International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 07 Issue: 04 | Apr 2020 www.irjet.net p-ISSN: 2395-0072

BIOMETRIC EXAMHALL AUTHENTICATION

Rajesh N B1, Subhiksha R2, Oormila T S3
1
Assistant Professor, Electrical and Electronics Engineering Velammal College of Engineering and Technology
Viraganoor, Madurai, Tamil Nadu.
2, 3
Student, Electrical and Electronics Engineering Velammal College of Engineering and Technology Viraganoor,
Madurai, Tamil Nadu.
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract-Impersonation in exam halls is increasing day by for authentication and identification purposes. In order to
day. This is due to careless and time-consuming traditional overcome the limitations of unimodal biometric system
candidate checking and authentication system in exam halls. multimodal biometrics came into existence. It combines two
This project is designed to reduce impersonation in exam hall or more biometric data recognition results such as a
by verifying biometric features of the candidate. Our system combination of a subject's fingerprint, palm and iris that
consists of fingerprint scanner connected to arduino and iris increases the reliability of personal identification system
and palm vein data stored in mat lab database. The system is that discriminates the approved and the fraudulent. Besides
designed to pass only users by verifying their fingerprint, iris improving the accuracy, Multi-biometric systems are being
and palm vein and block non verified users. The fingerprint progressively establish in many large-scale biometric
system was designed to scan the fingerprint and ID number applications because they have several advantages such as
which were properly saved into the database of the system and lower error rates and larger population coverage compared
confirm the eligibility of candidate for examination. to uni-biometric systems. In this paper multimodal fusion of
Automated iris recognition is provided for the verification and iris and palm vein images along with fingerprint sensor is
identification of people as iris is highly distinctive to an proposed. Fingerprint based authentication is one of the
individual. The first image acquisition is processed and beneficial biometric technique and are easily accessible,
concerned with localizing the iris from a captured image. Then recognition requires minimal efforts on the part of the users,
it is matched with candidate data base entries. Also, the it does not capture information other than strictly necessary.
contactless palm vein authentication technology used here Automated iris detection is yet another option for human
captures an infrared ray image of the user’s palm and the identification and non-invasive verification. Interestingly,
sensor can capture the palm image regardless of the position the spatial patterns of the human iris are highly distinctive to
and movement of the palm. The matlab program then matches an individual. Among these, touch less palm vein
the translated vein pattern with the recorded pattern, while authentication technology is integrated because of its high
taking a pattern matching method to determine the position precision that comparing the vascular pattern under the
and orientation of the palm. If the details are authorized, the skin, which are unique to each individual. The system is
microcontroller now sends a signal to a motor driver. The designed to pass only candidates validate by their biometric
motor driver now operates a motor to open a gate. It ensures verification and block non-validate users.
only authorized users are allowed to enter the examination
section and unauthorized users are not allowed to enter 2. RELATED WORK
without any human intervention.
The application of biometric technology related to
Key Words: Matlab, impersonation, iris, palm vein impersonation has been used by Federal Bureau of
technology, Arduino. Investigation (FBI) in the 1960s. According to research on
biometric methods has gained renewed attention in recent
1. INTRODUCTION years brought on by an increase in security concern. A more
secured and accurate biometric based model is needed for
Recognition of person based on biometric feature is an
emerging phenomenon in our society. In the exam hall, implementation. Systems like sensing the fingerprints, iris,
authentication has always been a major challenge and face and palm of the candidate who logged in for the exam
verification of the authentic candidate is not an easy task and has been separately done by the researchers. These things
it consumes a lot of time and process. Traditional systems to are the unique features of all humans and hence can detect
verify a person’s identity are based on knowledge (secret the unauthenticated candidate easily. Some of such devices
code) or possession (ID card), however codes can be used in biometric system are
forgotten or overheard and ID cards can be lost or stolen
giving impostors the possibility to pass the identity test. The
use of features inseparable form of person’s body
significantly decreases the possibility of fraud. Biometrics
acts as a source for identifying a human being. This is used

