Two Polish companies are working on a technology based on the use of artificial intelligence to remotely estimate soil parameters using satellite pictures. Such a system will allow a more efficient selection of the best places for cultivation and thus reduce the number of fertilizers.
According to the creators of the technology, its application will shorten the waiting time for the results of key research for farmers from 3 weeks to 4 days.
According to the Food and Agriculture Organization of the United Nations (FAO), agricultural production will need to increase by as much as 70 percent by 2050 to feed the Earth’s growing human population.
As the specialists from the Polish company KP Labs point out in their message, one of the answers to this challenge is precision farming, which uses the latest technologies and detailed knowledge, e.g., on soils. It uses automation for planning and decision-making to increase yields and farm efficiency.
For these reasons, KP Labs and QZ Solutions are working on a system for satellite-based soil analysis.
“To emphasize the importance that the discovery of our companies can bring to the further history of agriculture, we decided to name our project Genesis – meaning the beginning. It is intended to mark a turning point in the traditional approach to crop planning and become a tool that will revolutionize this industry in the future,” says Zbigniew Kawalec, CEO of QZ Solutions.
Commissioned by the European Space Agency (ESA), the project consists of a pilot program on soil analysis using hyperspectral data.
“The idea of using hyperspectral imaging supported by artificial neural networks will allow the comparison of material collected on-site – using the traditional method – with that acquired and processed remotely,” the developers of the technology described.
If it turns out that thanks to the technology used it is possible to map parameters such as potassium, magnesium, phosphorus content, or pH level in soil, the next stage of the project will be initiated. It will take place in space, where data processing will take place directly in orbit. The whole process of detecting soil parameters will be automated by algorithms present on board the Intuition-1 satellite.
“The idea of the project is based on using our Intuition-1 satellite for this purpose. Our plans are very ambitious, and in the first instance, piloting ‘on the ground’, we want to check whether it is at all possible to remotely detect soil parameters using machine learning techniques applied to hyperspectral data analysis,” explains Michał Zachara, COO at the KP Labs.
“If successful, we plan to transfer our solution right on board Intuition-1, thus proving that soil parameters can be detected ‘from space’. Genesis will not only contribute to agriculture but also prove that machine learning can deliver key insights from raw hyperspectral data in specialized applications – representing a big step into the future of agriculture,” he adds.