These are video presentations from the CAAUK conference, held October 26th-27th in Edinburgh, and covers the 2nd day in the Main Hall.
Where imagination fails: New possibilities of presenting the unexcavated
Josef Souček, Lucia Michielin
Italian “cultural heritage” legislation identifies valorisation of sites as one of the pillars of archaeological research. This results in the rising importance of creating proficient public archaeology plans that will engage the public on a multitude of levels. The Aeclanum excavation, which was born from a partnership between the University of Edinburgh and the Apolline Project, recognises the necessity of including and engaging the public in its newest discoveries. A major part of the ancient town remains unexcavated, however, there is a large amount of new data on the site obtained from a vast geophysical survey carried out over the last two years. Results from this survey are difficult to communicate to the public, but new technology can dramatically improve this situation by employing GPS-based augmented reality. The aim is to create a series of 3D reconstructive models based on the geophysical results. These models will then be input into a system that will allow the public access to these models, when triggered by GPS coordinates or by detecting a predefined image. In this way the visitors will be able to see a rough interpretation and volume of structures on the spot where they used to stand. All of this will offset the need for further extensive excavations that could put the archaeological remains at the risk of deterioration.
Demonstrator for data integration
Ceri Binding, Douglas Tudhope
As part of a semantic data integration case study for ARIADNE, we developed a web demonstrator capable of cross searching data records originating from multiple archaeological datasets combined with selected metadata extracted from multilingual grey literature documents. This was achieved by the use of a common ontological model (CIDOC Conceptual Reference Model – CRM) with names of materials and object types mapped to Getty Art & Architecture Thesaurus (AAT) concepts.
Data cleansing and vocabulary alignment proved problematic with respect to periods and dates. Although dating is obviously a significant aspect of archaeological work the recording of such information in a controlled format is often highly variable in practice, this is inevitable as each local study is probably less concerned with wider interoperability issues. In addition textual representations of date spans obtained via Natural Language Processing (NLP) can have a multitude of incompatible formats leading to subsequent data retrieval issues. Records referring to e.g. Pictish, Iron Age, L’âge du fer, Long Iron Age, Roman, Early Medieval, 405 AD, Early 5th Century, V Secolo a.C. could all actually be contemporary or overlap – but different queries on this data would retrieve different results. An obvious solution was to relate each of the terms to an absolute date span to make the records uniformly temporally comparable. The resulting multilingual data records were directly comparable and searchable by material type, object type and date range, or any combination of these.
One issue encountered in the cleansing/alignment of date spans (and named periods in particular) was the potential for regional differences. In the previous project UK records containing named periods e.g. “Iron Age” were enriched using start/end dates obtained from the Historic England/FISH periods list – however date spans for these periods may well be different in other countries. A recent project (Scottish Archaeological Periods & Ages – ScAPA) has produced an equivalent list for Scottish archaeological periods and associated date ranges with the aim of allowing more fine-grained control in data enrichment of UK records and for use in cross search between separate regional research frameworks.
http://www.ariadne-infrastructure.eu
http://ariadne-lod.isti.cnr.it/description.html
http://hypermedia.research.southwales.ac.uk
Filling the Gaps in a Fragmentary Bronze Object: the Contribution of 3D Scanning and Printing Techniques
M. Page, C. Hochart, S. Bentouati N. Timbart, F. Bridey, C. Boust, G. Obein, A. Razet, M.V. Ortiz Segovia
The project interrogates the possibilities of 3D scanning and 3D printing in the fields of archaeology and conservation-restoration. A bronze rim from the Louvre collections, coming from Susa (actual Iran) and dated from the 20th century BC, is undergoing a restoration work. The rim was made of several pieces, initially joined together on a wooden wheel. We focused on one of the pieces, a hollow rim element of u-shape cross-section, itself broken and covered by sediments.
