BIOMETRICS IMPLEMENTATION ON

RASPBERRY PI
Submitted to

SAVEETHA INSTITUTE OF MEDICAL AND TECHNICAL SCIENCES

In partial fulfilment for the award of the degree of

BACHELOR OF ENGINEERING

IN

COMPUTER SCIENCE AND ENGINEERING

By

K.vijayakanakamahalaxmi(191711440)
Pathipati yasasvi(191711291)
y.swetha(191711467)
Supervisor
Mr.R.Senthilkumar

SAVEETHA SCHOOL OF ENGINEERING

SAVEETHA INSTITUTE OF MEDICAL AND TECHNICAL
SCIENCES, CHENNAI – 602 105
 BONAFIDE CERTIFICATE
Cerfied that this project report “BIOMETRIC IMPLEMENTATION ON
RASPBERRY PI” is the bonafied work of “k.vijaya, p.yasasvi, y.swetha
(regno:191711440, 191711291, 191711467)” who carried out the project work
under my supervision.

SIGNATURE SIGNATURE

Dr. CHOKKALINGAM.SP
HEAD OF THE DEPARTMENT
Professor, Dept.of CSE&IT
Saveetha School of Engineering
SIMATS, Chennai - 602 105
MR.R.SENTHILKUMAR
PROJECT SUPERVISOR
Associate Professor, Dept. of CSE
Saveetha School of Engineering
SIMATS, Chennai – 602105

INTERNAL EXAMINER EXTERNAL EXAMINER

 DECLARATION BY THE CANDIDATE

I declare that the report entitled“BIOMETRICS IMPLEMENTATION ON

RAPBERRY PI”submitted by me for the degree of Bachelor of Engineering is
the record of the project work carried out by me under the guidance
of“Mr.R.Senthilkumar”and this work has not formed the basis for the award
of any degree, diploma, associateship, fellowship, titled in this or any University
or other similar institution of higher learning.

SIGNATURE

k.vijaya(Reg.No. 191711440)
p.yasasvi(regno.191711291)
y.swetha(regno.191711467)
 ACKNOWLEDGEMENT

This project work would not have been possible without the contribution of many people. It

gives me immense pleasure to express my profound gratitude to our honorableChancellor,Dr.

N. M. Veeraiyan, Saveetha Institute of Medical and Technical Sciences, for his blessings

and for being a source of inspiration. Isincerely thank our Vice Chancellor,Dr.

JawaharNesan, for his visionary thoughts and support.I am indebted to extend my gratitude

to our Director Madam,Mrs. Ramya Deepak, Saveetha School of Engineering,

forfacilitating us all the facilities and extended support to gain valuable education and

learning experience.

I register my special thanks to Dr. D. Dhanasekaran, Principal, Saveetha School of

Engineering and Dr. Chokkalingam.S

P, HoD, Department of Computer Science and Engineering, for the support given to

me in the successful conduct of this project.I wish to express my sincere gratitude to my

supervisor,Mr.R.Senthilkumar, for her inspiring guidance, personal involvement and

constant encouragement during the entire course of this work.

I am grateful to Project Coordinators, Review Panel External, Internal Members and

the entire faculty of the Department of Computer Science and Engineering, for their

constructive criticisms and valuable suggestions which have been a rich source to improve

the quality of this work.

Submitted by
k.vijaya
p.yasasvi
y.swetha
 ABSTRACT

Developments in the field of Information Technology also make

Information Security a devoted part of it. In order to deal with security,
Authentication plays an imperative role. In this paper, Biometrics is
used for authentication. This paper also describes how biometrics can
leverage cloud’s boundless computational resources and striking
properties of flexibility, scalability, and cost reduction in order to
reduce the cost of the biometrics system requirements of different
computational resources (i.e. processing power or data storage) and to
enhance the performance of biometrics systems’ processes (i.e.
biometric matching). Here, Raspberry Pi is used to build a low-cost
biometric system. Raspberry Pi (RPi) is a credit-sized mini-computer
with great capabilities similar to a PC. In this study it is used as a
remote enrollment node. The application of Raspberry Pi and cloud
computing has given a new direction of research into the field of
Internet-of-Things (IoT). Using the biometric technology, a new system
of IoT based biometrics is proposed. To maintain the security of
biometric traits over the Internet channel from RPi client to the cloud,
cryptographic algorithms are applied like RSA and enhanced AES-256.
The encrypted biometric information is stored on the cloud and the
authentication can be done by Biometric service hosted on Azure cloud.
Thus, this papers covers the following topics: attracting power of
biometrics into the authentication services, biometrics leveraging the
power of cloud, Raspberry Pi- a low-cost IoT device,
 KEYWORDS

