Face Recognition

Face Recognition Technology
ACKNOWLEDGEMENT
It is with greatest pleasure and pride that I present this report before you. At
this moment of triumph, it would be unfair to neglect all those who helped me
in the successful completion of this seminar.
First of all, I would like to place myself at the feet of God Almighty for his
everlasting love and for the blessings & courage that he gave me, which made
it possible to me to see through the turbulence and to set me in the right path.
I would also like to thank our Head of the Department, Mrs. Poonam Pathak
for all the help and guidance that she provided to me.
I am grateful to my seminar guide Ms. Sonam Singh for her guidance and
whole hearted support and very valued constructive criticism that has driven to
complete the seminar successfully.
I would take this opportunity to thank my friends who were always a source of
encouragement.
ABHISHEK SRIVASTAVA
BBDNIIT
Page 1
ABSTRACT
Wouldnt you love to replace password based access control to avoid having
to reset forgotten password and worry about the integrity of your system?
Wouldnt you like to rest secure in comfort that your healthcare system does
not merely on your social security number as proof of your identity for granting
access to your medical records?
Because each of these questions is becoming more and more important,
access to a reliable personal identification is becoming increasingly essential
.Conventional method of identification based on possession of ID cards or
exclusive knowledge like a social security number or a password are not all
together reliable. ID cards can be lost forged or misplaced; passwords can be
forgotten or compromised. But a face is undeniably connected to its owner. It
cannot be borrowed stolen or easily forged.
Face recognition technology may solve this problem since a face is undeniably
connected to its owner expect in the case of identical twins. Its
nontransferable. The system can then compare scans to records stored in a
central or local database or even on a smart card.
BBDNIIT
Page 2
TABLE OF CONTENTS
CHAPTER NO.
TITLE
PAGE NO.
1.
Abstract
2.
Acknowledgement
3.
Table of contents
4.
List of figures
5.
Introduction
What are biometrics
1.2
Why we choose face recognition over other biometrics 7
6.
Face recognition
7.
Capturing of image by standard video cameras
11
8.
Components of face recognition system
13
9.
Performances
15
Implementation of face recognition technology
16
6.1
Data acquisition
16
6.2
Input processing
16
6.3
Face image classification and decision making
18
10.
11.
Face recognition algorithm

7.1
BBDNIIT
1.1
Eigenface
19
19
Page 3

CHAPTER NO.
7.2
TITLE
Neutral network
PAGE NO.
20
12.
How face recognition system works
22
13.
Methods of face recognition
24
9.1
Two dimensional face recognition-PCA
24
9.2
Three dimensional face recognition
25
14.
The software
27
10.1
Detection
27
10.2
Alignment
27
10.3
Normalization
27
10.4
Representation
27
10.5
Matching
28
15.
Advantages and disadvantages
29
11.1
Advantage
29
11.2
Disadvantage
29
16.
Application
30
12.1
Government Use
31
12.2
Commercial Use
31
17.
Conclusion
33
18.
References
34
BBDNIIT
Page 4

LIST OF FIGURES
Sl. No.
Figure Title
Page No.
1.
Comparison of various biometrics
2.
Face recognition process
10
3.
12
4.
Process of capturing image
12
5.
Components of face recognition systems
14
6.
Block diagram showing input Processing
17
7.
18
8.
Performance of various face recognizer through
21
data table
9.
How face recognition systems work an example
22
10.
Another figure explaining working of face recognition
23
Systems
11.
Working of two dimensional face recognition system
24
12.
Working of three dimensional face recognition system
26
13.
Person checking whether there is a match of a face
28
or not
14.
Surveillance camera for comparing images used as
30
application
BBDNIIT
Page 5
1. INTRODUCTION
The information age is quickly revolutionizing the way transactions are
completed. Everyday actions are increasingly being handled electronically,
instead of with pencil and paper or face to face. This growth in electronic
transactions has resulted in a greater demand for fast and accurate user
identification and authentication. Access codes for buildings, banks accounts
and computer systems often use PIN's for identification and security
clearances.
Using the proper PIN gains access, but the user of the PIN is not verified.
When credit and ATM cards are lost or stolen, an unauthorized user can often
come up with the correct personal codes. Despite warning, many people
continue to choose easily guessed PINs and passwords: birthdays, phone
numbers and social security numbers. Recent cases of identity theft have high
lighten the need for methods to prove that someone is truly who he/she claims
to be.
Face recognition technology may solve this problem since a face is undeniably
connected to its owner expect in the case of identical twins. Its
nontransferable. The system can then compare scans to records stored in a
central or local database or even on a smart card.
1.1 What is biometrics?

