Practical Image and Video Processing
Using MATLAB®
Introduction and overview
What will we learn?
What is image processing?
What are the main applications of image processing?
What is an image?
What is a digital image?
What are the goals of image processing algorithms?
What are the most common image processing operations?
Which hardware and software components are typically
needed to build an image processing system?
What is a machine vision system (MVS) and what are its
main components?
Why is it so hard to emulate the performance of the human
visual system (HVS) using cameras and computers?
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Motivation
Vision is our most developed sense
Most frequently used
Influences our overall perception
Substantial amount of brain processing is devoted to it
The ability to guide our actions and engage our
cognitive abilities based on visual input is a remarkable
trait of the human species…
… but much of “how exactly we do what we do” remains to
be discovered.
“A picture is worth a thousand words.”
The ability to automatically extract semantic info. from
an image is an open and actively investigated research
problem (extraction and descriptive interpretation of image contents).
Examples of applications
Medical Applications
PET, CAT scans, MRI and fMRI, etc.
Industrial Applications
OCR, Gauging, Presence Verification etc.
Consumer Electronics
Digtial cameras, Smart phones, HDTVs (Video compression) etc.
Military Applications
Soldier / Vehicle Detection
Missile Guidance
Object Recognition
Reconnaissance Tasks (UAVs)
Range Cameras / FLIR Cameras
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Examples of applications
Law Enforcement and Security
Surveillance applications
Biometric Techniques (Fingerprint, Face, Iris and hand recognition)
Internet, particularly the Web.
Huge amount of visual contents
Sharing; Annotation (Tagging);
Search by image;
Content based image retrieval (CBIR) ;
Image / Video Compression
Basic concepts
What is an image?
A visual representation of an object, a person, or a scene
produced by an optical device such as a mirror, a lens, or a
camera.
A few remarks:
This representation is typically 2D, although it usually
corresponds to one of infinitely many projections of a real world,
3D object or scene.
This definition implicitly assumes the existence of a light source
illuminating the scene, which is a requirement for the image to
be produced.
An image means something, in other words, it is not a random
arrangements of dark and bright points.
Basic concepts
What is a digital image?
A digital image is a representation of a two-
dimensional image using a finite number of points,
usually referred to as picture elements, or pixels.
A few remarks:
Each pixel is represented by one or more numerical
values:
For monochrome (grayscale) images, a single value
representing the intensity of the pixel (usually in a [0, 255]
range) is enough;
For color images, three values (usually representing the
amount of red (R), green (G), and blue (B)) are required.
Basic concepts
What is digital image processing?
It is the science of modifying digital images by means
of a digital computer.
A few remarks:
Since both the images and the computers that process
them are digital in nature, here we will focus
exclusively on digital image processing.
The changes that take place in the images are usually
performed automatically and rely on carefully
designed algorithms to carry out such tasks.
Image Processing is in clear contrast with Image
Editing / Image Manipulation where images are
processed manually (e.g., using softwares like
photoshop or corel draw etc)
Basic concepts
What is digital image processing?
It is the science of modifying digital images by means of a
digital computer.
Image processing refer to all techniques and applications
whether the output is :
A modified (processed) version of input image e.g. Noise Reduction
An encoded version of its main attributes e.g., Edge Detection
Non pictorial description of its contents e.g., License Plate Recognition
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Basic concepts
What are the goals of image processing algorithms?
Image processing algorithms are usually designed to improve the
suitability of the image in order to either:
enable human interpretation (e.g., medical diagnosis), or
make it more suitable to further analysis and automatic extraction of
some of its contents (e.g., improving the contrast b/w characters and
background before it is examined by an OCR algorithm).
Sometimes these goals can be at odds with each other.
Example:
Sharpening an image to allow inspection of additional fine-
grained details (better for human viewing)
vs.
Blurring an image to reduce the amount of irrelevant information
(better for a machine vision solution).
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Basic concepts
3 levels of image processing operations:
Low- level: primitive operations (e.g., noise
reduction, contrast enhancement, etc.) where both the
input and output are images.
Mid-level: extraction of attributes (e.g., edges,
contours, regions, etc.) from images.
High-level: analysis and interpretation of the
contents of a scene (e.g, OCR)
Examples of image processing Operations
Sharpening: Technique by which the edges and fine details are
enhanced for human viewing
Can be done in Spatial as well as in Frequency domain
Examples of image processing Operations
Noise removal
Different noise removal filters (techniqes) depending on noise type
Used before further processing
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Examples of image processing Operations
Deblurring:
Improper lens focusing
Insufficient shutter speed for fast moving object
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Examples of image processing Operations
Edge extraction
a fundamental pre-processing step used to separate objects from one another
before identifying their contents
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Examples of image processing Operations
Binarization: reduction of no. of gray levels to only
two.
Used for simplification and speeding up interpretation
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Examples of image processing Operations
Blurring: We blur an image to minimize the
importance of texture and fine detials (e.g. when objects can
be better recognized by their shapes)
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Examples of image processing Operations
Contrast enhancement:
Used to improve an image for human viewing
Used to improve an image for other image processing tasks e.g. to make edge
extraction easier
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Examples of image processing Operations
Object segmentation and labeling
Pre-requisite for object recognition and classification
Segmentation Labeling Feature Extraction Classify / Compare /
Cluster / recognize
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Generic Digital Image Processing System
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Generic Digital Image Processing System
Hardware
Acquisition devices: responsible for capturing and digitizing
images / videos
Exs: scanners, sensors, cameras, camcorders, etc.
Processing equipment: responsible for running software that
allows the processing and analysis of acquired images
Exs: computers, workstations, specialized hardware, etc.
Display and hardcopy devices: responsible for showing the
image contents for human viewing
Exs: monitors, printers, etc.
Storage devices: responsible for long-term storage of images
Exs: magnetic disks, optical disks, etc.
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Generic Digital Image Processing System
Software
Modules that perform specialized tasks, e.g.:
MATLAB and its toolboxes.
Java, ImageJ, and its plugins.
OpenCV
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Machine Vision Systems
Ex: License Plate Recognitions
contrast improvement, Representation Recognition
brightness corrction etc and description and interpretation
Digital Feature
Image Vector
MVS vs. HVS
Why is it so hard to emulate the performance
of the human visual system (HVS) using
cameras and computers?
Very large database of images and associated
concepts
Very high speed
Ability to work under a wide range of conditions
Most MVS must impose numerous constraints on the operating
conditions of the scene to improve their chances of success.
THANK YOU
By Oge Marques Copyright © 2011 by John Wiley & Sons, Inc. All rights reserved.
Distribution of Marks:
Evaluation Component Weightage
End Semester Exam 60%
Mid Semester Exam 20%
Quizzes / Assignments 10%
Case Study / Paper 10%
Books:
Marques
Gonzale
Jain
