Matlab “Scale Down” a Vector with Averages

问题

Quite hard to explain what I am looking for, I have an image represented as a m by n matrix in Matlab and I am trying to scale it down to 4x4 the same way an image would be scaled (average the nearest values)

So for example

1 2 3 4 5 6 7 8
1 2 3 4 5 6 7 8
0 2 3 4 9 9 7 8
0 2 3 4 9 9 7 8
1 2 3 4 5 6 7 8
1 2 3 4 5 6 7 8
1 2 3 4 5 6 7 7
1 2 3 4 5 6 7 7

Would become

1.5 3.5 5.5 7.5
1.0 3.5 9.0 7.5
1.5 3.5 5.5 7.5
1.5 3.5 5.5 7.0

回答1:

Looks like imresize gives something slightly different from what you expected. For your input data, the following will work:

A = filter2([1 1; 1 1] / 4, X, 'same')
A = A(1:2:end, 1:2:end);

EDIT: Actually, it's probably faster to do the following:

i = 1:2:size(A,1)-1;
j = 1:2:size(A,2)-1;

B = 0.25 * (A(i,j) + A(i+1,j) + A(i,j+1) + A(i+1,j+1));

回答2:

Any reason why you can't use the imresize() function? It does exactly what you want..

I =

 1 2 3 4 5 6 7 8
 1 2 3 4 5 6 7 8
 0 2 3 4 5 6 7 8
 0 2 3 4 5 6 7 8
 1 2 3 4 5 6 7 8
 1 2 3 4 5 6 7 8
 1 2 3 4 5 6 7 8
 1 2 3 4 5 6 7 8



I2 = imresize(I,[4 4])

I2 =

1.4688    3.4919    5.5117    7.4922
1.0742    3.5289    5.5117    7.4922
1.4688    3.4919    5.5117    7.4922
1.5137    3.4877    5.5117    7.4922

回答3:

If you don't have the imresize.m from the Image Processing Toolbox, you can see how octave does it here (it uses convolution).

来源：https://stackoverflow.com/questions/7741808/matlab-scale-down-a-vector-with-averages