The attempt of dividing between remote sensing and photogrammetry could lie 
in assigning these transformations to that range, where they are mainly 
operating (feature, place): 
- remote sensing: transformation of features 
T RS - T(G,M) = T(I,M) = T(M) 
-photogrammetry: transformation in space 
T pH = T(G,M) = T(G,M(G)) = T(G). 
On this basis can be done the following assignements: 
- remote sensing 
classifikation of multispectral images, 
transformation in an abstract feature space (e. g. integral transformations) 
- photogrammetry 
correction of geometric differences in multitemporal images, 
digital geometric transformations. 
The new possibilities of a fully digital data structure must also lead to 
basically new approaches within photogrammetry, because the specific photo- 
grammetric instruments lead to special photogrammetric software for universal 
computers - completed by special processors - with specific input and output 
devices. In this direction possible data structures as well as algorithmic 
solutions of photogrammetric tasks have to be viewed.