CI PA 2003 XIX th International Symposium, 30 September - 04 October, 2003, Antalya, Turkey 
510 
presents the Java3D visualization tool using the flat shading 
approach, as outlined above). 
At the current stage, the tool allows Internet based visualization 
with standard means. However, further investigations are 
necessary to provide a superior presentation, enabling a better 
understanding of the artifact by the user and limiting the data to 
be transferred over the network at the same time. Edge 
visualization and contours might be a good approach here, too. 
For a thorough inspection, the complete set of data is necessary, 
however. Thus, the Internet may serve to inspect the collection 
and select those pieces for which a complete set of data will be 
ordered. 
5. DATABASE DESIGN AND GIS CONCEPT 
In addition to the scanning data describing the geometry of the 
artifacts, further information concerning the artifacts are stored 
in a database. Due to the similarity of the findings, the database 
design can be kept rather simple. The main table of the 
relational database holds one record for each artifact. 
The items can be classified into different categories: 
• Items of identification like the identification number. 
• Person keeping the artifact, place where it is kept, and 
inventory number. 
• Basic description (type of artifact, material, weight). 
• Link to the scanning data and derived graphical products 
like sections, views, etc. 
• Geometric quantities derived from the scanning data, like 
volume, surface, unit weight, etc. 
• Information about the place where it was found (easting, 
northing, height, accuracy of position and height). 
• Information on the actual condition (recent damages, traces 
of fire, etc.). 
• Further textual information. 
Additional tables may contain, for example, classes of types of 
artifacts, damages, etc., to which codes in the main table are 
pointing. The main table keeps all major information describing 
the single artifact and links to further information like the 3D 
model and views. It can be used for all kinds of queries using 
standard SQL format. 
The availability of information concerning the place of finding 
the artifacts allows using basic GIS functionality. The quality of 
the recorded coordinates for these places may differ consider 
ably. Some findings can be referenced just to a certain parcel 
where others may have an absolute accuracy of better than 10 
meters since GPS was used when the artifacts were collected. 
Therefore it is mandatory to store and use information about the 
accuracy of these observations and take it into account when 
analyzing or comparing spatial information. The spatial infor 
mation can be used in different ways, a simple one is the 
plotting of distribution maps of the findings, resulting from a 
database query, on the fly. Combinations with current topo 
graphic maps can assist in orientation. Using further external 
information like e.g. DEMs, geological or hydrological maps in 
combination with the distribution of the artifacts and the 
functionality of a GIS allow further conclusions about the 
history of the findings and their usage by man. 
6. CONCLUSIONS 
Considerable numbers of artifacts can be documented with high 
accuracy and resolution when the methods described are used. 
Virtual artifact collections can be documented completely, even 
if the artifacts are stored at different places, including objects 
that were considered as being unretouched pieces. The virtual 
collection thus achieved can be distributed easily on storage 
media such as CD ROMs or even through the Internet. It can be 
examined by anybody and compared to any real or virtual 
artifacts of similar origin. 
All visualization products are results of automatic and objective 
procedures, thus avoiding the individual subjective interpreta 
tion which is inevitably part of hand drawn figures. 
7. ACKNOWLEDGEMENTS 
We are very grateful to Prof. Dr. Johannes Preuss from the 
Department of Geography of Johannes Gutenberg University 
Mainz who initiated the project and Dr. Konrad Weidemann, 
Director General of Roemisch-Germanisches Zentralmuseum, 
Mainz, who gave us much support. The Government of 
Rhineland-Palatinate supplied the funding for the acquisition of 
the scanner. FH Mainz, University of Applied Sciences, 
supported the project with funds as well. Stefan Bartsch 
(Bartsch, 2003) and Stefan Tschoepe (Tschoepe, 2003) 
contributed considerably with the visualization solutions 
developed in their diploma theses. 
8. REFERENCES 
Bartsch, Stefan, 2003: Intemet-basierte Visualisierung von 
Laserscanningdaten. Diploma thesis at FH Mainz, University of 
Applied Sciences. Unpublished. 
Foley J.D., van Dam A., Feiner S.K., Hughes J.f., 1990: 
"Computer Graphics: Principles and Practice". 2 nd Edition, 
Reading, MA: Addison-Wesley, 1990 
Geomagic, 2003: http://www.geomagic.com 
GOM, 2003: http://www.gom.com 
i3mainz, 2003: http://scanning.fh-mainz.de 
Qslim, 2003: 
http://graphics.cs.uiuc.edu/~garland/software/qslim.html 
SolidView, 2003: http://www.solidview.com/ 
Tschoepe, Stefan, 2003: Entwicklung von Software zur 
Weiterverarbeitung vermaschter 3D Punktwolken. Diploma 
thesis at FH Mainz, University of Applied Sciences. 
Unpublished.