Chapter 1and2gonzalezandwoods 160204072955

Chapter-1 and 2
Subject: FIP (181102)

Prof. Asodariya Bhavesh
ECD,SSASIT, Surat
Digital Image Processing, 3rd edition by
Gonzalez and Woods
Optics and Human Vision
The physics of light
http://commons.wikimedia.org/wiki/File:Eye-diagram_bg.svg
Light
Light
Particles known as photons
Act as waves
Two fundamental properties
Amplitude
Wavelength
Frequency is the inverse of wavelength
Relationship between wavelength (lambda) and frequency (f)
c/ f
Where c = speed of light = 299,792,458 m / s
4
What is Digital Image Processing?
Digital image processing focuses on two major tasks
Improvement of pictorial information for human interpretation
Processing of image data for storage, transmission and
representation for autonomous machine perception
Some argument about where image processing ends and
fields such as image analysis and computer vision start
What is DIP? (cont)
The continuum from image processing to computer vision
can be broken up into low-, mid- and high-level processes
Low Level Process Mid Level Process High Level Process

Input: Image Input: Image Input: Attributes
Output: Image Output: Attributes Output: Understanding
Examples: Noise Examples: Object Examples: Scene
removal, image recognition, segmentation understanding,
sharpening autonomous navigation
In this course we will stop

here
History of Digital Image Processing
Early 1920s: One of the first applications of digital imaging
was in the news-
Images taken from Gonzalez & Woods, Digital Image Processing (2002)
paper industry
The Bartlane cable picture
transmission service
Images were transferred by submarine cable between London
Early digital image
and New York
Pictures were coded for cable transfer and reconstructed at
the receiving end on a telegraph printer
History of DIP (cont)
Mid to late 1920s: Improvements to the Bartlane system
resulted in higher quality images
New reproduction
processes based
on photographic
techniques
Increased number
of tones in
reproduced images
Improved
digital image Early 15 tone digital image
1960s: Improvements in computing technology and the
onset of the space race led to a surge of work in digital
image processing
1964: Computers used to
improve the quality of
images of the moon taken
by the Ranger 7 probe
Such techniques were used
in other space missions
including the Apollo landings
A picture of the moon taken by

the Ranger 7 probe minutes
before landing
1970s: Digital image processing begins to be used in
medical applications
1979: Sir Godfrey N.

Hounsfield & Prof. Allan M.
Cormack share the Nobel
Prize in medicine for the
invention of tomography,
the technology behind
Computerised Axial
Tomography (CAT) scans
Typical head slice CAT image
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
Key Stages in Digital Image Processing:
Image Aquisition
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Image Enhancement
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Image Restoration
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Morphological Processing
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Segmentation
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Object Recognition
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Representation & Description
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Image Compression
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Colour Image Processing
Image Morphological
Image
Segmentation
Enhancement
Image Object
Representation
Problem Domain
& Description
Colour Image Image
Visible Spectrum
(Images from Rafael C. Gonzalez and Richard E.

Wood, Digital Image Processing, 2nd Edition.
Light
Diagram of a light wave.
22
Conventional Coordinate for Image Representation

Digital Image Types : Intensity Image
Intensity image or monochrome image

each pixel corresponds to light intensity
normally represented in gray scale (gray
level).
Gray scale values

10 10 16 28
9 6 26 37

15 25 13 22

32 15 87 39
Digital Image Types : RGB Image
Color image or RGB image:

each pixel contains a vector
representing red, green and
blue components.
RGB components
10 10 16 28
9 656 70 26
56 43
3756 78
32 99 54 96 67
70
15 256013 902296 67
21 54 47 42
85 85 43 92
32 15 87 39
54 65 65 39
32 65 87 99
Image Types : Binary Image
Binary image or black and white image

Each pixel contains one bit :
1 represent white
0 represents black
Binary data
0 0 0 0
0 0 0 0

