Communications of the ACM
Turning Old Maps into 3D Digital Models of Lost Neighborhoods
The researchers used 13 Sanborn maps produced in 1961 of two adjacent neighborhoods on the near-east side of Columbus, OH. Machine learning techniques were able to extract data from the maps and create digital models.
Credit: Center for Urban and Regional Analysis/Ohio State University
Researchers at Ohio State University (OSU), the Mid-Ohio Regional Planning Commission, and Chicago-based marketing solutions provider Epsilon have converted old Sanborn Fire Insurance maps into three-dimensional digital models of historic neighborhoods with a new machine learning (ML) technique.
OSU's Yue Lin created ML tools that derive details about individual buildings—including location, construction materials, and primary use—from the maps.
The team used the method to digitally reconstruct two mostly demolished neighborhoods on the near east side of Columbus, OH, based on 13 Sanborn maps drafted in 1961.
Analysis validated the ML model's replication accuracy, with the digital version's building footprints and construction materials matching those in the maps by roughly 90%.
From Ohio State News
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA
No entries found