INTRODUCTION 
From the early days of photogrammetry, automatic stereoplotting from 
photographs was regarded as an improbable dream that would never materialize. 
Stereoperception, the basis of the photogrammetric measuring and plotting tech- 
nique, was considered to be a unique physiological attribute of human vision that 
would be difficult to simulate, and so any attempt to eliminate the human operator 
from stereoplotting was greeted with great skepticism. However, with techno- 
logical developments automatic stereoplotting has today become a fact, and its 
unexpected accuracy and speed have already, at this early stage of its develop- 
ment, exceeded the most optimistic expectations. How far this development may 
go in the immediate future, in view of these initial results, is left to individual 
speculation. But there is little doubt that we are witnessing at present an impor- 
tant change in the photogrammetric technique. 
In May 1960 the Photogrammetric Research Section of NRC was invited 
to evaluate from the point of view of photogrammetric applications an automatic 
stereoplotting device developed by G.L. Hobrough of the Hunting Survey Corpora- 
tion Limited of Toronto. For this purpose a series of experiments, described in 
the following report, was carried out. Whenever possible an attempt was made 
to obtain numerical data on the accuracy achievedby the automatic stereoplotting 
device as well as comparative figures from identical operations performed by a 
human operator. This latter information is of importance since the plotter used, 
as wellas the automatic stereoplotting attachment, influences errors. 
  
TEST MATERIAL 
Two different Wild RC-8a stereopairs were used in the experiments. 
One pair was taken from a very low altitude of about 500 m (photo scale of 
1:3000) over rather flat cultivated land with differences of up to 35 m in ground 
elevation within one overlap (Test area A, see Fig. 1). The slope of the terrain 
ranges from 2.5% to 20%. 
The second stereogram (photo scale of 1:50, 000) was identical to that 
used in the Second International Mapping Experiment. Test area B, shown in 
Fig. 2, covers the partly hilly and partly flat terrain around the small town of 
Renfrew along the Bonnechere River. Forests and occasional steep slopes add to 
the variety of the topographical features. The elevation differences range up to 
100 m and the slopes from 0.4% to 70%. The international test area was purposely 
selected because it is very well-known to the mapping centers that participated 
in the experiments organized by Commission IV-3 of the International Society 
for Photogrammetry. 
AUTOMATIC STEREOPLOTTING DEVICE 
The equipment we are concerned with is an improved version of a de- 
vice called Auscor or Stereomat, which has been described by G.L. Hobrough" 
  
  
% Hobrough, G.L. Automatic stereo plotting. Photogrammetric Engineering, XXV, 
No 5, 1959, 763-769.