ABSTRACT

The activities of COSCH community and the disciplines it represents were as diverse as they could possibly be in research into cultural heritage. To achieve common goals it was of utmost importance to have a common understanding of these diverse activities and disciplines. Work on the COSCH Knowledge Representation, or COSCHKR, was undertaken to develop a common semantic base representing different disciplines and to facilitate communication within the Action. The COSCHKR is an ontology-based inference model, guided by inference rules that provide a semantic bridge between various interdisciplinary activities involved in non-invasive technical documentation of material cultural heritage. The model is intended to support humanities experts by recommending optimal spatial and spectral techniques. The model may also be used by technology experts to compare their own solutions with the ones recommended through COSCHKR, and to understand why they may differ.

In this chapter we present the methods adopted for designing the COSCHKR and the steps in the development of the inference model. The difficulties in maintaining a common level of understanding within the diverse disciplines during the knowledge acquisition process are discussed. We present mechanisms and methods of information collection, its structuring, and aligning, to formulate different axioms and theorems within the model. The design and development of COSCHKR was based on an iterative procedure where the gathered knowledge was first verified with the group of experts before it was processed. This verification mechanism was important for the reliability of the model, ensuring technical consistency. This chapter highlights the importance of these iterative mechanisms in the validation of knowledge gathered and then information populated inside the knowledge base.

Keywords: COSCH ontology, knowledge representation, cultural heritage recording, Semantic Web technologies, inference system

Introduction

Spectral or spatial recording of material cultural heritage (CH) is an interdisciplinary task involving scientists and experts from the humanities, information technologies, and engineering. A mutual understanding and agreement about the complexities of undertaking such recoding tasks is necessary in order to deliver appropriate spectral or spatial data of physical cultural heritage objects adapted to the needs of cultural heritage experts who rely on and work with those data. The COST Action TD1201: Colour and Space in Cultural Heritage (COSCH) (Boochs 2012; Boochs et al. 2013), brought together these experts to enhance mutual understanding among the disciplines in order to better record and preserve material cultural heritage.

ABSTRACT

The study of cultural heritage requires the expertise of many disciplines as different as history and optical science. Each discipline brings its discrete research methods and know-how. Complementary knowledge is gained, gradually, through effective collaboration, which is not always easy. The articulation and communication of respective research questions, ways of addressing these questions, and expected outcomes, are a challenge in all collaborative projects. This concluding chapter discusses how the COST Action, Colour in Space in Cultural Heritage, has addressed and managed this challenge, through the case studies described in the present book and other activities. The authors draw conclusions from COSCH research activities and note some general points that bear on the success of diffusion of interdisciplinary research and its future.

Keywords: interdisciplinary research, communication, cultural heritage, science, optical technologies, COSCH

The study and documentation of cultural heritage require the expertise of many disciplines as different as history and optical science. Each discipline brings its discrete research methods and know-how. Complementary knowledge is gained, gradually, through effective collaboration. The articulation and communication of respective research questions, ways of addressing these questions, and expected outcomes, are challenges common to all collaborative projects. This concluding chapter discusses how the COST Action, Colour in Space in Cultural Heritage (COSCH) has addressed and managed these challenges, through the case studies, described in the preceding chapters, and other research and training activities conducted between 2012 and 2016. We reflect on earlier projects that also shared the goal of “bridging the gap” between disciplinary approaches and mindsets, specialist research methods and tools, while showing the distinctive approach of the COSCH network. We draw conclusions from COSCH's four-year engagement in the interdisciplinary debate by noting general points that, in our view, bear on the success of future research and its diffusion.

On the Perceived Notion of “Two Cultures”

“The scientist has the habit of science, the artist the habit of art.”1 The concept of “two cultures” has persisted since it was forcefully proposed in the 1950s. It is not only widely understood, but also cultivated in the English-speaking academia, as evidenced by a recent, double issue of Interdisciplinary Science Reviews (James 2016) that elaborates on its legacy.

