How to split a matrix in "submatrices"?

How can i divide a matrix into submatrices of given dimensions in automatic way?
example:
data=reshape(1:88,11,8); %create a matrix
[mpixel,npixel]=size(data); %matrix's dimensions
m=2; %rows of the submatrix
n=3; %columns of the submatrix
v_w=floor(mpixel/m); %number of "vertical" submatrices
h_w=floor(npixel/n); %number of "horizontal" submatrices
cropdata=data(1:v_w*m,1:h_w*n);%crop matrix that can be diveded using that mxn submatrices
%
newdata=mat2cell(cropdata,[m m m m m],[n n]);
This code work but if I have bigger matrix I must continue to explicit [m m m...],[n n n...] in the last command and it's crazy. How can i use the command mat2cell with a smarter way?

4 comentarios

Adam
Adam el 12 de Dic. de 2017
Why do you want to break your data up rather than just indexing into it and leaving it in one neat array?
Francesco Boari
Francesco Boari el 12 de Dic. de 2017
Editada: Francesco Boari el 12 de Dic. de 2017
Cause I need to have all the small matrices for other processing, in authomatic way. I can't call each time a part of the original matrix. The problem is a little bit more complicated then that in the example.
Image Analyst
Image Analyst el 12 de Dic. de 2017
Why not just use blockproc() and let that handle the getting of the submatrices?
Francesco Boari
Francesco Boari el 13 de Dic. de 2017
I see the documentation of blockproc(), it processes block by block but at the end the output it's the input image/matrix.

Iniciar sesión para comentar.

 Respuesta aceptada

Guillaume
Guillaume el 13 de Dic. de 2017
As others have said, there may be no need to explicitly split the matrix, but if you must what is wrong with:
newdata = mat2cell(cropdata, repelem(m, v_w), repelem(n, h_w));

1 comentario

Francesco Boari
Francesco Boari el 13 de Dic. de 2017
Thanks! It works. Meanwhile i have found a different way leaving, as someone said, the original matrix and indexing.

Iniciar sesión para comentar.

Más respuestas (0)

Categorías

Etiquetas

Preguntada:

el 12 de Dic. de 2017

Comentada:

el 13 de Dic. de 2017

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by