Full text: Reprints of papers (Part 4a)

   
  
  
   
  
  
  
  
  
  
  
  
  
  
  
  
  
   
   
   
  
   
     
  
  
  
  
   
   
    
     
XV / 
y Ea xk 
T bd (— pa t p4 t ps — pe) + 2d (ps — Ps) t 
y? 
Tap — 201 — 202 + Pa * Pa + Ds + Ds) "p, «95 
where p, is the measured y-parallax in the point (x, y) before the 
adjustment. Compare [2]. 
4. The residual y-parallaxes p are transformed into radial distor- 
y 
; 
tion with the aid of eq. (1) or (4). The y-coordinates can be determined 
graphically or numerically or in connection with the measurements in 
the plotter. 
5. Averages of the four possible determinations of the distortion for 
each radius are computed and are plotted as a distortion curve, where 
the different radii are the abscissae. À zero-point of the distortion curve 
is given by the radius d. Transformation of the curve to another zero- 
point can be performed with wellknown methods. 
4. The accuracy of the procedure 
Obviously, the accuracy of the numerically determined residual y- 
parallaxes after the adjustment of the relative orientation is of funda- 
mental importance for the accuracy of the determined distortion 
amounts. 
In [3], the standard error of the y-parallax corrections from the ad- 
justment of the relative orientation in six points has been given in equa- 
tion (8), page 21. If we assume the standard error of the y-parallax 
measurements in all points to be u we find the standard error of the 
residual y-parallaxes as 
  
  
  
p E e d m 1  ' 
m, =, V + = + MES PUEROS TEE (10) 
x, 3 3b? ' b2d2 3b bd?  4d* 242 
There are rather small variations within the model. For wide angle 
pictures and 60 96 overlap we find, under normal conditions, a mean 
value of the standard errors of the neat model area as about 1.25 y. 
The standard error of the y-parallax measurements can be determined 
from the adjustment of the relative orientation. For this purpose at 
least nine measured y-parallaxes should be used. See under point 3 
above. 
Normally u = 0.01 mm or less can be used for the precision plotters. 
The standard error of the residual y-parallaxes would consequently be- 
come about 0.013 mm. 
If the distortion amounts are determined from eq. (1) we find the 
standard error of the distortion amounts as 
  
Or 
we 
de: 
ob 
Ic
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.