Face recognition using Extended local binary pattern

1. Abstract:
Now a day’s keeping data or object safe is getting complicated due to digital
password, pins, pattern are easily getting hacked. For the beyond the digitalized
security lock everyone looking towards a better solution to get the data/object safe
by using various biometric techniques like finger prints, iris scan, as well as hand
geometry, the most efficient and more widely used one is the face recognition This
is because it is inexpensive, non-intrusive and natural. Therefore, researchers have
developed dozens of face recognition techniques over the last few years . Efficiency
of matching the face is always low because of the code or the data base utilized to
run the code. In this project we are mainly focusing on three parts, namely face
representation, feature extraction and classification. Face representation represents
how to model a face and determines the successive algorithms of detection and
recognition. The most useful and unique features of the face image are extracted in
the feature extraction phase. In the classification the face image is compared with
the images from the database. In our project, we empirically evaluate face
recognition which considers both shape and texture information to represent face
images based on Local Binary Patterns for person-independent face recognition.
The face area is first divided into small regions from which Local Binary Patterns
(LBP), histograms are extracted and concatenated into a single feature vector. With
that we are using PCA, LDA for better results.

2. Introduction:

Biometrics is a method of human identification by analyzing their physical and

behavioral characteristics. This method is user friendly because it uses physical
traits such as face recognition, recognition of palm, fingerprint and iris scanning.
Face recognition received great amount of attention due to the high recognition
rate since face recognition take into consideration both physical and behavioral
characteristics.
In this method we are using LBP, LDA, PCA for image preprocessing, feature
extraction and matching process. In the preprocessing stage, the image is
normalized by mean of image cropping and resize before being input to the next
stage. The LBP algorithm comes into play in the feature extraction stage. Finally,
the matching process will decide whether a particular face image match to the
database image. The metric used in the matching process is chi-square distance.
Face Recognition:
Face recognition is to use the facial characteristic for recognition and judgment.
• It is one of the biometric techniques.
• Two types of comparison in face recognition:
 Verification:
It means a 1:1 match that compares a face images against a template
face images whose identity being claimed.
 Identification:
It means a 1: N problem that compares a query face image against
all image templates in a face data base.

WHY FACE RECOGNITION?

• It does not require any physical contact with an image capturing device.
• It does not require any advanced hardware as it can be used with existing
image capture devices. Face recognition consists of three main steps:
1. Face Detection.
2. Feature Extraction.
3. Classification.

3. LITERATURE SURVEY:
Early face recognition algorithms used simple geometric
models, but the recognition process has now matured into a science of
sophisticated mathematical representations and matching processes. The face
recognition problem is made difficult by the great variability in head rotation,
angle, facial expression, lighting intensity.

4. Problem statement:

In real world safe guarding a something most likely the data/ object is not
possible with locks and passwords. Because the technology we much more
increase that helps the criminals to hack the system with different pattern,
password is happened. Now a days everyone is looking for a better option of safe
guarding not by pattern or password somewhat like biometric techniques in that
face recognition gives more efficient.

5. Objective:
 1. Increasing the efficiency in face verification (1:1) by using Extended Local
Binary Pattern (ELBP) under various conditions.
2. Increasing the efficiency in face identification (1: N) by using Extended
Local Binary Pattern (ELBP) under various conditions.

6. Methodology:

By using Local Binary Pattern (LBP), the face

images are mapped into a face subspace for
analysis. Different from Principal Component
Analysis (PCA) and Linear Discriminant
Analysis (LDA) which effectively see only the
Euclidean structure of face space, LPP finds
an embedding that preserves local
information, and obtains a face subspace that
best detects the essential face manifold
structure. The Laplacian faces are the optimal
linear approximations to the eigen functions of
the Laplace Beltrami operator on the face
manifold. In this way, the unwanted variations
resulting from changes in lighting, facial expression, and pose may be eliminated
or reduced.
Theoretical analysis shows that PCA, LDA, and LBP can be obtained from
different graph models. We compare the proposed Laplacianface approach with
Eigenface and Fisher face methods on three different face data sets. Experimental
results suggest that the proposed Laplacianface approach provides a better
representation and achieves lower error rates in face recognition.
Principal Component Analysis (PCA) is a statistical method under the broad
title of factor analysis. The purpose of PCA is to reduce the large dimensionality of
the data space (observed variables) to the smaller intrinsic dimensionality of
feature space (independent variables), which are needed to describe the data
economically. This is the case when there is a strong correlation between observed
variables. The jobs which PCA can do are prediction, redundancy removal, feature
extraction, data compression, etc. Because PCA is a known powerful technique
which can do something in the linear domain, applications having linear models
are suitable, such as signal processing, image processing, system and control
theory, communications, etc.
The main idea of using PCA for face recognition is to express the large 1-D vector
of pixels constructed from 2-D face image into the compact principal components
of the feature space. This is called eigenspace projection. Eigenspace is calculated
by identifying the eigenvectors of the covariance matrix derived from a set of
fingerprint images (vectors).

