DigitalGlobe’s WorldView 2 satellite captured this image of Boston, Massachusetts in July 2014. Photo: DigitalGlobe.

Radiant Solutions announced a contract with the U.S. National Geospatial-Intelligence Agency (NGA) to provide more than 1 million labeled objects within high-resolution satellite images that will be used to accelerate the development of machine learning algorithms that can extract valuable information from imagery at scale.

NGA is partnering with Defense Innovation Unit Experimental (DIUx) to launch the 2018 DIUx xView Detection Challenge to spur innovation that will ultimately support national security and humanitarian missions. With the challenge, individuals and global academic, commercial and research organizations will use 30 cm commercial satellite imagery from DigitalGlobe to train algorithms to automatically identify dozens of objects that are relevant to pressing global challenges, such as stopping the spread of disease and improving infrastructure in the developing world.

The DIUx Detection Challenge follows in the footsteps of earlier training datasets Radiant Solutions released under the SpaceNet and Intelligence Advanced Research Projects Activity (IARPA) Functional map of the World (FMOW) challenges, for which participants applied machine learning to satellite imagery to address a range of important applications. The xView dataset is licensed to NGA under a Creative Commons non-commercial license and contains more than 1 million labeled objects across 60 object classifications, such as damaged buildings, construction equipment and tents.

“Access to satellite imagery at increasingly high spatial and temporal resolutions, coupled with rapid advances in machine learning algorithms for object detection, has created a disruptive opportunity for NGA and other agency customers to see, understand and anticipate humanitarian and defense activities at a global scale,” said Tony Frazier, president of Radiant Solutions. “Our vision is that xView becomes a trusted resource for accelerating open innovation in machine learning.”