How can I obtain a 3D histogram from 3 separate arrays?

I have the following code:

img_rgb = imread('IMD037.bmp');
img_gray=rgb2gray(img_rgb);
[rows, cols] = size(img_gray);
%%%Converting image in HSV domain and then calculate Histogram
img_hsv = rgb2hsv(img_rgb);
% % splitting HSV image into h, s & v channels
h = img_hsv(:, :, 1);
s = img_hsv(:, :, 2);
v = img_hsv(:, :, 3);
% quantization levels
h_level=16; 
s_level=4;
v_level=4;
% %%%%%%%%%image quantization 
max_h=max(h(:));
min_h=min(h(:));
max_s=max(s(:));
min_s=min(s(:));
max_v=max(v(:));
min_v=min(v(:));
range_h=max_h-min_h;
range_s=max_s-min_s;
range_v=max_v-min_v;

%%%%%%%%%%%%%%FOR Hue 
scale_h=(h_level-1)/range_h;
%q=round(x*scale)/scale;
for row = 1:size(h, 1) 
  for col = 1 : size(h, 2)
      quantizedValueForH(row, col) = ceil(h(row, col)*scale_h); 
  end
end

%%%%%%%%%%%%%%%For Saturation
scale_s=(s_level-1)/range_s;
%q=round(x*scale)/scale;
for row = 1:size(s, 1) 
  for col = 1 : size(s, 2)
      quantizedValueForS(row, col) = ceil(s(row, col)*scale_s); % 16*h(i,j)/max of h
  end
end
%%%%%%%%%%%%%%%%For Value
scale_v=(v_level-1)/range_v;
%q=round(x*scale)/scale;
for row = 1:size(s, 1) 
  for col = 1 : size(s, 2)
      quantizedValueForV(row, col) = ceil(s(row, col)*scale_v); % 16*h(i,j)/max of h
  end
end

%%%%%Histogram 
figure
%vec_bin_edges=0:15;
%hist3([quantizedValueForH quantizedValueForS], {0 15});
hist(quantizedValueForH)
figure
hist(quantizedValueForS)
figure
hist(quantizedValueForV)

Answer 1

[Image Analyst]

You can use scatter3 like in my attached demo. But this is not a true 3D visualization because I just put a "dot" at the (R, G, B) location if there is a pixel with that color. To do a true 3D you'd have to associate each location with an intensity that represents how many pixels had that exact color. Nonetheless, it's probably good enough for what you want.

If you want a better 3D color gamut visualizer, like this:

, let me know. It's a bit tricky but can be done.

[Rida]

Thanks for your answer Sir

I am actually trying to count probability of each of the entry in H,S,V

such as when quantizedValueForH=0, quantizedValueForS=0, quantizedValueForV=0 the value at of another 16x4x4 matrix at 1,1,1 will be incremented. If the value is quantizedValueForH=4,quantizedValueForS=0,quantizedValueForV=1 the value at 5,1,2 will be incremented and so on, therefore I tried to make 3D histogram but that's not the right way I guess.