© 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 6274
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 04 | Apr 2020 www.irjet.net p-ISSN: 2395-0072

2.1. FINGER PRINT (PAD): 3. METHODOLOGY

The presentation attack detection system acquires 3.1. BIOMETRIC EXAM HALL AUTHENTICATION
fingerprint image of a person within short wave infrared USING ARDUINO:
(SWIR) spectrum to determine the security that any person
could eventually fabricate or using gummy finger or face This project is to develop biometric based exam hall
mask to impersonate someone else. To tackle it, analysis of authentication systems that assist in the elimination of
the bona-fide properties below the skin within SWIR examination impersonation. Our system consists of a
spectrum is needed to be performed. fingerprint sensor connected to arduino microcontroller
circuit. In registration mode, the system allows to register up
2.2. FINGERPRINT SENSOR WITH PIC18F4520: to 120 users and save their identity with respective id
number. In addition to that our project includes automatic
The system operates by having each student’s finger is iris recognition and palm vein technology. The biomedical
scanned while the system cross check the fingerprint literature indicates that irises are as distinct as retinal blood
database that was captured during registration to verify if vessel signatures or patterns. Therefore prior to performing
the scanned fingerprint is valid. This proposed system has iris pattern matching, it is important to localize that portion
matching tendency of 25% of the threshold value set of the of the acquired image that corresponds to the iris. The
operation. knowledge about the veins is difficult to replicate because
the veins are internal to the human body. At last the
2.3. CONVOLUTION NEURAL NETWORK:
fingerprint, iris and palm image is fused to a single image
A bi-modular biometric acknowledgment framework is and determine the characteristics value for the three. After
recommended to rank-based combination calculation Bio obtaining the values the matlab checks out whether the
Maximum Inverse Rank (BMIR) for recognizing IITD iris values are same. If same, it will display output as
databases and CASIA datasets for palm print and unique authenticated otherwise it displays unauthenticated. Also,
mark are utilized for investigation with existing multi another check-in is there to enter the hall that is the
modular frameworks. Iris recognition components are more fingerprint sensor connected to arduino. It will check the
harmful and cause discomfort to human as they have to keep candidate’s fingerprint with their respective ID and motor
their eyes close to the illumination of the machine for will open the door if the candidate is authenticated
accurate scanning. otherwise buzzer will sounds on.

2.4. RASPBERRYPI: The component for this system comprises of hardware and
software. They are as follows:
Raspberry Pi's blend with the cloud has contributed to a
trend-setting period. Internet-of-things (IoT), or machine-to- 3.2. HARDWARE REQUIREMENTS
machine (M2 M) or machine-to-human (M2H)
communication. The biometric characteristics are 3.2.1. POWER SUPPLY:
transmitted to a remote location via an unsecured path,
updated AES-256 is applied to ensure a safe transmission. The 9-volt battery is a popular battery size that was
developed for early transistor radios, consisting mainly of
2.5. PALMVEIN SCANNER: alkaline batteries made of six individual 1.5 V LR61 cells in a
wrapper.
Palm vein authentication has a high level of authentication
accuracy due to the uniqueness and complexity of vein 3.2.2. DC MOTOR:
pattern also it is difficult to forge. The system is contactless
& hygienic for use in public areas. This scanner offers high A DC motor is one of a class of rotary electric motors which
cost and the information should always be encrypted. The converts electrical direct current into mechanical energy.
images made during the control shouldn’t be saved. The most common forms depend on magnetic-field forces.
Here, the arduino output is fed into a motor, which in turn
drives and then opens and closes the door.

