[Ask Help] - Erode/Erosion Matrix

I've script and matrix 20x20 below, and i would like to search Erode/Erosion.

a = [1 1 1 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0;

1 1 0 0 1 0 1 1 1 0 1 1 1 0 0 1 1 1 0 0;

1 1 1 1 1 1 0 1 1 0 0 1 0 0 0 0 1 0 0 0;

0 0 0 1 1 0 1 1 1 0 1 0 1 0 0 0 0 0 0 0;

0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 0 1 1 1;

0 0 1 1 1 0 1 1 1 0 1 0 0 0 1 0 0 0 1 1;

1 0 0 0 1 0 1 1 0 1 1 1 0 1 1 1 0 1 0 1;

1 1 0 1 1 1 1 0 1 0 1 0 0 0 1 0 0 0 0 1;

1 0 0 0 1 0 0 1 0 0 1 0 0 0 0 0 1 1 1 0;

0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 0 1 0 0;

0 0 0 0 0 1 0 1 0 0 0 1 1 1 0 0 1 1 1 0;

0 1 1 0 1 0 1 1 0 0 0 1 1 1 1 0 1 1 1 0;

1 0 0 1 0 0 0 0 1 1 1 0 0 1 0 0 0 0 0 0;

1 1 1 0 0 0 0 0 1 1 1 0 1 1 1 0 0 0 0 0;

0 1 1 1 0 1 0 0 1 1 1 0 0 1 1 0 0 1 1 1;

0 1 1 0 0 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1;

0 0 0 0 1 1 1 0 0 0 1 0 0 0 1 0 1 1 1 1;

1 1 0 0 0 1 1 0 0 1 1 1 0 0 0 1 0 0 1 0;

1 1 1 0 0 1 0 0 0 1 1 1 0 0 1 1 1 0 0 0;

1 1 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 0];

b = [1 1 1;

0 1 0;

1 1 1];

What should i do?

Thanks....

[腾龙公司开户网址]

我弄了一下也搞不出来

Answer 1

[Walter Roberson]

a = [1 1 1 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0;
       1 1 0 0 1 0 1 1 1 0 1 1 1 0 0 1 1 1 0 0;
       1 1 1 1 1 1 0 1 1 0 0 1 0 0 0 0 1 0 0 0;
       0 0 0 1 1 0 1 1 1 0 1 0 1 0 0 0 0 0 0 0;
       0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 0 1 1 1;
       0 0 1 1 1 0 1 1 1 0 1 0 0 0 1 0 0 0 1 1;
       1 0 0 0 1 0 1 1 0 1 1 1 0 1 1 1 0 1 0 1;
       1 1 0 1 1 1 1 0 1 0 1 0 0 0 1 0 0 0 0 1;
       1 0 0 0 1 0 0 1 0 0 1 0 0 0 0 0 1 1 1 0;
       0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 0 1 0 0;
       0 0 0 0 0 1 0 1 0 0 0 1 1 1 0 0 1 1 1 0;
       0 1 1 0 1 0 1 1 0 0 0 1 1 1 1 0 1 1 1 0;
       1 0 0 1 0 0 0 0 1 1 1 0 0 1 0 0 0 0 0 0;
       1 1 1 0 0 0 0 0 1 1 1 0 1 1 1 0 0 0 0 0;
       0 1 1 1 0 1 0 0 1 1 1 0 0 1 1 0 0 1 1 1;
       0 1 1 0 0 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1;
       0 0 0 0 1 1 1 0 0 0 1 0 0 0 1 0 1 1 1 1;
       1 1 0 0 0 1 1 0 0 1 1 1 0 0 0 1 0 0 1 0;
       1 1 1 0 0 1 0 0 0 1 1 1 0 0 1 1 1 0 0 0;
       1 1 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 0];
b = [1 1 1;
       0 1 0;
       1 1 1];
diffcount = nlfilter(a, [3 3], @(block) nnz(block~=b))
mindiff = min(diffcount(:));
[r, c] = find(diffcount == mindiff)
bestmatches = arrayfun(@(R,C) a(R:R+2, C:C+2), r, c, 'uniform', 0)

There is no exact match, but there are places where there is only 1 difference.

Answer 2

[DGM]

If the goal is to look for a certain neighborhood, you can use bwlookup() for this.

A = [1 1 1 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0;
    1 1 0 0 1 0 1 1 1 0 1 1 1 0 0 1 1 1 0 0;
    1 1 1 1 1 1 0 1 1 0 0 1 0 0 0 0 1 0 0 0;
    0 0 0 1 1 0 1 1 1 0 1 0 1 0 0 0 0 0 0 0;
    0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 0 1 1 1;
    0 0 1 1 1 0 1 1 1 0 1 0 0 0 1 0 0 0 1 1;
    1 0 0 0 1 0 1 1 0 1 1 1 0 1 1 1 0 1 0 1;
    1 1 0 1 1 1 1 0 1 0 1 0 0 0 1 0 0 0 0 1;
    1 0 0 0 1 0 0 1 0 0 1 0 0 0 0 0 1 1 1 0;
    0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 0 1 0 0;
    0 0 0 0 0 1 0 1 0 0 0 1 1 1 0 0 1 1 1 0;
    0 1 1 0 1 0 1 1 0 0 0 1 1 1 1 0 1 1 1 0;
    1 0 0 1 0 0 0 0 1 1 1 0 0 1 0 0 0 0 0 0;
    1 1 1 0 0 0 0 0 1 1 1 0 1 1 1 0 0 0 0 0;
    0 1 1 1 0 1 0 0 1 1 1 0 0 1 1 0 0 1 1 1;
    0 1 1 0 0 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1;
    0 0 0 0 1 1 1 0 0 0 1 0 0 0 1 0 1 1 1 1;
    1 1 0 0 0 1 1 0 0 1 1 1 0 0 0 1 0 0 1 0;
    1 1 1 0 0 1 0 0 0 1 1 1 0 0 1 1 1 0 0 0;
    1 1 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 0];
b = [1 1 1;
    0 1 0;
    1 1 1];

% find matches
f = @(x) isequal(x,b);
lut = makelut(f,3);
B = bwlookup(A,lut);

% compare the input/output
% note that the 'I' pattern is present at the two match locations
imshow(imfuse(A,B))

If instead of looking only for exact matches, you wanted to count the number of matching pixels in any given nhood, you can just use a different definition of f.

f = @(x) nnz(x==b); % the number of matched pixels (9 is a full match)

If you don't want to use bwlookup(), you can do the prior matching with a basic linear filter. Note that the inputs to imfilter() must be numeric class (not logical). Cast them as necessary. This could also be done with conv2() instead of imfilter().

seb = 2.^([1 4 7; 2 5 8; 3 6 9]-1); % index weighting array
B = imfilter(A,seb) == sum(sum(seb.*b)); % logical map of matches

% compare the input/output
imshow(imfuse(A,B))

If the goal is to erode the image using b as a structuring element, I guess you could do that instead.

% do erosion instead of pattern searching
C = imerode(A,b);
imshow(C)