Newsgroups: comp.graphics,sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!rochester!rit!isc-newsserver!jch3254
From: jch3254@osfmail.isc.rit.edu (HANDLEY)
Subject: Re: Unhalftoning Halftones
Message-ID: <1995May11.123222.22682@ultb.isc.rit.edu>
Sender: news@ultb.isc.rit.edu (USENET News System)
Nntp-Posting-Host: grace.isc.rit.edu
Organization: Rochester Institute of Technology, Rochester, NY
References: <3o756l$eot@mack.rt66.com> <1995May3.130026.1020@csginc.com> <D8CEDy.HJ@ucc.su.OZ.AU>
Date: Thu, 11 May 1995 12:32:22 GMT
Lines: 83
Xref: glinda.oz.cs.cmu.edu comp.graphics:76330 sci.image.processing:14545

I just wrote a small report on inverse-halftoning methods. 
The  problem is really one of reconstruction.
There are basically two approaches.
One, low pass filter with a small kernel,
then apply an adaptive smoother i.e.,
one that uses local variance information.
If the variance is high, do nothing or little
to the current gray value, but if the variance
is low, drive the current gray value toward
the local mean.  This is supposed to
preserve edges.  This procedure can 
be iterated with different smoothing
parameters.

If the halftoning method is known (or
can be estimated from the image),
one can use it with the above
method.  The rationale is that
we are trying to estimate the original
gray image, but all we have is a
binary version.  So if we reconstruct
the image and then halftone it using 
the known method, we should get the
same binary image we had to begin with.
If not, our reconstructed image is wrong
and needs to be adjusted.  This
is the basis of "projection" methods.

There are some ad hoc methods, too.
Below is a bibliography.  I'd recommend
Miceli and Parker or Wong as being
particularly readable.
If anyone knows of others, please let me know.

M. Analouli and J. Allebach, ``New results on reconstruction of
continuous-tone from halftone,'' {\em Proc. Int. Conf. Acoustics, Speech
Signal Processing}, March 1992, pp. 313--316.

Z. Fan, ``Unscreening of stored digital halftone 
images by logic filtering,'' {\em %
United States Patent 5,027,078}, 25 June 1991.

Z. Fan, ``Retrieval of images from digital halftones,'' {\em %
Proc. Int. Symp. Circuits Syst.}, May 1992, 2477--2480.

S. Forchhammer and K. S. Jensen, ``Filters involving derivatives
with application to reconstruction from scanned halftone images,'' {\em IEEE
Transactions on Image Processing}, April 1995, Vol. 4, No. 4, pp. 448--459.

S. Hein and A. Zakhor, ``Halftone to continuous-tone conversion
of error-diffusion coded images,'' {\em IEEE Transactions on Image Processing%
}, February 1995, Vol. 4, No. 2, pp. 208--216.

C. M. Miceli and K. J. Parker, ``Inverse halftoning,'' {\em %
Journal of Electronic Imaging}, April 1992, Vol. 1, No. 2, pp. 143--151.

P.G. Roetling, ``Unscreening of stored digital halftone images,''
U.S. Patent 4,630,125, 16 Dec. 1986.
 
M. Y. Ting and E. A. Riskin, ``Error-diffused image compression
using a binary-to-gray-scale decoder and predictive tree-structured vector
quantization,'' {\em IEEE Transactions on Image Processing}, November 1994,
Vol. 3, No. 6, pp. 854--857.

P. W. Wong, ``Inverse halftoning and kernel estimation for error
diffusion,'' {\em IEEE Transactions on Image Processing}, April 1995, Vol.
4, No. 4, pp. 486--489. 
 

Hope this helps.

John


--
John C. Handley   jch3254@rit.edu
Center for Imaging Science
Rochester Institute of Technology
Rochester, NY  14623-5604
http://www.cis.rit.edu



