sum values of histogram bins

Hi, I have data on particle intensities. I am binning the particles into bins based upon the intensities. This code is working.

histBins=[10, 25, 50, 75, 100, 150,  200];
a=hist(d,histBins);

I will now like to sum up the intensity of all the particles in each bin. Is there a way to do this?

Thanks, John

Answer 1

[Sean de Wolski]

Building on The Cyclist's answer, use accumarray to do the summing:

x = randn(100,1);
edges = -10:10;
[~, idx] = histc(x,edges);
binsums = accumarray(idx,x)

Answer 2

[Image Analyst]

No need to mess with histograms, and resulting loss of precision. If you want "sum up the intensity of all the particles", just sum the particles themselves, not the histogram bins.

sumOfAllParticleIntensities = sum(d);

If you want the sums of all particle intensities bin by bin between the bin edges that you gave, then you can do this:

histBins=[10, 25, 50, 75, 100, 150,  200];
x = 200*randn(100,1);
for k = 1 : length(histBins)-1
  indexes = x > histBins(k) & x <= histBins(k+1);
  binByBinSums(k) = sum(x(indexes));
end
% Print to command window:
binByBinSums

This will be more accurate than doing anything based on counts and the values of the bin centers.

Answer 3

[the cyclist]

If you change your code to use the histc() command instead of the hist() command, then you can do

[count,idx] = histc()

will give you the index to which bin each element was sorted into. You can use that index sum the intensities in each bin.

[histc() specifies the bins based on edges, rather than centers.]