generate a 3d scatter plot with z-support lines / pin plot / needle plot ...

48 Ansichten (letzte 30 Tage)
Andres
Andres am 10 Dez. 2024 um 11:02
Kommentiert: Star Strider am 11 Dez. 2024 um 13:54
I'd like to plot scattered 3d data. To enhance readability, I want to add support lines in z-direction ("needles") to the data points with matching colors, see the image below.
Is there a specific name for such a plot?
To generate the chart, I used the code below with a for loop. I wonder if you have ideas how to simplify the code.
X = rand(20, 3);
f = figure;
hsc = scatter3(X(:,1),X(:,2),X(:,3), 20, X(:,3));
cmap = colormap('cool');
axis tight
xlabel("x"), ylabel("y"), zlabel("z")
title('needle plot')
box on
hold on
% draw support lines ("needles")
zlo = zlim*[1;0];
c_inter_fun = @(x) ...
interp1(linspace(zlo, zlim*[0;1], length(cmap)).' , cmap, x);
for k = 1:numel(hsc.XData)
plot3(hsc.XData(k)*[1 1], ...
hsc.YData(k)*[1 1], ...
[zlo, hsc.ZData(k)], ...
Color=c_inter_fun(hsc.ZData(k)));
end

Akzeptierte Antwort

Star Strider
Star Strider am 10 Dez. 2024 um 11:24
See if the stem3 function will do what you want.
You will probably still have to use a loop of you want the individual stems and markers to be different colours, since the stem plots do not offer those options.
X = rand(20, 3);
f = figure;
hsc = stem3(X(:,1),X(:,2),X(:,3));
cmap = colormap('cool');
axis tight
xlabel("x"), ylabel("y"), zlabel("z")
title('needle plot')
box on
cm = cool(size(X,1));
f = figure;
hold on
for k = 1:size(X,1)
hsc = stem3(X(k,1),X(k,2),X(k,3), 'filled', Color=cm(k,:));
end
hold off
cmap = colormap('cool');
axis tight
xlabel("x"), ylabel("y"), zlabel("z")
title('needle plot')
box on
grid on
view(-27,30)
% hold on
% draw support lines ("needles")
% zlo = zlim*[1;0];
% c_inter_fun = @(x) ...
% interp1(linspace(zlo, zlim*[0;1], length(cmap)).' , cmap, x);
% for k = 1:numel(hsc.XData)
% plot3(hsc.XData(k)*[1 1], ...
% hsc.YData(k)*[1 1], ...
% [zlo, hsc.ZData(k)], ...
% Color=c_inter_fun(hsc.ZData(k)));
% end
.
  10 Kommentare
Andres
Andres am 11 Dez. 2024 um 13:08
Bearbeitet: Andres am 11 Dez. 2024 um 13:08
If you just want to sort the colormap values along the z-value, you don't need to interpolate.
Two sorts
[~, si] = sort(X(:,3));
[~, si] = sort(si);
cm = cmap(si,:);
or one unique
[~, ~, si] = unique(X(:,3));
cm = cmap(si,:);
would do.
As I want to link the colormap values linearly to the z-value, the explicit interpolation seems unavoidable.

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (1)

Jacob Mathew
Jacob Mathew am 10 Dez. 2024 um 11:13
Hey Andres,
You can use stem3 to achieve the effect you want. Here is an example:
X = rand(20, 3);
figure;
hold on;
% Create stem plot (needles)
stem3(X(:,1), X(:,2), X(:,3), 'Color', 'k');
% Customize plot
colormap('cool');
axis tight;
xlabel('x');
ylabel('y');
zlabel('z');
title('Needle Plot');
box on;
view(45,45); % To view the plot from a angle since plot is not interactive below
You can refer to the documentation of stem3 below:

Kategorien

Mehr zu Data Distribution Plots finden Sie in Help Center und File Exchange

Produkte


Version

R2024b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by