Newsgroups: sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!pipex!uunet!newsgate.watson.ibm.com!hawnews.watson.ibm.com!hawnews!admin!danebury!ricky
From: ricky@hursley.ibm.com (Rick Turner)
Subject: Re: Color contrast enhancement for color-blind?
Originator: news@hercules.hursley.ibm.com
Message-ID: <1994Nov28.084851.67201@hursley.ibm.com>
Sender: news@hursley.ibm.com (Usenet news handler)
Date: Mon, 28 Nov 1994 08:48:51 GMT
Reply-To: shadowfax@vnet.ibm.com
Disclaimer: This posting represents the poster's views, not necessarily those of IBM.
References: <1994Nov15.233052.17076@news.uit.no> <3banku$c7s@newsbf01.news.aol.com>
Nntp-Posting-Host: danebury.hursley.ibm.com
Organization: IBM UK Laboratories Ltd.
Lines: 13

Take a look in the remote sensing literature for an algorithm known as
'balanced contrast enhancement technique'. It does something very similar
to what you are after. There was a paper on this in the International
Journal of Remote Sensing in about 1989/1990, but I cannot remember the
author (I *may* have a copy at home if you cannot locate it).

Also, if you don't care what the colors that come out of the algorithm 
are, take a look at the 'color decorrelation stretch' (also known as the
'principle components decorrelation stretch'). It separates regions based
on color very, very well but the resulting images tend to be a bit startling.
If you are simply interested in segmentation, this may help.

Rick
