Seeing beyond the visible: Estimating soil parameters from hyperspectral images (ICIP 2022)

Associated SPS Event: IEEE ICIP 2022 Grand Challenge

KP Labs, together with ESA (European Space Agency) and partner QZ Solutions, has created an extraordinary challenge, as they will revolutionize the future of farming with the help of in-orbit processing.

Maintaining farm sustainability through improving the agricultural management practices by the usage of recent advances in Earth observation and artificial intelligence has become an important issue nowadays. It can not only help farmers face the challenge of producing food at an affordable price, but can also be crucial step toward the planet-friendly agriculture.

Farmers need timely information about the soil parameters to optimize their fertilization process – this may ultimately lead to selecting a better mix of fertilizers, and to reducing the overall amount of them. The current approach for quantifying soil parameters is very time-consuming, and mostly relying on manual labor– soil samples need to be gathered in the field and mixed, to then be forwarded to specialized labs for further chemical analysis. Also, the number of sampling points in the field is limited and often scattered across large areas, compromising the eventual accuracy of the test results. Overall, the in-situ analysis is not scalable and extremely time-inefficient.

Why not exploit the cutting-edge airborne and satellite hyperspectral imaging technology for more sustainable agriculture, helping to shape a better future for our planet?

The objective of this challenge is to advance the state of the art for soil parameter retrieval from hyperspectral data in view of the upcoming Intuition-1 mission. A campaign took place in March 2021 over agricultural areas in Poland with extensive ground samplings collocated with airborne hyperspectral measurements from imagers mounted onboard an airplane. The hyperspectral data contains 150 contiguous hyperspectral bands (462-942 nm, with a spectral resolution of 3.2 nm), which reflects the spectral range of the hyperspectral imaging sensor deployed on-board Intuition-1.

Intuition-1 is a 6U-class satellite mission designed by KP Labs to observe the Earth using a hyperspectral instrument and an on-board computing unit capable of processing data using artificial intelligence in orbit. It will be the world’s first satellite with a processing power capable of advanced processing of hyperspectral images in orbit. It is due for launch in Q1 2023.

For further details, visit the Challenge page. Contact Jakub Nalepa, Silesian University of Technology and KP Labs, Poland, for more information.