“KARABOURNAKI-RECORDING THE PAST”: THE

DIGITIZATION OF AN ARCHAEOLOGICAL SITE

D. Tsiafakis, N. Tsirliganis, G. Pavlidis, V. Evangelidis, C. Chamzas
Cultural & Educational Technology Institute, Xanthi, Hellas

Abstract – “Karabournaki-Recording the Past” is a project regarding the digital

documentation of an archaeological site using as a case study the site of Karabournaki
located in the area of Thessaloniki (Greece). Focus of the project is to design, develop,
and implement a multimedia cultural database system capturing the full amount of the
available information regarding the site, including extended search and visualization
capabilities that can deliver its multilingual content over the Internet. The meta-data
that are produced with the completion of the project contribute significantly to the study
and publication of any archaeological site as well as its preservation, succeeding the final
goal of making it universally accessible.

INTRODUCTION
Archaeology as a discipline is usually based on the systematic collection, analysis
and interpretation of data through excavation and other forms of fieldwork. Data
collected in an excavation are often complex, multivariate and three dimensional,
making the development of new ways to efficiently and accurately record and
manipulate those data, an absolute necessity for archaeologists.
Traditionally the collection of data was based mostly on a systematic description
and recording on paper (notebooks, forms etc), while the visual components of the
archaeological information were captured in a variety of 2D means like photographs,
architectural and topographical drawings, maps etc. Three major unsolved problems
rise out of those traditional methods:
1. Vast amounts of data recorded using these methods is the major concern of
every large-scale excavation (usually lasting many years or decades) in respect
not only to their storage and preservation, but also to their handling, analysis,
study, and publication. Although exhaustive systems of recording, labeling and
storing have been devised and employed through the years, the problem of
handling, combining and retrieving data “randomly” (and at will) remains a
much wanted necessity in archaeology.
2. The majority of data are visual and three-dimensional (3D). Therefore, it is
obvious that the limitations imposed by the nature of two-dimensional (2D)
formats put significant constraints to our perception of both geographical space
and context.
3. Finally, there is always the problem of effectively presenting spatial and 3D
data to scholars and the general public, especially when interactivity and
accessibility are listed among the basic intentions of the publication.
The emergence of information technologies offers presently powerful solutions to
these problems, with additional benefits such as less consumed time greater analytical
insight.
Facing the problem of recording the total amount of the available excavation
material of the archaeological site at Karabournaki in combination to its presentation
and publication, an Integrated System of Digital Management of an Excavation was
developed [1]. Karabournaki is located in North Aegean, in the area of Thessaloniki.
The site dates from the Late Bronze Age down to the Roman times, with a flourishing
period during the Archaic times (7th - 6th centuries B.C.) and it preserves the
architectural remains of a settlement including a harbor and its cemeteries.[2],[3].

RECORDING THE PAST IN NEW WAYS

Contemporary methods in digital recording, data management, visualization and
dissemination offer today the ability to completely record and disseminate an
archaeological site and its findings. In the following paragraphs, we show our
approach to solving the problems involved in an attempt to integrate all methods and
data into a complete digital archaeological site.

Digital recording
First step in every digital representation application is the digitization that leads to
a “sampled” and “discretized” version (or representation) of the real world.
Digitization is targeted upon every data form, but since visual data are playing the
most significant role in comprehension, digital representation of visual data is a field
of continuous research and development. Digitization of an archaeological site
involves many aspects. The “Karabournaki-Recording the Past” system, as a
collaborative project of experts in various fields, had to take under consideration all
these aspects. The product involves three major categories of digitization that
correspond to the total amount of data coming out of the excavation:
• Landscape (site) and structures digitization: complete digitization of the
archaeological site using contemporary 3D scanning methods applicable to
landscapes, structures and buildings. These methods involve the usage of either
laser devices, either photographic/photogrametric methods and remote sensing,
either hybrid-combined methods. Traditional methods were also employed:
usage of (digitized) standard 2D topographical maps and 3D digital
reconstruction through specifically designed software tools. Furthermore, since
the site is described both in drawings and documents, 2D visual data were also
included in the digitization process.
• Objects (findings) digitization: complete digitization of all objects in 2D and
3D. The present practice guides that objects are recorded through multiple
photographs and entries in the excavation notebooks, as well as other
referencing documents. Our first approach was to digitize all 2D visual
material concerning the findings of the excavation. The next step was to record
the objects in 3D. As will be explained in a following section, 3D technologies
offer the ability to completely reconstruct objects with missing parts using
scans of the findings and archaeological data. Fig. 1 shows two methods of 3D
object scanning (laser and photographic) and a method of surface
physicochemical properties extraction.
• Documents (notebooks and forms) digitization: complete digitization of the
documentary material in the form of 2D images. This process could also
include the digital reproduction of all texts, so that full text search capabilities
can be integrated. In its present form the system processes all documentary
material as 2D digital images.
To produce and maintain a completely digital archaeological site, it is imperative
that after the digitization of the existing material, digital recording of new facts and
findings is continued through the usage of specifically designed software tools that
permit the archaeologists to go on with the digitization without any help from
technical staff. To this very end, special purpose, user-friendly software tools are
provided, with the additional ability to record data on-site: the archaeologists can
record current findings using a PC or a laptop on-site, and have the ability to easily
integrate the new data into the overall data management system.
Summarizing, for the needs of Karabournaki, a 2D and 3D digitization strategy
have been adopted, accompanied by 3D digital reconstruction (where possible) and
GIS functionality. According to this strategy, all visual material is being digitized
using high resolution 2D scanners and 3D laser and photographic techniques [4].
Archaeometry data are being produced by using contemporary analysis tools, like
XRF scanners [5], and object surface data are being acquired and recorded. These
data involve physical or chemical properties measured in a point-wise manner, on
significant areas of the objects' surface.

