This work is based on sixteen years of interdisciplinary cooperation between medieval archaeologists from University of Florence (Italy) and ICT researchers from CNRS LSIS of Marseille (France), aiming towards a connection between 3D spatial representation and archaeological knowledge for interpreting the Mediterranean Middle Ages.
In the last decade, we witnessed significant improvements in photogrammetric techniques, starting with the SIFT descriptor in 2004, used to automatically match thousands of homologous points. This opened the way to increased automatization of the photogrammetric process, autocalibration and automatic 3D dense point cloud generation. These new photogrammetric tools, which now can easily replace the terrestrial laser scanner still used at the turn of this century in archaeology, but offer no improvements and remain unable to link semantic data with this plethora of accurate geometric data.
At the same time, the stratigraphic analysis applied to upstanding building in the framework of medieval archaeology is on one hand absolutely mandatory and on the other hand seems to have reached a crucial point in its evolution. Italian Medieval Archaeology, in parallel with a close relationship with historical research, contributed to the development of archaeological methodologies at European and world levels. A specific and particularly outstanding contribution was provided in the field of non-destructive urban and territorial analysis. Such was indeed the (successful) attempt to embed stratigraphic theory in the study of historical buildings, extending the principles of site-formation-process to the architectural-formation-process. The translation of the Harris paradigm in upstanding structures, i.e. in full 3D context, brings some inconsistencies which become more and more visible with the development of new 3D surveying tools.
Recent advances in Knowledge Representation, as the development of ontologies also in Cultural Heritage, changes the way to manage knowledge and extends the possibilities to connect with other research fields. The presented work aims at merging photogrammetric survey and temporal relations in order to propose a new representation of temporal relations stemming from temporal qualitative networks thanks to the Allen approach. This is done by developing a common framework describing knowledge used in photogrammetry as well as in stratigraphy fully based on ontologies. An ontology describing the photogrammetric process and the measured artefact (ashlar block, observed relevant surface and then US and connected concepts) is aligned on the well-known ontology used now since a long time in Cultural Heritage: CIDOC CRM.
We have developed an ontology to represent both the photogrammetric process and the measured objects, ashlar blocks and US. The objects are modelled according to the point of view of the measurement process but indeed these artefacts or concepts as US can also be seen from a Cultural Heritage or conservation point of view. This is the reason why
we have aligned our ontology on CIDOC CRM.
The use of ontologies in archaeology and Cultural Heritage, as we have already stated, is becoming increasingly widespread. Indeed, this formalism is particularly suitable for heterogeneous data and offers concepts and tools to manage incompleteness, updates and revisions of the involved knowledge.
Therefore, the way ontologies are structured is far from a traditional relational database and managing them can be difficult for a person without a solid background in computer science. In addition, even if a lot of research papers are published in the field of ontology for Cultural Heritage, the use of ontologies is not yet a common or widespread tool for managing archaeological excavations. A lot of research shows the ontologies are very suitable for archaeology, but only a very few archaeological missions use ontologies to manage data coming from excavations. This is normal, even if we can demonstrate that a new approach is very interesting, changing technology is always a high cost decision and ensuring the continuity of the data, analysis, and student training is a real challenge.
We decided to develop some tools in order to be display data stored in ontologies as a virtual relational database in order to have a simplified view of the stored data. These tools can provide a static point of view on these ontologies and allow simple manipulations similar to those possible on a relational database. The first one is an editor which displays classes, instances and properties present in a ABox as if the data were structured by table in a relational database.
By reading the ABox, the editor is able to display the classes present in the ontology, all their instances and offers a simple
and dynamic way to display data properties as normal fields of these instances and the possibility to modify them (an also, as we are in ontology context, to add or remove fields). This of course gives us a lot of other possibilities as the archaeological work is always in progress and a modification of the data structure is always possible and easy to implement through this interface.
For the Shawbak Castle we build an editor dedicated to physical and stratigraphic relations between Unit of Stratigraphication(US). This tool is dedicated to stratigraphic relations between US, reading an ABox and displaying an exhaustive list of the US present in the ABox with all the physical and stratigraphic relations between them. This tool allows to add, cancel, modify the relations between US and visualize them through a graph. We are still working on a consistency check tool to control on the fly the digit. And finally, we developed a tool for 3D visualization of the artifacts stored in the ABox. Indeed, currently ontologies are used also as a serialization tool to store all the data relative to the survey: oriented photographs, 2D and 3D points, relevant points for stone-by-stone surveys, computed ashlar blocks and archaeological concepts often non-measured directly as Stratigraphic Unit, Corpo di Fabbrica, etc. The tool is able to read the ABox and thanks to the procedural attachment, can instantiate the corresponding Java instance and produce a 3D representation of the instances stored in the ABox. This tool was used to produce the two images shown below according to the specific request.
Moreover, a first possible interpretation of these data should be the generation of a graph of temporal relations where the nodes, instead of having a coded position as in the Harris graphs, should have as the coordinate of the USM center of mass projected in USM main plane.
As illustrated the graph is superimposed on an unrealistic representation of the stone-to-stone reading of the building.