Sq is obtained in microns for the radius r' in mm in the photographs. The 
root mean square value of the standard error of unit weight is about 7 microns. 
After sufficient similar tests under operational conditions, tolerances for the 
standard errors of unit weight of image coordinates can be established ac 
cording to the principles discussed above. 
The Relative Orientation. 
The standard error of unit weight of the y-parallaxes after an adjustment 
according to the method of least squares can be used as an indication of how 
well the relative orientation can be made. Because the accuracy of the relative 
orientation is of fundamental importance for the accuracy of the entire photo- 
grammetric procedure including the determination of the elements of the 
exterior orientation, tolerances for the relative orientation should be esta 
blished. Comprehensive tests have been made in order to determine the basic 
value of the standard error of unit weight for a specified combination of 
camera and instrument, see for instance Hallert-Ottoson-Ternryd 1960 and 
Hallert-Ottoson-Ohlin 1964. 
The following values have been found on the image scale. For normal angle 
photographs, c = 210 mm and first order stereoscopic projection instruments: 
s 0 = 4 microns. For wide angle photographs, c — 153 mm and first order 
stereoscopic projection instruments: s 0 = 6 microns. For superwide angle 
photographs, c = 88 mm and autographs Wild A9: s 0 = 10 microns. 
If the >’-parallax measurements are assumed to be made in 9 model points 
there are 4 redundant observations and the tolerances of the standard errors 
of unit weight become on the 5 percent level 1.54s 0 . For wide angle photo 
graphs and first order stereoscopic projection instruments this means that the 
standard error of unit weight should not exceed 9.2 microns. 
In analytical photogrammetry (wide angle photographs c = 153 mm and 
stereocomparator Wild StK 824) s 0 has been found to be about 4.5 microns 
and the tolerances after adjustment of measurements in 9 points 7 microns. 
If the measurements are made in 15 points the tolerances become 1.4s 0 , i. e. 
6.3 microns. 
If the relative orientation is not adjusted according to the method of least 
squares, the root mean square value of the residual y-parallaxes in the model 
can approximately be used as an indication of the accuracy of the orientation. 
For wide angle photographs c = 153 mm and first order stereoscopic projec 
tion instruments, comprehensive tests have indicated this root mean square 
value to be 12 microns, see Hallert-Ottoson-Ternryd 1960 and Hallert-Ottoson- 
Ohlin 1964. Assuming the residual y-parallaxes to be approximately normally 
distributed (see Hallert-Ottoson-Ternryd 1960), the tolerances of the root mean 
square values of the residual y-parallaxes become 18.4 and 17 microns after 
measurements in 9 and 15 model points respectively.