A biometric is a unique, measurable characteristic of a human being that can
be used to automatically recognize an individual or verify an individuals
identity. Biometrics can measure both physiological and behavioral
characteristics. Physiological biometrics (based on measurements and data
derived from direct measurement of a part of the human body) include:
a. Finger-scan
b. Facial Recognition
c. Iris-scan
d. Retina-scan
e. Hand-scan
BBDNIIT
Page 6

Behavioral biometrics (based on measurements and data derived from an
action) include:
a. Voice-scan
b. Signature-scan
c. Keystroke-scan
A biometric system refers to the integrated hardware and software used to
conduct biometric identification or verification.
1.2 Why we choose face recognition over other

biometric?
There are number reasons to choose face recognition. This includes the
following
a. It requires no physical interaction on behalf of the user.
b. It is accurate and allows for high enrolment and verification rates.
c. It does not require an expert to interpret the comparison result.
d. It can use your existing hardware infrastructure; existing cameras and
image capture devices will work with no problems
e. It is the only biometric that allow you to perform passive identification in a
one to many environments (e.g.: identifying a terrorist in a busy Airport
terminal
BBDNIIT
Page 7
FIGURE 1 Comparison of various biometrics
The above two figure shows the Comparison of various biometric features on
the following two basis :
(a) based on zephyr analysis
(b) based on MRTD compatibility
The zephyr analysis shows that how much various biometrics are accurate,
cost effective effortless. The other figure shows the comparison of various
biometrics in weighted percentage, here in this we see that the face
recognition method is best suited among all the other biometrics and hence in
many industries it is used for securing passwords and also used in various
security purpose.
BBDNIIT
Page 8
2. FACE RECOGNITION
THE FACE:
The face is an important part of who you are and how people identify you.
Except in the case of identical twins, the face is arguably a person's most
unique physical characteristics. While humans have the innate ability to
recognize and distinguish different faces for millions of years, computers are
just now catching up.
For face recognition there are two types of comparisons the first is verification.
This is where the system compares the given individual with who that
individual says they are and gives a yes or no decision. The second is
identification. This is where the system compares the given individual to all the
other individuals in the database and gives a ranked list of matches.
All identification or authentication technologies operate using the following four
stages:
a. Capture: A physical or behavioral sample is captured by the system during
enrollment and also in identification or verification process
b. Extraction: unique data is extracted from the sample and a template is
created.
c. Comparison: the template is then compared with a new sample.
d. Match/non match: the system decides if the features extracted from the new
samples are a match or a non match.
Face recognition technology analyze the unique shape, pattern and positioning
of the facial features. Face recognition is very complex technology and is
largely software based. This Biometric Methodology establishes the analysis
framework with tailored algorithms for each type of biometric device. Face
recognition starts with a picture, attempting to find a person in the image. This
can be accomplished using several methods including movement, skin tones,
or blurred human shapes.
The face recognition system locates the head and finally the eyes of the
individual. A matrix is then developed based on the characteristics of the
Individuals face. The method of defining the matrix varies according to the
algorithm (the mathematical process used by the computer to perform the
BBDNIIT
Page 9

comparison). This matrix is then compared to matrices that are in a database
and a similarity score is generated for each comparison.
Artificial intelligence is used to simulate human interpretation of faces. In order
to increase the accuracy and adaptability, some kind of machine learning has
to be implemented.
There are essentially two methods of capture. One is video imaging and the
other is thermal imaging. Video imaging is more common as standard video
cameras can be used. The precise position and the angle of the head and the
surrounding lighting conditions may affect the system performance. The
complete facial image is usually captured and a number of points on the face
can then be mapped, position of the eyes, mouth and the nostrils as an
example. More advanced technologies make 3-D map of the face which
multiplies the possible measurements that can be made. Thermal imaging has
better accuracy as it uses facial temperature variations caused by vein
structure as the distinguishing traits. As the heat pattern is emitted from the
face itself without source of external radiation these systems can capture
images despite the lighting condition, even in the dark. The drawback is high
cost. They are more expensive than standard video cameras.
FIGURE 2 Face recognition process
BBDNIIT
Page 10
3. CAPTURING OF IMAGE BY STANDARD VIDEO

