Digital Image Processing:

Revision

Brian Mac Namee

[email protected]

Course Website: http://www.comp.dit.ie/bmacnamee

2
of
27
Key Stages in Digital Image Processing

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 3
of
Key Stages in Digital Image Processing:
27 Image Aquisition
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 4
of
Key Stages in Digital Image Processing:
27 Image Enhancement
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 5
of
Key Stages in Digital Image Processing:
27 Image Restoration
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 6
of
Key Stages in Digital Image Processing:
27 Morphological Processing
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 7
of
Key Stages in Digital Image Processing:
27 Segmentation
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 8
of
Key Stages in Digital Image Processing:
27 Object Recognition
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 9
of
Key Stages in Digital Image Processing:
27 Representation & Description
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
10
of
Key Stages in Digital Image Processing:
27 Image Compression

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
 11
of
Key Stages in Digital Image Processing:
27 Colour Image Processing
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression
12
of
27
ACHTUNG!

THIS IS NOT A LIST OF

WHAT IS COMING UP IN
YOUR EXAM – DO NOT
COMPLAIN!
13
of
27
Introduction to Image Processing
What is a digital image?
What is digital image processing?
History of digital image processing
Digital image processing application areas
Key stages in digital image processing
14
of
27
Digital Image Processing Fundamentals

The human visual system

Light and the electromagnetic spectrum
Image acquisition
Image sampling and quantisation
Spatial and intensity level resolution
15
of
27
Image Enhancement
Enhancement in the spatial and frequency
domains
Point processing
– Log transformation
– Power law transformation
Histograms
– What is an image histogram?
– Histogram equalisation
16
of
27
Spatial Filtering
Spatial filtering process
– Can you explain how it works?
Smoothing filters
Problems at image edges during filtering
– Padding and different padding techniques
Difference between correlation and
convolution
17
of
27
Spatial Filtering (cont…)
Spatial differentiation
f
– 1st derivative  f ( x  1)  f ( x)
x
2
 f
– 2nd derivative 2
 f ( x  1)  f ( x  1)  2 f ( x)
 x
Differentiation based filters
0 1 0 -1 -2 -1 -1 0 1 You don’t need to know
1 -4 1 0 0 0 -2 0 2 the maths used to derive
0 1 0 1 2 1 -1 0 1 these filters
Laplacian Sobel

How to do sharpening using these filters

18
of
27
Frequency Domain Filtering
The Fourier transform
– Be able to explain the big idea behind it
– You do not need to know the maths for it
– Importance of the inverse Fourier transform
How filtering in the frequency domain works
Low pass filters
– What are they for?
– Ideal low pass filter You don’t need to know
the equations for these,
– Butterworth low pass filter but you must be able to
– Gaussian low pass filter draw them and explain
what they do
19
of
27
Frequency Domain Filtering (cont…)
High pass filters
– What are they for?
– Ideal high pass filter You don’t need to know
the equations for these,
– Butterworth high pass filter but you must be able to
draw them and explain
– Gaussian high pass filter what they do
The Fast Fourier Transform and its
importance
20
of
27
Image Restoration: Noise Removal
Image enhancement vs. image restoration
What is meant by noise removal?
What is meant by a noise model?
g ( x, y )  f ( x, y )   ( x, y )
– Common noise models
• Gaussian • Exponential
• Rayleigh • Uniform
• Erlang • Impulse (salt & pepper)
Filtering to remove noise
– Simple mean filter
– Other mean filters
21
of
Image Restoration: Noise Removal
27 (cont…)
Order statistics filters
– Median filter
– Max and min filter
– Midpoint filter
– Alpha trimmed mean filter
Removing noise in the frequency domain
– Particularly good for removing periodic noise
– Band reject filters
• Ideal band reject filter
• Butterworth band reject filter
• Gaussian band reject filter
22
of
27
Image Segmentation: Thresholding
The segmentation problem
Importance of good thresholding
Problems that can arise with thresholding
The basic global thresholding algorithm
Single value thresholding vs. multiple value
thresholding
Basic adaptive thresholding
23
of
27
Morphological Image Processing
Basic morphological concepts and operations
– Hitting, fitting and missing
– Erosion and dilation
– Opening and closing
Morphological algorithms
– Boundary extraction
– Region filling
24
of
27
Your Exam!
Follows the same format as previous years
4 questions in each section, attempt any 3
questions from each section
Read the questions carefully
DO THE PAST PAPERS – available online
25
of
27
Summary
There are two main jobs in image processing
– Enhancement of images for human viewing
– Preparation of images for machine processing
Both of these are hard areas to work in!
We have covered a lot of the first area, and a
little of the second
The subject of machine vision is huge, growing
and really interesting
26
of
27

Thank you very

much for listening
and good luck in
your exams