i'm trying to make histogram equalization code for normalized image without builtin function in MATLAB like 'histeq'
I have normalized gray image that has values between [0,1]
input_img = mat2gray(img);
and I have found an example code wihout histeq function
for i=1:size(GIm,1)
for j=1:size(GIm,2)
value=GIm(i,j);
freq(value+1)=freq(value+1)+1;
probf(value+1)=freq(value+1)/numofpixels;
end
end
sum=0;
no_bins=255;
%The cumulative distribution probability is calculated.
for i=1:size(probf)
sum=sum+freq(i);
cum(i)=sum;
probc(i)=cum(i)/numofpixels;
output(i)=round(probc(i)*no_bins);
end
for i=1:size(GIm,1)
for j=1:size(GIm,2)
HIm(i,j)=output(GIm(i,j)+1);
end
end
figure,imshow(HIm);
title('Histogram equalization');
but this code is applied to an image that has a grayscale of [0,255]
how should I apply this code to my grayscale normalized image?
my image size 450x450
Use the code you have to compute a 256-bin histogram by scaling your image to fit it:
input_img_255 = round(input_img * 255);
and then use input_img_255 with the code you have. round should be good enough to use the rounded double values as indices as in your code. If you want to be more exact you can use:
input_img_255 = uint8(round(input_img * 255));
so that the values are integers.
Related
As the title says, the output I’m getting out of this function is incorrect. By incorrect I mean that the data are overflowing. How do I normalise the matrix correctly? Currently almost all of the pictures I get are white.
I called the function from another MATLAB file like this:
mask = [3,10,3;0,0,0;-3,-10,-3];
A = imread(“football.jpg”);
B = ConvFun(A,mask);
function [ image ] = ConvFun( img,matrix )
[rows,cols] = size(img); %// Change
%// New - Create a padded matrix that is the same class as the input
new_img = zeros(rows+2,cols+2);
new_img = cast(new_img, class(img));
%// New - Place original image in padded result
new_img(2:end-1,2:end-1) = img;
%// Also create new output image the same size as the padded result
image = zeros(size(new_img));
image = cast(image, class(img));
for i=2:1:rows+1 %// Change
for j=2:1:cols+1 %// Change
value=0;
for g=-1:1:1
for l=-1:1:1
value=value+new_img(i+g,j+l)*matrix(g+2,l+2); %// Change
end
end
image(i,j)=value;
end
end
%// Change
%// Crop the image and remove the extra border pixels
image = image(2:end-1,2:end-1);
imshow(image)
end
In a convolution, if you want the pixel value to stay in the same range
, you need to make the mask add up to 1. Just divide the mask by sum(mask(:)) after defining it. This is however, not the case you are dealing with.
Sometimes that is not the needed. For example if you are doing edge detection (like the kernel you show), you don't really care about maintaining the pixel values. In those cases, the plotting of unnormalized images is more the problem. You can always set the imshow function to auto select display range: imshow(image,[]).
Also, I hope this is homework, as this is the absolutely worst way to code convolution. FFT based convolution is about 100 times faster generally, and MATLAB has an inbuilt for it.
I have a vector of values and I want to change the values somehow that its histogram is closer to the uniform distribution using MATLAB. I am aware of histeq in MATLAB that takes an image as input and assumes the densities are in 0-255 range. I am looking for a more general version of histeq.
You are looking to do a full scale contrast stretch, correct? If so this function will work. You can change K to be the largest value in your vector if you are not using 8 bit integers.
function [result] = myfscs(image)
K=255;
A= min(image(:));
B= max(image(:));
P=K/(B-A);
L=A*K/(B-A);
J = (P .* image - L);
result = uint8(J); % doesn't have to be a uint8 returned
I am trying to implement a simple function that its the same with the default hist() by MatLab.
We have two same images with different brightness and we have to convert them to grayscale and then use the default function of MatLab hist() to get the histograms (so far so good!).
Then we have to implement the function hist my_hist() , and when i am trying to count the frequency of the intensity the results are not the same.
