Digital Image Processing

Lecture # 2
Connected Component Analysis

1
Spatial & Gray Level Resolution
Spatial Resolution

3
Spatial Resolution

4
 Intensity Level Resolution
 Intensity level resolution refers to the number of intensity levels used to represent the
image

 The more intensity levels used, the finer the level of detail in
an image
 Intensity level resolution is usually given in terms of the
number of bits used to store each intensity level

5
 Intensity Level Resolution

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

6
Intensity Level Resolution

7
 Intensity Level Resolution
256 grey levels (8 bits per pixel) 128 grey levels (7 bpp) 64 grey levels (6 bpp) 32 grey levels (5 bpp)

16 grey levels (4 bpp) 8 grey levels (3 bpp) 4 grey levels (2 bpp) 2 grey levels (1 bpp)
8
 Resolution: How much is enough?

 How many samples and gray levels are required

for a good approximation?
 Quality of an image depends on number of pixels and gray-
level number
 The more these parameters are increased, the closer the
digitized array approximates the original image
 But: Storage & processing requirements increase rapidly as a
function of N, M, and k

9
 Resolution: How much is enough?

 Depends on what is in the image and what you

would like to do with it

10
 Image Resizing
 Pixel replication
[1 2 3 4 5]
[1 1 2 2 3 3 4 4 5 5] (One step)
[1 1 1 2 2 2 3 3 3 4 4 4 5 5 5] (Two step)

11
 Enlarging an Image

Example:
zoom this
image 4x to
get this
image.

.
.
Start with a blank image 4 times the Fill
. in every 4th pixel in every 4th
linear dimensions of the original. row with the original pixel values.

13
 Enlarging an Image

…
…
…
…
…
…
Detail showing every 4th pixel in every 4th row with the original pixel values.

14
Enlarging an Image

Replicate the values

15
 Image Interpolation

 Nearest neighbour interpolation

 Simple but produces undesired artefacts
 Bilinear Interpolation
 Contribution from 4 neighbors
 Bicubic Interpolation
 Contribution from 16 neighbors

16
Interpolation: Comparison

We’ll enlarge this image

by a factor of 4 …

… via bilinear interpolation

and compare it to a nearest
neighbor enlargement.
 Interpolation: Comparison

Original
Image

To better see what happens, we’ll look at the parrot’s eye.

Interpolation: Comparison

Pixel replication Bilinear interpolation

Interpolation: Comparison

Pixel replication Bilinear interpolation

 Relationships between pixels

 Neighbors of pixel are the pixels that are

adjacent pixels of an identified pixel

y-1 y y+1
x-1

x
x+1

22
4- Neighbors of a Pixel –N4(p)

coordinates of each of y-1 y y+1

x-1
the blue pixels
What are the

x
x+1

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

23
 Diagonal Neighbors of a Pixel –ND(p)

y-1 y y+1
x-1

x
x+1

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

24
8- Neighbors of a Pixel –N8(p)

y-1 y y+1
x-1

x
x-1

N8 ( p)  N4 ( p)  N D ( p)
(x-1,y), (x+1,y), (x, y-1), (x, y+1)
(x-1,y-1), (x+1,y-1), (x-1, y+1), (x+1, y+1)
25
Determine different regions in the
image

26
 Connectivity
 Establishing boundaries of objects and components in an image
 Group the same region by assumption that the pixels being the same
color or equal intensity
 Two pixels p & q are connected if
 They are adjacent in some sense
 If their gray levels satisfy a specified criterion of
similarity

27
 Connectivity
V: Set of gray levels used to define the criterion of similarity

4-connectivity

If gray level ( p, q) V , and q  N 4 ( p)

Set of gray levels V = {1}

28
 Connectivity
V: Set of gray levels used to define the criterion of similarity

8-connectivity

If gray level ( p, q) V , and q  N8 ( p)

Set of gray levels V = {1}

29
 Connectivity
V: Set of gray levels used to define the criterion of similarity

m-connectivity (Mixed Connectivity)

If gray level

( p, q) V , and q satisfies one of the following:

a. q  N4 ( p) or
b. q  N D ( p) And N 4 ( p)  N 4 (q) has no pixels
whose values are from V

30
 Example: m – Connectivity

 Set of gray levels V = {1}

Note: Mixed connectivity can eliminate the multiple path connections that often
occurs in 8-connectivity

