RANDBLOCK

scramble matrices using blocks (v2.3, mar 2011)

RANDBLOCK - randomize blocks of a matrix

R = RANDBLOCK(M,S) randomizes the matrix M by dividing M into non-overlapping blocks of the size specified by S, and shuffling these blocks. M can be a N-D matrix.

The number of elements in S should match the number of dimensions of M, or S can be a scalar specifying a S-by-S-by-S-by ... block size. S should contain positive integers. The size of M in any dimension should be an integer number of times the specified size of the block in that dimension (e.g., if size(M,1) equals 6, S(dim) can be 1,2,3, or 6).

[R,I,J] = RANDBLOCK(...) also returns indices I and J, so that R equals A(I)and R(J) equals A.

M can be a numerical or cell array.

Examples:
% Shuffle blocks of 3 elements of a 15-element vector
M = 1:15 ;
randblock(M,3)

% Scramble a 2D matrix
M = reshape(1:24,4,[]) ;
randblock(A,[3 2]) % randomize the position of the four 3-by-2 blocks
randblock(A,2) % randomize the position of the six 2-by-2 blocks

% Scramble a 3D volume
M = reshape(1:64,[4 8 2]) ;
randblock(A,[2 4 2]) % randomize the position of the four
% 2-by-4-by-2 sub-volumes
% Scramble a cell matrix
M = {'1','a','bb','1c' ; '2a',[3 4 5],'2c','2dd'} ;
randblock(M,[2 2]) % randomize the position of the four 2-by-2 blocks

% Scramble a RGB image Z, and retrieve the original using the
% indices (see the SCREENSHOT)
Z = peaks(200) ; Z = cat(3,Z,flipud(Z), -(circshift(Z.',[100,100]))) ;
Z = (Z - min(Z(:))) ; Z = Z./ max(Z(:)) ;
[Z2,I,J] = randblock(Z,[25,25,size(Z,3)]) ; % Scramble 25-by-25 blocks
subplot(2,2,1) ; image(Z) ; title('Original image') ;
subplot(2,2,2) ; image(Z2) ; title('25x25 scrambled image') ;
subplot(2,2,3) ; image(Z2(J)) ; title('Z2(J) = original image') ;
subplot(2,2,4) ; image(Z(I)) ; title('Z(I) = scrambled image') ;
set (gcf,'Name','[Z2,I,J] = randblock(Z,[25,25,size(Z,3)])') ;

Version 2.3 (mar 2011)