Module 4

Image Enhancement
Frequency Domain

Resmi N.G.
Reference:
Digital Image Processing
Rafael C. Gonzalez
Richard E. Woods
Module 4 - Syllabus
Image Enhancement in Frequency Domain
Basics of Filtering in Frequency Domain Filters -
Smoothing Frequency Domain Filters : Ideal Low
Pass
Filter; Gaussian Low Pass Filter; Butterworth Low
Pass
Filter;
Sharpening Frequency Domain Filters: Ideal High
Pass Filter; Gaussian High Pass Filter; Butterworth
High Pass Filter;
Homomorphic Filtering.
06/21/2025 2
 Frequency Domain Methods
Basics of filtering in frequency domain
Basic Filters and Properties
 Notch filter
 Lowpass Filter

 Highpass Filter

Smoothing Frequency Domain Filters

 Ideal Lowpass Filters
 Butterworth Lowpass Filters

 Gaussian Lowpass Filters

Sharpening Frequency Domain Filters

 Ideal Highpass Filters
 Butterworth Highpass Filters

 Gaussian Highpass Filters

Homomorphic Filtering
06/21/2025 3
 Basics of Filtering in Frequency
Domain

06/21/2025 4
06/21/2025 5
Basic Steps for Filtering in Frequency Domain

06/21/2025 6
 Frequency Domain Methods
Basics of filtering in frequency domain
Basic Filters and Properties
 Notch filter
 Lowpass Filter

 Highpass Filter

Smoothing Frequency Domain Filters

 Ideal Lowpass Filters
 Butterworth Lowpass Filters

 Gaussian Lowpass Filters

Sharpening Frequency Domain Filters

 Ideal Highpass Filters
 Butterworth Highpass Filters

 Gaussian Highpass Filters

Homomorphic Filtering
06/21/2025 7
Basic Filters and Properties
Notch Filter
It is a constant function with a hole at the origin.
Sets F(0,0) to zero.

Lowpass Filter
It attenuates high frequencies and passes low frequencies.

Highpass Filter
It attenuates low frequencies and passes high frequencies.

06/21/2025 8
 Frequency Domain Methods
Basics of filtering in frequency domain
Basic Filters and Properties
 Notch filter
 Lowpass Filter

 Highpass Filter

Smoothing Frequency Domain Filters

 Ideal Lowpass Filters
 Butterworth Lowpass Filters

 Gaussian Lowpass Filters

Sharpening Frequency Domain Filters

 Ideal Highpass Filters
 Butterworth Highpass Filters

 Gaussian Highpass Filters

Homomorphic Filtering
06/21/2025 9
Smoothing Frequency Domain Filters
Low Pass Filter (Smoothing Filter)
The result in the spatial domain is equivalent to that
of a smoothing filter as the blocked high frequencies
correspond to sharp intensity changes, i.e. to the fine-
scale details and noise in the spatial domain image.

06/21/2025 10
06/21/2025 11
 High Pass Filter(Sharpening Filter)
A highpass filter attenuates the low-frequency
components without disturbing the high frequency
information in the Fourier Transform.

It yields edge enhancement or edge detection in the

spatial domain, because edges contain many high
frequencies. Areas of constant gray level consist
mainly of low frequencies and are therefore
suppressed.

06/21/2025 12
06/21/2025 13
Band Pass Filter
 A bandpass filter attenuates very low and very high
frequencies, but retains a middle range band of
frequencies. Bandpass filtering can be used to enhance
edges (suppressing low frequencies) while reducing the
noise(attenuating high frequencies).

 Bandpass filter is a combination of both lowpass and

highpass filters. These filters attenuate all frequencies
below a specific frequency and above a specific frequency,
while retaining the frequencies between the two cut-offs.

06/21/2025 14
 Frequency Domain Methods
Basics of filtering in frequency domain
Basic Filters and Properties
 Notch filter
 Lowpass Filter

 Highpass Filter

Smoothing Frequency Domain Filters

 Ideal Lowpass Filters
 Butterworth Lowpass Filters

 Gaussian Lowpass Filters

Sharpening Frequency Domain Filters

 Ideal Highpass Filters
 Butterworth Highpass Filters

 Gaussian Highpass Filters

Homomorphic Filtering
06/21/2025 15
Ideal Low Pass Filters
Transfer Function
1 if D(u , v) D0
H (u , v) 
0 if D(u , v)  D0
D0 is a specified non  negative quantity .
D(u,v)is the distance from point (u,v)to the origin of
the frequency rectangle.
1

   

2 2 2
D(u , v)  u  M  v N
 2 2 
06/21/2025 16
 Ideal – because all frequencies inside a circle of radius D0
are passed without any attenuation, whereas all
frequencies outside the circle are completely attenuated.

