Digital Image Processing

Image Restoration:
Noise Removal
2
of
46
Contents
In this lecture we will look at image
restoration techniques used for noise removal
– What is image restoration?
– Noise and images
– Noise models
– Noise removal using spatial domain filtering
– Periodic noise
– Noise removal using frequency domain
filtering
3
of
46
What is Image Restoration?
Image restoration attempts to restore
images that have been degraded
– Identify the degradation process and attempt
to reverse it
– Similar to image enhancement, but more
objective
4
of
46

• The purpose of image restoration is to "compensate

for" or "undo" defects which degrade an image.
• Degradation comes in many forms such as motion
blur, noise, and camera misfocus.
• In cases like motion blur, it is possible to come up
with a very good estimate of the actual blurring
function and "undo" the blur to restore the original
image.
• In cases where the image is corrupted by noise, the
best we may hope to do is to compensate for the
degradation it caused
5
of
46

Degradation of images can have many causes

• defects of optical lenses;
• nonlinearity of the electro-optical sensor;
• graininess of the film material;
• relative motion between an object and camera
• wrong focus,
• atmospheric turbulence in remote sensing or astronomy,
• etc.
• The objective of image restoration is to reconstruct the
original image from its degraded version.
6
of
46
Noise and Images
The sources of noise in digital
images arise during image
acquisition (digitization) and
transmission
– Imaging sensors can be
affected by ambient
conditions
– Interference can be added
to an image during
transmission
7
of
46
Noise Model
We can consider a noisy image to be
modelled as follows:
g ( x, y )  f ( x, y )   ( x, y )
where f(x, y) is the original image pixel,
η(x, y) is the noise term and g(x, y) is the
resulting noisy pixel
If we can estimate the model the noise in an
image is based on, this will help us to figure
out how to restore the image
8
of
46
Noise PDF (Gaussian noise)

Where

P(Zi) = normalized histogram

 9
of
46
Noise Models
There are many different
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Gaussian Rayleigh
models for the image
noise term η(x, y):
– Gaussian
• Most common model Erlang Exponential

– Rayleigh
– Erlang
– Exponential Uniform
– Uniform Impulse

– Impulse
• Salt and pepper noise
 10
of
46
Noise Example
The test pattern to the right is
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

ideal for demonstrating the

addition of noise
The following slides will show
the result of adding noise Image
based on various models to
this image
Histogram to go here

Histogram
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

of
11

46

Gaussian
Rayleigh
Erlang
Noise Example (cont…)
 12
of
46
Noise Example (cont…)
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Histogram to go here

Exponential Uniform Impulse

13
of
46
Filtering to Remove Noise
We can use spatial filters of different kinds
to remove different kinds of noise
The arithmetic mean filter is a very simple
one and is calculated as follows:
ˆf ( x, y )  1
 g ( s, t )
mn ( s ,t )S xy
1/ 1/ 1/ This is implemented as the
9 9 9
1/ 1/ 1/
simple smoothing filter
9 9 9
Blurs the image to remove
1/ 1/ 1/
9 9 9 noise
14
of
46
Other Means
There are different kinds of mean filters all of
which exhibit slightly different behaviour:
– Geometric Mean
– Harmonic Mean
– Contraharmonic Mean
15
of
46
Other Means (cont…)
There are other variants on the mean which
can give different performance
Geometric Mean:
1
  mn
fˆ ( x, y )    g ( s, t )
( s ,t )S xy 
Achieves similar smoothing to the arithmetic
mean, but tends to lose less image detail
16
of
46
Other Means (cont…)
Harmonic Mean:
mn
fˆ ( x, y ) 
1

( s ,t )S xy g ( s, t )
Works well for salt noise, but fails for pepper
noise
Also does well for other kinds of noise such
as Gaussian noise
17
of
46
Other Means (cont…)
Contraharmonic Mean:
 g ( s, t )
( s ,t )S xy
Q 1

fˆ ( x, y ) 
 g ( s ,
( s ,t )S xy
t ) Q

Q is the order of the filter and adjusting its

value changes the filter’s behaviour
Positive values of Q eliminate pepper noise
Negative values of Q eliminate salt noise
 18
of
46
Noise Removal Examples
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image
Original Corrupted
Image By Gaussian
Noise

After A 3*3 After A 3*3

Arithmetic Geometric
Mean Filter Mean Filter
 19
of
46
Noise Removal Examples (cont…)
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image
Corrupted
By Pepper
Noise

Result of
Filtering Above
With 3*3
Contraharmonic
Q=1.5
 20
of
46
Noise Removal Examples (cont…)
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image
Corrupted
By Salt
Noise

Result of
Filtering Above
With 3*3
Contraharmonic
Q=-1.5
 21
of
46
Contraharmonic Filter: Here Be Dragons

Choosing the wrong value for Q when using

Images taken from Gonzalez & Woods, Digital Image Processing (2002)

the contraharmonic filter can have drastic

results
 22
of
46
Order Statistics Filters
Spatial filters that are based on ordering the
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

pixel values that make up the

neighbourhood operated on by the filter
Useful spatial filters include
– Median filter
– Max and min filter
– Midpoint filter
– Alpha trimmed mean filter
23
of
46
Median Filter
Median Filter:
fˆ ( x, y)  median{g ( s, t )}
( s ,t )S xy