ABSTRACT

A study was undertaken by an interdisciplinary group of COSCH researchers between 2014 and 2016, to record and examine silver coins believed to be ancient Roman in origin. The aim was to evaluate the suitability of various advanced, non-invasive optical and spectrometric techniques for analysing the physical characteristics and elemental composition of numismatic objects. In order to compare results, the same two silver coins were used throughout the study: two denarii portraying Empress Faustina I, wife of Antoninus Pius, believed to be posthumous deification issues of AD 141. The questions addressed included the characterization and authentication of test coins, as well as technical issues in the multimodal recording of material cultural heritage with metadata and paradata. The team investigated whether the methods chosen for this case study responded to the needs of numismatists, and whether they could feasibly be applied in museum practice, to support research and conservation of historic coins, and to enhance the documentation and dissemination of numismatic objects through heritage science.

Keywords: numismatic studies, Roman silver coins, examination, optical techniques, multimodal recording, COSCH

Background and Purpose of the Study

This chapter introduces an experimental study of two silver coins, believed to be ancient Roman, and discusses some of its interdisciplinary outcomes. We undertook to investigate, in collaboration with other international researchers, how historic coins are examined and documented, and what is the role of modern, non-invasive optical and spectrometric technologies in today's numismatic research and museum practice. The network Colour and Space in Cultural Heritage (COSCH) facilitated research contacts between experts representing a wide range of interdisciplinary interests and expertise in optical, metrological and imaging technologies and material sciences. Benefiting from this combined expertise and access to specialist equipment, at a level that is rarely available to museums, this case study was designed to explore which of the advanced technologies available to us could enhance numismatic research and conservation, and museum practice in general. Therefore, our main objectives were:

• to assess current research and documentation practice: how are historic coins in numismatic museum collections being examined and recorded, presented and disseminated in the 2010s? What is the role of digital methods?

• to apply a wide range of optical and spectrometric techniques to measure and examine, and present the test coins;

• to compare the results of different technologies applied for the same purpose, in particular to compare the geometric measurements achieved through several 3D techniques;

Definition

Three-dimensional depth sensing can be defined as the ability or capability to tell the distances from a reference point, or observer, to all objects in a 3D scene. In technological terms, it requires a system able to take measurements, or employ computational methods, or a combination of those, to estimate distances from the system itself to all objects in its surroundings. In cultural heritage applications 3D depth sensing is at the heart of most 3D digitization systems (in some cases termed 3D optical documentation). Several restrictions apply, including the proper handling of valuable and fragile historic objects.

Description

Technically, 3D depth sensing is the process of acquiring the distances of all points in a 3D scene from a fixed point of reference, which in most cases is the sensing device itself. There are a number of different methods and technologies for sensing the depth in 3D space. Usually these methods are categorized into light-dependent and light-independent: a light-dependent system senses light, in any way, in order to assess the three spatial dimensions, whereas light-independent employs methods not directly sensing light, but using geometrical and topographical principles. In addition, mainly within the category of light-dependent methods, a further sub-classification is based on the principle used or the technology being applied. These are the active and passive methods: active in a sense that the measuring system formulates the conditions of the measurements in order to acquire depth, whereas passive methods rely purely on computation applied to raw “unsupervised” measurements.

Numerous methods can be found in the literature, and here is a list of the most cited in their category:

• 1 Light-dependent methods

• a Active methods

• i Laser triangulation

• ii Time-of-flight scanning (or LiDAR, LADAR, range scanning)

• iii Structured-light scanning

• iv Shape from Photometry

• v Shape from Shading

• vi Shape from Shadow

• vii Tomography (of any kind)

• viii Holography

• b Passive methods

• i Photogrammetry

• ii Structure from Motion

• iii Shape from Silhouette

• iv Shape from Stereo

• v Shape from Texture

• vi Shape from Focus (zooming)

• vii Microscopy (of any kind)

• 2 Light-independent methods

• a Topographic methods

• b Empirical methods

• c Contact sensing methods.