It seems that it sums-up the frequency of 254 & 255 to 254 and 255 is zero!I dont know what the problem is, any help would be appreciated.
here is the code for the command line:
%Read the images and convert them from rgb to grayscale
i=imread('pic1.jpg');
j=rgb2gray(i);
x=imread('pic2.jpg');
y=rgb2gray(x);
%Display the two images
figure
imshow(j)
figure
imshow(y)
%Display the histogram of the two images
[a,b] = hist(j(:),0:1:255);
figure
plot(b,a)
[c,d]=hist(y(:),0:1:255);
figure
plot(d,c)
%Call of the built-in function
my_hist('pic1.jpg','pic2.jpg')
And here is the code of the self built function:
function []= my_hist( x,y)
%Read the images and convert them from rgb to grayscale
pic1=imread(x);
i=rgb2gray(pic1);
pic2=imread(y);
j=rgb2gray(pic2);
%Initialize two vectors to be the axis for histogram
plotx=0:255;
ploty=zeros(1,256);
%Take the dimensions of the first image pic1
[m,n] = size(i);
%With 2 loops we go through the matrix of the image and count how many
%pixels have the same intensity
for k=1:m
for l=1:n
num=i(k,l)+1;
ploty(num)=ploty(num)+1;
end
end
%Display the histogram for the first image pic1
figure
plot(plotx,ploty);
%Initialize two vectors to be the axis for histogram
plotx2=0:255;
ploty2=zeros(1,256);
%Take the dimensions of the second image pic2
[m2,n2] = size(j);
%With 2 loops we go through the matrix of the image and count how many
%pixels have the same intensity
for o=1:m2
for p=1:n2
num2=j(o,p)+1;
ploty2(num2)=ploty2(num2)+1;
end
end
%Display the histogram for the second image pic2
figure
plot(plotx2,ploty2);
end
And here are the images pic1 and pic2.
This is a problem due to your image being of integer type uint8 which can only range from 0-255:
>> a= uint8(255)
a =
255
>> a=a+1
a =
255
Convert your data to say type uint16 with
j = uint16(j);
y = uint16(y);
and your problem should be gone:
>> a=uint16(a)
a =
255
>> a=a+1
a =
256
I have a problem, which is how it is applied Fourier Transform (ftt) on the image (or how enhance image used Fourier Transform)
when i run my program
the input was fingerprint image
the output was white image
the problem is the output should be fingerprint image after enhance used Fourier Transform not white image
F=fft2( I );
factor=abs(F).^F;
block =ifft2(factor);
R= fftshift(block);
I hope finding some help
The exponentiation of F.^F seems to be a big number, so it is above the upper value and matlab slice it to be the upper value.
% Calculating fft2
fft2im = fft2(double(im));
% Taking the spectrum with log scaling
fft2im = log(1+(abs(fft2im)));
% Putting DC in the middle:
spectrum = fftshift(fft2im);
% finding maximum in spectrum:
maximum = max(max(spectrum));
% scaling maximum to 255 and minimum to 0:
spectrum = 255*spectrum/maximum;
% Casting to uint8 to be able to display:
spectrum = uint8(spectrum);
imshow(spectrum);
I am using fft2 to compute the Fourier Transform of a grayscale image in MATLAB.
What is the common way to plot the magnitude of the result?
Assuming that I is your input image and F is its Fourier Transform (i.e. F = fft2(I))
You can use this code:
F = fftshift(F); % Center FFT
F = abs(F); % Get the magnitude
F = log(F+1); % Use log, for perceptual scaling, and +1 since log(0) is undefined
F = mat2gray(F); % Use mat2gray to scale the image between 0 and 1
imshow(F,[]); % Display the result
Here is an example from my HOW TO Matlab page:
close all; clear all;
img = imread('lena.tif','tif');
imagesc(img)
img = fftshift(img(:,:,2));
F = fft2(img);
figure;
imagesc(100*log(1+abs(fftshift(F)))); colormap(gray);
title('magnitude spectrum');
figure;
imagesc(angle(F)); colormap(gray);
title('phase spectrum');
This gives the magnitude spectrum and phase spectrum of the image. I used a color image, but you can easily adjust it to use gray image as well.
ps. I just noticed that on Matlab 2012a the above image is no longer included. So, just replace the first line above with say
img = imread('ngc6543a.jpg');
and it will work. I used an older version of Matlab to make the above example and just copied it here.
On the scaling factor
When we plot the 2D Fourier transform magnitude, we need to scale the pixel values using log transform to expand the range of the dark pixels into the bright region so we can better see the transform. We use a c value in the equation
s = c log(1+r)
There is no known way to pre detrmine this scale that I know. Just need to
try different values to get on you like. I used 100 in the above example.