CSE4227 Digital Image Processing

Lecture 02 – Chapter 2: Digital Image

Fundamentals

Dr. Kazi A Kalpoma

Professor, Department of CSE
Ahsanullah University of Science & Technology (AUST)

Contact: [email protected]
Google Class code: 6znzyyh

CSE | AUST Spring 2023

Today’s Contents

 Image Sensing and Acquisition

 Sampling and quantization

 Image Representation

 Spatial and Intensity Resolution

Chapter 2 from R.C. Gonzalez and R.E. Woods, Digital Image Processing
(3rd Edition), Prentice Hall, 2008 [ Section 2.3, 2.4 ]
Image Formation

Digital Camera

Film

The Eye
IMAGE SENSING AND ACQUISITION
 Image Acquisition

 Three main elements

1. Illumination source
2. Scene
3. Sensor (imaging system)
 Image Acquisition

1. Illumination source:
– Can be light energy or
– EM spectrum
– Even less tradition sources like
• Sound, heat
2. Scene:
– Any object: visible or hidden
– Source itself
3. Sensor:
– Should be capable of sensing the energy
Light And The Electromagnetic Spectrum

In 1666 Sir Isaac Newton discovered that sunlight passed

through a prism splits into a continuous spectrum of colors

Light is just a particular part of the electromagnetic

spectrum that can be sensed by the human eye
The electromagnetic spectrum is split up according to the
wavelengths of different forms of energy
 A discrete bundle (or quantum) of electromagnetic (or light) energy,
PHOTON is proportional to frequency.
A photon’s life choices
 Absorption

 Diffusion light source

 Reflection
λ
 Transparency

 Refraction
?
 Fluorescence

 Subsurface scattering

 Phosphorescence
Chapter 2 of Szeliski
 Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Reflection
• Transparency
λ
• Refraction
• Fluorescence
• Subsurface scattering
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffuse Reflection
light source
• Reflection
• Transparency
λ
• Refraction
• Fluorescence
• Subsurface scattering
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Specular Reflection
• Transparency
λ
• Refraction
• Fluorescence
• Subsurface scattering
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Reflection
• Transparency
λ
• Refraction
• Fluorescence
• Subsurface scattering
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Reflection
• Transparency
λ
• Refraction
• Fluorescence
• Subsurface scattering
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Reflection
• Transparency
λ1
• Refraction
• Fluorescence
λ2
• Subsurface scattering
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Reflection
• Transparency
λ
• Refraction
• Fluorescence
• Subsurface scattering
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Reflection
• Transparency
t=1
• Refraction
• Fluorescence
• Subsurface scattering t=n
• Phosphorescence
• Interreflection
 A photon’ s life choices
• Absorption
• Diffusion
light source
• Reflection
• Transparency
λ
• Refraction
• Fluorescence
• Subsurface scattering
• Phosphorescence
• Interreflection

(Specular Interreflection)
 Image Sensors
 Incoming energy lands on a sensor material responsive to that type
of energy and this generates a voltage
Collections of sensors are arranged to capture images

Charge-Coupled Device (CCD) or

Complementary Metal Oxide Semiconductor (CMOS)
Used for convert a continuous
image into a digital image

Contains an array of light sensors

Converts photon into electric charges

accumulated in each sensor unit

CCD KAF-3200E from Kodak.

(2184 x 1472 pixels,
Pixel size 6.8 microns2)
 Image Sensors used to transform illumination energy to
digital images.
 3 main sensor arrangements:
1. Single sensor
2. Line sensor
3. Array sensor

Single sensor Array sensor

Line sensor

 Sensing element can be a photodiode

 Filter absorbs extra energy or acts as pass-band
Image Acquisition using Sensor Array

 Images are typically generated by illuminating a scene and

absorbing the energy reflected by the objects in that scene.
Image Digitization
 Computers use discrete form of the pictures

 The process transforming continuous analog image into

discrete approximated image is called digitization.

Transforming
Analog image Digital image
Process

 Two steps:
– Sampling and
– Quantization
Image Sampling And Quantisation

Sampling:
 Digitizing coordinates
 A process which converts the continuous analog space into a
discrete space

Quantization:
 Digitizing amplitudes (gray scale values)
 A process of converting a continuous analogue signal into a
digital representation of that signal
Image Sampling

1-Dimensional Sampling

Digitizing the coordinate values

Image Quantization

7
6
5
4
3
2
1
0

Digitizing the amplitude values

Image Sampling And Quantisation

7
6
5
4
3
2
1
0

A 7 7 7 1 2 3 4 4 5 5 5 5 5 4 3 3 2 1 0 0 7 7 7B

Digitization
Digital Image

 Remember that a digital image is always only

an approximation of a real world scene
Representation of Digital Image

Matrix Representation
Representation of Digital Image
Representation of Digital Image
Representation of Digital Image

Formally,
• Sampling:
– Dividing (x, y) plane into grid with coordinate (zi, zj),
where,
zi , z j  Z
• Quantization:
– Assign gray level value from Z to f(zi, zj)
Spatial Resolution

