re 
al 
d 
8, 
Georgopoulos, Andreas 
  
THE POTENTIAL OF APPLYING 3-D VISUALISATION METHODS 
FOR REPRESENTING DESTROYED SETTLEMENTS 
B. Nakos and A. Georgopoulos 
Faculty of Rural and Surveying Engineering 
National Technical University of Athens, Greece 
9, Heroon Polytechniou Str., Zographos, GR-157 80 
bnakos@central.ntua.gr - drag@central.ntua.gr 
Working Group V/3 
KEY WORDS: CAD, Modelling, Reconstruction, Rendering, Visualisation 
ABSTRACT 
Modern technological advances have enriched the means for electronic representation of 3-D data. Proper data 
acquisition enables the production of visualisations of the built environment as close to reality as possible. In this paper 
such an effort is described. The theoretical approach is followed by the description of a specific application concerning 
the representation of an old traditional settlement, which was destroyed by an earthquake. The data collection and the 
processing of the 3-D data are described in detail. Finally sample renderings are presented and compared to old amateur 
photographs available. The results are discussed and assessed for their representation value and usefulness. 
KURZFASSUNG 
Die modernen technologischen Fortschritte haben zweifellos die Mitteln fur die elektronische Representierung von 3 -D 
Daten bereichert. Geeignete Datenerfassung erleichtet die Produktion von Visualisierungen der gebauten Umwelt so 
nah der Realitat wie moglich. In diesem Bericht wird ein solcher versuch beschreibt. Nach der theoretischen Einfuhrung 
folgt die Beschreibung der spezifischen Anwendung, die sich mit der Representierung von einer traditionellen Siedlung, 
die von einem Erdbeben zerstort wurde, beschaftigt. Die Datenerfassung und die Verarbeitung der 3-D Daten werden 
ausfuhrlich beschrieben. Die Ergebnisse werden ausgewertet fur ihre Representationswert und Benutzbarkeit. 
1 INTRODUCTION 
Modern technological advances have enabled researchers to provide three-dimensional representations of space in a 
number of different ways. Usually, the three-dimensional representations vary in scale, in data sources and in 
representation media. By examining the scale, it diverges from very large to very small values. In very large scale 
representations an object (i.e. a building or a part of a building) is presented with all of its details —see for example 
relevant application in Guerra and Balletti (1999). In large scale representations groups of several objects are presented 
with high degree of detail -see for example the application of Georgopoulos et al. (1997). Furthermore, in small scale 
representations a part or even all the earth's surface is presented, and additional media can enrich the representation by 
adding a virtual character, consisting for example a virtual globe (Riedl, 1999). 
Data acquisition to this end may be realised either through accurate mensuration processes or by using other kinds of 
available pertinent information. Depending on the representation scale the three-dimensional data may be extracted 
from a wide range of sources consisting of field surveys, vertical or oblique metric or non-metric photographs and 
existing images, plans or maps, or even from an existing GIS (Schilcher et al., 1999). The nature of photogrammetric 
techniques, especially for large or very large representations, allows for a rather complete data acquisition, providing 
both qualitative and quantitative spatial information. A comprehensive survey of the use of photogrammetry in large 
scale representation can be found in Lammi (1999). Hence, it is considered as a most suitable way for providing the 
necessary raw data for the three-dimensional visualisation of space in large scale. 
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B5. Amsterdam 2000. 285