RMS of Adjusted Object 
Coordinates (method 2) 
  
RMS (mm) 
  
  
  
0.25 0.5 1 
GPS Accuracy (m) 
  
Figure 2. RMS of X, Y, and Z Coordinates of All Object Points 
(method 2) 
Comparing Figures 1 and 2 show that if GPS can provide 
accuracy at the level of 0.25 to 0.5 m for camera exposure 
stations coordinates, the RMS values for all object coordinates 
are better or equal to those obtained from the full ground 
control version. These results confirm that constraint 
information from the tower points can replace the ground 
control points and eliminates the need for the second strip of 
photography which has been adopted conventionally to improve 
the geometry of the strip. To see how this technique recovers 
the roll angle of the aircraft, Figures 3, 4, and 5 show the 
adjusted roll angle of each photo obtained from the 3 methods 
(Goes =0.25 m) 
As seen in these Figures, the adjusted roll angle recovered from 
method 2 (tower points included) is almost the same as what 
has been obtained from method 3 (full ground control version). 
  
N 
© 9 
Roll Angle 
(seconds of arc) 
  
E. X 
5 Oö 0 
Photo Number 
  
  
  
Figure 3. Adjusted Roll Angle from Method 1 
  
d — ND 
eoo © 
Roll Angle 
(seconds of arc) 
N 
© 
Photo Number 
  
  
  
Figure 4. Adjusted Roll Angle from Method 2 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B3. Vienna 1996 
  
Roll Angle 
(seconds of arc) 
  
Photo Number 
  
  
  
Figure 5. Adjusted Roll Angle from Method 3 
In the reliability analysis, redundancy numbers, internal 
reliability factors, and external reliability factors were computed 
for various observations (image coordinates, GPS coordinates 
of exposure stations). Table 2 and 3 show the reliability 
measures for image coordinates and GPS coordinates of 
exposure stations for the various methods. 
Table 2. Reliability Measures for Image Coordinates 
  
  
  
  
  
  
  
  
  
Method 1 2 3 
Reliability 
Measure X y X y X y 
Ti 0.28 0.11 | 0.34 0.54 | 0.23 0.39 
O0 8.34 12.52 | 8.41 5.73 | 9.96 6.61 
$0; 72/1185 7.11 4393 1 3.95 5 21 
  
Table 3. Reliability Measures for GPS Coordinates of 
Exposure Stations 
  
  
  
  
Method 1 2 
Reliability 
Measure X Y Z X XY 2 
ri 0.74 0.80 0.84 | 0.74 0.85 0.82 
00i 4.65 447 436 | 4.66 434 442 
$i 236 198 172 1237 1.67 1,85 
  
  
  
  
  
These values are rated as good (0.5 « r;, 89; « 4.0), acceptable 
(0.15 1, « 0.5, 4.0< So. <10.0), bad (0.04< ri <0.1, 
10.0 € 5,; < 20.0), and not acceptable (r; < 0.04, 
20.0< $0, (Forstner,1985). The best values for the reliability 
measures have been obtained from method 2 which imply that 
including tower points in the GPS controlled strip triangulation 
improves the reliability of both image coordinates especially the 
y coordinates of image points and GPS observations. 
4. 3. Results With Real Data 
Encouraging results from the simulated data have convinced us 
to apply this new technique on real data. The results from the 
real data was not available at the time of writing this paper. 
These results will be presented during the conference. 
    
       
  
   
   
    
   
   
   
    
   
   
   
    
   
    
    
      
   
   
   
   
   
     
    
     
   
   
   
   
    
GPS contro 
constraints 
the ground « 
camera. The 
if kinemati 
camera exp 
without an 
transformat 
photograph 
and to inc 
technique 1 
multiple stt 
the cost of t 
Ackermann 
Photogrami 
pp. 261-271 
Alobaida, 
Mapping S 
Dissertatio! 
The Ohio S 
Cannon, ^ 
Positioning 
Colomina, 
Triangulati 
Engineerin 
1624. 
Deren, L., 
Adjustmen 
Engineerin 
1746. 
Ebadi, H., 
Assisted 
Internation 
Forstner, | 
Photogran 
pp. 1137-1 
FrieB, F 
Empirical 
Workshop 
Bonn, Ger 
Lachapell 
Precision 
Ambiguit: 
Lapine, I