MATLAB TUTORIAL FIVE

Image Enhancement in Frequency Domain.

GuassianNoise ,  Salt & Pepper Noise,  Laplacian Noise

clc;

clearall;

closeall;

a = imread('Arjunlamborghini.jpg');

subplot(2,2,1);

imshow(a); title('Original image');

b = imnoise(a,'gaussian',0.1);

subplot(2,2,2);

imshow(b);  title('Gaussian Noise added');

c =imnoise(a,'salt& pepper',0.1);

subplot(2,2,3);

imshow(c);  title('Salt & Pepper Noise added');

d =fspecial('laplacian',0.2);

e=imfilter(a,d,'replicate');

subplot(2,2,4);

imshow(e);  title('Laplacian Noise added');

Histgrams of GuassianNoise ,  Salt & Pepper Noise,  Laplacian Noise images

clc;

clearall;

closeall;

a1 = imread('lamborghini.jpg');

a = rgb2gray(a1);

b = imnoise(a,'gaussian',0.1);

c =imnoise(a,'salt& pepper',0.1);

d =fspecial('laplacian',0.2);

e=imfilter(a,d,'replicate');

subplot(2,2,1);

imhist(a);  title('Histogram of Gray image');

subplot(2,2,2);

imhist(b);  title('Histogram of Gaussian Noise image added');

subplot(2,2,3);

imhist(c);  title('Histogram of salt & Pepper noise added image');

subplot(2,2,4);

imhist(e);  title('Histogram of Laplacian noise added image');

Rayleigh Noise & Histograms

clc;

clearall;

closeall;

a=1;b=0.25;

i = imread('lamborghini.jpg');

f = rgb2gray(i);

f1 = double(f);

[r, c]=size(f);

R = a+(-b*log(1-rand(r,c))).^0.5;  %RAYLEIGH NOISE EQUATION

mmax=max(max(R));

mmin=min(min(R));

const=100/(mmax-mmin);

for x = 1:1:r;

for y = 1:1:c;

noise(x,y)=const*(R(x,y)-mmin);

end

end

noisy_image=f1+noise;

subplot(2,3,1);

imshow(i),title('ORIGINAL IMAGE');

subplot(2,3,2);

imshow(f),title('GRAYSCALE IMAGE');

subplot(2,3,3);

imshow(uint8(noisy_image)),title('IMAGE WITH ADDED RAYLEIGH NOISE');

subplot(2,3,4);

imhist(f), title('HISTROGRAM OF GRAYSCALE IMAGE');

subplot(2,3,5);

hist(noise), title('HISTOGRAM OF RAYLEIGH NOISE')

subplot(2,3,6);

hist(noisy_image),title('HISTOGRAM OF IMAGE WITH RAYLEIGH NOISE')

 Uniform Noise & Histograms

clc;

clearall;

closeall;

a=1;b=0.25;

i = imread('lamborghini.jpg');

f = rgb2gray(i);

f1 = double(f);

[r, c]=size(f);

U = a + (b-a)*rand(r,c); % UNIFORM NOISE EQUATION

mmax=max(max(U));

mmin=min(min(U));

const=100/(mmax-mmin);

for x = 1:1:r;

for y = 1:1:c;

noise(x,y)=const*(U(x,y)-mmin);

end

end

noisy_image=f1+noise;

subplot(2,3,1);

imshow(i),title('ORIGINAL IMAGE');

subplot(2,3,2);

imshow(f),title('GRAYSCALE IMAGE');

subplot(2,3,3);

imshow(uint8(noisy_image)),title('IMAGE WITH ADDED UNIFORM NOISE');

subplot(2,3,4);

imhist(f), title('HISTROGRAM OF GRAYSCALE IMAGE');

subplot(2,3,5);

hist(noise), title('HISTOGRAM OF UNIFORM NOISE')

subplot(2,3,6);

hist(noisy_image),title('HISTOGRAM OF IMAGE WITH UNIFORM NOISE')

Exponential Noise & Histograms

clc;

clearall;

closeall;

a=1;b=0.25;

i = imread('lamborghini.jpg');

f = rgb2gray(i);

f1 = double(f);

[r, c]=size(f);

E = -log(1.12-rand(r,c)); %EXPONENTIAL NOISE EQUATION

mmax=max(max(E));

mmin=min(min(E));

const=100/(mmax-mmin);

for x = 1:1:r;

for y = 1:1:c;

noise(x,y)=const*(E(x,y)-mmin);

end

end

noisy_image=f1+noise;

subplot(2,3,1);

imshow(i),title('ORIGINAL IMAGE');

subplot(2,3,2);

imshow(f),title('GRAYSCALE IMAGE');

subplot(2,3,3);

imshow(uint8(noisy_image)),title('IMAGE WITH ADDED EXPONENTIAL NOISE');

subplot(2,3,4);

imhist(f), title('HISTROGRAM OF GRAYSCALE IMAGE');

subplot(2,3,5);

hist(noise), title('HISTOGRAM OF EXPONENTIAL NOISE')

subplot(2,3,6);

hist(noisy_image),title('HISTOGRAM OF IMAGE WITH EXPONENTIAL NOISE')

Lognormal Noise & Histograms

clc;

clearall;

closeall;

a=1;b=0.25;

i = imread('lamborghini.jpg');

f = rgb2gray(i);

f1 = double(f);

[r, c]=size(f);

L = a*exp(b*randn(r,c)); %LOGNORMAL NOISE EQUATION

mmax=max(max(L));

mmin=min(min(L));

const=100/(mmax-mmin);

for x = 1:1:r;

for y = 1:1:c;

noise(x,y)=const*(L(x,y)-mmin);

end

end

noisy_image=f1+noise;

subplot(2,3,1);

imshow(i),title('ORIGINAL IMAGE');

subplot(2,3,2);

imshow(f),title('GRAYSCALE IMAGE');

subplot(2,3,3);

imshow(uint8(noisy_image)),title('IMAGE WITH ADDED LOGNORMAL NOISE');

subplot(2,3,4);

imhist(f), title('HISTROGRAM OF GRAYSCALE IMAGE');

subplot(2,3,5);

hist(noise), title('HISTOGRAM OF LOGNORMAL NOISE')

subplot(2,3,6);

hist(noisy_image),title('HISTOGRAM OF IMAGE WITH LOGNORMAL NOISE')

Blur Images , Restoration and Histograms

clc;

clearall;

closeall;

a = im2double(imread('lamborghini.jpg'));

a1= rgb2gray(a);

PSF = fspecial('motion', 21, 11);

blurred = imfilter(a1, PSF, 'conv', 'circular');

noise_var = 0.0001;

estimated_nsr = noise_var/var(a1(:));

wnr3 = deconvwnr(blurred, PSF, estimated_nsr);

subplot(2,3,1);

imshow(a1),title('Grayscale Image');

subplot(2,3,2);

imshow(blurred),title('Blurred Image');

subplot(2,3,4);

imhist(a1),title('Histogram of Grayscale Image');

subplot(2,3,5);

imhist(blurred),title('Histogram of Blurred Image');

subplot(2,3,3);

imshow(wnr3)

title('Restoration of Blurred Image');

subplot(2,3,6);

imhist(wnr3);

title('Histogram of Restoration Image');