Colour and Space in Cultural Heritage: Key Questions in 3D Optical Documentation of Material Culture for Conservation, Study and Preservation

Colour and Space in Cultural Heritage: Key Questions in 3D Optical Documentation

of Material Culture for Conservation, Study and Preservation

Frank Boochs 1, Anna Bentkowska-Kafel 2,*, Christian Degrigny 3,

Maciej Karaszewski 4, Ashish Karmacharya 1, Zoltan Kato 5, Marcello Picollo 6, Robert Sitnik 4, Alain Trémeau 8, Despoina Tsiafaki 9, Levente Tamas10

1 i3mainz, University of Applied Sciences Mainz, Lucy Hillebrand Str. 2, 55128, Mainz, Germany – [email protected]; [email protected]

2 Department of Digital Humanities, King's College London, 26-29 Drury Lane, London WC2B 5RL, United Kingdom – [email protected]

3 Haute Ecole Arc de Conservation-restauration, Espace de l’Europe 11, CH-2000 Neuchâtel, Switzerland – [email protected]

4 Warsaw University of Technology, Faculty of Mechatronics, 8 Boboli Str., 02-525 Warsaw, Poland, [email protected]; [email protected]

5 Institute of Informatics, University of Szeged, Arpad ter 2, Hungary – [email protected]

6 Nello Carrara Institute for Applied Physics of the Italian National Research Council (IFAC-CNR), Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy – [email protected]

7 Laboratoire Hubert Curien UMR 5516 - Université Jean Monnet, Bat. E, 18 rue Benoit Lauras, 42000 Saint-Etienne, France – [email protected]

9 Athena Research & Innovation Center in Information, Communication & Knowledge Technologies, Athena Building R.C., University Campus of Kimmeria, P.O. Box 159, GR-67

100 Xanthi, Greece – [email protected] 10 Levente Tamas, Technical University of Cluj-Napoca, Dorobantilor st. 73,

400609, Romania – [email protected]

Abstract. The paper introduces some key interdisciplinary questions concerning the development of optical measuring techniques and electronic imaging applied to documentation and presentation of artefacts, as identified through the work of Colour and Space in Cultural Heritage (www.COSCH.info), a trans-domain European Action (TD1201) in the area of Materials, Physics and Nanosciences (MPNS) supported, since 2013, by the European Cooperation in Science and Technology http://www.cost.eu/domains_actions/mpns/Actions/TD1201. Some 125 international researchers and professionals participate in COSCH activities which have been organised around six main subjects: (1) spectral object documentation; (2) spatial object documentation; (3) algorithms and procedures; (4) analysis and restoration of cultural heritage surfaces and objects; (5) visualisation of cultural heritage objects and its dissemination; and (6) the semantic development of the COSCH Knowledge Representation.

The Authors outline and illustrate the approaches adopted by COSCH. They indicate future work that is needed to resolve the identified scientific, technical and semantic questions, as well as challenges of interdisciplinary communication, to ensure a wider adoption of specialist technologies and enhanced standards in 3D documentation of material cultural heritage — being a basis for its understanding, conservation, restoration, long-term preservation, study, presentation and wide dissemination.

Keywords: cultural heritage, 3D documentation, optical measurement, multispectral imaging, fusion, visualisation, art conservation, interdisciplinary collaboration, COSCH KR

1. Introduction: Why COSCH?

Colour and Space in Cultural Heritage (www.COSCH.info) represents one of the latest international and interdisciplinary efforts in perfecting documentation of extant material cultural heritage. The emphasis is on non-invasive documentation, based on accurate recording of colour and geometry of objects, using multispectral and 3D imaging. Despite recent advancements in colour science and optical measuring techniques there are still many unresolved questions concerning the lack of shared standards in colour and shape measurements. There is no common understanding of best practice. Access to high quality 3D records, based on precise surveying methods, is limited — this hinders the understanding of benefits and limitations of such techniques, constricting their wider adoption in professional museum and heritage practices, and other areas of study. COSCH draws on earlier interdisciplinary research and relies on strong engagement of users from across cultural heritage domains. COSCH is designing new case studies and developing new digital tools for enhanced, shared understanding and application of spatial and spectral documentation.

