Digital Image Transforms
Dr. Pratishtha Verma
Department of Computer Engineering
NIT Kurukshetra
Need for Image Transforms
Spatial Frequency
Images and their intensity profiles
(a) Image that has one sinusoidal signal
cycle
(b) Image that has multiple sinusoidal
signal cycles
(c) Image that has a non-sinusoidal signal
These intensity profiles are very useful in image processing. The intensity profile is a
collection of either sinusoidal signals such as sine, cosine, and harmonic signals, or
non-sinusoidal waves.
Types of
Image
Transforms
Basis of Transforms
Introduction to
Fourier
Transform
Discrete Fourier Transform
Images are digitized, it is necessary to
have a discrete formulation of the Fourier
transform. This is achieved by the discrete
Fourier transform (DFT), which takes
regularly spaced data values, and returns
the value of the Fourier transform by
replacing the integral by a summation.
Fast Fourier Transform
The 2D DFT can be calculated by applying the 1D DFT for each column of f(x,y),
and then performing a 1D DFT for each row on the resulting complex data.
However, the DFT algorithm is computationally intensive. Hence, a faster version,
the Fast Fourier Transform (FFT), is used.
Image Enhancement
Discussion of image enhancement starts with the concept
of image quality. Many factors are responsible for image
quality:
1. Contrast
2. Brightness
3. Spatial resolution
4. Noise