Write an image processing algorithm that measures the surface area of the green pills in pixels

Considering the image ‘pills2’, write an image processing algorithm that measures the surface area of the green pills in pixels. The pills pixels must be isolated as white pixels while the rest of the image must be colored in black.

I = imread('pills2.bmp');

R = I(:,:,1);
G = I(:,:,2);
B = I(:,:,3);


hsv = rgb2hsv(I);
H = hsv(:,:,1);
figure,imshow(H), title('Hue');

How can I do next?

Answer 1

[Mathieu NOE]

hi

this is my 2 cents suggestion - but I am not truly an image processing expert

For what I know your "green" definition must be more precise, because it covers quite a large range of RGB triplets

see :

So you have to play with the individual R,G,B triplets or find a smarter way to compare with a RGB table that list allcombinations considered as green

I = imread('pills2.bmp');

R = I(:,:,1);
G = I(:,:,2);
B = I(:,:,3);

thresholdValueR = 0.4*(max(max(R)));
thresholdValueG = 0.5*(max(max(G)));
thresholdValueB = 0.65*(max(max(B)));

binaryImage = R < thresholdValueR & G > thresholdValueG & B < thresholdValueB;

figure(2),
subplot(2,1,1),imshow(I), title('H');
subplot(2,1,2),imshow(binaryImage), title('Binary Image');
[a,b] = size(binaryImage == 1);
nb_of_G_pixels = a*b