Internet-of-Things (IoT)
Raspberry Pi (RPi)
cloud
biometrics
biometric security
cryptography
AES-256 encryption

PROJECT OBJECTIVE

The conventional authentication technologies like RFID tags and

authentication cards has a lot of weakness, biometric method of
authentication is a prompt replacement for this. Biometrics such as
fingerprints, voices and ECG signals are unique human characters that
cannot be tampered or replicated. This facilitates real time system
implementations. And it is also proven to be more accurate with less
than 2 seconds of processing time, facilitating the authentication system
to be faster and reliable.

The proposed project uses finger print module that can detect the finger
print of a user and facilitate the authentication and attendance system.
The project will have two hardware devices, first is the handheld device
that will be there in all the places (like classrooms, security doors etc)
and other will act as the local server for all the handheld devices present
.
 PROJECT DESCRIPTION

1. Raspberry Pi: It is a small size minicomputer that is capable of doing small computing and
networking operations which can be done by a computer system. Also it comes with GPIO
pins which is the main element in the field of internet of things. For instance, Raspberry Pi 3
model has 1.2 GHz 64-bit quad core ARMv8 CPU, and RAM of 1GB. And also it has 40
GPIO pins, Full HDMI port, 4 USB ports, Ethernet port, 802.11n wireless LAN
connectivity, Bluetooth 4.1 connectivity, low energy bluetooth, 3.5mm audio jack, video
Camera interface (CSI), the Display interface (DSI), and Micro SD card slot.

2. Raspbian OS: The raspbian operating system is an open source and free operating system
which is a Debian based OS. Raspbian provides the basic set of programs and software
utilities, also comes with more than 35,000 raspbian packages which are precompiled
software.

3. Python: Python is the programming language that is used to operate the Raspberry Pi. It is
considered as one of the powerful programming languages out there to operate a
microcontroller. Basically, Python programming language is used as a scripting language
for Linux. Generally Python program contains a series of commands and the program will
be executed by the computer from top to bottom.

4. Zigbee: The robot uses Zigbee to establish the connection. It works on 2.4GHz ISM band
with 20~250kbits/s data rate and has a transmission range of upto 1.5 km.

5. Arduino Uno: The digital and analog input/output pins are equipped in boards that may be
interfaced to various expansion boards and other circuits. Serial communication interface is
a feature in this board, including USB which will be used to load the programs from
computer.

6. Finger Print Module: This module helps in the identification through fingerprint images.
And in general it can be split up into the following tasks, finger print scanning-finger print
classification-finger print comparison. During the classification process, finger print images
are optionally allocated to a certain category based on the global orientation of the ridges
while the location of the minutiae is marked as well. The comparison is divided into the
following six step.
 Scanning of a finger print image: The quality of the scanned
finger print image is the decisive factor for the identification
purpose. This can be achieved by using a high-definition finger
print scanner which can tolerate the skin types, colours,
damages and dryness factors.
 Image quality improvement: Here an optical improvement is
applied to the structure on the scanned finger print image
 Image processing: This is the preparatory phase for the feature
extraction and classification purposes
 Feature classification: All the finger prints show certain
global similarities facilitating a rough classification. There are
3 principal finger classes exist. In this process the image is
classified under anyone of the principal classes.
 Feature extraction: In this process the location of the minutiae
(ridge bifurcation & ridge endings) in the finger print is
detected and extracted. At real time, the quality of the finger
print image impacts this process a lot. So proper care should be
taken to avoid the negative influence caused due to poor
quality image.
 Verification process: In this process two features are
compared. The algorithm functioning strongly depends on the
quality of the extracted minutiae and the comparison image.
PROJECT IMPLEMENTATION

1. First, the finger print of the users should be stored in the server to
recognize the user. You can program the system in such a way that
whenever you press ‘E’ the enrolment action will be activated. And
then the finger print of any user can be stored.
2. During the authentication or attendance process, you can press the
‘ST’ button to start the process. A green LED glows and indicates
the status of the system. Then the handheld device can used for the
authentication purpose.
3. And once the authentication process is done you can press ‘SP’
button to stop the process and the green LED will stop glowing.
4. The finger print modules is connected to Arduino and the serial
communication is done through the ZigBee module, which
establishes the connection with Raspberry Pi
5. Once the authentication or attendance system process is over, the
data will be uploaded to the cloud server for storage and analysis
purposes
 SYSTEM HARDWARE
REQUIREMENTS

