Architectural 3D modeling in historical buildings knowledge and restoration processes

Architectural 3D modeling in historical buildings knowledge and restoration processes Mario CENTOFANTI 1, Stefano BRUSAPORCI 2 Dipartimento di Architettura e Urbanistica, Facoltà di Ingegneria, Università degli Studi dell’Aquila, L’Aquila, Italy (1) mario.centofanti@univaq.it (2) stefano.brusaporci@univaq.it Abstract The validation processes of restoration projects in all institutions are still based on communication and training requisites which continue to influence choices during the preparation of projects, and not only in purely formal terms, but also with respect to content. These requisites are paper documents and digital media, i.e. reproductions of paper documents in PDF format. Furthermore, project representation and communication are substantially established in two-dimensional form (plans, elevations, sections) with rare inroads into three-dimensional formats for explicative and integrative purposes (axonometric and/or perspective images), with the usual separateness, which is also logical and conceptual, between graphic representation and documentation, during the surveying and analysis phase and during the design phase. Even considering recent experiences related to the management of the reconstruction phase in the historical centres affected by the earthquake in L'Aquila on 6 April 2009, the intrinsic possibilities of using of architectural 3D modelling is to be examined, with the purpose of building an information system dedicated to architectural specificities, and to building the project proposal, fully directing its communicative complexity to the synthetic and interactive support of 3D modelling. Key words: Architecture, 3D modeling, Restoration, 3D GIS. Fig. 1: L’Aquila. Software for visualization, surfing and measurement of point clouds. 1. Relationship between survey drawing and project design in architectural restoration The theoretical precondition for restoration projects is the inevitable relationship with the existing reality. In this sense, the close interdependence between the survey and the project is particularly significant, as it still represents a general value of the project. However, in the field of restoration, it acquires its own poignant meaning. The restoration graphical model is a complex system which, for each phase is founded on assembly drawings and part and detail drawings, where, at every stage, there should be systematic specularity between the survey drawing and the project drawing. Four significant moments in the conservation-restoration process can be identified. Phase 1: preliminary and final plans aimed at controlling architectural and spatial restoration and possible re-utilisation of the building. This is based on an essentially geometric-dimensional survey. Phase 2: project ready for execution-building with all its content including technological features. This is related to first-level surveying of the construction and the state of deterioration, carried out through inspections, sampling, surveys and non-destructive testing. Phase 3: monitoring during construction. This again involves second level surveying of the construction and the state of deterioration, this time in full view. In response to the site monitoring survey, the working project is constantly updated in accordance with the specific contexts. Phase 4: at the end of the works, a Fig. 2, 3, 4: L’Aquila. Point clouds integrated with measurable true view image based models . graphical model should be created of the building altered by the restoration work. This final post-works graphical model, together with the products of the previous phases, is the documentation needed for the database, the support for subsequent monitoring of the building and for planning any future conservation work based on certain knowledge. But above all, for comparison and contrast with the pre-works restoration model, it must provide the necessary support for formulating of a judgement on the overall quality of the restoration work. It is quite clear that, in disciplinary terms, the issue stems from identification of the characteristics for building the restoration graphical model, which intrinsically should be concise, analytical, iconic, interpretive but also dynamic, capable of supporting and sustaining continuous updating of the surveying process. It should nevertheless remain intelligible as the information it holds increases. For all stages it should be comparable, represented clearly in part drawings and other useful forms, it should be included in the relationships between assembly drawings and detail drawings. It should be systematically specular to the project design at every stage. 