baopham1340 / tesseract-ocr

Automatically exported from code.google.com/p/tesseract-ocr
Other
0 stars 0 forks source link

Sometimes Strange Output: FAILURE! Couldn't find a matching blob #1381

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago
What steps will reproduce the problem?
1.use eng.arial.01.tif and eng.arial.01.box to get eng.arial.01.tr, use this 
command: tesseract.exe -psm 10 eng.arial.01.tif eng.arial.01 nobatch box.train 
. But I get the error as the summary.
2.use eng.arial.02.tif and eng.arial.02.box to get eng.arial.02.tr, use this 
command: tesseract.exe -psm 10 eng.arial.02.tif eng.arial.02 nobatch box.train 
. It works.
3.I used the same .box file (01.box is same to 02.box),and 02.tif add one 
character based 01.tif

What is the expected output? What do you see instead?
In both .tif/.box picture, we marked character block is same, only the first 
character on the left side is marked. the right side of the picture has 
different content, but not marked on the right. But they are one success, 
another failed, why?
What version of the product are you using? On what operating system?
tesseract 3.0.2 windows

Please provide any additional information below.
The outside of marked area(.box) will effect the result about marked character? 
I can't understand the reason,please help me. Thank all!

Original issue reported on code.google.com by houandli...@gmail.com on 14 Nov 2014 at 7:26

Attachments:

GoogleCodeExporter commented 9 years ago
Please follow wiki instruction[1].

[1] https://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3

Original comment by zde...@gmail.com on 2 May 2015 at 1:15