2. Spectral Object Documentation: Questions of Colour In art conservation the need to define colour and its changes over time became

evident in the late 19th century, when Russel and Abney studied the effects of different types of light on watercolour by using spectrophotometric techniques [1]. By the 1920s significant efforts to reach international agreement on quantifying light and colour had been undertaken, resulting in the establishment of the CIE 1931 chromaticity diagram and tristimulus colour coordinates. In the 1930s–40s, at the National Gallery, London, Rawlins pioneered the introduction of the colorimetric techniques to the art world [2]. Since the 1930s, the number of scientific laboratories within museums has increased remarkably, and conservation scientists have made use of a wide range of analytical devices to study works of art. Among the scientific methodologies introduced to art conservation, visible reflectance spectroscopy has been used not only to calculate the colour of polychrome artworks, but also to identify pigments/dyes and to create specific databases of standard spectral curves for the various pigments/dyes [3].

Colorimetric measurements that were mainly performed on small areas (spot size or 1D technique) have generally been applied to evaluating the differences between colours of neighbouring areas in artworks before and after conservation; monitoring the fading phenomena caused by exposure to a specific environment; studying colour alteration produced by the application or removal of top layers of varnish; and stabilising colour against various light sources.

Great improvement in the study and calculation of colour in this field was the ability to move from 1D methodologies to those that cover wide areas (2D technique). The latter are called imaging techniques, and they could be grouped in multi-band, multi-spectral, and hyper-spectral methodologies depending on the number of bands selected over a given spectral interval and on their bandwidths. In the past, some European Research Projects focused on 2D multi-band and multi-spectral imaging capture. Notably, VASARI, MARC I and MARC II, CRISATEL, EPOCH and NOESIS emphasised the importance of colour definition in art objects [4]. Indeed, accurate colour reproduction is one of the most important requirements for imaging techniques for the documentation and study of artworks. The analysis of chromatic characteristics is important not only to ensure the reproduction of the artwork is reliable, but also to provide quantitative and objective information about the condition of the object.

The project Colour Research for European Advanced Technology Employment (CREATE), funded from 2006 to 2010 through Framework 6 Marie Curie, incorporated research from eight European universities and collaborators from both arts and science colour groups. In order to gain a better understanding of the

importance of colour and its measurements and evaluation, this group shaped new research by melding nascent colour research in the visual arts with computational and material sciences, to form a new international interdisciplinary cohort of colour researchers [5].

Still, there are no well-established and commonly accepted standards for precise, non-contact documentation of artworks that could implement and combine the above mentioned techniques. Therefore the COSCH Action aims to propose general recommendations for colour measurements by using imaging systems. This will provide the art conservation community with guidelines for the most common applications. However, the main focus is the standardisation of the acquired imaging data and the calibration procedures to be followed in diverse imaging systems. In particular, great emphasis has been placed on a well-defined and structured round robin test in order to compare colour measurements on calibrated standards and laboratory mock-ups, performed by using diverse imaging devices developed by participating research groups.

A final report is expected by the end of 2014. It will provide the art conservation community, as well as the scientists working with colour and spectral imaging acquisition, with reliable documentation of the best measurement procedures to be followed by users, starting from the most immediate applications, such as acquisition of accurate RGB images using an RGB camera. In addition, the standard procedures for the acquisition and processing of images will be documented and made available, as will the definitions of technical terminology.

3. Spatial Object Documentation: Questions of Shape COSCH considers 3D imaging of artefacts as spatial object documentation which

consists of surface geometry and/or volumetric imaging, acquired by different techniques and devices. The main aims of COSCH related to 3D documentation are to: collect information about a whole range of 3D imaging techniques and devices; validate these techniques and assess their usefulness for documentation of cultural heritage objects; create guidelines on how to choose a technique depending on the properties of the artefact and technical requirements. In the COSCH project all information about 3D imaging techniques and devices involves knowledge about:

• measurement method principle, including a list of devices for its implementation; • cost and time constrains, including time required for preparation, acquisition and

processing; • required experience of the operator; • technical parameters such as the measurement resolution and uncertainty; • calibration method and the entire procedure of traceability of 3D data; • limitations of the method and environmental conditions necessary to achieve

good results; • examples of application, with measurement results and comments.

