... and of course you only need to turn it into text if you want to cut
and paste - if just to read or you can type copy the bits you want to
quote, then leave it as photographed.
David
> > > Trying to remember the name of the pic to text app, but memory is
> > > failing me at the moment - will have a browse through my CDs and that
[quoted text clipped - 16 lines]
>
> Nikki

Signature
If we're not supposed to eat animals, why are they made out of meat?
> > I have also experimented with digital camera to OCR. You have to have a
> > good image of the page to start with and convert it to a 300dpi tiff
> > image so that Omnipage or similar will read it. Not very successful in
> > terms of saving time, but good for whole page graphic images.
> I've been following this post with interest, waiting for the simple
> solution to appear: how about scanning photocopies?
Well, that would be OK if you had access to a photocopier. But I work
in countries where the documents I use are often in hard copy and no
photocopying is easily available. (It might be on another floor where
I can't take the original or under the control of someone who isn't
inclined to be cooperative. Or broken.)
In that case, the digital camera can work. The other benefit of the
digital camera here is that the document is often in a large heavy
book, and putting it on a photocopier for OCR purposes just doesn't
work, for obvious reasons. Also, the photocopier can add to the scan
problems if it runs characters together. Photographing it carefully,
even at comparatively low res, can sometimes yield better results.
> Works for me (using Omnipage) provided it's a decent copy, the rest is
> down to typeface style, print quality, and crap scribbled on the page
> by student morons. All this will pose an identical issue regardless of
> your method of capture - even if you design a megasatellite to take
> those digital pictures.
You don't have to take them at that high a resolution to begin with.
If the image is good enough at lower res, you can feed to a laptop in
that formay and then step it up to 300 later. Mind you, this is
cumbersome and you really have to want an OCR version to go through
with the various steps involved. For a non-touch-typist, it can be.
Denis Wright
Nikki - 24 Sep 2004 22:43 GMT
> > > I have also experimented with digital camera to OCR. You have to have a
> > > good image of the page to start with and convert it to a 300dpi tiff
[quoted text clipped - 30 lines]
>
> Denis Wright
OK, I accept that point. Mea culpa. Problem is: I've been hanging
around too many universities - which fund themselves on photocopying
these days and there's a machine behind every shelf. Guess what I was
saying is I'm surprised how well photocopies will scan. I'm using
Omnipage Pro.
Wish I could afford that good a camera!
Nikki
quietguy - 26 Sep 2004 00:35 GMT
One big problem with photocopying books (yep, I went to Uni too) is that
you either have to break the books spine or put up with distorted test
at the spine edge of the page - NBG for OCRing
David
... The other benefit of the
> > digital camera here is that the document is often in a large heavy
> > book, and putting it on a photocopier for OCR purposes just doesn't
> > work, for obvious reasons. Also, the photocopier can add to the scan
> > problems if it runs characters together. Photographing it carefully,
> > even at comparatively low res, can sometimes yield better results.
Rifty - 27 Sep 2004 16:22 GMT
> One big problem with photocopying books (yep, I went to Uni too) is that
> you either have to break the books spine or put up with distorted test
> at the spine edge of the page - NBG for OCRing
...which is of course exactly where the camera has an advantage.
Rifty

Signature
Academic and Computing Help
http://rifty.net