112 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1976 
TABLE 2. MODEL ACCURACY FOR METRIC CAMERAS (PHOTO-THEO, SMK-40) AND NON-METRIC 
CAMERAS FROM Practical TESTS. 
  
  
  
  
Vertical 
Camera Arrangement Convergent Photographs Photos 
Calibration Selfcalibration Test Field 
Reference (6) (4) (9) (10) 
Camera Photo-Theo Linhof Hasselblad Contarex Hasselblad SMK 40 SMK 40 
Picture scale 1:22 1:25 1:44 1:45 1:36 1:43 1:45 
Principal distance 204 141 82 53 82 60 60 
Base to height ratio 1:45 115 1:13 Y:1 1:15 1:15 1:65 
Object size 2.7x39m 2.0x9.5m 2.4x29.4m Ll1xl.6m 2.0x 2.0m 2x 2m 4.2x4.2m 
Number of control points 30 27 22 23 120 64 16 
Residual parallaxes (mm) 0.08 0.10 0.14 0.13 
Co-ordinate errors (mm) 0.11 0.10 0.20 0.15 0.19 0.16 0.10 
Co-ordinate Errors Reduced to a Picture Scale of 1 : 20 
m, (mm) 0.10 0.08 0.09 0.07 0.10 0.08 0.49 
m, (mm) 0.08 0.06 0.08 0.06 0.06 0.05 0.39 
m, (mm) 0.06 0.05 0.06 0.06 0.06 0.07 0.39 
m. (mm) 0.15 0.12 0.12 0.11 0.16 0.10 0.65 
  
area of the control points until the coordinate 
errors exceed the influence of the unavoid- 
able measuring errors by a factor of two. 
These reflections point out that the question 
of metric or non-metric cameras is even 
i 
aS m; in [mm] / 
| 
oo | / 
0.15 | | / of 
  
  
  
  
/ 
ti 
BES 
  
SN 
— 
0.05 
  
  
-—r 
  
  
  
  
  
  
  
  
Z in [mm] 
o 
  
Ns 
500 700 900 1100 1300 1500 1700 1900 
Fıc. 8. Estimated planimetric errors in radial 
direction for photogrammetric point determina- 
tion. It is assumed that the camera is calibrated 
by a test field located within the marked area (Z 
= 900 to 1100 mm). As the depth of the test field is 
very narrow the precision of the parameters of 
the inner orientation is relatively low. Neverthe- 
less the influence of these errors is completely 
compensated within the area ofthe test field. The 
measuring precision decreases for points closer 
to or further from the camera (camera: Hassel- 
blad with Planar 1 : 2.8/80; the calibration was 
performed with 74 control points and included 
the radial distortion and the principal point of 
symmetry; 0, = * 4.4um, m, = + 53 um, my 
= + 37 um, maximum correlation coefficient 0.95, r 
r indicates the distance ofthe image point from the 
principal point, m, the effect of the measuring 
precision). 
inferior to the problem of using a favorable 
camera arrangement. 
CONCLUSIONS 
The aim of this paper was to show toler- 
ances for the parameters of inner orientation 
and to compare these figures with the poten- 
tials of various cameras. The differences in 
precision between metric and non-metric 
cameras are smaller than one would expect. 
To a great extent the model accuracy de- 
pends on factors other than the use of metric 
or non-metric cameras. Nevertheless, the 
application of non-metric cameras is coupled 
with a number of problems which are not 
expressed by such figures. The user of non- 
metric cameras has to calibrate the camera by 
himself and should investigate the repro- 
ducibility of the principal point and the 
principal distance. Sophisticated computer 
programs are needed for the calibration of 
the cameras and for the data reduction of the 
photographs, whereas for metric cameras the 
manufacturer delivers the calibration report. 
Photographs taken with metric cameras can 
be restituted on analog plotters due to the 
smaller lens distortion and the controlled 
relative orientation. Consequently it is more 
a question of comfort than a matter of preci- 
sion whether metric or non-metric cameras 
should be used. But this comfort requires a 
considerable capital investment. 
REFERENCES 
1. Karara, H. M. and Abdel-Aziz, Y. I., “Accuracy 
Aspects of Non-Metric Imageries", Photo- 
grammetric Engineering 40, 1107-1117, 1974 
2. Jordan-Eggert-Kneissel (Editor), Handbuch