The techniques currently available are being compared and validated by performing comparative measurements of the same artefact using different devices. All information and validation results constitute valuable specialist knowledge that will enable COSCH to create guidelines concerning the choice of technical tools and the optimisation of 3D documentation process for the use by the cultural heritage communities, including art historians, conservators, curators and educators. An exemplary description of the structured light method, with automation of multi-view acquisition, is presented in Table 1.

Table 1. Structured light method with automation of multi-view acquisition. An exemplar by R. Sitnik and M. Karaszewski, Warsaw University of Technology

Questions Explanation

Measurement method principles

Structured-light 3D scanning with colour measurement (visible range). Automated acquisition of directional measurements

Cost Equipment: measurement head 3,000€, positioning system 30,000€, PC computer for processing 2,000 €; Total 35 000€. Working day: energy approx. 10 €, operator supervision 40€. Total 50€ Service: 0€

Timeframe Preparation: 10 days (first time), 10 minutes (each subsequent time) Acquisition: 26 days (4,822 measurements) Processing: 5 days

Experience required of the operator

Due to automated measurement process, little experience is required. Operator’s supervision is needed as a safety measure, in case of errors in collision-avoidance algorithms.

Preparation process Calibration check was performed on a daily basis to assess highest quality of measurement. It was done automatically by measuring calibration unit from few positions and checking results. Calibration was performed routinely after moving the robot platform to another position. The calibration procedures are also automated and no user assistance is required.

Limitations The object size was limited in height by the robot’s operating range (2500 mm). Its diameter was not a limiting factor (for larger objects, more robot platform position changes are required). The weight of an object was not an issue because it was not moved during the digitisation procedure.

The surface of an object should be diffusive, reflecting or transparent surfaces cannot be measured. The shape of a surface is also irrelevant as long as it does not have deep cavities which cannot be measured by the structured light system. If the surface of an object is devoid of distinct geometrical or colour features, it is not possible to achieve good integration of directional measurements into one, complete model.

The second category of difficulties is related to the positioning system, including the supply of power for the positioning system (3phase AC, hardly available in, for example, remote places), the weight of the whole system (almost 500 kg in total – limiting the use of such a system for ground-level objects only), the size of the robot platform and a requirement for free space around the object for manoeuvring (therefore it cannot be used in tight spaces). Finally, some environmental conditions such as temperature and humidity are limiting factors, which have to be controlled; ambient darkness is required.

List of potential difficulties

This particular system currently cannot measure objects higher than 2,000mm because the positioning system has too small a range. In fact it cannot measure, for example, small objects high up in a building façade – the robot platform cannot be raised. It cannot be used within a gallery due to its size. It is cumbersome to manoeuver and requires much space, therefore it may not be suitable for digitisation of some excavation sites, tombs etc.

The structured-light technology with automated acquisition cannot be used for transparent or reflective surfaces. It cannot be used for multispectral imaging (without some modifications) or for objects without distinct geometrical or colour features. As the maximum achievable resolution is about 10,000 pts/mm2 (with an inaccuracy lower than +/- 5µm) it cannot be used when higher point density or accuracy is required.

This technology is suited for digitisation of individual objects (with high resolution) rather than for documentation of whole sites or groups of objects.

Measurement environment limitations (laboratory

Laboratory conditions are required for the highest quality of results. However, when correctly operated outdoors and within allowable inaccuracies, this

conditions, outdoor, indoor, in-situ, etc.)

particular system achieves a resolution of 2,500 pts/mm2 with some means for environmental control (darkened tent, air conditioning units).

Measurement method parameters (resolutions, uncertainties, texture imaging, etc.)

Resolution: 2,500 points / mm2 Uncertainty: +/- 10um Working volume (single measurement): 50mm x 50mm x 20mm Texture imaging: yes (in visible range), RGB, directly mapped onto measurement points

Example of application: A sandstone vase from Museum of King Jan III’s Palace at Wilanów, Warsaw, by J. A. Karinger and J. A. Siegwitz, 18thc.

