[Radiance-general] New falsecolor scale in 3.8

[Gregory J. Ward]

Hi Axel,

It's a fair question.  I was inspired during last year's attendance  
of the Color Imaging Conference, where I find myself again this  
year.  It occurred to me that the original scale in false color  
didn't actually traverse that many "named colors," making it more  
difficult than it needed to be to differentiate values.  Having seen  
a number of color scales, including spectral scales like the original  
one I used and thermal scales, I decided that some mix of the two  
might work better.  So, I essentially chose colors from a palette  
moving through the cooler spectral colors, then shifting to a thermal  
scale in the green-to-red (which became a green-to-brown)  
transition.  This then continues on through red and orange to yellow,  
avoiding white at the end as I found it to be confusing.  I can't say  
the results are terribly aesthetic, and the images look a bit bloody  
in some cases, but it is much easier to pick values off the resulting  
scale.

The curves you plotted are simply a result of interpolating the  
colors I chose along the way.  I made some effort and went through  
numerous iterations to balance out the perceptual movement in an  
attempt to keep the scale uniform, again with debatable success.

-Greg

> From: "Axel Jacobs" <a.jacobs at londonmet.ac.uk>
> Date: November 7, 2006 6:28:32 PM MST
>
> Dear Greg,
>
> I'm just looking at the new mapping of the falsecolor scale in  
> RADIANCE 3.8:
>
> http://luminance.londonmet.ac.uk/pickup/colour_scale-3.8.png
>
> I makes me wonder where you got the inspirations for those bizzare  
> lines
> from. Hope you're not taking any illegal substances?
>
> Cheers
>
> Axel