On Mar 5, 1:12 am, rthomas remi.tho...@gmail.com wrote:
Hi,
I think this project is dead.
- Many people come here to get an Open Source OCR, do not contribute
and go.
- The few contributors can't commit the code because the project is
locked.
- Tesseract needs a complete rewriting.
-
On Apr 12, 10:31 pm, MARTIN Pierre hicksc...@gmail.com wrote:
i've been making some very small changes to the Tesseract project. Most of
them are only related to the way things are organized (Visual studio project,
etc).
However, i plan to be attacking the big deal very soon... In this
On Feb 20, 9:02 pm, Jon Andersen jande...@gmail.com wrote:
My project athttp://RecordAGrave.comis about recording headstones from
graves and posting the text and images on the Net so that people can
research their family history. I would appreciate some advice on how to
pre-process these
Thanks for reporting on your findings.
On Thursday, May 17, 2012 5:50:36 AM UTC-4, Galt wrote:
I have used this simple training text and the output is highly
accurate.
...
I ran this on a 75 page book
and virtually every page was perfect,
even though the scans themselves were not.
On Wednesday, October 10, 2012 12:59:49 PM UTC-4, Nick White wrote:
So I've been tossing an idea around in my head for a while now, and
I think it deserves discussion.
As I understand it, the box/tif steps basically reduce varying
character shapes to basic simplifications, for each font,
On Friday, October 12, 2012 8:50:25 AM UTC-4, Nick White wrote:
Hi Tom, thanks for your thoughts.
A key reason for not using scans when training is when the character
set is quite large, so it would take many pages of real scans to get
a few samples of each. Plus I found the process of
On Wednesday, October 24, 2012 11:37:18 PM UTC-4, Phlip wrote:
Tesseractors:
We are using Tesseract for an outside-of-the-box situation - not
scanning neatly typed documents.
Our situation is a fuzzy, low-contrast picture. But - even when I use
many image enhancements, such as leveling
Embossed cards are designed to be printed. Have you considered taking an
impression and scanning the impression? Or just scanning (magnetically)
the magnetic strip on the card?
There have been other discussions of training Tesseract for OCR-A (and I
think OCR-B). Farrington 7B is another in
Nick, thanks for going through and reviewing the comments for useful info.
On Thursday, November 29, 2012 1:47:20 PM UTC-5, Nick White wrote:
On Thu, Nov 29, 2012 at 07:16:05PM +0100, zdenko podobny wrote:
What about to block posting comments on wiki (after its deletion)?
It is more
On Sunday, February 24, 2013 11:53:52 AM UTC-5, zdenop wrote:
On Sun, Feb 24, 2013 at 12:20 AM, Nick White
nick@durham.ac.ukjavascript:
wrote:
On Fri, Feb 22, 2013 at 03:20:49PM +, Nick White wrote:
On Sun, Jun 03, 2012 at 10:27:23PM +0100, zdenko podobny wrote:
it looks like
On Monday, April 29, 2013 8:39:57 PM UTC-4, Michael Sander wrote:
Yes, I'm doing something similar in python. Do you know of a list of a
ligatures so I can convert them to ascii? I know fi and fl are the most
popular, but there are probably many more.
The list of Unicode ligatures is here:
On Friday, May 10, 2013 7:45:13 AM UTC-4, Renard Wellnitz wrote:
Regarding publishing the code. I want to do this. Iam not yet sure if i
want to use the GPL or Apache 2.0.
Also since i modified tesseract and leptonica sources a bit and have
several submodules i need to write a proper
On Thursday, May 16, 2013 10:52:51 PM UTC-4, Mike Masinick wrote:
So, I have several hundred thousand scans of sports cards that look
similar to the attached. I want to scan the text at the top of the page
and extract at least the 8 digit number. Ideally more of the text as well,
but the
On Tue, Jul 2, 2013 at 12:50 PM, zdenko podobny zde...@gmail.com wrote:
Hi,
I thought about increasing version number in svn to be able to distinguish
devel version from released version, but I was not sure about version
number. Up to this time it was Ray Smith who changed it after
On Friday, July 12, 2013 8:24:12 AM UTC-4, sdk wrote:
I have registered for Prima Tools. However, since I am not affiliated to
any institution, I am not sure whether they will approve registration. I
haven't heard back yet.
