Hi!

 

I have a problem with Dempster-Shafer fusion of classifications. When I do 
it in a 'normal' window, it says there was a Python error. I have to do it 
in the batch mode instead.

 

First, I chose 2 classification rasters (they have 2 classes: 0 and 1 in 
each of them), then 2 confusion matrices in txt format which I’ve done in 
ComputeConfusionMatrix using ground truth points (Do the matrices' files 
have to have the same names as rasters they refer to? How the program will 
know which matrix is for which classification?). I put 5 and 10 for label 
for the NoData and Undecided classes.

 

Then there is the mass of belief measurement. What is it exactly? Can 
someone explain it to me, please? I used all of them and here is what I got:

- 'accuracy' --> program crashes but still produces an output which looks 
exactly like one of the input classifications.

- 'kappa' --> program crushes, the output has only two values: NoData and 
Undecided.

- 'recall' and 'precision' (output looks the same) --> program crashes, 
output isn’t correct (I was expecting to get a result that’s better than 
the input classifications)

 

What does the mass of belief measurement mean and how do I get it to work 
without crushing? Every tool in Orfeo Toolbox is very poorly described. I 
don’t know if I can trust the results if I don’t know how it works or why 
it keeps crushing. 

 

I’m using OTB 5.10.1 on Windows 10.


Have a nice day!

Monika

-- 
-- 
Check the OTB FAQ at
http://www.orfeo-toolbox.org/FAQ.html

You received this message because you are subscribed to the Google
Groups "otb-users" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/otb-users?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"otb-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to