U of R professors awarded $325K NEH Grant for open-source tool to analyze historic images

Statistics professor Taylor Arnold and digital humanities professor Lauren Tilton have received a nearly $325K ($324,693) grant from the National Endowment for the Humanities to support a project to build open-source software for collecting and analyzing digital images.

Arnold and Tilton created and co-direct Photogrammar, an interactive photo collection focusing on the Great Depression era. The open-access, web-based tool allows users to easily navigate and engage with a collection of 170,000 photographs taken between 1935 and 1943.

The NEH Digital Humanities award will support a project to make the Photogrammar software available to allow anyone with a set of digital images and associated information to create — with no prior programming experience — their own digital public humanities projects.

“The goal of the software is to use interactive data visualization and AI to open up new ways of exploring and understanding digitized collections of images,” said Tilton. “We draw on methods from data science, spatial analysis, natural language processing, and computer vision to provide additional context and information to digital images — context that helps people browse and interpret the materials.”

“We are excited to create this open-source tool that will allow anyone to have this same experience with their own collections,” said Arnold. “We envision people using this software for a variety of different applications, from documentary photography, historic newspapers, and digitized medieval manuscripts.”

In addition to the software, the grant-funded project will produce six case studies that will model and highlight how the software can be used in a variety of different domains, data sizes, and types of institutions including archives, libraries, and museums. Extensive tutorials and documentation will be developed to assist in making the free software broadly accessible to the general public by 2025.