Scribd
Upload
89 views16 pages

Image Enhancement in Frequency Domain

This document discusses image enhancement in the frequency domain. It introduces Fourier analysis and the Fourier transform. The key points are: 1. The Fourier transform represents an image as a sum of sinusoids of different frequencies, allowing analysis and processing in the frequency domain. 2. The discrete Fourier transform (DFT) represents a discrete image as a sum of complex exponentials, enabling digital computation via fast Fourier transform (FFT). 3. In the frequency domain, low frequencies correspond to smooth areas of the image while high frequencies contain edges and details. This property allows filtering techniques to enhance or remove certain image features.

Uploaded by

Mahreen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
89 views16 pages

Image Enhancement in Frequency Domain

This document discusses image enhancement in the frequency domain. It introduces Fourier analysis and the Fourier transform. The key points are: 1. The Fourier transform represents an image as a sum of sinusoids of different frequencies, allowing analysis and processing in the frequency domain. 2. The discrete Fourier transform (DFT) represents a discrete image as a sum of complex exponentials, enabling digital computation via fast Fourier transform (FFT). 3. In the frequency domain, low frequencies correspond to smooth areas of the image while high frequencies contain edges and details. This property allows filtering techniques to enhance or remove certain image features.

Uploaded by

Mahreen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

Image Enhancement in Frequency Domain

Fourier Analysis

• The French mathematician Joseph Fourier stated


(in 1768) that any periodic function can be
expressed as the sum of sines / cosines of
different frequencies each multiplied by a
different coefficient( this sum is called Fourier
Series).
•
 
 
                        
 
 
 

 
 
                                     
 
 
 

 
 
                                    
 
 
 
Fourier Transform

• Non-Periodic Functions (but with finite area under the


curve can also be expressed as the integral of sines and/or
cosines multiplied by a weighting function. This
representation is called Fourier Transform.
• Fourier Transform is more useful than Fourier Series.
• A function/image expressed either by Fourier Series or
Fourier Transform can be reconstructed (recovered)
completely via an inverse process, with no information loss.
• Fast Fourier Transform (FFT) is the digital computation of
Fourier Transform and was developed in late 1950’s. It has
revolutionized the field of Signal/Image Processing.
Fourier Analysis
One-Dimensional Fourier Transform

• The Fourier Transform of a single variable,


continuous function is defined as:

• The Inverse Fourier Transform can be


obtained
Two-Dimensional Fourier Transform

The Fourier Transform of a single variable,


continuous function is defined as:

The Inverse Fourier Transform can be obtained


Discrete Fourier Transform

• The DFT of a discrete function of one


variable, f(x) where x=0,1,2…..,M-1 is given as:

• The Inverse DFT can be obtained as:

04/13/2023 8
Discrete Fourier Transform

• To compute F(u), we start by substituting u=0 in the


exponential term and then summing for all values of x. Then
we substitute u=1 in the exponential and repeat the
summation over all values of x. This process is repeated for
all M values of u to obtain the complete Fourier Transform.
It takes M^2 summations and multiplications to compute
the DFT.

• Like f(x), the transform is also a discrete quantity, and it


has the same number of components as f(x).

04/13/2023 9
Discrete Fourier Transform

• The Euler representation of a complex number is:

• By substituting this expression in Eq. 4.2.5 and using the fact


that cos(-x)=cosx. For u=0,1,2,…….M-1
Discrete Fourier Transform

• Fourier Transform in polar form can be represented as:


The Two-Dimensional DFT

• The DFT of a function (image) f(x,y) of size M X N is given by:

• This expression must be computed for u=0,1,2……M-1 and


v=0,1,2…..N-1 and the variables u and v are the transform or
frequency variables while x and y are the spatial or image
variables.
The Two-Dimensional DFT

• The value of the transform at (u,v=0,0) is equal to the average


gray level of the image as shown below:
The Two-Dimensional DFT

• Similarly, the Inverse DFT given that we have F(u,v), can be


represented as:

• Where x=0,1,2,3…………..,M-1 and y=0,1,2,3……..,N-1


Properties of the Frequency Domain
• Since frequency is representing rate of change, so we can
associate frequencies in the F.T with patterns of intensity
variations in an image.
• As discussed previously, the slowest varying frequency
component i.e. u=v=0 corresponds to the average gray level of an
image.
• As we move away from the origin of the transform, the low
frequencies correspond to the slowly varying components of an
image.
• As we move further away from the origin, the higher
frequencies begin to correspond to faster and faster gray level
changes in the image. So, the edges of the objects and other
abrupt changes in the image (e.g. noise) lie away from the origin.
Basics of Filtering in the Frequency Domain

You might also like