I am using jTessBoxEditor to train tesseract-ocr for Uyghur language.
Uyghur language is using Arabic Scripit, but there is some special charaters. so we cannot use Arabic training data.
jTessBoxEditor can create tif file and box file. But when run training progress I got lot of errors like below
FAIL!
APPLY_BOXES: boxfile line 5/ا ((2322,3167),(2332,3195)): FAILURE! Couldn't find a matching blob
FAIL!
APPLY_BOXES: boxfile line 7/م ((2285,3165),(2308,3175)): FAILURE! Couldn't find a matching blob
FAIL!
.
.
.
I want to send you, Font, uyghur text file ...
Can you help me?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I am using jTessBoxEditor to train tesseract-ocr for BNazanin language.
BNazanin language is using Persian Scripit, but there is some special charaters . so we cannot use Persian training data .
jTessBoxEditor can create tif file and box file. But when run training progress I got lot of errors like below
APPLY_BOXES: boxfile line 12/و ((2860,4650),(2878,4674)): FAILURE! Couldn't find a matching blob
FAIL!
APPLY_BOXES: boxfile line 461/ن ((3018,3748),(3035,3767)): FAILURE! Couldn't find a matching blob
FAIL!
Hello,
I am using jTessBoxEditor to train tesseract-ocr for Uyghur language.
Uyghur language is using Arabic Scripit, but there is some special charaters. so we cannot use Arabic training data.
jTessBoxEditor can create tif file and box file. But when run training progress I got lot of errors like below
FAIL!
APPLY_BOXES: boxfile line 5/ا ((2322,3167),(2332,3195)): FAILURE! Couldn't find a matching blob
FAIL!
APPLY_BOXES: boxfile line 7/م ((2285,3165),(2308,3175)): FAILURE! Couldn't find a matching blob
FAIL!
.
.
.
I want to send you, Font, uyghur text file ...
Can you help me?
attached font and uyghur text file.
If those warnings are not too many, you can safely ignore them and continue on with the training.
When I get some spare time, I'll try to take a look at the attachment.
Last edit: Quan Nguyen 2014-12-02
I removed the middle section which had repetitive characters and seemed to have few warnings and better output with the resultant traineddata.
Hello,
I am using jTessBoxEditor to train tesseract-ocr for BNazanin language.
BNazanin language is using Persian Scripit, but there is some special charaters . so we cannot use Persian training data .
jTessBoxEditor can create tif file and box file. But when run training progress I got lot of errors like below
APPLY_BOXES: boxfile line 12/و ((2860,4650),(2878,4674)): FAILURE! Couldn't find a matching blob
FAIL!
APPLY_BOXES: boxfile line 461/ن ((3018,3748),(3035,3767)): FAILURE! Couldn't find a matching blob
FAIL!
I want to send you, Font, BNazanin text file ...
Can you help me?
Thanks
Thank you for your help.
Best Regards
Gheyret