Date: 2/10/2021
Synopsis
PROJECT TITLE: “Securing Online Transaction Using Face
Recognition”
Batch No: #08
Sl. No. USN Name E-mail id Contact number
01 1RN18CS004 Adithi V Kempadala
[email protected] 9663366371
02 1RN18CS109 Sushmitha N
[email protected] 7795096130
03 1RN19CS410 Rashmi C
[email protected] 98454 55333
Internal Guide Details : Karnam Sunil Kumar
Company Name (Optional) :RNS Institute of Technology
Address : Channasandra, Bangalore
External guide Name :
Designation :
Contact number :
E_mail_ID :
Approved/ Permission letter :
CONTENTS
SL NO Content Name Page No
1. Objective/ Aim of the project 3
2. Introduction 4
3. Technical flaws about existing system 5
4. Technical details about the proposed project
a. Control / Data flow diagram 5
5. Usefulness of Project 6
6. Requirements specification 7
7. Current Status of Development 7
2
OBJECTIVE OF THE PROJECT
As the technology is growing the security to the personal information stored
in online website are at higher risk. The hackers / scammers are finding one
or the other way to cheat the people , one of them is during transacting
online though the password and otp are sent to our personal info the security
during transaction is not ensured.
There has been no full proof method to “authenticate” a customer in an
online credit card transaction. Authentication is the verification of a credit
card owner made during a credit card purchase. In the physical world
authentication is achieved through a physical signature which is manually
checked at the point of sale. Without effective authentication, following
problems may arise like fraudulent transaction, lengthy process of
transaction including lack of confidence for customers, higher cost of
transactions etc. So in our project we will be providing face recognition
method to authenticate a user during transactions.
3
INTRODUCTION
Since information of users can leak easily, many authentication approaches
are proposed to secure data and systems. Authentication method should be
able to “declare anyone that they are who they claim to be”. However, it is
not important in authentication process that what identity is authorized for.
Moreover, they can be disclosed and stolen by intimates or colleagues. Some
systems use an object such as passport or ID card to identify a user. It is
practical in use but also is very easy to be stolen or copied. Therefore,
biometrics which is unforgettable and more unique becomes a choice for
secure process. Biometric technologies can be fingerprint, face, iris, palm
print or even hand geometry. Among these technologies, face image seems
to be one of the most popular used features due to its usability, collectability
and acceptability. For usability and collectability, face technology is non-
intrusive technology as test subjects are not required for a direct contact. For
acceptability, many studies claim that their correct recognition rates are
more than 95%.
Here we are using Local Binary Pattern (LBP) as our base face recognition
algorithm. This algorithm is efficient and has been customized for our
specific purpose and more efficient face recognition. It is a simple yet very
efficient texture operator which labels the pixels of an image by thresholding
the neighborhood of each pixel and considers the result as a binary number.
We implement the LBPH using OpenCV. OpenCV was started at Intel in
1999 by Gary Bradsky and the first release came out in 2000. Right now,
OpenCV supports a lot of algorithms related to Computer Vision and
Machine Learning and it is expanding day-by-day.
Currently OpenCV supports a wide variety of programming languages like
C++, Python, Javaetc and is available on different platforms including
Windows, Linux, OS X, Android, iOS etc. Also, interfaces based on CUDA
and OpenCL are also under active development for high-speed GPU
operations. OpenCV-Python is the Python API of OpenCV. It combines the
best qualities of OpenCV C++ API and Python language.
4
Technical Flaws
Issues and Challenges Followings are the issues and challenges which are
faced in system.
1) Less Security in proposed system.
2) Lack of awareness of personal information in transaction.
3) Lack of awareness of future development.
4) High-cost maintenance
Data Flow Diagram
5
Usefulness of Project
Using facial recognition means that the banking customer has only one
face which can allow them to access to all their bank accounts. In
cybercrime, a hacker can infiltrate a person’s social media profile and
learn key information which can enable them to answer security
questions such as ‘what was the name of your first pet?’ The biometric
software only authenticates a customer based on who they are. A
hacker can only manage to access a bank account based on what they
know about the customer. So this makes it easier for a banking
customer to transact online on a mobile device by using a device that
has authorized access and secondly by using a live facial image of them
via the phone’s camera.
Facial Recognition software has a liveness detection which prevents
hackers from using a picture of the customer for impersonation
purposes. It also applies to other biometric modalities such as
fingerprints where the liveness detection does exactly that – it assesses
the ‘liveness’ of the facial image to guarantee that it is a live image and
not a (still) picture or a spoof as it is known. The recognition system
also allows customers to access their bank accounts from computers. It
is possible via the webcams which are built-in particularly in laptops.
While logging into their account, facial biometrics is an additional
layer of security.
6
System Requirements
1.1.1 Hardware
• Processor: Intel Duo Core
• RAM: 8 GB
• GPU: NVIDIA Tesla K80
1.1.2 Software
• Operating System: Ubuntu (14) (64bit), Windows 10
• Programming Languages: C, Python, JavaScript, CSS,
php5, html5
• Deep Learning Framework: Darknet
• Server: Apache2
Current Status of Project
1. As of know we are creating dataset for the analysis and analyzing the
LBHP theory for implementation.
2. HTML code for website is done.
3. CSS code is completed.
4. PHP code under completion
