The Defense Advanced Research Projects Agency (DARPA) and the US Geological Survey (USGS) are exploring the use of artificial intelligence and machine learning (AI/ML) tools for critical minerals assessments.
In order to solicit innovative solutions, DARPA collaborated with USGS, MITRE, and NASA’s Jet Propulsion Laboratory on an industry competition: The USGS Map Georeferencing Challenge and the Feature Extraction Challenge of the Artificial Intelligence for Critical Mineral Assessment.
At the virtual awards ceremony for the competition held on December 7th, Sanborn earned fourth place and an Honorable Mention. Sanborn successfully demonstrated ways to automate the georeferencing of scanned and raster maps, as well as feature extraction. Starting with a set of 1,000 maps of unknown location and coordinate system, for example, the Sanborn Team, led by Becky Soltanian, VP of Research and Development, developed routines to scan, recognize text and features, and subsequently fit coordinate points referenced to known locations using a variety of tools in an automated process.
“We enjoyed participating in this competition,” said Soltanian, “It allowed the Sanborn Team to demonstrate our skills in applying AI/ML tools to develop an automated workflow, while helping address an important resource mapping challenge facing the federal government.”
“With the large volumes of imagery, lidar and other data available for analysis these days, helping our customers process data more efficiently with AI/ML tools is a key value-added service that Sanborn can provide,” said Soltanian.