£e (D 
st 
hc 
The process can be terminated if the following conditions are ful- 
filled; 
For a block containing n models, the scale-tilt and height procedure 
involves the solution of a system of n linear equations and next a 
system of 3n non-linear equations (apart from the Z;-coordinates 
of the points to be determined). The next procedure involves the 
solution of a system of 2n non-linear equations and then a system of 
n linear equations. 
Tilt height procedure 
(a) TILT-DETERMINATION 
Finally, we consider points j and jtl in the overlap of the models 
(1) an (itl). For the points in model (i+1) and equation of type (6) 
is written. Eliminating sina..,, and next the ratio A / Ay from 
the equations, equations (6)'änd (5) give 
i i i 
[2314 41794) * 232 O4 44 7749) * 233 Ca ga 7) 1433) 7 
(akt ex -x ) + alten - y ) 
35 Cari; gj 32 Vili iij 
(7) 
i+1 
+ 533 Giu gea agg 
in which 
l1 oit = 1 
235 953 : 
We see that equations (6) and (7) contain only tilt elements (two 
per model). 
The coefficients of the til: elements are formed by xq - and y4 
- coordinate differences. It is therefore recommended to use pairs 
of points parallel to and perpendicular to the line of flight. We 
know, that a system of linear equations can only be solved if the 
coefficient matrix of the system is non-singular. It is also evident 
that the models must now have a forward overlap of 507 (bundle mo- 
dels). 
We consider a block containing only parallel strips and a sidelap of 
15 to 20%. To level the models with respect to the object photo- 
graphed, in each strip a slope 0441 perpendicular and in 
the first strip a slope parallel to the line of flight must be 
known. Any extra slope gives redundancy.