ial 
'ec- 
nr- 
| zn 
3.3. Geometrical aspects of remote sensing 
Ever since remote sensing has established itself as a separate 
field and has pursued wide development, contributions have been 
submitted at ISP-congresses and symposia to almost all commis- 
sions, at increasing rates. The attempts of internal structu- 
ring of remote sensing too have allotted the geometrical as- 
pects of remote sensing to Commission III. A working group 
has been established under the chairmanship of Prof. Konecny, 
Hannover. The activities of the Working group cover vast 
experimental programs and numerous special studies, the results 
of which will be presented and discussed at the Helsinki con- 
gress. 
The first central theme of the working group concerns the geo- 
metrical accuracy of radar mapping. SLAR imagery has been sub- 
mitted to blockadjustment. In the Phoenix Radar Test planimetric 
accuracy values of 3 times the limits of resolution have been 
obtained (50 m). The planimetric accuracy of SLAR is consi- 
dered sufficient for map scale 1:250 000. The further develop- 
ment attempts to meet accuracy specifications of large map 
scales, in particular also with respect to heights. 
The second problem area is the derivation of geometrically 
correct thematic maps from scanner imagery, especially from 
thermal infrared scanners and from multispectral scanners. 
Test results have shown that the geometrical restitution reaches 
values around the limits of resolution of the various Scanners, 
both from satellites or airplanes. Such investigations are 
related to digital image processing, especially to matching 
different images and to change detection. Special technical 
problems are differential orientation and differential rectifi- 
cation of image elements, geometrical image distortion, and 
actual mapping with analytical plotters. 
The recent investigations have demonstrated the principal accu- 
racy potential of various sensors. Apart from radar mapping the 
development therefore attempts ultimately thematic mapping by 
fully- or semi-automated methods of adequate geometrical accu- 
racy from scanner imagery. 
The developments are not likely to have direct effect on con- 
ventional techniques for medium or large scale topographic 
mapping. And much is still in a state of research, experimenta- 
tion and testing. Principles and technology of acquisition of 
geometrical and physical information by remote sensing have, 
however, reached an advanced stage of development. Imagery and 
data are or become available in great quantities. Thus the 
necessary conditions are given for the further development of 
economic application.