Issue with dispaly of 3D images created from multiple 2D slices

Hi,

I have multiple 2D grayscale image slices (all of them are of same dimensions). I need to stack them to dispaly 3D image. I tried cat function and then implay. But this just plays video, I am not able to see 3D image.

I came across many threads relevant to this. But I am not sure which one will serve the purpose.

I have attached few 2D grayscale slices here. Could someone please suggest the possible method to get 3D reconstruction from these images?

Answer 1

[Anjaneyulu Bairi]

To effectively stack multiple 2D images and visualize them as a 3D image, follow these steps:

Image Acquisition:

• Sequentially read each 2D image and store them within a 3D matrix. This matrix acts as a volumetric representation where each slice corresponds to an individual image.

Volume Visualization:

• Utilize the "volshow" function on the constructed 3D matrix. This function facilitates the rendering and interactive visualization of the volumetric data, allowing for comprehensive analysis and exploration of the 3D image.
• Alternatively you can use "sliceViewer" for visualization.

Refer the below code for reference:

imageFolder = 'images_dir'; % Specify your image directory
imageFiles = dir(fullfile(imageFolder, '*.png')); 
numSlices = length(imageFiles);

% Read the first image to get dimensions
sampleImage = imread(fullfile(imageFolder, imageFiles(1).name));

% Check if the image is RGB and convert to grayscale if necessary
if size(sampleImage, 3) == 3
    sampleImage = rgb2gray(sampleImage);
end

[rows, cols] = size(sampleImage);

% Initialize a 3D matrix
volumeData = zeros(rows, cols, numSlices, 'like', sampleImage);


for k = 1:numSlices
    imageName = fullfile(imageFolder, imageFiles(k).name);
    img = imread(imageName);
    
    % Convert to grayscale if the image is RGB
    if size(img, 3) == 3
        img = rgb2gray(img);
    end
    
    volumeData(:, :, k) = img;
end

% Visualize the 3D volume
volshow(volumeData);

% Using sliceview to visualize
sliceViewer(volumeData);

Output of "volshow" function looks like:

Hope it helps!

[MechenG]

Thank you very much!. Actually these are focal stack images. I guess I need to follow different strategy?

Answer 2

[Mathieu NOE]

For a 3D rendering (like MRI) I suggest you try this : vol3d v2 - File Exchange - MATLAB Central

%% create a 3D rendering 
out = [];
for k=1:5
    img = rgb2gray(imread(['0' num2str(k) '.png']));
    out = cat(3,out,img);
end

h = vol3d('cdata',out,'texture','3D');
view(3);  
axis tight;  daspect([1 1 .4])
alphamap('rampup');
alphamap(.06 .* alphamap);

[Mathieu NOE]

if you just want to stack figures , you can do this

%% create stacked images (I am simply repeating the same image 5 times)
% plot each slice as a texture-mapped surface (stacked along the Z-dimension)
figure
hold on
for k=1:5
    img = imread(['0' num2str(k) '.png']);
    [X,Y] = meshgrid(1:size(img,2), 1:size(img,1));
    Z = ones(size(img,1),size(img,2));
    surface('XData',X-0.5, 'YData',Y-0.5, 'ZData',Z.*k, ...
        'CData',img, 'CDataMapping','direct', ...
        'EdgeColor','none', 'FaceColor','texturemap')
end
view(3), box on, axis tight square
set(gca, 'YDir','reverse', 'ZLim',[1 k])
hold off

[MechenG]

Thank you very much, Mathieu. I will try this all the images.

[MechenG]

Actually these are focal stack images. Sorry I forgot to mention this. Do I need to follow different stretegy?

[Mathieu NOE]

maybe this ?

(26) Focus Stacking in Matlab | LinkedIn

Creating extended depth of field image through focus stacking - File Exchange - MATLAB Central

Autofocusing-of-camera-using-a-stack-of-images - File Exchange - MATLAB Central

[MechenG]

Thanks! I will look into this.