3D digitalization techniques
3D DIGITALIZATION TECHNIQUES
David Lo Buglio (Université Libre de Bruxelles)
For many years, architectural survey has seen the development of digitization techniques, methods and their progressive democratization. The emergence of these technologies has been accompanied by a questioning about the contribution and the integration into fields connected with the architectural heritage. New prospects have been opened up for the communication of physical data in the form of digital information. The emergence of drawing and visualization tools (3D interface, augmented reality, etc.) has enabled to obtain systems of representation which offer a reading of architectural spaces by immersion.
At the same time, the development of survey techniques (lasergrammetry, photogrammetry, etc.) gave the possibility of obtaining a digital print which can provide simultaneously:
- three-dimensional metric data,
- geometric information about the morphology of the object,
- information about the textures and materials of which it is made.
The combination of all these resources is of benefit to those working in the disciplines concerned by management, analysis and conservation of the architectural heritage.(1)
The data collected from these techniques are often in the form of dense point clouds where each point corresponds to a measurement of the object. These point clouds come from automated acquisition/processing methods that can be divided into two main families: point acquisition by terrestrial laser scanning (lasergrammetry) or acquisition by multi-stereo correlation (in the case of photogrammetric acquisition).
The two methods of digitization mainly used:
- LASERGRAMMETRY
The terrestrial laser scanning is an acquisition technique that uses a light source (the laser) as a means to measure the space surrounding the survey station. The most common method of laser surveying, also known as the "time-of-flight method", is based on the calculation of the time required for the light wave to cover the distance from the object to the sensor of the station. The amount of data (or coordinates) collected by a laser scanner can easily range from several tens to several hundred thousand points per second. - PHOTOGRAMMETRY
Photogrammetry is a measurement technique that uses photographs as a study medium. This technique is based on the principle of triangulation of points homologous to a set of photographs illustrating the same object. The triangulation work makes it possible to find the positions of the cameras in a reference space. From the identification of their position and coordinates, it becomes possible to obtain the measurements of the object, to reconstruct it and to texture it precisely from the information contained in the photographic sources (or rather by their projections).
As shown in the second illustration, photogrammetric or laser scanner acquisition provides support for automatic restitution using polygonal meshes. Moreover, this digitization is also enriched by the presence of a texture projected from a photographic survey (left part of the image). The combined result of the survey, the polygonal mesh and the texture presents a great visual and metric coherence with the object analyzed.(2)
The democratization of photogrammetry and other acquisition techniques permits the creation of digitization composed of several millions of data having a great visual and metrical coherence with the observed artefact . However, these techniques constitute, in the field of architectural study, new arguments to produce figurative documentation at the expense of a representation integrating the codes and the language of architecture. This architectural representation paradigm shift raises the question of cultural and cognitive issues of architectural representation.
Point cloud from the lasergrammetric survey of the Tholos of Delphi, Greece.
La maison du peuple: aerial photgrammetric survey of the Sablon district and position of cameras, Brussels. AlICe lab, ULB.
La maison du peuple: aerial photgrammetric survey of the Sablon district and position of cameras, Brussels. AlICe lab, ULB.
Results of the point cloud and polygonal mesh of the S5 column (south 05) of the cloister of the Abbey of Saint-Michel-de-Cuxa. 1: S5 column surveyed by photogrammetry, 2: Point cloud made up of 1,595,897 vertices,3: Mesh of 313,755 polygons, 4: Shading system: Ambient Occlusion per vertex.
(1) Livio De Luca and David Lo Buglio, ‘Geometry vs Semantics: Open Issues on 3D Reconstruction of Architectural Elements’, in 3D Research Challenges in Cultural Heritage, ed. Marinos Ioannides and Ewald Quak, Lecture Notes in Computer Science 8355 (Springer Berlin Heidelberg, 2014), 36–49;
DL. De Luca et al., ‘An Iconography-Based Modeling Approach for the Spatio-Temporal Analysis of Architectural Heritage’ (SMI ’10 Proceedings of the 2010 Shape Modeling International Conference, Washington: IEEE Computer Society, 2010), 78–89;
Marinos Ioannides and Ewald Quak, eds., 3D Research Challenges in Cultural Heritage, vol. 8355, Lecture Notes in Computer Science (Berlin, Heidelberg: Springer Berlin Heidelberg, 2014).
(2) David Lo Buglio, Vanessa Lardinois, and Livio De Luca, ‘What Do 31 Columns Tell about a “theoretical” 32nd?’, ACM Journal on Computing and Cultural Heritage 8, no. 1 (2015), https://doi.org/10.1145/2700425.
(3) Marc Pierrot-Deseilligny, Livio De Luca, and Fabio Remondino, “Automated Image-Based Procedures for Accurate Artifacts 3D Modeling and Orthoimage,” in XXIIIth International CIPA Symposium, 2011.