ABSTRACT

Accuracy, artefact, feature, precision, reconstruction, resolution, texture, uncertainty are words central to many discussions of the documentation of cultural heritage. This terminology, whilst broadly understood across the disciplines, is often misunderstood due to its specific use in particular cases. An interdisciplinary dialogue conducted over a period of years and comprising experts in a range of fields—art history, colour science, engineering, semantics, mathematics, cultural heritage, museum studies, and others—has yielded a challenging discussion document that considers the thorny issue of a shared understanding of a set of keywords. On occasion our perceived shared language is not shared at all but reveals—at times through subtle nuance, and yet at times through gaping chasm—the disciplinary subjectivities we hold unbeknownst to ourselves. Mutual understanding of some of these key terms is central to any newly engaged, participatory transdisciplinary endeavour that seeks to develop critical methods for the documentation, analysis, preservation, and sharing of cultural heritage objects outside the traditional disciplinary silos. This chapter charts the interdisciplinary discussion towards a common understanding of terminologies used in cultural heritage. It is a discussion that recognizes critical differences or common misuse, and aims to contribute to a shared understanding that may be useful for all knowledge domains in the field. The chapter summarizes the work of a number of Think Tanks conducted by Early Career Investigators participating in the COSCH network.

Keywords: cultural heritage, interdisciplinary research terminology, metrology, humanities, digital documentation, COSCH

Introduction

The digital documentation of material cultural heritage is a multidisciplinary task and often involves experts in spectral and spatial recording, and Information and Communication Technology (ICT). When generating digital representations of objects of cultural heritage, a shared understanding is necessary to develop suitable and user-oriented solutions. However, reaching a shared understanding is complex. It begins with a clear definition of the project's aim(s), including the explanation of concepts, terms, and procedures that may be unknown to some project partners. These first steps may seem obvious and should be the basis not only for multidisciplinary projects, but for all projects. However, these tasks are more multifaceted in multidisciplinary projects as, in addition to unfamiliar disciplinary terminology, each project partner might use terms that are common to various domains, but may have inherently different meanings.

COSCH Case Studies that have employed this technology: Roman coins, Germolles

Definition

XRF spectrometry is a non-destructive analytical technique widely used to determine the elemental composition of materials. It has been demonstrated to be effective in fields including biology, medicine, geology, material science, environmental studies, and cultural heritage, among many others. The suitability of this technique for conducting elemental surveys that require no extraction of samples makes it very versatile in cultural heritage research.

Description

X-ray fluorescence is induced when photons with enough energy, emitted from an X-ray source, interact with a material. High-energy photons induce ionization of inner shell electrons through the photoelectric effect, creating electron vacancies in these shells. These vacancies are almost immediately filled with electrons from outer shells resulting in emission of fluorescent radiation, which is characteristic for each element. The lines observed in the XRF spectrum enable identification of the chemical elements present. XRF analysis is therefore a powerful analytical tool for the spectrochemical characterization of most elements present in an object.

An XRF spectrometer contains several components including: an excitation source or X-ray tube, a sample chamber or open shutter system to deliver X-rays to the sample, a detector to determine the characteristic X-rays generated, an analyser that converts the measured energies to their correspondent electronic transitions, and a display device to visualize the measured spectra. Early XRF instruments contained a sample chamber, typically of limited size, that allowed only for the measurement of samples extracted from cultural heritage objects. In response to the needs of heritage scientists, modern instruments employ open beam systems that allow simultaneous X-ray irradiation and detection of fluorescence emitted by actual historic objects of all sizes without requiring sampling. The detector is composed of two charged electrodes that have a non-conducting or semi-conducting material positioned between them. The X-rays ionize these materials causing them to become conductive. The unbound electrons are accelerated toward the detector anode to generate a signal that can be measured. Based on the detector used, X-ray techniques can be divided into energy (EDXRF) and wavelength dispersive (WDXRF).