That's an interesting racket that the University of Salford has
On Wednesday, July 31, 2013 3:20:43 AM UTC-4, billsmi...@gmail.com wrote:
I am doing some research on tesseract-ocr, the most famous OCR software in
the world. However, I am stuck in some code(for example, the feature
extraction algorithms in the function ExtractIntFeat). I took long time
On Wednesday, September 11, 2013 11:14:22 AM UTC-4, ch...@sc3.net wrote:
What I'm struggling with, is that the position information that Tesseract
gives me doesn't seem to allow me to position the characters for display or
for creating a stream to insert back into my PDF file.
If I want
On Tuesday, November 5, 2013 12:38:57 PM UTC-5, rkomar wrote:
So, I would say that totally different results with code
compiled with different options or different compilers
is a sign of bad code. I don't think it should just be
shrugged off as unavoidable.
I agree with Rob here.
The discussion of vectorization vs parallelization is interesting, but
doesn't have anything to do with the question that was asked. Perhaps a
new thread is in order?
Niral - An order of magnitude difference in performance is unlikely to be
attributable to 32-bit vs 64-bit. You left out all
On Tuesday, November 12, 2013 12:18:59 PM UTC-5, Vishalkpp wrote:
Sorry for missing the attachment. Please find attached the image. The
numbers around the 2D diagram are the annotations I would like to address.
It's hard to extrapolate from a single example, but this feels more like a
It's great to have another open source tool in the toolbag. (It's GPL v3
BTW for those who don't appreciate the irony of distributing a GPL license
in proprietary Microsoft Word format.)
I'll echo what the others have said about openness and freedom, or lack
thereof. Not only is the Aletheia
On Friday, December 6, 2013 5:46:39 AM UTC-5, Casper Kent wrote:
I am trying to use Tesseract to process coupon images. The recognition
result is great. However the processing speed varies hugely between my
Windows test environment and the Sparc/Solaris production environment. A
batch
In watching Bryan Tarpley's Franken+ presentation (
http://emop.tamu.edu/node/54) it's pretty obvious from the example that
there are (at least) two clusters of glyphs for the letter 'o': a tall
skinny glyph and a round glyph.
p.s. You probably want to add a .gitignore file so that you aren't
committing binaries to the repository. Also, it seems like Franken+, as a
standalone tool, really could use its own repo. They're lightweight and
free and that would give you a separate bug tracker, wiki, etc, as well as
the
On Friday, December 13, 2013 11:25:42 AM UTC-5, Nick White wrote:
I've drafted such a page, and I'd be keen to get feedback on it. Is
it clear? Is it a good idea? I haven't filled out all of the Image
processing sections yet, but (presuming people don't hate the idea
in general) I will do
On Wednesday, March 12, 2014 7:57:38 AM UTC-4, John Green wrote:
*What I'm doing: *As part of a longer pipeline, at one step I am
reasoning over very small but highly characteristic strings like drug
dosage, 60 mg. Edit distance (Levenshtein or a variation) and n-grams,
even unigrams,
p.s.
On Saturday, June 28, 2014 12:39:21 AM UTC-4, scott...@gmail.com wrote:
3) Attempted to increase the strength of dictionary matches as discussed
on the FAQ (
https://code.google.com/p/tesseract-ocr/wiki/FAQ#How_to_increase_the_trust_in/strength_of_the_dictionary?),
both via API
On Wed, Aug 27, 2014 at 4:51 PM, zdenko podobny zde...@gmail.com wrote:
there is no git command for it (well maybe we could track down the
revision number and tag it, but...)
Isn't tagging releases good software engineering practice regardless of all
this other discussion?
Looks to me like
On Tuesday, September 16, 2014 9:11:35 PM UTC-4, Albrecht Hilker wrote:
Tesseract requires test to be 300 DPI.
Resize your images with factor 3 and try again.
