multiple colormaps

[pcpuck]

I would like to use different colormaps in different subplots. This simple example doesn't seem to work:
%%%%%%%%%%%%%%%%%%%%%%%
P=peaks(40);
C=del2(P);

subplot(1,2,1)
surf(P,C)
colormap white
colorbar

subplot(1,2,2)
surf(P,C)
colormap hot
colorbar
%%%%%%%%%%%%%%%%%%%%%%%%
Matlab always uses the last defined colormap, but I would like to use a different one for each subplot.

Can anyone help, please.

/Peter

 

[vorwald]

Since color map is a figure property, it takes a little work to get multiple color maps on a single figure.
Look at: Example - Simulating Multiple Colormaps in a Figure
at

http://www.mathworks.com/access/helpdesk/help/techdoc/creating_plots/chaxes17.shtml

 

[vorwald]

Here's your code rewritten

Cheers

%%%%%%%%%%%%%%%%%%%%%%%
colormap('default');

P=peaks(40);
C=del2(P);

ax1=subplot(1,2,1);
surf(P,C)
% colormap white % Sets the colors to black and white on the figure
cbar1=colorbar;

ax2=subplot(1,2,2);
surf(P,C)
colormap hot % Changes all the colors on the figures
cbar2=colorbar;

% A colormap is an m-by-3 matrix of real numbers between 0.0 and 1.0.
% Each row is an RGB vector that defines one color. The kth row of
% the colormap defines the k-th color, where
% map(k, = [r(k) g(k) b(k)]) specifies the intensity of
% red, green, and blue. Color is obtained by mapping the color data (C)
% into the color map indices using

% colormap_index = fix((color_data-cmin)/(cmax-cmin)*cm_length)+1

% In the normal case, min(cdata) maps into 1, and max(cdata) maps into
% size(colormap,1). To have different color maps in different plots on a
% figure, Cmin and Cmax of each plot modified so that min(cdata) and max(cdata)
% maps into the start and end of the appropriate subsection of the colormap.
% We can do this by solving the above equation for cmin and cmax for two pairs
% of (color_data, colormap_index). For a composite color map, we know the
% range of the color map (colormap_index_min and colormap_index_max) and the
% range of the color data (color_data_min and color_data_max), the above
% equation can be solved for cmin and cmax.

% cmin = ( (colormap_index_max-1) color_data_min
% - (colormap_index_min-1) color_data_max )
% / (colormap_index_max - colormap_index_min)
% cmax = ( (colormap_index_max-1) color_data_min
% - (colormap_index_min-1) color_data_max
% + cm_length (color_data_max - color_data_min) )
% / (colormap_index_max - colormap_index_min)

% To use different color maps for different plots in a figure, you need to
% modify the color map so that it contains all the color maps for the
% plots on the figure. Also, you need to modify cmin and cmax (which come
% from CLim) of each plot so the plot maps cdata the proper index range of the color
% map.

% The composite color map can be created by stacking the desired color maps
% on top of each other. Keep track of the range of the color maps

cmap1=bone;
cmap2=jet;
colormap([cmap1;cmap2]);
cm_length = size(get(gcf,'colormap'),1);
range(1,1:2) = [1 size(cmap1,1)];
range(2,1:2) = [1 size(cmap2,1)]+range(1,2);
colormap_index_min = range,1);
colormap_index_max = range,2);
color_data_range(1, = get(ax1,'CLim');
color_data_range(2, = get(ax2,'CLim');
color_data_min = color_data_range,1);
color_data_max = color_data_range,2);


cmin = ( (colormap_index_max-1) .* color_data_min ...
- (colormap_index_min-1) .* color_data_max ) ...
./ (colormap_index_max - colormap_index_min);
cmax = ( (colormap_index_max-1) .* color_data_min ...
- (colormap_index_min-1) .* color_data_max ...
+ cm_length * (color_data_max - color_data_min) ) ...
./ (colormap_index_max - colormap_index_min);

% Set the cmin and cmax of the plots on this figure

set(ax1,'CLim',[cmin(1),cmax(1)]);
set(ax2,'CLim',[cmin(2),cmax(2)]);

% The colorbars are managed differently since the
% color data mapping is set to direct. This means that the values
% of the color data are indices in the color map.

set(get(cbar1,'children'),'CData',[colormap_index_min(1):colormap_index_max(1)]');
set(get(cbar2,'children'),'CData',[colormap_index_min(2):colormap_index_max(2)]');
%%%%%%%%%%%%%%%%%%%%%%%%

 

[mhovers]

I am ploting multiple 3D surfaces in the same figure. Each surface has a different data range. I want each to have its own colormap. The examples in MatLab help deal with having multiple axes, and setting CLim differently for each axes after concatonating colormaps.

With only 1 axes and multiple surfaces how can I have each surface with its own colormap. Or at least each surface with its data range mapped into a color range in the one colormap so that each surface shows the variaiton in its data?

thanks,