Two types of analysis are possible in XRF. Qualitative analysis allows identification of the elements present in the object, and has been used on paper artworks, paintings, and ceramics. Quantitative analysis involves the determination of the relative amount of each element present in an object.

ABSTRACT

The fragmentary condition of objects is often an issue in the study of material cultural heritage. In archaeology, and in pottery studies in particular, the fragmentary condition of excavated objects impacts on research into their history and presentation. Ceramic vessels and vase fragments are the most numerous archaeological findings and a primary source of information about various aspects of ancient life: private, public, religious, economic and technological, social and artistic. The subject of this chapter is a fragmentary clay drinking cup, kantharos, a vessel attributed to the god Dionysus and a typical drinking vase used at symposia (gatherings). This particular kantharos was unearthed during the excavations at the ancient settlement of Therme, today's Karabournaki near Thessaloniki, Greece. The vase dates to the Archaic period between the seventh and the sixth century BC.

Although the kantharos is preserved to a large extent, its fragmentary condition challenges complete reconstruction. Evidence for reliable reconstruction is insufficient: the lower part that would originally have consisted of a base and foot, is missing. The process of virtual reconstruction through 3D visualization, described in this chapter, has contributed significantly to the study and presentation of the vase. The authors consider the advantages and limitations of technologies used. The process of creating this particular computer model may be applied to other fragmentary vases that come either from the excavation at Karabournaki or any other archaeological site or collection. This research may be of interest to experts in 3D technologies, as well as archaeologists and art historians, both academic scholars and students, museum curators and conservators, educators and other multidisciplinary audiences.

Keywords: Greek pottery, kantharoi, fragmented objects, archaeological excavation data, virtual reassembly, 3D approximation, visualization, COSCH

Introduction

The focus of this chapter is on how completion of pottery sherds through three-dimensional (3D) digital visualization contributes to archaeological research. Pottery is usually discovered in fragments and in vast numbers. Displayed in museums or deposited in storerooms, the clay sherds play an essential role in reconstructing the past. They are primary sources of information about various aspects of ancient life: private, public, and religious; economic and technological; social and artistic. They are, indirectly, also essential in archaeology for establishing the chronology of the finds under investigation (Biers 1992). For this reason pottery studies are a principal area of archaeological expertise (Orton and Hughes 2013).

The COSCH Case Study that has employed this technology: Roman coins (fig. 15.1)

Definition

Focus stacking is a technique which extends the depth of focus of an image by combining different images with low depth of field. It is mostly used in close-up and macrophotography since in those cases the lens, even at high f-stop, is not able to render everything in focus. This method, although not used often in cultural heritage, is useful to document objects and to illustrate details. It can also be used in combination with Structure from Motion (SfM) in order to create accurate 3D models of small artefacts.

Description

Focus stacking allows the extension of the focus range of a single view, using the lens at its optimal f-number (typically in the range f/8 to f/11). If the lens is stopped down more than its optimal f-number, the depth of field is larger, but the image is softer (less sharp) due to diffraction. The starting point for focus stacking is a series of images, with different regions in focus. Although none of the images has the object of interest entirely in focus, collectively they contain all of the data required to generate an image that has all parts in focus. The in-focus regions of each image may be selected manually or detected automatically, for example, via edge detection or a sharpness metric or Fourier analysis. The in-focus regions are then blended together, pixel by pixel, to generate the final image.

This process is also known as z-stacking or focal plane merging.

Significant Applications

Example 1: Rock Art Studies

Plisson and Zotkina (2015) tested the potential of focus stacking for 3D recording of submillimetric details of prehistoric petroglyphs and paintings. They tested their technique at open air sites in France, Portugal, and Russia. The study shows that focus stacking and photogrammetry are complementary.

Plisson, H., Zotkina, L. V. 2015. “From 2D to 3D at Macro-and Microscopic Scale in Rock Art Studies.” Digital Applications in Archaeology and Cultural Heritage 2.2: 102‒19.

