Color objects with areas larger than a set value

I 1960 images like this. I am trying to identify the regions where the areas have gotten larger than a set value by coloring them in. For example in the fourth Image the bottom cluster of circles would have a large area and that would be colored in. I pasted the code I am trying to implement below. Is there a way to do this using a loop or would a mask be needed first?

video = VideoWriter('Example.avi')
open(video);

for img = 1:1000
    
    file = strcat('bw_frame', num2str(img),'.jpeg');
    I = imread(file);
    I = imbinarize(I);
    [L, n] = bwlabel(I);
    stats = regionprops(L, 'Area');
    
    for count = 1:length(n)
        val = stats.Area;
        if val(count) > 1000
            rgb2 = label2rgb(L);
        end
    end
    
    writeVideo(video,rgb2)
    disp(img)
end
close(video);

Answer 1

[DGM]

I'm not familiar with the video IO tools, but isolating/coloring the large objects should be easy enough

I = imread('dots.jpg');
I = im2bw(I); % imbinarize isn't in my version
L = bwlabel(I);
stats = regionprops(L, 'Area');

maxarea=700;
hilitecolor=[0.6 0 0.8];

biglab = find([stats.Area]>maxarea)
bigmap = ismember(L,biglab);
outpict = repmat(I.*~bigmap,[1 1 3]) ...
	+ bsxfun(@times,bigmap,reshape(hilitecolor,[1 1 3]));

I'm sure there are some efficiency tweaks that can still be made to this, but it works.

[Alessandro De La Garza]

Thank you! that ended up working perfectly!