1 1 1 1

1 1 1 1
Image Types : Index Image
Index image
Each pixel contains index number
pointing to a color in a color table
Color Table
Index Red Green Blue

component component component
No.
1 0.1 0.5 0.3
2 1.0 0.0 0.0
1 4 9
6 4 7 3 0.0 1.0 0.0
4 0.5 0.5 0.5
6 5 2
5 0.2 0.8 0.9
Index value
Cross Section of the Human Eye

Human Eye
29
Anatomy of the Human Eye
30 Source: http://webvision.med.utah.edu/
Human Visual System
Human vision
Cornea acts as a protective lens that roughly focuses
incoming light
Iris controls the amount of light that enters the eye
The lens sharply focuses incoming light onto the retina
Absorbs both infra-red and ultra-violet light which can damage the
lens
The retina is covered by photoreceptors (light sensors)
which measure light
31
Photoreceptors
Rods
Approximately 100-150 million rods
Non-uniform distribution across the retina
Sensitive to low-light levels (scotopic vision)
Lower resolution
Cones
Approximately 6-7 million cones
Sensitive to higher-light levels (photopic vision)
High resolution
Detect color by the use of 3 different kinds of cones each of
which is sensitive to red, green, or blue frequencies
Red (L cone) : 564-580 nm wavelengths (65% of all cones)
Green (M cone) : 534-545 nm wavelengths (30% of all cones)
Blue (S cone) : 420-440 nm wavelengths (5% of all cones)
33
Cone (LMS) and Rod (R) responses
34 http://en.wikipedia.org/wiki/File:Cone-response.svg
Photoreceptor density across retina
35
Comparison between rods and cones
Rods Cones
Used for night vision Used for day vision
Loss causes night blindness Loss causes legal blindness
Low spatial resolution with higher noise High spatial resolution with lower noise
Not present in fovea Concentrated in fovea
Slower time response to light Quicker time response to light
One type of photosensitive pigment Three types of photosensitive pigment
Emphasis on motion detection Emphasis on detecting fine detail
36
Color and Human Perception
Chromatic light
has a color component
Achromatic light
has no color component
has only one property intensity
37
Image Formation in the Human Eye

(Picture from Microsoft Encarta 2000)

Brightness Adaptation
Actual light intensity is (basically)

log-compressed for perception.
Human vision can see light between

the glare limit and scotopic
threshold but not all levels at the
same time.
The eye adjusts to an average value

(the red dot) and can simultaneously
see all light in a smaller range
surrounding the adaptation level.
Light appears black at the bottom of

the instantaneous range and white
at the top of that range.
39
Weber Ratio I/I
Weber Ratio
Human Visual Perception
Light intensity:
The lowest (darkest) perceptible intensity is the scotopic threshold
The highest (brightest) perceptible intensity is the glare limit
The difference between these two levels is on the order of 1010
We cant discriminate all these intensities at the same time! We adjust
to an average value of light intensities and then discriminate around the
average.
Log compression.
Experimental results show that the relationship between the
perceived amount of light and the actual amount of light in a scene
are generally related logarithmically.
The human visual system perceives brightness as the logarithm of the actual
light intensity and interprets the image accordingly.
Consider, for example, a bright light source that is approximately 6times
brighter than another. The eye will perceive the brighter light as
approximately twice the brightness of the darker.
42
Brightness Adaptation and Mach Banding
When viewing any scene:
The eye rapidly scans across the eld of view while coming
to momentary rest at each point of particular interest.
At each of these points the eye adapts to the average
brightness of the local region surrounding the point of
interest.
This phenomena is known as local brightness adaptation.
Mach banding is a visual effect that results, in part, from local
brightness adaptation.
The eye over-shoots/under-shoots at edges where the brightness
changes rapidly. This causes false perception of the intensities
Examples follow.
43
Brightness Adaptation and Mach Banding
44
Brightness Adaptation(Hermann Grid)
45
46
Optical illusion

