I am trying to smooth the temporal history of each pixel in my matrix- in other words, trying to smooth each pixel through both 'space' (mxn) and 'time'(third dimension). I am using the function movmean to create an average of each pixel in time of a 1000x1000x8 matrix.
I am currently using the following code to take an average, using a window size of 5, operating along the third dimension:
av_matrix = movmean(my_matrix,5,3)
This is creating an average as expected, but I'm wondering if the window is just operating in the mxn direction and not taking the average along the third dimension as well.
To compute a moving average along the n dimensions of an n-dimensional array (the "window" is an n-dimensional rectangle), the simplest way is to use convolution (see convn).
You need to be careful with edge effects, that is, when the convolution kernel (or n-dimensional window) partially slides out of the data. What movmean does is average over the actual data points only. To achieve that behaviour you can
compute the sum over the kernel via convolution with the 'same' option; and then
divide each entry by the number of actual data points from which it was computed. This number can also be obtaind via convolution, namely, applying the kernel to an array of ones.
So, all you need is:
my_matrix = randi(9,5,5,3); % example 3D array
sz = [3 3 2]; % 3D window size
av_matrix = convn(my_matrix, ones(sz), 'same') ... % step 1
./convn(ones(size(my_matrix)), ones(sz), 'same'); % step 2
Check:
The following examples use
>> my_matrix
my_matrix(:,:,1) =
6 8 2 1 8
4 6 7 9 8
4 5 1 4 3
5 5 8 7 9
3 6 6 4 9
my_matrix(:,:,2) =
8 8 5 3 6
8 9 6 9 1
9 5 6 2 2
1 7 4 1 2
5 4 7 4 9
my_matrix(:,:,3) =
6 5 8 6 6
1 6 8 6 1
5 5 1 6 7
1 1 2 9 8
1 2 6 1 2
With edge effects:
>> mean(mean(mean(my_matrix(1:2,1:2,1:2))))
ans =
7.125000000000000
>> av_matrix(1,1,1)
ans =
7.125000000000000
Without edge effects:
>> mean(mean(mean(my_matrix(1:3,1:3,1:2))))
ans =
5.944444444444445
>> av_matrix(2,2,1)
ans =
5.944444444444445
This is about matlab.
Let's say I have a matrix like this
A = [1,2,3,4,5;6,7,8,9,10;11,12,13,14,15]
Now I want to know how to get a mean value of a small matrix in A.
Like a mean of the matrix located upper left side [1,2;6,7]
The only way I could think of is cut out the part I want to get a value from like this
X = A(1:2,:);
XY = X(:,1:2);
and mean the values column wise Mcol = mean(XY);.
and finally get a mean of the part by meaning Mcol row-wise.
Mrow = mean(Mcol,2);
I don't think this is a smart way to do this so it would be great if someone helps me make it smarter and faster.
Your procedure is correct. Some small improvements are:
Get XY using indexing in a single step: XY = A(1:2, 1:2)
Replace the two calls to mean by a single one on the linearized submatrix: mean(XY(:)).
Avoid creating XY. In this case you can linearize using reshape as follows: mean(reshape(A(1:2, 1:2), 1, [])).
If you want to do this for all overlapping submatrices, im2col from the Image Processing Toolbox may be handy:
submatrix_size = [2 2];
A_sub = im2col(A, submatrix_size);
gives
A_sub =
1 6 2 7 3 8 4 9
6 11 7 12 8 13 9 14
2 7 3 8 4 9 5 10
7 12 8 13 9 14 10 15
that is, each column is one of the submatrices linearized. So now you only need mean(A_sub, 1) to get the means of all submatrices.
I am interested in sorting the columns of a matrix in terms of the values in 2 other vectors. As an example, suppose the matrix and vectors look like this:
M = [ 1 2 3 4 5 6 ;
7 8 9 10 11 12 ;
13 14 15 16 17 18 ]
v1 = [ 2 , 6 , 6 , 1 , 3 , 2 ]
v2 = [ 3 , 1 , 2 , 7 , 9 , 1 ]
I want to sort the columns of A in terms of their corresponding values in v1 and v2, with v1 taking precedence over v2. Additionally, I am interested in trying to sort the matrix in place as the matrices I am working with are very large. Currently, my crude solution looks like this:
MM = [ v1' ; v2' ; M ] ; ## concatenate the vectors with the matrix
MM[:,:] = sortcols(MM , by=x->(x[1],x[2]))
M[:,:] = MM[3:end,:]
which gives the desired result:
3x6 Array{Int64,2}:
4 6 1 5 2 3
10 12 7 11 8 9
16 18 13 17 14 15
Clearly my approach is not ideal is it requires computing and storing intermediate matrices. Is there a more efficient/elegant approach for sorting the columns of a matrix in terms of 2 other vectors? And can it be done in place to save memory?
Previously I have used sortperm for sorting an array in terms of the values stored in another vector. Is it possible to use sortperm with 2 vectors (and in-place)?
I would probably do it this way:
julia> cols = sort!([1:size(M,2);], by=i->(v1[i],v2[i]));
julia> M[:,cols]
3×6 Array{Int64,2}:
4 6 1 5 2 3
10 12 7 11 8 9
16 18 13 17 14 15
This should be pretty fast and uses only one temporary vector and one copy of the matrix. It's not fully in-place, but doing this operation completely in-place is not easy. You would need a sorting function that moves columns as it works, or alternatively a version of permute! that works on columns. You could start with the code for permute!! in combinatorics.jl and modify it to permute columns, reusing a single column-size temporary buffer.
