SOT: Scanning old manuals

[Paul Koning]

>>>>> "Kevin" == Kevin Parker <Parker> writes:

 Kevin> Has anyone had any experience using Adobe Capture to scan in
 Kevin> old manuals etc.

I've used it on two documents.

As with other OCR tools, it requires editing after the fact to clean
up the spots that it gets wrong.  Unlike other OCR tools, this is
unacceptably painful -- the Acrobat edit tools are absurdly
cumbersome.

If the source material is pretty clean, the recognition quality is
good.  The main hassle tends to be that the font flips back and forth,
which makes the result look funny but doesn't interfere with being
able to use text search on the file.  If the source material isn't
really clean, then the scan quality drops, and some of the text will
not be recognized as text at all.  There isn't any editing during the
scan process (as in other OCR tools), and the after-the-fact editing
tools suck so you can't reasonably repair this sort of thing.  I made
a brief attempt at using Capture on a fairly poor quality original (a
line printer output of pretty average line printer quality).  The
result was absolutely unuseable.

So... the price is right (for the free limited edition), but I
couldn't imagine paying for the commercial version.  It's sometimes
useable, if you have a clean document and either don't mind the mixed
fonts and typos (one vs. ell, etc.), or if you have a small enough
document and a lot of patience to fix the errors one line at a time.

       paul