After scanning all pieces with a structured light instrument, the broken parts were digitally assembled. The scanning process was complicated because of the relief of the object as well as its fine texture: we had to find a compromise between measurement time and quality. The digital reconstitution proved itself useful to the conservator by the fine determination of edges and fractures. From the digital model, we inferred the shape of missing parts and selected texture from extant zones which was applied to the digital filling.
3D scanning gave us the shape and texture of the object, but these are not sufficient to fully characterize its appearance: roughness, gloss and colour should be accounted for if we are to print fillings for the rim element. They were recorded with a micro-topographical device, a gloss-meter and a photographic installation. These data (shape, texture, colour and gloss) were then converted into digital information and sent to the 3D printer developed by Océ Print Logic Technologies, which allows the realistic reproduction of the appearance. This specific 3D printing process was carefully managed for the matte and coloured rendering to be possible in relief, by the addition of micro-structures which reduced the glossy aspect of the inks. Printed fillings could then be inserted during a restoration process to improve the legibility of the artefact. This supposes of course a fine study of compatibility between the print and the conservator’s code of ethics.
Large-Scale Photogrammetric Recording on Commercial Projects
Roberta Marziani, Vijaya Pieterson, Rebecca Hall
The use of 3D modelling in archaeology is being increasingly explored, however it is currently more widely adopted in research and academia than in commercial archaeology. By and large, archaeological site archives generated by commercial units are still more likely to contain hand drawn records than anything produced by more advanced technological means.
Wessex Archaeology is developing its recording methods by using more up-to-date digital recording, primarily in the form of photogrammetry. On sites where grave deposits number in their hundreds, digitals models are being produced as a substitute for hand drawing graves, as well as for other archaeological features, such as structures.
Our adoption of this methodology has encountered a variety of challenges, from software, to staff training, to the sheer number of models to process. Wessex Archaeology are constantly developing the approach to photogrammetric survey on commercial archaeology sites and discovering new ways in which this technology can benefit the fieldwork teams, the company, and the client.
Recording in this manner has proven to be more productive, more efficient, and more accurate. To generate a 3D model of a grave, the feature needs only a series of high quality photographs and survey points produced by a member of the field team. The process takes significantly less time than hand drawing the feature and the output for each model provides more archaeological data in a form that is easier to understand than standard hand-drawn records, due to the presence of photographic imagery and full 3D data. This allows us a far better understanding of the archaeology, and produces more comprehensive reports.
Digital technology also offers us greater opportunities for dissemination to the wider public. Thus despite some of the challenges employing photogrammetry on a very large scale, it is now Wessex Archaeology’s preferred recording method for graves.
This will examine some of the large-scale projects on which Wessex Archaeology has undertaken extensive photogrammetric recording in the last year, and discuss some of the logistical and technical challenges we have faced, and their solutions. It will also look at the conclusions we have drawn from our experiences to date.
Social Media in Digital Heritage Research – data extraction, management and analysis
Marta Krzyzanska, Chiara Bonacchi
The increasingly prominent role of social media in contemporary society has produced a vast amount of social data, often publicly available and easily accessible through the APIs of platforms such as Facebook and Twitter. This ‘big data’ has recently been leveraged to research a range of social phenomena, including the role of heritage in the construction of political identities (Bonacchi et al; forthcoming 2018). In this paper, we will discuss methodologies for conducting ‘big social data’ research, and compare a range of techniques of data extraction, management and analysis. We will examine the advantages and limitations of using platforms’ APIs, and present alternative procedures for data collection, such as Scrapy. We will also compare different formal methods of analysis, and their implementations via different software solutions and programming languages including R and Python. The methods we will review topic modelling, sentiment analysis and network analysis, and the paper will critically reflect on their applicability to different types of data, in particular those extracted from Facebook and Twitter. Finally, we will present a strategy for the management and integration of large amounts of unstructured web datasets that utilises the non-relational Mongo Database and cloud computing. This paper draws on examples from the ongoing Digital Heritage research that we are conducting as part of the ‘Ancient Identities in Modern Britain’ project (ancientidentities.org).
