I am working on a project where I have to find deltaE value of CIElab image... The formula for deltaE is sqrt((L1-L​2)^2+(a1-a​2)^2+(b1-b​2)^2). And also I want to know how to find L1,L2,a1,a2,b1,b2

My project is to detect anemia using eye image... So far I have found segmenting red color and mean values... My code is

%Segmentation of eye image
a = imread('eye.jpg');
imshow(a);title('H&E image');
text(size(a,2),size(a,1)+15,...
'Image Courtesy of Alen Partin,Johns Hopkins University',...
'FontSize',7,'HorizontalAlignment','right');
cform = makecform('srgb2lab');
lab_i = applycform(a,cform);
ab = double(lab_i(:,:,2:3));
nrows = size(ab,1);
ncols = size(ab,2);
ab = reshape(ab,nrows*ncols,2);
nColors = 3;
[cluster_idx,cluster_center] = kmeans(ab,nColors,'distance','sqEuclidean',...
'Replicates',3);
 pixel_labels = reshape(cluster_idx,nrows,ncols);
imshow(pixel_labels,[]),title('image labeled by clusterindex');
segmented_images = cell(1,3);
rgb_label = repmat(pixel_labels,[1 1 3]);
for k = 1:nColors
color = a;
color(rgb_label~=k) = 0;
segmented_images{k} = color;
end
imshow(segmented_images{1}),title('objects in cluster 1');
imshow(segmented_images{2}),title('objects in cluster 2');
%To make background black null
i2 = segmented_images{2};
background = i2<50;
i2(background) = 0;
impixelinfo % To check whether background black is zero
%Mean values
mean(mean(i2))% Mean R,G,B
% converting i2 Image to lab
cform = makecform('srgb2lab');
lab_Image = applycform(im2double(i2),cform);
imshow(lab_Image)

Now I have to find the L1,L2,a1,a2,b1,b2 values which are needed to find deltaE = (sqrt((L1-L2)^2 + (a1-a2)^2 + (b1_b2)^2)

So please help me with code... Thanks in advance:)

[jonas]

The L, a and b components are stored in your image matrix.

L=lab_i(:,:,1)
a=lab_i(:,:,2)
b=lab_i(:,:,3)

If you want to calculate a delta, then you just need reference values as well.

[Sivasurya J]

What do you mean by reference values and how do I know this...

[Sivasurya J]

When I try the above code the values are running... I want to know single average or mean value

[jonas]

Delta is the distance between two points in the color space, where L*, a and b are coordinates. Obviously, you need two different colors to do this, whereof one is (L1,a1,b1) and one is (L2,a2,b2). What exactly is it that you want to calculate? Difference between images? Difference between pixels?

[Sivasurya J]

Difference between pixels within an image

[jonas]

It's still not clear to me what exactly you want to achieve with this analysis. Are you trying to map an image by finding regions where the color changes? Perhaps this would be more clear if you upload eye.jpg

[Sivasurya J]

We are just going to detect anemia by non-invasive method by segmenting conjunctiva of an eye image... Now, I uploaded my eye image

[Guillaume]

Define what 1 and 2 refers to in your formula. Which pixel is 1 and which pixel is 2?

[Sivasurya J]

Is it possible to create a for loop to iterate between each adjacent pixels in an image?

[Stephen23]

@Sivasurya J: you don't need a loop for this. As you can easily compare the Lab data with itself: once offset be one row, and once offset by one column:

...
lab = rgb2lab(rgb);
dy = lab(2:end,:,:)-lab(1:end-1,:,:);
dx = lab(:,2:end,:)-lab(:,1:end-1,:);
dEy = sqrt(sum(dy.^2,3));
dEx = sqrt(sum(dx.^2,3));

[Sivasurya J]

It shows error as 'This statement is not inside any function. ( It follows the END that terminates the definition of the function "sRGB2CIEDeltaE".)'

[Sivasurya J]

Can you specify the line in my code which you say a better deltaE code..?

[Sivasurya J]

i2 = imread('eye.jpg'); imshow(i2);

redChannel = i2(:,:,1); greenChannel = i2(:,:,2); blueChannel = i2(:,:,3); mask = redChannel == 25& greenChannel == 25 & blueChannel == 25; cform = makecform('srgb2lab'); lab_Image = applycform(im2double(i2),cform); imshow(lab_Image) LChannel = lab_Image(:,:,1); aChannel = lab_Image(:,:,2); bChannel = lab_Image(:,:,3); LMean = mean(LChannel(mask)); aMean = mean(aChannel(mask)); bMean = mean(bChannel(mask)); [rows,columns,numberOfColorBands] = size(i2); LStandard = LMean*ones(rows,columns); aStandard = aMean*ones(rows,columns); bStandard = bMean*ones(rows,columns); deltaL = LChannel-LStandard; deltaa = aChannel-aStandard; deltab = bChannel-bStandard; deltaE = sqrt(deltaL.^2 + deltaa.^2 + deltab.^2);

I am getting output for this code when I am using eye.jpg(which I uploaded above) image But it shows NaN value for red.jpg(I uploaded it here) image I can't figure out the reason and I want to know what is mask? Why we are using it? and How we are assigning values(like 0 or 50) for it?

[Stephen23]

@Sivasurya: you wrote "It follows the END that terminates the definition of the function "sRGB2CIEDeltaE""

Apparently you have a function called sRGB2CIEDeltaE: is this an existing function (in which case: what does it do?), or is this the function that you are trying to write now?

[Sivasurya J]

The segmentation of image(which is the first step of my code) is based on K-means algorithm where RGB is converted to CIE, but it doesn't give delta value:(

[Sivasurya J]

@jonas:- If I am having two images... one is my reference image(another eye image). Now I need to find DeltaE value. How could you help me with code..? a = imread('eye.jpg'); imshow(a); cform = makecform('srgb2lab'); lab_Image = applycform(im2double(a),cform); LChannel = lab_Image(:,:,1); aChannel = lab_Image(:,:,2); bChannel = lab_Image(:,:,3); [LMean,aMean,bMean] = GetMeanLABValues(LChannel,aChannel,bChannel,mask); lRef = LMean(indexOfReferenceImage); aRef = aMean(indexOfReferenceImage); bRef = bMean(indexOfReferenceImage); deltaLs = LMean - lRef; deltaas = aMean - aRef; deltabs = bMean - bRef; deltaEs = sqrt(deltaLs.^2+deltaas.^2+deltabs.^2); This image shows error at Ref... Can you help me?

[Image Analyst]

There is a lot wrong with much/most of what you've written above. I don't have time to explain it all now - hopefully tomorrow. In the meantime, I don't see your eye image and reference image posted, so please post them.