How to resize a group of ones in an array of zeros?

If i have an array of zeors (512,512) with a couple of ones in there making up a shape, how do i rezise that group of ones making that shape either smaller or larger by a inputted factor

ex

0 0 0 0 0

0 1 1 1 0

0 0 1 1 0

0 0 0 0 0

0 0 0 0 0

i though of doing a for loop changing (i,j+1), (i,j-1),(i+1,j),(i-1,j) to a value of one for (1: the number inputted to resize). However the problem with this is that this doesnt correctly resize the ones image, for an example, inputting a 2 doesnt result in a doubled image. Also this doesnt work to decrease the size of the ones image.

[Adam Danz]

It would be helpful if you provided an expected output to your input example and then supply several more input/output pairs.

It would also be useful if you constrained the problem a bit.

• Is there just 1 pattern of 1s within the matrix or several disconnected patterns of 1s?
• How should the pattern be expanded/contracted?
• Does the matrix of 0s change size, too? If not, what happens when the 1s become larger than the 0s?
• Factors are usually multiplicative. Is the factor always a positive integer? If not, how should we treat rounding?

[Mohammad Saleh]

A real example would be

DT=zeros(512 ,512);
DT(256,256)=1;
DT(256,257)=1;
DT(256,258)=1;
DT(257,257)=1;
DT(257,258)=1;

So i would want to be able to increase or decrease the size which the ones make by using an input such that an input of 2 would double it, 3 would tripple it, -2 would decrease the size by half, etc. Also, i do not want to change the size of the matrix, but only to expand or decrease the shape of the ones by overtaking more zeros with ones, or replacing ones with zeros.

[Adam Danz]

That's a little better but it's still not clear how you'd like to expand/decrease the shape.

Again, a few (more than 1) input & expected output pairs (with much smaller matrices) would be more grounded and would leave less of the burdon of interpretation up to us. Two volunteers have already invested time into answers that won't work because the question was unclear. It would be unwise to invest more time without a clear definition of how the expansion/contraction would work.

Answer 1

[KALYAN ACHARJYA]

If i have an array of zeors (512,512) with a couple of ones in there making up a shape, how do i rezise that group of ones making that shape either smaller or larger by a inputted factor

data=[0 0 0 0 0
0 1 1 1 0
0 0 1 1 0
0 0 0 0 0
0 0 0 0 0]
result=imresize(data,2)
%....................^ times larger, here double

Command Window:

data =

     0     0     0     0     0
     0     1     1     1     0
     0     0     1     1     0
     0     0     0     0     0
     0     0     0     0     0
result =
    0.0088   -0.0190   -0.0747   -0.1003   -0.0959   -0.0959   -0.1003   -0.0747   -0.0190    0.0088
   -0.0212    0.0466    0.1822    0.2377    0.2132    0.2062    0.2169    0.1619    0.0413   -0.0190
   -0.0813    0.1778    0.6960    0.9139    0.8315    0.8106    0.8513    0.6350    0.1619   -0.0747
   -0.0813    0.1708    0.6751    0.9742    1.0681    1.1353    1.1760    0.8716    0.2222   -0.1025
   -0.0212    0.0257    0.1196    0.4186    0.9229    1.1804    1.1910    0.8716    0.2222   -0.1025
    0.0066   -0.0346   -0.1170    0.1009    0.6191    0.8765    0.8732    0.6350    0.1619   -0.0747
    0.0022   -0.0101   -0.0346    0.0209    0.1566    0.2238    0.2227    0.1619    0.0413   -0.0190
         0    0.0016    0.0049   -0.0143   -0.0560   -0.0769   -0.0769   -0.0560   -0.0143    0.0066
         0    0.0005    0.0016   -0.0048   -0.0187   -0.0256   -0.0256   -0.0187   -0.0048    0.0022
         0         0         0         0         0         0         0         0         0         0

Answer 2

[Adam Danz]

Maybe interp2() would help.

% Input data
m = [0 0 0 0 0
    0 1 1 1 0
    0 0 1 1 0
    0 0 0 0 0
    0 0 0 0 0];
% Interpolate
mInt = double(interp2(m,1) > 0); 
%                       ^ this value (k) determines interp resolution. 
%                       There will be 2^k-1 number of data points added
%                       between each sample.
% compare original vs interp'd
figure()
subplot(1,2,1)
heatmap(m)
title('Original data')

subplot(1,2,2)
heatmap(mInt)
title('Interp''d data')