How to convert RGB to hsv?

I have a RGB data with each size like R is 1 x 100 double. how can i convert these RGB to hsv. I find the formula for it from Wikipedia and it available at [here]:

I also found some materials that could answer my questions such as 1 but it is in another language than Matlab.

Answer 1

[KALYAN ACHARJYA]

For one frame, you can check here, as you mentioned the reference (wiki), I have implmented the same from here

I assumed, you can find R,G,B

Defining hue in terms of RGB

hue_rgb=atan2(sqrt(3)*(G-B),R-G-B)

Answer 2

[Walter Roberson]

for H = 1:ncols
    if T3(H) == 0
        hprime(H) = 0;
    elseif T1(H) == R(H)
        hprime(H) = mod((G(H)-B(H))/C(H), 6);
    elseif T1(H) == G(H)
        hprime(H) = ((B(H)-R(H))/C(H))+2;
    elseif T1(H) == B(H)
        hprime(H) = ((R(H)-G(H))/C(H))+4;
    else
        error('undefined hue at index %d', H);
    end
end

[Fredrick Graham]

Thanks for your comment and code.

I am not sure what i should add for ncols since i know i have 3 vectors of 1 x 100. Do you think i need to get read of for loop and use only if statement? Please let me know you comment. I would like to generate this hue and normilize it. thanks

[Walter Roberson]

%R, G, B are row vectors
ncols = length(R);
hprime = zeros(1, ncols);

T1 = max([R; G; B]);
T2 = min([R; G; B]);

"Do you think i need to get read of for loop and use only if statement?"

No, that cannot work. It is possible to get rid of the for loop, but it would have to be replaced with logical indexing. For example,

mask = T1 == G;
hprime(mask) = ((B(mask)-R(mask))./T3(mask))+2; 

[Walter Roberson]

T1 = max([R, G, B])

You need to decide whether you are working with row vectors or column vectors.

For any one position, you need to take the maximum of the corresponding R, G, and B values.

If your R, G, B are 1 x 800 row vectors, then [R, G, B] is 1 x 2400 and max() would find the single value among all of the R, G, B values that is largest overall, instead of being the corresponding maxima.

If your R, G, B are 800 x 1 column vectors, then [R, G, B] is 800 x 3, and max() would find the maximum per-column, which would be a 1 x 3 result -- the largest R, largest G, largest B.

[Fredrick Graham]

@ Walter Roberson thanks again fot your explanation. It is ok I think to go with column vestor. when i am running now, i am getting error as:

undefined hue at index 1

Please let me know if now my code is make sense.

Thanks

[Walter Roberson]

What difficulty did you run into with the code that I had already posted to take care of the issue?

T1 = max([R; G; B]);

[Fredrick Graham]

Thanks

Answer 3

[Image Analyst]

Why not simply use the built-in rgb2hsv() function:

% load R and G and B data
clear
close all
clc
load('RGB.mat')

subplot(4,1,1)
plot(t,R,'r')
ylim([110, 160]);
xlabel('Time (sec)')
ylabel('Red')
grid on;

subplot(4,1,2)
plot(t,G,'g')
ylim([110, 160]);
xlabel('Time (sec)')
ylabel('Green')
grid on;

subplot(4,1,3)
plot(t,B,'b')
ylim([110, 160]);
xlabel('Time (sec)')
ylabel('Blue')
grid on;

% Convert RGB to HSV:
RGBMatrix = [R; G; B];
% Make MxNx3 array rather than a Mx3 array (where rgb2hsv will think it's a colormap rather than a list of RGB values).
RGBMatrix = repmat(RGBMatrix, [1, 1, 2]);
RGBMatrix = permute(RGBMatrix, [3, 2, 1]);
hsvVec = rgb2hsv(RGBMatrix);
hueVector = hsvVec(1, :, 1);

subplot(4,1,4)
plot(t,hueVector,'b')
xlabel('Time (sec)')
ylabel('Hue')
grid on;