With the help of machine learning and old Sanborn Fire Insurance maps, researchers now have a method for creating 3D digital models of historic neighborhoods.

However, the computerized models will be something other than a curiosity – – they will give specialists an asset to direct investigations that would have been almost unthinkable previously, for example, assessing the monetary misfortune brought about by the destruction of noteworthy areas.

“The story here is we presently can open the abundance of information that is implanted in these Sanborn fire chart books,” said Harvey Mill operator, co-writer of the review and teacher of topography at The Ohio State College.

“It empowers an entirely different way to deal with metropolitan verifiable examination that we would never have envisioned before AI. It alters the game.”

The review was distributed today (June 28, 2023) in the diary PLOS ONE.

This examination starts with the Sanborn maps, which were made to permit fire insurance agency to evaluate their responsibility in around 12,000 urban areas and towns in the US during the nineteenth and twentieth hundreds of years. According to Miller, who is the director of Ohio State’s Center for Urban and Regional Analysis (CURA), they were frequently updated on a regular basis in larger cities.