CAMERAS
The image is optical in characteristics and may be thought of as a collection of
a large number of bright and dark areas representing the picture details. At an
instant there will be large number of picture details existing simultaneously
each representing the level of brightness of the scene to be reproduced. In
other words the picture information is a function of two variables: time and
space. Therefore it would require infinite number of channels to transmit
optical information corresponding to picture elements simultaneously. There is
practical difficulty in transmitting all information simultaneously so we use a
method called scanning.
Here the conversion of optical information to electrical form and its
transmission is carried out element by element one at a time in a sequential
manner to cover the entire image. A TV camera converts optical information
into electrical information, the amplitude of which varies in accordance with
variation of brightness.
An optical image of the scene to be transmitted is focused by lens assembly
on the rectangular glass plate of the camera tube. The inner side of this has a
transparent coating on which is laid a very thin layer of photoconductive
material. The photo layer has very high resistance when no light is falling on it
but decreases depending on the intensity of light falling on it. An electron
beam is formed by an electron gun in the TV camera tube. This beam is used
to pick up the picture information now available on the target plate of varying
resistance at each point.
The electron beam is deflected by a pair of deflecting coils mounted on the
glass envelope and kept mutually perpendicular to each other to achieve
scanning of the entire target area. The deflecting coils are fed separately from
two sweep oscillators, each operating at different frequencies. The magnetic
deflection caused by current in one coil gives horizontal motion to the beam
from left to right at a uniform rate and brings it back to the left side to
commence the trace of the next line. The other coil is used to deflect the beam
from top to bottom.
BBDNIIT
Page 11
FIGURE 3
FIGURE 4
BBDNIIT
Process of capturing image
Page 12

As the beam moves from element to element it encounters different resistance
across the target plate depending on the resistance of the photoconductive
coating. The result is flow of current which varies in magnitude as elements
are scanned. The current passes through the load resistance Rl connected to
conductive coating on one side of the DC supply source on the other.
Depending on the magnitude of current a varying voltage appears across the
resistance Rl and this corresponds to the optical information of the picture.
4.
COMPONENTS OF FACE RECOGNITION

SYSTEMS
The problem of face recognition has been one of the most prominent areas of
machine vision for about a decade. Current system have been advanced to be
fairly accurate in recognition under constrained scenarios, but extrinsic
imaging parameters such as pose, illumination and facial expression still
cause much difficulty in face recognition.
Recently, components approaches have shown promising result in various

object detection and recognition such as face detection, person detection and
face recognition. The component based face detector described localizes a set
of facial components using a two level hierarchy of classifiers. On top of this
detector, we built a component based face recognition system in which the
gray values of the extracted components were combined and then classified
by asset of Support Vectors Machines, one for each person in the data base.
The components of face recognition system can be divided in these three

following parts:
a. An automated mechanism that scans and captures a digital or an analog
image of a living personal characteristics. (enrollment module)
b. Another
entity which handles compression, processing,

compression of the captured data with stored data (database)
storage and
c. The third interfaces with the application system (identification module)
BBDNIIT
Page 13
FIGURE 5 Components of face recognition systems

User interface captures the analog or digital image of the person's face. In the
enrollment module the obtained sample is preprocessed and analyzed. This
analyzed data is stored in the database for the purpose of future comparison.
The database compresses the obtained sample and stores it. It should have
retrieval property also that is it compares the entire stored sample with the
newly obtained sample and retrieves the matched sample for the purpose of
verification by the user and determine whether the match declared is right or
wrong.
The verification module also consists of a preprocessing system. Verification
means the system checks as to who the person says he or she is and gives a
yes or no decision. In this module the newly obtained sample is preprocessed
and compared with the sample stored in the database. The decision is taken
depending on the match obtained from the database. Correspondingly the
sample is accepted or rejected.
Instead of verification module we can make use of identification module. In this
the sample is compared with all the other samples stored in the database. For
BBDNIIT
Page 14

each comparison made a match score is given. The decision to accept or
reject the sample depends on this match score falling above or below a
predetermined threshold.
5.
PERFORMANCE
False acceptance rate (FAR)