Simultaneous Contrast
Simultaneous contrast refers to the way in which two
adjacent intensities (or colors) affect each other.
Example: Note that a blank sheet of paper may appear white
when placed on a desktop but may appear black when used to
shield the eyes against the sun.
Figure 2.9 is a common way of illustrating that the perceived
intensity of a region is dependent upon the contrast of the
region with its local background.
The four inner squares are of identical intensity but are
contextualized by the four surrounding squares
The perceived intensity of the inner squares varies from bright on the
left to dark on the right.
48
Simultaneous Contrast
49
Image Sensing and acquisition
Single sensor
Line sensor
Array sensor

Image Sensors : Single Sensor

Image Sensors : Line Sensor
Fingerprint sweep sensor

Computerized Axial Tomography
Image Sensors : Array Sensor
Charge-Coupled Device (CCD)
w Used for convert a continuous

image into a digital image
w Contains an array of light sensors
w Converts photon into electric charges

accumulated in each sensor unit
CCD KAF-3200E from Kodak.

(2184 x 1472 pixels,
Pixel size 6.8 microns2)
Gate
Vertical Transport Register
Gate
Photosites
Image Sensor: Inside Charge-Coupled Device
Gate
Horizontal Transportation Register
Amplifier Output Gate

Output
Image Sensor: How CCD works
i h g Image pixel
f e d
c b a
i h g
f e d
i h g
Horizontal transport
register c b a
f e d
Vertical shift c b a Output
Horizontal shift
Digital Image Acquisition Process

Generating a Digital Image

Image Sampling and Quantization
Image sampling: discretize an image in the spatial domain

Spatial resolution / image resolution: pixel size or number of pixels

How to choose the spatial resolution
Spatial resolution = Sampling locations

Original image
Sampled image
Under sampling, we lost some image details!

How to choose the spatial resolution : Nyquist Rate
Original image
Sampled image
1mm
2mm
No detail is lost!
Minimum Nyquist Rate:
Spatial resolution
Period Spatial resolution must be less or equal
(sampling rate)
half of the minimum period of the image
or sampling frequency must be greater or
= Sampling locations
Equal twice of the maximum frequency.
Aliased Frequency
x1 (t ) sin( 2t ), f 1
1 x2 (t ) sin( 12t ), f 6
0.5
-0.5
-1
0 0.5 1 1.5 2
Sampling rate:
1
5 samples/sec
0.5
-0.5
-1
0 0.5 1 1.5 2
Two different frequencies but the same results !

Effect of Spatial Resolution
256x256 pixels 128x128 pixels
64x64 pixels 32x32 pixels

Spatial Resolution
It is a measure of the smallest discernible detail in an
image
Can be stated in line pairs per unit distance, and dots(pixels)
per unit distance
Dots per unit distance commonly used in printing and
publishing industry (dots per inch)
Newspaper are printed with a resolution of 75 dpi,
magazines at 133 dpi, and glossy brochures at175 dpi
examples


Can we increase spatial resolution by interpolation ?
Down sampling is an irreversible process.

Image Quantization
Image quantization:
discretize continuous pixel values into discrete numbers
Color resolution/ color depth/ levels:

- No. of colors or gray levels or
- No. of bits representing each pixel value
- No. of colors or gray levels Nc is given by
Nc 2 b
where b = no. of bits

Quantization function
Nc-1
Nc-2
Quantization level
1
0
Light intensity
Darkest Brightest
Intensity Resolution
It refers to the smallest discernible change in intensity
level
Number of intensity levels usually is an integer power of
two
Also refers to Number of bits used to quantize intensity
as the intensity resolution
Which intensity resolution is good for human perception
8 bit, 16 bit, or 32 bit
Effect of Quantization Levels or Intensity resolution
256 levels 128 levels
64 levels 32 levels
Effect of Quantization Levels (cont.)
or Intensity resolution
16 levels 8 levels
In this image,
it is easy to see
false contour.
4 levels 2 levels
How to select the suitable size and pixel depth of images
The word suitable is subjective: depending on subject.
Low detail image Medium detail image High detail image
Lena image Cameraman image
To satisfy human mind

