RICERCA ITALIANA     

Integrated Survey Methods and Techniques for constructions and utilization of 3D virtual models dedicated to architectural and urban complexes.

Università degli Studi di Roma "La Sapienza"

Abstract

Documentation, conservation, and knowledge of architectural and urban patrimony is an open and central problem affecting the management of cultural heritage assets in our territory.
Too often, however, such a richness isn't balanced by an adequate planning and shows a substantial lack in the capacity of providing all needed information for researchers and experts work, either in normal or emergency conditions.

The inadequacy of the traditional approach concerning documentation and conservation can be explained as the result of a lack or disorganization of information. On one side in fact a relevant information base (geometrical, structural, etc.) is missing even for well known sites; on the other existing data sets are often dispersed among various actors being seldom centrally reunited in a repository for storage and retrieval.
Part of the problem is related to the absence of data banks capable of delivering
any time critical information to the intended intervention effort, up to the point that often the partial or total damage of an artefact has represented its definitive loss.
In this field, however, we are experimenting an actual revolution due to the appearance of the 3D Scanning technique which is becoming very well known. It consists in a regular exploration of an object operated by a laser or other electromagnetic beam emitted and received back by an appropriate instrument, the 3D Scanner. The measure of the time between emission and reception (time of flight) owing to light speed invariability, or by geometrical triangulation, can then be used to calculate the sharp three-dimensional position of every point visited by the beam. This new technology is being used to generate 3D Virtual Models of every material object, practically without any limit concerning its complexity, shape, dimension, and assuring an extraordinary definition level.
Even this powerful technology, though, is not very useful without deep testing trials focused on procedure definition and optimisation: first experiences demonstrate all 3D scanning potentials, both in the field of improving and speeding up traditional procedures (for example 3D point photogrammetric survey) as well as in achieving results that before were nearly impossible. Besides it shows dramatic integration features both "before" the 3D Model construction (with data deriving from traditional procedures, with photographic and photogrammetric images, with termographic or radar analysis, etc.) as well as "after", by means of a substantially unlimited database available for GIS systems exploration or for Internet 3D Browsing new approaches. However, though, the appears to be the most suitable core for different data organisation; it can be considered also the ideal "portal" for any interactive exploration.
This methodology inserts another extraordinary powerful element in architectural and urban documentation and restoration field, the CNC equipments, that can produce exact material reproductions of any studied element using the 3D Virtual Model information: this new possibility will actually offer great opportunities for conservation as well as for dissemination activities.
The main goal of this project is then to develop and test all new opportunities offered by 3D Scanning and other innovative technologies on real study cases: defining operative protocols and data integration methods always applied to documentation and conservation of our common architectural and urban patrimony. <<<

Principal Investigator

Mario DOCCI Universita' degli Studi di ROMA

Research Objectives

There is strong dissonance between the extraordinary architectural and urban patrimony we can find in our country and the generalized lack in interventions concerning its conservation and valorisation as well as in the capacity of providing all needed information for researchers and experts work, either in normal or emergency conditions.
Possible reasons for this serious inconvenience can easily be found in a general lack of adequate databases, structured to extract anytime critical information concerning possible interventions. This situation affects not only the "background" Cultural Heritage Patrimony, but even the most famous monuments, whose most relevant characteristics (geometry, structure, etc.) are only sporadically and partially documented: as a result partial or massive destruction of a monument often coincides with its definitive loss.
A solution to this problem seems to be actually at hand owing to new sophisticated acquisition and manipulation data technologies (3D scanning, CNC equipments with CAD/CAM interface) as well as to the development of software oriented to real-time managing of different information sets (database): these data can be afterwards recalled and selected according to any need.
Presently, though, despite a number of unique experiences, organic studies aiming at testing these technologies and at defining a framework methodological approaches for documentation and conservation activities have not been carried out yet. Similarly, all aspects concerning integration of differently gathered data, adequate database construction, ways for their transmission, sharing and using, have still been not deeply investigated.
Present project will then aim at studying all these unsolved problems, with the objective of developing and experimenting 3D scanning and other technologies' new opportunities: this activities will be carried out on real study cases, tending to a definition of innovative operational protocols as well as new integration methodologies for different data sets in the field of documentation and conservation of architectural and urban patrimony.

In this framework we can however distinguish three priorities which, merging together, actually concur to reaching the general project goal:
- A Scientific and technological goal, aiming at developing, according to previous ideas, innovative procedures for construction and manipulation of 3D Virtual Models coming from 3D scans or other acquisition techniques.
The research activities will so concentrate on acquisition and reproduction techniques, in order to make them faster, reliable and flexible in reconstruction and integration with different information.

