I am trying using tesseract OCR 3.01 and visual studio 2010 to extract bold
and italic words from an image. For example, if I input a clear image with
text like so:
"The quick *brown* fox *jumps* over the *lazy* dog."
I would like to get an output just like so:" The quick brown fox jumps over
Hi Cian,
It would be best to use the pre-release version 3.02 (from SVN) if
possible, since a couple of important fixes have been made to the hOCR
support.
--Sven
On Mon, Jul 30, 2012 at 8:13 AM, Cian Mc Govern
wrote:
> Hi all,
>
> I'm using Tesseract with the hOCR output format. When I invoke Te
On Mon, Jul 30, 2012 at 3:13 PM, Cian Mc Govern wrote:
> Hi all,
>
> I'm using Tesseract with the hOCR output format. When I invoke Tesseract
> on an image, the results are returned in hOCR format with a UTF-8 character
> encoding. However, if I then convert the same image to TIFF format from
> PN
hi Angelica...
this header file will be used to compile tesserect in linux.
I compiled it on vs2010 but its done without problem.
On Sun, Jul 29, 2012 at 7:22 PM, Angélica Mascaro wrote:
> Hi, Ankur,
> Ok, i've seet the paths of includes (just now I noticed the
> "tesseract/api" project was a .
Hi all,
I'm using Tesseract with the hOCR output format. When I invoke Tesseract on
an image, the results are returned in hOCR format with a UTF-8 character
encoding. However, if I then convert the same image to TIFF format from
PNG/JPEG etc. the character encoding of the hOCR output is Latin1
On Sat, Jul 28, 2012 at 8:54 PM, Angélica Mascaro wrote:
> On project properties -> Linker -> Additional Dependencies i`ve put all the
> libs that tesseract generated (ccmain.lib ccstruct.lib ccutil.lib
> classify.lib cube.lib cutil.lib dict.lib image.lib libtesseract_tessopt.lib
> libtesseract_tr
first you need to follow steps at
http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3 for your
language and fonts just go through it, they seems hard at first but they
are really easy.
On Monday, 30 July 2012 12:47:49 UTC+5:30, Txoov Chij Her wrote:
>
> if i have a new language and n
if i have a new language and new font How is covert this font for support
OCR program.
please tell me too.
On Monday, July 30, 2012 1:35:49 PM UTC+7, shah dipen wrote:
> Hello everyone,
> I was able to complete OCR on single file, but when i
> used more than 32 font fil
Hello everyone,
I was able to complete OCR on single file, but when i
used more than 32 font files and combined generated data the new
eng.traineddata file was not able to perform even 1% more better than
original trained data file. I have performed all steps on Tesseract-3
9 matches
Mail list logo