Thanks, I appreciate the suggestions!
On Friday, November 2, 2012 1:48:45 PM UTC-4, sventech wrote:
>
> Cutting off the borders and possibly adding white borders might help.
> Normalizing out the text that bleeds through the page would also help.
> The text is clear, so you might not need to ret
I want to build the tesseract 3.02.02 project so that I can modify some
code to tune it to some specific task.
Version: tesseract 3.02.02
Ubuntu 12.04, Eclipse Juno
I put the tesseract into the Eclipse project.
Include directories
/usr/local/include
/usr/local
/usr/include/leptonica
and all fil
Version: tesseract 3.02.02
Ubuntu 12.04, Eclipse Juno
I am trying to use ViewerDebugging.
Following the instructions in
http://code.google.com/p/tesseract-ocr/wiki/ViewerDebugging
I installed javac
download piccolo-1.2.jar, piccolox-1.2.jar, and make ScrollView.jar
Then I use export to set the
Hi Dmitri,
How do we clear the adaptive classifier?
Can I please know, what is the API or function for clearing the adaptive
classifier?
Best Regards,
- ganesh
On Friday, November 16, 2012 3:39:22 AM UTC+8, Dmitri Silaev wrote:
>
> Sriranga,
>
> All you can specify in the command line can be s
There is a newer wrapper for 3.x version:
http://code.google.com/p/tesseractdotnet/w/list
I think it was made by the developer of VietOCR
--Sven
On Thu, Nov 15, 2012 at 5:06 PM, zdenko podobny wrote:
> On Fri, Nov 9, 2012 at 1:43 PM, Troy Frazier wrote:
>
>> Is it possible to search an image
On Fri, Nov 9, 2012 at 1:43 PM, Troy Frazier wrote:
> Is it possible to search an image for a particular word using the Tessnet
> wrapper? I see that it is possible to limit your scan to certain
> characters, but what I would like to do is to input a word and have all
> instances of that word be
Hi Rey,
The Shared Questionnaire System (SQS) is doing something here under Apache
license:
http://dev.sqs2.net/projects/
in Java, XSLT and JavaScript
And queXF assumes you create the forms yourself (under GPLv2)
http://quexf.sourceforge.net/
for tesseract's license Check here:
http://www.apache.
Regarding "user_patterns_suffix" have a look at tesseract manual page [1].
I am not sure if there is possibility to force tesseract choose ocr
output from dictionary (I never tried it ;-) )
But you can increase dictionary strength with variables
language_model_penalty_non_freq_dict_word and
la
Can you please use 3.02 version instead of 3.01 and write exact error
message?
There is possibility to copy text from windows console - select relevant
text/lines with pressed left mouse button then click with right mouse
button outside of selected text but in console window - highlight will
di
Yes, I think the text size (x-height) was too small. Also, the English
language data may be trained with more fonts, given that Google created it.
--Sven
On Thu, Nov 15, 2012 at 6:43 AM, sascha4j wrote:
> after converting the image with imagmagick the result is better. not 100%
> but nearly.
>
after converting the image with imagmagick the result is better. not 100%
but nearly.
the options for imagemagick were
convert -colorspace gray -resize 200% -unsharp 0x8+1.5+0.05
Am Donnerstag, 15. November 2012 10:26:21 UTC+1 schrieb sascha4j:
> Hi,
>
> i try to ocr some scanned text w
Hi,
i try to ocr some scanned text with tesseract-ocr.
for some images the result is quite good.
but for this one ( see attached file) the result is poor.
any hints why ? and what i could do to get a better result?
i use tesseract 3.0.2 with german language.
greetings
sascha4j
Thanks very much for your responses zdenop,
I'm not used to dev in open source projects like this, perhaps you may help
me to understand, for example if I implement a feature to add character
rect&confidence to the hocr output, how this is translated to the main
project (if it is good enough
On Thu, Nov 15, 2012 at 10:15 AM, José Luis Rey wrote:
> Thanks very much for your responses zdenop,
>
> I'm not used to dev in open source projects like this, perhaps you may
> help me to understand, for example if I implement a feature to add
> character rect&confidence to the hocr output, how
Hi friends
I am using a static TessBaseAPI object in my application. This object gets
initialized and reads, processes the training data at the startup of the
application.
Then, this application processes multiple scanned images through the
TESS_API TessBaseAPI::ProcessPages() function, using
15 matches
Mail list logo
