How do I plot the nearest neighbour algorithm in a matrix?

Hi,

X = [0 0 0 0 0; 0 0 1 0 0; 0 0 0 0 1]

I would like to produce a coloured grid that displays the matrix elements equal to one as red. Yellow for the elements at a neighbourhood of 8 connectivity to the non-zero elements and green for the elements further away from 1.

Should I be using brute force or knnsearch or is there a better/ faster way to go about it?

Thanks

[Walter Roberson]

.... Then don't use that code.

You have forgotten that chessboard distance does not stop at distance 1 or 2.

Answer 1

[Walter Roberson]

d = bwdist(X);
t = ones(size(X));
t(d < 1.5) = 3;
t(d == 0) = 2;
cmap = [0 1 0;
        1 0 0;
        1 1 0];
image(t);
axis image
colormap(cmap)

[Image Analyst]

[Walter Roberson]

Blue, green, what's the difference? ;-)

[Walter Roberson]

Another approach:

t = imdilate(X, ones(3,3)) + 1;
t(X==1) = 3;
cmap = [0 1 0;
        1 1 0;
        1 0 0];
image(t)
axis equal
colormap(cmap)

Answer 2

[Chad Greene]

How about bwlabel? Then you can use imagesc to display the matrix and set the colormap to red, yellow, and green by

colormap([0.9 0 0;1 1 0.08; 0.08 0.69 0.1])

[Walter Roberson]

bwlabel is not enough in this case to figure out whether you are beside a 1.