How to calculate PSNR of compressed images, and how to compare PSNR of images compressed by two different techniques.

I have to compare image compression techniques like VQ, JPEG, WAVELET, and fractal. For this, the parameter to be compared is PSNR. Please tell me how I can calculate PSNR OF AN IMAGE which is COMPRESSED by different compression techniques. plz explain with example.

Answer 1

[Image Analyst]

See my demo:

% Demo to calculate PSNR of a gray scale image.
% http://en.wikipedia.org/wiki/PSNR
% Clean up.
close all; 
clear all; 
workspace;

%------ GET DEMO IMAGES ----------------------------------------------------------
% Read in a standard MATLAB gray scale demo image.
grayImage = imread('cameraman.tif');
[rows columns] = size(grayImage);

% Display the first image.
subplot(2, 2, 1);
imshow(grayImage, []);
title('Original Gray Scale Image');
set(gcf, 'Position', get(0,'Screensize')); % Maximize figure.

% Get a second image by adding noise to the first image.
noisyImage = imnoise(grayImage, 'gaussian', 0, 0.003);

% Display the second image.
subplot(2, 2, 2);
imshow(noisyImage, []);
title('Noisy Image');

%------ PSNR CALCULATION ----------------------------------------------------------
% Now we have our two images and we can calculate the PSNR.
% First, calculate the "square error" image.
% Make sure they're cast to floating point so that we can get negative differences.
% Otherwise two uint8's that should subtract to give a negative number
% would get clipped to zero and not be negative.
squaredErrorImage = (double(grayImage) - double(noisyImage)) .^ 2;
% Display the squared error image.
subplot(2, 2, 3);
imshow(squaredErrorImage, []);
title('Squared Error Image');
% Sum the Squared Image and divide by the number of elements
% to get the Mean Squared Error.  It will be a scalar (a single number).
mse = sum(sum(squaredErrorImage)) / (rows * columns);
% Calculate PSNR (Peak Signal to Noise Ratio) from the MSE according to the formula.
PSNR = 10 * log10( 256^2 / mse);
% Alert user of the answer.
message = sprintf('The mean square error is %.2f.\nThe PSNR = %.2f', mse, PSNR);
msgbox(message);

[rashmi]

PSNR = 10 * log10( 256^2 / mse); WHY YOU USE 256^2 //// IF OUR IMAGE SIZE IS 225X225 .... THAN ALSO WE USE 256^2?

[Image Analyst]

See the definition. It's 256 because that's the maximum gray level. It doesn't have anything to do with what size your image is. Why do you think it does? The image size comes in when you calculate MSE because you have to get the mean, but after that it's not used.

[Image Analyst]

Akanksha, please read Cris's blog on PSNR: http://www.cb.uu.se/~cris/blog/index.php/archives/490#more-490. If we're done here, please mark my answer as Accepted.

[Karbala'a Unvi. Science]

Dear Sir, It was a great code that you did here and it did helped me a lot in my project... It is the one that I was looking for Yours Z

[Said BOUREZG]

Image Analyst have do all thing, nothing to say it after this answer, thank you sir for this code.

[zeeshan akhtar]

@Image Analyst A slight modification ( I think) will be using 255^2 instead of 256^2. For graysacle image value varies form [0 255];

[prasannakumar madigani]

its not the pixel range it is the size of the image

[Image Analyst]

zeeshan, you're right. It should be 255. Using that agrees with Wikipedia and with MATLAB's built-in psnr function. New code is attached.

[DIMITRIOS]

can you please send in the codes for eight.tif gaussian, original and salt and peper image also their psnr?

[DGM]

imread()

im2gray()

imnoise()

psnr()