7. Applications:
 Retail Stores
Many retail stores are increasingly adopting face recognition to identify repeat
customers and offer them special services. They are also using it to derive data and
analyse the performance of their stores. Demographic details like gender and age
can be used to reveal the kind of customers that frequent the store. Businesses can
then optimise their products to drive more sales. The way things are headed, every
store in the future will greet you by name!
 Casinos
Casinos are actively using facial recognition systems in order to recognize their
customers. They then use it to track them, identify them on the web, and let them
know how often they’re going to the casino. Recognition allows them to identify
repeat customers and estimate frequency of their visit. They can then use it to help
gambling addicts, reward loyal customers, identify fraudsters, and even enforce
bans.
 Dating sites
With explosion in dating sites/apps and users showing keen interest in such
services, businesses have come up with all sorts of crafty matching mechanisms to
help users find their soulmates. Among such sites are the likes of
findyourfacemate.com. The makers of this site believe (more like claim) that
people are most attracted to those that look like them. They use neural nets to map
faces to a mathematical space. People whose faces are nearby in this space are then
suggested by the app as potential mates. Similarly, Doggleganger can also match
you up with a DOG that looks like you! crazy!.
 Government Agencies
This one comes as no surprise. In countries like US and China, government
institutions like FBI etc. have been using face recognition for a long time to
identify criminals. The Chinese have gone full ninja with this technology, with
some reports suggesting that the government can identify criminals even before
they commit the crime! They’ve been known to snoop on crowds at sporting events
and other places. More power to the authorities.
 Social Media
Facebook and its gazillion photos of 2 Billion users uses Facial Recognition to
identify everyone. There was a huge uproar and protest over users getting tagged in
strangers' photos and the massacre of privacy. Following these, FB has decided to
turn it off by default... for now.
 Shopping
Companies like MasterCard are researching about ways to enable payment
verification through the face. The advantages are numerous but the primary
motivation is preventing fraud and identity theft. Among other advantages is that
one won't have to remember passwords or put the credit card information on the
web. This will greatly reduce losses banks incur due to international credit card
theft/hacking mafia.
In the future, you will smile to pay for everything your wife forces you to buy!
 Hotels and Restaurants
Hotels and restaurants are beginning to use this service to identify customers even
before they enter the door. This allows them to offer specialized services to each
customer by taking into account their past preferences. Restaurants can also cater
to their guests by serving them their favorite meal by preparing it as soon as they
arrive.
 Bars picking out underage drinkers
Teenagers using fake Ids to get a drink have been a problem for quite some time.
With age recognition technology it is possible to determine age from a person's
face. This will prevent teenagers from getting fake ids and hopefully reduce black
market activity.
 Schools have started using it to take attendance
Many schools have started installing this technology to automatically track
attendance of pupils. This gives more time to teachers for teaching and maximizes
productivity. It also prevents students from faking attendance or other mischief. It
is also possible to keep a full-eye on the campus and find and track bullies.
  Access Control in Offices
Many offices and government buildings are now using face recognition based
authentication for seamless access control. This prevents, among other things, tail
gating where people who forgot their IDs run behind their colleagues to pass the
access point. It also increases security by immediately reporting potential breach
attempts. Its also possible to comprehensively track an individual inside a
campus/building and identify malicious visitors.

