hello... how to write a code for 0.7 membership threshold

I am working on cardiac MRI LV images. I need a membership threshold to remove the unwanted area for images and give me only bright area inside in contours.

Answer 1

[Vidhi Agarwal]

Hi,

To isolate the bright regions within contours in cardiac MRI images, you can use a technique called fuzzy c-means clustering to determine membership thresholds. To achieve the aim, follow the given below steps:

1. Use fuzzy c-means clustering to segment the image into different regions based on intensity.
2. Use the membership values from the fuzzy c-means clustering to create a binary mask that isolates the bright regions.

Sample Code with above consideration is given below:

% Read the cardiac MRI image 
image = imread('cardiac_mri.jpg'); % Replace with your image file 
if size(image, 3) == 3 

    image = rgb2gray(image); % Convert to grayscale if necessary 

end 

% Enhance the contrast of the image 
image = imadjust(image); 

% Reshape the image for clustering 
imageData = double(image(:)); 

% Apply fuzzy c-means clustering 
numClusters = 3; % You can adjust the number of clusters 
[centers, U] = fcm(imageData, numClusters); 

% Find the cluster with the highest intensity 
[~, maxClusterIdx] = max(centers); 

% Get membership values for the brightest cluster 
membershipValues = U(maxClusterIdx, :); 

% Reshape membership values to the original image size 
membershipImage = reshape(membershipValues, size(image)); 

% Threshold the membership image to create a binary mask 
threshold = 0.7; % Adjust the threshold as needed 
binaryMask = membershipImage > threshold; 

% Postprocess the binary mask 
binaryMask = imfill(binaryMask, 'holes'); % Fill holes 
binaryMask = bwareaopen(binaryMask, 50); % Remove small objects 

% Display the results 
figure; 
subplot(1, 3, 1), imshow(image), title('Original Image'); 
subplot(1, 3, 2), imshow(membershipImage, []), title('Membership Image'); 
subplot(1, 3, 3), imshow(binaryMask), title('Isolated Bright Region'); 

For given input image and above consideration the results are:

For better understanding of Fuzzy C-Means Clustering refer to the following documentation:

• web(fullfile(docroot, "help/fuzzy/fcm.html"))

Hope that helps!