How can I show the objects found in segmentation?

Hello, everyone,

I'm working on a segmentation code but what I'm missing is to know how can I plot/show the segments of the image I have found in another figure, this is the code I have so far:

I=imread('Prueba1.png');
%Show original image;
figure
imagesc(I)
colormap('gray')
axis equal
title('Imagen Original')

I=double(rgb2gray(I));
figure
imagesc(I)
colormap('gray')
axis equal
title('Imagen Escala Grises')

IN=255-I; 

f=bwlabel(IN,8);
figure
imagesc(f)
colormap('gray')
axis equal

g=regionprops(f,'Area','Centroid','Image');
[r,c]=size(g);
sprintf('Total of Objects: %d',r)

[r1,c1]=size(f);
figure
for o=1:r
%This is something I was trying but it just shows all of the figures:
    imagesc(g(o,1).Image)
    colormap('gray')
    axis equal
    drawnow
end

I think that I have to find where it starts and then draw it there, but I'm not quite sure about this, do you know anyway of doing this?

Thanks for your help.

P.S. This is the image I'm working on:

What I want to do is to have another image similar to the Gray Scale image here but that as the program is running I can see it's plotting. Thanks again.

Answer 1

[Gopichandh Danala]

In your code there is some noise when i saved the image, so I did multithresh it works anyway for segmentation of your image

Then I used label to segment each object seperately

Here is the sample code:

 I=imread('IMO.jpg');
  %Show original image;
  figure
  imshow(I,[])
  title('Imagen Original')

I=double(rgb2gray(I));
figure
imshow(I,[])
title('Imagen Escala Grises')

% segmentation
levels = multithresh(I,1);
seg = imquantize(I, levels);
% figure, imshow(seg,[])

% remove background
seg2 = seg;
seg2(seg == 2) = 0;
% figure, imshow(seg2,[]);

% label image for each object
BW = bwlabel(seg2);
figure, imshow(BW,[]);
title('Labeled image');

% display or segment each object and display
for i = 1:max(BW(:))
    tempImg = BW;
     tempImg(BW ~= i) = 0;
    figure, imshow(tempImg,[]);
    title(strcat(num2str(i),'-th Object'));
end

If you have any doubts comment here.

Gopi

[Annie]

Hello, Gopi,

Thanks for your answer, I made the changes you show me in the code, but I have a little doubt on how this part works:

% display or segment each object and display
for i = 1:max(BW(:))
    tempImg = BW;
     tempImg(BW ~= i) = 0; % <=This part is the one I have doubt
    figure, imshow(tempImg,[]);
    title(strcat(num2str(i),'-th Object'));
end

As I understand what it does is that every cell/pixel of the tempImg matrix that is different from the total number of objects is marked as zero (?) So that when BW = i then that is going to be shown?

Thanks again,

Annie

[Gopichandh Danala]

Here I am creating a tempImg for each object displaying it, repeating this process for all available objects in BW

[Image Analyst]

He's making a copy of the labeled image. Then he's setting all pixels that do not have a value in that copied image to zero, so that it will have only the one label left. You might prefer the ismember() method I used in my answer - that's what ismember() is built for.

Answer 2

[Image Analyst]

I just answered the question here: https://www.mathworks.com/matlabcentral/answers/331538-display-each-object-in-a-binary-image-separedtly#answer_260094

Again....

See my Image Segmentation Tutorial: http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862&sort=downloads_desc

Or, simply use ismember() to extract the blob you want. For example

[L, num] = bwlabel(BW);
for k = 1 : num
    thisBlob = ismember(L, k);
    figure
    imshow(thisBlob, []);
end