Retaining specific colors while tracking objects

Hi!

I'm using the watershed algorithm, combined with either 'turbo' or 'jet' to color detected objects.

thewatershedimage = label2rgb(watershedimage,'jet', [1 1 1], 'shuffle');

When I use it on a timeseries of images, I would like it to retain the same color for each object instead of shuffling it each time and assigning a new color.

Is there a straightforward way to do this?

Thanks!

[KSSV]

Can't you specify the color while plotting. ?

[Veena Chatti]

It's the "imshow(thewatershedimage)" command, not "plot(thewatershedimage)."

thewatershedimage = label2rgb(watershedimage,'jet', [1 1 1], 'shuffle');

The purpose of the "shuffle" is to ensure that the neighbors don't have a color gradient, and that the colors are shuffled for each of 200 objects within an image. When I track the object in multiple frames, I would like it to retain the same color it assigns to the object the first time it shuffles.

Answer 1

[Shashank Gupta]

Hi Veena,

I tried taking a series of images with some motion and without changing anything in the function, the watershed function was able to give me the expected results. It is retaining the colors in the image, if you still facing the issue, would you like to describe your case more in detail? I tried small piece of code, check this out.

for i=1:10
    center1 = -40;
    center2 = -center1;
    dist = sqrt(2*(2*center1)^2);
    radius = dist/2 * 1.4;
    lims = [floor(center1-1.2*radius) ceil(center2+1.2*radius)];
    [x,y] = meshgrid(lims(1):lims(2));
    bw1 = sqrt((x-center1-i*5).^2 + (y-center1-i*5).^2) <= radius-i*5;
    bw2 = sqrt((x-center2-i*5).^2 + (y-center2-i*5).^2) <= radius-i*5;
    bw3 = sqrt((x-center2-75).^2 + (y-center2-75).^2) <= radius-75;
    bw = bw1 | bw2 | bw3;
    % imshow(bw)
    % title('Binary Image with Overlapping Objects')
    D = bwdist(~bw);
    D = -D;
    L = watershed(D);
    L(~bw) = 0;
    rgb = label2rgb(L,'jet',[1 1 1],'shuffle');
    figure();imshow(rgb)
end

Cheers.