Three-dimensional (3D) reconstruction techniques have revolutionized many different fields, from architecture or archaeology to environmental studies, engineering, or others. Among the techniques that have materialized with the closest relation to current needs is close-range photogrammetry, offering the potential to reconstruct the geometric shape or texture of small to medium-sized objects with the help of overlapped photos or structure from motion algorithms. This paper describes the creation and analysis of a three-dimensional (3D) model created with the aid of close-range photogrammetry techniques utilizing an open-source mobile application. The main focus is to evaluate the feasibility, accuracy, and productivity of mobile-supported photogrammetry techniques, which can be an efficient and affordable substitute for the currently existing 3D reconstruction techniques. A detailed image capturing procedure was also carried out with the help of the standard smartphone camera, taking overlapped photos from different angles to cover the area sufficiently with the help of parallax. The generated data was then processed with the help of an open-source mobile application for photogrammetry, allowing the process of feature points identification, image matching, dense cloud creation, and surface recreation. The reconstructed 3D model was then compared on the basis of geometric precision, resolution, and texture. The result shows that close-range photogrammetry, conducted with the help of open-source mobile software, is able to provide acceptable levels of accuracy for non-meter purposes, especially for the application areas of cultural heritage, environmental scanning, and educational visualizations. The current study also brings attention to the promising mobile technology of photogrammetry, with the possible future application of democratizing 3D technology availability.

3D modelling, Close Range Photogrammetry, open source mobile app, image – based modelling

geodesy engineering

Presentation: poster