AT RR
ti rs nsc
4
The number of templets, necessary for a certain region, is much smaller than the
number of photographs. The scale of these templets may be much smaller than the photo-
scale and may be adapted to the map-scale. In this case, the application of perfections,
as described by P. O. Fagerholm in his dissertation: Mechanical Radial Triangulation, is
certainly justified.
After the determination of the adjusted coordinates of the block-cornerpoints each
individual block coordinate system
can be transformed into the map
coordinate system by a linear trans-
formation. The remaining discrep-
ancies will probably be of the order
of 0,2 or 0,3 mm on map scale. They
may be distributed over the block by a
simple graphical interpolation, if
necessary, without disturbing the
relative accuracy.
Par.4. Conclusions.
Compared with a normal slott-
ed templet triangulation the above
method has a number of advantages p
of which the most important are:
Fig. 4.
a. a considerable decrease in the is
number of templets;
b. automatic elimination of influences of tilts and heights differences;
c. templets can be made on map-scale instead of on photo-scale so that the size of the
lay-out is smaller;
d. the result will have a greater relative and absolute accuracy.
If a spatial triangulation is executed, the application of the suggested method requires
some extra work, compared with a normal slotted templet triangulation.
This consist of:
a. The calculation of machine-coordinates for each strip;
b. Some linear transformations to form blocks;
c. The calculation of final coordinates, eventually followed by a graphical interpolation
of remaining errors.
Compared with other existing adjustment-methods (see publications 1—5) the above
method gives a considerable saving in time. It is especially justified in those cases where
the ground control is sparse, so that the methods described in 2—5 cannot be used. Compared
with the method, given in (1) it has the advantage that a lot of tedious computation work is
saved.
For small scale topographical maps the application of a rigorous adjustment method
by means of electronic computers is not justified for economical reasons.
By using any approximate adjustment method we have to accept a loss in accuracy
but this is not harmful if a certain relative accuracy is still maintained.
This is so for the suggested method and for this reason it could be characterized as
“rational”.
(1) W. Schermerhorn, Experience with aerial triangulation, Photogrammetria 1940 nr. 1.
(2) H. Bonneval, I’Aérotriangulation spatiale à l’I.G.N., Archives Int. de Photogram-
métrie Tome XI—II, p. 483.
(3) M. Zeller, Practical Experience etc. Photogrammetria 1951/52 nr. 4.
(4) Brazier and Williams, The adjustment of a block, Photogrammetrie Record Vol. 1
nr. 4, 1954.
(5) B.St.G. Irwin, An experiment in aerial triangulation, Photogrammetric Record Vol.
1 nr. 5, 1955.
