Image Segmentation and

Extraction of object from an

image

CH.Venkanna
EC104018
M.Tech (E.I)

1
Introduction to image segmentation
The purpose of image segmentation is to partition an
image into meaningful regions with respect to a
particular application
The segmentation is based on measurements taken
from the image and might be greylevel, colour, texture,
depth or motion

2
Introduction to image segmentation
Usually image segmentation is an initial and vital
step in a series of processes aimed at overall image
understanding
Applications of image segmentation include
Identifying objects in a scene for object-based
measurements such as size and shape
Identifying objects in a moving scene for object-based
video compression (MPEG4)

4
Introduction to image segmentation
Example 1
Segmentation based on greyscale
Very simple ‘model’ of greyscale leads to inaccuracies in
object labelling

5
Different Methods of image
segmentation

There are different types of segmentation methods are

proposed. These are
Thresholding
Clustering
Relaxation labelling
EM algorithm(Proposed Algorithm)
The Expectation/Maximization (EM)
algorithm

In general we may imagine that an image comprises L

segments (labels)
Within segment l the pixels (feature vectors) have a
probability distribution represented by pl ( x |  l )
  represents the parameters of the data in segment l
l
 Mean and variance of the greylevels

 Mean vector and covariance matrix of the colours

 Texture parameters

7
The Expectation/Maximization (EM)
algorithm
The incomplete data are just the measured pixel greylevels or
feature vectors
We can define a probability distribution of the incomplete data
as pi ( x;1 , 2 ..... L )
The complete data are the measured greylevels or feature
vectors plus a mapping function f (.) which indicates the
labelling of each pixel
Given the complete data (pixels plus labels) we can easily work
out estimates of the parameters
 l : l  1..L
But from the incomplete data no closed form solution exists
The Expectation/Maximization (EM)
algorithm
The algorithm is as follows:

Initialize an estimate of  l : l  1..L

Repeat
Step 1: (E step)
Obtain an estimate of the labels based on the
current parameter estimates

Step 2: (M step)
Update the parameter estimates based on the current
labelling
Until Convergence
9
The Expectation/Maximization (EM)
algorithm
A recent approach to applying EM to image
segmentation is to assume the image pixels or feature
vectors follow a mixture model
Generally we assume that each component of the
mixture model is a Gaussian
A Gaussian mixture model (GMM)

p( x | )  l 1 l pl ( x |  l )
L

1 1
pl ( x |  l )  exp( ( x   l )  l ( x  l ))
T 1

(2 ) det( l )
d /2 1/ 2
2


L
l 1
l  1
10
The Expectation/Maximization (EM)
algorithm
Our parameter space for our distribution now includes
the mean vectors and covariance matrices for each
component in the mixture plus the mixing weights
   1 , 1 , 1 ,.......... L ,  L ,  L 
We choose a Gaussian for each component because
the ML estimate of each parameter in the E-step
becomes linear

11
The Expectation/Maximization (EM)
algorithm
Define a posterior probability P (l | x j ,  l ) as the
probability that pixel j belongs to region l given the
value of the feature vector x j
Using Bayes rule we can write the following equation
 l pl ( x j |  l )
P(l | x j , l ) 

L
k 1
 k pk ( x |  k )

This actually is the E-step of our EM algorithm as

allows us to assign probabilities to each label at each
pixel
12
The Expectation/Maximization (EM)
algorithm
The M step simply updates the parameter estimates
using MLL estimation
1 n
 l
( m 1)
  P(l | x j , l( m ) )
n j 1
n

 j
x P
j 1
(l | x j ,  l
(m)
)
l( m 1)  n

 P(l | x ,
j 1
j l
(m)
)

 
n

 P
j 1
(l | x j ,  l
(m)
) ( x j   l
( m)
)( x j   l
( m) T
)
 l( m 1)  n

 P (l
j 1
| x j ,  l
(m)
)
13
Fig : Original image
Fig : (Right column) The proposed image segmentation, (left column)
direct segmentation
Extraction of object from an image
After partition an image into meaningful regions
and sometimes it is required extract object from an
image
Extraction founds many applications in Biometric
image processing and face recognition.
Face Feature Extraction(voronoi Diagram method)
 Detecting eyes
 Detecting nose and mouth
Fig : Extraction of eyes ,nose and mouth from an image
Extraction of alphabet from
number plate

Fig : Original image and segment image,extracting

alphabet .
Conclusion
The EM algorithm is general image segmentation
technique in computer vision with many
applications.
Extraction found many applications in security system
and biometric image processing.

19