# Calculate the mean of two dimensions in a 3D array

120 views (last 30 days)
Mar on 4 May 2018
Commented: Rohit shaw on 10 Oct 2021
I have a 3D array and I want to calculate the mean of the first two dimensions along the 3rd dimension but I am getting the following error
"Assignment has more non-singleton rhs dimensions than non-singleton subscripts".
My code looks as follows
MeanMatrix = zeros(1,1,30);
ImageROI = data(50:55,100:105,:);
for i=1:30
MeanMatrix(:,:,i) = mean(ImageROI(:,:,i));
end

Guillaume on 4 May 2018
Edited: Guillaume on 4 May 2018
mean on a 2D matrix (your imageROI(:, :, i)) will return a vector. You can't assign a vector to a scalar (which is what MeanMatrix(:, :, i) is since it has 1 row and 1 column).
In any case, the loop is completely unnecessary:
MeanMatrix = mean(mean(ImageROI, 1), 2);
Rohit shaw on 10 Oct 2021
Hello, could you kindly tell me what would the command be if I have to use this command across 86 timesteps. I have a 3d array of 720x360x86. These are temperature values for the world.

KL on 4 May 2018
Edited: KL on 4 May 2018
If you want the mean of all rows in every page(3rd dimension)
mean(ImageROI,1)
for columns in every page
mean(ImageROI,2)
and mean of pages,
mean(ImageROI,3)
if you're looking to calculate the mean of all elements in a page, mean2 would be something
mean2(ImageROI(:,:,pageNo)) %with a loop
without a loop, probably with reshape,
mean(reshape(ImageROI,[],size(ImageROI,2),1))