Excellent at noise removal, without the

smoothing effects that can occur with other
smoothing filters
Particularly good when salt and pepper
noise is present
24
of
46
Max and Min Filter
Max Filter:
fˆ ( x, y)  max {g (s, t )}
( s ,t )S xy

Min Filter:
fˆ ( x, y )  min {g ( s, t )}
( s ,t )S xy

Max filter is good for pepper noise and min

is good for salt noise
25
of
46
Midpoint Filter
Midpoint Filter:
ˆf ( x, y )  1  max {g ( s, t )}  min {g ( s, t )}
2 ( s ,t )S xy ( s ,t )S xy 
Good for random Gaussian and uniform
noise
26
of
46
Alpha-Trimmed Mean Filter
Alpha-Trimmed Mean Filter:
ˆf ( x, y )  1
 g r ( s, t )
mn  d ( s ,t )S xy
We can delete the d/2 lowest and d/2 highest
grey levels
So gr(s, t) represents the remaining mn – d
pixels
 27
of
46
Noise Removal Examples
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Result of 1
Corrupted Pass With A
By Salt And 3*3 Median
Pepper Noise Filter

Result of 2 Result of 3
Passes With Passes With
A 3*3 Median A 3*3 Median
Filter Filter
 28
of
46
Noise Removal Examples (cont…)
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Image
Corrupted Corrupted
By Pepper By Salt
Noise Noise

Result Of Result Of
Filtering Filtering
Above Above
With A 3*3 With A 3*3
Max Filter Min Filter
 29
of
46
Noise Removal Examples (cont…)
Image Image Further
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Corrupted Corrupted
By Uniform By Salt and
Noise Pepper Noise

Filtered By Filtered By
5*5 Arithmetic 5*5 Geometric
Mean Filter Mean Filter

Filtered By Filtered By
5*5 Median 5*5 Alpha-Trimmed
Filter Mean Filter
30
of
46
Adaptive Filters
The filters discussed so far are applied to an
entire image without any regard for how
image characteristics vary from one point to
another
The behaviour of adaptive filters changes
depending on the characteristics of the
image inside the filter region
We will take a look at the adaptive median
filter
31
of
46
Adaptive Median Filtering
The median filter performs relatively well on
impulse noise as long as the spatial density
of the impulse noise is not large
The adaptive median filter can handle much
more spatially dense impulse noise, and
also performs some smoothing for non-
impulse noise
The key insight in the adaptive median filter
is that the filter size changes depending on
the characteristics of the image
32
of
46
Adaptive Median Filtering (cont…)
Remember that filtering looks at each
original pixel image in turn and generates a
new filtered pixel
First examine the following notation:
– zmin = minimum grey level in Sxy
– zmax = maximum grey level in Sxy
– zmed = median of grey levels in Sxy
– zxy = grey level at coordinates (x, y)
– Smax =maximum allowed size of Sxy
33
of
46
Adaptive Median Filtering (cont…)
Level A: A1 = zmed – zmin
A2 = zmed – zmax
If A1 > 0 and A2 < 0, Go to level B
Else increase the window size
If window size ≤ repeat Smax level A
Else output zmed
Level B: B1 = zxy – zmin
B2 = zxy – zmax
If B1 > 0 and B2 < 0, output zxy
Else output zmed
34
of
46
Adaptive Median Filtering (cont…)
The key to understanding the algorithm is to
remember that the adaptive median filter
has three purposes:
– Remove impulse noise
– Provide smoothing of other noise
– Reduce distortion
 35
of
46
Adaptive Filtering Example
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image corrupted by salt Result of filtering with a 7 Result of adaptive

and pepper noise with * 7 median filter median filtering with i = 7
probabilities Pa = Pb=0.25
 36
of
46
Periodic Noise
Typically arises due to
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

electrical or electromagnetic
interference
Gives rise to regular noise
patterns in an image
Frequency domain
techniques in the Fourier
domain are most effective
at removing periodic noise
37
of
46
Band Reject Filters
Removing periodic noise form an image
involves removing a particular range of
frequencies from that image
Band reject filters can be used for this purpose
An ideal band reject filter is given as follows:
 W
1 if D(u, v)  D0  2
 W W
H (u, v)  0 if D0   D(u , v)  D0 
 2 2
1 if D(u, v)  D0  W
 2
 38
of
46
Band Reject Filters (cont…)
The ideal band reject filter is shown below,
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

along with Butterworth and Gaussian

versions of the filter

Ideal Band Butterworth Gaussian

Reject Filter Band Reject Band Reject
Filter (of order 1) Filter
 39
of
46
Band Reject Filter Example
Image corrupted by Fourier spectrum of
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

sinusoidal noise corrupted image

Butterworth band Filtered image

reject filter
40
of
46
Summary
In this lecture we will look at image restoration
for noise removal
Restoration is slightly more objective than
enhancement
Spatial domain techniques are particularly
useful for removing random noise
Frequency domain techniques are particularly
useful for removing periodic noise