You are on page 1of 33

Computer Vision and Image

processing (CoSc4151)
Chapter One
Introduction to computer vision and image
processing
Jimma University
Jimma Institute of Technology
For Computer Science students
CV&IP Abel W.
Contact information
Name of instructor: Abel Worku (BSc.)
Email: abelworku1221@gmail.com
Office no: 5308 (Varnero building).

CV&IP 04-Nov-16 2
Course objectives
At the end of this course, students are expected to:
 Define fundamental concepts of computer vision and
image processing system
 Apply pre-processing operations in image enhancement
 Analyze 2-dimensional signals in the frequency domain
through the Fourier transform
 Describe the basic concepts of color image processing
 Apply image restoration, reconstruction and
compression operations
 Design and implement image processing algorithm
with Matlab.
CV&IP 04-Nov-16 3
Course breakdown
Chapter 1. Introduction to computer Chapter 2. Digital image
vision and image processing fundamentals

 What is a computer vision?  Basic concept of image


 What is image?
 Digital image Representation
 Related fields in CV.
 Digital image acquisition process
 Computer Vision Vs image
 Image sampling and quantization
processing

 Application of CV and IP
 Representation of different image
types
 Different Image processing
examples  Mathematical Tools used in
Digital Image Processing
 Fundamental steps in image
processing
CV&IP 04-Nov-16 4
Cont’d…
3. Spatial domain 4. Frequency domain
image processing image processing
 Spatial Processing of Digital  Introduction to frequency domain
Images  What is sound?

 Basic Intensity Transformation  Frequency spectra


Functions  Fourier series
 Continuous One-
 Histogram of images Dimensional Fourier
Transform and Its Inverse
 Spatial filtering
 Discrete One-Dimensional
 Histogram processing Fourier Transform and Its
Inverse
 Combining Spatial
 Frequency information of images
Enhancement Methods
 Image enhancement in frequency
domain

CV&IP 04-Nov-16 5
Cont’d…
Chapter 5. Image restoration and Chapter 6. Image compression
reconstruction
 Basic definition of image
 Introduction compression
 Models of images and  Data redundancy
noise.  Elements of information
theory
 Estimation of noise  General mechanism and
parameters types of data compression
and image restoration
 Restoration in the  Huffman coding
presence of noise  Arithmetic coding
 Dictionary based coding
 Bit-plane coding
CV&IP 04-Nov-16 6
Cont’d…
Chapter 7. Color image processing basics

 Color Fundamentals
 Color Models (color
space or color system)
 Basics of Full-Color
Image Processing
 Color Transformations

CV&IP 04-Nov-16 7
Lab activities
Activity one: Introduction to matlab basics.
Activity two: Basic arithmetic operations over an image.
Activity three: Image Cropping, Resizing, Flipping, and Rotation
Activity four: Gray-level transformations
Activity five: Basic image filtering and compression
Activity six: Image histograms
Activity seven: Histogram processing
Activity eight: Color image processing basics
Activity nine: Basic image processing using C# (OPTIONAL)

CV&IP 04-Nov-16 8
Methodology, grading and references
Methodology: lecture, lab practice using matlab and
C#, class activity
Grading
 Class work, lab-report, quizzes, attendance ...………………...…. 35%
 Mid-exam ……………………………………………………….. 25%
 Final-exam ……………………………………………………… 40%
References/ literatures:
1. 1. R. C. Gonzalez, R. E. Woods, S. L. Eddins,’Digital Image
Processing using Matlab,’ Pearson-Prentice-Hall, 2004
2. William K. Pratt, Digital Image Processing, 3rd edition, 2001

CV&IP 04-Nov-16 9
Introduction to computer vision and
image processing

What is vision?
what is computer vision?
What is image processing?

CV&IP 04-Nov-16 10
Vision
Vision is “to know what is where, by looking.’’
 Where?
 What?

CV&IP 04-Nov-16 11
Why is Vision Interesting?
 Psychology
• ~ 50% of cerebral cortex is for vision.
• Vision is how we experience the world.
 Engineering
• Want machines to interact with world.
• Digital images are everywhere.

CV&IP 04-Nov-16 12
What is computer vision?
 CV is a science that deals how computer can be
made to gain high-level of understanding from the
digital image or video.
 From engineering point of view it seeks to automate
the tasks performed by human vision system.
 It concerns about the theory how artificial systems
can extract information from images.
 CV studies and describes the processes implemented
in software and hardware behind the artificial vision
systems.
CV&IP 04-Nov-16 13
Cont’d…
 The ultimate goal of computer vision is to model,
replicate, and more importantly exceed human vision
using computer software and hardware at different
levels.
 It needs knowledge in computer science, electrical
engineering, mathematics, physiology, biology, and
cognitive science.
 It is inverse of computer graphics (2D-3D).

