Computational Vision in Photogrammetry for Georeferencing: Modern Resources Evaluation for UAV Image Processing

Abstract

Environmental Science aims to understand the world; it can be done by geographic information system (GIS). Also, georeferencing allows adjusting and aligning raster data combined with other GIS data. Thus, with these techniques it is possible to interpret these kinds of data and its relationships, patterns, and trends. This study aims to investigate the usage of moderns engineering procedures called computational vision, in photogrammetric image processing obtained by unarmed aerial vehicles (UAV) via a small front camera in an embedded system combined with proprietary software that uses computational vision resources. Even though, the priority is open source softwares. The research will start with a study about the state of the art of computational vision algorithms and photogrammetry with drones inspection along with the processing and checking of a set of images that will be generated and processed into georeferencing. Indeed, it was used a proprietary software that uses computational vision resources, at the Universidade de Trás-os-Montes e Alto Douro, in Portugal. Also, it will compare with the conventional methodology with modern computational resources to verify the gain obtained. To conclude, it was verified the positional quality of the generated georeferencing, and satisfying results were reported, which reiterates the potential of these modern computational resources in contemporary photogrammetry.