This question already has an answer here:
Multiply 2D Matrix with vector to span third dimension - MATLAB
(1 answer)
Closed 7 years ago.
Here what I would like to achieve:
I have a matrix C
C=[1 2 3; 4 5 6; 7 8 9];
And a vector a
a=[1 2];
I would like to make such an operation, that each element of the a vector is multiplied with C ( scalar multiplication ) and out comes a 3-dimensional array D:
(:,:,1) =
1 2 3
4 5 6
7 8 9
(:,:,2) =
2 4 6
8 10 12
14 16 18
It would certainly work with a loop, but, since I'll need this operation on multiple occasions, a oneliner would be a great saver.
This is a beautiful example for the use of bsxfun and reshape. While #thewaywewalks proposes first calling bsxfun and reshaping the result, I'd suggest the opposite. This makes one of the key concepts of bsxfun - the singleton dimension expansion - more clear:
out = bsxfun(#times,C,reshape(a,1,1,[]))
ans(:,:,1) =
1 2 3
4 5 6
7 8 9
ans(:,:,2) =
2 4 6
8 10 12
14 16 18
With reshape(a,1,1,[]), you make a be in the third dimension. If you now apply bsxfun, it will multiply the matrix C with each element of a.
Some reshape'ing and some bsxfun will do:
out = reshape(bsxfun(#mtimes, C(:), a(:).'), [size(C),numel(a)] )
As suggested in hbaderts answer one could also use bsxfun's capability of dimension expansion, and provide a permuted vector of factors:
out = bsxfun(#mtimes,C,permute(a,[3,1,2]))
out(:,:,1) =
1 2 3
4 5 6
7 8 9
out(:,:,2) =
2 4 6
8 10 12
14 16 18
I have another method for the benchmark compare...
IMO it's the neatest way, at least for the syntax/readability term:
out = reshape(kron(a,C),[size(C),numel(a)]);
out(:,:,1) =
1 2 3
4 5 6
7 8 9
out(:,:,2) =
2 4 6
8 10 12
14 16 18
Another possibility is to use matrix multiplication of C as a column vector times a as a row vector (this gives all element-wise products), and then reshape the result:
out = reshape(C(:)*a, size(C,1), size(C,2), numel(a));
EDIT (BENCHMARKING): Since several solutions (including mine below) have been suggested, here is some rough benchmarking to compare the different solutions, using larger arrays:
a=1:10;
N=1000; timers=zeros(N,6);
for ii=1:N; C=rand(400);
tic; out = repmat(C,[1,1,numel(a)]).*reshape(repelem(a,size(C,1),size(C,2)),[size(C),numel(a)]); timers(ii,1)=toc;
tic; out = bsxfun(#times,C,reshape(a,1,1,[])); timers(ii,2)=toc;
tic; out = reshape(C(:)*a, size(C,1), size(C,2), numel(a)); timers(ii,3)=toc;
tic; out = bsxfun(#mtimes,C,permute(a,[3,1,2])); timers(ii,4)=toc;
tic; out = reshape(bsxfun(#mtimes, C(:), a(:).'), [size(C),numel(a)] ); timers(ii,5)=toc;
tic; out = reshape(kron(a,C),[size(C),numel(a)]); timers(ii,6)=toc;
end;
mean(timers)
ans =
0.0080863 0.0032406 0.0041718 0.015166 0.0074462 0.0033051
... suggesting that #hbaderts solution is fastest, then #Adiel's, then #Luis Mendo's, then #thewaywewalk's (1), then mine, then #thewaywewalk's (2).
My solution:
Another option, using repmat and reshape (no bsxfun):
out = repmat(C,[1,1,numel(a)]).*reshape(repelem(a,size(C,1),size(C,2)),[size(C),numel(a)])
out(:,:,1) =
1 2 3
4 5 6
7 8 9
out(:,:,2) =
2 4 6
8 10 12
14 16 18
This is the element-wise multiplication of two arrays. The first is your original matrix C repeated numel(a) times in the third dimension:
repmat(C,[1,1,numel(a)])
ans(:,:,1) =
1 2 3
4 5 6
7 8 9
ans(:,:,2) =
1 2 3
4 5 6
7 8 9
The second is the same size as the first, with each slice containing the corresponding element of a:
reshape(repelem(a,size(C,1),size(C,2)),[size(C),numel(a)])
ans(:,:,1) =
1 1 1
1 1 1
1 1 1
ans(:,:,2) =
2 2 2
2 2 2
2 2 2
the easiest way for me to explain what i want is with an example:
a = 1:20
b = [2,7,12,18]
Now I want c to be [1,3,4,5,6,8,...,19,20] with length 16: length(a) - length(b) of course.
Is there a way for me to get c?
You can delete array elements using x(3)=[]
c=a;
c(b)=[];
What you want is called set difference in most languages. In MATLAB, you can use the setdiff function:
a=1:20;
>> b=[2,7,12,18];
>> setdiff(a,b);
ans =
Columns 1 through 11
1 3 4 5 6 8 9 10 11 13 14
Columns 12 through 16
15 16 17 19 20