CI P A 2003 XIX 1 ' International Symposium, 30 September - 04 October, 2003, Antalya, Turkey
342
Another interesting fact is that the scale of the photography and
the final plot reaches 1:10 ratio, especially for orthophotograph
production.
Aspro-
Dilos
Larisa
Larisa
valta
theatre
scene
(sections
and
facades)
Final orthophoto
scale
1:100
1:100
1:50
1:25
Area fsq. metersl
1400
1700
5400
1250
Rollei
Rollei
Rollei
Rollei
Metric
Camera & Lens
Metric
60006
Metric
60006
Metric
60006
60006
50mm
50mm
50mm
50mm
and
80mm
Control points
10
30
54
321
Photographs taken
48
72
144
131
Photography and
control
measurements
[hours/people]
4/3
8/3
9/4
20/3
Average photo
scale
1:520
1:480
1:450
1:180
Photographs used
8
30
79
86
Scanning
resolution [dpi]
1800
1800
1800
1200
Scanning
[man days]
1
2
6.5
5
Triangulation and
model preparation
[man days]
3
4
9
22
AT: RMS control
2.2,2.7,
1.8,2.2,
1.2,1.3,
Variable
X,Y,Z [cm]
5.6
3.9
2.4
<2.0
DTM collection
7
7
15
[man days]
A
Orthophotos
[man days]
l
2
10
8
Mosaics
[man days]
1.5
5
17
13
3d plotting
[man days]
-
3
-
-
Preparation and
printing of hard
copies, retouch of
1.5
3
18.5
20
images
[man days]
Total time
[man days]
10
26
76
68
Number of plots
TAO]
1
1
9
4
Pixel size [m]
0.01
0.01
0.05
0.03
Table 1. Comparison of projects. Man days is the workload for
each task. Tasks such as DTM collection, plotting
etc, can be done simultaneously or in shifts, hence
reducing deliveiy time. AT: Aerial Triangulation.
What’s common in all cases is the fact that the accuracy implied
from the scale of the final product is not meet. It is very
expensive, and unviable commercially to try to reach accuracies
of 1.25 cm in 1:50 orthophotographs. Archaeologists and
architects are not familiar with the relationship between scale
and accuracy and therefore consider these terms independent
from each other. Therefore they happily accept a really big
discount for a product printed at scale 1:50 with accuracy
standards of 1:100.
If we had decided to follow the photogrametric practise at scale
1:25, saying that control points should have three times better
accuracy than the expected from the final product, meaning that
control points should have been measured with ±0.2 cm, the
cost of 321 control points would have raised more than the total
budget and would have been a great challenge for any
researcher.
We shall not forget side products such as the complete DTM of
the area. Contours might not be the best possible way to
represent the 3 rd dimension, especially when man-made features
are present. What the end users considered interesting when
unofficially presented to them, was a red-blue orthorectified
representation of their sites. This and a photo realistic
representation (Ioannidis et. al., 2000) are the best applications
of a good DTM.
From Larisas project became apparent that orthophotograph
production might not be faster, nor more economic than line
plotting but it is much better as a scene representation to the
final user.
Time needed for film processing and scanning is considerable
when in comparison with the whole project. A digital camera
with high resolution might be the answer to this problem.
Resolutions are now starting to compete with film and time for
downloading the image from the CCD to memory pose two
serious holdbacks. Until they reach the same resolution level, in
order to keep the same number of processing photographs and
models, usage of such camera is not considered.
Another advantage of the model helicopter is the speed of the
photographic procedure. The helicopter can shoot a twelve-
shoot film in less than 10 minutes. The most competitive
platform, the balloon, is much slower. In addition requires three
people handling it and they have to walk simultaneously over
the site in order to position it. On the other hand height can be
better adjusted and layout is similar or better. With a 6 beaufort
wind speed though, the balloon is unlikely to be able to
complete the task and therefore we still consider the model
helicopter as the better platform for fast acquisition of
photographs in such cases.
References:
Baratin, L., Bitelli, G., Unguendoli, M., Zanutta, A., 2000.
Digital Orthophoto as a tool for the restoration of monuments.
IAPRS, Vol. XXXIII, Part B5, Amsterdam 2000. pp. 62-69.
Dorffner, L., Kraus, K., Tschannerl, J., Altan, O., Kulur, S.,
Toz, G., 2000. Hagia Sophia - Photogrammetric record of a
world cultural heritage. IAPRS, Vol. XXXIII, Part B5, WG V/5,
Amsterdam 2000. pp. 172-178.
Ioannidis, C., Potsiou, C., Soile, S., Badekas, J., 2000. Detailed
3d representations of archaeological sites. IAPRS, Vol. XXXIII,
Part B5, WG V/5, Amsterdam 2000. pp. 642-649.
Karras, G., E., Mavrommati, D., Madani, M., Mavrelis, G.,
Lymperopoulos, E., Kambourakis, A., Gesafidis, S., 1999.
Digital orthophotography in archaeology with low-altitude non
metric images. ISPRS Vol XXII Part 5W11, Workshop
“Photogrammetric measurement, object modeling and
documentation in architecture and industry”, Thessaloniki,
Greece, 7-9 July, 1999, pp 8-11.
