during periods of unfavourable weather, making them insufficient in areas with high cloud coverage and at certain times.
The AILAI project from World from Space addresses a key shortfall in AI/ML technology for digital agriculture: cloud interference in satellite imaging. Traditional multispectral satellite data, crucial for crop monitoring, is often obstructed by clouds.
“Our solution combines radar and multispectral data, using advanced AI to generate clear, cloud-free images for reliable global crop monitoring, regardless of weather conditions,” says World from Space CEO, Jan Labohý. “Our project enhances AI performance using advanced methods, balancing improvement of accuracy with time of computations and costs. We access extensive Earth observation data from the EU's Copernicus program, ensuring a rich training dataset. Our expertise helps maintain high-quality data labels, essential for accurate AI training in Earth monitoring.”
With the world's population projected to reach 9.7 billion by 2050, and the agricultural sector facing multiple challenges such as food insecurity, climate change, and water scarcity, the World from Space team see precision agriculture as crucial for sustainable food production. “We believe that satellite technology is a game-changer for the sustainability of food production”, says Labohý. “Therefore, we pursue the aim to democratize access to the benefits of satellite monitoring to all farmers around the globe.”
Find out more about World from Space and their work here and follow them on LinkedIn here!
More about the ELISE 2nd Open call and awarded companies is available here.
