MODULE -1 DIGITAL IMAGE

FUNDAMENTALS

By,
Mr. Manjunathraddi Bentur
Asst. Professor Dept. of ECE
1
Why do we need image processing?
It is motivated by two major applications-

Improvement of pictorial information for human

perception.
For autonomous perception.

2
Image Processing Goals

Image Processing is a subclass of signal

processing concerned specifically with Pictures.
It Aims to improve the image quality for
•H u m a n Perception: subjective
•Computer interpretation: objective
Compress images for efficient
storage/transmission

3
 What is Digital Image Processing?
 Computer m a nipul ation of pictures, or images
t h a t have been converted into numeric form.
Typical operations include:
Contrast Enhancement
Remove Blur from image
Smooth out graininess, speckle or noise
Magnify, minify or rotate a n image(im age
warping)
Geometric Correction
Image Compression for efficient
storage/transmission. 4
What is Digital Image???

An image may be defined as a 2D function f(x,y),

where x and y are the spatial coordinates and f is the
intensity value at x and y.

If x,y and f are all discrete, then the image is called as

Digital Image.

5
The origins of Digital Image Processing:

6
7
Examples of Fields that use Digital Image
processing.
The principal energy source for images is the
electromagnetic energy spectrum.
EM waves = stream of massless (proton) particles,
each traveling in a wavelike pattern at the speed of
light. Spectral bands are grouped by energy/photon
Gamma rays, X-rays, UV, Visible, Infrared, Microwaves, radio waves

Other sources: acoustic, ultrasonic, electronic

8
Gamma-Ray Imaging:
Major uses of Gamma-Ray
Imaging include nuclear
medicine, astronomical
observations.
Nuclear medicine:
patient is injected with
radioactive isotope that emits
gamma rays as it decays.
Images are produced from
emissions collected by
detectors.

9
Contd.
Fig 1a shows an image of complete bone scan obtained
by using gamma ray imaging.
Images of this sort are used to locate sites of bone
pathology such as infections or tumors.
Fig 1b shows an another modality called Positron
Emission Tomography. This image shows tumor in the
brain and one in the lung. Easily visible as small white
masses.

10
X-Ray Imaging:
Oldest source of EM
radiation for imaging
Used for CAT scans
Used for angiograms where
X-ray contrast medium is
injected through catheter to
enhance contrast at site to be
studied.
Industrial inspection

11
 X-Ray Imaging (Contd…)
CT Image: Computed Tomography(good for hard
tissues such as bones.
In CT each slice of human body is imaged by means of
X-ray and then a number of such images are piled up to
form a volumetric representation of a body or specific part

12
Ultraviolet Images :

13
Imaging in the Visible band and infrared band:
Infrared band applications:
Industrial inspection
inspect for missing parts
missing pills
unacceptable bottle fill
unacceptable air pockets
anomalies in cereal color
incorrectly manufactured replacement
lens for eyes
Imaging in visible band
Face detection &recognition
Iris Recognition
Number Plate recognition
14
Imaging in the Visible and Infrared Bands:

15
16
17
Imaging in the Microwave Band:
The dominant application of imaging in the microwave
band is radar.
The imaging radar has the ability to collect data over
any region at any time regardless of weather or ambient
lighting conditions.

18
Imaging in the Radio Band:
The major application is in the field of medicine which includes
MRI.
MRI: Magnetic Resonance Imaging very similar to CT Imaging,
provides more detailed images of the soft tissues of the body. It
can be used to study both structure and function of a body.
Difference between CT and MRI is the imaging radiation. Ct
uses ionizing radiation such as X-ray whereas MRI uses a
powerful magnetic field.
Thermal Image: Thermo graphic camera used to capture images
in night vision.

19
Examples in which other Imaging modalities are used:

20
Fundamental Steps in Image Processing

21
Components of a general purpose Image
Processing

22
 Structure of Human Eye

Shape is nearly spherical

Average diameter = 20mm Three
membranes:
-Cornea and Sclera
-Choroid
-Retina

23
 Structure of Human Eye:
Cornea
Iris
Cornea:
Ciliary muscle
-Tough, transparent tissue that covers the
anterior surface of the eye
Sclera:
-Opaque membrane that encloses the
remainder of the optical globe
Choroid:
-Lies below the sclera
-Contains network of blood vessels
that serve as the major source of
nutrition to the eye. Sclera
-Choroid coat is heavily pigmented and Choroid
hence helps to reduce the amount of
extraneous light entering the eye and
the backscatter within the optical globe
24
 Lens and Retina
 Lens
-Both infrared and ultraviolet light are absorbed appreciably by
proteins within the lens structure and, in excessive amounts, can
cause damage to the eye

Retina
-Innermost membrane of the eye which lines the inside of the
wall’s entire posterior portion. When the eye is properly focused,
light from an object outside the eye Is imaged on the retina.

