How to merge regions from region adjacencies graph.

I have the following code and I want to merge regions by using mean color or average intensity value etc.

clear all
clc
[filename, pathname] = uigetfile({'*.*'},'Browse');
name=[pathname,filename];
rgb = imread(name);
figure,imshow(rgb);

%%%%Preprocessing Step
I = rgb2gray(rgb);
hy = fspecial('sobel');
hx = hy';
Iy = imfilter(double(I), hy, 'replicate');
Ix = imfilter(double(I), hx, 'replicate');
gradmag = sqrt(Ix.^2 + Iy.^2);
L = watershed(gradmag);
figure,imshow(L); %gradient img
Lrgb = label2rgb(L);

%Mark the Foreground Objects
 se = strel('disk',1);
Io = imopen(I, se);%opening
Ie = imerode(I, se);  
Iobr = imreconstruct(Ie, I); %opening-by-reconstruction

Iobrd = imdilate(Iobr, se);%closing-by-reconstruction
Iobrcbr = imreconstruct(imcomplement(Iobrd), imcomplement(Iobr));
Iobrcbr = imcomplement(Iobrcbr);
fgm = imregionalmax(Iobrcbr);%calculate regional maxima to obtain foreground markers
I2 = I;  %superimpose
I2(fgm) = 255;
%clean the edges of the marker blobs
se2 = strel(ones(1,1));      %ones(5,5)
fgm2 = imclose(fgm, se2);
fgm3 = imerode(fgm2, se2);
fgm4 = bwareaopen(fgm3,60);  
I3 = I;
I3(fgm4) = 255;
 figure, imshow(I3)
 title('Modified regional maxima superimposed on original image (fgm4)');

%Compute Background Markers
bw = im2bw(Iobrcbr, graythresh(Iobrcbr));
figure, imshow(bw), title('Thresholded opening-closing by reconstruction (bw)');

%too close to the edges of the object
D = bwdist(bw);
DL = watershed(D);
bgm = DL == 0;

%Compute the Watershed Transform of the Segmentation Function

gradmag2 = imimposemin(gradmag, bgm | fgm4);
L = watershed(gradmag2);
    figure, imshow(L);
    imtool(L)

I4 = I;
I4(imdilate(L == 0, ones(4,4)) | bgm | fgm4) = 255;
imtool(I4);
figure, imshow(I4);

[n e] = imRAG(L);
hold on;
for i=1:size(e, 1)
plot(n(e(i,:), 1), n(e(i,:), 2), 'linewidth', 2, 'color', 'black');
end
plot(n(:,1), n(:,2), 'bo');

By using this code, I can get the following region adjacencies graph. Now I want to merge these oversegmented regions. How can i continue this coding to merge similar regions. Please help me ...!

The size of the image is 732x486 outdoor scene RGB image. How to merge similar regions ...

Answer 1

[Image Analyst]

Maybe you want to use this algorithm instead: http://www.mathworks.com/matlabcentral/fileexchange/37197-dem--diffused-expectation-maximisation-for-image-segmentation

[kyawt kyawt]

hi sir, when i using this DEM code, I have the following problem how can i solve

To create the software library and run the demos: 1) unpack the compressed zip file in your working directory and cd to such directory

 you will find the following directories:
  - /visualization:   some visualization tools
  - /mex:     mex files 
  - /img:     color images to be processed
  - /results:     to store segmentation results
       - /doc:     the EL paper

2) download and install the netlab package in the same directory (<http://www.ncrg.aston.ac.uk/netlab/index.php>)

3) build the mex files, here we are providing only precompiled mexmaci64 (MacOS, 64 bit Matlab) >> cd mex >> mexall [you may need to modify the mexall.m script under Windows OS]

I face the following problem. Moreover, how to download and install netlab package. when downloading in the following link http://www.ncrg.aston.ac.uk/netlab/index.php I have not found ,which netlab package. Can you help me Sir, Thank a lot for your suggestion

>> cd mex >> mexall

C:\PROGRA~1\MATLAB\R2010A\BIN\MEX.PL: Error: 'mexutil.o' not found.

??? Error using ==> mex at 222 Unable to complete successfully.

Error in ==> mexall at 2 mex repmatC.c mexutil.o

[Image Analyst]

Sorry but I haven't actually tried the code. I suggest you contact the author.

[kyawt kyawt]

ok thanks sir !