How to display the two biggest blobs in terms of area?

1 vue (au cours des 30 derniers jours)
sanyukta
sanyukta le 22 Juil 2019
Commenté : Image Analyst le 23 Juil 2019
I have a binary image 'i' which I have attached here. From this image, I want to find the two biggest blobs in terms of area. When I tried this code, I am able to find the biggest blob which I have attached and the next biggest blob is removed.
labeledImage=bwlabel(i);
measurements=regionprops(labeledImage,'Area');
allAreas=[measurements.Area];
[biggestArea, indexOfBiggest]=sort(allAreas,'descend')
biggestBlob=ismember(labeledImage,indexOfBiggest(1));
biggestBlob=biggestBlob>0;
figure,imshow(biggestBlob,[]);
I tried to use the bwareafilt() but MATLAB R2013a does not support it. Kindly help me to display the next biggest blob.
Thanking you.

Réponse acceptée

David K.
David K. le 22 Juil 2019
In the line
biggestBlob=ismember(labeledImage,indexOfBiggest(1));
you are finding the biggest by looking at the first index in a sorted array. So to get the second biggest you do
secondBlob=ismember(labeledImage,indexOfBiggest(2));
Then you can combine them simply by doing
twoBiggest = biggestBlob | secondBlob;
This is because both blobs are binary - 1s and 0s - so by doing an OR operation it makes every white part in both images seen in both images.
  1 commentaire
Image Analyst
Image Analyst le 23 Juil 2019
I think
twoBiggest = ismember(labeledImage, indexOfBiggest(1:2));
would be simpler. It gets you both of the two biggest blobs all in one line of code.

Connectez-vous pour commenter.

Plus de réponses (1)

Image Analyst
Image Analyst le 22 Juil 2019
It'w very bad practice to call your image i (the imaginary variable). Call it binaryImage, mask, or BW instead. You might want to call imclearborder() to get rid of that outer blob, unless you want it.
mask = imclearborder(mask);
I don't believe you need to use
biggestBlob=biggestBlob>0;
because ismember() will return a logical image already.
Other than that, David's solution should work.
  1 commentaire
sanyukta
sanyukta le 23 Juil 2019
Thank you so much for this valuable suggestion.

Connectez-vous pour commenter.

Catégories

En savoir plus sur Images dans Help Center et File Exchange

Produits


Version

R2013a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by