Re: Questions on some basic definitions and concepts.
Re: Questions on some basic definitions and concepts.
- Subject: Re: Questions on some basic definitions and concepts.
- From: dpascale <email@hidden>
- Date: Mon, 24 Mar 2014 00:19:05 -0400
Joseph,
As I said there is confusion on the terminology and how it is used and I
would not bet on a unique set of descriptive words.
Munsell is a system which is implemented as a 3D color space (Berns), so
whether we call it a space or a system is both OK.
I have seen L*a*b* being described as a system (Hunt), and as a color space
(Berns).
Lets say we agree that Munsell and L*a*b* are color spaces. Then, according
to what you wrote, they should be different implementations of a color
model. Unfortunately, Munsell does not have mathematical formulas similar to
what we have behind L*a*b*. So Munsell is both a color system and a color
space and Pantone a color system but not a color space. Munsell and L*a*b*
can represent any color using a set of three orthogonal coordinates, so they
both are 3D color spaces. There is a one-to-one match between Munsell and
L*a*b*, even if this match cannot be mathematically defined (conversions
require successive approximations)..
Not a simple semantic problem...
As for your search for the first reference for daylight illuminants, go to
books.google.com
search for
contributions to color science
the article
Spectral Distribution of typical daylight as a function of correlated
color temperature
is in there (plus a lot of other interesting stuff!).
Danny
----- Original Message -----
From: "Joseph Chamberlain" <email@hidden>
To: <email@hidden>
Sent: Sunday, March 23, 2014 11:30 PM
Subject: Re: Questions on some basic definitions and concepts.
Danny,
Thank you for your help and the detailed nature of your reply.
Just to clarify, I was under the impression that the Munsell system was a
color model.
When you say that Munsell, Lab, XYZ and xyY are absolute color spaces, I
was aware of it. I was under the impression that Lab, XYZ and xyY were
color models and Munsell a color system. Is this not the case ? In other
words my impression was that color models were defined by mathematical
formulas while color spaces provided different implementations of these
color models. Based on your explanation my interpretation was wrong. Is
this the case ?
Thank you again,
Joseph.
On March 10, 2014, at 8:58 AM, dpascale <email@hidden> wrote:
Hello Joseph,
Here are answers which try to supplement the ones already given.
D55 as daylight:
D55 is not the only reference to "Daylight". Any illuminant with a "D"
followed by a number, such as D50, D52, D65, D93, or D100, is daylight.
The one you use depends on application; generic graphic arts applications
usually select D50 while commercial paint applications are more D65. D55
is often seen as a reference for flash photography. The exact "daylight"
(or even another Illuminant such as Illuminant-A) to use for a given
application will usually be specified in a standard (such as ISO 3664,
ISO 12646, etc)
Note 1: The Standard Observer (2-degree or 10-degree) can also be
different between applications, but this is just another variable....
Color model, space, system, etc:
There is overlap, confusion and some distinctions that can be made in
these terms. Pantone and Munsell are often referred as "systems" (Munsell
Color System, Pantone Matching System) while RGB and CMYK (any flavour)
are most often referred to as "color spaces", but sometimes also as
"color systems". However, only RGB, CMYK and Munsell are color spaces,
which can be specified in terms of coordinates. Pantone is more of a
color collection, in this case as a series of colors which can be
obtained by a mix of basic colorants. A further distinction is that
Munsell is an absolute color space, like L*a*b*, XYZ, xyY, which can
describe any color. This is not true of most RGB and CMYK spaces (and
certainly not Pantone). The fact that L*a*b* equivalents of Munsell can
be derived does not mean that they are the same space, but since both
spaces are absolute, you can always find corresponding values.
Note 2: Munsell equivalents of very saturated colors, near the
chromaticity diagram periphery, are estimates at best and have not been
thoroughly verified.
Note 3: L*a*b* equivalents of Munsell colors will be different for
different Illuminant/Standard Observer combinations.
Note 4: Historically the Munsell system was "renotated" (i.e smoothed) in
association with xyY coordinates under Illuminant C. The two systems
remain distinct.
12 bit vs 14 bit color space:
In principle there should be no differences in the gamut and 3D shape of
these spaces (if generated with the same camera and software of course).
There may be differences in smoothness within the space. However, I doubt
that any 3D color space visualization tool will enable you to see
differences, unless the data processing is not well done.
Danny
Danny Pascale
www.babelcolor.com
----- Original Message ----- From: "Joseph Chamberlain"
<email@hidden>
To: <email@hidden>
Sent: Monday, March 10, 2014 7:41 AM
Subject: Questions on some basic definitions and concepts.
Hi again, everyone.
I have done quite a bit of research but haven't found the answer to two
questions. Any help or guidance you can provide will be very
appreciated.
Here they are:
1. A textbook with a chapter on color theory (the book itself is not
about color technology) has a reference for daylight but I haven't been
able to find the source for this reference. Here it is:
" A full-spectrum light source, as recommended, is one with a color
temperature close to 5500 K (D55) that is specifically balanced
throughout the visible spectrum. This light source is defined as
equivalent to daylight in Washington D.C. (Bureau of Standards) in the
month of June, during the hours of 12:00 noon to 1:00 pm, when there is
a slightly overcast sky; the CRI of this light source should be 90 or
higher. "
Do any of you know the source for this reference ?
2. I have been trying to understand the difference between a color model
and a color space but haven't found any text or any source that provides
a good definition for either one or explains both in a manner that is
clear and simple to understand. I am trying to understand the following:
2.a. Is Lab a color model or a color space ?
2.b. Is the Munsell Color system a color model ? In this case is the Lab
a color space based on the Munsell color system (color model) ?
2.c. Is RGB a color space ?
2.d. How can I draw graphs comparing the color spaces defined by RGB,
Adobe RGB and ProColor ? In case I have a raw file created by a 12-bit
and another created by a 14-bit camera, how can I draw a graph depicting
the difference in the color spaces of these two cameras ?
Thank you again in advance for your help.
Joseph.
_______________________________________________
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
_______________________________________________
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