Last line is poorly recognized

[Дарья Арбузова]

Hello!

I'm surprised I couldn't google this problem.

I'm using tesseract to recognize a tif-file, which is an image of a doc-file, but turns out the line with a lot of empty space before/after it is pain. It's being recognized in all caps and with bad quality.

Maybe the problem here is to allocate the line?

Any thoughts?

Thank you!

Daria

[Дарья Арбузова]

Maybe I should specify a different pagesegmode?

четверг, 9 октября 2014 г., 13:45:17 UTC+4 пользователь Дарья Арбузова написал:

[Zunair Fayaz]

Try exporting your doc with 300dpi or more

[Дарья Арбузова]

Zunair, I already used 600 dpi.

I'm now wondering whether it happens only with Russian text.

четверг, 16 октября 2014 г., 19:07:49 UTC+4 пользователь Zunair Fayaz написал:

[bulk...@gmail.com]

С русским у тессеракта совсем беда :)