Lecture 4 Basic Relationships between Pixels
by
Paresh Kamble
DIGITAL IMAGE
PROCESSING
Neighbors of a Pixel
Y
X
Y
Neighbors of a Pixel
f(0,0)
f(1,0)
f(x,y) =
f(2,0)
f(3,0)
I
I
f(0,1)
f(1,1)
f(0,2)
f(1,2)
f(0,3)
f(1,3)
f(0,4) - - - - f(1,4) - - - -
f(2,1)
f(2,2)
f(2,3)
f(2,4) - - - - f(3,1)
f(3,2)
f(3,3)
f(3,4) - - - I
I
I
I
----I
I
I
I
-----
Neighbors of a Pixel
f(0,0)
f(1,0)
f(0,1)
f(1,1)
f(0,2)
f(1,2)
f(0,3)
f(1,3)
f(0,4) - - - - f(1,4) - - - -
f(x,y) =
f(2,0)
f(2,1)
f(2,2)
f(2,3)
f(2,4) - - - - f(3,0)
f(3,1)
f(3,2)
f(3,3)
f(3,4) - - - I
I
I
I
I
- - has
- - 4 horizontal and vertical
A Pixel p at coordinates ( x,- y)
I
I
I
I
I
neighbors.
---- Their coordinates are given by:
(x+1, y)
&
f(2,1)
(x-1, y)
(x, y+1)
(x, y-1)
f(0,1)
f(1,2)
f(1,0)
This set of pixels is called the 4-neighbors of p denoted by
N4(p).
Neighbors of a Pixel
f(0,0)
f(1,0)
f(0,1)
f(1,1)
f(0,2)
f(1,2)
f(0,3)
f(1,3)
f(0,4) - - - - f(1,4) - - - -
f(x,y) =
f(2,0)
f(2,1)
f(2,2)
f(2,3)
f(2,4) - - - - f(3,0)
f(3,1)
f(3,2)
f(3,3)
f(3,4) - - - I
I
I
I
I
- - has
- - 4 diagonal neighbors.
A Pixel p at coordinates ( x,- y)
I
I
I
I
I
--- Their coordinates are given- by:
(x+1, y+1)
(x+1, y-1)
&
(x-1, y-1)
f(2,2)
f(2,0)
(x-1, y+1)
f(0,2)
f(0,0)
This set of pixels is called the diagonal-neighbors of p
denoted by ND(p).
Adjacency, Connectivity
Adjacency: Two pixels are adjacent if they are
neighbors and their intensity level V satisfy some
specific criteria of similarity.
e.g. V = {1}
V = { 0, 2}
Binary image = { 0, 1}
Gray scale image = { 0, 1, 2, ------, 255}
In binary images, 2 pixels are adjacent if they are
neighbors & have some intensity values either 0 or 1.
In gray scale, image contains more gray level values in
range 0 to 255.
Adjacency, Connectivity
4-adjacency: Two pixels p and q with the values
from set V are 4-adjacent if q is in the set of N4(p).
e.g. V = { 0, 1}
1
1
1
1
1
0
p in RED color
q can be any value in GREEN color.
0
0
1
Adjacency, Connectivity
8-adjacency: Two pixels p and q with the values
from set V are 8-adjacent if q is in the set of N8(p).
e.g. V = { 1, 2}
0
0
0
1
2
0
p in RED color
q can be any value in GREEN color
1
0
1
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
OR
(ii) q is in ND(p) & the set N4(p)
N4(q) have no
pixels whose values are from V.
e.g. V = { 1 }
0a
0d
0g
1b
1e
0h
1c
0f
1i
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
e.g. V = { 1 }
(i) b & c
0a
0d
0g
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
e.g. V = { 1 }
(i) b & c
0a
0d
0g
Soln: b & c are m-adjacent.
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
e.g. V = { 1 }
(ii) b & e
0a
0d
0g
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
e.g. V = { 1 }
(ii) b & e
0a
0d
0g
Soln: b & e are m-adjacent.
