I'd like to be able to pull the confidences from the character recognition 
engine - ideally at the character level but the word level would be 
sufficient.  Looking at the hOCR specification 
(https://docs.google.com/document/d/1QQnIQtvdAC_8n92-LhwPcjtAUFwBlzE8EWnKAxlgVf0/preview)
 
it seems like this information should be embedded in the hOCR output.  
However, the most recent version of ocropus (0.7) doesn't appear to contain 
this information in the hOCR output.  The only data I see in the hOCR 
output are the bbox coordinates.  Are there any command line switches I can 
include to provide additional information in the hOCR output, specifically 
the confidence measures?  Or can this only be achieved programmatically?  
If so, has anyone had any luck patching the code to provide this additional 
feature?

-Elliot

-- 
You received this message because you are subscribed to the Google Groups 
"ocropus" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/ocropus/9e536aa5-9918-4071-9dd0-87f5b5a1c535%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to