how to make 3D image from x(50x1doub​le),y(100x​1double),z (80x1double), intensity(50x100x80)

I want to make a 3D image of electric field near the gold nanoparticle, similar to the figure below (ACS omega 8.24 (2023): 21493-21505). I have x(50x1double),y(100x1double),z (80x1double), electric field (50x100x80). The electric field only locates near the gold nanoparticle, while the most other area electric field is zero. As such, I need to make it transparent where electric field is zero, to get a clear image. Thank you so much for your advice!

ACS omega 8.24 (2023): 21493-21505

Answer 1

[Image Analyst]

Not sure about the transparency but check out the Volume Viewer app and the Volume Segmenter app, both on the apps tab of the tool ribbon under the Image Processing group.

[Image Analyst]

If you need or want far more powerful 3-D visualization than MATLAB offers, see Avizo by Thermo Fisher Scientific

Answer 2

[Meg Noah]

Without having your actual data, perhaps something like ...

efieldimage = repmat(linspace(1,255,255),128,1);

numPoints = 60; % Number of points around the circumference
Length = 2;
Radius= 6;
% Create the mesh grid for the cylinder
[Theta, z] = meshgrid(linspace(0, 2*pi, numPoints), linspace(0, Length, numPoints));
Theta(2:2:end) = Theta(2:2:end) + pi/numPoints;
x = Radius*cos(Theta);
y = Radius*sin(Theta);
for k = 1:10
    idx = find(z == z(k,1));
    x(idx) = ((k-1)/10)*(Radius)*cos(Theta(idx));
    y(idx) = ((k-1)/10)*(Radius)*sin(Theta(idx));
    idx = find(z == z(end-k,1));
    x(idx) = ((10-k)/10)*(Radius)*cos(Theta(idx));
    y(idx) = ((10-k)/10)*(Radius)*sin(Theta(idx));
end
z(1:10,:) = z(11,1);
z(end-9:end, :) = z(end-10,1);

s = surf(x, y, z);
s.CData = efieldimage;              % set color data to data
s.FaceColor = 'texturemap';    % use texture mapping
s.EdgeColor = 'none';          % remove edges
s.FaceLighting = 'gouraud';    % preferred lighting for curved surfaces
s.SpecularStrength = 0.1;      % change the strength of the reflected light
s.FaceAlpha = 0.25; % transparency
s.SpecularExponent = 25;
s.DiffuseStrength = 1.0;
axis off
view(80,68)

●

Answer 3

[Walter Roberson]

The basic 3D volume viewer is volshow or the application VolumeViewer https://www.mathworks.com/help/images/ref/volumeviewer-app.html

Unfortunately, both of those are for uniform volumes with equidistant centers in x y and z .

So... to use the data you will need to use something like

xu = unique(x); xu(isnan(xu)) = [];
assert(numel(xu)>=2, 'not enough unique x');

yu = unique(y); yu(isnan(yu)) = [];
assert(numel(yu)>=2, 'not enough unique x');

zu = unique(z); zu(isnan(zu)) = [];
assert(numel(zu)>=2, 'not enough unique x');

xdelta = min(diff(xu));
ydelta = min(diff(yu));
zdelta = min(diff(zu));
delta = min([xdelta, ydelta, zdelta]);

xvec = xu(1):delta:xu(end);
yvec = yu(1):delta:yu(end);
zvec = zu(1):delta:zu(end);

Egrid = interp3(x, y, z, electric_field, xvec.', yvec, reshape(zvec,1,1,[]));

Now you can run the volume viewer on Egrid

The code gets simpler if only you knew that x or y or z were equal spaced. As it is, we have to assume that the marginal vectors are irregularly spaced and have duplicate values and fold back on themselves multiple times (because we are not told otherwise.) If the vectors are all equally spaced, the code can be mostly replaced with a single call to imresize3