Average standard error(um) 
um) 
Average standard emor (| 
Fig.5 Comparison of the Standard Deviation vs. the 
Uncorrected Case and the Corrected Case. 
  
e 9 9 5 e e e 
  
° o o © 2. 2. 
Target array 
  
  
  
  
object 
distance:0.5m 
exposed 
1.0m position 
2.0m 
Fig.4 Test Field of Simulation Test. 
  
600 
S00[- 
UN—CORR. | 
EE 
4001 CORR. 
  
  
Distance 
a) Plumb line method 
  
  
  
  
600 
500 SD 
Un—corre 
a 
4001 Correct 
  
  
  
  
Distance (m) 
b) Collimator method 
  
  
    
Three dimensional position errors illustrated in 
Fig.6 show that the more targets place at fringe of 
film, the more errors increase gradually. Also, the 
case that systematic errors were corrected was 
distributed lowly by about 30-40ym in comparison 
with uncorrected case. Thus it is proved that 
calibration by collimator method or plumb line 
method is effective considerably. 
standard error(um) 
bo. 00 66,64 83,27 99.91 116,65 133.19 
¢ 
   
      
     
       
      
y 
  
    
  
    
    
  
Q 
SR 
SA 
SA 
OS OUEN 77 7 
XE 7 LIGA 
VS 
CSS T 
         
etenderd error(um) 
     
   
   
  
   
  
  
  
   
SNS 
SEEKERS 
zz 
SEE A 
CELL 
SX [77772 
SS 
    
standard error(um) 
dro. 00 587.04 564.08 621.13 678.17 
é 
  
   
      
   
2 
RE 
SL PH 
SCIE 
m 
Fig.6 Comparison of Errors vs. the Uncorrecte 
Distortion and the Corrected Distortion 
by Collimator Method.