Chapter four

Frequency Domain Image

Processing

01/19/2025 1
 Introduction To Frequency
Domain
In the frequency domain, a digital image is converted
from spatial domain to frequency domain.
In the frequency domain, image filtering is used for image
enhancement for a specific application.
Whereas in frequency domain, we deal with the rate at
which the pixel values are changing in spatial domain.
For simplicity, Let’s put it this way.
.

01/19/2025 1
 Cont’d…
In simple spatial domain, we directly deal with the image matrix.

Where as Frequency Domain, We first transform the image to

its frequency distribution, Then black box system perform what
ever processing it has to performed, and the output of the black
box in this case is not an image, but a transformation.

After performing inverse transformation, it is converted into

an image which is then viewed in spatial domain

01/19/2025 1
Cont’d…

It can be pictorially viewed as.

01/19/2025 1
 Transformation
A signal can be converted from time domain
into frequency domain using mathematical
operators called transforms.
There are many kind of transformation that
does this.
Some of them are given below.
Fourier Series
Fourier transformation
Laplace transform
Z transform
Out of all these, we will thoroughly discuss a
Fourier series and Fourier transformation.
01/19/2025 1
 Frequency Components

Any image in spatial domain can be represented in a

frequency domain.
But what do this frequencies actually mean?
We will divide frequency components into two major
components.
High frequency components
High frequency components correspond to edges in
an image. (that change greatly in intensity over short
image distances.)
Low frequency components
Low frequency components in an image correspond
to smooth regions.(that change smoothly in intensity
over short image distances.)
01/19/2025 1
 Fourier Transform
Fourier:-
Fourier was a mathematician in 1822. He give Fourier series
and Fourier transform to convert a signal into frequency
domain.
Any function that is not periodic can be expressed as the
integral of sines and /or cosines
The Fourier Transform is an important image processing tool
which is used to decompose an image into cosine and sine
components
Fourier transform is a decomposition of a signal into some
basic functions.
The output of the transformation represents the image in
Fourier or frequency domain.
The input image is spatial domain equivalent.
In the Fourier domain image, each point represents a
particular
01/19/2025
frequency contained in the spatial domain image.
1
 Cont’d…
The Fourier transform has many wide
applications that include, image compression
(e.g JPEG compression), filtering and image
analysis.

01/19/2025 1
 Fourier Series
It further states that periodic signals can be broken down
into further signals with the following properties.
The signals are sines and cosines
The signals are harmonics of each other

In the above signal, the last signal is actually the sum of
all the above signals. This was the idea of the Fourier.1
01/19/2025
 Cont’d…
Any periodic function can be expressed as the sum of
sines and /or cosines of different frequencies, each
multiplied by a different coefficients
The Fourier series can be denoted by this formula.

The inverse can be calculated by this formula.

01/19/2025 1
 Diff erence Between Fourier
Series And Transform

The difference between them is Fourier series is

applied on periodic signals and Fourier transform is
applied for non periodic signals.
Which one is applied on images?
Images are non – periodic. And since the images
are non periodic, so Fourier transform is used to
convert them into frequency domain.

01/19/2025 1
 Discrete Fourier Transform
Since we are dealing with images, and in fact digital
images, so for digital images we will be working on
discrete fourier transform,
It include three things.
Spatial Frequency
Magnitude
Phase(degree)
The spatial frequency directly relates with the
brightness of the image.
The magnitude of the sinusoid directly relates with the
contrast.
Contrast is the difference between maximum and
minimum pixel intensity.
Phase
01/19/2025contains the color information. 1
Cont’d…

2 dimensional discrete Fourier transform is given

The discrete Fourier transform is actually the

sampled Fourier transform, so it contains some
samples that denotes an image.
In the above formula f(x,y) denotes the image, and
F(u,v) denotes the discrete Fourier transform.

01/19/2025 1
 Cont’d…

The formula for 2 dimensional inverse discrete

Fourier transform is given below.

The inverse discrete Fourier transform converts the

Fourier transform back to the image

01/19/2025 1
 Frequency Information Of
Images
The intensity of an image changes a lot, then it is a
high frequency image.
The brightness level changes quickly from one pixel
to the next.
A low-frequency image may be one that is relatively
uniform in brightness or changes very slowly.

01/19/2025 1
 Frequency Information Of
Images
Most of the images have
both low and high-
frequency parts.
 color of the left image
change very gradually,
which is considered
smooth, low frequency
pattern
On the right stripped
image, it have a high-
frequency image pattern,
this image changes very
rapidly from one
brightness to another

01/19/2025 1
 Cont’d…
Frequency in image is the rate of change of intensity
values.
Thus, a high frequency image is the one where the
intensity values change quickly form one pixel to the
next.
On the other hand a low frequency image may be that
one that relatively uniform in brightness or where
intensity changes very slowly.
Most image contains both high frequency and low
frequency components.

01/19/2025 1
 Image Enhancement In
Frequency Domain
Image enhancement in the frequency domain is
straightforward.
We simply compute the Fourier transform of the image to be
enhanced, multiply the result by a filter (rather than convolve
in the spatial domain), and take the inverse transform to
produce the enhanced image.
The idea of blurring an image by reducing its high frequency
components.
Sharpening an image by increasing the magnitude of its high
frequency components is intuitively easy to understand.
However, computationally, it is often more efficient to
implement these operations as convolutions by small spatial
filters in the spatial domain.
Understanding frequency domain concepts is important, and
leads to enhancement techniques that might not have been
thought
01/19/2025
of by restricting attention to the spatial domain. 1
 Cont’d…
The concept of filtering is easier to visualize in the
frequency domain.
Therefore, enhancement of image f(m,n) can be done
in the frequency domain, based on its DFT F(u, v).
This is particularly useful, if the spatial extent of the
point spread sequence h(m, n) is large. In this case,
the convolution

01/19/2025 1
 CONT’D…
We can therefore directly design a transfer function
H (u, v) and implement the enhancement in the
frequency domain as follows:
The transfer function is multiplied with all pixel. then
convert to Fourier transform

01/19/2025 1
Basic Steps Of Filtering In
Frequency Domain
1. Multiply the input image by (-1) x+y to
center the transform to u = M/2 and v =
N/2
2. Compute F(u,v), the 2-D DFT of the image
from (1)
3. Multiply F(u,v) by a filter function H(u,v)
4. Compute the inverse DFT of the result in
(3)
5. Obtain the real part of the result in (4)
6. Multiply the result in (5) by (-1) x+y to
cancel the multiplication of the input
image.

01/19/2025 1
Cont’d…

01/19/2025 1
 ?
01/19/2025 1