(a) (b) (c)

Fig. 1. 3D scanning of object geometry, texture and surface properties

Digital data management

All digitization processes produce a large amount of digital data in many digital
file forms. This amount of information no matter how precise and complete is, can be
rendered virtually useless if one is not able to manage it efficiently. Traditional
methods of data management involve the design and usage of relational models to
represent and combine the data, by imposing relations on each other imitating human
cognitive processes [6]. For the purposes of Karabournaki, we have adopted a
relational data model to represent the data. In this model, information unit-entity is the
finding, object. Side information is provided by bibliographic entities, data from
conservation and archaeometry, as well as other complementary data entities. An
important aspect of our implementation is that it is based on open-source MySQL
database that significantly cuts the costs of the integrated system. MySQL is a
database engine that uses the standard relational model to represent data. Access to
the content is achieved by using SQL queries. Data management as well as data
retrieval is web based, portable and straightforward. Data input in Karabournaki
database is done in a twofold way: either explicitly, by accessing the central database
through an account with administration privileges, or implicitly, by using specifically
designed software data input forms that provide with the flexibility to collect data
even during on-site works (using a laptop). Data input forms include versions to
handle data produced from archaeological, archaeometrical and conservational
research and works.

Digital data visualization

Since the eighties of past century Computer Aided Design (CAD) systems have
been systematically used in archaeology [7]-[10] not only as a sophisticated means of
digitization, but also as a tool for reconstructing archaeological context. CAD and 3D
modeling allows archaeologists for the first time to record, integrate, organize and
handle the visual parts of archaeological information (plans, drawings, maps etc).
CAD has a number of advantages over traditional architectural and topographic
design methods like the unique ability to work on different layers in the same
drawing, but its true power lies in its ability to integrate and display multivariate data
on a single canvas. The integration of data in a 2D or 3D model enables the researcher
to gain a stronger understanding of the archaeological context and to seek spatial
relationships between the different elements. Moreover, 2D and 3D mapping have
evolved to a necessary preparatory stage in order to advance to the application of
Geographical Information Systems (GIS) technology or the use of Virtual reality [11]
as a means to approach ancient landscapes. In other words GIS promises the
archaeological community the ability to record, analyze, seek, retrieve and represent
vast amounts of complex spatial data in a homogeneous manner. The adoption [12] of
such a powerful spatial tool in intra site analysis can revolutionize the manner we
reconstruct archaeological context.

(a) (b)

(c) (d)
Fig. 2. (a-b) 3D virtual reconstructions, (c) GIS and (d) 3D-GIS of objects

In “Karabournaki-Recording the Past”, accepting the notion in these and many

other works, a 3D visualization approach was employed; an approach that is twofold:
• visualization of the site itself Fig. 2, using standard CAD design and GIS
techniques, fed by digitization data either from 2D scanning or 3D scanning
methods, and
• visualization of the objects Fig. 2, using an extension of a standard GIS system,
that has been specifically implemented. This system is targeted upon the
construction of a virtual “3D-GIS” environment for cultural objects, where the
object itself is treated as a geographical entity. In its present form, it is a
browser client plug-in, that gives the user all the known 3D viewing
capabilities, while extending the interactivity by providing with the option of
selecting surface areas and reading archaeometry or complementary data stored
in the database that concern the selected area. This way, the system is promoted
to a GIS-like environment for objects.
Digital data dissemination
In order to integrate the overall database management system into a web-enabled
information dissemination system, one has to design an Internet front-end with
extended search capabilities to be able to exploit the reach information content to the
maximum. The scheme adopted for the purposes of Karabournaki was based on
MySQL database using PHP as the main programming language for constructing the
dynamic web content and accessing the database. The whole system is dynamic,
displaying data straight from the database, using only minimal static information
needed mainly for customizations. It is based upon well known and tested
technologies that are successfully combined into a single and meaningful integrated
interface, which is able to provide with extended search and visualization capabilities.
The structure of the final web application as well as all involved technologies is
shown in Fig. 3.