2. Specularity between the restoration model and the design model as an invariant Specularity between the restoration model and the design model is an invariant in the surveying-planning process for architectural conservation. This invariance, usually occurring in traditional 2D restoration models, should have full confirmation in architectural 3D modelling. In terms of surveys done using digital technology, which focuses the process on 3D models, the concept of specularity should be proposed on different and appropriate bases. Different purposes require the 3D model to be capable of meeting multiple demands: it should allow a varied body of information to be collected and, at the same time, be processable and modifiable by the many professionals involved; it should be available for consultation by the competent bodies for approving the project, and browsable, queriable, measurable but not editable; it should ensure high quality, photorealistic viewing for communication of design choices. These are many aspects of the same digital media, and in order to meet these demands, they should have appropriate characteristics, not only metrical but also constitutive and qualitative. In this sense, the issue of defining standards and formats for modelling is inescapable. Fig. 5, 6: L’Aquila. Cubic projections of photographic images related to point clouds. Current practice shows that, in faced with a wealth of metric and photographic information provided by laser scanning and photogrammetry, returns are mostly made using 2D vectors and rasters: point clouds are sectioned and projected orthogonally for layouts, elevations and sections; meshes are made for elevation views, so they can be textured to produce orthophotos; in many cases elevation images are obtained simply by displaying the points in the cloud with RGB colours. The only concessions to 3D are measurable, image-based models (e.g. true-view systems by FARO Technologies Group), or point cloud viewers (like the one by Leica Geosystems), free and easy-to-use interfaces, primarily aimed at providing designers with metric data without involving them in the processing of point clouds, and thus marking the separation between measurers/surveyors and restorers. 3D is limited to the metric data of point clouds, thereby being influenced by 2D data. In the same way, the project goes back to traditional surveying means: tables of macroscopic surface deterioration, damage and lesions, interventions. The situation is similar for information systems, used for tried and tested 2D applications, mostly on the urban scale, and therefore unaffected by the most recent acquisitions and developments in architecture with particular reference to 3D GIS for architecture. Notwithstanding the usefulness of point clouds for rigorous 2D drawings, their intrinsic potential is not exploited, to favour the creation of 3D models for surveying architectural structures and planning restoration projects. There are rare excursions into the virtual field, but mostly for publicity purposes. Conversely, the wealth of information knowledge, the importance and uniqueness of the scope provided by L'Aquila's historical town centre and those in the earthquake zone, should allow development and testing of a procedure that would place the 3D model at the centre of the surveying process and the restoration project. In terms of the purpose of restoration, it is useful to refer to photorealistic 3D models that are also defined on the scale of the construction. This second aspect does not refer so much to the issue of information systems at the so-called LOD (Level of Detail) regarding the number of polygons used to approximate a surface - but rather to the issue of the number of models to be used to describe the building system. This leads to the issue of managing architectural models that are laborious in computational terms, on commonly used software platforms. A simpler solution would be to organise models in a 'cascade' system, i.e. linking together increasingly defined models. Fig. 7: Historical window’s surveying 3D model. Fig. 8: 3D Model of a building in L’Aquila historical centre. (modeling: Francesca Cerasoli). Fig. 9: 3D model for the restoration project of Aquilanum Collegium. There is a wide series of tools available to surveyors for 3D images. The most common programs - often available with point cloud management programs - enable extraction from 2D-processed point clouds sections and orthogonal views of the clouds themselves - as well as synthesis of MESH or NURBS surfaces, or even models built on 3D primitives to be arranged using Boolean operations, to be defined semantically in terms of their architectural significance. By analogy, reference can be made to the approach provided by BIM software, an acronym with two definitions applying to the model and to the creation and management of related information, the second of which is more common: Building Information Model and Building Information Modelling. BIMs place 3D models at the centre of the entire process, on the basis of the idea of building the model as a tool for project development, a common and interoperable interface between the various professionals involved, such as architects, structural engineers and plant engineers. Unlike other types of software, developed for architectural applications from reverse engineering applications - specific to mechanical engineering and design - BIMs are devised specifically for architectural