23 
cantilever extension and bridge, in which 25 pass points per photo were used 
for the triangulation; and (b) Test Cases 11, 13 and 14 in which rejection 
criteria of 0.008, 0.008 and 0.005mm, respectively, were used to eliminate 
points in which plate coordinates had the largest random errors. 
Control and pass point layouts and a summary of discrepancies in Test 
Cases 9-1^ are illustrated in Figure 3.04 and Table 3.02. 
C. TESTS USING ACTUAL PHOTOGRAPHIC DATA 
A limited amount of testing was performed with the Arizona Test Strip 7. 
This strip of photography extends approximately 17 miles and is at a scale 
of 1:20,000. Photo coordinates were measured on the Nistri TA-3 stereo 
comparator and were corrected for lens distortion and differential film 
distortion. 
Computations were performed only through the stages of relative 
orientation and triplet assembly so that plate coordinate residuals and the 
standard error of unit weight of the triplet relative orientations could be 
examined Results are tabulated in Table 3.03 for 15 photographs of Strip 7*