Hempenius, Makarovië, Van der Weele, Tests of Restitution Instruments 
The results can be interpreted easily if the disturbances are presented in the following 
combinations of pairs: 
a) , the á (I, r) and d (l,t) group in the projection, 
b) . the d (l, t) and D (L, T) group in the output (drawing table), and 
c) . the ô (A, r) and D (L, T) group in the output. 
In total there are six combinations of pairs (ô',d'), (ô",d"), (d',D), (d",D), (ô',D) 
and (ô", D). It is convenient to represent them graphically as shown in fig. 13a, 13b and 
13c. In case <5'~<5" and d'three combinations (for the mean values of <5 and d) are 
sufficient. 
Intersection points of lines, representing individual groups of distortions, define the 
scale parameters, at which the distortions are equal. The actual limits of the scale para 
meters should also be indicated in the graphs. 
Such graphs may be considered as comparable for similar plotting instruments when 
assuming that the scanning time had been adapted to the required care. 
APPENDIX 
The fundamental approach of the experimental analysis proposed may be applied to 
any type of plotting apparatus if the information flow is clearly defined and the succes 
sion of individual tests is arranged accordingly. 
In the analytical plotters the projection system and measuring device are replaced by 
digital or analog computers. The output of the computer are the instructions for the cor 
rective motions applied the photographs (or movable optical system) and the input for the 
plotting device. 
In automatic plotters (e.g. Stereomat) the human operator in the feedback loop is 
replaced by optical, electronic and mechanical substitutes (for contouring). The analysis 
of this sophisticated component of the system is rather complicated. It involves several non 
dynamic phenomenae. The dynamic performance could in principle be investigated similarly 
when using as the input a sinusoidally undulated steep slope model or a vertical sine*- 
edge model. 
In the near future even more sophisticated plotting systems are expected. This requi 
res, however, a higher standard of analysis methods for the design, comparative testing 
and the maintenance. The combined application of the information theory and system 
analysis appears to offer adequate theoretical tools for this purpose. 
LITERATURE 
[1] Hempenius, S. A., Specifications for mirror stereoscopes. Int. Archives of Photogram- 
metry, Delft, 1962, Vol. XIV, p. 45. 
[2] Hothmer, J., Errors in gimbal axes, Photia, 1955—56, p. 249. 
[3] Lychen, L. E., Numerical adjustment of X-inclination and latitude distortion in 
stereoscopic plotters. Int. Archives of Photogrammetry, Stockholm 1956, Vol. XII, 
Part 4A. 
[4] Hallert, B., A new method for the determination of the distortion and inner orienta 
tion of cameras and projectors. Photia, 1954—55, p. 107. 
[5] Doebelin, E. G., Dynamic analysis and feedback control. Ohio State University, USA, 
1962, Me. Graw-Hill Book Comp, Inc., New York. 
[6] Makarovic, B., Dynamic performance of photogrammetric plotting instruments. I.T.C. 
Delft, 1964, Publication A 35. 
[7] Neidhardt, P., Einführung in die Informationstheorie. Verlagtechnik, Berlin, 1957. 
[£] Shannon, C. E., Communications in the precence of noise. Proc. IRE, 1949, V. 37, 
p. 10—21.