+CM

MATLAB git submodule package with common colormaps used by Max at NML.

Code

This should be a relatively small submodule, with a few functions and one main class.
Generally, the idea is to generate some colormap of interest with whatever number of discrete points are desired. That should be done using cm.map or cm.umap. To handle interpolation between values in the desired colormap, you can instantiate a cm.cmap object, using the colormap you've created in its constructor.

Functions

• map - Returns custom colormap data.
• umap - Returns custom colormap data based on a single fixed color.

Classes

• cmap - This just handles interpolating between discrete colormap indexed values.

Example 1

First, I can get a list of what default colormaps are available:

cm.map('help');

Now, I pick a value from the list and generate a colormap. Any color data map returned by cm.map will be size [256, 3]. However, maybe I want to interpolate between only a few discrete values, along that custom map (for example lets say I have 8 discrete classes). To do that, I'll get the color data map first, and then use a cm.cmap object to handle interpolation:

cmapdata = cm.map('rosette');
cmobj = cm.cmap([1, 8], cmapdata);

That's it! Now, to get a color value for some intermediate index (let's use 5 in this example), I can simply do:

color = cmobj(5); % [r, g, b] as uint8

Note that this strategy will work for non-integer values on the domain [1, 8] as well.

Example 2

To generate and visualize a quick-and-dirty colormap using a single color value that is scaled from dark to light, try the following:

my_custom_map = cm.umap("#FF0000");
fig = figure; 
ax = axes(fig);
C = reshape(my_custom_map, 1, 256, 3);
imagesc(ax, [1 256], [0 1], C);

You can swap in different colors in the first line, as desired.

Troubleshooting

All graphics are only white or black?

If you're setting a color property but only get objects that are white/black or discrete "R" / "G" / "B" values, it's likely the case that the graphics object expects a value between 0 and 1 for color values. In that case, referring to the end of the first example section, you would modify like:

dcolor = double(color) ./ 255.0;

Now, the dcolor variable should scale on the correct range and fix the issue.