Yao-Yi Chiang Leverages AI to Analyze Maps for $500K NSF Grant
Department of Computer Science & Engineering associate professor Yao-Yi Chiang is the principal investigator on the University of Minnesota led Imola (Intelligent Map recOgnition LAb) project, which recently received a $500K grant from the National Science Foundation (NSF). Professor Stefan Leyk from the University of Colorado Boulder’s is also a principal investigator on the project. The project titled, “"Collaborative Research: HNDS-I: Building Long-term, National-scale Spatiotemporal Data Collections from Historical Map Archives", will develop advanced computational methods and scientific approaches to extract historical geographic information from scanned maps into a database that allows researchers to study the evolution of transportation networks and their implications on human’s interaction with the environment.
“We have been building AI methods for extracting information from historical maps for about 20 years,” said Chiang. “This project focuses on road networks, because there are no other comprehensive resources for historical road networks in the United States. People can only look at the maps, but systematically analyzing and studying these data is still not possible.”
Chiang and Leyk will work to develop the intelligent system called DaVinci to automate the extraction process and produce a large spatiotemporal database of historical road networks. Over the next three years, they will continue to develop this new technology and analyze tens of thousands of maps.
"DaVinci will autonomously interpret maps like a human by analyzing geographic features and datasets to automatically extract historical road networks,” said Chiang. “This method eliminates the need for large, manually curated training datasets. The project will also include a case study and tutorials to demonstrate how the extracted data can be utilized in scientific research, enhancing our understanding of human-environment interactions."