The proposed system is a multimodal biometric system, face and

fingerprint are the biometric traits under consideration.
The hardware requirements for the remote enrolment node are listed
below: In this research, Raspberry Pi 2 Model B is used which costs US
$35.
It has 4 USB ports, a HDMI port for connection with the display, micro
SD card slot for booting and data storage as RPi doesn’t have on-board
storage.
Also it has 10/100 Mbit/s Ethernet port for internet connection.
To make RPi portable in this paper wireless USB Wi-Fi adapter is used.
The OS used is Raspbian (Debian wheezy). RPi needs power supply of
5V-800mA (4.0 W) . In order to make the proposed portable, RPi is
supplied power through power bank. RPi 2 has 1GB RAM and CPU
speed is 900 MHz quad core ARM Cortex-A7. It is a Broadcom2835
System-on-chip hardware. One powerful features of the Raspberry Pi is
the row of GPIO pins along the edge of the board as shown in Figure 1.
These pins are a physical interface between the Pi and the outside
world.
 PROPOSED SYSYTEM

The proposed IoT based biometrics architecture consists of 3 modules

as shown in Figure 2: Raspberry Pi as a remote enrolment node,
enhanced AES-256 for security and the Azure cloud for storage,
scalability and performance concerns. The FS88 scanner in interfaced
using LibScan API and libusb libraries on the RPi. After the interfacing
of webcam the RPi is ready for enrolment process. As soon as the
motion sensor detects motion, a desktop application pops up which
initiates the enrollment/login process. After the image capturing
process, the biometrics are encrypted on the RPi on Mono Developer
(C# language) using the proposed AES-256 algorithm along with
Round structure and dynamic S-box generation based on pseudo noise
sequence generator as explained below. The last step is to upload the
encrypted images on the Azure cloud where they are stored in blobs
within the register/login containers depending on whether the user is
already registered or he is a new user. After the decryption process on
the cloud, the original biometric traits are retrieved.

Fig.no.:1
 Fig.no.:2

RASPBERRY pi fingerprint sensor

Fig.no.:3
 Figureno.4

Figure no.5 raspberry pi face recognigation

Figure no 6 RASPBERRY PI
 RESULT

This section focuses on the results after the implementation of the IoT
based system. Results are produced in following way: Raspberry Pi as a
remote enrolment node, proposed AES-256 analysis and the uploading
of encrypted images to the register/login containers.
 CONCLUSION

In this project, presented a low-cost IoT based biometrics architecture.

Raspberry Pi was successfully implemented as a remote wireless
enrolment node. Also the encryption module was efficiently executed
on RPi. The encrypted biometric traits was sent from RPi client to the
Azure cloud for decryption. The proposed system can be used for
security and access control mechanisms like unlocking a door, logging
details of a person entering and exiting a building, attendance
management, accessing a particular service etc. This system can be
applied at all places where authentication is required. Avalanche effect
depicted that the IoT based biometric system is highly secured.

FUTURE WORK

The future work includes decryption on cloud, incorporation of

recognition module on the Azure cloud and hence authentication will be
carried on cloud which will increase the performance and scalability of
the biometric system. Enhanced AES-256 algorithm is used to handle
the privacy and security of the biometric data. Since decryption process
is executed on the cloud, the original biometric information of users are
within the boundaries of cloud service providers. To solve this issue
homomorphic encryption can be used. Unlike traditional encryption,
homomorphic encryption allows data processing on the encrypted data.
This will assure that users’ privacy is not compromised.
Biocryptographic systems can be used where biometrics and
cryptography are employed together to provide privacy enhanced
biometric authentication capabilities.
 REFERENCE
/
1. Anil K. Jain, Karthik Nandakumar, and Abhishek Nagar, Review
Article: Biometric Template Security, Journal on Advances in Signal
Processing Volume 2008, Article ID 579416.
2. Debnath Bhattacharyya, Rahul Ranjan1, Farkhod Alisherov A., and
Minkyu Choi, Biometric Authentication: A Review, International
Journal of u- and e- Service, Science and Technology, Vol. 2, No. 3,
September, 2009.
3. http://biometrics.gov/Documents/Glossary.pdf, National Science
and Technology Council’s (NSTC) Subcommittee on Biometrics,
Biometrics Glossary, 2006.