Sorry guys, I really missed the discussion.
I guess there are a few options and we have to combine a few things here
and there.
So what is the best combination as of now?
Happy hacking.
Krishnakant.
On Friday 11 July 2014 12:49 PM, Willem van der Walt wrote:
The abbyy engine does auto rotation detection or whatever the correct
term is. I found that one need to use the contrast setting of
scanimage for best results.
On Fri, 11 Jul 2014, Tony Baechler wrote:
I vaguely recall Tesseract having an option for this, but it isn't
automatic. Convert from ImageMagick should do that as well, but it
isn't
automatic either. The short answer is trial and error if memory
serves. I
remember thinking that maybe the reason for the terrible OCR is due
to the
pages not being aligned and rotating the images, but I didn't get any
better
results. I haven't played with the other OCR engines. I think
FineReader
is better about this. I'm possibly wrong here, but as I understand
it, the
Windows software for the blind does the image rotation before passing
it to
the OCR engine and detects the page misallignment during the scanning
process. The Internet Archive seems to use FineReader and scans
millions of
books in all kinds of conditions, so perhaps it can handle the rotation
automatically.
On 2014-07-10 06:49 AM, Sam Hartman wrote:
Is there a way to get tesseract or openocr or anything open-source to
deal with rotations?
The commercial software along with anything targeted for the blind
tends
to
1) deal with 90 or 180-degree rotaions--I put the book down on the
glass
in the wrong orientation
and
2) Deal with small rotations (it wasn't perfectly aligned) relatively
well.
I find these features really important when scanning things myself.
Less so when OCRing images from the web etc.
_______________________________________________
Blinux-list mailing list
Blinux-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/blinux-list
--
Have a good day,
Tony Baechler
tony@xxxxxxxxxxxx
_______________________________________________
Blinux-list mailing list
Blinux-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/blinux-list
--
This message is subject to the CSIR's copyright terms and conditions,
e-mail legal notice, and implemented Open Document Format (ODF)
standard.
The full disclaimer details can be found at
http://www.csir.co.za/disclaimer.html.
This message has been scanned for viruses and dangerous content by
MailScanner,
and is believed to be clean.
Please consider the environment before printing this email.
_______________________________________________
Blinux-list mailing list
Blinux-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/blinux-list
_______________________________________________
Blinux-list mailing list
Blinux-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/blinux-list