Re: Neutral grey under different lighting
Re: Neutral grey under different lighting
- Subject: Re: Neutral grey under different lighting
- From: Graeme Gill <email@hidden>
- Date: Wed, 05 Sep 2007 12:10:00 +1000
Fleisher, Ken wrote:
On 9/4/07 9:21 AM, "Graeme Gill" <email@hidden> wrote:
Fleisher, Ken wrote:
Two spectral distributions that do not appear identical are not metamers
only if they never appear identical. If there is an illuminant that can
cause them to appear the same, then they are metamers regardless of whether
you are currently viewing them under that condition.
I think you'll find that it is the stimuli (the light hitting the
observers eyes) that are metamers or not. The mechanism used to create the
stimuli is irrelevant (ie. whether it uses reflective samples and certain
light sources, or whether it is an emissive source etc.) So of course
two spectral stimuli that do not appear identical are not metamers, and
can never appear identical, because you can't change them without them
being different spectral stimuli.
It is metamerism that allows (for instance) a color displayed on a CRT
display to visually match a color displayed on an LCD display, even though
they don't match spectrally. No object colors involved.
I don't think that's correct. Metamers are objects that produce the same
tristimulus values when viewed under a reference illuminant for a particular
standard observer. Metamers are the objects, not the resulting spectra. You
cannot have two illuminants that are metamers. Refer back to the CIE
definition that Roger posted:
> CIE Publication 15.2-1986, Section 5.2, page 35, describes metamerism in the
> following way : "Two specimens having identical tristimulus values, XYZ, for
> a given reference illuminant and reference observer are metameric if their
> spectral radiance distributions differ within the visible spectrum".
You're inferring more from this than is intended I think. Note the above
refers to "tristimulus values, XYZ". This is a stimuli, not a reflectance
value. "spectral radiance distributions" is a stimuli, not a reflectance value.
Just because two particular samples under a given illuminant have an appearance
match due to metamerism, does not imply that the property applies to the samples
themselves, it is the light stimuli emitted by the samples under that illuminant
that is metameric.
It is easy to understand this confusion, since almost all references to
metamerism use an example of different reflectance samples that match
under one illuminant, and don't under some other, but metamerism
is a more fundamental phenomenon than that being something connected
only with reflective surface.
But if they are metamers, then they are always metamers. The spectral
properties of the two objects don't change just because you view them under
a different illuminant
The spectral reflectance of the objects didn't change, but the spectral
stimuli did, and it is the matching of the different spectral stimuli
that is metamerism.
> (we are ignoring any effects of fluorescence, of
course). If they will produce the same XYZ under D50 but not A, then just
because you are viewing it under illuminant A doesn't mean they no longer
will produce the same XYZ under D50. They are still a metameric pair. The
property is constant.
It's not a property of an object, it's a properly of the light stimuli. That's
because it's a result of the operation of the receptor (the eye), and that's
why it's the basis of all tristimulus color reproduction, including emissive
displays.
If metamerism was a term only for the property of two selected reflectance
colors appearing to be identical under one illumination, then it would not be
referred to as the basis for all tristimulus color reproduction, and some
other term would have to be used to cover the more universal property
known as "metamerism".
Graeme Gill
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden