Prev: RC-Element
Next: FFT processing
From: shilpa khinvasara on 22 Jul 2010 02:35
sir,
u already know that i am doing quantization of image into 18,3,3 bis with the following code
now how do i count the pixels in each bin and plot its histogram
i have corrected the code as per ur suggestion



clc;
clear all;
close all;

% Open a standard MATLAB demo image.
rgbImage = imread('test.jpg');
rgbImage=im2double(rgbImage);

% Display the original image.
figure;
subplot(2, 4, 1);
imshow(rgbImage, []);
title('Original RGB image');

% Convert to HSV color space
hsv = rgb2hsv(rgbImage);

% Extract out the individual channels.
h = hsv(:,:,1);
s = hsv(:,:,2);
v = hsv(:,:,3);

% Display the individual channels.
subplot(2, 4, 2);
imshow(h, []);
title('Hue Image');
subplot(2, 4, 3);
imshow(s, []);
title('Saturation Image');
subplot(2, 4, 4);
imshow(v, []);
title('Value Image');

% Take histograms
[hCount, hValues] = hist(h(:), 18);
[sCount, sValues] = hist(s(:), 3);
[vCount, vValues] = hist(v(:), 3);

% Plot histograms.
subplot(2, 4, 5);
bar(hValues, hCount);
title('Hue Histogram');
subplot(2, 4, 6);
bar(sValues, sCount);
title('Saturation Histogram');
subplot(2, 4, 7);
bar(vValues, vCount);
title('Value Histogram');
h=h*360;
[M,N,O]=size(hsv);

H=floor(h * (1-eps) / 20);

V = zeros(M,N);
vplaces = v > 0.25;
V(vplaces & s <= 0.2) = 3;
V(vplaces & s <= 0.126) = 2;
V(vplaces & s <= 0.67) = 1;
S=zeros(size(hsv,1),size(hsv,2));
for i=1:M
for j=1:N
if s(i,j)>0.2&&s(i,j)<=1&&v(i,j)>0.25&&v(i,j)<=1
q(i,j)=9*H(i,j)+3*S(i,j)+V(i,j)+4;
end
end

end

pls let me know the corrections.
thanks a lot lot
From: Walter Roberson on 22 Jul 2010 03:01
shilpa khinvasara wrote:
> sir,
> u already know that i am doing quantization of image into 18,3,3 bis
> with the following code
> now how do i count the pixels in each bin and plot its histogram

I already gave you the code needed to count the pixels in each bin and
plot its histogram. It was a single call to hist .

> S=zeros(size(hsv,1),size(hsv,2));

You failed to populate the S array before you use it in the loop that
follows.

> for i=1:M
> for j=1:N
> if s(i,j)>0.2&&s(i,j)<=1&&v(i,j)>0.25&&v(i,j)<=1
> q(i,j)=9*H(i,j)+3*S(i,j)+V(i,j)+4;
> end
> end
> end

You also do not pre-allocate the "q" array, and you do not have any
indication of what that "q" array should contain for locations at which
the "if" statement does not hold true.

I see no reason why you would want to use a nested "for" loop to build
the "q" array values. The technique used to construct "vplaces" can be
used here to select the locations of interest and calculate your q
values for those locations in one or two fully vectorized commands.
From: kaushalk on 22 Jul 2010 04:46
thanx a lot
but still i am unable to understand that how to count pixels in each of quantzed image and plot its histogram


 | 
Pages: 1
Prev: RC-Element
Next: FFT processing