kcobra / tesseract-ocr

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

Japanese ref image to create language train data is not recognized correctly #1414

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago
What steps will reproduce the problem?
1. run tesseract with the attached image for OCR output
2.
3.

What is the expected output? What do you see instead?
All characters in the image file to be recognized correctly as the image was 
used to create traindata. But there are many mistakes in the output file.

What version of the product are you using? On what operating system?
Windows XP, 3.02

Please provide any additional information below.

In some cases, the output of default japanese language(jpn) traindata output is 
better than the one I created. 

Original issue reported on code.google.com by sivakuma...@gmail.com on 30 Jan 2015 at 8:34

Attachments: