Re: Normalizing CIEXYZ Values
Re: Normalizing CIEXYZ Values
- Subject: Re: Normalizing CIEXYZ Values
- From: Graeme Gill via colorsync-users <email@hidden>
- Date: Sun, 06 Nov 2022 17:54:32 +1100
Jeff Nova via colorsync-users wrote:
I have a math question for you. I would like to display the viewing
conditions' illuminant and surround on a 1931 2deg chromaticity plot.
These are un-normalized CIEXYZ values that need to be normalized within the
plot range (0-1 both axes) first.
Example:
sRGB IEC 2.1 profile:
Illuminant - 19.6445 20.3718 16.8089
Surround - 3.92889 4.07439 3.36179
How to normalize them? Please forgive my ignorance; I am learning every
day.
Hello Jeff,
when dealing with reflective or transmissive media, it
is usual to relate measurements to 100% reflectance or transmittance
of the light source. Typically the instrument does this for you -
it measures the reflected/transmitted spectrum then divides by the
actual instrument 100% illuminant spectrum, then computes the resultant XYZ
by integrating over an (by ICC convention) D50 light source.
That light source XYZ value will have a Y value of 1.0 or 100%, meaning
that the measured values will also cover this range.
Within the ICC profile itself reflectance or transmittance XYZ values have a nominal range
of 0..1, although it is convention in other file formats (such as CGATS) to use a % range
of 0..100. Some measurements might exceed 100% due to fluorescence etc.
Viewing conditions will typically remain in absolute XYZ terms so as to convey
the absolute light level, but if only the chromaticity is of interest, you could
divide the XYZ by the Y value.
As noted by Wayne Bretl, xy chromaticity values don't typically require
this normalization since it is taken care of by the very nature of the
calculation.
Cheers,
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