I stopped using Omnipage years ago, thanks to a tip from Susan Rhoads of Elfinspell.com that there was a new kid on the block. This was Abbyy Finereader, and I bought a copy of version 5, with the Cyrillic option which was extra. They were a Russian company. The rumour was that the software was developed by the KGB to help pirate western research.
Anyway it did a fine job of the bit of Russian text I wanted to scan (so I could run it through a machine translator and get an idea of what it said!). It did a much better job of English than Omnipage did. And I was hooked.
I’ve used it ever since. The first need for good scanning is a good scanner. If you buy a $50 scanner you don’t get good character recognition. If you buy a $300 you will get half the errors, just from that alone. Software is actually less important than scanner quality.
The years passed, and FR5 became FR6 became FR7 became FR8. I upgraded each time. Improvements there were, but they were incremental. I upgraded to FR9, and this was a big step forward in OCR quality; but a big step back in user-interface.
Today I found by accident that FR 10 is out. They’ve added features that make it easier to work with PDF’s. (Who said “Google books”?) I tend to download a PDF, open it with Adobe Acrobat and do character recognition. Then I save it, and thereby make the PDF searchable. But it wasn’t easily possible to do this in FR9, because it degraded automatically the image quality (which you didn’t want) and got all tetchy if there were un-recognised pages. Apparently you could do some obscure setting about the former; but the latter was so bad that I kept my copy of FR8 installed as well.
Curiously Abbyy decided not to tell people like me that FR10 is out. After all, we’d want to upgrade, and they’ve decided not to allow us to, at least not until December. Ah, the curious mind of marketing people!
Still that’s not long now. I’m looking forward to it. The OCR in Acrobat is rubbish anyway. I hope I can throw it away and go back to using FR instead.