3 
In the resulting system, the first three of the above functions are per 
formed by the image correlator. Tasks 4 and 5 are shared between the ( 
computer and the correlator. Tasks 6 and 7 are performed by the com 
puter, using information which the correlator supplies. 
BASIC AP-2 SYSTEM 
Figure 1 is a photograph of the basic AP-2 system, and Figure 2 
is a block diagram showing the general organization of the system. As 
indicated in both figures, the system consists of three major elements: 
viewing unit, computer, and coordinatograph. The viewing unit contains 
two precision carriages, capable of independent motion in the horizontal 
plane, on which are mounted the two photographs of a stereo pair. It 
also contains the binocular viewing optics which permit the operator to 
view a selected area of each photograph. The computer is a stored- 
program, digital machine, containing both a whole-number or general- 
purpose section and an incremental or DDA (digital differential analyzer) 
section. 
The prime function of the computer is to control the motions of the 
two photocarriages, with respect to their independent viewing optics, so 
that a properly-fused, stereoscopic image is presented to the operator 
at all times, with a floating mark accurately indicating the specific model 
point called for by the handwheels and the footwheel. In determining the 
required photocarriage positions, the computer takes into account the pre 
cise angles and positions at which the two photographs were taken, as well 
as basic camera geometry. It also corrects for earth curvature, atmos 
pheric refraction, camera lens distortion, film shrinkage, and any system 
atic errors that can be defined. In the case of convergent or panoramic 
photography, it controls the visual magnification and image rotation in 
each optical channel as a further aid in stereoscopic viewing. 
The computer also plays a major role in the initial set-up and 
orientation procedures. In interior orientation, the principal point of 
each photograph is calculated automatically on the basis of camera cali 
bration data and measurements of the fiducial locations. Relative orien 
tation may be achieved through conventional procedures or as a least- 
squares solution utilizing from 5 to 12 parallax measurements. Absolute 
orientation may be determined by the computer on the basis of measured 
coordinates of three or more previously entered control points.