Free OCR is great. The only problem is that it doesn’t fill the text in the document from margin to margin. Does anyone know of a free OCR that can do this?
I have the same question Lisa has but I'll try to answer one possibility:
OCR detects rendered text and turns it into selectable text. To do this it has to overlay, "underlay" or replace the image with actual text.
One reason why it may not have the same layout as the original is that it will probably use a different font - some system standard font which probably has a different width then the one used in the image.
What do you mean 'from margin to margin'? Perhaps you can use Calibre to fine-tune the result a bit, or reconvert it. It lets you choose the margin size.
Getting Started with FreeOCR
OCR Using Microsoft OneNote 2007