The built-in thresholder should be enough for your samples. (No Gimp
required)
Or better yet, rescan at the higher resolution so
On Thursday, October 16, 2014 11:04:58 AM UTC-4, Zunair Fayaz wrote:
need best practice to OCR on documents with + sign that helps align the
documents.
Those are typically referred to as registration marks
Any known practice?
I would have thought they'd be pretty easy to detect using
Has anyone tackled training for the IPA since this initial query?
I'm considering using Tesseract to OCR the first edition of the Oxford
English Dictionary (as input to a crowdsourced proofing process) and trying
to decide whether it's worth training it to recognized the pronunciations.
I'm
On Thursday, April 23, 2015 at 9:03:02 AM UTC-4, Stathis L. wrote:
Point is it's a 800-900 page tiff that I want to create and these
solutions don't really work, as making a list is not a question and VietOCR
runs out of memory. Anything else?
Are you replying to Quan or Zdenko? It helps
On Sunday, April 19, 2015 at 8:23:20 AM UTC-4, James Worldprogram wrote:
During processing of tiff files, which are having *600 - 700 pages* from
Tesseract OCR engine with hocr option, we monitored that files are taking
around *40 - 50 minutes*.
So, 14-15 pages/minute or 4-5 seconds/page.
On Tuesday, April 28, 2015 at 2:17:40 AM UTC-4, Lenilson Castro wrote:
I'm beginning an academic work about OCR technology on mobile devices and
i was looking for some fonts for a bibliographic revision about that.
Does anybody know where do i can find good referênces about OCRs history
all non-ascii characters) in the digital 2nd edition OED,
which is probably a very good starting point for generating training
texts.
Nick
On Thu, Apr 23, 2015 at 01:32:12PM -0700, Tom Morris wrote:
Has anyone tackled training for the IPA since this initial query?
I'm considering using
applicable.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2732221/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3265968/pdf/nihms241943.pdf
http://cda.ornl.gov/publications_2011/Publication%2028596_S.%20Xu.pdf
On Wednesday, May 6, 2015 at 2:41:13 PM UTC-4, Tom Morris wrote:
You might consider looking
You might consider looking at some of the papers on text detection in
natural images and using the techniques from the later stages in the
pipeline. These are similar what Dmitri outlined, but reviewing what
others have done might give you ideas on additional ways to filter and
group
On Thursday, May 14, 2015 at 12:54:26 AM UTC-4, smwikipedia smwikipedia
wrote:
I am trying to add some pre-processing post-processing for my scenario.
I read some papers and it seems the steps could be:
- noise attenuation
- orientation correction
- binarization
-
On Friday, March 13, 2015 at 8:29:13 AM UTC-4, Jan wrote:
I noticed that when I use tesseract to create a searchable pdf (I use
pdfsandwich fot this), some characters are not displayed and are replaced
by blank spaces instead. If I, however, ocr the same file with tesseract
only in order
While some people think OCRopus has a more sophisticated page layout
analysis, it's not true that Tesseract does no layout analysis.
The man page describes the different types of segmentation that it does:
https://tesseract-ocr.googlecode.com/git/doc/tesseract.1.html
Tom
On Monday, May 18,
A few notes on equation detection:
- In the 3.02 announcement
https://groups.google.com/forum/#!searchin/tesseract-ocr/equation/tesseract-ocr/EXyGqT9osrw/y1xMAujqZy4J,
the feature is listed as *experimental *equation detector (emphasis added)
- there's no documentation on what's actually in
p.s. For anyone interested in the topic, here's a masters thesis on math
equation detection and segmentation (and the work was done using Tesseract):
https://vtechworks.lib.vt.edu/handle/10919/46724
https://vtechworks.lib.vt.edu/bitstream/handle/10919/46724/Bruce_JR_T_2014.pdf?sequence=1
--
On Monday, June 22, 2015 at 7:56:51 AM UTC-4, Gunasekaran Velu wrote:
I have attached the image as well as Tesseract OCR result for attached
image screen shot. the below OCR some words are missing from OCR how can i
improve the image quality to detect the missing words.
