RE: Metamerism
RE: Metamerism
- Subject: RE: Metamerism
- From: "Fred Bunting" <email@hidden>
- Date: Wed, 31 Oct 2001 17:38:32 -0800
- Thread-topic: Metamerism
Bruce Lindbloom wrote:
>
... I have assembled what I think is an exhaustive list
>
of metameric
>
cases. Of the twelve cases, seven involve two observers, and
>
therefore are
>
theoretically possible, but of little practical interest.
>
The remaining
>
five (1a, 2a, 2b, 2c and 3a) are the ones that matter.
I agree with your five cases that involve a single observer ... although
there are some interesting subtleties. The important thing, for those
people trying to understand the precise meaning of the term
"metamerism", is that the spectral compositions of the two light stimuli
*reaching the observer* are different, but produce the same response.
E.g. it is possible to take two spectrally different prints, and
illuminate each with a different light source, and get the identical
color stimulus reaching the eye. While this would trivially produce a
match, this would not be metamerism.
Where I disagree is when you say that cases involving two observers are
of "little practical interest".
I think what's confusing things is trying to construct a notion of what
it means to form a "match" in "two different brains" (as you put it
well). Your definition in your notes starts to go there, but would
involve a *lot* more discussion.
Instead, always remember that metamerism always boils down to a
stimuli-response event (what the scientists call 'psychophysics' ... the
psychological response to a physical stimulus), where the response is an
observer saying "match!" or "no match!". So each observer must be shown
a *pair* of samples.
So consider the case where Observer A is shown two prints under a
certain illuminant, and Observer B is shown the *same* pair of prints
under the same illuminant. So both observers are being shown the same
pair of stimuli. If Observer A says "match!" this is metamerism ... the
two stimuli are metameric for her. If Observer B says "no match!" then
the two stimuli are not metameric for him.
This has many practical implications. If Observer A is the press
operator, and Observer B is her color-anomalous client, the practical
implications are obvious. If Observer A is you, and Observer B is your
digital camera, then your camera may capture those lace curtains as a
different color from the napkins, when you "know" they are the same
color. If Observer A is your scanner, and Observer B is you or a
colorimeter representing you, then this explains why it's hard to do
printer profiling using a scanner.
Fred Bunting