Tess3.01 hocr output not working with pdfbeads

I should begin by saying that I am grateful and happy to have a very nice searchable pdf of an old book thanks to Tess. I found this on the web: https://github.com/steelThread/mimeograph/commit/b29af3338e8f15b22392b4e313c8688d9950e13b pdfbeads currently doesn't work with hOCR output generated

Re: List of Config Paramenters

IMO the best way is to search for '_MEMBER' (or ' _VAR_H ' in "*.h" ???) as suggested (indirectly :-) ) in Visual Studio 2008 Developer Notes for Tesseract-OCR -> Handy free tools [1]. [1] http://tesseract-ocr.googlecode.com/svn/trunk/vs2008/doc/tools.html#id2 -- Zdenko On Mon, May 21, 2012 at

Re: Using Cube Engine And Right to Left Language

I dont know if or how you can use cuve for training. But if you want to use cube programmatically have a look at api/baseapi.h For the wrong order problem I would suggest to unpack the ara.trainddata file with combine_tessdata -u tessdata/ara.traineddata /outputfolder/ara. there should be a ara.c

Re: List of Config Paramenters

I found a partial list in tesseractclass.h . On Friday, 17 June 2011 11:54:05 UTC+2, Derek wrote: > > Check here: > http://code.google.com/p/

Re: Using Cube Engine And Right to Left Language

Stane, Thank you for replay. Can you explain how to use cube in training by command line or code, it will help alot. When i train my own tiff's file and use this trained file it give me the word in backword like ((word)) = ((drow)) with Arabic word. Thanks again. -- You received this message b

Re: List of Config Paramenters

I found a partial list in tesseractclass.h . On Friday, 17 June 2011 11:54:05 UTC+2, Derek wrote: > > Check here: > http://code.google.com/p/

Word White and Blacklisting with RegEx

I'm interested in implementing a word white (or black) list described by a regular expression. In my application I generally only need to detect single words with a predefined structure. Character whitelists definitely help, but I can't help but wonder if a word whitelist would be even better.

Re: Using Cube Engine And Right to Left Language

As far as i know there are no tool to train something for the cube engine. you just can use the ara.cube.* files in tessdata, to use the cube engine for the Arabic language. -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group,

Re: what are the protos (related to mftraining); how to set the max number

Maybe you can change the MAX_NUM_INT_FEATURES in baseapi.h currently its set to 512. Or as you suggested you can train two to separated traineddata files and run tesseract 3.02 with it. -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to

Re: Tess3.01 not recognizing my curly double quotes.

Hi Galt, I've been suffering a very similar problem with some of the text I'm training, which has several diacritics above and below glyphs. It isn't infrequent to find quite a few lines of garbage which are some of the diacritics taking a line, which then causes the following and preceding lines