Evaluating Software Tools to Orthorectify Archival Aerial Photographs

Abstract

Archival aerial photographs are among the most common, voluminous, and frequently used collections in map and geography libraries because of the unique and important spatial information that they contain. They are, however, an underutilized resource because they are challenging to work with and due to misconceptions about their quality. Therefore, developing efficient workflows to convert archival aerial photographs into research-quality digital spatial data is an important next step in making them accessible to a broad range of potential users. The present study tested three software to georeference, orthorectify, and mosaic 15 digitized archival photographs. Using positional accuracy and aesthetics of output mosaics as measures of software performance, all three completed the task impressively well. Output image spatial resolution and positional accuracy were found to be comparable to common public-domain contemporary aerial photography datasets like those produced by the National Agricultural Imagery Program. We also assessed software ease of use by geospatial professionals who were not trained photogrammetrists and found all three to be accessible with care and training resources. The software tools to efficiently convert archival aerial photographs into research-quality digital spatial data have finally matured to the point where collection managers can reasonably consider collection scale conversion projects.

Keywords:

• GIS
• aerial photography
• georeference
• orthrectification
• computer-vision

Acknowledgments

The authors would like to knowledge the assistance that they received from Cody Benkleman and Shaojun He of ESRI, and Cristoph Heck and Mohsen Miri of Trimble.