On 9/6/07, Ross Presser <[EMAIL PROTECTED]> wrote:
> I used the Any2DjVu Server at
>
>    http://any2djvu.djvuzone.org/any2djvu.php
>
> to convert this 18Mb full color TIFF to a 38 kilobyte DJVU file.  (By the
> way, why did you save it with no compression? ZIP compression reduced the
> original from 18mb to 15.5 mb ... not much, but a little savings.)
>
>    http://rpresser.googlepages.com/example.djvu (38k)
>
> There are a number of different freely available DJVU readers; Google is
> your friend here.
>
> When it is initially viewed as a DJVU file, it appears exactly the same as
> the color TIFF. But the View menu allows you to see only the black and white
> layer, suppressing the color background. (It also suppresses the color
> photographs.)  Now the black text is easily readable on a white background.
>
> Viewing only the color layer, we can see the color photos are still there,
> as well as some soft residue from the text.
>
> I wasn't able to easily find a way to export only the b/w layer using the
> viewer I have. But I was able to select it and copy it to the clipboard,
> then paste it into a new grayscale graphic (in another graphic editor) and
> save that.  (It claimed to be 3 unique colors, therefore grayscale rather
> than true black and white.) The result is here:
>
>    http://rpresser.googlepages.com/example-bw.tif  (79k)
>
> Note the incredible efficiency of DJVU here. The entire layered DJVU file,
> with all the color information and appearing identical to the original when
> viewed, is half the size of JUST THE TEXT, even when compressed using G4
> compression -- one of the best available for bw images.
>
> The final step in this work, if desired, would be to extract the color
> photos from the color layer, clean them up, and paste them back into the b/w
> layer.


Thanks, Ross, for your help. It is quite impressive impressive the
amount of size reduction that you achieved!

My ultimate goal is to get a pdf from the scanned document. I exported
from the djvu viewer with ascii and then I used the command

$ convert -monochrome -compress Group4 example.ppm example.pdf
Segmentation fault
$

getting always a "Segmentation fault". Any ideasto overcome the
"Segmentation fault"?

Paul



> On 8/30/07, Paul Smith <[EMAIL PROTECTED]> wrote:
> > On 8/30/07, Ross Presser < [EMAIL PROTECTED]> wrote:
> > > The best solution to this problem may be in the domain of JPEG2000 /
> djvu
> > > image segmentation algorithms.  Look up some scanned books on
> archive.org,
> > > particularly in djvu format. The djvu plugin lets you suppress the
> > > background completely ... so identifying what is "background" is part of
> the
> > > djvu construction.
> >
> > Thanks, Ross and Anthony. My figure is available at:
> >
> > http://rapidshare.com/files/52236170/example.tif.html
> >
> > (Note please that the file has the size of 18MB.)
> >
> > Paul
> >
> >
> >
> >
> > > On 8/29/07, Paul Smith <[EMAIL PROTECTED]> wrote:
> > > >
> > > > Dear All,
> > > >
> > > > I have got a colored scanned TIFF image from a newspaper article. Can
> > > > ImageMagick be used to convert the color of the image background to
> > > > white, with no loss?
> > > >
> > > > Thanks in advance,
> > > >
> > > > Paul
> > > > _______________________________________________
> > > > Magick-users mailing list
> > > > [email protected]
> > > >
> > >
> http://studio.imagemagick.org/mailman/listinfo/magick-users
> > > >
> > >
> > >
> > _______________________________________________
> > Magick-users mailing list
> > [email protected]
> >
> http://studio.imagemagick.org/mailman/listinfo/magick-users
> >
>
>
_______________________________________________
Magick-users mailing list
[email protected]
http://studio.imagemagick.org/mailman/listinfo/magick-users

Reply via email to