RE: Metamerism and Color Constancy
RE: Metamerism and Color Constancy
- Subject: RE: Metamerism and Color Constancy
- From: Marco Ugolini <email@hidden>
- Date: Tue, 4 Sep 2007 10:02:28 -0700 (GMT-07:00)
Peter MacLeod wrote:
>I think you have to be careful here. Metamerism, whether caused by a
>change in illuminant or by two different samples under the same
>illuminant, is a color match that happens because the integration of the
>spectra with the color matching functions yields identical tristimulus
>values. In the case of illuminants it's called "illuminant metamerism"
>in much of the literature. For example, you could think about a very
>spiky fluorescent light that matches a particular color sample against
>how that same sample looks under real D65 (think of a split-field
>apparatus like the ones used in the standard color matching experiments,
>where the two sides have an identical sample but different illuminants),
>but not all color samples, in which case the colors that matched were
>cases of illuminant metamerism, while the colors that didn't match
>represent illuminant metameric failure. That's not the same thing as
>color constancy. The "Metamerism (color)" wikipedia page describes this
>pretty well: "Color constancy is the everyday perception that the color
>of objects remain unchanged across significant changes in illumination
>color and luminance level. Color constancy is related to memory colors
>and discounting the illuminant."
Peter, you start by saying that " we have to be careful", but I fail to understand what exactly we should be wary of, or precisely which distinction you're attempting to draw here.
Could you please state your point as succinctly and clearly as possible?
One thing I would point out as inaccurate or possibly misleading: you write: "Metamerism, whether caused by a change in illuminant or by two different samples under the same illuminant...". Reading that, I assume that you are talking about "change in illuminant" for *one* sample as falling under the rubric of "metamerism". I disagree with that, and I won't bore everyone by restating the reasons why. You can re-read my previous posts for that.
To use your example, a sample whose colorimetric readings do not change under fluorescent and D65 illuminants has a good *color constancy* under both illuminants. "Metamerism" has nothing to do with it.
By the way, your Wikipedia quote does *not * appear to confirm what you are saying.
Best.
Marco Ugolini
_______________________________________________
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