How to implement Boundary discriminative noise detection method for salt and pepper noise removal of images

Respected sir/madam

I have to implement the boundary discriminative noise detection method for high density impulse noise removal from images using matlab.can you help me with the implementation? Noise as high as 80-90% needs to be removed. can it be implemented for colour images also?

[Sajid Khan]

function [ bin_noisy ] = BDND( image_noisy, wind_init )

[row, col] = size(image_noisy);

start_wind = ceil(wind_init/2);

end_wind = floor(wind_init/2);

bin_noisy = false(row, col);

loc_median = ceil(wind_init*wind_init/2);

image_padded = pad_image( image_noisy, wind_init );

for i = start_wind: row+end_wind

      for j = start_wind: col+end_wind  

%         if(i == 317 + start_wind - 1 && j  == 296 + start_wind - 1)

%             i

%         end

          patch_selected = image_padded(i - end_wind: i + end_wind, j - 
  end_wind: j + end_wind);

          pix_center = patch_selected(start_wind, start_wind);

          patch_selected = patch_selected(:);

          [array_sorted, ind] = sort(patch_selected);

          diff_patch = diff(array_sorted);

          pre_med = diff_patch(1:loc_median);

          post_med = diff_patch(loc_median+1:end);

          [~,max_pre] = max(pre_med);

          [~,max_post] = max(post_med);

          b1 = array_sorted(max_pre);

          b2 = array_sorted(loc_median+max_post);

          bin_noisy(i - end_wind, j - end_wind) = (pix_center<=b1 || pix_center>b2);

      end

end

Answer 1

[Image Analyst]

I don't know that method by name. I have posted salt and pepper removal code here before. Didn't you run across it in your search?

http://matlab.wikia.com/wiki/FAQ#Can_you_program_up_the_algorithm_in_this_article_for_me_and_explain_it_to_me.3F