Contrast stretching in MATLAB

     Increase the dynamic range of gray values in the input image.

Read contrast.jpg image which is a low contrast image. Apply the transformation such that r₁=r_min, S₁=0, r₂=r_max, S₂=L-1. Plot the transformation function and display the contrast stretched image.

CODE:

clc; clear all;

%% Reading an image

a1=imread('contrast.jpg');

a=double(a1);

[row,col]=size(a);

%% Transformation function (r1,s1)=(rmin,0), (r2,s2)=(rmax,255)

t=0:255;

x1=0*(t>=0 & t<=100); 

x2=7.285*(t-100).*(t>100& t<135);          

x3=255*(t>=135 & t<=255);

x=x1+x2+x3;

%% Obtaining contrast stretched image

for n=1:row

    for m=1:col

        out(n,m)=x(a1(n,m)+1);

    end

end

out1=imadjust(a1,stretchlim(a1),[]);

plot(x)

grid on;

xlabel('Intensity in input image');

ylabel('Intensity in output image')

title('Transformation function')

figure()

subplot(131), imshow(a1), title('Original image')

subplot(132), imshow(uint8(out)), title('Contrast Stretching(using code)')

subplot(133), imshow(out1), title('Contrast Stretching(using stretchlim)')

OUTPUT RESULT:

Transformation Output

Basic Intensity Transformation Functions

1) Image Negative: s=L-1-r where r is the input intensity level and s is the output intensity level.

Read northpole.jpg image, obtain its negative. Display the original image and its negative.

clc; clear all;

a=imread('northpole.jpg');

b=255-a;                    % s=L-1-r

subplot(211), imshow(a), title('Northpole Image')

subplot(212), imshow(b), title('Negative of Northpole')

2)   Log Transformation:   s=T(r)=clog(1+r) where c is a constant and r is the input gray level.

    Read fourier.jpg image and apply log transformation and display the fourier Spectrum.

clc; clear all;

in=imread('fourier.jpg');

c=input('Enter the constant value, c = ');

a=im2double(in);

a=a*255;

out=c*log10(1+a);                   % s=T(r)=clog(1+r)

out=out/max(max(out));              % Normalization

subplot(121), imshow(in), title('Original Image')

subplot(122), imshow(out), title('Log Transformed Image(c=1)')

3)   Power Law Transformations: s=cr^γ. To account for an offset, when the input is zero use s= c(r+ε)^γ.

 Read crt.jpg, apply the transformation function with γ=1/2.5. Display the gamma corrected image.

clc; clear all;

c=1;

Gamma=input('Enter the Gamma value = ');

x=imread('crt.jpg');

x1=double(x);      

y=c*(x1.^Gamma);                % s=c*(r^ γ)

subplot(211),imshow(x), title('CRT input image')

subplot(212),imshow((y),[]), title('Corrected image(Gamma=0.4)')

4)     Contrast Manipulation:

Read mr.jpg image which is a magnetic resonance image with an upper thoracic human spine with a fracture dislocation. Apply power law transformation with exponents γ=0.6,  γ=0.4,  γ=0.3. Display the outputs and comment on the results.

clc; clear all;

c=1;

Gamma=input('Enter the Gamma value = ');  % Must be vector, Ex:[0.6 0.4 0.3]

x=imread('mr.jpg');

x1=double(x);      

y=c*(x1.^Gamma(1));                             % s=c*(r^ γ)

y1=c*(x1.^Gamma(2));

y2=c*(x1.^Gamma(3));

subplot(141),imshow(x), title('MRI scanned image')

subplot(142),imshow((y),[]), title('Corrected image(Gamma=0.6)')

subplot(143),imshow((y1),[]), title('Corrected image(Gamma=0.4)')

subplot(144),imshow((y2),[]), title('Corrected image(Gamma=0.3)')