- A Methodological goal, aiming at focusing and developing new methodologies and utilisation fields for acquisition and reproduction techniques: as a result, suitable operational protocols as well as new usage hypothesis well be defined.
After this initial analysis some foreground methods will be selected in order to the following project objectives:
3D Virtual Model features construction and control - Numeric data coming from the "point cloud" is the interpretation basis leading to geometric properties of any surveyed surface. Transformation of raw data and their refinement have to be both investigated and improved with special attention to ‘decimation' (controlled reduction of needed points for complete surface description), interpolation and smart recognition of simple surfaces and discontinuities protocols.
Predictive study of surface characteristic - Together with geometric parameters, 3D scanners measure the response grade in any single point (reflectance): this datum, strictly linked to superficial condition and nature of the object, seems potentially predictive both for conservation state and material first estimation.
Geometric information integration and Multidimensional Database construction - The 3D Virtual Model can also be integrated point by point with additional information deriving from other investigations (photographic, spectrophotometric, etc.), so to become the central core of an actual Multidimensional Database concerning the monument.
Suitable storing and consultation procedures result as a major objective for this project: as applying architectural-and-urban-complexes-oriented GIS systems, as finally investigating new on-line sharing approaches (like the 3D Browsing technology) able to integrate videogames features with concurrent, real-time interactive exploration modes.
Knowledge/Dissemination - This Multidimensional Database, whose core is the 3D Virtual Model, demonstrate dramatic potentials in the dissemination activities: in the scientific field (many experts can contemporarily share object's information in any part of the world) as well in the tourist and teaching one. We can in fact imagine interactive workstation which can be used by the public to explore parts generally unreachable or not visible owing to restoration, maintenance, etc.
A major task will finally concern reproduction potentials of material replicas techniques using CNC equipments CAD/CAM interfaced: in this case the 3D Virtual Model archive, owing to its accuracy, can lead to material models production again for dissemination purposes (teaching, museums exhibitions, etc.) as well as conservation ones, providing anytime substitutions for damaged or destroyed object's areas.
Monitoring - This aspect is going to play a more and more relevant role in architectural and urban conservation policies because it provides an extraordinary monitoring system able to control monuments' general conditions as well as to immediately show possible critical situations (deformation or subsidence new areas, superficial degradation, etc.).

- An application task , finally, where some developed procedures will be tested and validated on real study cases: these will be chosen within the Mediterranean area by all URs depending of their relevance to project's goals. <<<

First Results

From this task we expect a relevant increase of basic knowledge concerning the instruments themselves and their management; on the other hand, though, a productive feed-back process between research institutions and manufacturers will be set up in order to reach an overall improvement of existing equipments as well as the development of new ones.This task represents the core of the project: the main part of results, both on the theoretical and operational side, are then expected during this phase. Beyond the set up of methodologically correct procedures, a number of virtual 3D models will be acquired, constructed and elaborated according to the research line expressed by each Research Unit.
All these elaborations could form a first national kernel for data responding to a common methodological standard being in the meanwhile open to further additions or integrations.
Some of these examples will be used for testing the previously illustrated on-line dissemination techniques.Refining and validation of the procedures developed within the project.
Application to relevant real cases.
Partial reports on local research on the Internet and in the proceedings of meetings. Publication of the final research report on journals and conference proceedings. <<<

Timescale

24 months

National and international background

Documentation, conservation, and knowledge of architectural and urban patrimony is an open and central problem affecting the management of cultural heritage assets in our territory.
Too often, however, such a richness isn't balanced by an adequate planning and shows a substantial lack in the capacity of providing all needed information for researchers and experts work, either in normal or emergency conditions.

The inadequacy of the traditional approach concerning documentation and conservation can be explained as the result of a lack or disorganization of information. On one side in fact a relevant information base (geometrical, structural, etc.) is missing even for well known sites; on the other existing data sets are often dispersed among various actors being seldom centrally reunited in a repository for storage and retrieval.
Part of the problem is related to the absence of data banks capable of delivering
any time critical information to the intended intervention effort, up to the point that often the partial or total damage of an artefact has represented its despair.

The question is quite complex, and addresses not only a solution, but an organic investigation capable to put in full light its specific terms. In order to design the heterogeneous scientific framework of this research proposal it is useful to start from this very opened question.

The terms of the problem.
At the centre a few technological and methodological problems can be listed: they relate to the data acquisition, management, organization and employ. And although each of these questions involves a specific scientific frame the analysis of these will lead to a comprehensive understanding of the state of the art.

Data Acquisition
This specific problem is strictly connected with the sampling technologies.
The traditional survey procedures, direct, instrumental and photogrammetric (either computer based or not), have recently been subject to a true revolution connected with the development of 3D acquisition new tools (3D Scanning). This innovative technique consists in the 3D scanning of an artefact, with the aid of different technologies (using the emission of either a laser or a radar beam and the measurement of the so called ‘time of flight' -the time spent between the emission and the reception; with emission of light blade and double triangulation) which determine the actual position of any point of the artefact sampled by the beam. This leads to the generation of points clouds, called range maps, which represent a regular spatial sampling of the acquired surface. The width of the sampling area is generally limited to one square meter for the more accurate (triangular and radar based) systems whose range of action spans from a few centimetres to a few meters, and even more in case of the time of flight systems, whose range of action can reach up to a few hundred meters, but that are at the same time less accurate.

