206 
COMPOSITION CODING OF REMOTE SENSING DATA 
FOR DEVELOPING GEOINFORMATION TECHNOLOGIES OF 
OF RESEARCH OF NATURAL ENVIRONMENT. 
N. F = AFANASYEV. 
Aroal gamat i on "Aerogeol ogy ” , Al 1 -Un i on Research 
Institute of cosmoaerogecl ogi cal methods 
CVNIIKAM) of USSR. 
I n t roduL t i on. Remote sen si r.g of 
ti e henef i t s of agr i oui ture . sel vi 
■ »y drology j geology, ecology, are based 
of subjeet-s-ref1ected radiation. As 
natura! environment to 
culture, meteorclcgy, 
on usage of intensity 
a result of modulation 
o f p r oh i n g r a d 5 at i on n a t ur a 1 en v i r on men t f or ms an 
information field reflecting its state at the moment of 
3urvey. However. due to the fact that signals intensity 
depends on integral effects of interaction between 
environmental objects, direct interpretation of intensities 
field is valid only for a limited number of cases when the 
study objects arevisible, In most practically important 
cases of non-visible objects it is necessary to define 
direct and indirect <intermediate' features of objects which 
can be correspondingly interpreted. 
Additional information capacity of remote sensing is 
achieved by two following ways; 
First, raising selectivity of sensing systems by choosing 
such values of spectral and geometric resolving capacity 
which reveal the features of study objects in intensity and 
spatial structure of fields recorded. Second, decreasing 
information redundancy and applying the theory of image 
estimation and identification. Positive results with the 
help of these methods can be obtained, however, by 
processing only homogeneous information structures. On the 
contrary, the geosystem has a heterogeneous structure, and 
the geoinformation structure (structure of information 
fields/ it engenders is also heterogeneous. Thus,