Xref: utzoo comp.sys.sgi:9071 comp.sys.apollo:8484
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!wuarchive!udel!haven!cvl!mimsy!oasys!dtrc!lipman
From: lipman@dtrc.dt.navy.mil (Lipman)
Newsgroups: comp.sys.sgi,comp.sys.apollo
Subject: Perceptually equidistant color scale
Message-ID: <6627@oasys.dt.navy.mil>
Date: 26 Mar 91 20:57:57 GMT
Sender: news@oasys.dt.navy.mil
Reply-To: lipman@dtrc.dt.navy.mil (Robert Lipman)
Followup-To: comp.sys.sgi
Organization: David Taylor Research Center, Bethesda, MD
Lines: 24

Hi, we're doing some color contouring of scalar values on
2d and 3d meshes of polygons.  Currently, we use a rainbow
color scale that varies in equal increments of hue in an
HLS color system.  Obviously, the same distance between any
two colors does not have the same perceptual distance.

I know that a lot of work has been done with color perception
and color science.  Tektronix has their TekColor color system
that is perceptually equidistant.

What we are looking for is a set of HLS or RGB values that would
define a rainbow color scale that is perceptually equidistant.
It is assumed that there would be a linear variation of HLS or
RGB between the specified values in the set.

Our main concern is how the colors look on the monitor.  I know
that once we print them or videotape them, they'll look
different.

Can anybody help?

Thanks in advance, e-mail please

Bob Lipman