Measuring Integrated Intensity using regioprops

Hi. I have an image that I threshold to create a binary image.

I then use regioprops to create regions

   labeledImage = bwlabel(Binary, 8);     % Label each blob so we can make measurements of it
   blobMeasurements = regionprops(labeledImage, ROI, 'all');     %ROI is original image

I undestand I can obtain the max or mean intesnities of each region such a s:

   allBlobIntensities = [blobMeasurements.MaxIntensity];

However, I need to calculate the integrated intensity of each region (and then take the mean of all of these).

I don't see a way to do it using regionprops as there is no

blobMeasurements.IntegratedIntensity

Any suggestions?

thanks Jason

[Massimo Zanetti]

What is exactly the "integrated intensity"? Do you have any formula?

[Jason]

Hi, its the sum of all pixels values within a a region. I then want to take the average over all regions.

[Youtao Liu]

Hi Jason, may I ask you how did you circle those blobs with varied line? Thanks!

[Jason]

Hi, I used this:

%DrawRegions from Binary to raw image:
    labeledImage = bwlabel(Binary, 8);     % Label each blob so we can make measurements of it
    blobMeasurements = regionprops(labeledImage, ROI, 'all');     %ROI is original image
    numberOfBlobs = size(blobMeasurements, 1);
    hold on;
    boundaries = bwboundaries(Binary);  
    numberOfBoundaries = size(boundaries);
    subplot(4,5,[3,4,8,9])
        hold on
    for k = 1 : numberOfBoundaries
  thisBoundary = boundaries{k};
  plot(thisBoundary(:,2), thisBoundary(:,1), 'g', 'LineWidth', 1);
    end
    hold off;

[Youtao Liu]

Hi, Jason, thanks a lot! appreciated!

Answer 1

[Image Analyst]

Ask for PixelValues to get you the values of every pixel in the blob. Then sum them up to get your values. Something like

[labeledImage, numBlobs] = bwlabel(binaryImage);
props = regionprops(labeledImage, grayImage, 'PixelValues');
for k = 1 : numBlobs
    thisBlobsValues = props(k).PixelValues;
    integratedGrayValues(k) = sum(thisBlobsValues);
end

where grayImage is the original image that you want to get the intensities (gray levels) from.

Answer 2

[Massimo Zanetti]

You may ask the REGIONPROP function to provide you the pixel indexes of the regions detected by the algorithm. Then you can use such indexes to do whatever you want in the target image (including summing the values in correspondence of the regions). See below.

%binary image containing blobs (here rectangles for simplicty)
labelImg = zeros(30,30);
labelImg(12:18,4:9) = 1;
labelImg(21:26,16:24) = 5;
%region props (optional argument to get pixel idx in a list)
meas = regionprops(logical(labelImg),'PixelIdxList');
%get a list which elements are vectors of linear indexes (each vector
%contains indexes of the connected components)
indexes = {meas.PixelIdxList};
numComp = numel(indexes);
intDens = zeros(numComp,1);
for k = 1:numComp
    %sum the pixel values of each component (integrated density)
    intDens(k) = sum(labelImg(indexes{k}));
end