ik 
al, 
EEE 
  
EEE 
  
  
  
  
  
d 
  
140 THE CANADIAN SURVEYOR May, 1960 
For the Second Test identical pbotograpbs and identical ground control were used 
by all participants, and this arrangement greatly facilitated the evaluation of the results 
and the final conclusions. The scope and the general program of the experiment was 
decided upon at the meeting of Commission IV in Brussels in May 1958. 
Space does not permit a complete quotation of the specifications for this small-scale 
mapping experiment that were supplied to all participants in this project. In the main 
the experiment consisted of plotting one stereomodel covering 47 km? at a scale of 
1:25,000, with mapping requirements valid for maps at 1:50,000. In addition, the 
participants were requested to determine about 75 elevation points and 13 horizontal 
points. ‘The test was designed to study the following questions: 
(1) systematic errors in photogrammetric plotting, 
(2) accuracy of photogrammetric contouring, and 
(3) comparison between anaglyphic and non-anaglyphic plotters. 
After a primary evaluation, individual results were returned to each participant for 
comments. Again, lack of space does not allow an exposition of some of these replies, 
which were particularly interesting and valuable. 
During the analysis of the results Professor Richard Finsterwalder of Munich 
agreed to study the topographic-cartographic aspect of the project. The analysis of 
the contouring accuracy was conducted by Mr. G. C. Tewinkel of Washington, a 
member of the Commission Board. 
Participants were also requested to perform vertical parallax measurements on the 
oriented stereogram. These measurements were sent for evaluation to Stockholm to 
Prof. B. Hallert, who will report on the results separately. 
Test AREA 
It is very important that a test area should be readily accessible to persons con- 
ducting detail studies for the final evaluation of results. Because of this requirement 
an area about 47 km? in size was selected around the small town of Renfrew, which is 
near Ottawa, the capital of Canada. Its varying topographical character, rivers, occa- 
sional forests, varying land-use, and its abundance of artificial and natural features, made 
this area suitable for this study. Fifteen triangulation points and 78 elevation points 
were established within the test area. All triangulation points were targeted before 
the flight, and well-defined points along the roads and railways were used as elevation 
points. In order to avoid any misidentification of ground control points a short des- 
cription of each was accompanied by a special photograph of 10X magnification, with 
an arrow marking the exact location of a horizontal or vertical control point, as shown 
in Fig. 2. Only the x and y coordinates of four triangulation points (Nos. 1, 3, 13, and 
15) and the elevations of five vertical control points (Nos. 32, 49, 89, 111, and 70) were 
given to the participants. The remaining points were used only for accuracy analysis. 
AERIAL PHOTOGRAPHS 
Several stereopairs were taken at the scale of 1:50,000 over the test area on August 
27, 1958, by a commercial operator using Kodak Super XX film with a Wild RC-8 
camera equipped with an Aviogon lens (f = 153 mm, 230 x 230 mm). The quality of 
the negatives seemed to be adversely affected by excessive graininess. 
The most suitable stereopair was selected and about sixty contact diapositives were 
produced on a parallel-light printer, on glass plates 3 mm thick. Printing conditions 
were rigidly controlled and all possible care was taken to ensure the identicalness of the 
diapositive material. For those who requested it, Kelsh diapositives were produced on 
the parallel light printer through the film base. The reduced diapositives for the 
Balplex plotter were produced outside the National Research Council laboratories. 
In addition, for a better evaluation of some variable factors in photogrammetric 
procedures, 25 stereopairs were taken of a similar test area with an RC-5 camera