CIP A 2003 XIX th International Symposium, 30 September - 04 October, 2003, Antalya, Turkey 
It is clear that imagery needed for large-scale mapping (1:500 or 
1:1000) should be supplied by aerial photography, where 
satellite imagery stands by as an alternative for 1:5000 mapping. 
Since the imagery should be in digital form as satellite imagery, 
aerial photos taken by analogue cameras would be digitized at 
photogrammetric scanners or could directly be acquired by 
digital metric cameras such as ADS40 of Leica Geosystems and 
DMC of Z/I Imaging. 
When deciding to use either aerial photos or satellite imagery, 
cost, time and accuracy issues dominate the selection. While 
aerial photo flight costs vary dramatically as country or 
company basis, satellite prices are rather stable. Since, both 
airborne and spacebome sensors are optical devices and require 
clear skies, image availability time mostly depends on 
meteorological conditions. From the point of the accuracy view, 
aerial photography is still one step ahead when speaking of 
large-scale mapping. 
Since both image types are both daylight and good weather 
dependent, distinctive fact for the choice is their cost and 
accuracy. Cost is limited with the project budget, while 
accuracy is determined by the expected exactness of the final 
product. Providing that the requested satellite images of the area 
are present in the archives of satellite image distributor, it is 
possible to get them at very low costs comparing to the 
programmed imagery upon request. This is an important note 
that should be kept in mind because the area is not experiencing 
any significant change since October 2000. 
4.3 Photogrammetric Work 
Photogrammetry has experienced an evolution parallel to the 
technological development, starting with analogue, continuing 
with analytic and reaching to digital technique. Taking into 
account that the main objective of the photogrammetric work 
mentioned herein is to supply sufficient geographical 
infrastructure for the proposed geo-based archeological 
information system of Zeugma site, all of the photogrammetric 
outputs are desired to be in digital form. For this reason, final 
products of the photogrammetric labor, such as orthoimages, 
topographic maps and Digital Elevation Model (DEM) of the 
area, should be in digital form. 
First step of the photogrammetric process will be digitizing the 
aerial photography if it is not obtained directly in digital form. 
During this process, 12 mm scanning resolution is considered to 
suffice. Subsequently, combined block adjustment will take 
place for aerial triangulation 
Third phase covers the compilation process during which the 
operator will collect every natural and men made details. 
Though most of the commercial photogrammetric software 
allow DEM generation by automatic image matching and 
determination of contour lines by using the DEM, it must be 
reminded that elevation model created will be not a DEM but 
digital surface model (DSM) including natural and artificial 
features on the terrain. This is due to the fact that if the 
automatically selected points are features such as a building or a 
tree, floating mark will be on top of the object whereby the 
operator will set it down to the ground. Therefore DSM created 
by either automatic image matching or LIDAR should be 
reduced to DEM using appropriate software. Final phase of the 
photogrammetric work is obtaining the orthoimages by using 
the DEM of the area. 
Another way of generating DEM is utilizing Light Detection 
and Radar (LIDAR) technology, which is also called as airborne 
laser scanning or mapping. Over the last years, LIDAR 
technology has become the accurate, timely and economical 
way to capture elevation data by means of DSM (Hill et al 
2000). LIDAR system basically consists of a laser scanner and a 
GPS/INS. Fundamental principle of the system is based on 
measuring the direction and distance between the feature and 
the sensor by the mirror angle and accurate laser beam travel 
time computation while GPS/INS system is determining the 
coordinates of the sensor at the moment of recording. The 
accuracy of the system mostly depends on the positional 
information acquired by GPS/INS. Various accuracy validation 
studies have shown that LIDAR technology has reached to a 
vertical accuracy of 10-20 cm and horizontal accuracy of 1 m, 
which can be enhanced by additional ground survey 
observations (Murakami et al 1999, Hill et al 2000). 
4.4 Proposed Work Summary 
In order to produce 1:500 or 1:1000 and 1:5000 scale digital 
maps of the region, including the area on which establishment 
of an open-air museum is planned, whole area has to be 
photographed at a scale of 1:4000 and 1:16000 either with 
analogue or digital airborne cameras. Film type could be either 
panchromatic or color, however considering the ease of 
interpretation and feature classification color photography is 
preferred. In order conduct a less ground survey, flight should 
be conducted with kinematic GPS support and GCP coordinates 
should be determined accurately according to TUTGA. For our 
case, if the flight is conducted by kinematic GPS technique, 
instead of 12 GCP, only four will be enough but two additional 
cross flight strips are required to avoid possible cycle slips. 
As it is depicted in Figure 3, for 1:4000 photography, in 
addition to five north-east flight strips, two cross strips have to 
be flown while only one flight line is will be enough for 
1:16000 flight. 
Figure 3. Sketch of the GCPs and flight strips. 
As an alternative panchromatic or natural color 0.6 m resolution 
Quickbird imagery could be used for 1:5000 map production. 
For aerial triangulation, combined bundle block adjustment 
should be applied with proper software and digital compilation 
process should be carried on.