Largest connected component(circular)

I need an algorithm to detect the largest connected component which has circular shape in an image. I hope your kind help.

Thank you.

Answer 1

[Sean de Wolski]

Ibw = imfill(imread('coins.png')>100,'holes'); %example binary image
CC = bwconncomp(Ibw); %connected components
Objsz = cellfun(@numel,CC.PixelIdxList);  %size of blobs
[garbage, ObjszOrder] = sort(Objsz,'descend'); %descended sorting
stats = regionprops(CC,'stuff you need to write');
for ii = 1:length(ObjszOrder)
  if(iscircularenough(stats(ii)))
      %object ii wins!
      break
  end
end

I've done most of the work for you here. You have to determine what criteria you want to use to test circularity and what you need to extract from regionprops for this. The most common measurement of circularity is perimeter to area. Use wikipedia or the explanation of 'regionprops:equivdiameter' for the correct equation.

[nayomi ranamuka]

metric = 4*pi*area/perimeter^2.

This metric is equal to one only for a circle and it is less than one for any other shape. So I used this property to determine whether a component of image is circular or not. Thanks for your help.