Find array elements that meet a condition an put them in a secondary array

Hello all, I have a 3D matrix A. For example suppose that A is as follows:

A(:,:,1) =
  0.1835    0.0811    0.4359
  0.3685    0.9294    0.4468
  0.6256    0.7757    0.3063
  0.7802    0.4868    0.5085
A(:,:,2) =

    0.5108    0.3786    0.9390
    0.8176    0.8116    0.8759
    0.7948    0.5328    0.5502
    0.6443    0.3507    0.6225

I want to find the elements in A that are >0.5 and put them in an array B. Each row of array B should be [i,j,k,value]. In above example my array B would be like:

    3 1 1 0.6256 
    4 1 1 0.7802
    2 2 1 0.9294
    3 2 1 0.7757
    ...

After this I want to print array B as it is in a text file ('Example.txt'). Thank you for your help.

Answer 1

[Star Strider]

The find and ind2sub will do exactly what you want:

A(:,:,1) = [0.1835    0.0811    0.4359
            0.3685    0.9294    0.4468
            0.6256    0.7757    0.3063
            0.7802    0.4868    0.5085];

A(:,:,2) = [0.5108    0.3786    0.9390
            0.8176    0.8116    0.8759
            0.7948    0.5328    0.5502
            0.6443    0.3507    0.6225];

idx = find(A > 0.5);
[i,j,k] = ind2sub(size(A), idx);

B = [i j k A(idx)]

B =
            3            1            1       0.6256
            4            1            1       0.7802
            2            2            1       0.9294
            3            2            1       0.7757
            4            3            1       0.5085
            1            1            2       0.5108
            2            1            2       0.8176
            3            1            2       0.7948
            4            1            2       0.6443
            2            2            2       0.8116
            3            2            2       0.5328
            1            3            2        0.939
            2            3            2       0.8759
            3            3            2       0.5502
            4            3            2       0.6225

Answer 2

[Image Analyst]

Here are two different ways, a for loop way, and a vectorized way:

A(:,:,1) =[...
  0.1835    0.0811    0.4359
  0.3685    0.9294    0.4468
  0.6256    0.7757    0.3063
  0.7802    0.4868    0.5085];
A(:,:,2) =[...
    0.5108    0.3786    0.9390
    0.8176    0.8116    0.8759
    0.7948    0.5328    0.5502
    0.6443    0.3507    0.6225]

% In tuitive brute force "for" loop way:
numRows = nnz(A(:)>0.5);
out = zeros(numRows, 4);
currentRow = 1
for z = 1 : size(A, 3)
  for row = 1 : size(A, 1)
    for col = 1 : size(A, 2)
      if A(row, col, z) > 0.5
        out(currentRow, 1:4) = [row, col, z, A(row, col, z)];
        currentRow = currentRow + 1;
      end
    end
  end
end
out

% Vectorized way:
bigA = A(A(:)>0.5);
actualSubscripts = find(A(:) > 0.5)
[row, col, z] = ind2sub(size(A), actualSubscripts)
out = [row, col, z, bigA]

Answer 3

[Azzi Abdelmalek]

ll=find(A>0.5)
[ii,jj,kk]=ind2sub(size(A),ll)
out=[ii,jj,kk,A(ll)]

Answer 4

[Andrei Bobrov]

t = A > .5;
[m,n,~] = size(A);
[ii,jj] = find(t);
B = [ii,rem(jj-1,n)+1,ceil(jj/n),A(t)];

Answer 5

[H R]

Thank you for your help.

As a secondary filtering on B, I would like to only retain the rows when the first element of any row in B >2, and the second element of B in any row <3 and put them in a new array C. In other words, my array C has the following properties [i>2,j<3,k, value]

[Star Strider]

This seems to work (continuing my previous code):

Crows = (B(:,1)>2) & (B(:,2)<3);                        % Logical Vector: 1=Meets Criteria 
C = [B(Crows,1) B(Crows,2) k(Crows) B(Crows,4)]         % New Filtered Matrix

C =
            3            1            1       0.6256
            4            1            1       0.7802
            3            2            1       0.7757
            3            1            2       0.7948
            4            1            2       0.6443
            3            2            2       0.5328