Example 2: Venus of Frasassi

The Venus of Frasassi is an 87 mm high Palaeolithic sculpture carved in a stalactite. It was found in a cave in the Frasassi Gorge (hence the name) in Central Italy and is housed in the National Archaeological Museum of the Marche in Ancona. The figurine was recorded using focus stacking pictures combined with photogrammetry to achieve a 3D model.

ABSTRACT

The aim of this study was to examine and document the wall paintings in the Château de Germolles. Situated in Burgundy, France, Germolles is the best preserved residence of the Dukes of Burgundy and was listed as a monument of national importance in 1989.

The medieval wall decoration of the Château de Germolles was rediscovered under the nineteenth-century plasters during World War II. Medieval accounts of the château provide a detailed list of the materials acquired to make the mural decoration, but this list is incongruous when compared with the current appearance of the paintings. The discrepancy between the archival and material evidence, and also the need to understand the complexity of the painting technique used were the main motivations for undertaking the case study described in this chapter. Imaging alongside more traditional examination techniques were utilized to record and document the mural decoration. The objectives of the case study were to distinguish the original materials from those applied during restoration, identify those materials, and correlate them with the archives. We also tried to understand the medieval painting techniques used and assess the condition of the paintings and stabilization requirements. Finally we aimed to find a sustainable solution for the management of the various types of data collected. Various techniques and investigations offered valuable insights into the materials and the painting technique used. To improve visitor experience, based on the information gained in the course of this study, a 3D virtual representation of the original decoration is currently proposed for display to the public visiting the Château de Germolles.

Keywords: Château de Germolles, dukes of Burgundy, Middle Ages, wall paintings, tin leaf decoration, spatial and spectral imaging techniques, COSCH

Introduction

The Château de Germolles dates from the fourteenth century and is the best preserved residence of the Dukes of Burgundy. The wall paintings decorating the ducal piano nobile are a unique manifestation of the courtly love that permeated the courts of French and Italian dukes and princes of the time. Concealed sometime in the nineteenth century, the wall paintings were accidentally rediscovered during World War II. They were non-professionally uncovered in the 1970s and restored in 1989–95.

COSCH Case Studies that have employed this technology: Roman Coins, Kantharos, White Bastion

Definition

Structured light 3D scanners project a known pattern of light (stripes, dots), typically regular and periodic, onto the object. The result is captured using one or more cameras, and the 3D information of the object is recovered by software using different triangulation or projection geometries. Very dense and accurate point clouds may be obtained. The configuration and approach used enables adjustment of the surface resolution, using multiple exposures, or adjusting the field of view of the system.

Description

A structured light 3D scanning system is non-contact and consists of a projector and at least one camera. Data acquisition and analysis are controlled by dedicated software running on personal computers; both are integral parts of the measuring system. The calculation of 3D data is based on the triangulation principle; typical triangulation angles are approximately 30°. The camera is mounted in a calibrated position relative to the projector, and the scanning system projects light patterns onto the object surface. The contrast of the light patterns influences the quality of scan results. Therefore, working in direct sunlight is often not possible or advisable. For good results, working indoors or in shaded areas outdoors is recommended. Scanning of reflective or (semi-)transparent surfaces is problematic for any optical scanning system. Difficult surfaces include shiny metal, glass, marble, bones, teeth, and many plastic materials. Covering the surface with whitening spray eliminates reflection or transparency problems, but is not applicable to all objects. Scanning critical objects that may not be spray-coated is often possible with additional effort, by further scan positions or angles, user interaction or software filtering.

The object size covered in a single scan is limited by the projector's brightness as well as the feasible distance between projector and camera. The field of view for a single scan setup typically ranges from 30 mm to 2000 mm; the camera system's depth of field determines the system's measuring depth. Furthermore, there is a link to depth resolution, as larger areas imply lower resolution/accuracy because spatial resolution is dependent mostly on the number of samples per unit distance, so denser sampling means, in most approaches, a better depth detail under comparable angle of view (disparity) conditions.