Photo © Wojciech Holnicki

Radius: 500mm, Height: 2,000mm. A surface decoration in relief, showing mythological scenes. The colour of the object varies; signs of weathering can be seen. The whole object was measured with the highest possible resolution for the scanner used (2,500 pts/mm2). 4,822 directional measurements were collected. The output format is a cloud of points with RGB colour information and surface normal. The model data size is about 1TB. More details about the process of automated digitisation can be found in Karaszewski, 2012 [6].

Exemplary results

(a) Whole dataset (b) Zoom in (c) Close-up detail

Computer model of a garden vase, 18th century, King Jan III’s Palace at Wilanów. Screenshots © Warsaw University of Technology

4. Algorithms and Procedures

One of the most challenging issues in digitisation of cultural heritage is the fusion

of information from several different sources. Today, acquisition platforms include a range (3D or multi 2D) camera (colour/infrared/spectral) sensors that are usually capturing the surrounding environment independently, although the information from different sources can be used in a complementary way. In order to fuse the information from independent devices it is highly desirable to share calibration, i.e. to

transform the measured data into a common coordinated frame. To achieve this goal either extrinsic or intrinsic-extrinsic calibrations must be performed, depending on whether prior knowledge of the camera’s intrinsic parameters is available. When the extrinsic parameter estimation must be performed for a range-camera sensor pair, the rigid movement between the two reference systems must be determined.

For example, in geo-archaeology, the extrinsic parameter estimation for 3D Lidar and perspective camera makes sense, especially for environmental mapping applications (Fig. 1). However, this problem is far from trivial. Due to the different functionality of 3D and camera sensors, the calibration is often performed manually, or by considering special calibration targets on images (e.g. checkerboard patterns), or point feature extraction methods. These tend to be laborious, especially if the calibration procedure has to be done more than once during data acquisition. In practice, it is often desirable to have a flexible, one step calibration as sensors are not necessarily fixed to a common platform.

Example of input Lidar range image

First perspective camera image.

Second perspective camera image.

Perspective camera images projected onto the Lidar 3D point cloud after calibration

Fig. 1. Fusion of a Lidar and two perspective cameras after calibration [7]. The calibration problem is a consequence of a multitude of spatial and

spectral/colour acquisition techniques available, and a huge list of parametric factors for each instrument that affect the calibration accuracy. For example, for spatial acquisition we can mention laser scanning, point-based spatial recording, spatial fringe projection, spatial photogrammetry and tactile measurement. In this context, it is essential to: (1) identify what are the main parametric factors which have an influence on the calibration process; (2) determine what are the parametric factors common to acquisition systems and (3) identify what are the main calibration/registration/fusion approaches that can be used to fuse heterogeneous data. To address these problems we are developing two complementary documentation ontologies: (1) the COSCH Knowledge Representation (COSCH KR) which represents the formal knowledge of experts and engineers working in the field as a hierarchy of concepts using a shared vocabulary to denote the types (e.g. technologies, instruments,

algorithms), properties (e.g. sensitivity, accuracy, speed) and interrelationships (e.g. acquisition geometry vs. accessibility of sites/objects, acquisition geometry vs. physical properties) of those concepts; (2) the COSCH Algorithm Selection Module (COSCH ASM) which formally represents the knowledge of computer vision experts as a hierarchy of algorithms using a shared vocabulary, understood by non-experts, to denote the types (e.g. calibration, registration, fusion), properties (e.g. accuracy) and interrelationships (e.g. between parametric factors) of those algorithms.

5. Analysis and Restoration of Surfaces of Cultural Heritage Objects It is a well-known phenomenon that surface characteristics are subject to the

interaction with optical radiation and the process of data capture. Indeed glossy materials, such as metals, or shiny surfaces, will be difficult to scan using structured-light imaging techniques. Due to their ageing, the monitoring of the condition of cultural heritage artefacts, using spatial and spectral imaging, should be carried out in a standardised way. Simultaneously it is essential to adapt the technologies to the needs, the nature and size of the artefact, as well as the budget available. The level of accuracy will not be the same for all artefacts. The end-user must be able to save, store and re-use data collected in different formats.

