🚀 go-pugleaf

In article <1ie20sINNe78@charnel.ecst.csuchico.edu>,
csyew@ecst.csuchico.edu (YewSim Cheah) wrote:
> 
> I would like to sort an array of RGB Palette in intensity order,
> but do not know how.  Does anybody please help ?
> 
NIH Image for the Macintosh can sort RGB palettes in intensity order. It is
available, including complete source, from zippy.nimh.nih.gov, in the
directory /pub/image.

--wayne

> I would like to sort an array of RGB Palette in intensity order,
> but do not know how.  Does anybody please help ?
> 
I sort first by min(r,g,b), then by saturation and then intensity.  The goal 
(I assume) is to get a sorted set of indices that displays reasonably as B&W. 
Intensity=(r+g+b)/3.

+----------------------------------------------------------------------------+
|Ross Smith,  Research Computing Resource, Department of Cell Biology, NYU-MC|
|E-Mail:  SMITH@NYUMED.BITNET (BITNET),   SMITH@MCCLB0.MED.NYU.EDU (Internet)|
+----------------------------------------------------------------------------+

In article <1993Jan9.134848.7175@mcclb0.med.nyu.edu> smith@mcclb0.med.nyu.edu writes:
>> I would like to sort an array of RGB Palette in intensity order,
>> but do not know how.  Does anybody please help ?
>> 
>I sort first by min(r,g,b), then by saturation and then intensity.  The goal 
>(I assume) is to get a sorted set of indices that displays reasonably as B&W. 
>Intensity=(r+g+b)/3.
>
>+----------------------------------------------------------------------------+
>|Ross Smith,  Research Computing Resource, Department of Cell Biology, NYU-MC|
>|E-Mail:  SMITH@NYUMED.BITNET (BITNET),   SMITH@MCCLB0.MED.NYU.EDU (Internet)|
>+----------------------------------------------------------------------------+


NOT! >Intensity=(r+g+b)/3.

The actual Intensity value is:


 Intensity:=(r*0.299)+(g*0.587)+(b*0.114)

giving intensity as [0..1]. These are to a few decimal, but more precise
coeff. are hard to find.

	Regards,
		S.


-- 
------------------------------------------------------------
 STEVEN PIGEON, a.k.a pigeons@jsp.umontral.ca, planet Earth.
==Departement d'informatique et de Recherche Operationnelle=
              == Universite de Montreal==

In article <1993Jan9.134848.7175@mcclb0.med.nyu.edu> smith@mcclb0.med.nyu.edu writes:
>I sort first by min(r,g,b), then by saturation and then intensity.  The goal 
>(I assume) is to get a sorted set of indices that displays reasonably as B&W. 
>Intensity=(r+g+b)/3.

I would recommend Y = 0.253r + 0.684g + 0.0634b. Try to display
(r,g,b)=(0,0,127) and = (0,127,0) and tell me whether you think they have the
same intensity.
In general I would recommend to be careful if you have to make any color to
b&w conversions. HSV etc. do a very lousy job and one should stick to
something like YES, YIQ, YCC, XYZ for the intensity part ( those four color
spaces have the same Y component, although I am not absolutely certain for the
Kodak YCC )

Reiner

>+----------------------------------------------------------------------------+
>|Ross Smith,  Research Computing Resource, Department of Cell Biology, NYU-MC|
>|E-Mail:  SMITH@NYUMED.BITNET (BITNET),   SMITH@MCCLB0.MED.NYU.EDU (Internet)|
>+----------------------------------------------------------------------------+

pigeons@JSP.UMontreal.CA (Pigeon Steven) writes:
>NOT! >Intensity=(r+g+b)/3.
>
>The actual Intensity value is:
>
>
> Intensity:=(r*0.299)+(g*0.587)+(b*0.114)

Why is this the case? Is it related to the sensitivity of the eye
for the different primary cones, or is it to do with physics?

Does it mean that green is the 'dimmest' colour and blue is
the 'brightest'? Does it also mean the RGB model is not ideal?

--
  ________  Lionel Tun, lionel@cs.city.ac.uk  ________
 / /_  __/\       Computer Vision Group      /\ \__  _\
/___/_/_/\/ City University, London EC1V 0HB \ \___\_\_\
\___\_\_\/        071-477 8000 ext 3889       \/___/_/_/

lionel@cs.city.ac.uk (Lionel Tun) writes:

>pigeons@JSP.UMontreal.CA (Pigeon Steven) writes:
>>NOT! >Intensity=(r+g+b)/3.
>>
>>The actual Intensity value is:
>>
>>
>> Intensity:=(r*0.299)+(g*0.587)+(b*0.114)

>Why is this the case? Is it related to the sensitivity of the eye
>for the different primary cones, or is it to do with physics?

Both.  The formula reflects the sensitivity of the eye to different
wavelengths (the luminosity function), in combination with the
physical emission spectra of the standard NTSC phosphors.

There are a lot of phosphors in use besides the "standard" ones,
so if it is important, the weights need to be measured.

Nit:  the above formula is for luminance.  What is "intensity?"

>Does it mean that green is the 'dimmest' colour and blue is
>the 'brightest'? Does it also mean the RGB model is not ideal?

Backwards.  Green has the larges weight because it makes the largest
contribution to luminance.

-- 

	Jeff Mulligan (jbm@vision.arc.nasa.gov)
	NASA/Ames Research Ctr., MS 262-2, Moffett Field CA, 94035-1000
	(415) 604-3745