id 
lytical photogrammetry. Improved error models 
developments are (1) improved error models to 
compensate for systematic errors; (2) universal 
use of radial and tangential weight functions; 
(3) auto-regression processing of statoscope 
observations; (4) doppler control of large projects; 
(5) incorporation of inertial surveying methods; 
and (6) global positioning systems. 
In the course of the general discussion the fol- 
lowing comments were made. 
Dr. G. Kupfer (F.R.G.) observed that in the S.W.A. 
photography of the Rheidt test field the vertical 
accuracy obtained was as expected when field 
calibration was used. To this Mr. Brown replied 
that in field calibration it is essential that the 
platen is calibrated as part of the radial distortion. 
Dr. F. Ackermann (F.R.G.) said that in Ober- 
schwaben no improvement was achieved by using 
self-calibration. He added that the reason for 
this was to be sought in the fact that in the Test 
Oberschwaben the image quality of S.W.A. pho- 
tographs was much poorer than that of W.A. 
ones. 
Dr. D. Merchant (U.S.A.) observed that vacuum 
causes a deformation of the platen, and that the 
deformation should thus change with the flying 
height. He asked whether it was not necessary 
to perform a calibration separately for every 
flight altitude. Mr. Brown answered yes and no, 
saying that the quality of new platen is so good 
that deformation caused by vacuum is not 
significant. 
Dr. H. Ebner (F.R.G.) wondered whether an effort 
should not be made to standardize the correction 
terms used in self-calibration. Mr. Brown gave a 
positive answer, but added that further practical 
experience is necessary so that the best para- 
meters can be found. In particular, the connection 
between the physical sources and systematic 
errors should be further studied. Very little has 
been heard about the systematic errors of the 
Wild cameras, especially RC 10. 
Mr. W. Pryor (U.S.A.) said tests using the Zeiss 
aerial camera indicate that the vacuum platen 
System was the cause of large image position 
deformations. Newly designed platen had greatly 
reduced such deformations. He asked whether 
this meant that all aerial cameras should be 
tested for the deformations caused by the vacuum 
platen in use. He also asked whether it would be 
possible to develop camera calibration parameters 
for each type of aerial camera and, if so, should 
all aerial cameras be tested, using Mr. Brown's 
methods, before use for aerial analytical trian- 
gulation. Mr. Brown answered that he considered 
it appropriate to calibrate cameras before use. 
He said the quality of new platen was good, but 
that they are deformed with the times, so much 
so that they are rejected after the lapse of 4 to 6 
years at the U.S.G.S. 
Presented Papers 
Authors: Mr. E. Sigmark (Sweden) 
Mr. A. Andersson (Sweden) 
Title: Block Triangulation by Independent 
Three-Dimensional Method PP 111/55 
A description of the Swedish block triangulation 
system "STEREO" was given. Here use was 
made of the method of independent models for 
adjustment in planimetry and height according 
to the method of least squares. Practical results 
were demonstrated and discussed. 
Dr. P. Stefanovic (Netherlands) asked Mr. Sigmark 
whether this program could be used for the 
detection of gross errors. The answer given was 
to the effect that this obviously was not possible 
in an automatic way. 
Authors: Dr. B. Shmutter (Israel) 
Dr. A. Perlmuter (Israel) 
Title: Block Triangulation PP 111/54 
Drs. Shmutter and Perlmuter discussed a sim- 
plified version of a bundle adjustment compatible 
with a small computer. 
This triangulation procedure is applicable to 
photographs measured on a comparator as well 
as to stereomodels evaluated in an analogue 
instrument. The idea of the method is to replace 
the simultaneous solution of all unknowns by 
consecutive solutions of groups of unknowns, 
which is achieved by performing series of spatial 
resections and intersections of points taken 
one at a time. 
The paper illustrates the method with two com- 
puted examples, a block consisting of single 
photographs and a block consisting of models 
evaluated in an Autograph A 8. 
Author: Mr. G. de Masson d'Autume (France) 
Title: L'interpolation par une régle flexible 
(spline) et ses applications en pho- 
togrammétrie numérique PP ////39 
Mr. de Masson d'Autume, in his paper given by 
a colleague, presented a method for "Interpolation 
by a Flexible Rod (Spline): Applications to Digital 
Photogrammetry”. 
Interpolation by flexible rod (spline) on N points 
can be carried out using an interpolation function 
which gives the value of a particular point as a 
linear function of the values of the N basic 
points. A very simple algorithm, using precalcu- 
lated interpolation matrices, enables one to cal- 
culate the coefficients. The interpolation algo- 
rithm can be applied to various problems of 
digital photogrammetry, the most important ap- 
plications being the parametrization of the sys- 
tematic deformations of a bundle of rays, inter- 
polation in a random distribution of points, 
filtering of grossly inaccurate data and finally 
an extremely simple method for the adjustment 
of aerial triangulation blocks. 
Tests of the latter method yield results com- 
parable to those obtained using independent 
models or the bundle adjustment. 
Wednesday, July 14, 1976, from 14.00 to 15.30 
Session 4 of Commission lll 
Topic: Digital Terrain Models, Digital Map- 
ping 
Chairman: Dr. K. Kraus (Austria) 
The following Presented Papers related to the 
main topic of the session had been handed in, 
but could not be presented due to the lack of 
time. 
Authors: Dr. M. Allam (Canada) 
Dr. C. Wong (Canada) 
Title: Gridding Topographical Surfaces 
PP 111/05 
101