The attached
In addition to Art's training data, you might also want to test the IKU
language data for Tesseract 3.04 that Google released a few hours ago:
https://github.com/tesseract-ocr/tessdata/blob/master/iku.traineddata
It was generated from the source language data here:
On Thursday, June 25, 2015 at 4:23:32 PM UTC-4, subo...@gmail.com wrote:
I've been playing around with the OpenCL option for Tesseract. It appears
on very first runs, it profiles the system and tries to determine which is
the fastest compute device and uses that device for future runs. Is
That's cool that there's already a starting point for the IKU language
training.
To help you understand the various files in Art's repo and the process used
to create them, here's the wiki page which describes the training
process:
It's difficult to tell what the problem is without any example images. Are
you saying that there are ligatures in the image and you don't want them
recognized as such or that there are not ligatures, but the characters are
touching due to low resolution or poor quality scan or over inking or
On Tuesday, May 26, 2015 at 12:17:32 AM UTC-7, Stathis L. wrote:
Does anybody know of a software editing the hOCR output of tesseract
(besides a simple text editor) ?
This Firefox add-on might do what you want:
https://addons.mozilla.org/en-us/firefox/addon/hocr-editor/
Tom
--
You
On Monday, June 29, 2015 at 3:57:08 AM UTC-4, Jeff Breidenbach wrote:
Not available currently, and pretty major effort required to make it
happen,
both in Leptonica and Tesseract's PDF output module. No plans to work
on this. For other formats we try hard to not re-encode during PDF
When I look at the word frequencies in eng.cube.word-freq, they look more
like what I would expect from analyzing a web corpus rather than a corpus
of printed materials (of any era).
The list starts off okay:
#1 the 13675
#2 of 15222
#3 and 15473
#4 to 15694
#5 a 17149
but then we have:
#29
Be sure to check https://github.com/tesseract-ocr/langdata before assuming
that the language that you need isn't supported. Dozens of new languages
were added a couple of weeks ago.
Tom
On Monday, July 6, 2015 at 9:09:06 AM UTC-4, Brennan Nunamaker wrote:
For clarification: With text, I
On Friday, July 31, 2015 at 3:26:16 AM UTC-4, Merv wrote:
The task I am facing is extraction of rotated text (any angle) from a
scanned image. Kindly find the link : http://1drv.ms/1OS8elW which has
the sample PDF document that I need to OCR. The sample contains a blue
sticker on it and I
On Tuesday, July 14, 2015 at 2:47:40 AM UTC-4, James Owers wrote:
You should consider also using the PAGE format. You can use this tool for
conversion: http://www.primaresearch.org/tools/TesseractOCRToPAGE
Most PAGE format tools aren't available as open source and use a custom
license
Your goals don't sound unreasonable, but I'd suggest using an approach that
focuses on pre and post processing before diving in and hacking on
tesseract itself. That will allow you to easily continue to track
improvements in base tesseract without having to worry about re-integrating
your
On Monday, July 13, 2015 at 1:23:10 AM UTC-4, Elan wrote:
I am fairly new to tesseract, I have done some playing around with
training new fonts, and loading config files etc. I have an issue with the
images I am trying to OCR.
In many cases, there is a dotted horizontal line about 5-10
On Thursday, July 16, 2015 at 1:07:11 AM UTC-4, Robertas Kaunas wrote:
Hello, everybody. I am trying to read two columns between whose are big
gap. So which setting should I modify to let tesseract reading word from
second column as in one line with word in first column?
Look at the page
Thanks for the analysis and feedback, Jeff.
Unfortunately, I don't know much about QPDF (and SourceForge's storage
problems are preventing me from learning any more), but doing #3 externally
using a tool like QPDF, perhaps in conjunction with doing #1 in Tesseract
itself, sound like reasonable
On Tuesday, October 27, 2015 at 4:49:11 PM UTC-4, Daniel Kraft wrote:
>
>
> On 2015-10-27 16:10, Allistair wrote:
>
> > I was able to get it reading everything by cropping it to the same
> > amount as Working but then rotating it anti clockwise by just a few
> > degrees - I tried this because
On Wednesday, October 28, 2015 at 4:46:14 AM UTC-4, jinh...@google.com
wrote:
>
>
> First, I have very little knowledge about ocr/tesseract.
