Save plots in a for cycle

2 views (last 30 days)
Miguel Albuquerque
Miguel Albuquerque on 6 Jul 2022
Hey guys, I have this code that correlates two signals with a large amount of samples(30 million). I want to see the plot in the code(surf) for all the iterations, maybe, not all, but perhaps skipping one iteration. Meaning I see the plot in one iteration, the next I dont see, the following I see, and so on. How can I make this, so that the plots are saved in PNG perhaps,to a folder in my pc identified with different number so I can see which iteration it was.
I wanted to do this in a good matlab way, since there are a lot of plots.
Thanks in advance
reference_reshaped = reshape(Reference_signal, 1000, 4, []);
surveillance_reshaped = reshape(Surveillance_signal, 1000, 4, []);
counter=1;
sz = size(reference_reshaped);
for i = 1:sz(3)
r = reference_reshaped(:,1,i);
s = surveillance_reshaped(:,1,i);
[afmag3,delay3,doppler3] = ambgfun(r,s,Fs,[250000 250000]);
afmag3(afmag3>1 )= 1;
counter=counter+1
figure;
surf(delay3,doppler3,afmag3,'LineStyle','-.');
shading interp;
axis([-1e-5 1e-5 -2000 2000]);
grid on;
view([140,35]);
colorbar;
ylabel('Doppler (Hz)');
xlabel('Delay(s)');
title('Ambiguity Function Sref');
end
  2 Comments
Miguel Albuquerque
Miguel Albuquerque on 6 Jul 2022
Thanks in advance,
I cant attach the signals, they are too big, but in format complex double.
I want that the plot on the code is saved to a folder in each iteration. If possible, on first iteration the plot is saved, second iteration the plot is not saved, 3 iteration plot is saved,... So I could reduce the number of plots by half.
If not possible save the plot on each iteration.

Sign in to comment.

Accepted Answer

Jan
Jan on 6 Jul 2022
Edited: Jan on 6 Jul 2022
If you want to create every 2nd image only, simply use:
for i = 1:2:sz(3)
% ^^ then i is [1, 3, 5, ...]
To save the image, simply use exportgraphics and sprintf to create the file name.
Creating a new figure without deleting the old one will let Matlab crash due to exhausted resources. It is much faster to create the figure once only and to modify the graphics object only:
reference_reshaped = reshape(Reference_signal, 1000, 4, []);
surveillance_reshaped = reshape(Surveillance_signal, 1000, 4, []);
counter = 1;
sz = size(reference_reshaped);
FigH = figure;
AxesH = axes(FigH);
SurfH = surf(AxesH, [], [], [],'LineStyle','-.'); % [EDITED]
shading interp;
axis([-1e-5 1e-5 -2000 2000]);
grid on;
view([140,35]);
colorbar;
ylabel('Doppler (Hz)');
xlabel('Delay(s)');
title('Ambiguity Function Sref');
for i = 1:sz(3)
r = reference_reshaped(:,1,i);
s = surveillance_reshaped(:,1,i);
[afmag3,delay3,doppler3] = ambgfun(r,s,Fs,[250000 250000]);
afmag3(afmag3>1 )= 1;
counter=counter+1;
set(SurfH, 'XData',delay3, 'YData', doppler3, 'ZData', afmag3); % [EDITED]
exportgraphics(AxesH, sprintf('Image%07d.png', counter));
end
If you have 15 Million images and the creation and saving takes 0.1 seconds, the code runs for more than 17 days. Checking the images manually will need some years afterwards.
  3 Comments
Miguel Albuquerque
Miguel Albuquerque on 7 Jul 2022
Thanks a lot , it works fine

Sign in to comment.

More Answers (0)

Tags

Products


Release

R2021b

Community Treasure Hunt

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

Start Hunting!

Translated by