Generate histogram of HSV quantized image

Hello,

I have tried to quantize the HSV color space but I can not, can anyone help me please?

I want to Quantization HSV Space by assigning 8 level each to hue, saturation and value to give a quantized HSV space with 8x8x8=512 histogram bins.

I have used the following code for that but I am getting an error in rescale function. Can anyone help me to sort out this problem?

rgbImage = imread('1.jpg');

hsvImage = rgb2hsv(rgbImage);

montage(hsvImage,'Size', [1, 3]); title('Before Quantization'); colorbar;

threshRGB = multithresh(hsvImage, 8);

threshForPlanes = zeros(3, 8);

for i = 1:3

threshForPlanes(i, :) = multithresh(hsvImage(:, :, i), 8);

end

quantPlane = zeros(size(hsvImage));

for i = 1:3

value = [0 threshForPlanes(i, 2:end) 1];

quantPlane(:, :, i) = imquantize(hsvImage(:, :, i), threshForPlanes(i, :), value);

end

quantPlane = double(rescale(quantPlane, 0, 1));

figure; montage(quantPlane, 'Size', [1,3]); title('After Quantization');

colormap(parula(8)); colorbar;

% Error: Undefined function 'rescale' for input arguments of type 'double'.

quantPlane = double(rescale(quantPlane, 0, 1));

Thank you.

Answer 1

[Image Analyst]

Why are you calling multithresh() and imquantize() to quantize it. Just call histogram() if you want the histogram. Your subject line says you want to "Generate histogram of HSV quantized image". Quantizing it will not create a histogram. The histogram() function will.

[Priyanka Mehta]

yes, it is true. but then how to compare this histogram with the histogram of another image.

please help me.

[Image Analyst]

There are lots of ways. For example, subtract the means to see how the average brightness is different. You'd have to specify what you want to learn.

[Priyanka Mehta]

As an input I will give one image. its matching images should be retrived from the dataset. this should be done using histogram of HSV image.

[Image Analyst]

You should Google "CBIR" and see what algorithms people are using.