OCR: Omnipage and Finereader

Scanning and OCR is on my mind at the moment.  A new version of Abbyy Finereader — version 11 — is out.  Since I have some 750 pages of Ibn Abi Usaibia to do, any improvement in accuracy is welcome, however slight. 

Originally I did my OCR using Omnipage.  It is many years since I was led (by Susan Rhoads of Elfinspell.com) to look at Finereader 5.  This was immensely superior, and I have never used any other product since.  But I see that Omnipage 18 is now out.  Stirred by a bit of curiosity, I’ve been wondering what this would be like.

Finereader is not without its faults.  Foremost among them, for what I want to do, is that it cannot make a PDF searchable without making the PDF much, much larger, messing with the images, and so forth.  This is so bad, in fact, that I use Adobe Acrobat Pro 9 for that task, despite the much inferior OCR.

Omnipage seems to be aware of the issue, and a look at their site suggests that they realise that a lot of this activity goes on.

I decided, therefore, to buy both and see what they’re like.  I will let you know!

But … software vendors are thieves and robbers!  If you go to the Abbyy site, the cost of a downloaded upgrade to Finereader Pro 11 is “€ 89 / £ 65 (download)”.  The full version is “€ 129 / £ 99″ — and if you want just the download, it’s exactly the same price, despite the fact that it costs them less!  But go to Amazon.co.uk, the complete boxed set is just £63.16 — less than the upgrade.  Needless to say, that’s what I ordered.

Omnipage are no better.  Go to the Nuance site, and Omnipage 18 (standard version) is £79.99, whether download or boxed.  Again they swindle the download users.   But go to Amazon.co.uk, and the complete boxed set is £46.90!

I didn’t buy the Omnipage Pro version, but stuck with the standard one.  It’s a lot more money, and I wasn’t convinced that I’d use the extra features — especially since I don’t know if the OCR is any good at all.  Here a trial version would have helped — Finereader make trial versions available online.  This is smart marketing on their part, because magazine reviews of such a specialised area of software are invariably useless.

My current interest in Russian texts of Methodius means that I was interested to see that Omnipage offer a separate Russian version.  Finereader used to have a specific “Cyrillic option” version — indeed I owned a copy, back in the FR5 days — but this seems to have vanished from their product list.  Kudos to Finereader: Russian support is included in the main product!  I only wish their obscure “fraktur” recognition module was included too!  This recognises old “Gothic”-style typefaces, and some of us would find it handy.  But I could only find it in their SDK for Linux.  And it doesn’t seem that you can even buy the latter off-the-shelf.

2 Responses to “OCR: Omnipage and Finereader”


  1. David Crocker

    I was wondering how you made out with OmniPage and fraktur. Do you have any further insight? Regards, David

  2. Roger Pearse

    I think Omnipage wasn’t bad (Fraktur I have never been able to scan). But it didn’t seem any better than Finereader, so I have stuck with the latter.



css.php