How to apply a mask only on saturation in a hsv image

I have an rgb image converted n hsv. I need just the saturation level (sIm)

hsvIm = rgb2hsv(im); % Convert RGB to HSV
hIm = hsvIm(:,:,1); % Get the hue
sIm = hsvIm(:,:,2); % Get the saturation
vIm = hsvIm(:,:,3); % Get the value
   

Now i would need to apply a threshold on the sIm and i am trying using this function which i created modifying once which I generated by image toolbox

function [BW,maskedhsvImage] = satMask(sIm)
% Define thresholds for channel 2 based on histogram settings
channel2Min = 0.000;
channel2Max = 0.404;

I = sIm;
% Create mask based on chosen histogram thresholds
sliderBW = (I(:,:,2) >= channel2Min ) & (I(:,:,2) <= channel2Max);
BW = sliderBW;

% Set background pixels where BW is false to zero.
maskedRGBImage(repmat(~BW,[1 1 1])) = 0;
end

The probem is that when i call the function i have this error.

% Index in position 3 exceeds array bounds (must not exceed 1).
% Error in satMask (line 23)
% sliderBW = (I(:,:,2) >= channel2Min ) & (I(:,:,2) <= channel2Max);

Where i am wrong?

Answer 1

[DGM]

It's not at all clear what the goals even were, but considering that other people are looking at this question and upvoting it, I guess someone has an idea.

I'm going to omit the broken junk that used to be an exported function from ColorThresholder. If you want to rewrite it as a function, that's up to you.

RGBimage = imread('peppers.png');
HSVimage = rgb2hsv(RGBimage);

% Define thresholds
Srange = [0.000 0.7];

% Create mask
mask = (HSVimage(:,:,2) >= Srange(1)) & (HSVimage(:,:,2) <= Srange(2));

% Do you want to ...?
% Set background S to 0
maskedS = HSVimage(:,:,2);
maskedS(~mask) = 0;
outpict1 = HSVimage;
outpict1(:,:,2) = maskedS;
outpict1 = im2uint8(hsv2rgb(outpict1));

% Set background of all HSV channels to 0
outpict2 = HSVimage;
outpict2(repmat(~mask,[1 1 3])) = 0;
outpict2 = im2uint8(hsv2rgb(outpict2));

% Set background of parent RGB image to 0
outpict3 = RGBimage;
outpict3(repmat(~mask,[1 1 3])) = 0;

% the results
OP = [outpict1 outpict2 outpict3];
imshow(OP)

You'll have to decide what you think it means to "apply a mask" and specifically, what it means to "apply a mask only on saturation".

See also:

https://www.mathworks.com/matlabcentral/answers/153866-convert-an-rgb-image-to-grayscale-but-keep-one-color#answer_1080133

https://www.mathworks.com/matlabcentral/answers/588391-creation-of-a-mask-on-rgb-image#answer_1084878

https://www.mathworks.com/matlabcentral/answers/488338-how-to-divide-split-rgb-image-into-two-parts-and-second-part-gray-and-concatenate-them-in-a-single-i