carroll@s.cs.uiuc.edu (05/30/89)
Just a simple question -
I have a grid on which I am computing values. Assume they get
normalized to 0 to 1. I want to plot these on a color display, with a
nice color gradiation. The display is 256 entry color table with 8 bits R,G,B
in each entry. I'd like to use around 100, or 128, or something in that range
(the exact number is not important). I tried to get some good color maps,
but haven't come up with anything I really like. Surely _someone_ out there
has done something like this - what did you use? Thanks!
Alan M. Carroll "And there you are
carroll@s.cs.uiuc.edu Saying 'We have the Moon, so now the Stars...'"
CS Grad / U of Ill @ Urbana ...{ucbvax,pur-ee,convex}!s.cs.uiuc.edu!carrollhil@garlic.ADS.COM (Hilarie Nickerson) (06/06/89)
In article <207400004@s.cs.uiuc.edu> carroll@s.cs.uiuc.edu writes: > >Just a simple question - > I have a grid on which I am computing values. Assume they get >normalized to 0 to 1. I want to plot these on a color display, with a >nice color gradiation. The display is 256 entry color table with 8 bits R,G,B >in each entry. I'd like to use around 100, or 128, or something in that range >(the exact number is not important). I tried to get some good color maps, >but haven't come up with anything I really like. Surely _someone_ out there >has done something like this - what did you use? Thanks! Assuming that you want to show the range for a single variable, the color gradations most easily understood by people cover one dimension in HSV coordinates, preferably Saturation or Value (H is for Hue). In plain English: Map the values 0-->1 into gray-->hue (your choice) [saturation scale] or dark-->light (using one hue) [value scale] The saturation scale is less subjective (I think :-) and easier to generate by trial and error adjustment of RGB values. References: Couldn't locate exact issue, but look in fall '88 (?) Computer Graphics and Applications for article on data visualization. HSV -> RGB conversion is widely available, or use trial and error to find saturation gradations. More stuff, no reference: If it is important to be able to identify a particular sub-range in the data, try 0-->important sub-range--> 1 into hue1-->gray-->hue2. If you really want to have lots of different hues, most people find "rainbow" order (Red, ..., Violet) easiest to follow (there IS a reference for this, I just don't know where I saw it). Hilarie Nickerson hil@ads.com