Extracting Dominant Color of an Image using K-Means Clustering

clc

clear all

close all

warning off

[filename, pathname] = uigetfile('*.*', 'Pick an Image');

if isequal(filename,0) || isequal(pathname,0)

    disp('User pressed cancel')

else

    filename=strcat(pathname,filename);

    rgbImage=imread(filename);

    imshow(rgbImage);

    title('Original Image');

    k=input('Enter the k value: ');

    close

    subplot(1,2,1);

    imshow(rgbImage);

    title('Original Image');

    redChannel=rgbImage(:, :, 1);

    greenChannel=rgbImage(:, :, 2);

    blueChannel=rgbImage(:, :, 3);

    data=double([redChannel(:), greenChannel(:), blueChannel(:)]);

    numberOfClasses=k;

    [m n]=kmeans(data,numberOfClasses);

    m=reshape(m,size(rgbImage,1),size(rgbImage,2));

    n=n/255;

    clusteredImage=label2rgb(m,n);

    subplot(1,2,2);

    imshow(clusteredImage);

    title('Clustered Image');

    frequency=[];

    temp=0;

    for i = 1:k

        for a=1:size(rgbImage,1)

            for b=6:size(rgbImage,2)

                if(m(a,b)==i)

                    temp=temp+1;

                end

            end

        end

        frequency=[frequency temp];

        temp=0;

    end

    figure;

    pie(frequency)

    colormap([n])%define the color scheme

    [ma na]=max(frequency);

    disp(n(na,:));

end

 

Note if you want to visualize the dominant color separately and get the values in command window(which are in double and scaled down by dividing by 255), use the below code--

clc

clear all

close all

warning off

[filename, pathname] = uigetfile('*.*', 'Pick an Image');

if isequal(filename,0) || isequal(pathname,0)

    disp('User pressed cancel')

else

    filename=strcat(pathname,filename);

    rgbImage=imread(filename);

    imshow(rgbImage);

    title('Original Image');

    k=input('Enter the k value: ');

    close

    subplot(1,2,1);

    imshow(rgbImage);

    title('Original Image');

    redChannel=rgbImage(:, :, 1);

    greenChannel=rgbImage(:, :, 2);

    blueChannel=rgbImage(:, :, 3);

    data=double([redChannel(:), greenChannel(:), blueChannel(:)]);

    numberOfClasses=k;

    [m n]=kmeans(data,numberOfClasses);

    m=reshape(m,size(rgbImage,1),size(rgbImage,2));

    n=n/255;

    clusteredImage=label2rgb(m,n);

    subplot(1,2,2);

    imshow(clusteredImage);

    title('Clustered Image');

    frequency=[];

    temp=0;

    for i = 1:k

        for a=1:size(rgbImage,1)

            for b=6:size(rgbImage,2)

                if(m(a,b)==i)

                    temp=temp+1;

                end

            end

        end

        frequency=[frequency temp];

        temp=0;

    end

    figure;

    pie(frequency)

    colormap([n])%define the color scheme

    [ma na]=max(frequency);

    disp(n(na,:));

    figure;

    patch([0 0 10 10],[0 10 10 0],n(na,:));

    title('Dominant Color');

end

 

Join us on Telegram: https://t.me/matlabcastor

Join us on Facebook Group: https://www.facebook.com/groups/matlabcodes