25
 Receptors
Two classes of light receptors on retina: cones and rods
Cones:
 6-7 million cones lie in central portion of the retina, called the
fovea.
Highly sensitive to color and bright light.
Resolve fine detail since each is connected to its own nerve end.
 Cone vision is called photopic or bright-light vision.
Rods
75-150 million rods distributed over the retina surface.
Reduced amount of detail discernable since several rods are
connected to a single nerve end.
Serves to give a general, overall picture of the field of view.
Sensitive to low levels of illumination.
Rod vision is called scotopic or dim-light vision.
26
Distribution of Cones and Rods
Blind spot: no receptors in region of emergence of optic nerve.
Distribution of receptors is radially symmetric about the fovea.
Cones are most dense in the center of the retina (e.g., fovea)
Rods increase in density from the center out to 20°and then
decrease

27
Density of rods and cones for a cross section of right eye
Image Formation in the Eye:

28
Brightness Adaptation and Discrimination:
The eye’s ability to discriminate
between intensities is important.
Experimental evidence suggests
that subjective brightness
(perceived) is a logarithmic
function of light incident on eye.
Notice approximately linear
response in log-scale below.

29
Contd …
FIGURE 2.5 Basic exp e
rime ntal setup used to char
act erize brightness
discrimination.

FIGURE 2.6
Typical Weber
ratio as a function
of intensity*

30
 Two Phenomena to illustrate that the perceived
brightness is not a simple function of intensity:
1. Mach Band Effect:

31
Two Phenomena to illustrate that the perceived brightness is
not a simple function of intensity(contd….)

2. Simultaneous Contrast:

32
Optical Illusions:

33
Light and the Electromagnetic Spectrum:

34
Image Sensing And Acquisition:

35
Image Acquisition Using a Single Sensor:

36
Image Acquisition Using Sensor Strips:

37
Image Acquisition Using Sensor Array:

38
Sampling and Quantization:

39
40
Representing Digital Image:

41
Contd…

Let A be a digital Image

The pixel intensity levels (gray scale levels) are in the interval of [0, L-1].

b is the no. of bits required to store digitized image of size M by N,

then b is given as b = M x N x k 42
Contd…
The range of values spanned by the gray scale is called the
dynamic range of an image.
It is also defines as the ratio of maximum measurable intensity
to minimum detectable intensity level.
Upper limit- saturation
Lower limit- Noise
Images whose gray levels span a significant portion of the gray
scale are considered as having a high dynamic range.
Contrast (highest- lowest) intensity level
When an appreciable number of pixels exhibit this property, the
image will have high contrast.
Conversely, an image with low dynamic range tends to have a
dull, washed out gray look.
43
Contd…
 Spatial and Intensity Resolution:
Spatial resolution is the smallest discernible(noticeable)
detail in an image.
It is also stated as dots per unit inch(dpi), line pairs per unit
distance.
Gray-level resolution similarly refers to the smallest
discernible (noticeable) change in gray level.
Measuring discernible changes in gray level is a highly
subjective process
The number of gray levels is usually an integer power of 2 Thus,
an L-level digital image of size M*N has a spatial resolution of
M*N pixels and a gray-level resolution of L levels

45
46
False Contouring:

47
The effect caused by the use of an insufficient number of gray levels in smooth
areas of a digital image, is called false contouring

48
49
 Image Interpolation:
Interpolation is the process of using known data to estimate values
at unknown locations.
Interpolation is the process of determining the values of a function
at positions lying between its samples.
It is the basic tool used extensively in tasks such as zooming,
Shrinking, rotating and geometric corrections.
Zooming and Shrinking are basically image re-sampling methods.
The process of interpolation is one of the fundamental operations
in image processing. The image quality highly depend on the used
interpolation technique.

50
Nearest neighbor interpolation:
Suppose that we have an image of size 500 × 500 pixels and we
want to enlarge it 1.5 times to 750 × 750 pixels.
Conceptually, one of the easiest ways to visualize zooming is
laying an imaginary 750 × 750 grid over the original image.
Obviously, the spacing in the grid would be less than one pixel
because we are fitting it over a smaller image.
In order to perform gray-level assignment for any point in the
overlay, we look for the closest pixel in the original image and
assign its gray level to the new pixel in the grid.
When we are done with all points in the overlay grid, we
simply expand it to the original specified size to obtain the
zoomed image.
This method of gray-level assignment is called nearest
neighbor interpolation.
51
Bilinear interpolation:
A slightly more sophisticated way of accomplishing gray-level assignments is
bilinear interpolation using the four nearest neighbors of a point.
Let (x', y') denote the coordinates of a point in the zoomed image and let
v(x’, y') denote the gray level assigned to it.
For bilinear interpolation, the assigned gray level is given by:
v(x, y) = ax + by + cxy + d
where; the four co-efficients are determined from the four equations in four
unknowns that can be written using the four nearest neighbors of point (x, y)
It is possible to use more neighbors for interpolation.
Using more neighbors implies fitting the points with a more complex surface,
which generally gives smoother results.