CV&IP 04-Nov-16 14
Limitations of Human Vision
 Limited memory-cannot remember a quickly
flashed image
 Limited to visible spectrum
 Illusion (false or unreal perception)

CV&IP 04-Nov-16 15
CV&IP 04-Nov-16 16

Copyright A.Kitaoka 2003


What is Image?
 stored visual information that has a similar appearance to
some subject, usually a physical object or a person.

 Produced either by capturing or rendering.

CV&IP 04-Nov-16 17
What is an image? Cont’d…
 We can think of an image as a function, f, from R2 to R:
• ( x, y) be the coordinate in the space or in a plane
• f( x, y ) gives the intensity or gray-level at position ( x, y )
• Realistically, we expect the image only to be defined over a rectangle,
with a finite range:
A color image is just three functions pasted together. We can
write this as a “vector-valued” function:
 r ( x, y ) 
f ( x, y )   g ( x, y ) 
 
CV&IP 
 b ( x, y ) 
 04-Nov-16 18
Related fields with CV
1. Artificial intelligence (AI)
 Deals about robotic navigation
 Detailed understanding of environment is required to
navigate the robot, needs CV (providing high level
information about the env’t) and act as a vision sensor for
the robotic system.
 AI and CV shares topics like pattern recognition, so CV is
sometimes a part of AI.

CV&IP 04-Nov-16 19
Related fields with CV
2. Solid state Physics
 Most CV systems relay on image sensors which detect
electromagnetic radiation.
 Those sensors are designed by quantum physics.
 The process by which light interacts with surface is
explained by physics.
 Physics also explain optics, which is the most part of
imaging system.
 Image sensors require physics to provide complete
understanding of image formation process.

CV&IP 04-Nov-16 20
Related fields with CV
3. Neurobiology
 Specifically the study of biological vision system.
 The CV researches are very close to the study of
biological vision.
 The principles present in CV are very close to
biological vision system.

CV&IP 04-Nov-16 21
Related fields with CV
4. Signal processing
 Deals about methods of processing signals for specific
application.
 Image processing is signal processing which is applied to
images to increase the quality of information.
 CV uses image processing to get the ultimate goals in
understanding, interpreting, analyzing images.
Others related fields:- statistics, optimization, geometry,
computer graphics mathematics and others.

CV&IP 04-Nov-16 22
Computer Vision Vs image
processing
Image processing ;- studies about image-to-image
transformation. The input and output of image
processing are both images.
Some image processing operations include
 Image compression
 Image restoration
 Image enhancement
• Most computer vision algorithms usually assumes a
significant amount of image processing has taken
place to improve image quality.
CV&IP 04-Nov-16 23
Computer Vision Vs Image
Processing (cont’d)
Computer vision is the construction of explicit,
meaningful descriptions of physical objects from their
images.
 The output of computer vision are a description or an
interpretation or some quantitative measurements of the
structures in the 3D scene.
 Image processing and pattern recognition are among many
techniques that computer vision employs/uses to achieve
its goal

CV&IP 04-Nov-16 24
Application of CV and IP
 Robotics (Localization, obstacle removal, Vision based GPS,
Manipulation in manufacturing….)
 Medicine (tumor detection, organ reconstruction,
segmentation, vision guided surgery…)
 Security (defense, fingerprint, face recognition, iris….)
 Transportation (autonomous vehicle, safety, driver
monitoring,…)
 Industrial automation ( defect identification, barcode reading,
sorting document understanding (OCT)….)
 Image/video databases (image retrieval based on its
contents….)
 Human Computer Interface (gaze estimation, Face
expression recognition head and hand gesture recognition….)

CV&IP 04-Nov-16 25
Application of vision

Safety Health Security

CV&IP
Comfort Fun 04-Nov-16
Access
26
Optical character recognition (OCR)
Technology to convert scanned docs to text
• If you have a scanner, it probably came with OCR software

Digit recognition, AT&T labs License plate readers


http://www.research.att.com/~yann/ http://en.wikipedia.org/wiki/Automatic_number_plate_reco
gnition
CV&IP 04-Nov-16 27
Face detection

• Many new digital cameras now detect faces


• Canon, Sony, Fuji, …
CV&IP 04-Nov-16 28
Robotics

NASA’s Mars Spirit Rover http://www.robocup.org/


http://en.wikipedia.org/wiki/Spirit_rover
CV&IP 04-Nov-16 29
Medical imaging

Image guided surgery


3D imaging
Grimson et al., MIT
MRI, CT

CV&IP 04-Nov-16 30
Target Recognition (military)
• Department of Defense (Army, Airforce, )

CV&IP 04-Nov-16 31
Fundamental steps in image
processing

CV&IP 04-Nov-16 32
Thank you!!!

CV&IP 04-Nov-16 33

You might also like