Detect largest bounding box and applying centroid.

I have written this code to use bounding box and centroid properties. Due to noise of the image i am getting many unwated boxes and centroids. so i want to edit this code so that only the biggest bounded box is shown and centroid is shown only on that box. i need help in this regards.

binaryY = im2bw(rgb2gray(F));
   Ibw1 = imfill(binaryY,'holes');
K1 = medfilt2(Ibw1);
subplot(2,2,2);
imshow(K1, []);
%centrod
Ilabel1 = bwlabel(K1);
stat = regionprops(Ilabel1,'centroid');
subplot(2,2,3);
imshow(Im1); hold on;
for x = 1: numel(stat)
    plot(stat(x).Centroid(1),stat(x).Centroid(2),'r+');
end
  % Make sure image is not artificially stretched because of screen's aspect ratio.
[L,num]=bwlabel(K1);
bboxes=regionprops(L,'BoundingBox');
subplot(2,2,4)

imshow(Im1),title('Bounding Box');
hold on;
for k=1:length(bboxes)
    CurrBB=bboxes(k).BoundingBox;
    rectangle('position',[CurrBB(1),CurrBB(2),CurrBB(3),CurrBB(4)],'EdgeColor','g','LineWidth',2);
end
hold off;

[Walter Roberson]

regionprops the Area as well. max [stat.Area] to find the index of the largest .

[Sarmad Paracha]

Thank you for your response .

but using regionprops for the area i can find max area of the blobs but how can i use that to remove all the other boxes and centroids... Please help .A sample may help me alot. Thanks

Answer 1

[Image Analyst]

Near the top, do this:

Ibw1 = bwareafilt(Ibw1, 1); % Extract largest blob only.

Then continue with the rest of your code, though I don't believe the median filter is needed.

[Sarmad Paracha]

Oh man Thank You very much.... That solved my problem ...