———— 
EIE rec AMEN 
  
| 
| 
| 
ja 
| ^ 
E 
|. 
| 
E 
ix 
|. 
TE 
2.4 Accuracy of the I.I. S, New Stereo-Camera 
To examine the accuracy of the I. I.S. New Stereo-Camera, the following 
experiments were conducted: 
1) Limitation of measur ement accuracy of plane 
2) Measurement of distortion, curvature of the field and astigmatism of lens 
3) Measurement of resloving power of photograph. 
2.4.1 Accuracy of Measurement 
In the case of photogrammetry, the accuracy of measurements of X and Y.co- 
ordinates is generally higher than that of Z values. From the accuracy of X and q ? 
Y values, accuracy of measurements perpendicular to this plane can be calculated 
by the error theory of the ellipsoid. 
Fig. 4 shows a grid test board (2.1m 
X 1.8 m) with a grid drawn with 15 cm 
interval very carefully. The grid test 
board is set parallel to the vertical 
plane containning the base line connect- 
ing both nodal points of the lens and at 
a 1.8 m distance from the camera. 
This is the same situation as in the 
case of the actual short range photo - 
graphy applied to our application works. 
  
  
cut 
umm uuu AM EIN 
PETI 
A 
À SITE 
| Bm 
| eA TEENS 
Yanaagalaling 
  
    
Fig. 4 Grid Test Board (1.5 x 1.8 m) % 
(Grid Interval 15 cm) 
The grid points are measured four times 
each with an A7 Autograph and their co-ordinates are typed by an EK3 automatic 
co-ordinates printer and are compared with the direct measurement values. 
Fig.5 shows the four orientation points and the 
Fig. 5 Grid Test for Orientation 
Error (in mm) 
Xms and Yıns of each points. These values are MD @ a, 
A + 670 Orientation Error. 
004m» ^ 048 Xms = 0321s 
the mean square error of deviations between the Yıns = 0530s 
4 (5) 
values measured with a A7 Autograph and the values . 
42 042 
measured directly by scaling. ah s 
P7022 
pu oh + 
y 
031 0