Contour not working: The delineation erases (or minimizes) previous data from a plot

Contour function issue (I think)

Hi everyone, I am trying to get a delaniation (by using contour function) on a certain area in a figure.

Long story short I have the range on neural activity in the auditory cortex (several recordings and animals)

I am trying to find the areas with most change by using 90-95% percentile and making a X-score mask at p= 0.05.

When imaging the Range: It works flawlessly.

When imaging the Zscore: It works flawlessly.

When adding the contour of the zscore to the Range it does not work and the range seems to empty:

I get no warning and no error messages...

Here is the plotting code:

nRecs3 = numel(Recordings3);

figure;

sgtitle('Mouse 168: Range per condition','fontweight','bold','fontsize',16); %overall title

for iR3 = 1:nRecs3

sp = subplot(1, nRecs3, iR3);

im = imagesc(sp, squeeze(RANGE_3(:,:,iR3)'));

hold on; colorbar;

set(gca,'DataAspectRatio',[1,1,1]);

cont = contour(sp, ZMASK_3(:,:,iR2)', 1, 'r');

title(RecordingNames3{iR3})

xlabel('Anteroposterior (mm)'); %x axis label

ylabel('Mediolateral (mm)'); %y axis label

end

Does anyone know what the issue may be and how can I fix it?

As some extra info I do the same as previously (except looking at bottom percentile) mentioned with the RATIO and that seems to work just fine:

Answer 1

[DGM]

Both the image object created by imagesc() and the contour object created by contour() rely upon the current axes colormap and caxis settings. As such, a contour plotted over an pseudocolored image will be invisible since both objects use the same scaled mapping.

You can fix this either by creating a duplicate overlaid axes:

https://www.mathworks.com/matlabcentral/answers/194554-how-can-i-use-and-display-two-different-colormaps-on-the-same-figure

... or by converting the pseudocolor image generated by imagesc() to an RGB image:

https://www.mathworks.com/matlabcentral/answers/295620-how-can-i-overlap-two-images-pictures-in-matlab#answer_964200

... or you can try to directly edit the undocumented properties of the contour() object to break its dependence on the axes colormap.

[X Y Z] = peaks(100);

imagesc(Z); hold on

[~,hc] = contour(Z,[-1 0 1]);
drawnow

levelcolors = [1 0 0; 0 1 0; 0 0 1]; % colors for each level
for k = 1:numel(hc.EdgeLoopPrims)
    hc.EdgeLoopPrims(k).ColorData = im2uint8([levelcolors(k,:) 1].');
end

[Ada]

Hey, thanks for all, in the end my supervisor helped me by manually swapping the pseudocolor to an RGB image as you suggested!

    nRecs3 = numel(Recordings3);
    figure;
    sgtitle('Mouse 168: Range per condition','fontweight','bold','fontsize',16); %overall title
    for iR3 = 1:nRecs3
      sp = subplot(1, nRecs3, iR3); 
      im = imagesc(sp, squeeze(RANGE_3(:,:,iR3)'));  
      imMax = max(im.CData,[],'all');
      colorbar; hold on;
      im.CData = ind2rgb(round(im.CData/imMax*255),map);  % convert pseudocolor to RGB for contour compatibility
      set(gca,'DataAspectRatio',[1,1,1]);  
      cont = contour(sp, ZMASK_3(:,:,iR3)', 1, 'r'); hold on;
      title(RecordingNames3{iR3})
      xlabel('Anteroposterior (mm)'); %x axis label 
      ylabel('Mediolateral (mm)');  %y axis label
    end