>
...
Please help.
>
If only you worked for Google, you could probably get help directly from
the Google software engineers.
Oh, wait. You DO work
On Thursday, November 12, 2015 at 8:51:06 AM UTC-5, sriranga(82yrsold)
wrote:
>
>
> Alternatively, kindly forward copy of your commandlines used for
> *grctraining*
> - since http://ancientgreekocr.org/grctraining.git
>
What happens when you use the same command line flags on both machines? You
show no flags being used on the Mac. What version of Tesseract are you
using? Did you build the binaries yourself (what compilers?) or did you
get them from somewhere (where?)? What image are you trying to recognize?
On Sunday, October 18, 2015 at 2:00:39 AM UTC-4, Mayu Shukla wrote:
>
>
> But i am not able to understand where to look for and which config file in
> config folder,since there are many. which one tesseract calls and how it
> works?
>
The config file is the one you specify on the command line.
On Thursday, September 10, 2015 at 2:31:03 AM UTC-4, hmmwhat...@gmail.com
wrote:
>
> On Friday, September 4, 2015 at 9:38:20 PM UTC-7, Jeff Breidenbach wrote:
>>
>> But I would like to see an example PDF - one of the simpler ones - just
>> to see how the vector graphics were done. Please do not
Are you doing any pre-processing besides cropping? If those images are
representative and the colors are constant, I'd replace the orange
background with black and then invert the image to give black digits with
no border on a white background. Also use the page segmentation mode for a
July 2015 22:55:25 UTC+5:30, Tom Morris wrote:
>>
>> I'd suggest looking at OpenCV. It looks more like a computer vision task
>> than an OCR task. Some of the specific issues like dials not fully aligned
>> in the window are things the OCR systems aren't designed to
On Wednesday, September 9, 2015 at 1:30:38 PM UTC-4, Max Heiber wrote:
>
> Here's an example where the Chinese characters are very large and clear,
> but Tesseract gets the wrong result. Could you advise on what image
> processing could help Tesseract's accuracy?
>
What have you tried so far?
On Friday, September 11, 2015 at 3:29:45 AM UTC-4,
worldwar...@googlemail.com wrote:
>
>
> Thank you for the fix in GitHub!
>
You're welcome! Hopefully the other hints will help you report/fix the
next bug you find.
Tom
--
You received this message because you are subscribed to the Google
I don't know where all this complexity came from. PDF rasterizers have
existed since the format was invented. GhostScript is one popular open
source alternative. It could either be used directly or through a tool
that embeds it such as ImageMagick.
Tools like Apache PDFBox can be used to add
Perhaps this is just a misunderstanding or bad documentation. The
--print-parameters dump shows the input parameters, and the user_words_file
/ user_patterns_file parameters, if they're not set on the command line,
will always be empty.
The actual file name that gets loaded gets computed on
On Wednesday, September 9, 2015 at 9:09:39 AM UTC-4,
worldwar...@googlemail.com wrote:
>
> I can confirm this bug! I tried to compile compile tesseract-3.04.00 on a
> QNAP TS119 NAS with a Marvell Kirkwood with many tool from
> optware/cs08q1armel/cross/unstable/ . I get exactly the same error
On Monday, September 21, 2015 at 9:29:39 AM UTC-4, Stef wrote:
>
> I'm trying to use user wordlists with the bazaar config but it seems to
> have no effect on the OCR result in my case. Therefore I printed the
> current parameters to verify whether the user-words list is used. This
> confirmed
You might try pytesseract instead to see if you have better luck:
https://pypi.python.org/pypi/pytesseract
Tom
On Friday, October 2, 2015 at 1:57:56 PM UTC-4, Chang Alden wrote:
>
> I have python-tesseract-0.9.1-py2.7 installed but I am getting this error.
>
> Traceback (most recent call last):
I think Dmitri's suggest to start simple is a good one, but, if you need
it, don't forget that you've got a lot of other information that can be
leveraged to help. The notes all have a fixed aspect ratio (and size?).