The point of transition between H(u,v) = 1 and H(u,v) = 0

is called the cut-off frequency.

06/21/2025 17
06/21/2025 18
Ideal Low pass Filter

Produces “Ringing” effect.

Cannot be realized in electronic components.
Not very Practical

06/21/2025 19
Butterworth Low Pass Filters
The transfer function of a BLPF of order n, and with cut-
off frequency at a distance D0 from the origin, is defined
as 1
H (u , v)  2n
 D(u , v) 
1  
 D 0 
1
 n

   

2 2
u  M  v  N
  2 2  
1  2 
 D0 
 

06/21/2025 20
06/21/2025 21
 Provides a smooth transition between low and high
frequencies.

Butterworth filter of order 1 has neither ringing nor

negative values.

BLPF of order 2 has mild ringing and small negative

values.

Reduced ringing effect than ILPF.

06/21/2025 22
Gaussian Low Pass Filters
 D 2 ( u ,v )
2 2
H (u , v) e
D(u,v)is the distance from the origin of the Fourier
Transform.
 is a measure of the spread of the Gaussian curve.
When  D0 ,
 D 2 ( u ,v )
2 D02
H (u , v) e
where D0 is the cut  off frequency.
06/21/2025 23
 When D (u , v) 0, H (u , v) 1
 D02
2 D02 1
When D (u , v) D0 , H (u , v) e e 2
0.607

06/21/2025 24
Gaussian Low Pass Filters
Very smooth filter function.
Inverse DFT of the Gaussian lowpass filter is Gaussian.
No “Ringing” effect.

06/21/2025 25
Applications of Low Pass Filters
In the field of machine perception
Character Recognition

In printing and publishing industry.

Cosmetic processing prior to printing

For processing satellite and aerial images.

06/21/2025 26
 Frequency Domain Methods
Basics of filtering in frequency domain
Basic Filters and Properties
 Notch filter
 Lowpass Filter

 Highpass Filter

Smoothing Frequency Domain Filters

 Ideal Lowpass Filters
 Butterworth Lowpass Filters

 Gaussian Lowpass Filters

Sharpening Frequency Domain Filters

 Ideal Highpass Filters
 Butterworth Highpass Filters

 Gaussian Highpass Filters

Homomorphic Filtering
06/21/2025 27
Sharpening Frequency Domain Filters
Ideal High Pass Filters
Transfer Function of high pass filter is given by

H hp (u, v) 1  H lp (u, v)
H lp (u , v) is the transfer function of corresponding
low pass filter.
That is, when low pass filter attenuates frequencies, high
pass filter passes them and vice versa.

06/21/2025 28
 Opposite of ideal lowpass filter.

0 if D(u, v) D0
H (u , v) 
1 if D (u , v)  D0
Sets to zero all frequencies inside a circle of radius D0
while all frequencies outside the circle are passed without
attenuation.

Not physically realizable with electronic components.

Produces ringing effect.

06/21/2025 29
06/21/2025 30
 D0 = 15,30,80

06/21/2025 31
Butterworth High Pass Filter
1
H (u , v)  2n
 D0 
1  
 D (u , v ) 

06/21/2025 32
 D0 = 15,30,80
Represents a transition between the sharpness of IHPF
and the total smoothness of Gaussian filter.
06/21/2025 33
Gaussian High Pass Filter
 D 2 ( u ,v )
2 D0 2
H (u , v) 1  e

06/21/2025 34
 D0 = 15,30,80

06/21/2025 35
Homomorphic Filtering
The illumination-reflectance model can be
used to develop a frequency domain
procedure for improving the appearance of
an image by simultaneous intensity range
compression and contrast enhancement.

06/21/2025 36
06/21/2025 37
06/21/2025 38
06/21/2025 39
 H(u,v) – homomorphic filter transfer function.

06/21/2025 40
06/21/2025 41
 Thank You

06/21/2025 42