As someone previously mentioned, Acrobat X is very good at doing OCR on "non-text" PDFs (and scanning to PDF is very easy to do these days). Not sure how well it works with JPGs converted to PDFs and then run through this OCR process. It does great with "normal" scanned documents. I'd be interested in hearing your results if you go this route.
The real advantage to this, of course, is that it makes unsearchable documents searchable. BTW: The JPGs have been produced from what source? Craig Ede
