Full text: Proceedings of the international symposium on remote sensing for observation and inventory of earth resources and the endangered environment (Volume 1)

    
  
   
   
  
  
  
  
  
  
  
  
  
  
  
  
  
  
    
  
  
  
   
   
   
     
er) 
to 
exposure or in the processing conditions of the film should affect 
the secondary textural parameters very little. 
The following secondary parameters were used for the further 
evaluations: 
area at b 
perimeter at b 
area integrated from a to b 
area integrated from b to c 
perimeter integrated up to b 
perimeter integrated from b onwards 
number integrated up to b 
number integrated from b onwards 
maximum of perimeter up to b 
maximum of perimeter from b onwards 
maximum of number. 
—_ OO WoT Ud wh += 
pp 
The correlations between these secondary parameters as well as 
their dependence on the angle of view were studied. Many combinations 
of these parameters were found useful for classifying the samples 
according to tree species. Some combinations showed very little 
dependence on the angle of view. 
As an example, a classification according to tree species was 
possible using secondary parameter No 9 (maximum of perimeter up 
to b). As this parameter is not independent of the angle of view, it 
was necessary to apply an additive correction using a linear function 
of the image coordinates of the sample area. This correction 
function was obtained by linear regression from a large number of 
sample areas of all tree species under consideration. The frequency 
of occurrence of the corrected parameter No 9 for three different 
tree species is shown in Fig. 6. On the other hand, the ratio of 
secondary parameters No 9 and No 3 showed little dependence on the 
angle of view. This ratio could be used for tree species recognition 
without applying any correction for the angle of view (Fig. 7). 
The use of textural parameters turned out to be indispensable for 
a separation of spruce and pine. On the other hand, it was not 
possible to separate spruce and fir using textural parameters, while 
this separation was easily achieved with a spectral signature. 
Conclusions 
The use of the Quantimet and, of course, of similar systems of 
other manufacturers, appears to be a promissing technique for 
extracting textural parameters from aerial photographs. These
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.