8. Advantages:
1. Security levels will be significantly improved
2. The integration process is easy and flawless
3. High accuracy allows avoiding false identification
4. Facial Recognition System is fully automated
5. Time fraud will be excluded

9. Tools Required:
MATLAB (matrix laboratory) is a multi-paradigm numerical computing
environment and proprietary programming language developed by MathWorks.
MATLAB allows matrix manipulations, plotting of functions and data,
implementation of algorithms, creation of user interfaces, and interfacing with
programs written in other languages, including C, C++, C#, Java, Fortran and
Python.
Although MATLAB is intended primarily for numerical computing, an optional
toolbox uses the MuPAD symbolic engine, allowing access to symbolic computing
abilities. An additional package, Simulink, adds graphical multi-domain simulation
and model-based design for dynamic and embedded systems.
Most MATLAB functions can accept matrices and will apply themselves to each
element. For example, mod(2*J,n) will multiply every element in "J" by 2, and
then reduce each element modulo "n". MATLAB does include standard "for" and
"while" loops, but (as in other similar applications such as R), using the vectorized
notation often produces code that is faster to execute. This code, excerpted from
the function magic.m, creates a magic square M for odd values of n (MATLAB
function meshgrid is used here to generate square matrices I and J containing 1:n).
[J,I] = meshgrid(1:n);
A = mod(I + J - (n + 3) / 2, n);
B = mod(I + 2 * J - 2, n);
M = n * A + B + 1;
 Structures
MATLAB has structure data types. Since all variables in MATLAB are arrays, a
more adequate name is "structure array", where each element of the array has the
same field names. In addition, MATLAB supports dynamic field names (field
look-ups by name, field manipulations, etc.). Unfortunately, MATLAB JIT does
not support MATLAB structures, therefore just a simple bundling of various
variables into a structure will come at a cost.
 Functions
When creating a MATLAB function, the name of the file should match the name
of the first function in the file. Valid function names begin with an alphabetic
character, and can contain letters, numbers, or underscores. Functions are often
case sensitive.
 Function handles
MATLAB supports elements of lambda calculus by introducing function handles,
function references, which are implemented either in .m files or anonymous/nested
functions.
Most MATLAB functions can accept matrices and will apply themselves to each
element. For example, mod(2*J,n) will multiply every element in "J" by 2, and
then reduce each element modulo "n". MATLAB does include standard "for" and
"while" loops, but (as in other similar applications such as R), using the vectorized
notation often produces code that is faster to execute. This code, excerpted from
the function magic.m, creates a magic square M for odd values of n (MATLAB
function meshgrid is used here to generate square matrices I and J containing 1:n).
[J,I] = meshgrid(1:n);
A = mod(I + J - (n + 3) / 2, n);
B = mod(I + 2 * J - 2, n);
M = n * A + B + 1;
 Structures
MATLAB has structure data types. Since all variables in MATLAB are arrays, a
more adequate name is "structure array", where each element of the array has the
same field names. In addition, MATLAB supports dynamic field names(field look-
ups by name, field manipulations, etc.). Unfortunately, MATLAB JIT does not
support MATLAB structures, therefore just a simple bundling of various variables
into a structure will come at a cost.
 Functions
When creating a MATLAB function, the name of the file should match the name
of the first function in the file. Valid function names begin with an alphabetic
character, and can contain letters, numbers, or underscores. Functions are often
case sensitive.
 Function handles
MATLAB supports elements of lambda calculus by introducing function handles,
or function references, which are implemented either in .m files or
anonymous/nested functions.
 Graphics and graphical user interface programming
MATLAB supports developing applications with graphical user interface (GUI)
features. MATLAB includes GUIDE (GUI development environment) for
graphically designing GUIs. It also has tightly integrated graph-plotting features.
For example, the function plot can be used to produce a graph from two vectors x
and y. The code:
x = 0:pi/100:2*pi;
y = sin(x);
plot(x,y)
produces the following figure of the sine function:
Matlab plot sin.svg
A MATLAB program can produce three-dimensional graphics using the functions
surf, plot3 or mesh.
10. Plan of Work:
Description of work Duration Remarks
Data Collection 20/12/2018 to 12/01/2019
Literature Survey 15/01/2019 to 28/01/2019
Problem Identification 29/01/2019 to 02/02/2019
Methodology Design 03/02/2019 to 08/02/2019
Software Implementation 09/02/2019 to 09/03/2019
Testing and Analysis of 10/03/2019 to 12/03/2019
results
Presentation in outside 15/03/2019 to 17/03/2019
Thesis Preparation 18/03/2019 to 05/04/2019
Final Submission 06/04/2019 to 10/04/2019

