You are here

CuneiForm OCR

4 posts / 0 new
Last post
rbon
Offline
Last seen: 6 days 2 hours ago
Joined: 2011-04-30 13:26
CuneiForm OCR

CuneiForm OpenOCR
this program is the best OCR freeware/open source, it's very good but not paragonable with FineReader or OmniPage tools.
This tool is multilanguage: more than 20 languages as English, Russian, German, French, Spanish, Italian, Swedish, Ukrainian and others.
It's came with a dictionary (for each language): this dictionary increase with new word contained inside document recognized.
- Author: Cognitive Technologies
- Web site: http://cognitiveforms.ru/products/cuneiform/
- international download page point to OpenOCR.org http://en.openocr.org/
- release: 12.0.0.58851
- license: BSD
- Softpedia: http://www.softpedia.com/get/Office-tools/Other-Office-Tools/CuneiForm.s...
- Wikipedia: http://en.wikipedia.org/wiki/CuneiForm_%28software%29
- tips: download the english version because it's multilanguage
- direct download link: http://cognitiveforms.ru/downloads/setup_openocr_cuneiform_en.exe

solanus
solanus's picture
Offline
Last seen: 9 years 5 months ago
Joined: 2006-01-21 19:12
Not as good as TOPOCR

Just did a quick test comparing to TopOCR, and there's no contest.
Using a screenshot from a random webpage, TopOCR had no problem recognizing the text - a few misspellings, but not bad, and displayed it for review and correction before saving.
Cuneiform: when opening the file, it required to convert it to another resolution, and then when doing the OCR, it complained that MS Word was not installed (there's no option to use an alternate program, such as LO), then it wouldn't preview, it just saved the file as an rtf. When I opened it, it had awful formatting, and the text was all gobbledegook - not a single recognizable word, not even close.

Can you get it to work? I tried different file formats, different resolutions, all fail.

I made this half-pony, half-monkey monster to please you.

rbon
Offline
Last seen: 6 days 2 hours ago
Joined: 2011-04-30 13:26
Not as good as TOPOCR

Hi,
my problem was: a pdf multipage abtained from a photocopier/scanner.
This pdf contain a multicolumn table.
1. I have print the pdf and scanned again in TIF format
2. I have used:
a. TopOCR 3.1 portable (after use I have moved the file "topocr.ini" from
C:\Windows to C:\TopOCR_Portable)
b. CuneiForm OCR
3. TopOCR do not mantain a structure of the table: it create only a single column
4. CuneiForm create a Word file with the table (not perfect)
5. I have tried to import this table in Excel: this not work properly because the
Word file contain some (many) "tab"
6. I have copied this table from Word to PsPad and changed "tab" in cr+lf
7. finally I have imported the ascii file in Excel.

Probably TopOCR work well with pages with text, but if You have to import tables CuneiForm OCR work.

dbojan
Offline
Last seen: 8 years 11 months ago
Joined: 2009-03-05 06:54
CuneiForm portable

CuneiForm is free, TopOcr is not.
Anyway to make CuneiForm portable, download it.
Make text file in the notepad and rename it to register.bat.

regsvr32 "%~dp0APuma.dll"

With just one line.
Put it in the main dir, the one where is APuma.dll.
When you want to copy it another computer, copy whole dir. Right click on register.bat, select "Run as administrator". There you go.
To unregister, create file unregister.bat

regsvr32 "%~dp0APuma.dll"

which will unregister dll.

There is also puma.net dll, which has sample application (Puma.Net.Sample.exe) in it.
http://pumanet.codeplex.com/

I have uploaded a version here:
https://onedrive.live.com/?cid=D3221CFE195F835B&id=d3221cfe195f835b!434

--
http://dbojan.tk

Log in or register to post comments