projects. The concept of information is central in that it involves reference models consisting of objects semantically defined as components of the construction. Therefore, these digital elements are in themselves an expression of the structural components of the building, with regards to its metrical characteristics and in relation to materials and their structural and energy characteristics, etc. Nevertheless, ignoring general basic objects - lines, splines, surfaces, volumes created through Boolean operations, etc. – and referring to components such as vertical or horizontal closures, doors, windows, etc., requires definition of these elements in libraries, i.e. these components must be standardised. This aspect is difficult to reconcile with the specific needs of surveying historic architecture, where each element has its own historical and architectural value. The very reasons for the emergence and development of BIM is due to the search for software suitable for developing architectural designs and managing information about the entire life cycle of the building, from its inception to its construction then maintenance, with particular emphasis on economic aspects measurements and timing. The goal is to optimise the cost-time benefits of the planning process based on the industrialisation, prefabrication and standardisation of commercial components. These requisites do not coincide with the demands of the process of careful listening and dialogue between the surveyor/restorer and the historic building, based on a system of knowledge that is always open and constantly evolving - e.g. the specifications of a restoration site. Notwithstanding, BIM can offer interesting elements. In particular, the main types of software on the market are Autodesk Revit, Archicad / Allplan, Bentley Systems, Digital Project. The first two, and the most popular, have parametric modelling limitations, while the latter two allow integration with complex 3D modelling. The emphasis is on interoperability, i.e. the possibility for the 3D model to be the project's common development interface, which can be processed in different applications depending on different professional needs. This concept, dear to the BIM software houses, is not only limited to these types of programs, since even 3D models constructed with traditional modellers can be usefully employed in architecture and, say, structural calculation. More generally, 3D models provide the option of being a support and interface for architectural information systems. In fact, for the surveying process, the model enables the collection, correlation and systematisation of the large and heterogeneous mass of information concerning surveying and historical-documentary analysis. The 3D model contains metric, geometric, architectural and construction information, as well as information on materials, colour, deterioration, etc. By its very nature, the 3D model offers a spatially structured interface allowing intuitive navigation and querying, in order to facilitate analysis of historical architecture. Hence, a model defined in terms of the construction features of the building may be linked to a database that is editable and implementable in time, thus enabling definition of an architectural information system. In this sense, the 3D model can be seen as a support for a process of involving the 3D recataloguing of surveying and historical-archive documentation Notwithstanding the issues involved in importing 3D models into a GIS environment, which have previously been studied and partly resolved by using proprietary software (see Centofanti et al, 2011). Whereas BIM already offers an opportunity to focus on the model as a useful design tool, nevertheless, the validation and authorisation phases are still centred on 2D representations, where digital features (raster or PDF) are merely simpler means of disseminating and reproducing drawings. This consideration suggests a broadening of horizons, whereby the 3D model might also become the instrument/interface for communicating analyses and the project at all stages. Interoperability (as a possibility for several operators developing and modifying the model) and use of the model as an interface for communication (so the model is a queriable but not modifiable document) are the focuses of what can be considered two prominent issues: definition of model formats and construction standards and the syntactic composition and exchange of the model itself. For example, in order to communicate the restoration project to the monitoring bodies, a queriable model could be exported in a searchable format such as PDF or VRML, linked to an open source webGIS system. Fig. 10: Surveying 3D model of a roof. The model is defined in all constructive components. Fig. 11: 3D model for equipment and lighting planning. Lastly, we look at the issue of defining standards for 3D architectural surveying models, i.e. not only the characteristics of metric accuracy, but also the level of definition and the number of constituent components, with numbers and characteristics that describe both the