52
Bicubic Interpolation:

Involves sixteen neighbors to estimate intensity

V(x, y) = ΣΣaij 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

53
SOME BASIC RELATIONSHIPS
BETWEEN PIXELS

54
 SOME BASIC RELATIONSHIPS BETWEEN PIXELS:

An image is denoted by f(x, y).

Lowercase letters such as p and q are used to represent particular
pixels in an image.
The structure of a digital image allows stating some basic
relationships between pixels that can be useful in some practical
cases.
The pixels are organized in a regular structure and can have a
limited number of values.

55
Digital Image Coordinates:

56
Neighbors of a Pixel:

57
Neighbors of a Pixel:

58
Neighbors of a Pixel:

59
 Adjacency:
 Two pixels are adjacent if:
•They are neighbors in some sense (e.g. N4(p), N8(p), …)
•Their gray levels satisfy a specified criterion of similarity
V (e.g. equality, …)
For example, in a binary image two pixels are connected if they are
4-neighbors and have same value (0/1)
Let v: a set of intensity values used to define adjacency and connectivity.
In a binary Image v={1}, if we are referring to adjacency of pixels with
value 1.
In a Gray scale image, the idea is the same, but v typically contains more
elements, for example v= {180, 181, 182,....,200}.
If the possible intensity values 0 to 255, v set could be any subset of
these 256 values.

60
Adjacency(Contd..)

61
Adjacency(Contd..)
Two subsets S1 and S2 are adjacent, if some pixel in S1 is adjacent
to some pixel in S2.
Adjacent means, either 4-, 8- or m-adjacency.

62
Path:

63
Connected:
Two pixels p and q are said to be connected in S if there exists a path between
them consisting entirely of pixels in S.
For any pixel p in S, the set of pixels that are connected to it in S is called a
connected component of S.
If it only has one connected component, then set S is called a connected set.

Region:
 We call R, a region of the image if R is a connected set.

Boundary of a region R:
The boundary (also called border or contour) of a region R is the set of pixels in
the region that have one or more neighbors that are not in R.
Edge:
The boundary of a finite region forms a closed path and is thus a "global" concept.
Edges are formed from pixels with derivative values that exceed a preset threshold.
Idea of an edge is a "local" concept that is based on a measure of gray-level
discontinuity at a point.
Edges are intensity discontinuities and boundaries are closed paths 64
Distance Measures:
For pixels p,q,z with coordinates (x,y), (s,t), (u,v), D is a distance function or
metric if:
D(p,q) ≥ 0 (D(p,q)=0 iff p=q)
D(p,q) = D(q,p) and
D(p,z) ≤ D(p,q) + D(q,z)

The Euclidean Distance between p and q is defined as:

Pixels having a distance less than or equal to some value r from (x,y) are the
points contained in a disk of radius r centered at (x,y)

65
City-block distance:
The D4 distance (also called city-block distance) between p and q is defined
as:

the pixels having a D4 distance from (x, y) less than or equal to some value r
form a diamond centered at (x, y).

66
Distance Measures
Example: the pixels with D4 distance ≤ 2 from (x, y) (the center point) form
the following contours of constant distance:

The pixels with D4 = 1 are the 4-neighbors of (x, y).

67
Chessboard distance (D8, distance)
The D8, distance (also called chessboard distance) between p and q is defined as:

The pixels with D8 distance from (x, y) less than or equal to some value r
form a square centered at (x, y).

68
Distance Measures:

 For example:
The pixels with D8 distance ≤ 2 from (x, y) (the center point) form
the following contours of constant distance:

69
Distance Measures :

 m-distance:
Defined as the shortest m-path between the points.
The distance between two pixels will depend on the values of the pixels
along the path, as well as the values of their neighbors.

Example :
consider the following arrangement of pixels and assume that p, p2, and p4
have value 1 and that pl and p3 can have a value of 0 or 1:
Suppose that we consider adjacency of pixels valued 1 (i.e,, V ={1} ).

70
Case1: If pl and p3 are 0
The length of the shortest m-path (the Dm distance) between p and p4 is 2. (p, p2, p4).

Case2: If pl is 1 and p3 is 0
Then p2 and p will no longer be m-adjacent and the length of the shortest m-path
becomes 3 (p, pl, p2,p4).

Case 3: If p3 is l and pl is 0
The length of the shortest m-path also is 3.

71
Case 4:
If both pl and p3 are 1
The length of the shortest m-path between p and q is 4. (p,p1, p2, p3, p4)

72
Linear And Nonlinear Operations:

 Let H be an operator whose input and output are images.

 H is said to be a linear operator if; for any two images f and g and any
two scalars a and b:
 H(af + bg) = aH(f) + bH(g).
 The result of applying a linear operator to the sum of two image that have been
multiplied by the constants shown) is identical to applying the operator to the
images individually, multiplying the results by the appropriate constants, and
then adding those results.

73
Thank You

74