The present imaging of wall paintings at the Château de Germolles in Burgundy is a good example of the way such techniques can support documentation of cultural heritage. The 14th-century painted decoration of the only surviving residence of the Duke of Burgundy, Philip the Bold and Margaret, Countess of Flanders, was rediscovered, in the 1970s, well-preserved beneath a layer of plaster that had been applied in the course of modernisation of the first floor of the château at the beginning of the 19th century (Fig. 2a).

a b Fig. 2. Decoration of the dressing room of Margaret of Bavaria, daughter-in-law of Margaret

of Flanders; (a) after the removal of the 19th-century plaster, and (b) after the restoration work © Château de Germolles.

Some of the best preserved decoration shows large white letters, P (for Philip) and

M (for Margaret), with teasels, a symbol of fidelity, in between. These motifs are characteristic of the spirit of courtly love that was fashionable at princely courts at the end of the 14th century. The 1989–1991 conservation treatment consisted of filling in the holes left by the chipping process. The tratteggio retouching technique was applied to achieve visual integration between the original and repaired portions of the paintings (Fig. 2b).

The conservation work has not been documented. Therefore the public visiting the château have difficulties in appreciating the original parts of the rare decoration, which is recorded in the contemporary medieval accounting notes. They provide evidence for the teasels to be gilded. Large quantities of linseed oil have been used suggesting that the oil gilding technique was indeed employed.

Some preliminary investigation during a COSCH Short Term Scientific Mission, using technical photography, enabled some documentation of the decoration using UV and IR radiations, as well as VIS light. The fluorescence observed under UV radiation has been used to locate the presence of linseed oil (Fig. 3c) and of gold remains, as shown in Fig. 3d.

a b c d Fig. 3. A teasel of Margaret of Bavaria in her dressing-room, (a) photographed in VIS light; (b) closer view with the remains of the original paint layer; (c) remains of linseed oil that

fluoresces under UV radiation and (d ) of gold © Francesca Piqué. The gold layer is difficult to see except when using raking light. Another way is to

use structured-light imaging. The walls, not being flat, 3D scanning of the decoration may help with identifying more of the original decoration (Fig. 4).

a b c Fig. 4. SLI of the teasel represented in Fig. 3. (a) equipment used, (b) imaging in progress

and (c) result obtained (c), © Mounir Haddadi. The results of this imaging campaign should be used eventually to propose a 3D

representation of the original wall paintings at Germolles.

6. Visualisation of Cultural Heritage Objects and its Dissemination Despite considerable international efforts in the past decades, 3D documentation of

the material culture is not applied as widely as one would expect. Many issues remain still unresolved. Among the most important is the fact that many specialists in the arts and humanities are not too familiar with methods available, and some do not know how to choose between different techniques. The definition of the requirements is not always clear, or is difficult to communicate clearly to scientists and technical specialists, who may be able to implement the desired visualisation of an object or site. Therefore the need for bridging the communication gap between the experts representing the respective domains of cultural heritage and technology is essential. This is one of the COSCH objectives.

As far as visualisation is concerned, COSCH deals with the identification, planning, implementation and testing of typical applications within the cultural heritage domain. It is well known that the objects can be visualised in many different ways and for different purposes. Applications range, for example, from edutainment and relatively simple presentation to the general public, to accurate 3D records surveyed for specialist professional research and conservation. Each purpose bears on the quality expected from visualisation.

Within this framework COSCH is selecting a number of case studies indicative of 3D visualisation needs, that are either results of earlier work, or are being developed by COSCH groups (as the ones described above). Concepts, methods, processes, issues and standards in historical visualisation of cultural heritage will be taken into consideration for the evaluation of the selected case studies in applied visualisation. Precise digital reproduction plays a key role in the visualisation of material culture. Therefore the accuracy in representing the form, colour and texture will be the chief criteria for the evaluation, alongside accessibility and ease of use.

These projects will present visualisation as it is approached by COSCH. The results will provide information and explanation concerning the advantages and disadvantages of certain techniques depending on the requirements and the goals of different applications for scholarly use, education and appreciation by the general public.

Digital visualisation of material culture offers a major path for its dissemination. Image is essential in order for any cultural asset (object, monument, site) to become known to scholars, scientists and the general public. Visualisation is a communication process and COSCH approaches it in that way. Historical visualisation frequently involves representation — based on the available material and documentary evidence, as well as hypothetical interpretation — of not only the surviving original fabric that can be measured, but also of missing parts and/or those which have been altered in the course of history. The intellectual and methodological transparency of visualisation is therefore essential for its credibility.[8] Various ways of representing and disseminating knowledge of material culture is being explored within the COSCH framework.

