Finally, an OCR app that doesn't disappoint
- By John Breeden II
- Jan 23, 2004
OmniPage Pro 14 represents a first for the GCN Lab'it's an optical character recognition program that works, and works well.
It's been more than three years since the lab looked at OCR packages. The reason? They never measured up to our expectations. No application ever received a grade higher than C. And filled-in forms flummoxed every one.
OmniPage Pro 14 gets around this in a fairly ingenious way. It scans paper into Adobe Portable Document Format files and them converts them into one of 30 editable formats, such as Microsoft Word or Excel or PowerPoint. PDF files become something more than read-only. For example, if your agency has a complex PDF form with inaccuracies or outdated content, you can actually fix it.
For the most part this happened flawlessly in my tests. Formatting remained the same as in the original documents. Trouble cropped up only when an original had odd fonts that were not installed in the target program. Even then, the problems were minor and did not happen consistently.
You can scan documents with any standard scanner; I used the same scanner as in our last OCR reviews. OmniPage Pro 14 turned the test documents into PDF files with identical formatting. Once in PDF form, they could be handled and edited just like native PDF.
That's a pretty smart way to sidestep most of the problems with OCR accuracy. In 20 tests, OmniPage Pro 14 never garbled a word from the different documents. The worst job was formatting a very complicated form, but even then the errors were minor and fixable once I converted to an editable format.
Ironically, OmniPage Pro 14 also can add security to PDF files to keep them from being changed. There's a way around this, however. So long as you can print the PDF files, you can always rescan them and make a copy that can then be edited. You can add a watermark to documents to discourage this, but you can't completely have your cake and eat it, too.
Agencies that rely on PDF as if it were set in stone should watch out.
With such great performance and accuracy, anything more is just icing, but OmniPage 14 has several other nice features that might be handy depending on your needs.
A batch manager program will e-mail you when jobs are finished. If you're working with a huge file or many small ones, you can walk away and let the computer do the crunching until you're notified that the job is complete.
You can also convert documents to the spoken word, in Word format, using a component of OmniPage called RealSpeak. Section 508 accessibility compliance has become a must for government agencies, and RealSpeak goes a long way to make documents more accessible.
Though some disabled users will still need a full voice recognition program, users who have Word don't need any other software to hear the documents read aloud.
OmniPage Pro 14 restores my faith in OCR software, which I wrote off years ago. If you're unsatisfied with your current OCR setup, give OmniPage Pro 14 a try. You might just be converted as easily as a PDF into a Word file.