Hizyan Hanum Hi
Murk Hassan Memon Hi
let me try answer your question.
If you want me to further develop any specific part of my answer just let me know.
If find it useful please mark it as Accepted Answer, thanks in advance:
1.
Capture
A=imread('sample1.jpg');
title('initial image');
2.
sometimes the variance between RGB layers shows a clear boundary
figure(5);varA=surf(var(double(A),0,3));
.
B=varA.CData;
B(B<1000)=0;
B(B>=1000)=255;
figure(6);imshow(B)
.
3.
RGB split
Because with the variance the cells look 'dehydrated', the RGB split shows
ColorList={'Red' 'Green' 'Blue'};
N=255;gr=0:1/(N-1):1;
figure(1);imshow(A);
cMap=zeros(N,3);cMap(:,1)=gr;
figure(2);hr=imshow(ind2rgb(A(:,:,1),cMap));title(ColorList{1});
% filtering greens
cMap=zeros(N,3);cMap(:,2)=gr;
figure(3);hg=imshow(ind2rgb(A(:,:,2),cMap));title(ColorList{2});
% filtering blues
cMap=zeros(N,3);cMap(:,3)=gr;
figure(4);hb=imshow(ind2rgb(A(:,:,3),cMap));title(ColorList{3});
C=hg.CData;
figure(7);imshow(C);
C=C(:,:,2);C=255*C;
C(C>150)=255;
C(C<150)=0;
figure(8);imshow(C);
.
4.
total area, in pixels
numel(C)
=
198660
area in pixels
numel(nonzeros(~C))
=
6172
same as
[cx,cy]=find(~C);
numel(cx)
=
6172
in percentage
100*numel(nonzeros(~C))/numel(C)
=
3.106815664955200
there is a 3.1% are of red tagged cells
5.
and following the MATLAB example using RGB to L*a*b and K-means clustering
A=imread('sample1.jpg');
figure(1);imshow(A)
cform = makecform('srgb2lab'); % RGB to L*a*b color space
lab_he = applycform(A,cform);
ab = double(lab_he(:,:,2:3));
nrows = size(ab,1);
ncols = size(ab,2);
ab = reshape(ab,nrows*ncols,2);
nColors = 3; % sort out with K-means clustering
% repeat the clustering 3 times to avoid local minima
[cluster_idx, cluster_center] = kmeans(ab,nColors,'distance','sqEuclidean', 'Replicates',3);
pixel_labels = reshape(cluster_idx,nrows,ncols); % label all pixels !?
figure(2);imshow(pixel_labels,[]), title('image labeled by cluster index');
segmented_images = cell(1,3);
rgb_label = repmat(pixel_labels,[1 1 3]);
for k = 1:nColors % sub-images segmenting A by color
color = A;
color(rgb_label ~= k) = 0;
segmented_images{k} = color;
end
figure(3);imshow(segmented_images{1}), title('objects in cluster 1');
figure(4);imshow(segmented_images{2}), title('objects in cluster 2');
. If the tagging has caught all the cells of interest and only the cells of interest (assuming the small specks do not contain red tagging) then a more accurate area percentage would be
C=segmented_images{2};
C1=C(:,:,1);imshow(C1)
C1(C1>100)=255;C1(C1<100)=0;
100*numel(nonzeros(C1))/numel(C1)
=
2.281787979462398
thanks for time and attention, any feedback welcome awaiting answer
John BG
