calibration). The worst ones are of the order of 3cm
(digitizer + self calibration).
Data | Adjust. D, D, D ax M,
set Type (cm) (cm) (cm) im (pixel)
I a 1.48 1.38 2.06 | 6.43 (0.76)
I b 0.92 0.79 1.58 | 4.00 (0.47)
II a 4.47 | 4.13 6.18 | 19.43(0.78)
II b 2.96 2.40 | 4.65 | 12.87(0.51)
D, RMS value corresponding to the mean variance
of the check points coordinates(object space)
D, Mean Standard Deviation of the check points
coordinates (object space)
D Maximum Standard Deviation of the check points
coordinates (object space)
M, RMS value corresponding to the mean variance,
calculated with respect to D, (image space)
KAIMAKA AIXTYOY
C»
C»
e Ae (>
Cm eG uo Go ©
(m) e e
Figure 6. Projections of error ellipsoids for solutions Iib.
7. CONCLUSIONS
Many images of buildings of cultural interest are taken by
architects, historians or others, who are interested in
Architectural Photogrammetry or just in souvenir
photography. There is a great potential for such imagery
taken by accident, and it would be advantageous to use
this resource in addition to professional photogrammetric
imagery for the restitution of the world's heritage. Many
such images are acquired today by low-cost consumer
products, like camcorders.
Although camcorders are not intented for photogram-
metric use, they excibit many useful characteristics (low-
cost, universal use, portability, no need of special equip-
ment, inexpensive means for storage of huge amount of
data, etc) while previous studies have shown that with
proper system calibration sufficient accuracy for archite-
ctural tasks can be otained under practical situations.
The aim of this work is to contribute to the CIPA test "O.
Wagner Pavillion" with the purpose to describe the state-
of-the-art in Architectural Photogrammetry. This contri-
bution consists of demonstrating the user of camcorders
in architectural recording. It is shown that even such low-
cost, low-resolution, and low-geometrical-stability
systems can give results compatible to those obtained
under conventional photogrammetric techniques. The
accuracies obtained is 1 cm with minimal ground control.
Acknowledgements
This work has been performed during the stay of the first
author at ETH, Zurich, as a visiting professor. The access
to the images, DIPAD system and computer facilities is
gratefully aknowledged.
REFERENCES
Beyer, H., T. Kersten, A. Streilein. 1992. Metric accuracy
performance of solid-state camera systems, Proc. of SPIE
OE/Technology'92 - Videometrics conference, Nov, 15-
20, 1992, Boston, MA., SPIE Vol. 1820, pp. 103-110.
Gruen, A., 1985. Adaptive Least Squares Correlation: A
powerful image matching technique, South African
Joutnal of Photogrammetry, Remote Sensing and
Cartography, Vol. 14, No. 3, pp. 175-187.
Patias, P., D. Rossikopoulos, O. Georgoula. 1993. CIPA
Test: "O. Wagner Pavillion". Preliminary Report.
Presented at the CIPA XV International Symposium,
Bucharest, Romania, September 22-26, (Executive
Summary under publication in the proceedings).
Streilein, A. and S. Gaschen. 1994. Comparison of a S-
VHS Camcorder and a High-Resolution CCD-Camera for
Architectural Photogrammetry. Proc. of the ISPRS Inter-
congress Symposium, Melbourne, Australia, pp. 382-389.
Streilein, A., 1993. Application of a S-VHS Camcorder in
Architectural Photogrammetry. Proc. of the CIPA XV
International Symposium, Bucharest, Romania.
Streilein, A., 1995. Towards automation in Architectural
Photogrammetry: CAD-based 3D-feature extraction,
ISPRS Journal of Photogrammetry and Remote Sensing,
49(5), pp. 4-15.
Waldhausl, P. 1992. Defining the future of Architectural
Photogrammetry, International Archives of Photogram-
metry and Remote Sensing, Vol. 29, Bart B5, pp. 767-
770.
462
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B5. Vienna 1996
KEY V
ABSTE
World-
architec
a great
professi
The ain
Second,
The test
building
differen
images)
Europes
In all ca
concern
case, mi
differen
himself
and UN)
ZUSAN
Weltwei
wurden
photogrz
Zufallsb
metrie c
Welterb:
Das Ziel
Gebiet
arbeitenc
Als Tes
gewaehlt
wurden
Systemer
paketen
Laenderr
In allen ]
Schlussf:
Kleinfor
reichlich
Wertever:
fuer sich
