INFERLINK CORP — Department of Defense STTR Phase II: AF19B-T006

Description

Critical minerals are essential components in many modern technologies used by modern society and the global economy. Finding new sources of critical materials depends on having accurate data regarding the geology of potential sites. The United States Geological Survey (USGS) has an extensive collection of geological maps containing detailed information about various geological features, such as rock formations, faults, folds, and mineral deposits. However, most of these maps exist only as scanned images, requiring significant expert effort to manually convert the information into an analytic-ready format. The objective of this Phase II effort is to research, develop, and evaluate state-of-the-art machine learning algorithms to help automate critical mineral assessments. Specifically, the algorithms will include automated map georeferencing to accurately geolocate maps of unknown locations and automated map feature extraction to identify point, line, and polygon features in geological maps.  The expected outcome will be an end-to-end prototype system, called AIM (AI for Maps), for automated georeferencing and feature extraction from scanned geological maps.