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
OR
e.g. V = { 1 }
(iii) e & i
0a
0d
0g
1b
1e
0h
1c
0f
1i
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in ND(p) & the set N4(p)
N4(q) have no
pixels whose values are from V.
e.g. V = { 1 }
(iii) e & i
0a
0d
0g
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in ND(p) & the set N4(p)
N4(q) have no
pixels whose values are from V.
e.g. V = { 1 }
(iii) e & i
0a
0d
0g
Soln: e & i are m-adjacent.
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
OR
(ii) q is in ND(p) & the set N4(p)
N4(q) have no
pixels whose values are from V.
e.g. V = { 1 }
(iv) e & c
0a
0d
0g
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
m-adjacency: Two pixels p and q with the values
from set V are m-adjacent if
(i) q is in N4(p)
OR
(ii) q is in ND(p) & the set N4(p)
N4(q) have no
pixels whose values are from V.
e.g. V = { 1 }
(iv) e & c
0a
0d
0g
Soln: e & c are NOT m-adjacent.
1b
1e
0h
1c
0f
1I
Adjacency, Connectivity
Connectivity:
2 pixels are said to be connected if their
exists a path between them.
Let S represent subset of pixels in an image.
Two pixels p & q are said to be connected in S if their 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.
Paths
Paths: A path from pixel p with coordinate ( x,
y) with pixel q with coordinate ( s, t) is a
sequence of distinct sequence with
coordinates (x0, y0), (x1, y1), .., (xn, yn) where
(x, y) = (x0, y0)
& (s, t) = (xn, yn)
Closed path: (x0, y0) = (xn, yn)
Paths
Example # 1: Consider the image segment shown in figure.
Compute length of the shortest-4, shortest-8 & shortestm paths between pixels p & q where,
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-4 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-4 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-4 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-4 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-4 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-4 path:
V = {1, 2}.
4
3
2
p2
So, Path does not exist.
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-8 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-8 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-8 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-8 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-8 path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-8 path:
V = {1, 2}.
4
3
2
p2
So, shortest-8 path = 4
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-m path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-m path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-m path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-m path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-m path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-m path:
V = {1, 2}.
4
3
2
p2
2
3
3
1
3
1
2
2
2q
3
2
3
Paths
Example # 1:
Shortest-m path:
V = {1, 2}.
4
3
2
p2
So, shortest-m path = 5
2
3
3
1
3
1
2
2
2q
3
2
3
Regions & Boundaries
Region:
Let R be a subset of pixels in an image. Two
regions Ri and Rj are said to be adjacent if their union form a
connected set.
Regions that are not adjacent are said to be disjoint.
We consider 4- and 8- adjacency when referring to regions.
Below regions are adjacent only if 8-adjacency is used.
1
1
0
0
1
1
1
0
1
0
1
1
1
1 Ri
0
1
1 Rj
1
Regions & Boundaries
Boundaries (border or contour): The boundary
of a region R is the set of points that are adjacent to
points in the compliment of R.
0
0
0
0
0
0
0
1
1
1
1
0
0
1
1
1
1
0
0
0
1
1
0
0
0
0
0
0
RED colored 1 is NOT a member of border if 4connectivity is used between region and background.
It is if 8-connectivity is used.
Distance Measures
Distance Measures: Distance between pixels p, q &
z with co-ordinates ( x, y), ( s, t) & ( v, w) resp. is given
by:
a) D( p, q) 0 [ D( p, q) = 0 if p = q]
..called reflexivity
b) D( p, q) = D( q, p)
.called symmetry
c) D( p, z) D( p, q) + D( q, z)
.called transmitivity
.
..
Euclidean distance between p & q is defined asDe( p, q) = [( x- s)2 + (y - t)2]1/2
Distance Measures
City Block Distance: The D4 distance between p &
q is defined as
D4( p, q) = |x - s| + |y - t|
In this case, pixels having D4 distance from ( x, y) less
than or equal to some value r form a diamond
centered at ( x, y).
2
1
2
1
0
2
1
Distance Measures
Chess-Board Distance: The D8 distance between p
& q is defined as
D8( p, q) = max( |x - s| , |y - t| )
In this case, pixels having D8 distance from ( x, y) less
than or equal to some value r form a square centered
at ( x, y).
2
2
End of topic - 2
