[OT?] Wrestling HP OCR

[W. eWatson]

Sometimes I like to copy text out of books using my HP printer's OCR
feature. The biggest problem is getting faithful translations from the
binding area. It's difficult, of course, to press down adequately. The
result is that maybe 1/2" or so either side of the seam, the page is
dark and overlaps the letters. Anyone got a way to deal with that?

I usually put up with it, and copy the text into Notepad. It knocks out
strange characters, and I can usually see incorrect letters along the
margins. After I correct the .txt material, I copy it into Word.

 

[Suzanne S. Barnhill]

In some cases I find I get better scan results (from books as well as other
sources) if I photocopy the document first (on a dedicated photocopier at,
say, the UPS Store), then scan. Reducing the item to B&W seems to help.

 

[W. eWatson]

Continuing my story from a few months ago. I'm now scanning text with
common math symbols, /, (,),=, etc. The scanner seems to miss = signs
pretty often. Some of the others are hit and miss. Is there some way to
tell the HP C6100 that math symbols are present?

I'll likely check with HP tomorrow.