The probability that a system will incorrectly identify an individual or will fail to
reject an imposter. It is also called as type 2 error rate
FAR= NFA/NIIA
Where FAR= false acceptance rate
NFA= number of false acceptance
NIIA= number of imposter identification attempts
False rejection rates (FRR)
The probability that a system will fail to identify an enrollee. It is also called
type 1 error rate.
FRR= NFR/NEIA
Where FRR= false rejection rates
NFR= number of false rejection rates
NEIA= number of enrollee identification attempt
Response time:
The time period required by a biometric system to return a decision on
identification of a sample.
Threshold/ decision Threshold:
The acceptance or rejection of a data is dependent on the match score falling
above or below the threshold. The threshold is adjustable so that the system
BBDNIIT
Page 15

can be made more or less strict; depending on the requirements of any given
application.
Enrollment time:
The time period a person must spend to have his/her facial reference template
successfully created.
Equal error rate:
When the decision threshold of a system is set so that the proportion of false
rejection will be approximately equal to the proportion of false acceptance.
This synonym is 'crossover rate'. The facial verification process involves
computing the distance between the stored pattern and the live sample. The
decision to accept or reject is dependent on a predetermined threshold.
(Decision threshold).
6. IMPLEMENTATION OF FACE RECOGNITION

TECHNOLOGY
The implementation of face recognition technology includes the following four
stages:
Data acquisition
Input processing
6.1 Data acquisition:

The input can be recorded video of the speaker or a still image. A sample of 1
sec duration consists of a 25 frame video sequence. More than one camera
can be used to produce a 3D representation of the face and to protect against
the usage of photographs to gain unauthorized access.
BBDNIIT
Page 16
6.2 Input processing:

A pre-processing module locates the eye position and takes care of the
surrounding lighting condition and color variance. First the presence of faces
or face in a scene must be detected. Once the face is detected, it must be
localized and
Normalization process may be required to bring the dimensions of the live
facial sample in alignment with the one on the template.
Some facial recognition approaches use the whole face while others
concentrate on facial components and/ or regions (such as lips, eyes etc). The
appearance of the face can change considerably during speech and due to
facial expressions. In particular the mouth is subjected to fundamental
changes but is also very important source for discriminating faces. So an
approach to persons recognition is developed based on patio- temporal
modeling of features extracted from talking face.
Models are trained specific to a persons speech articulate and the way that
the person speaks. Person identification is performed by tracking mouth
movements of the talking face and by estimating the likely hood of each model
of having generated the observed sequence of features. The model with the
highest likelihood is chosen as the recognized person.
FIGURE 6
BBDNIIT
Block diagram showing input Processing
Page 17
6.3 Face image classification and decision making:
FIGURE 7
Synergetic computer are used to classify optical and audio features,

respectively. A synergetic computer is a set of algorithm that simulates
synergetic phenomena. In training phase the BIOID creates a prototype called
face print for each person. A newly recorded pattern is preprocessed and
compared with each face print stored in the database. As comparisons are
made, the system assigns a value to the comparison using a scale of one to
ten. If a score is above a predetermined threshold, a match is declared.
From the image of the face, a particular trait is extracted. It may measure
various nodal points of the face like the distance between the eyes ,width of
nose etc. it is fed to a synergetic computer which consists of algorithm to
capture, process, compare the sample with the one stored in the database.
We can also track the lip movement which is also fed to the synergetic
computer. Observing the likely hood each of the samples with the one stored
in the database we can accept or reject the sample.
BBDNIIT
Page 18
7. FACE RECOGNITION ALGORITHMS

