Trying to add color gradient to patch(isosurface))

Hi all,

im currently trying to visualize a pressure field in a 3D voxel field of a 32x32x32 image.

The way this currently works is through using patch(isosurface()) as is shown in the code below.

image = rand(32,32,32); 
filtered = imgaussfilt3(image,2.5);  
BW = imbinarize(filtered,0.517);   
patch(isosurface(BW),'FaceColor','white','EdgeColor','black')

The image above is what it produces.

What it does is plot the contour of the 1's and leave empty the 0's but I want it to do something else. The 1 values are going to be replaced by pressure field values (1 to 33) and I want the white+black contour to change into a color gradient. So far I couldnt get any colormap to do this. Is this even possible?

Answer 1

[darova]

Here is an example (not tested)

% BW - is your 01 3d matrix
% A - is your color value matrix (0-33)
cm = jet(33);           % colormap
fv = isosurface(BW,0.9);
fv.facevertexcdata = cm(A(:),:);        % get color according to 0-33
patch(fv,'facecolor','interp')

[Simon Penninag]

Thanks for your reply!

I think there's a misunderstanding though, maybe I didn't explain well enough.

I do not have a color value matrix, I have a matrix called pz which is the same 'BW' matrix with different values.

This is the matrix that I plotted in the image of the original post. Now these exact values are replaced with values between 0 and 33.

I want the color to be dependent on the value in each position.

[darova]

Everything is ok. I understood correctly. Here are corrections

% BW - is your 01 3d matrix
% A - is your color value matrix (0-33)
cm = jet(34);                                   % colormap
fv = isosurface(BW,0.9);                        % extract patch data
[m,n,k] = size(A);              
[x,y,z] = meshgrid(1:m,1:n,1:k);                % mesh 
v1 = fv.vertices;       
A1 = griddata(x,y,z,A,v1(:,1),v1(:,2),v1(:,3)); % interpolate A data (0-33) for x y z position
ind = 1 + round(A1);                            % indices
fv.facevertexcdata = cm(ind,:);                 % get color according to 1-34

patch(fv,'facecolor','interp')

[Simon Penninag]

It works like a charm! Thanks!

Answer 2

[ANKUR KUMAR]

Not sure about the exact what you wish to plot. But you can give it a try to plot using imagesc, use the gray colormap to plot.

imagesc(nanmean(filtered,3)
colorbar
colormap(gray)

[Simon Penninag]

I'm trying to reproduce the contour shown in the original post. It's a 3D image of the voxels that have the value '1' as oppossed to the value '0'. This '1' is going to become a value between 0 and 33 and needs a corresponding colour gradient.