They've got a relatively standard layout. The denomination is encoded
If you think those images are visually identical you should visit your
optician. :-)
The ImageMagick version is much blurrier, so I'd guess that the high
frequency noise from the pixelated OpenCV image is making Tesseract
unhappy. If you want to continue using OpenCV, try applying a Gaussian
cessor* being external program is - one should
>> have update the post-proc.text everytime for each ocred
>> I am puzzled why unicharmabigs does not work as internal program
>> correctly - when the post processor program works fine?
>> With regards,
>> sriranga(83yrs)
>>
Hi Sriranga. I haven't used the training tools, but since no one else has
answered, I'll give it my best attempt. Shree might have better insights.
First, a question of clarification. Are you having problems with the file
or are you just trying to determine whether it is working properly or
On Wednesday, November 25, 2015 at 12:54:25 PM UTC-5, Nasim Ali wrote:
>
> Nguyen (program creator) says the problem is with java, so I've decided to
> use qtboxcreator to create boxes and the subsequent work is handled by
> jTessBoxEditor.
>
To expand on this, it doesn't look like Java
I don't see what the IDE has to do with anything.
https://github.com/tesseract-ocr/tesseract/wiki/APIExample
On Wednesday, December 9, 2015 at 10:22:51 AM UTC-5, Matthias Schneider
wrote:
>
> Hi all,
>
> I'm currently trying to get Tesseract working with Embarcadero RAD Studio
> 10 C++Builder.
xts
that you're trying to correct.
Tom
> your suggested sentence
> "Novv is the time to go dovvn" also corrected. Please note I regenerated
> eng.traineddata in ubuntu 15.10.
> With regards, sriranga(83ys)
>
> On Wed, Dec 9, 2015 at 12:04 AM, Tom Morris <t
On Tuesday, December 22, 2015 at 2:04:26 AM UTC-5, Utkarsh Sinha wrote:
>
> I'm trying to find out why Tesseract is rejecting certain blobs from the
> image here. The text "nestle" and "nesquik" have overlapping baselines. I
> suspect the overlap might be causing it to stop recognizing anything
Although it wasn't tagged with a version, you'll also want the relevant
language data files from https://github.com/tesseract-ocr/langdata.
Note that this is a source release. There are binary releases available
from some of the distribution packagers such as Debian:
On Friday, June 3, 2016 at 10:09:04 AM UTC-4, Cristian wrote:
>
>
> I'm new on tesseract. I'm working on application that has to recognize the
> expiration date of some products like foods. The input will be an image
> (very good resolution) with only the date on it.
> Before putting my hand on
On Thursday, June 2, 2016 at 6:40:55 AM UTC-4, Вадим Авлочинский wrote:
>
> Can I use the Tesseract OCR lib to develop my applications and use it for
> commercial purposes?
>
The license is pretty liberal. Your lawyers can find it here:
On Wednesday, June 1, 2016 at 9:14:21 AM UTC-4, Rob Shanks wrote:
>
> I am trying to use Tesseract to recognise a 6 digit number from scanned
> documents. Because they are scanned the numbers can be faded but I know
> that they are 6 digit. The scanned documents have been destroyed long ago
>
On Thursday, June 2, 2016 at 2:38:32 AM UTC-4, Guilherme Galdino Siqueira
wrote:
>
>
> I have an issue with bad quality of recognition of text with high delay of
> finishing the process. I've tried to binarize the source image, reduce its
> size, change environment light to take the picture,
What version of Tesseract? Source or binary? If source, compiled with what
compiler & compilation options? What operating system? What processor
architecture? What's the stack trace for the crash?
On Friday, June 10, 2016 at 6:00:19 AM UTC-4, Matthias Schneider wrote:
>
> Hi,
>
> I'm trying to
Bojidar - Google isn't going to provide support for 3rd party use of a
project that they've open sourced.
