how to create and use mixed colormaps associated to positive and negative data values?

For example I have a 2D matrix with values from [-100,200]. I would like to plot positive values from white to black, and negative values from blue to red.

Answer 1

[Kelly Kearney]

I like using .cpt files for specific colormaps like this, with the cptcmap.m function:

cdata = [...
    -100   0   0 255   0 255   0   0
       0 255 255 255 200   0   0   0];
dlmwrite('mycmap.cpt', cdata, ' ');

axes;
imagesc(rand(100)*300 - 100);
cptcmap('mycmap', 'mapping', 'direct'); 
colorbar;

[TheStranger]

Very nice, thank you

Answer 2

[Abhishek Pandey]

Hello,

I understand that you want to plot a 2D matrix with colors from white to black for positive values and blue to red for negative values. Also, the values range from -100 to 200.

I believe you could do this by creating two different matrices, one which contains the positive values and other one containing the negative values. You can set the rest of the entries in these matrices to NaN.

For the matrix containing positive values, set the ‘colormap’ to ‘gray’ and plot it in a figure. After this, set the ‘colormap’ to ‘jet’ and plot the other matric on the same figure using the “hold on” command.

For more information about how to set the ‘colormap’, you can refer to the following documentation link:

http://www.mathworks.com/help/matlab/ref/colormap.html

I hope this helps!

- Abhishek

[Walter Roberson]

If the images are in the same axes then the colormap jet will change the colormap interpretation of the gray things already plotted, even in current versions of MATLAB. In R2014a or earlier, the same thing would happen if the two colormaps are in the same figure.

The File Exchange contribution freezeColors() can be useful for things like this.