In the previous section we have shown that the task of face recognition
encounters complex variations. In order to cope with such complication and
find out the true invariant for recognition, researchers have developed various
recognition algorithms. In this section, we will describe two representative
ones. The eigenface approach applies the Karhonen-Loeve(KL) transform for
feature extraction. It greatly reduces the facial feature dimension and yet
maintains reasonable discriminating power. The neural network approach,
though some variants of the algorithm work on feature extraction as well,
mainly provides sophisticated modeling scheme for estimating likelihood
densities in the pattern recognition phase.
7.1 Eigenface
As mentioned, one of the goals that the feature extraction routine wishes to
achieve is to increase the efficiency. One simple way to achieve this goal is
using alternative orthonormal bases other than the natural bases. One such
basis is the Karhonen-Loeve (KL). KL bases are formed by the eigenvectors of
the covariance matrix of the face vector X. In the high dimensional "face"
space, only the first few eigenvalues have large values. In other words, energy
mainly locates in the subspace constituted by the first few eigen vectors.
Therefore, a great compression can be achieved by letting those eigenvectors
with
large
eigenvalues to represent the face
vector
where u is the eigenvector and M is usually much smaller than original vector
dimension N. Since the eigenvectors associated with the first few eigen values
look like face images, KL bases are also referred to as eigenfaces. The
eigenface representation is well known in statistics literature as the principal
component analysis. It is optimal in the sense of efficiency: for any given M <
N, the KL representation has the minimum mean square error among all
possible approximations of X that uses M orthonormal vectors. However, it
does not mean that the KL representation is optimal in the sense of
discriminating power, which relies more on the separation between different
faces rather than the spread of all faces.
Pentland's Photobook is one implementation of the eigenface algorithm. It
compresses a facial image with 128x128 pixels (16,384 pixels) into a vector
with only 40 eigenfaces (80 bytes). It recognizes 95% of the 200 faces chosen
from a large database with 7562 facial images (3000 different persons)
BBDNIIT
Page 19
7.2 Neural Network

In principle, the popular back-propagation neural network may be trained to
recognize face images directly. For even an image with moderate size,
however, the network can be very complex and therefore difficult to train. For
example, if the image is 128x128 pixels, the number of inputs of the network
would be 16,384. To reduce complexity, neural network is often applied to the
pattern recognition phase rather than to the feature extraction phase. Sung
and Poggios face detection algorithm (Sung, 1995) down-samples a face
image into a 19x19 facial feature vector before they apply the elliptical mean
clustering to model the distributions of the "face samples" and the "non-face
samples". Rowley et al. (Rowley, 1998) also reduce the dimension of the facial
image to 20x20 by down sampling before the facial image is fed into their
multi-layer neural network face detector. One example of the neural classifier
is the Probabilistic Decision- based Neural Network (PDNN) (Lin, 1997).
PDNN does not have the fully connected network topology. Instead, it divides
the network into K subnets. Each subnet is dedicated to recognize one person
in the database. PDNN uses the Gaussian activation function for its neurons,
and the output of each "face subnet" is the weighted summation of the neuron
outputs. In other words, the face subnet estimates the likelihood density using
the popular mixture-of-Gaussian model. Compared to the AWGN scheme,
mixture of Gaussian provides a much more flexible and complex model for
approximating the true likelihood densities in the face space. The learning
scheme of the PDNN consists of two phases. In the first phase, each subnet is
trained by its own face images (no cross training). In this phase, the weights
and bias are trained by the Expectation-Maximization (EM) algorithm
(Dempster, 1976). The EM has been proven to be an efficient algorithm for ML
estimation. In terms of system implementation, the advantage of the EM
algorithm is that it does not require the learning rate parameter. The learning
rate parameter in the conventional neural network training scheme controls the
speed of adjusting the network weights. The learning rate is often a sensitive
parameter; an improper selection may cause the whole network fail to
converge.
The second phase of the PDNN learning is called the decision-based learning.
In this phase, the subnet parameters may be trained by some particular
samples from other face classes. The decision-based learning scheme does
not use all the training samples for the training. Only those who are
misclassified are used. If a sample is miss-classified to the wrong subnet, the
rightful subnet will tune its parameters so that its "territory"(decision region)
can be moved closer to the misclassified sample. This learning process is also
known as the Reinforced Learning. In the meantime, the subnet that
wrongfully claims the identity of the questionable sample will try to move itself
away from the sample. This is called the Anti-reinforced Learning.
BBDNIIT
Page 20