Information will be provided to the cultural heritage professionals — archaeologists, art historians, historians, conservators, architects, structural engineers and others — in the form of guides for the use of spatial and multispectral technologies, and for the definition of existing requirements. This information is being incorporated within the COSCH Knowledge Representation, COSCH KR.

7. COSCH Knowledge Representation The above mentioned areas of research contribute to the main objective of the

COSCH Action, i.e. the preparation of a novel, independent, reliable and global knowledge base facilitating the use of today’s and future optical measuring techniques for the documentation of European material cultural heritage. Such a base needs to integrate findings from various fields and will only be reliable, when founded on a common understanding. However, in a multi-disciplinary field of cultural heritage common understanding is a challenge. Every discipline has its own understanding and expectations of a particular subject and question. This often leads to the miscommunication among disciplines within the collaborating domains. The COSCH Action group involves experts from arts and humanities and heritage science, and experts in science and technologies of spatial and spectral data acquisition. Out of this heterogeneous group a common understanding has to be developed leading to the required knowledge base. In order to facilitate this process a framework has been proposed, which forces the discussion to follow clear definitions and semantic schema. The work is expected to result in the Knowledge Representation, COSCH KR, based on semantic technologies.

This ontology knowledge model focuses on mediating information exchange between disciplines. The model brings together dispersed information from heterogeneous sources and structures under a common semantic umbrella. The design and structure of the COSCH model exploit the recent developments in the semantic and knowledge technologies within the Semantic Web framework [9]. Information sources are structured through their logical definitions. The logical structure helps machines ‘understand’, thus assisting humans in the interpretation of knowledge. This logical structure is the first step towards a common understanding of the

interdisciplinary information. The intention is to exploit unifying semantics among knowledge sources, not only to bridge the persistent gaps in knowledge, but also to exploit the use of logic to infer new knowledge discovery from the existing knowledge model.

The COSCH KR is intended to serve both the heritage and technical expert groups, helping interpreting knowledge through a human-machine collaborative approach. Figure 5 illustrates a simplified structure of the COSCH KR.

Fig. 5. Simplified structure of COSCH KR

The structure comprises different knowledge areas, which form the top-level-

ontology of the knowledge model. They can be divided in two broad categories: the expert cultural heritage subject domains (presented above the dotted line in Fig. 5) and technological domains (below the dotted line); some interests and questions are shared by both groups. The intention is to define the semantic components that unify these mostly independent knowledge areas. These semantic components are the interrelationships presented in Fig. 5 through the arrows. The application purposes and requirements lie at the centre of the structure. The nature, behaviour and requirements of the application drive the knowledge processing within the knowledge model. These applications are highly dependent on the objects to be documented and thus have strong correlations with the objects. Consequently, these applications drive technological processes and selection of suitable instruments. Ultimately, the underlying interest in structuring the model, comprising both technical and human science expert fields, is to benefit from enhanced understanding of knowledge of each expert group and to infer a chain of process — from data acquisition to documentation of cultural heritage objects — that is mutually agreeable to both. This could be achieved through reasoning and inference capabilities that exist in the semantic technologies.

Previous attempts at structuring knowledge of cultural heritage have primarily focused on information exchange and dissemination through linked data. CIDOC-CRM, the most used conceptual model to document cultural heritage objects, focuses on how the highest level of data interoperability and information exchange could be maintained [10]. Research projects carried out by the Centre de Recherche et de Restauration des Musées de France (C2RMF) rely on CIDOC-CRM for content

management and retrieval of information. The CIDOC-CRM is used as a base knowledge model and maps its metadata schema for providing semantics to its high-end data [11]. The Europeana project promotes data sharing in cultural heritage through Linked Open Data (LOD) [12]. Moving a step further, the COSCH KR exploits the knowledge technology in discovering underlying knowledge through rule-based semantic interpretations. These discoveries primarily aim to assist the arts and humanities experts with the technical aspects. This kind of knowledge discovery is a relatively new domain in cultural heritage. If exploited, it has a huge potential. COSCH taps into this potential through the COSCH KR.