 Spatial resolution: M x N
Number of Rows (M) and Columns (N) can be any integer
 Bit size of an image: b  M  N k
 For Square image, bN k 2

An image that is 2048 pixels in width and 1536 pixels in height has a
total of 2048×1536 = 3,145,728 pixels or 3.1 megapixels.
Here, spatial resolution is 3.1-megapixel.
One could refer to it as 2048 by 1536 or a 3.1-megapixel image.
Intensity (Gray) Level Resolution

Intensity level resolution refers to the number of

intensity levels used to represent the image

Intensity (Gray) level resolution: L

 Number of gray levels, L, is usually power of 2:

k = Bit depth L2 k

Gray level range = [ 0 – L-1 ]

= [ 0 – 255 ]
Intensity (Gray) Level Resolution
 The more intensity levels used, the finer the level
of detail discernable in an image
Intensity level resolution is usually given in terms
of the number of bits used to store each intensity
level

Number of Intensity
Number of Bits Examples
Levels
1 2 0, 1
2 4 00, 01, 10, 11
4 16 0000, 0101, 1111
8 256 00110011, 01010101
16 65,536 1010101010101010
Image Sizes in bits for Different Spatial
and Gray Level Resolutions

The calculation is for square images

N = number of row/column
L = gray level resolution
K= bit required to represent gray levels
Answer the questions:

Q1: Suppose a pixel has 1 bit, how many gray

levels can it represent?
Q2: Suppose a pixel has 2 bit, how many gray
levels can it represent?
Q3: Suppose a pixel has p bit, how many gray
levels can it represent?
Q4: if we want to represent 256 intensities of
grayscale, how many bits do we need?
Q5: What is the Gray Level Resolution and Spatial
Resolution of a 3 bit 5x5 image? What will be the
image size in bits? How many gray levels can be
there in the image? What is bit depth?
Effect of Spatial and Gray Level
Resolutions

 14X12 resolution means

samples from 14X12
 What happens, if we sample
from 8X8 locations?
 Effect of Spatial and Gray Level
Resolutions

 14X12 resolution means

samples from 14X12
 What happens, if we sample
from 8X8 locations?

 Larger cell size

 Lower spatial resolution
 Lower features spatial accuracy
 Lower file size, faster display and faster processing time
 Effect of Spatial Resolutions

 Sub sampled from 1024X1024 up

to 32X32
 unchanged gray level (intensity)
 Delete alternate row and column while
sub sampling
 Effect of Spatial Resolutions
Resized to 1024 x 1024
From 1024X1024 512X512 256X256

Using column
duplication process

From 128X128 64X64 32X32

Can we increase spatial resolution ?

From 128X128 64X64 32X32

Using nearest
neighbor

Using bilinear
interpolation

From 128X128 64X64 32X32

Down sampling is an irreversible process.

Effect of Gray level Resolutions

 unchanged
spatial
resolution

 Gray level
256 levels 128 levels changed from
256 to 32

64 levels 32 levels
 Effect of Gray level Resolutions

 Gray level
changed
from 16 to 2

16 levels 8 levels

In this image,
it is easy to see
false contour.

4 levels 2 levels
Effect of Gray level Resolutions

• unchanged spatial resolution

• Gray level changed from 256
to 32

Ridge-like structure in
the smooth area
Effect of Spatial and Gray Level
Resolutions
• Gray level changed from
16 to 2

• Ridge-like structure
is more prominent
• Reason: insufficient
number of gray
levels used
Resolution: How Much Is Enough?

 The big question with resolution is always how

much is enough?
– This all depends on what is in the image and what
you would like to do with it

– Key questions include

• Does the image look aesthetically pleasing?
• Can you see what you need to see within the image?
Resolution: How Much Is Enough?
(cont…)

The picture on the right is fine for counting the

number of cars, but not for reading the number
plate
 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.
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  256013 902296   67
  21  54 47  42 
32 15 87 39
  85 85 43  92
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

No. component component component

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
 
6 5 2 4 0.5 0.5 0.5
5 0.2 0.8 0.9
Index value … … … …
 Image Scaling (Zooming & Shrinking)
• Zooming/up scaling/resizing upward can be
achieved by the following techniques:
Pixel Replication
Interpolation
– Nearest Neighbor Interpolation
– Bilinear Interpolation
– Bicubic Interpolation
3x3 matrix is interpolated to 6x6
matrix
 Class Work:
• Interpolate a 2x2 matrix [1, 4, 3, 7] to 4x4
• Interpolate a 3x3 matrix [1,4,3,7,2,3,5,6,0] to 9x9
• Interpolate a 2x2 matrix [1, 2, 3, 4] to 8x8
• Interpolate a 3x3 matrix [1,4,3, 7, 5, 6] to 6x6
https://theailearner.com/2018/12/29/image-processing-bilinear-interpolation/
https://theailearner.com/2018/12/29/i
mage-processing-bicubic-interpolatio
n//