Lin (1997) reports a performance comparison of various face recognition
algorithms (see Table 1). This comparison adopted the public Olivetti facial
database. There are 40 different persons in the database and 10 images per
person. There are variations in facial expression (open/close eyes,
smiling/frowning), facial details (with or without glasses),scale (up to 10%) and
orientation (up to 20 degrees). Five face recognition algorithms were
compared in this experiment.PDNN recognizer achieved 4% error rate. It
outperformed the eigenface-based recognizer, whose error rate is 10%. The
other three algorithms are SOM+CN (self organized map with convolution
neural network), HMM (Hidden Markov Model), and Pseudo 2D-HMM.
Figure 8 Performance of various face recognizer through data table
BBDNIIT
Page 21
8. HOW FACE RECOGNITION SYSTEMS WORK

An example
Visionics, company based in a New Jersey is one of the many developers of
facial recognition technology. The twist to its particular software, Face it is that
it can pick someone's face from the rest of the scene and compare it to a
database full of stored images. In order for this software to work, it has to know
what a basic face looks like. Facial recognition software is based on the ability
to first recognize faces, which is a technological feat in itself.
Figure 9 How face recognition systems work an example
BBDNIIT
Page 22

If you look at the mirror, you can see that your face has certain distinguishable
landmarks. These are the peaks and valleys that make up the different facial
features. Visionics defines these landmarks as nodal points. There are about
80 nodal points on a human face. Here are few nodal points that are measured
by the software.
distance between the eyes
width of the nose
depth of the eye socket
cheekbones
jaw line
chin
Figure 10
BBDNIIT
Another figure explaining working of face recognition Systems
Page 23

These nodal points are measured to create a numerical code, a string of
numbers that represents a face in the database. This code is called face print.
Only 14 to 22 nodal points are needed for face it software to complete the
recognition process.
9. METHODS OF FACE RECOGNITION

9.2 Two Dimensional Face Recognition-PCA
Principle Component Analysis (PCA) is one of the widely used 2d face
recognition algorithm it is based on information theory concepts, seeks a
computational model that best describes a face by extracting the most relevant
information contained in the face. The Eigenfaces approach is a PCA method,
in which a small set of characteristic pictures are used to describe the variation
between the face images. The goal is to find the eigenvectors (eigenfaces) of
the covariance matrix of the distribution, spanned by training a set of face
images. Later, every face image is represented by a linear combinations of
eigenvectors. Recognition is performed by projecting a new image onto a
subspace spanned by the eigenfaces and then classifying the face by
comparing its position in the face space in the positions of known individuals.
Figure 11 : Explanation of face recognition by 2D method
BBDNIIT
Page 24

The PCA-eigenfaces systems capture the image and change it to light and
dark areas. Both the initial facial image and the facial image in question are
also captured in a two dimensional form. Then, the two images are compared
according to the points of the two eigenface image. It picks out certain features
and calculates the distance between them. The points are facial features such
as eyes, nose, mouth, bone curves, and other distinct features. The eigenface
algorithm firstly forms overall average image. This is the image just adding all
images and dividing by number of images training set. And the eigenvectors of
the covariance matrix that is formed by combining all deviations of training
sets images from average image is formed in order to apply eigenface
algorithm. After finding overall average image, the order is to find eigenvector
of the covariance matrix. Visualization of eigenvectors is carried out simply
applying a quantization that is if they found eigenvectors have components
that are greater than 255 and smaller than 0 round them to 255and 0
respectively.
9.2 Three Dimensional Face Recognition

Three-dimensional face recognition (3d face recognition) is modality of facial
recognition methods in which the three-dimensional geometry of the human
face is used. Three dimensional face recognition has the potential to achieve
better accuracy than its two dimensional counterpart by measuring geometry
of rigid features of the face. This avoids such pitfalls of two dimensional face
recognition algorithms as change in lighting, different facial expressions,
make-up and head orientations. Another approach to use the three
dimensional model to improve the accuracy of traditional image based
recognition by transforming the head into the known view. Additionally, most
range scanners acquire both the three dimensional mesh and the
corresponding texture. This allows combining the output of pure three
dimensional matches with the more traditional two dimensional face
recognition algorithms, thus yielding better performance (as shown in FRVT
2006). The main technological limitation of three dimensional face recognition
method is the acquisition of three dimensional images , which usually requires
a range camera. This is also a reason why three dimensional face recognition
methods have emerged significantly later (in the late 1980s) than two
dimensional methods. Recently commercial solutions have implemented depth
reception by projecting the grid onto the face and integrating video capture of it
into a high resolution three dimensional model. This allows for good
recognition accuracy with low cost off-the-shelf components.
3D Face recognition system consists of four modules :
1. Device Module
2.
Data Processing Module
BBDNIIT
Page 25