8. Conclusion

It is hoped that the cross-disciplinary work of COSCH will contribute to improved communication between the subject-domain scholars, scientists and the technical sector. Once tested and evaluated by the community, the COSCH Knowledge Representation, under development, is seen as another significant step towards a better understanding of the material cultural heritage and its 3D documentation. This key objective has been widely recognised throughout Europe, as indicated through the work of a number of European past and ongoing projects and initiatives such as EPOCH, 3D-COFORM, 3D-ICONS and the exemplary work of some international colleagues [13].

9. References [1] Russell, W.J., Abney, W. de W.: Report to the Science and Art Department of

the Committee of the Council on Education, London, HMSO (1888) [2] Bacci, M. (et al.): Colour measurement on paintings, L. Tassi and M.P.

Colombini (Eds.): New trends in Analytical, Environmental and Cultural Heritage Chemistry Developments, RESEARCH SIGNPOST, Trivandrum (India), pp. 333–344 (2008)

[3] Barnes, N.F.: A spectrophotometric study of artists’ pigments. Technical Studies in the Field of Fine Arts, 7, pp. 120–130 (1938). Johnston, R.M., Feller, R.L.: The use of differential spectral curve analysis in the study of museum objects , Dyestuff, 44(9), pp. 1–10 (1963). Johnston, R.M.: Spectrophotometry for the analysis and description of color. Journal of Paint Technology, 39(509), pp. 346–354 (1967). Bullock, L.: Reflectance Spectrophotometry for Measurement of Colour Change. National Gallery Technical Bulletin, 2, pp. 49–55 (1978).

[4] Martinez, K., Hamber, A.: Towards a colormetric digital image archive for the visual arts. Proceedings of the Society of Photo-Optical Instrumentation Engineers, v. 1073, (January 1989).

[5] Parraman C. (Ed.): Colour Coded, Society of Dyers and Colourists, Bradford, West Yorkshire, UK (2010)

[6] Karaszewski, M., Sitnik, R., Bunsch, E., On-line, collision-free positioning of a scanner during fully automated three-dimensional measurement of cultural heritage objects, Robotics and Autonomous Systems, vol. 60 (9), 1205–1219 (2012)

[7] Levente T. and Kato Z.: Targetless Calibration of a Lidar - Perspective Camera Pair. In IEEE Proceedings of ICCV Workshop on Big Data in 3D Computer Vision (BigData3DCV), Sydney, Australia, pages 668–675, (December 2013)

[8] Bentkowska-Kafel A. and Denard H. (eds): Paradata and Transparency in Virtual Heritage. A volume of essays for the AHRC ICT Methods Network series, Digital Research in the Arts and Humanities, Ashgate (2012)

[9] Berners-Lee, T. et al.: The Semantic Web. Scientific American, pp. 34–43 (2001)

[10] Boeuf, P. L.: Definition of the CIDOC Conceptual Reference Model. ICOM/CIDOC CRM Special Interest Group (2013)

[11] Pillay, R. et al.: Archive Visualization and Exploration at the C2RMF. International Cultural Heritage Informatics Meeting (ICHIM07). Ontario, Toronto, Canada: Proceedings. Toronto: Archives & Museum Informatics (2007)

Ribés, A. (et al.): Calibration and SpectralReconstruction for CRISATEL: An Art Painting Multispectral Acquisition System. Journal of Imaging Science and Technology, 49, pp. 563–573 (2005); www.epoch-net.org; http://cordis.europa.eu/publication/rcn/12866_en.html (accessed 9.07.2014)

[12] Europeana, 2014. Making connections. Europeana Foundation. [13] See www.epoch-net.org; http://www.3d-coform.eu/; http://3dicons-project.eu/

(accessed 9.07.2014).

Acknowledgements. The Authors acknowledge the support of the European Cooperation in Science and Technology and wish to thank all who have contributed to the work of COSCH (COST Action TD1201) www.cosch.info. Z. Kato was partially supported by TAMOP-4.2.4.A/2-11-1-2012-0001, the Hungarian National Excellence Program.