speckle parren analysis and displacement measurement between every two concecutive frame

I have a series of frames containing speckle patterns for which I need to calculate the displacement. I require each subsequent image to be compared with the previous image to determine displacement. In other words, I need to calculate the displacement between every two consecutive images. Can anybody assist me with this issue?

[Manikanta Aditya]

Try checking the following reference about 'imregdemons' to know about estimating displacement between images:

• Estimate displacement field that aligns two 2-D or 3-D images - MATLAB imregdemons (mathworks.com)

Answer 1

[Rishi]

Hi Parastoo,

I understand that you want to know how to find the displacement between two consecutive frames.

You can find the displacement by using cross-corelation. This can be achieved using the 'normxcorr2' function. This function calculates the normalized cross-correlation of two matrices and returns a matrix containing correlation coefficients.

Here is a sample code for the same. You can modify it and use it for your own case.

num_images = 10; % Update this with the actual number of images

displacements = zeros(num_images - 1, 2);

% Loop over each pair of consecutive images
for i = 1:(num_images - 1)
    img1 = imread(sprintf('frame%d.png', i));
    img2 = imread(sprintf('frame%d.png', i + 1));

    % Convert images to grayscale if they are not already
    if size(img1, 3) == 3
        img1 = rgb2gray(img1);
    end
    if size(img2, 3) == 3
        img2 = rgb2gray(img2);
    end

    % Perform cross-correlation
    c = normxcorr2(img1, img2);
    
    % Find the peak in cross-correlation
    [max_c, imax] = max(abs(c(:)));
    [ypeak, xpeak] = ind2sub(size(c), imax(1));
    
    % Account for the padding that normxcorr2 adds
    corr_offset = [(xpeak-size(img1,2)) (ypeak-size(img1,1))];
    
    displacements(i, :) = corr_offset;
end

In the above code, displacement is the difference in the peak locations in the cross-correlation matrix 'c' from the center.

You can find more details about the 'normxcorr2' function from the below documentation:

https://www.mathworks.com/help/images/ref/normxcorr2.html

Hope this helps!