Scribd
Upload
25 views1 page

Ex 10

This document provides instructions for an image processing exercise involving adding noise to an image and then using different filters to reconstruct the noisy image. Students are asked to add Gaussian noise, salt and pepper noise, and Rayleigh noise to an image, then use an arithmetic mean filter, geometric mean filter, and harmonic mean filter to reconstruct each noisy image. They are instructed to write MATLAB scripts to add the noise and apply the filters, and include the resulting images and an explanation in their report.

Uploaded by

Huy Tran
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
25 views1 page

Ex 10

This document provides instructions for an image processing exercise involving adding noise to an image and then using different filters to reconstruct the noisy image. Students are asked to add Gaussian noise, salt and pepper noise, and Rayleigh noise to an image, then use an arithmetic mean filter, geometric mean filter, and harmonic mean filter to reconstruct each noisy image. They are instructed to write MATLAB scripts to add the noise and apply the filters, and include the resulting images and an explanation in their report.

Uploaded by

Huy Tran
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

SGN-3016 Digital Image Processing I

EXERCISE 10
15-16.11.2011
Reports should be submitted through online form on the following link:
http://www.cs.tut.fi/kurssit/SGN-3016/Reports/

1- Download the lena.jpg, which is 512x512 gray-level image.
Add the following noises to the image individually.
a) Gaussian noise: Use imnoise function to generate the noisy image.
b) Salt & pepper noise: Use imnoise function to generate the noisy image.
c) Rayleigh noise:
- Assume a uniform random number, w, in the interval (0,1). We wish to generate random numbers, z, with a Rayleigh CDF of the
form:

2
( ) /
1 for
( )
0 for
z a b
z
e z a
F z
z a


>
=
`
<
)

where b > 0. To find z we solve the equation

2
( ) /
1 = w
z = a + -bln(1-w)
z a b
e


- MATLAB: R = a + sqrt(-b*log((1-rand(M,N))
- Add the generated noise to the image to obtain the noisy image.

Select the noise parameters so that image quality deterioration is noticeable and the image quality approximately same in
each case.

Reconstruction of Noisy Images:

Create 3 3
a) Arithmetic mean filter:
# MATLAB: f = i m filte r (g , fs p e c ia l( 'a v e ra g e ', [m n ]) )

b) Geometric mean filter
# MATLAB: f = e x p ( im fi lte r ( l o g ( g ) , o n e s( m , n ) , ' r e p li c a te ' ) ) .^ ( 1 /( m *n ) )
! log(a*b*c) = log(a) + log(b) + log(c) + ...

c) Harmonic mean filter
# MATLAB: (m *n ) ./ i m fi lte r (1 ./ (g + e p s ) , o n e s ( m, n ) , 'r e p lic a te ') ;


# MATLAB : Use imagesc function to display the output images.

What filtering methods do you suggest for eliminating the specific noises created above? Describe the effects of
the filters on the each noisy image. Your report should contain the Matlab scripts, the resulting images and explanations.

You might also like