Full text: Systems for data processing, anaylsis and representation

  
  
The numbe 
Data Base of the grou 
a position . 
| can be con 
Referential Classific. Stage of the clas: 
are related 
  
  
  
  
  
  
  
  
  
Comparison 
Ordinary Classification where p(w 
the image. 
| x will be: 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
User x 3 
  
  
  
  
  
  
Figure 2, Block Diagram of the Referential Classification. As we me 
two stages, 
applying 
mentioned 
Accuracy (%) spectral co 
second stag 
Class Pixels coordinate 
the first st 
base and c 
Poplar 800 81 81 0 same coor 
W. Chestnut 1612 89 96 7 means that 
Forest 3791 74 88 13 yea £0 
Grass 2718 81 97 16 check resu 
W. Bodies 2126 78 94 16 ans On 
S. Beet 2821 60 85 25 
conflicting 
the data 
Total 13568 75.6 92.8 17.2 classificati 
required. I 
such pixel 
classificati 
informatio 
classificati 
The result 
Accuracy (2) site image 
100 quality an 
accuracy | 
BOF — Referential classificati 
Supervised ES. 167 
i ; : chestnut, 
eo [^ 7773 Differenc ctv 
40r the studie 
Further c 
20r classificat; 
accuracy ( 
0 | CT |! } | | | | Class than that 
Wors 
Poplar Chestnut Forest Grass Waters Beet tccas 
happens v 
collected 
n classifi 
Figure 3, Graphical Comparison of Supervised and Referential Classification Accuracies. fsified, 
  
  
Supervised Referential Improvement 
  
  
  
Table 1, Accuracies of Supervised and Referential Classification. 
  
  
  
  
  
  
  
  
  
  
334 
  
 
	        
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.