A QGIS plugin to explore landscape connectivity through circuit theory
Guillem Domingo Ribas
Archaeological studies traditionally make use of Least-Cost Path Networks to study landscape mobility. This method is the most common approach to predict or reconstruct ancient routes, whose path is seldom known. The reconstruction of communication lines in the landscape through LCP is often criticised as it can only compute a single route between two points. Nevertheless, different approaches to landscape mobility may help to overcome this constraint. Functional connectivity models have been developed by other fields such as ecology, evolution and conservation, which have various advantages over LCP.
One of these explores circuit theory as well as random walk theory and its application in landscape connectivity. It is an approach that has been broadly used in several areas with success. Landscape is displayed by networks of nodes or raster cells to create circuit representations of a territory, which will be used to calculate the approximate resistance values of the different features within it. From this framework, circuit-theoretic models will provide distinct possible paths. Thus, storing various feasible routes can be more effective than having one single optimal path with views to studying mobility.
The current research aims to create a new plugin for QGIS with Python which implements this alternative in the study of landscape archaeology. After the creation of the new tool, the plugin will be tested with the EUROEVOL dataset (https://www.ucl.ac.uk/euroevol), specifically for the Neolithic transition in Central Europe. This case study will be analysed with LCP and the new approach, whose preliminary results will be compared and presented in the congress. Moreover, the presentation will focus on the benefits and constraints for Archaeological research identified in both outcomes.
The plugin will be available in the QGIS repository and it will also become a beneficial tool for researchers in the aforementioned fields.
Classifying Digital Research Methods: Linked Ancient World Data in Context
Sarah Middle
Linked Data is an implementation of the Semantic Web, connecting digital objects across datasets in a machine-readable way. It facilitates searching across multiple collections and reveals relationships between resources that could not otherwise be determined. As such, Linked Data has the potential to transform research involving online material, including that done in the Humanities. However, its adoption in this area remains minimal, and little research has been conducted into the use of existing Linked Data tools/resources (as opposed to production of new ones). Within the Humanities, there has been a relatively high uptake of Linked Data in disciplines relating to the Ancient World. My PhD looks at how this approach might be integrated with existing methodologies for Ancient World research. This includes investigating the end user experience of Linked Data tools/resources and identifying specific research methods for which they are (or might be) used.
The user research component of my PhD will start with a survey of Ancient World researchers in spring 2018, about their experiences of using/producing Linked Data in the context of digital tools/resources more generally. Research methods associated with tools/resources by survey participants will be classified using the Taxonomy for Digital Research Activities in the Humanities (TaDiRAH – http://tadirah.dariah.eu/) – an ontology describing research methods, techniques and information resources. TaDiRAH itself utilises Linked Data technologies, and will be used as a framework to connect data about the tools/resources.
Visualising relationships between Linked Ancient World Data tools/resources based on their associated research methods will provide an insight into methodologies that might be particularly amenable to Linked Data integration. Analysing this data alongside users’ comments about their experiences of these tools/resources will demonstrate their relative effectiveness when performing those research methods, in comparison with non-Linked Data tools/resources.
This paper will start with a brief overview of Linked Data, and the survey construction methodology. I will then explore what my findings demonstrate about the current state of Linked Data use in Archaeology and related disciplines, with reference to example tools/resources, and conclude with recommendations for future tool/resource development.
Offshore Scotland and archaeology: the challenge of big data
Stephanie Said, Abby Mynett
Offshore development work, undertaken for the developers of windfarms, aggregate dredging, ports and cables and pipelines, now accounts for the bulk of commercial marine archaeological work carried out in the UK. This work depends very heavily upon access to the often huge quantities of marine geophysical and geotechnical data acquired by offshore survey companies for these schemes. With a particular emphasis on offshore Scotland, this paper will examine when and how this data is processed and interpreted by archaeologists and how the results are used by archaeologists to inform mitigation measures put in place by regulators and developers to protect and preserve archaeology during the construction phases of these schemes.
