Ali, Mahmoud El Nokrashy 
  
AX= a, (1° - x Ha (14 —1) x + ag (1° + 2x" ) + 24 Xy — A5 X + A6 Y 
(5) 
AY= a (-1)ysa (r'l)ysa,(r *2y ) cas xy asy 
To study the effect of additional parameters on the accuracy, a test using six stereo pairs with base of 2.05m and 
different object to camera distances were adjusted with and without using the mentioned additional parameters. Table 5 
shows the results of the six models. From the table it can be seen that the percentages of improvements in accuracy for 
the stereo pairs are about 33%, 24% and 42% in X, Y and Z directions respectively when using the additional 
parameters. 
3.7 Effect of Using Side Photography 
In establishing a GIS for a highway, the usual case is photographing in the direction of the van movements. Sometimes, 
targets of special importance can be located at one side of the road. Then it is possible to photograph the object points 
with the base of the two cameras parallel to the van motion. The advantages are the ability to increase the length of the 
base and also one may obtain the smaller object to van distances. Four stereo pairs were taken. The first was the usual 
case with base line of 2.05m and 35 m average object to camera distance. The other three pairs were using side 
photographs and same number of control and check points. The second pair had base line of 2.05m but with 25 m object 
to camera distance. The third had base line and object to camera distance of 2.50m and 25m respectively. The last pair 
has base line of 2.75m and same camera to object distance. Table 6 shows the average error for check points for the four 
models. From the table it is clear that the accuracy have been improved. The percentage of improvements for the fourth 
model compared with the first are 30%, 32% and 42% in X, Y and Z respectively. 
Table (5) Average Errors for Check Points With and Without Additional Parameters. 
  
  
  
  
  
  
  
  
  
  
  
  
  
Pair | Av. Objects With Without 
No. Distance m SX SY SZ ST SX SY SZ ST 
1 35 4.30 3.70 8.20 9.97 6.45 4.90 18.13 17.16 
2 50 3.02 3.51 7.76 9.38 6.00 4.58 14.32 16.19 
3 45 3.63 3.32. 7.31 8.81 5.55 4.34 13.49 15,22 
4 40 3.54 3.24 6.95 8.45 5.21 4.24 12.83 14.48 
5 35 3.33 3.13 6.63 8.05 5.04 4.14 12.24 13.87 
6 30 2.87 2.71 6.26 7.40 4.24 3.76 11.55 12.87 
  
  
  
  
  
  
  
Table (6) Average Errors for Check Points for Side Photography Case Study. 
  
  
  
  
  
  
  
  
  
Case of Study | Length of Base | Average Objects Average Error (cm) 
Line (m) Distance (m) SX SY SZ ST 
Normal 2.05 33 3.54 3.57 5.64 7.56 
Suggested 2.05 25 2.41 2.53 3.50 4.96 
Suggested 2.50 25 1.90 2.09 3.10 4.19 
Suggested 2.75 25 1.65 1.90 2.70 3.69 
  
  
  
  
  
  
  
3.8 Effect of Increasing the Number of Camera Stations. 
To estimate the effect of increasing the number of camera stations, tests have been carried out using two, three, four, 
five and six camera stations. The first pair was the usual case. For the second test, one camera station was added in the 
middle and its elevation is 0.50m above the other two cameras. For the third test one more cameras station was added 
with its elevation 0.5m higher than the first two.. The fourth test a fifth camera with its elevation 0.75m higher than the 
first two cameras was added in the middle. The fifth test a sixth camera station was added with its elevation similar to 
fifth camera station. Fig. 2 shows the arrangement of the cameras for the five tests. 
  
22 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B5. Amsterdam 2000. 
Table 
impro 
by 31 
4- CO: 
From tl 
1-* Tt 
phe 
2- Inc 
The 
can 
roa