31
 Paths
 Path: Let coordinates of pixel p: (x, y), and of pixel q: (s, t)

 A path from p to q is a sequence of distinct pixels with

coordinates: (x0, y0), (x1, y1), ......, (xn,yn)
where (x0, y0) = (x, y) & (xn,yn) = (s, t), and (xi,yi) is adjacent
to (xi-1,yi-1) 1≤i ≤n

32
Test Yourself
 CC labeling – 4 Connectivity
 Process the image from left to
right, top to bottom:
1.) If the next pixel to process is 1
i.) If only one of its neighbors
(top or left) is 1, copy its label.

ii.) If both are 1 and have the

same label, copy it.

iii.) If they have different labels

Pass 1
 Copy the label from the left.
 Update the equivalence table.

iv.) Otherwise, assign a new label.

 Re-label with the smallest of equivalent

labels Pass 2

35
CC labeling – 4 Connectivity

36
CC labeling – 4 Connectivity

37
 CC labeling – 8 Connectivity

Same algorithm but examine also the upper diagonal neighbors of p

38
 CC labeling – 8 Connectivity

Background pixel
Background pixel
Unlabeled Pixel
Unlabeled Pixel
Label 1

39
CC labeling – 8 Connectivity

Background pixel Background pixel

Unlabeled Pixel Unlabeled Pixel
Label 1 Label 1
Label 2 Label 2

Label 3

40
 CC labeling – 8 Connectivity

Background pixel Background pixel

Unlabeled Pixel Unlabeled Pixel

Label 1 Label 1
Label 2 Label 2
Label 3 Label 3

41
 CC labeling – 8 Connectivity

Background pixel Background pixel

Unlabeled pixel Unlabeled pixel

Label 1 Label 1

Label 2 Label 2

Label 3 Label 3

42
CC labeling – 8 Connectivity

Background pixel Background pixel

Unlabeled pixel Unlabeled pixel
Label 1 Label 1
Label 2 Label 2
Label 3 Label 3
Label 4 Label 4

43
 Distance Metrics

 Let pixels p, q and z have coordinates (x,y),

(s,t) and (u,v) respectively.

 D is a distance function or metric if

 D(p,q) ≥ 0 and
 D(p,q) = 0 iff p = q and
 D(p,q) = D(q,p) and
 D(p,z) ≤ D(p,q) + D(q,z)

44
City block distance (D4 distance)

D4 ( p, q)  x  s  y  t

 Diamond with center at

(x,y)
 D4 = 1 are the 4 neighbors
of pixel p(x,y)

45
Chessboard distance (D8 distance)

D8 ( p, q)  max( x  s , y  t )

 Square centered at p(x,y)

 D8 = 1 are the 8 neighbors
of pixel p(x,y)

47
 Euclidean Distance

De ( p, q )  ( x  s ) 2  ( y  t ) 2
q(s,t)

p(x,y)

A circle with radius r centered at (x,y)

49
 Arithmetic Operations

 Carried out between corresponding pixel

pairs s( x, y)  f ( x, y)  g ( x, y)
d ( x, y )  f ( x, y )  g ( x, y )
p ( x, y )  f ( x , y )  g ( x , y )
d ( x, y )  f ( x, y )  g ( x, y )

50
 Arithmetic Operations

 Conversion to range 0 – 255

 Difference of two 8-bit images: -255 to 255
 Sum of two 8-bit images: 0 to 510
 Solution?
Set all values < 0 to 0
Set all values > 255 to 255
Full range of arithmetic operation not captured
51
 Arithmetic Operations

 First perform the operation

Creates an image
whose minimum value
is 0
f m  f  min( f )

 Then perform Creates a scaled

image fs with values in
the range [0 K]

f s  K  f m max( f m )

52
Assignment-1
 Readings from Book (4th Edn.)
• Chapter – 2
 Acknowledgements
 Statistical Pattern Recognition: A Review – A.K Jain et al., PAMI (22) 2000
 Pattern Recognition and Analysis Course – A.K. Jain, MSU
Material in these slides has been taken from, the following resources

 Pattern Classification” by Duda et al., John Wiley & Sons.

 Digital Image Processing”, Rafael C. Gonzalez & Richard E. Woods, Addison-Wesley, 2002
 Machine Vision: Automated Visual Inspection and Robot Vision”, David Vernon, Prentice Hall,
1991
 www.eu.aibo.com/
 Advances in Human Computer Interaction, Shane Pinder, InTech, Austria, October 2008

56