Jose Ramon Martinez Batlle

Several collections of aerial photographs have been acquired in the Dominican Republic during the last 70 years. Although many of these sources are increasingly becoming available as scanned images, limited digital photogrammetric processing has been done, mainly because of the unaffordable prices of proprietary software licenses and the lack of clear workflows for processing historical photos. Many of the software tools developed in the last two decades, designated as structure-from-motion and multi-view stereo techniques, are bundled within OpenDroneMap, which is an open-source toolkit originally designed for processing aerial drone imagery. The aim of this research was to test the capabilities of this toolkit for the production of accurate orthomosaics, point clouds, and digital surface models from historical aerial photos. A workflow was designed and tested with 10 color aerial photos taken in 2003 of the Ocoa River system, which is located in the southern part of the Dominican Republic. The workflow comprised the preparation of images, selection of ground control points, reconstruction or bundle-block adjustment, dense matching with two different algorithms, and finally the generation of digital surface models (DSMs) and orthomosaics. The results showed that both the orthomosaics and DSMs showed good agreement with the reference data. The main conclusion was that the proposed workflow may be applicable to other historical aerial photographs, with high efficiency and at zero cost.