3. Feature Extraction Module
3. Matching Engine Module
Figure 12 : Explanation of face recognition by 3D method
The device module acquires initial three dimensional facial data by a3D
surface scanner VGA camera and transfers it to the processor. After receiving
the raw data (the distorted pattern on the target object)
The Data Processing Module performs the image filtering (noise reduction)
and then instantly reconstructs the three dimensional face, smoothing and
interpolating the data to avoid holes and optimize the mesh.
The Feature Extraction Module receives the optimized 3D surface for further
feature vector (biometric template) extraction. During biometric template
extraction ,a proprietary two-stage algorithm is used. At the first stage, the
surface semantic analysis is performed, resulting in the location of key
crania-facial landmarks (points) on the facial surface and the fitting of the
surface patches (eye-sockets, super ciliarys arches, forehead zone,
nasolabial zone, chin zone, etc.) is known, information about local surface
BBDNIIT
Page 26

curvature characteristics is extracted. This local curvature information is used
further to build a single geometric descriptor and packs this data into biometric
template. The output of module is a biometric template uniquely characterizing
the person, which is used in the next matching stage.
The Matching Engine works in the identification mode, which compares the
extracted biometric template with all of the stored templates in the database
and produces the similarity score for each of the stored templates. The
template with the best similarity score is the top match.
10. THE SOFTWARE

Facial recognition software falls into a larger group of technologies known as
biometrics. Facial recognition methods may vary, but they generally involve a
series of steps that serve to capture, analyze and compare your face to a
database of stored images. Here is the basic process that is used by the Face
it system to capture and compare images:
10.1 Detection
When the system is attached to a video surveillance system, the recognition
software searches the field of view of a video camera for faces. If there is a
face in the view, it is detected within a fraction of a second. A multi-scale
algorithm is used to search for faces in low resolution. (An algorithm is a
program that provides a set of instructions to accomplish a specific task). The
system switches to a high-resolution search only after a head-like shape is
detected.
10.2 Alignment
Once a face is detected, the system determines the head's position, size and
pose. A face needs to be turned at least 35 degrees toward the camera for the
system to register it.
10.3 Normalization
The image of the head is scaled and rotated so that it can be registered and
mapped into an appropriate size and pose. Normalization is performed
regardless of the head's location and distance from the camera. Light does not
impact the normalization process.
BBDNIIT
Page 27
10.4 Representation
The system translates the facial data into a unique code. This coding process
allows for easier comparison of the newly acquired facial data to stored facial
data.
10.5 Matching
The newly acquired facial data is compared to the stored data and (ideally)
linked to at least one stored facial representation. The heart of the facial
recognition system is the Local Feature Analysis (LFA) algorithm. This is the
mathematical technique the system uses to encode faces. The system maps
the face and creates a face print, a unique numerical code for that face. Once
the system has stored a face print, it can compare it to the thousands or
millions of face prints stored in a database. Each face print is stored as an 84byte file. Using facial recognition software, police can zoom in with cameras
and take a snapshot of a face.
Figure 13 Person checking whether there is a match of a face or not
BBDNIIT
Page 28

The system can match multiple face prints at a rate of 60 million per minute
from memory or 15 million per minute from hard disk. As comparisons are
made, the system assigns a value to the comparison using a scale of one to
10. If a score is above a predetermined threshold, a match is declared. The
operator then views the two photos that have been declared a match to be
certain that the computer is accurate.
11. ADVANTAGES AND DISADVANTAGES

11.1Advantages:
a. There are many benefits to face recognition systems such as its
convenience and social acceptability. All you need is your picture taken for
it to work.
b. Face recognition is easy to use and in many cases it can be performed
without a person even knowing.
c. Face recognition is also one of the most inexpensive biometric in the
market and its price should continue to go down.
d. New York Times says about face recognition technology that after this
usage there would be none with just a face in the crowd.
e. This technology can be used to identify whether there is some one
standing in front of a camera or not as well as identifying characteristics
about that person such as age or gender without identifying who that
person is ?
f. It also in identifying individuals considered to be threats through facial
recognition.
11.2Disadvantage:
a. Face recognition systems cant tell the difference between identical
twins.
b. Face recognition is not perfect and struggles to perform under certain
conditions. Ralph Gross, a researcher at the Carnegie Mellon Robotics
Institute, describes one obstacle related to the viewing angle of the
face: "Face recognition has been getting pretty good at full frontal faces
and 20 degrees off, but as soon as you go towards profile, there've
been problems."
BBDNIIT
Page 29

