Analysis of denoised image by plotting histogram

Analysis of denoised image by plotting Histogram. Image histogram can be defined as the graphical representation of pixel intensity values in a digital image as shown in below Fig 1. The assessment of the image quality of denoised image can also be done by plotting image histogram. This is a very unique way of analyzing the image quality of denoised image.
  1. The previous post explains the image assessment method using intensity profile.
  2. Here a new graphical representation of image quality assessment of denoised image is explained by plotting the histogram.
  3. The practical implementation of this image analysis technique is shown in three of my research paper of SAR image despeckling. The author can refer those three articles i.e. article1 , article2 and article3.
Image Histogram (imhist)
  1. Image histogram can be defined as the graphical representation of pixel intensity values in a digital image as shown in below Fig 1. 
  2. Matlab provides an predefined in-built function for plotting the image histogram i.e. imhist.
  3. Fig 1. depicts the image histogram. Axis X denotes the intensity values and Axis Y denotes the count i.e. number of pixels.

Image Histogram (imhist) Implementation (MATLAB code)


Fig 2. Reference, Noisy  and Denoised images


X = imread('cameraman.tif');     ---> X is the Reference Image
Y = imnoise(X,'gaussian',0,0.05);   ---> Y is the Noisy Image
Z = Denoising_Method(Y);   --->   Z is the Denoised Image

figure;

histo=imhist(X);      
plot(histo,'-r','LineWidth',2)      ---> Plotting histogram of Reference Image
hold on
histo1=imhist(Z);      
plot(histo1,'-b','LineWidth',2)      ---> Plotting histogram of Denoised Image
hold off
xlabel('Pixel Values');
ylabel('Number of Pixels');
legend('Reference Image', 'Denoised Image')
title('Histogram Plotting of Reference Image and Denoised Image')

Output of above code and result analysis

Fig 2. Plotting histogram of reference image and denoised image
  1. Fig 1 shows reference image, noisy image and denoised image.
  2. Fig 2 shows the plotting of histogram of reference image (red) and denoised image (blue).
  3. Fig 2. explains the overlapping scenario of pixel distribution of reference (red) and denoised images (blue).
  4. How to do Analysis: Since in the Fig 2., the overlapping of plot (pixel distribution) of reference and denoised images  are shown in red and blue line graph respectively. If they are overlapped, then the denoising results are good and if they are not overlapped properly then the denoising results are not good. Here in the Fig 2., The overlapping is good, so the denoising results are satisfactory.



For more image processing topics refer: https://www.gofastresearch.com/
Analysis of denoised image by plotting histogram Analysis of denoised image by plotting histogram Reviewed by IPR on April 20, 2020 Rating: 5

1 comment:

Powered by Blogger.