Interim Report, HD-51897-14, Image Analysis for Archival Discovery (Aida), January 2016

In the third six months of work on Image Analysis for Archival Discovery, the project team has made progress toward the goals outlined in our report from June 2015. As we reported in June 2015, we realized that our original plan to analyze 7 million pages from Chronicling America was overly ambitious for the grant period, and we revised our goal to complete a thorough case study of our methodology and code for all newspaper images in Chronicling America from the period 1836-1840. Activities undertaken, toward this and other grant goals, from June 2015–December 2015: Publication of the article, Developing an Image-Based Classifier for Detecting Poetic Content in Historic Newspaper Collections, D-Lib Magazine (July/August 2015). doi: 10.1045/july2015-lorang Completed pre-alpha version of complete code base Continued development of project documentation Made progress in case study of images from 1836-1840 Developed partnership with a researcher from the University of Virginia Discussed future project directions with Institute of Museum and Library Services staff Filed for no-cost extension to continue grant work through June 201