(a) (b)
Fig. 3. Overview of (a) the Internet application scenario schematic and (b) the
underlying system technologies

CONCLUSIONS
Complete and integrated digital recording and dissemination of an archaeological
site is a complex and multivariate problem. To solve this problem, several existing
technologies have to be combined and some new have to be developed. In
“Karabournaki – Recording the Past” a first attempt is being made to overcome the
difficulties of such a task and to provide with an integrated system with extended
management and dissemination capabilities over the Internet. In order to ensure
universal accessibility the system was designed as bilingual. So far, most of the
system is implemented, while the whole strategy is already developed. Finally, its
flexibility and capability of being expandable in data structure secures the integration
of relevant future archaeological demands and needs.

References
[1] D. Tsiafakis, A. Tsompanopoulos, G. Pavlidis, N. Tsirliganis, V. Evangelidis,
C. Chamzas, “Archiving Cultural Objects in the 21st century: Pottery from
Karabournaki” in the 16th International Congress of Classical Archaeology,
Harvard University Art Museums, August 23-26, 2003, in press.
[2] M. Tiverios, E. Manakidou, D. Tsiafakis, “Panepistimiakes anaskafes sto
Karabournaki Thessalonikis (2000-2002)”, Egnatia 7, in press.
[3] M. Tiverios, E. Manakidou, D. Tsiafakis, “Archaeological research at
Karabournaki in 2002. The ancient settlement”, in To Archaeologiko Ergo sti
Macedonia kai ti Thraki 16, 2002 (Thessaloniki 2004) 257-266.
[4] N. Tsirliganis, G. Pavlidis, A. Koutsoudis, E. Politou, A. Tsompanopoulos, K.
Stavroglou, C. Chamzas, “New Ways in Digitization and Visualization of
Cultural Objects”, in Proc. IEEE DSP 2002, Santorini, Hellas, July 1-3, 2002.
[5] N. Tsirliganis, G. Pavlidis, A. Koutsoudis, D. Papadopoulou, A.
Tsompanopoulos, K. Stavroglou, Z. Loukou, C. Chamzas, “Archiving 3D
Cultural Objects with Surface Point-Wise Database Information”, in Proc. 3D
Data Processing, Visualization & Transmission 3DPVT 2002, Padova, Italy,
Jun. 18-21, 2002.
[6] E. Politou, I. Tsevremes, A. Tsompanopoulos, G. Pavlidis, A. Kazakis, C.
Chamzas, “Ark of Refugee Heirloom - A Cultural Heritage Database”, in
Proc. Electronic Imaging & Visual Arts EVA 2002 Conference, Florence,
Italy, March 25-29, 2002.
[7] Daniels R., “The need for the solid modelling of structure in the archaeology
of buildings”, in Internet Archaeology 2 (http://intarch.ac.uk/journal/issue 2).
[8] Bateman J., “Immediate Realities: an anthropology of computer visualisation
in archaeology”, in Internet Archaeology 8 (http://intarch.ac.uk/journal/issue
8).
[9] CSA (Center of Study of Architecture), “CAD Guide for Archaeology and
Architectural History”, http://csanet.org/inftech/cadgd/cadgdtoc.html.
[10] Reily P., “Towards a Virtual Archaeology” in Lockyear K.-Rathz S.(eds),
Computer Applications and Quantitative Methods in Archaeology (BAR 565),
pp. 133-140, 1990.
[11] M. Gillings - G. Goodrick, “Constructs, simulations and hyperreal worlds: the
role of the Virtual Reality in Archaeological Research”, On the Theory and
Practice of Archaeological Computing, G.Lock-K.Brown (eds), Oxford
University Committee for Archaeology, 2000.
[12] Wheatley D., “Spatial Technology and Archaeological Theory revisited”, in
Lockyear K.-Sly T.J. (eds), Computer Applications and Quantitative Methods
for Archaeology (BAR 845), pp. 123-130, 1996.