Making two different color maps in 2014b

I'm trying to figure out how to use two different color maps in 2014b and use multiple color bars. Specifically I want to plot some image data and on top of that plot a set of contours.

Basically in the past I used freezecolors and cbfreeze. The same method behind freezecolors still works, changing the CData of your image to RGB values, but the colorbar is still a problem. CBfreeze doesn't seem to work any more because the colorbar objects are no long structs and now are specific objects.

I want to have separate colorbars for both the image data and contours but I can't seem to figure out the issue with the colorbars.

Thanks,

John

[Doug Hull]

In R2014b MATLAB there is a colormap per axes, not just per figure. This solves lot of problems from the past. Since you are still using freezecolors to get two colormaps in the same axes, that is still fine.

The real question is getting two colorbars for the same axes, correct?

Can you post a typical firgure from before?

[Jorge Ramirez]

I have the same question/problem. I am attaching a figure of what I could do before 2014b using freezeColors and cbfreeze. However, cbfreeze does not seem to work in 2014b. Thank you.

[Adam]

Wow, I'd never noticed R2014b snuck in that change. It's something I've wanted for quite a while, but must have missed in the changes listings!

Answer 1

[Sean de Wolski]

You can still use the freezecolors approach or one like I use in my meshCanopy function (same idea). You'll need to set the 'Limits' property of the colorbar to the range you want (and perhaps the tick/ticklabel etc.). For example:

I = imread('cameraman.tif');
n = 64;
meshCanopy(I,stdfilt(I),parula(n),80);
h = colorbar;

%%Now compare
h.Limits = [256 256+n];

Answer 2

[Nash Chu]

Actually you just need to add 'gca' when you define your colormap. like this:

figure
subplot(2,1,1)
pcolor(x,y,pxy);shading interp;
colormap(gca,jet);
colorbar;
subplot(2,1,2)
pcolor(x,y,pxy);shading interp;
colormap(gca,autumn);
colorbar;

[Steven Lord]

I recommend avoiding using gca in your program files. When experimenting in the Command Window it's okay, but inside a function (and especially inside a GUI) it's too easy for the current axes to change without you realizing it. [As one example you create an axes intending to manipulate it using gca but before you can your user gets a little bored and clicks on a different axes.]

Instead, call subplot with an output argument (which will be the handle of the subplot axes) and pass that handle into colormap.

h = subplot(2, 1, 1);
% stuff
colormap(h, jet);

Answer 3

[Jorge Ramirez]

Thank you, Sean. However, the problem is how to add two different colorbars corresponding to the two different colormaps. Your example shows only one colorbar. I am attaching an example graph of what I mean. I produced the attached graph with Matlab 2013b using freezecolors and cbfreeze. However, as I said, cbfreeze does not work anylonger on Matlab 2014b.

Thank you.

Jorge

[Sean de Wolski]

In that case just call colorbar twice and change the position as necessary:

I = imread('cameraman.tif');
meshCanopy(I,stdfilt(I),parula(64),80);
ax = gca;
h = colorbar('peer',ax);
h.Limits = [1 256];
h.Ticks = [];
h.Position(1) = h.Position(1)-0.1;
h = colorbar('peer',ax);
h.Limits = [256 256+64];
h.Ticks = [];

[John]

Hi Sean,

I think you got it, thanks for the help! The only thing I'm going to have to watch out for is if the two data sources here are in the same range. It looks like you get around that by adding an offset to the data in the manifold right? I believe you can work around this by changing the labeling on the colorbar. I'll post a piece of example code if I get it working to show what I'm talking about.

Answer 4

[Raul]

Still isn't clear how to use two different colormaps. How do you change the colormap?

Answer 5

[Chad Greene]

I made a solution to this for a problem I was working on. It's called newcolorbar.

[Zhengwei Lai]

thanks for your suggest