Apply 2-D logical mask to obtain masked third dimension values in 3-D array

This is difficult to describe in words so I have given an example below. I have a 3-D array. Some of the values are NaNs (although this is not directly relevant to my question). I want to apply a 2-D mask to the first two dimensions and arrange all the third dimensions behind the mask as sequential rows in a 2-D array. I'm looking for a way to do this without 'for' loops.

Here's the example:

% Set up the example 3-D array and mask

BW = logical([0,0,1; 0,1,1; 1,1,0; 1,0,0]); %create a mask

A = 9*ones(4,3,3); % create a 3 dimensional array
A(:,:,2) = 8*ones(4,3); % Make the second 'slice' 8s
A(:,:,3) = 7*ones(4,3); % Make the third 'slice' 8s
A(1,3,1) = NaN; % add a NaN
A(3,2,3) = NaN; % add another NaN
A(4,1,2) = NaN % add another NaN

A = 
A(:,:,1) =

     9     9   NaN
     9     9     9
     9     9     9
     9     9     9


A(:,:,2) =

     8     8     8
     8     8     8
     8     8     8
   NaN     8     8


A(:,:,3) =

     7     7     7
     7     7     7
     7   NaN     7
     7     7     7

% Apply the mask and organise the masked third dimension values into a 2-D array
my2Darray = zeros(1,size(A,3));

for row=1:(size(A,1))
    for col=1:(size(A,2))
        if (BW(row,col))
            mySpectrum = squeeze (A(row, col,:))';
            my2Darray = cat (1, my2Darray, mySpectrum);
        end
    end
end

my2Darray(1,:) = [] % Get rid of the first row of zeros

my2Darray = 6x3

NaN 8 7 9 8 7 9 8 7 9 8 7 9 8 NaN 9 NaN 7

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

The outcome is correct. The total number of rows in my2Darray should be equal to sum(sum(BW)).

But is there an elegant way to reshape this that doesn't require the 'for' loops, please?

Answer 1

[Hassaan]

% Define the 2D mask and 3D array with NaNs
BW = logical([0,0,1; 0,1,1; 1,1,0; 1,0,0]);
A = 9*ones(4,3,3);
A(:,:,2) = 8*ones(4,3); A(:,:,3) = 7*ones(4,3);
A(1,3,1) = NaN; A(3,2,3) = NaN; A(4,1,2) = NaN;

% Process the array
B = reshape(pagetranspose(A), [], size(A,3)); % Transpose and reshape to 2D array
mask = BW'; % Transpose the mask to match reshaped array
mask = mask(:); % Convert mask to a column vector
B(~mask, :) = []; % Apply the mask to filter the array

% Store the result in my2Darray
my2Darray = B;

% Display the final 2D array
disp(my2Darray);

   NaN     8     7
     9     8     7
     9     8     7
     9     8     7
     9     8   NaN
     9   NaN     7

------------------------------------------------------------------------------------------------------------------------------------------------

If you find the solution helpful and it resolves your issue, it would be greatly appreciated if you could accept the answer. Also, leaving an upvote and a comment are also wonderful ways to provide feedback.

Professional Interests

• Technical Services and Consulting
• Embedded Systems | Firmware Developement | Simulations
• Electrical and Electronics Engineering

Feel free to contact me.

[Steve Francis]

Thank you for a lightning response!

Unfortunately the final values in your my2Darray don't match the final values in mine. Probably a hiccup in the order of commands?

[Steve Francis]

Thanks for your edit, Hassaan. But didn't you just change your answer to copy my solution ? I'm not sure it has moved this discussion forward. Thank you for your interest, though.

[Hassaan]

@Steve Francis I used a slightly different approach didn't meant to plagiarise it. Anyhow a good solution on your end.

Answer 2

[Steve Francis]

(Can you answer your own question?)

This seems to work without the loops. Is there a better approach?

% Set up the example 3-D array and mask
BW = logical([0,0,1; 0,1,1; 1,1,0; 1,0,0]); %create a mask
A = 9*ones(4,3,3); % create a 3 dimensional array
A(:,:,2) = 8*ones(4,3); % Make the second 'slice' 8s
A(:,:,3) = 7*ones(4,3); % Make the third 'slice' 8s
A(1,3,1) = NaN; % add a NaN
A(3,2,3) = NaN; % add another NaN
A(4,1,2) = NaN; % add another NaN

B = pagetranspose(A); %transpose the first 2-D page and all subsequent pages
B = reshape(B, [], size(B,3)); % convert to 2-D array
mask=BW';
mask=mask(:); % mask is now a column vector
B(~mask,:) = []; % apply the mask
my2Darray=B

my2Darray = 6x3

NaN 8 7 9 8 7 9 8 7 9 8 7 9 8 NaN 9 NaN 7

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>