MATLAB: Bwarea & area, white/black ratio

bwarearegionpropswhite/black ratio

Hi,

Can someone please explain me what is the difference of bwarea & area.

I have an image which is black and white (after segmentation and fill) & I wish to calculate the area of an object (as I understand done by bwarea) relative to the image area (imageArea = regionprops(image_name, 'Area')), more or less calculating the white/black ratio.

Any other ideas is welcome.

Thanks.

Best Answer

They use different algorithms. regionprops gives the number of pixels while bwarea gives some kind of weighted area that depends on the shape of the boundary. This is described in the documentation for bwarea

The white to black ratio is

totalPixels = numel(binaryImages);
numWhitePixels = sum(binaryImage(:));
numBlackPixels = totalPixels - numWhitePixels;
ratio = numWhitePixels  / numBlackPixels;

Related Solutions

MATLAB: How to calculate the white area pixels amount

Try this:

numWhitePixels = nnz(binaryImage);

Other ways

numWhitePixels = sum(binaryImage(:));
whiteArea = bwarea(binaryImage);

MATLAB: Filling the region within objects with white

I don't know exactly what "fill the black region in the region of Interests" means. If you want to fill the black regions with black, thus making the white "islands" or "holes" black also (essentially making them disappear), then use bwareafilt() and specify the smallest white blob that you're willing to accept. For example, if you want to remove all blobs smaller than 1000 pixels, do this:

binaryImage = bwareafilt(binaryImage, [1000, inf]);

You could also use bwareaopen().

binaryImage = bwareaopen(binaryImage, 1000);

On the other hand, if you want to remove black holes in white regions, simply use imfill:

binaryImage = imfill(binaryImage, 'holes');

I also don't know what the "region of interest" is in that image. Can you indicate it, perhaps by outlining it in red? If the ROI is only part of the whole image, then you'll need to make a mask somehow, perhaps with poly2mask() or some other method depending on how ROI is defined, then use the mask to erase the non-ROI parts of the image.

binaryImage(~ROIMask) = false; % Erase outside of ROI region(s)

Best Answer

Related Solutions

MATLAB: How to calculate the white area pixels amount

MATLAB: Filling the region within objects with white

Related Question