Lecture No. 1 introduction.pptx

1. 1
Computer Imaging
 Digital Image Processing
 Computer Vision
Introduction
“One picture is worth more than ten thousand words”

2. 2
Computer Imaging
 Image analysis involves examination of image
data to facilitate solving an imaging problem
 A computer vision application can be
considered to be a deployed image analysis
system
 In a computer vision system the images are
intended to be “seen” by a computer

3. 3
Image Processing Computer Vision
Input  Image
Output  Image (for human
consumption)
Topics
Image restoration
Image enhancement
Image compression
Image morphing, etc.
Input  Image
Output  Image attributes (for
computer use)
Topics
Image segmentation
Image recognition (feature
extraction)
Motion analysis & tracking
3D shape reconstruction, etc.
Computer Imaging
Computer Imaging can be separated into two different but
overlapping areas:

4. 4
Digital images:
A pixel has a location (the spatial coordinate) and a
value.
Digital images can be from many types of sources.
Digital image processing
Digital image processing is closely related to
signal processing at one end, and to computer
vision at the other end.
Digital Image Processing (DIP)

5. 5
Computer vision is concerned with making
computers “understand” what is contained in an
image and other functions of human vision.
It is sometimes considered to be a topic of artificial
intelligence.
However, it is not possible to state exactly where
image processing ends and computer vision begins.
If you compare a digital image processing book
and a computer vision book, you will find a lot of
overlap.
Computer Vision

6. 6
Related Fields

7. 7
Why DIP?
 Hospitals (Medical Imaging: Radiology - MRI, CT Scan, X-Rays, Ultrasound, Retinopathy
and other Pathologies)
 Forensic Departments (like Punjab Forensic Science Agency: Firearms identification, lab. tests,
Image & video forensics)
 Software Houses (Games developments including Kinect, Touch applications)
 Electronic & Print Media (Artistic effects, Morphology)
 Security Agencies, Industries & Offices (Biometrics, Surveillance, Object & Speed
recognition)
 Pakistan Air Force & NESCOM (UAV: drone project with NESCOM)
 GIS (SUPARCO: Satellite Imagery; Terrain Classification; Meteorology; Weather)
 OCR (text, number plate recognitions etc.)
 MRD (Multidisciplinary Research & Development)

8. 8
Defect Detection in Microdisplay Chips
Captured
image
After
defect
detection
After blob
analysis
 Inspection speed vastly increased
 Much lower error rates
 More complete analysis
 Cost savings

9. 9
Image Processing
Computer imaging application involving a
human being in the visual loop
Consists of following major fields:
o Image enhancement
o Image restoration
o Image compression

10. 10
Distorted image Restored image
Geometric distorted image Restored image
Image Restoration

11. 11
Image Enhancement
Poor contrast image Enhanced image
Original image Sharpened image

12. 12
Restored image
Noisy image
Image Restoration

13. 13
13
Image Compression
Original image Compressed image(1/10)
Compressed image(1/20) Compressed image(1/30)

14. 14
1. Computer Aided Inspection of Grains
2. Capturing outlines of planar images
3. Background adjustment of a digital photograph
4. Area of an irregular image
5. Traffic signal alert
6. Safe drive alert
7. Transformation and restoration of B &W photograph into fine &
colored picture
8. Enhancement of satellite images taken in cloudy area
9. Computer Aided Automatic Grading of Physical Models
10. Image & video compression
11. Image audio transformation system for blind people
12. Diabetes retinopathy on digital fundus images
13. Firearm identification
Industrial & Research Projects

15. 15
MRI brain images classification
Classification of brain MRI images using LDyWT (Lifting
Dyadic Wavelet Transform), Genetic algorithms, and SVM.
Denoising Medical Images
To investigate for a denoising method for medical images using
dyadic wavelet transform (DyWT) and one-dimensional
singularity function model.
Blood Vessel Segmentation in Retinal Images
To investigate for blood vessel segmentation method in retinal
images using A Generative Model for Image Segmentation
Based on Label Fusion
Industrial & Research Projects

16. 16
Every picture tells a story
Can a computer infer what happened from the image?

17. 17
The goal of Computer Imaging

18. 18
Can computers match (or beat) human vision?
Yes and no (but mostly no!)
humans are much better at “hard” things
computers can be better at “easy” things

19. 19
Why Study Computer Imaging?
Millions of images being captured all the time
Lots of useful applications
The next slides show the current state of the art

20. 20
Optical character recognition (OCR)
Digit recognition, AT&T labs
http://www.research.att.com/~yann/
Technology to convert scanned docs to text
• If you have a scanner, it probably came with OCR software
License plate readers
http://en.wikipedia.org/wiki/Automatic_number_plate_recognition

21. 21
Face detection
Many new digital cameras now detect faces
Canon, Sony, Fuji, …

22. 22
Smile detection?
Sony Cyber-shot® T70 Digital Still Camera

23. 23
Face recognition
Who is she?
Sharbat Gula at
age 12 in an
Afgan refugee
camp in 1984
Traced in 2002
but is she the
same person?

24. 24
Vision-based biometrics
“How the Afghan Girl was Identified by Her Iris Patterns” Read the story
1984 2002

25. 25
Login without a password…
Fingerprint scanners on
many new laptops,
other devices
Face recognition systems now
beginning to appear more widely
http://www.sensiblevision.com/

26. 26
Object recognition (in mobile phones)
This is becoming real:
Microsoft Research
Point & Find, Nokia

27. 27
Earth viewers (3D modeling)
Image from Microsoft’s Virtual Earth
(see also: Google Earth)

28. 28
Phototourism
Automatic 3D reconstruction from Internet photo
collections
“Statue of Liberty”
3D model
Flickr photos
“Half Dome, Yosemite” “Colosseum, Rome”

29. 30
Smart cars
Mobileye
Vision systems currently in high-end BMW, GM, Volvo models
By 2010: 70% of car manufacturers.

30. 31
Vision-based interaction (and games)
Nintendo Wii has camera-based IR
tracking built in. See Lee’s work at
CMU on clever tricks on using it to
create a multi-touch display!
Digimask: put your face on a 3D avatar.
“Game turns moviegoers into Human Joysticks”, CNET
Camera tracking a crowd, based on this work.

31. 32
Vision in space
Vision systems (JPL) used for several tasks
• Panorama stitching
• 3D terrain modeling
• Obstacle detection, position tracking
• For more, read “Computer Vision on Mars” by Matthies et al.
NASA'S Mars Exploration Rover Spirit captured this westward view from atop
a low plateau where Spirit spent the closing months of 2007.

32. 33
Robotics
http://www.robocup.org/
NASA’s Mars Spirit Rover
http://en.wikipedia.org/wiki/Spirit_rover

33. 34
Medical imaging
Image guided surgery
Grimson et al., MIT
3D imaging
MRI, CT

34. 35
Current state of the art
You just saw examples of current systems.
Many of these are less than 5 years old
This is a very active research area, and rapidly changing
Many new apps in the next 5 years
To learn more about imaging applications and companies
David Lowe maintains an excellent overview of
vision companies
http://www.cs.ubc.ca/spider/lowe/vision.html

35. 36
Objectives
 To familiarize you with the basic techniques and
terminology of Computer Imaging
 To provide the sufficient background for FYPs on
computer imaging
 To ready you for learning the advanced topics of
Computer Imaging course at post graduate level, like
wavelets, camera models and calibration, motion
analysis and tracking, video understanding, image
mosaics, 3D-shape reconstruction, etc.
 To excite & stimulate you!