grayscale image clustering, show ouput image from matrix

Question

Tomas 2014-4-26

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/127303-grayscale-image-clustering-show-ouput-image-from-matrix

评论： Image Analyst 2018-3-13

Hello, I want a grayscale image clustering according intersity colors from 0-255.

my code:

I = imread('obraz1.png');
I=rgb2gray(I);
imshow(I);
title('Grayscale image','FontSize',16,'Color','k'); 
I=double(I);
maxi=max(I(:));
I=I./maxi;
[m n]=size(I);
P = [];
for i=1:m    
    for j=1:n        
        if I(i,j)<1 %%All except white
          P=[P; i j ];
    end
        end
    end
    size(P);
    MON=P; 
[IDX]= kmeans(MON,3,'emptyaction','singleton')

how do i display image after clustering ?

2 个评论
显示无隐藏无

Manjurekha M.Phil 2018-3-12

i need grayscale image clustering, show ouput image from matrix source code in matlab

Image Analyst 2018-3-13

OK. Good luck.

https://www.mathworks.com/help/stats/cluster-analysis.html

http://www.powershow.com/view/83184-NWNiZ/Advanced_analysis_Classification_clustering_and_other_multivariate_methods_flash_ppt_presentation

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Image Analyst 2014-4-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/127303-grayscale-image-clustering-show-ouput-image-from-matrix#answer_134780

I'd guess create a classified image by going down IDX and setting the value of the classified image to either 1, 2, or 3 for each pixel, depending on what class it is.

3 个评论
显示 1更早的评论隐藏 1更早的评论

Image Analyst 2014-4-28

在 MATLAB Online 中打开

I don't have the Statistics Toolbox so I can't be sure but don't you think it would go like this

classifiedImage = zeros(size(I), 'int32');
for p = 1 : length(IDX)
  row = P(p, 1);
  column = P(p, 2);
  % Set this pixel of the classified image
  % to the class it identified for that pixel.
  classifiedImage(row, column) = IDX(p);
end

I've never used kmeans since I don't have the toolbox but I just read the documentation for a minute and that's what I came up with. It seemed really really obvious to me, though it's probably not right since you worked on it for a whole week, so what I came up with in less than a minute can't be right. But for what it's worth, that's my best guess.

Tomas 2014-4-28