This book presents some of the outcomes of interdisciplinary research and debates conducted by the participants in the international network, Colour and Space in Cultural Heritage (COSCH) between 2012 and 2016. The book adds to a large body of literature on the applications of digital technologies to the study and preservation of cultural heritage. So why was another book on the subject needed? In this introduction the rationale for the book, its scope, and methodology are explained.

Cultural Heritage

After a certain period of time, human cultural activities and their products acquire the status of cultural heritage. The International Council on Monuments and Sites (ICOMOS 2002, 21) defines cultural heritage broadly as “an expression of the ways of living developed by a community and passed on from generation to generation.” Some standard definitions of cultural heritage, both tangible and intangible, in the context of 3D documentation, have been collated by the consortium, 3D-COFORM (Arnold et al. 2009, 16‒19).

No single book can cover the variety and richness of cultural heritage, or the methods of its documentation. The limitations are inevitable. The COSCH network limited its interests to material objects of the cultural heritage of Europe, some of which are covered in this book. The objects and sites come from different periods and bear witness to different cultures. They vary in scale and materials, significance and value. They include an ancient Greek vase, the Karabournaki kantharos (chapter 2); Roman silver coins, the denarii of Faustina the Elder (chapter 3); painted wall decoration of the medieval French château of Germolles (chapter 4); a fortress overlooking Sarajevo, steeped in the multicultural, turbulent history of Bosnia (chapter 5); a medieval wooden shipwreck, known as a cog, lifted from a river bed in Germany (chapter 7); and a variety of historic objects housed by the National Museum of Romanian History, Bucharest, including icons, illuminated manuscripts, and pottery (chapter 6). Other works of world art and architecture, as significant as the ancient, rock-cut statues of Buddha in the Bamiyan Valley in Afghanistan, demolished by the Taliban in 2001, are mentioned in the second part of the book. This small selection is representative of other art, crafts, and architecture, in various states of preservation and restoration that can be found in situ or housed in museums.

COSCH Case Studies that have employed this technology: Roman coins, Germolles, Kantharos, Bremen Cog

Definition

Structure from Motion (SfM) uses the principle that movement through a scene allows an understanding of the shape of objects within the scene in three dimensions, in the same way as walking through a room allows one to visualize the space and objects within it. In SfM the movement is represented by a series of systematic viewpoints; overlapping photographs taken from different locations around the object. This can be achieved from the ground in the field, in a photo studio, or from the air with a drone or other unmanned aerial vehicle (UAV).

Description

SfM is a method of photogrammetric recording. It is used for area-based recording and for object recording. A series of single images are photographed and the reconstruction of the 3D model uses similar steps to a photogrammetric workflow with orientation through image point comparison and bundle adjustment, measurement and analysis based on internal and external geometry, possibly using image masking, and output of a coloured point cloud or polygon mesh. Workflows include the combination of free software for photogrammetry, or licensed but affordable software, increasingly tailored for easy use. Whereas photogrammetry was previously used to measure a set of discrete points, typically using markers placed on objects within the scene, SfM extends the method (without the need for markers) by automatically finding feature correspondences and using dense matching techniques to reconstruct complete surfaces. The drawback is that where photographic coverage is poor the point clouds generated by SfM may be quite noisy and have “holes” in the surfaces represented by the point clouds, requiring subsequent smoothing and filling operations.

SfM can be used for the archaeological investigation of landscapes and built structures. The use of a kite or UAV can enable coverage of a large area in high resolution, allowing earthworks to be traced across the landscape. When used in conjunction with an archaeological excavation, its simplicity, low cost, high coverage and speed of recording, allows individual aspects of the 3D data to be analysed in ways that are simply not possible with 2D records. The amount of detail that can be captured makes this an ideal means of recording features that could be damaged or destroyed, such as fragile wooden objects, landscapes that are to be developed, or stratigraphic layers that must be removed.