Is it possible to resize a region in a grayscale image?

Is it possible to resize a white blob in a grayscale image without changing anything else? If so, how can I do so? I am trying to resize the largest blob while keeping the size of the other regions the same.

Answer 1

[Gopichandh Danala]

The steps for this is to extract the largest Blob, resize it as per requirement, again map it back to the original image to increase the size of it.

img = imread('blobDilate.jpg');
img = rgb2gray(img);
img(img < 50) = 0;
[rows, cols] = size(img);
figure, imshow(img), title('Original Image');

% Get the max Blob
largestBlob = bwareafilt(logical(img), 1);
figure, imshow(largestBlob)

% increase the size  [+50, +50]of the Blob and remove the added rows, cols
resizeImg = imresize(largestBlob, [rows + 50, cols + 50]); % You can change this according your requirement

% remove added rows, cols
removeRows = [1:25, size(resizeImg,1)-24:size(resizeImg,1)];
removeCols = [1:25, size(resizeImg,2)-24:size(resizeImg,2)];
resizeImg(removeRows,:) = [];
resizeImg(:,removeCols) = [];
figure, imshow(resizeImg)

% New Blob Area
props = regionprops(logical(resizeImg),'Area');
resizedBlobArea = [props.Area];
fprintf('%s %d\n', 'newResizedBlobArea: ', resizedBlobArea)

% final Image
finalImg = logical(img + uint8(resizeImg));
figure, imshow(finalImg), title('Largest Blob resized Image')

[Image Analyst]

You know, you can get the largest blob simply by doing

largestBlob = bwareafilt(binaryImage, 1);

instead of all that stuff with regionprops(). largestBlob will be a binary image of the same size as the original but with only 1 blob, the largest, in it.

[Gopichandh Danala]

Oh thanks Image Analyst, I didn't knew it before. I updated the above answer