Sumant - There are a number of developers on this list who've used
Tesseract in their own applications or created custom OCR solutions based
on it. One or more of them may be willing to
On Monday, June 13, 2016 at 10:22:32 AM UTC-4, Matthias Schneider wrote:
>
> I'm using latest dev version 3.05.00dev and I used peirick/leptonica (
> https://github.com/peirick/leptonica) to build libtesseract.dll and
> liblept.dll with Visual Studio 2015.
> However, the resulting DLLs I'm using
Rather than rolling your own, I'd suggest looking at the ScrollView app to
see if it has the information that you need:
https://github.com/tesseract-ocr/tesseract/wiki/ViewerDebugging
Tom
On Wednesday, June 22, 2016 at 7:16:35 PM UTC-4, Quan Nguyen wrote:
>
> You'll need to programmatically
I'd also get rid of the blue and make the text black on white. Presumably
the colors are constant, so this should be an easy pre-processing step.
Tom
On Friday, June 24, 2016 at 4:01:42 AM UTC-4, Stef wrote:
>
>
> You could try and scale up the image before OCR. See section "Scale text
> up"
On Wednesday, January 13, 2016 at 3:52:48 PM UTC-5, John Scancella wrote:
>
> I tried searching but couldn't find which versions of PDF/A (if any)
> tesseract supports. Specifically I have a requirement for PDF/A-2a
> generation, but I couldn't find anywhere if tesseract can write PDF/A-2a
>
Hi Stef. Is that info hiding in the wiki somewhere? If not, do you think
you could find a place to add it?
Tom
On Saturday, June 25, 2016 at 4:41:23 PM UTC-4, Stef wrote:
>
> The two numbers are the slope (1st number) and constant term (2nd number)
> of a linear equation describing the
waiting for
> server... waiting for server... it's like it goes in infinity loop.
> can you suggest solution for it?
>
>
> On Friday, June 24, 2016 at 10:35:27 PM UTC+5:30, Tom Morris wrote:
>>
>> Rather than rolling your own, I'd suggest looking at the ScrollView app
On Saturday, June 25, 2016 at 2:45:51 PM UTC-4, denny.m...@gmail.com wrote:
>
>
> could anyone kindly explain why the text "YTOVWG" in ocr3.jpg is not
> recognized
> but can be recognized in ocr4.jpg ?
>
As a guess, I'd say because the mixed fonts on a single line and large gap
before the
On Sunday, June 26, 2016 at 7:22:20 AM UTC-4, Arunabh Ghosh wrote:
>
> I have uploaded an image which has the character 'e' written in it. I had
> preprocessed the original image using opencv2 but still couldn't get
> tesseract to recognize the character.
>
> Any suggestions on how to get
On Thursday, January 28, 2016 at 4:24:26 AM UTC-5, Gunasekaran Velu wrote:
I am using following tesseract command to do the HOCR file for bmp image
>
> >tesseract.exe "test.bmp" Test -l eng -psm 6 hocr
>
> Marked area in the attached image does not come in to Test.hocr.html file.
> except
I just stumbled across https://github.com/jbarlow83/OCRmyPDF which claims
to use Tesseract and provide PDF/A support.
Tom
--
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send
On Saturday, January 30, 2016 at 5:46:31 PM UTC-5, Ilya Mezhirov wrote:
>
>
> I've written a binarizer that does Sauvola a lot faster than Leptonica. It
> also can double the resolution.
> Works strictly with JPEGs and outputs G4-compressed TIFFs.
>
> Check it out:
On Tuesday, February 23, 2016 at 5:40:33 PM UTC-5, Devon Yoo wrote:
>
>
> Is there a way to give TesseractEngine a hint of expected text format? For
> example, can I set a format like 00XXX00 XX-000 where 0 represents number
> and X represents alphabet?
>
See the answer to this question:
, syr, tgk, tir,
uig, urd, uzb, uzb_cyrl, yid
There are a total of 107 languages supported now.
On Thursday, February 25, 2016 at 3:41:00 PM UTC-5, Tom Morris wrote:
>
> On Thursday, February 25, 2016 at 3:32:10 AM UTC-5, 기옥주 wrote:
>>
>> I wonder what is improved ver. 3.04. more
1 - 100 of 227 matches
Mail list logo
