Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!ucbvax!hplabs!hpfcdc!hpfcdq!jlg
From: jlg@hpfcdq.HP.COM (Jeff Gerckens)
Newsgroups: comp.graphics
Subject: Re: Color Quantization (was Re: Discussion of Computer Pornography)
Message-ID: <390033@hpfcdq.HP.COM>
Date: 25 May 89 22:34:22 GMT
References: <310@celit.UUCP>
Organization: Hewlett-Packard - Fort Collins, CO
Lines: 22


Oops! Lost the text to which I am responding.  Oh, well. 

Daryl Poe has done research into using a mean-cut rather than 
a median-cut algorithm with some positive results.  He can be
reached at poe%fokker@hplabs.hp.com.

I don't think Daryl reads notes anymore, so he isn't likely
to respond without direct intervention. :-)

As for using the HSV color space, distances in the HSV space
are not significantly more "real" than those in the RGB space.
CIELUV is the reccommended space for acheiving uniformity and
perceptual linearity.  Color Science by Wysiecki and Stiles is
a good general reference on color spaces and conversions,
although not very good for quantization algorithms themselves.

- Jeff Gerckens, Hewlett-Packard Graphics Technology Division
  (jlg%hpfcrg@hplabs.hp.com)

  "... Red is Grey, and Yellow White, But we decide which is right.
	   And which is an illusion?" - Moody Blues