Dne 2.10.2011 1:36, piše B.J.:
I ran into this problem recently. Here is the solution (I'm using
Tesseract 3.01):
to use user-words list, in dict.h and dict.cpp, find user_words_suffix
and change the "" to "user-words"
//dict.h
STRING_VAR_H(user_words_suffix, "user-words", "A list of user-provided
words.");
//dict.cpp
STRING_INIT_MEMBER(user_words_suffix, "user-words",
"A list of user-provided words.",
getImage()->getCCUtil()->params()),
This assumes, then, that in your tessdata folder there is a file named
"eng.user-words" with your user made word list.
.bj.
I have 3.01 from svn too.
And that field's are empty. So I modified as you suggest. But I see no
difference in OCR. The confidence is still low and missreaded word is
still missreaded.
And if I remove 'eng.user-words' then tess just abort execution with
missing eng.user-words statments so I assume that file is oppened and used.
So is there someone smart enought to explain how that
('lang.user-words') works.
And other things.. Is there someone smart enought to change source on
svn to have that included but just to check if user-words exist not to
popup error? (as I know the lang.user-words is optional so keep is like
that.)
Thanks...
--
You received this message because you are subscribed to the Google
Groups "tesseract-ocr" 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/tesseract-ocr?hl=en
