How can i change the colours of segmentation on a chart pie?

Hey,

I have plotted a pie chart, but the colours of the segmentation does not appeal to me...is there any way I can alter the colours?

Thank you.

Answer 1

[Adam Danz]

Unforunately the pie() function does not allow you to directly specify the face color of each wedge as inputs.

Here are two ways to change the wedge colors after producing the pie plot.

Produce a demo pie chart & define colors

ax = gca(); 
pieData = [.3 .4 .3]; 
h = pie(ax, pieData); 
% Define 3 colors, one for each of the 3 wedges
newColors = [...
    1,       0.41016, 0.70313;   %hot pink
    0,       1,       0.49609;   %spring green
    0.59766, 0.19531, 0.79688];  %dark orchid

Option 1 : change the axes colormap

You can use on of Matlab's many pre-defined colormaps or you can create your own has we've done above. This works with pie() and pie3() objects.

% ax is the handle to the pie chart axes
% newColors is a nx3 matrix for n pie-wedges
ax.Colormap = newColors; 

% to use a pre-defined colormap (in this example, 'Spring')
% h is the output to pie()
ax.Colormap = spring(numel(h)/2);

Option 2: Change the FaceColor of the wedges

For pie() objects

% h=pie() output is a vector of alternating patch and text handles. 
% Isolate the patch handles
patchHand = findobj(h, 'Type', 'Patch'); 

% Set the color of all patches using the nx3 newColors matrix
set(patchHand, {'FaceColor'}, mat2cell(newColors, ones(size(newColors,1),1), 3))

% Or set the color of a single wedge
patchHand(2).FaceColor = 'r';

For pie3() objects

% Extract the surface and patch handles from the output to pie3()
surfaceHand = findobj(h, 'Type', 'Surface'); % edges
patchHand = findobj(h, 'Type', 'Patch');     % tops & bottoms

% Set the color all patches (tops & bottoms) and surfaces (edges) using the nx3 newColors matrix
set(surfaceHand, {'FaceColor'}, mat2cell(newColors, ones(size(newColors,1),1), 3));
set(patchHand, {'FaceColor'}, mat2cell(repelem(newColors,2,1), ones(size(newColors,1)*2,1), 3));

Answer 2

[Dave B]

Starting in R2023b we have a new piechart which is a little easier to change colors for. Here are some demos with our built in color palettes, and then one with a custom palette. You can use the colororder function to change them or set the ColorOrder property on the PieChart object.

vals = [1 1 2 3 5];

figure
piechart(vals)
colororder("sail")

figure
piechart(vals)
colororder("reef")

figure
piechart(vals)
colororder("meadow")

figure
piechart(vals)
colororder("earth")

figure
% use validatecolor as an easy way to turn hex values into RGBs
mypalette = validatecolor(["#3BA8ED" "#ED4747" "#2FEDD1" "#ED9418" "#24ED5D"], "multiple");
piechart(vals)
colororder(mypalette)

[Jeremias]

Hi Dave,

thank you for this solution which in principal worked perfectly for me.

Unfortunately I now have the problem that the colors appear "dimmed down" and only if I hover over one segment it is displayed in the color specified in my code. If I save my piechart as png or any other format I have tried it will not show the correct color but the dimmed version,

Do you know of any solution?

Thank you in advance!

[Dave B]

Actually I do! The piechart uses transparency by default, which allows it to show the sort-of highlight effect, but that makes the colors look a little muted.

You can control the amount of transparency with the FaceAlpha property, and if you set it to 1 it will produce full opacity and the colors won't look muted. Here's the same palette as above,

vals = [1 1 2 3 5];
mypalette = validatecolor(["#3BA8ED" "#ED4747" "#2FEDD1" "#ED9418" "#24ED5D"], "multiple");

figure
piechart(vals, 'FaceAlpha', 1)
colororder(mypalette)