HES▪SIGMA: a new geo-spatial data capture tool for managing the conservation of historic assets
E.A. Tracey, N. Smith, K. Lawrie, J. Russell
The British Geological Survey (BGS) has worked with Historic Environment Scotland (HES) to develop an integrated digital site assessment system that provides a refined survey process for historic assets.
HES has a statutory obligation to conserve the 336 properties in care of Scottish Ministers without compromising the cultural significance of the assets. The condition, statutory compliance, operations and interventions associated with these assets must be appropriately recorded, monitored, reported and reviewed. Similar to other heritage agencies, HES needs a system that can store and present conservation and maintenance information for historic assets and be used to plan effective programmes of maintenance and repair. To meet this need, BGS has worked with HES to develop new, customised tools for use within the existing BGS System for Integrated Geoscience Mapping (BGS▪SIGMA)—an integrated workflow underpinned by a geo-spatial platform for data capture and interpretation—resulting in the provision of an application specifically designed to allow the capture of digital data relating to monument condition.
The new system is named HES▪SIGMA. It is built on top of ESRI’s ArcGIS software and is underpinned by a bespoke relational data model. It provides users with the ability to populate custom-built data entry forms to record maintenance issues and repair specifications for architectural elements ranging from individual blocks of stone to entire building elevations. Photographs, sketches, and digital documents can be linked to architectural elements to enhance the usability of the data.
The benefits of digital data capture with HES▪SIGMA are huge: standardised data formats with supporting dictionaries to constrain the input parameters ensure a high degree of consistency and facilitate data extraction, querying and geospatial analysis; custom-built reporting tools produce fully formatted reports based on collected data; date and time stamps combined with the capability to re-survey monuments at regular intervals facilitates condition monitoring over time, all of which creates a versatile planning tool for scheduling works, specifying materials, identifying skills needed for repairs, and allocating resources.
The digital data collection principles on which HES▪SIGMA is built, i.e. the capability to capture attributed data geospatially, provides many potential applications in conservation, archaeology and similar fields.
Wemyss Caves 4D: a digital future for the Pictish Wemyss Caves, Fife Scotland
Mike Arrowsmith
Wemyss Caves 4D is a collaborative digital documentation and interpretation project involving archaeologists from the SCAPE Trust, University of St Andrews and the community group Save the Wemyss Ancient Caves Society (SWACS). The project website introduces a worldwide audience to renowned Pictish carvings inscribed into the walls of the Wemyss Caves in Fife, Scotland. Most of the caves are difficult to access and all are directly threatened by coastal erosion, structural instability and vandalism.
Laser scanning and convergence photogrammetric survey techniques were applied to eight hundred metres of coastline and eight caves to document the carvings within their setting. Individual carvings were recorded by community volunteers using Reflectance Transformation Imaging (RTI).
The massive resulting dataset was simplified and combined with historic content to present a coherent interactive digital resource that allows users to explore a realistic virtual model of the caves; examine the carvings; access additional information, and even travel back in time to see the caves as the Picts would have done.
Within each cave, context-relevant material (image gallery, video, RTI scan etc.) can be called up via a popup viewer. These viewers are self-contained modules which integrate into the main code, allowing for more content types to be seamlessly added to the site via the application programming interface (API).
Achieving a responsive design for the website necessitated amending the WebRTI viewer developed by the Visual Computing Lab of the Italian National Research Council and further modifications to other third party components such as the panorama viewer to enable communication between them and the core code.
Content is also organised through a filterable catalogue allowing the considerable resources to be accessed through a single click. The back-end database allows us to easily add new or rediscovered content.
This presentation will discuss the design and technical challenges of the website, in which we sought to create a balance between the interactive interface designed to give some simulation of the experience of being in the caves with a practical and intuitive way of accessing the rich and detailed content relating to the caves and carvings.