RE: Humans (and cameras and scanners) do not have a color gamut (?)
RE: Humans (and cameras and scanners) do not have a color gamut (?)
- Subject: RE: Humans (and cameras and scanners) do not have a color gamut (?)
- From: Wayne Bretl via colorsync-users <email@hidden>
- Date: Tue, 7 Jan 2020 15:20:50 -0700
The problem with this approach is that it stops understanding at the point
where “I can take an image from any camera and produce any colors I like.” In
fact, for a practical system, this is never done. The outputs of the sensor are
subject to at least linear matrixing (or direct control of a complementary
colorant) and the parameters are adjusted to get acceptable representation or
reproduction of the original scene. When this is done for good representation
of common broad object spectra, the representation of narrow spectra may
limited to lower saturation than what the eye would perceive, and the
reproduced color in the overall system will be limited to lower saturation (due
to one or two sensor channels going to zero). A realization that there is an
actual gamut limit to the numerical ratios of the sensor outputs and that this
results in a gamut limit of a reproduction system seems helpful to me.
Note: this type of limit was definitely present in color film, but not critical
for practical use because the dyes generally could not reproduce higher
saturation than the limits imposed by the sensor ratios anyway. In other words,
when common colors were produced with reasonable saturation, the sensor limits
occurred when object spectra narrowness corresponded with typical
maximum-saturated object surface color limits. With current wide color gamut
reproducers, cameras need to have broader spectral responses in the RGB
channels in order to not limit the signaled color gamut to less than the
display capability while simultaneously representing common broader-spectra
objects with correct saturation.
Raw image data is in some native camera color space, and while it is not a
colorimetric color space, and therefore may not report the same color as the
human eye, it will be translated to a color space. Narrow RGB sensitivities
that go to zero when the human eye cone sensitivities do not will impose a
limit on the gamut of reproduced colors. It is useful to understand this.
From: Andrew Rodney <email@hidden>
Sent: Tuesday, January 07, 2020 1:56 PM
To: WAYNE BRETL <email@hidden>; Andrew Rodney via colorsync-users
<email@hidden>
Subject: Re: Humans (and cameras and scanners) do not have a color gamut (?)
On Jan 7, 2020, at 1:51 PM, WAYNE BRETL <email@hidden
<mailto:email@hidden> > wrote:
(changing the subject now that the seemingly interminable is seemingly
terminated) :>)
Could you (and/or others) expand on this statement?
"Humans (and cameras and scanners) do not have a color gamut."
See:
http://www.color-image.com/2012/08/a-digital-camera-does-not-have-a-color-gamut/
Further (and getting back to Fairchild), I asked this question directly to Mark
D. Fairchild from RIT and I just would like to share his answer here:
"This one is easy for me … cameras absolutely do not have gamuts.
A color gamut is the range of colors produced by a device or system. I can take
an image from any camera and produce any colors I like. The colors produced are
not a property of the camera or limited by the sensors. Some argue that the
cameras can only detect a certain range of colors and that also has nothing to
do with “gamut” even if it were true. I can put any visible stimulus in front
of any camera (that responds to visible light) and I will get some response.
How that response is rendered to a display has an impact on the quality and
color accuracy of the camera, but it in no way creates a gamut (only the
display or some arbitrary limitation of the display such as R=G=B to make only
grayscale images limits the color gamut.).
So I fall strongly, and unequivocally, on the side that says cameras do not
have color gamuts. (FWIW, this isn’t even a discussion among the faculty in our
program, we all agree on this.)
The human eye also does not have a gamut. The spectral locus on chromaticity
diagram (which is also missing a dimension) simply shows the response of the
eye to monochromatic light. The limit is in the light, not the eye. The camera
can also respond to all that light."
Andrew Rodney
http://www.digitaldog.net/
_______________________________________________
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