1. For images of the same size, the low detail image may need more pixel depth.
2. As an image size increase, fewer gray levels may be needed.
Isopreference Curve
Curves tend to become more vertical as the detail in the
image increases
Image with a large amount of detail only a few intensity
levels may be needed
Image Interpolation
Used in image resizing (zooming and shrinking), rotating,
and geometric corrections
Interpolation is the process of using known data to
estimate values at unknown locations
Nearest Neighbor interpolation
It assigns to each new location the intensity of its nearest
neighbor in the original image
Produce undesirable artifacts, such as severe distortion of
straight edges
Bilinear Interpolation
We use the four nearest neighbors to estimate the intensity
V(x, y) = ax + by + cxy + d
Image Interpolation
Need to solve four equations
Better results than nearest neighbor interpolation, with a
modest increase in computational burden
Bicubic Interpolation
Involves sixteen neighbors to estimate intensity
V(x, y) = aij xi yj ( i, j = 0 to 3)
Need to solve sixteen equations
Gives better results than other methods
More complex
Used in Adobe Photoshop, and Corel Photopaint
Basic Relationship of Pixels
(0,0) x
(x-1,y-1) (x,y-1) (x+1,y-1)
(x-1,y) (x,y) (x+1,y)

y
(x-1,y+1) (x,y+1) (x+1,y+1)
Conventional indexing method

Neighbors of a Pixel
Neighborhood relation is used to tell adjacent pixels. It is

useful for analyzing regions.
(x,y-1) 4-neighbors of p:
(x-1,y)
(x-1,y) p (x+1,y)
(x+1,y)
N4(p) = (x,y-1)
(x,y+1)
(x,y+1)
4-neighborhood relation considers only vertical and

horizontal neighbors.
Note: q N4(p) implies p N4(q)

Neighbors of a Pixel (cont.)
(x-1,y-1) (x,y-1) (x+1,y-1) 8-neighbors of p:
(x-1,y-1)
(x-1,y) p (x+1,y)
(x,y-1)
(x+1,y-1)
(x-1,y)
(x-1,y+1) (x,y+1) (x+1,y+1) (x+1,y)
N8(p) = (x-1,y+1)
(x,y+1)
(x+1,y+1)
8-neighborhood relation considers all neighbor pixels.

Neighbors of a Pixel (cont.)
(x-1,y-1) (x+1,y-1) Diagonal neighbors of p:
(x-1,y-1)
p
(x+1,y-1)
ND(p) = (x-1,y+1)
(x+1,y+1)
(x-1,y+1) (x+1,y+1)
Diagonal -neighborhood relation considers only diagonal

neighbor pixels.
Connectivity
Connectivity is adapted from neighborhood relation. Two pixels are connected if they
are in the same class (i.e. the same color or the same range of intensity) and they are
neighbors of one another.
For p and q from the same class

w 4-connectivity: p and q are 4-connected if q N4(p)
w 8-connectivity: p and q are 8-connected if q N8(p)
w mixed-connectivity (m-connectivity):
p and q are m-connected if q N4(p) or
q ND(p) and N4(p) N4(q) =
Adjacency
A pixel p is adjacent to pixel q is they are connected.

Two image subsets S1 and S2 are adjacent if some pixel
in S1 is adjacent to some pixel in S2
S1
S2
We can define type of adjacency: 4-adjacency, 8-adjacency

or m-adjacency depending on type of connectivity.
Path
A path from pixel p at (x,y) to pixel q at (s,t) is a sequence

of distinct pixels:
(x0,y0), (x1,y1), (x2,y2),, (xn,yn)
such that
(x0,y0) = (x,y) and (xn,yn) = (s,t)
and
(xi,yi) is adjacent to (xi-1,yi-1), i = 1,,n
q
p
We can define type of path: 4-path, 8-path or m-path

depending on type of adjacency.
Path (cont.)
8-path m-path
p p p
q q q
m-path from p to q
8-path from p to q
solves this ambiguity
results in some ambiguity
Distance
For pixel p, q, and z with coordinates (x,y), (s,t) and (u,v),

D is a distance function or metric if
w D(p,q) 0 (D(p,q) = 0 if and only if p = q)
w D(p,q) = D(q,p)
w D(p,z) D(p,q) + D(q,z)
Example: Euclidean distance
De ( p, q) ( x - s )2 + ( y - t ) 2
Distance (cont.)
D4-distance (city-block distance) is defined as
D4 ( p, q) x - s + y - t
2
2 1 2
2 1 0 1 2
2 1 2
Pixels with D4(p) = 1 is 4-neighbors of p.