c. Other conditions where face recognition does not work well include poor
lighting, sunglasses, long hair, or other objects partially covering the
subjects face, and low resolution images.
d. Another serious disadvantage is that many systems are less effective if
facial expressions vary. Even a big smile can render the system less
effective. For instance: Canada now allows only neutral facial
expressions in passport photos.
12. APPLICATIONS
The natural use of face recognition technology is the replacement of PIN,
physical tokens or both needed in automatic authorization or identification
schemes. Additional uses are automation of human identification or role
authentication in such cases where assistance of another human needed in
verifying the ID cards and its beholder.
Figure 14: Surveillance camera for comparing the images of various

people
BBDNIIT
Page 30

There are numerous applications for face recognition technology:
12.1Government Use
a. Law Enforcement: Minimizing victim trauma by narrowing mug shot
searches, verifying identify for court records, and comparing school
surveillance camera images to know child molesters.
b. Security/Counterterrorism: Access control, comparing surveillance images
to know terrorist.
c. Immigration: Rapid progression through Customs.
d. In protecting weaponries by giving its authority to selected members only.
12.2Commercial Use
a. Day Care: Verify identity of individuals picking up the children.
b. Residential Security: Alert homeowners of approaching personnel
c. Voter verification: Where eligible politicians are required to verify their
identity during a voting process this is intended to stop proxy voting where
the vote may not go as expected.
d. Banking using ATM: The software is able to quickly verify a customers
face.
e. Physical access control of buildings areas, doors, cars or net access.
f. Facial recognition systems are used to unlock software on mobile devices.
An independently developed Android Marketplace app called Visidon
Applock makes use of the phone's built-in camera to take a picture of the
user. Facial recognition is used to ensure only this person can use certain
apps which they choose to secure.
g. Face detection and facial recognition are integrated into the iPhoto
application for Macintosh, to help users organize and caption their
collections.
BBDNIIT
Page 31

h. Because of certain limitations [clarification needed] of fingerprint
recognition systems, facial recognition systems are used as an alternative
way to confirm employee attendance at work for the claimed hours.
i. Another use could be a portable device to assist people with
prosopagnosia in recognizing their acquaintances.
j. Helps an employer as by advantage of facial recognition technology he
dont have to remember large number of passwords and company is also
not afraid of any loss of data from any system.
k. The face recognition system can be used for one more application. It can
communicate as a wealth of social information :
a.
Age, gender, personal identity (physical structure)
b.
Mood and emotional state (facial expression)
c.
Interest / intentional focus (direction of gaze)
BBDNIIT
Page 32
13.CONCLUSION
Face recognition technologies have been associated generally with very costly
top secure applications. Today the core technologies have evolved and the
cost of equipments is going down dramatically due to the integration and the
increasing processing power. Certain applications of face recognition
technology are now cost effective, reliable and highly accurate. As a result
there are no technological or financial barriers for stepping from the pilot
project to widespread deployment.
Face recognition is a both challenging and important recognition technique.
Among all the biometric techniques, face recognition approach possesses one
great advantage, which is its user-friendliness (or non-intrusiveness). In this
presentation I have covered factors that may affect the performance of the
recognizer, and several face recognition algorithms. I hope this presentation
can provide the readers a better understanding about face recognition, and I
am encouraging the readers to increase their interest in this topic. The
readers who are interested in this topic can go to the references for more
detailed study.
BBDNIIT
Page 33
REFERENCES
1. Electronics for You: - Part 1 April 2001
Part 2 May 2001
2. Electronics World: - December 2002
3. IEEE Intelligent Systems - May/June 2003
4. Modern Television Engineering- Galati R.R
5. www.facereg.com
6. www.Imagestechnology.com
7. www.iee.com
BBDNIIT
Page 34

Face Recognition

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Face Recognition

Uploaded by

Copyright:

Available Formats

Face Recognition Technology