Such new technology, thanks to its amazing sampling speed (up to several points per second) allows for the first time in history, to build 3D virtual models of complex and extremely large objects (projects carried on by the research units include Rome's Colosseo and Pantheon, and the Hagia Sophia in Istanbul), with extreme accuracy (with an error tolerance of a few millimetres). Such models can furthermore be geo-referenced contextually with the aid of GIS systems, which determine the hearth coordinates of the scanner's station points.

It is evident that the potential of such set of technologies has scarce range of employ in the absence of a deep inquiry of their applications in architectural/monumental and urban contexts. The 3D scanning technology was in fact mainly developed for military and industrial scopes. Here continuity and geometrical legibility prevail, whereas on the contrary, surface non-regularity and discontinuity is a typical trait of the artefacts to be found on architectural/monumental and urban contexts, raising questions about specific hardware and software improvements.

Data management
The data gathered either with conventional and innovative methods merge towards the construction of the 3D virtual model of the artefact under investigation. The process is all but automatic, and involves a protocol of steps whose definition is still uncertain.

The raw data deriving for the cloud of points, which can be seen as the reference for the geometric features of the artefact, need to be transformed and refined. This process is currently carried on by percentage ‘decimation' of the existing points, with a limited possibility to weight independently different surface areas (an average planar surface can be represented with a smaller number of points than a more geometrically complex surface). Similarly the possibility to take advantage of ‘intelligent' automatic discontinuity recognition functions is also limited.

Similar opened problems pertain the procedures for generating high resolution 3D models referring to a number of different single scans, that need to share a unique and common reference system. The alignment of several "range maps" can produce wrong measurements of the final model, dramatically reducing their potential degree of accuracy.

The full understanding within the integration of 3D models with different data sets, such as digital photographs, photogrammetry, spectrophotometry, colour sampling data and so on also needs deeper investigation.

All the research units participating to the present proposal have been investigating strategically the set of problems that this document addresses and share the understanding of the central role taken in this process by the building of the 3D virtual model, as the nucleus for merging the date needed to build a multidimensional database of an artefact.

Data Organisation - In the last ten-year periods, the methods of the historical, architectural and urban research have really changed, even as for registration systems: from easy and general notes to sheets that bind the scholars to collect data in an impartial way. Moreover, architectural and urban researches are more and more contextual and have to expand on a vast scale and to manage a big number of data. The large amount of plants, drawings, prospects and views makes hard the analyses fort the scholars. That is why it has become necessary to establish a system to manage easily the data and to process them quickly in an whole and open frame: this system, that we have called Multidimensional Database, has to be able to record many kind of multimedia data (texts, numbers, bi- and tri-dimensional drawings) divided in single objects. The model has to include spatial and temporal dimensions to allow the projection in different temporal fields in Unit or Units Site space.
The idea to create a multi-scale system comes from the need of recording, managing and analysing data coming from different kind of research. Such system has to give to the final user the possibility to view and analyse the data both in macro (context) or micro scale (single monument analysis).

Data consultation and utilisation - Problems referring to this field are connected with any scientific research representing a major independent task.
In our case, though, it becomes the actual benchmark for testing all developed approaches.
There is a first consultation problem: presently all features are very scarce comparing with digital systems potentials; as well user interfaces are not enough user friendly. However some innovative instruments are coming up: advanced GIS systems, 3D Browsers and advanced visualisation and interaction tools. These last systems open real interesting research perspectives referring to architectural and urban contexts.
The state of the art in the development of such technologies sees the combination of a relevant improvement of graphical performance with the enhancement in the capability to edit and manage events, behaviours, constrains, animations, the interaction with archives and databases. With the consequence of highlighting the potential added value of the merging between a true 3D database, advanced visualization capabilities. and hypermedia interactive narrative patterns. Such platforms ease the access to complex data structures (text, images, animations, drawings, photographs and so on), that are extremely versatile and can be assembled in a very intuitive fashion. Due to their capacity to give access to a vast range of information, simply by clicking on a 3D model component (videogames derived technology), 3D model based information systems represent an improvement of traditional approaches towards information-access, and disclose new directions to the idea of direct data manipulation of and of the ‘exploration' of an information set.
Together with this "virtual" and digital approach, another important research direction is recently emerging: it concentrates on reproduction potentials of material replicas techniques using CNC equipments CAD/CAM interfaced: in this case the 3D Virtual Model archive, owing to its accuracy, can lead to material models (stone, plastic materials, etc. ) production again for dissemination purposes (teaching, museums exhibitions, etc.) as well as conservation ones, providing anytime substitutions for damaged or destroyed object's areas. <<<