Distance (cont.)
D8-distance (chessboard distance) is defined as
D8 ( p, q) max( x - s , y - t )
2 2 2 2 2
2 1 1 1 2
2 1 0 1 2
2 1 1 1 2
2 2 2 2 2
Pixels with D8(p) = 1 is 8-neighbors of p.

89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
Boundary (Border or Contour)
of a region R is the set of points that are adjacent to
points in the complement of R.
of a region is the set of pixels in the region that have at
least one background neighbor.
Inner Border
Outer Border
111
Moire Pattern Effect : Special Case of Sampling
Moire patterns occur when frequencies of two superimposed

periodic patterns are close to each other.

Human vision: Spatial Frequency vs Contrast
Human vision: Distinguish ability for Difference in brightness
Regions with 5% brightness difference

Chapter 1and2gonzalezandwoods 160204072955

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Chapter 1and2gonzalezandwoods 160204072955

Загружено:

Авторское право:

Доступные форматы

Chapter-1 and 2

Subject: FIP (181102)

The physics of light

Low Level Process Mid Level Process High Level Process

In this course we will stop

A picture of the moon taken by

1979: Sir Godfrey N.

(Images from Rafael C. Gonzalez and Richard E.

Diagram of a light wave.

(Images from Rafael C. Gonzalez and Richard E.

Intensity image or monochrome image

Gray scale values

Color image or RGB image:

Binary image or black and white image

Index Red Green Blue

(Images from Rafael C. Gonzalez and Richard E.

(Images from Rafael C. Gonzalez and Richard E.

(Picture from Microsoft Encarta 2000)

Actual light intensity is (basically)

Human vision can see light between

The eye adjusts to an average value

Light appears black at the bottom of

(Images from Rafael C. Gonzalez and Richard E.

(Images from Rafael C. Gonzalez and Richard E.

(Images from Rafael C. Gonzalez and Richard E.

Fingerprint sweep sensor

Charge-Coupled Device (CCD)

w Used for convert a continuous

w Contains an array of light sensors

w Converts photon into electric charges

CCD KAF-3200E from Kodak.

Vertical Transport Register

Amplifier Output Gate

Vertical shift c b a Output

(Images from Rafael C. Gonzalez and Richard E.

(Images from Rafael C. Gonzalez and Richard E.

Image sampling: discretize an image in the spatial domain

(Images from Rafael C. Gonzalez and Richard E.

Spatial resolution = Sampling locations

Under sampling, we lost some image details!

Two different frequencies but the same results !

256x256 pixels 128x128 pixels

64x64 pixels 32x32 pixels

(Images from Rafael C. Gonzalez and Richard E.

(Images from Rafael C. Gonzalez and Richard E.

Down sampling is an irreversible process.

Color resolution/ color depth/ levels:

where b = no. of bits

256 levels 128 levels

The word suitable is subjective: depending on subject.

Low detail image Medium detail image High detail image

Lena image Cameraman image

To satisfy human mind

(x-1,y-1) (x,y-1) (x+1,y-1)

(x-1,y) (x,y) (x+1,y)

(x-1,y+1) (x,y+1) (x+1,y+1)

Conventional indexing method

Neighborhood relation is used to tell adjacent pixels. It is

4-neighborhood relation considers only vertical and

Note: q N4(p) implies p N4(q)

(x-1,y-1) (x,y-1) (x+1,y-1) 8-neighbors of p:

8-neighborhood relation considers all neighbor pixels.

(x-1,y-1) (x+1,y-1) Diagonal neighbors of p:

Diagonal -neighborhood relation considers only diagonal