Hello everyone!
I'm implementing, on Android, some digital Scrabble tile recognition, and I
got some weird recognition behaviour.
After a few transformations, the images I got are these ones:
..............................
and the recognition sometimes recognize a O as a I, or a D as a I.
The settings I use are:
setPageSegMode(TessBaseAPI.PageSegMode.PSM_SINGLE_CHAR);
setVariable(TessBaseAPI.VAR_CHAR_WHITELIST, "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
setVariable("load_system_dawg", "0"); // But it seems not to change anything
setVariable("load_freq_dawg", "0"); // But it seems not to change anything
I would find pretty weird that with as much as obvious and digital (= perfect)
characters (O is obviously not a I !!), there aren't any Tesseract settings or
base images transform I can make to improve recognition.
Do you have some ideas to improve this system?
I was thinking of training Tesseract on this font, but it seems to be a huge
work!
Thanks a lot!! :)
--
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" 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].
Visit this group at https://groups.google.com/group/tesseract-ocr.
To view this discussion on the web visit
https://groups.google.com/d/msgid/tesseract-ocr/84897f24-ef06-4c40-890e-3af4060c2f92%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