architectural configuration and the construction configuration of the building. In this sense, the following would be appropriate: definition of standards for 3D models in terms of describing the geometric-dimensional morphological-figurative, technical and construction characteristics of the constructions on an architectural scale; establishing standards for 3D models of historic centres, which are suitable for describing the complexity of the spatial, historical and environmental relationships within clusters of buildings; studying multi-scale relationships between 3D models of historic centres, buildings, architectural details, optimised for use on a software platform for normal operation. In conclusion, some issues remain regarding the effectiveness of the 3D model for restoration projects. First are the definition, validation and dissemination of shared standards relating to the configuration of the 3D model. Secondly, identification of standards, formats and procedures for the exchange, modification and processing of 3D models in various fields of interest (architectural, structural, installations, etc.). Finally, investigation into methodologies linking 3D models in a proprietary or open-source GIS environment For the development of technical and operational specifications, it would be useful to implement the same cultural change in restoration projects as is already seen in the architectural concept with reverse modelling. Fig. 12, 13: Architectural Informative System of Correr-Dolfin villa made by the integration between 3D models and ESRI GIS (SIArch-Univaq Sistema Informativo Architettonico Università dell’Aquila, cfr. Centofanti et alii, 2011). References [1] BENEDETTI, Benedetto, GAIANI, Marco, REMONDINO, Fabio, (editor). Modelli digitali 3D in archeologia: il caso di Pompei. Pisa, Edizioni della Normale, 2010. ISBN 9788876423536. [2] BIANCHINI, Carlo, INGLESE, Carlo. Il Rilievo come Sistema di Conoscenza: prime sperimentazioni su Santa Maria della Rupe a Narni. Disegnare idee immagini. N. 41, 2010, p. 30-41. ISSN 1123-9247. [3] BRUSAPORCI, Stefano. Modelli digitali per la rappresentazione dell’architettura. DISEGNARECON. Vol. 4, n.8, 2011, p. 107-115. ISSN: 1828-5961. [4] BRUSAPORCI, Stefano. Modelli 3D per il progetto di restauro. In PAPA, L. M. (editor). Il disegno delle trasformazioni. Napoli, CLEAN, 2011. ISBN: 9788884972156. [5] CARBONARA, Giovanni, FIORANI, Donatella. Analisi, descrizione e interpretazione storico-critica nel disegno a fini di restauro. In MARTONE, Maria (editor). La formazione nella rappresentazione edilizia. Roma, Kappa, 2000, p. 39-49. [6] CENTOFANTI, Mario, CONTINENZA, Romolo, BRUSAPORCI, Stefano, TRIZIO, Ilaria. The Architectural Information System SIArch3D-Univaq for analysis and preservation of architectural heritage. ISPRS -The international archives of the photogrammetry remote sensing and spatial information sciences. Vol. XXXVIII5/W16, 2011. ISSN: 1682-1777. [7] CHIAVONI, Emanuela, PAOLINI, Priscilla, (editor). Metodi e tecniche integrate di rilevamento per la realizzazione di modelli virtuali dell’architettura della città. Roma, Gangemi, 2009. ISBN 9788849214154. [8] CIGOLA, Michela. Sistemi informativi per la gestione e valorizzazione del patrimonio urbano. In BRUSAPORCI, Stefano, (editor). Sistemi informativi integrati per la tutela la conservazione e la valorizzazione del patrimonio architettonico e urbano. Roma, Gangemi, 2010, p. 212-217. ISBN: 9788849218602. [9] DE LUCA, Livio, BUSSAYARAT, Chawee, STEFANI, Chiara, VÉRON, Philippe, FLORENZANO, Michel. A semantic-based platform for the digital analysis of architectural heritage. Computers & Graphics. Vol. 35, n. 2, april 2011, p. 227-241. ISSN 0097-8493. [10] DOCCI, Mario, CHIAVONI, Emanuela, FILIPPA Monica, (editor). Metodologie integrate per il rilievo, il disegno, la modellazione dell'architettura e della città. Roma, Gangemi, 2011. ISBN 9788849222081. [11] GAIANI, Marco, APOLLONIO, Fabrizio Ivan, CORSI, Cristiana, BALDISSINI, Simone. An integrated 3D geodatabase for Palladio’s work. International Journal of Architectural Computing. Vol. 2(8), 2010. [12] IPPOLITI, Elena, MESCHINI, Alessandra. Dal “modello 3D” alla “scena 3D”. Prospettive ed opportunità per la valorizzazione del patrimonio culturale architettonico e urbano. DISEGNARECON. Vol. 3, n.6, 2010, p. 77-91. ISSN: 1828-5961. [13] MIGLIARI, Riccardo, 9788878909410. (editor). Prospettiva dinamica interattiva. Roma, Kappa, 2008. ISBN [14] MINGUCCI, Roberto, GARAGNANI, Simone, CINTI LUCIANI, Stefano. Building Information Modeling: la tecnologia digitale al servizio del progetto architettonico. DISEGNARECON. Vol. 4, n.7, 2011, p. 5-19. ISSN: 1828-5961. [15] REMONDINO, Fabio, MANFERDINI, Anna Maria. Reality-based 3D modeling, segmentation and webbased visualisation. Procedings of Euromed 2010, LNCS 6436, Springer Verlag, p. 110-124, 2010. [16] TRIZIO, Iaria. GIS-technologies and Cultural Heritage: stocktaking, documentation and managment. In Rethinking Cultural Heritage. Experiences from Europe and Asia. Dresden, Technische Universität Dresden, 2007, p. 75-91. ISBN 9783867800164.