© 
e 
29) 
305 
(model) consists of. 30 points with six 
control points has been done. The interior 
and exterior orientation parameters for 
both left and right photos were assumed. 
the geometrical dimensions were the base 
length (P). = 5.00 m, and the object 
distance (H) - 10.00 m. Random errors were 
added to each photo coordinate (0 to + 
0.05 mm) to simulate the measuring errors 
and random imzge deformations (e.g. lens 
distortion, film deformation, ...etc.). 
Table (1) shows the standard deviation for 
object (check points} space coordinates 
(in cm). 
  
  
  
  
  
  
  
Stand.| Metric Non-metric 
dev. case Case 
um 0.1535 1.1270 
c 8.1115 3.1103 
em 0.3983 0.4976 
cr 0.4440 1.5252 
Xyz 
Table (1) 
A practical test was carried out by 
Ebrshim (1992) using a phototheodolite 
19/1318 metric camera. The control field 
consisted of 50 check points, among which 
8 control points were chosen. The object 
distance was taken as H- 10.00 m. image 
coordinates were taken for 15 stereopsirs 
of photographs using Zeiss Jena 
stereocomparator 1818 in the 
photogrammetric Lab. in Assiut University. 
The standard deviation for the object 
space coordinates were determined using 
the proposed method. The results were 
compared with those calculated by the 
collinearity method as a standard method. 
Table (2) shows an example of the results 
of the standard devistion for check and 
control points (in cm) using both the 
  
  
  
developed method and the collinearity 
method. 
Stand. Check points Controi points 
dev. Rew | Collin.] — New] Collin. 
e 8.057 0.059 0.00017] 0.047 
e 0.061 0.071 0.00018[ 0.050 
e? 10.186 | 0.193 ]0.00060| 0.232 
eun 0.185 0.214 0.00085] 0.242 
  
  
  
  
  
  
  
Tabie (2) 
401 
CONCLUSION 
The developed method described here has 
proved an increase in the accuracy of the 
orientation parameters for both metric and 
non-metric cameras, which affected 
directiy the accuracy of the 
calculated space coordinates for both 
check and control points. This increase in 
accuracy may be referred to the reduction 
in the number of the unknowns in the 
equations of the developed method when 
compared to the collinearity method, (5 
unknowns instead of 9 in the case of 
non-metric cameras, and 5 unknowns instead 
of 6 in the case of metric cameras). This 
means that the number of iterations and 
consequently the computing time is 
reduced. 
BIBLIOGRAPHY 
1i- Abdel-Aziz and Karara, 1874. accuracy 
aspects of non-metric imageries. 
Photogrammetric Engineering. p.p. 
1107-1117 
2- Ebrahim, M.A., 1992. Using close-range 
photogrammetry for some engineering 
applications 
3- Hottier, Ph., 1978. Accuracy of 
close-range Analytical restitutions 
Practical experiments and prediction. 
Photogrammetric engineering and remote 
sensing. Vol.42, No.3. p.p. 345-375 
4- Torlegárd, A.K., 1981. Development of 
D EE photogrammetry and its 
future. 5 anniversary meeting of the 
Finnish society of photogrammetry. 
p.p. 49-72 
5- Verdina, J., 1971. Projective geometry 
and point transformation. Allyn and 
Bacon, Inc. 
Wolf ,P.R.,1974 .Elementsof photogrammetry 
Mograw-Hill, Inc. 
© 
|