Count amount of contours in images

I am struggling to find a good contour detection function that would count the number of contour in bw images that I have processed using some previous tools. I have uploaded an example of such image. In this image, ideally, I wish to have a function which counts four closed contour.

I don't mind if it also detects the really tiny contours in between, or the entire shape itself as extra contours. As long as it counts the medium sized ones, I can fix the rest by applying area threshold. My problem is that any function I have tried detects only one contour - the entire shape, as it cannot separate it to the su-conours which are connected to one another.

Any suggestions?

Answer 1

[Image Analyst]

Very very easy

% Invert and get rid of the border.
binaryImage = imclearborder(binaryImage);
% Count the number of blobs:
[labeledImage, numberOfBlobs] = bwlabel(binaryImage);

Actually it looks like there are around 10 or so. If you want to remove blobs smaller than, say 500 pixels, add a call to bwareaopen():

% Invert and get rid of the border.
binaryImage = imclearborder(binaryImage);
% Get rid of blobs smaller than 500 pixels.
binaryImage = bwareaopen(binaryImage, 500);
% Count the number of blobs:
[labeledImage, numberOfBlobs] = bwlabel(binaryImage);

[Guy Talmor]

Thanks for your quick answer. I don't want to get rid of the borders, why should I? I don't want to create one closed contour out of this. And anyway, it didn't change the image at all even when I tried it (with 8 pixels connectivity). I also tried bwlabel and it counted only one blob - the whole structure which is exactly my problem.

What I would like to have is a function which recognizes that there are several closed sub-contours in the the image that share the same edges (overlap). This overlapping is essentially the problem. If they were all separated I could have easily used bwconncomp and this would have counted the number of separated connected components (just as the blob detection does).

Any other thoughts?

[Image Analyst]

You need to get rid of the border because the black background, when inverted, becomes white, and you don't want that to be considered as a blob.

Alright, it looks like I'll have to program up a full blown demo for you. Unfortunately you didn't post a nice binary image - you posted a full color image with some extra stuff surrounding it. I'll have to crop then convert to gray scale, then threshold. Give me a few minutes.

[Image Analyst]

OK, here's the full blown demo. Try the attached script. Change the file name and folder if necessary.

If it solves your problem, then please mark the answer as Accepted. Thanks in advance.

[Guy Talmor]

Excellent! it works.

Thank you

[Siti Aisyah]

Could you help me on how to do the filtering on Figure A like how the person do in figure B? the are that I want is the one i highlighted in figure C. I have set of images as in figure A with different size and area of the interested region