REFERENCES
1. Evaluation of LBP-based Face Recognition Techniques
Siti Anis Amirah Mohd Faudzi, Norashikin Yahya Electrical and
Electronics Engineering Department, Universiti Teknologi PETRONAS
2. Face Recognition using Local Binary Patterns (LBP) By Md. Abdur
Rahim, Md. Najmul Hossain, Tanzillah Wahid & Md. Shafiul Azam Pabna
University of Science and Technology, Bangladesh
3. Review of Face Recognition Techniques Kamini Solanki Ph.D Scholar
Computer Science Dept. RK University Rajkot India, Prashant Pittalia, PhD
Computer Science Dept. NICM Gandhinagar India
4. A Survey of 2D Face Recognition Techniques
Mejda Chihaoui , Akram Elkefi, Wajdi Bellil and Chokri Ben Amar
REGIM: Research Groups on Intelligent Machines, University of Sfax,
National School of Engineers (ENIS),
Sfax 3038, Tunisia;
5. A Literature Survey on Face Recognition Techniques Riddhi Patel, Shruti
B. Yagnik Department of Computer Engineering, Gujarat Technological
University Gujarat, India Department of Information Technology, Gujarat
Technological University
Gujarat, India.
6. A Review Paper on Biometrics: Facial Recognition
Sakshi Goel, Akhil Kaushik, Kirtika Goel, (CSE Department, IIMT Meerut,
UP Email: sakshi.kur15@gmail.com) (CSE Department, SIT Meerut, UP
Email: akhil.cse.07@gmail.com) (CSE Department, SIT Meerut, UP Email:
Kirtikagoel_88@yahoo.co.in
7. Future of Face Recognition: A Review
Shwetank Arya, Neeraj Pratap, Karamjit Bhatia
Department of Computer Science, GKV, Haridwar, India
8. SURVEY OF AUTOMATIC FACIAL EXPRESSION RECOGNITION
BASED ON CLASSIFICATION SCHEMES
P. Poornima, S. Radhapriya Research Scholar, 2Assistant Professor,
Department of Computer Science, Government Arts College, Coimbatore
9. Automatic Analysis of Facial Actions: A Survey
Brais Martinez, Member, IEEE, Michel F. Valstar, Senior Member, IEEE,
Bihan Jiang, and Maja Pantic, Fellow, IEEE
10. A Review Paper on FACIAL RECOGNITION AnkurBansal,
MukeshAgarwal, Anima Sharma, Anindya Gupta B.Tech,CSE, JECRC,
Jaipur, Rajasthan, India, (ankurbnsl02@gmail.com) Associate Professor,
CSE, JECRC, Jaipur, Rajasthan, India, (mukeshsir@gmail.com) Senior
Lecturer,CSE, JECRC, Jaipur, Rajasthan, India,
(animasharma2004@yahoo.co.in) Senior Lecturer, CSE, JECRC, Jaipur,
Rajasthan, India, (eranindyagupta@gmail.com)
11. METHODOLOGY AND EXTENSIONS OF LOCAL BINARY
PATTERN: A SURVEY
CHANDRAJA D
Department of Electrical Engineering (UG Student), BITS Pilani Hyderabad
E-mail: chandraja.dharmana@gmail.com
12. A Survey on Different Face Detection Algorithms in Image Processing
Doyle Fermi, Faiza N B, Ranjana Radhakrishnan, Swathi S Kartha, Anjali S
U.G. Student, Department of Computer Engineering, Model Engineering
College, Thrikkakara, Cochin, India
Assistant Professor, Department of Computer Engineering, Model
Engineering College, Thrikkakara, Cochin, India
13. Detailed Survey of Different Face Recognition Approaches Yogish Naik
Department of Studies in Computer Science and MCA, Kuvempu
University, Karnataka
14. Detecting Faces in Images: A Survey
Ming-Hsuan Yang, Member, IEEE, David J. Kriegman, Senior Member,
IEEE, and Narendra Ahuja, Fellow, IEEE
15. Face Recognition: A Survey Muhammad Sharif, Farah Naz, Mussarat
Yasmin, Muhammad Alyas Shahid and Amjad Rehman Department of
Computer Science, Comsats Institute of Information technology WahCantt
MIS Department CBA Salman bin Abdulaziz University Alkharj KSA
16. A Survey on Sentiment Classification in Face Recognition JingyuQian1
Engineering School, University of Michigan, Michigan MI 48105, US
egbert_qian@163.com