How to find neighbour pixels of a central pixel in a binary image?

I am working on a binary image so how to find its neighbours. I have successfully found neighbour pixels in gray scale image but after its conversion from gray to binary its not giving me the pixels of binary image, its still showing me the pixels of gray scale image. Can anyone provide help on this?

Thanks in advance.

[tauseef ahmad]

hi, Ekta sharma, how you found neighbour pixel in gray image? can you send me the code at [e-mail]

[kmla]

Hi Ekta Sharma, are you find the code to find neighbors pixels of a central pixel in a binary image? can you send it to me [e-mail]

Answer 1

[Walter Roberson]

What do you need to do with the neighbours? Do you specifically need to know their indices? If so then in what order?

If you are doing a calculation such as region finding then you would normally use regionprops() or one of the image morphological operators.

If you are doing a calculation such as mean you would normally use conv2()

If you are doing a calculation such as local min or local max then one of the morphological operators can handle that.

If you are doing more complex arithmetic calculations on a block centered around the pixel then nlfilt()

[Ekta Sharma]

Thanks a lot for you reply Sir. Actually I want to apply chain code for that purpose I need this

[Walter Roberson]

http://www.mathworks.com/matlabcentral/fileexchange/29518-freeman-chain-code

[Walter Roberson]

To find the neighbouring pixels for each non-zero pixel in a grayscale image:

[nr, nc] = size(YourImage);
[r, c] = find(YourImage);
aboves = [r - 1, c];
aboves(r == 1, :) = [];
belows = [r + 1, c];
belows(c == nr, :) = [];
lefts = [r, c - 1];
lefts(c == 1, :) = [];
rights = [r, c + 1];
rights(c == nc, :) = [];
aboveleft = [r - 1, c - 1];
aboveleft(r == 1 | c == 1, :) = [];
aboveright = [r, c + 1];
aboveright(r == 1 | c == nc, :) = [];
belowleft = [r + 1, c - 1];
belowleft(r == nr | c == 1, :) = [];
belowright = [r + 1, c + 1];
belowright(r == nr | c == nc, :) = [];

neighbors = [aboveright; lefts; belows; aboveleft; belowright; aboves; rights; belowleft];

Now, neighbors will be two columns, each row of which is the row and column number of a pixel that is a neighbor of a non-zero pixel in the grayscale image. Pixels might occur up to 8 times in the list, due to being neighbors of different non-zero pixels.

Good luck doing anything useful with this list.