How to combine two plots into one

I am using MATLAB R2018 b. I have two plots with same x but, the limits of y axes are different. Both plots has a line and some data points (different color of the line and shape of the markers). These two plots are output from two different scripts. I want to combine these two plots to compare my results.

I am unable to use the dock figure option which was suggested for one such similar questions in the community.

Any help would be appreciated.

Answer 1

[Star Strider]

Plot both data series on the same axes either in the same plot call, or use the hold function with two plot calls.

[SS]

These two plots are result of two codes. I cannot use "hold" command.

[Star Strider]

In what form do you have the plots? Are they in .fig files?

[SS]

Yes, they are in .fig form.

[Star Strider]

It is straightforward to recover the data from .fig files.

Here is some code I wrote for a similar problem:

I = openfig('YourFigure.fig');
Ax = gca;
lgnd = Ax.Legend;
Lines = findobj(Ax,'Type','line');

for k1 = 1:fix(numel(Lines)/6)                              % Consider 6 ‘Lines’ In Each Iteration
    for k2 = 1:6
        X(k2,:) = Lines(k2+(k1-1)*6).XData;                 % Get ‘X’ Values For This Set
        Y(k2,:) = Lines(k2+(k1-1)*6).YData;                 % Get ‘Y’ Values For This Set
    end
    Xm(k1,:) = X(1,:);                                      % X-Vector (For Plot)
    Ym(k1,:) = mean(Y);                                     % Y-Vector (For Plot) - Uses ‘mean’, ‘median’ Or Others May Be Appropriate As Well
end

figure
semilogy(Xm', Ym')
grid
legend(Ax.Legend.String)
xlabel(Ax.XLabel.String)
ylabel(Ax.YLabel.String)

Do that for each of your .fig files, save the ‘X’ and ‘Y’ arrays separately for each file (name them ‘X1‘, ‘Y1’ and ‘X2’, ‘Y2’ for each figure as appropriate, since you only have two figures), then analyse the ‘X’ and ‘Y’ data as you wish. You can then plot them together as well, as the figure call in my code illustrates.

Even though I do not have your .fig files, this code should be robust enough to work with them.

Consider using the save function to save the data your recover from your .fig files in a .mat file so you will not have to extract the data from them again if you want to do more analyses with them. You can then use the load funciton to load them back into your workspace.

[SS]

I am sorry. I didn't succeed with it. I am attaching the two figures. Could you please, help me with it.

[Star Strider]

It took a few minutes to adapt my previous code to your figures.

Try this:

fignames = {'Track_50_PDF_Ux.fig'; 'Track_100_PDF_Ux.fig'};
for k2 = 1:numel(fignames)
    SSF = openfig(fignames{k2});
    Ax = gca;
    Lines = findobj(Ax,'Type','line');
    for k1 = 1:numel(Lines)
        X{k2,:}(k1,:) = Lines(k1).XData;                        % Get ‘X’ Values For This Set
        Y{k2,:}(k1,:) = Lines(k1).YData;                        % Get ‘Y’ Values For This Set
    end
    figure
    plot(X{k2}(1,:), Y{k2}(1,:),'p',    X{k2}(2,:), Y{k2}(2,:),'-')
    grid
    xlabel(Ax.XLabel.String)
    ylabel(Ax.YLabel.String)
end

The ‘X’ and ‘Y’ cell arrays contain the information you want. The data for the first plot are in ‘X{1}’ and ‘Y{1}’, and ‘X{2}’ and ‘Y{2}’ for the second plot. (You can similarly save the x-label strings and y-label strings if you want to. I demonstrate how to get them in the plot calls.) See the documentation on Access Data in Cell Array for details on how to work with cell arrays if you are not familiar with them.

[SS]

Thanks for the reply. I am new to MATLAB. I tried it but, it didn't generate the combined figure. It is generating the original figure and another figure with different shape of the markers of the former. Is there something that, I have to add to the above code?

[Star Strider]

Add this after my previous code:

figure
hold all
for k = 1:numel(X)
    plot(X{k}(1,:), Y{k}(1,:),'p',    X{k}(2,:), Y{k}(2,:),'-')
end
hold off
title('Combined Data')
xlabel(Ax.XLabel.String)
ylabel(Ax.YLabel.String)

The complete code is now:

fignames = {'Track_50_PDF_Ux.fig'; 'Track_100_PDF_Ux.fig'};
for k2 = 1:numel(fignames)
    SSF = openfig(fignames{k2});
    Ax = gca;
    Lines = findobj(Ax,'Type','line');
    for k1 = 1:numel(Lines)
        X{k2,:}(k1,:) = Lines(k1).XData;                        % Get ‘X’ Values For This Set
        Y{k2,:}(k1,:) = Lines(k1).YData;                        % Get ‘Y’ Values For This Set
    end
    figure
    plot(X{k2}(1,:), Y{k2}(1,:),'p',    X{k2}(2,:), Y{k2}(2,:),'-')
    grid
    xlabel(Ax.XLabel.String)
    ylabel(Ax.YLabel.String)
end

figure
hold all
for k = 1:numel(X)
    lgndstr{k,1} = sprintf('Figure %d Data',k);
    lgndstr{k,2} = sprintf('Figure %d Line',k);
    plot(X{k}(1,:), Y{k}(1,:),'p',    X{k}(2,:), Y{k}(2,:),'-')
end
hold off
title('Combined Data')
xlabel(Ax.XLabel.String)
ylabel(Ax.YLabel.String)
legend(lgndstr{1,:}, lgndstr{2,:})

producing this plot:

That should do what you want.

[SS]

Thank you.

[Star Strider]

As always, my pleasure!

[SS]

Hi again. How can I have the axes lines on the right and top of the plot? To make it look like a box?

[Star Strider]

Hi!

Add this line somewhere after the plot call loop:

set(gca, 'Box','on')

That will produce the full box outline. See the Axis Properties section on Box.

[SS]

Thanks.

[Star Strider]

As always, my pleasure.