Techniques, tools, platforms and algorithms in close range photogrammetry in building 3D model and 2D representation of objects and complex architectures

The aim of this paper is to identify a suitable pipeline in order to build high-resolution 3D models and 2D orthophotos of objects and architectural structures of particular historical and cultural importance by means of the photogrammetric method. An accurate reconstruction of architectural elements can be exploited both for a detailed analysis of the artefacts, and also to determine the original position of detached architectural fragments. Structure from Motion (SfM) and Multi View Stereo (MVS) approaches have already been successfully applied in many applications to build 3D models. Moreover, the obtained reconstruction can be exploited in order to build orthographic projections of the object from different directions: the orthophotos generated in this way ensure a very high geometric resolution (even sub-millimetre) and accuracy. Orthophotos are particularly useful for the digital documentation and analysis of archaeological and architectural objects. To such aim, the direction to be used for the computation of the orthophoto, which in certain cases might be nontrivial, should be carefully determined. This work describes a simple procedure for the definition of such projection direction, while also enabling the express the object position according to certain specific requirements. In addition, the use of low-cost smartphone and action cameras was investigated to carry out photogrammetric surveys. Finally, this paper shows the results obtained with an imagery acquisition tool to be mounted on the top of a drone, developed ad hoc in order to properly acquire images for instance of the base of high bridges that cannot be achieved through a terrestrial survey. The developed tools and the overall proposed method is validated on three case studies, two related to terrestrial surveys, whereas the latter consider an example of UAV (Unmanned Aerial Vehicle) photogrammetry with the developed imagery acquisition device.