© 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 6275
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 04 | Apr 2020 www.irjet.net p-ISSN: 2395-0072

3.2.3. ARDUINO: programming approach is to use the Arduino IDE, which

utilizes the C programming language. It gives you access to a
The Arduino Uno variant of the Arduino family is the most massive Arduino Library. Once it’s been opened, it opens
growing. The Arduino Uno is an ATmega328-based micro into a blank sketch where we can start programming
controller board with 14 digital input / output pins (of which immediately. First, we should configure the board and port
settings to allow us to upload code. Connect Arduino board
6 can be used as PWM outputs), 6 analog inputs, a 16 MHz
to the PC via USB cable.
ceramic resonator, a USB link, a power jack, an ICSP header,
and a reset button. The Arduino Uno is great choice for 3.3.2. MATLAB:
beginners. It contains everything you need to support the
micro controller; simply attach it to a device with a USB MATLAB® is a high-performance language for technical
cable, or power it to get going with an AC-to-DC adapter or computing. It integrates computation, visualization, and
battery. programming into an easy-to-use environment where
familiar mathematical notation expresses problems and
3.2.4. FINGERPRINTSENSOR: solutions. MATLAB is an interactive system whose basic
The basic function of fingerprint scanner is to get an image of element of data is an array which needs no dimensioning.
a person's fingerprint and find a match for this print in the This allows you to solve many technical computing problems
database. The capacitance scanner is better, because the in a fraction of the time it would take to write a program in a
images are more reliable and precise. scalar non-interactive language like C or FORTRAN,
particularly those with matrix and vector formulations.
3.2.5. RELAY:
The following is the block diagram about the software
It is a relay style single pole double throw (SPDT) with 5 pins
requirements of the system:
in a box type cube and rated to operate at 5VDC.Load
Current Max.: 7Amps 250V AC or 12Amps 24V DC. Coil
Resistance: 65-75 Ohms. The following block diagram
represents the hardware components:

Fig: 3.3.2.1. Software development

4. PROPOSED SYSTEM

Biometric based exam hall authentication system helps to

eliminate the unauthenticated candidate in the exam hall
with the help of fingerprint sensor and matlab simulation
with the database registered before. The overall system can
be viewed in three stages. They are,
Fig: 3.2.5.1. Hardware development
4.1. PRIMARY UNIT:

3.3. SOFTWARE REQUIREMENTS 4.1.1. SENSING:

3.3.1. ARDUINO IDE: The first unit is sensing unit. Here we use fingerprint sensor
connected to arduino microcontroller circuit. It will check
Arduino is an open-source company of hardware and the candidate fingerprint with their respective I’d which was
software computers. The development boards are known as registered before.
Arduino Modules, which are open-source prototyping
platforms. The simplified microcontroller board comes in a
variety of development board packages. The most common

© 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 6276
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 04 | Apr 2020 www.irjet.net p-ISSN: 2395-0072

4.1.2. IMAGE ACQUISTION: individual are compared against the stored data is used to
determine the user identity.
In this step, the images of palm-vein, fingerprint and iris are
acquired by the matlab simulation. The images are converted 4.3. FINAL UNIT:
to the round off size of 256*256.
The last step is to detect the matching of the fingerprint
4.2. PROCESSING UNIT: which is sensed by the sensor at present with the respective
Id’s by the arduino. Also the final encrypted information is
It is the technique in which the input data is obtained as then stored in the database and decrypted image is matched
image by the arduino in hardware session and as matrix with the current query in matlab simulation. The verification
form in mablab process. In arduino adafruit fingerprint module for the current query template to identify an
library is used. individual for authorization.

Biometric exam hall authentication technique implements

the flow of sequential process in matlab.

They are as follows

This step consists of

4.2.1. ENROLLMENT PROCESS:

To register a fingerprint, the enroll(id) is called in arduino

program. The ID passed is passed to link the scanned
fingerprint also each fingerprint has an unique ID number. If
the scanner is able to read the fingerprint, it will ask to
remove and then replace the same finger on the scanner.
This happens because granting access of the matching
fingerprint.

4.2.2. GRAY SCALE CONVERSION:

The image is converted to gray scale because OpenCV

functions tend to work on the gray scale image representing
vectors, matrices, images and linear algebra functions in
matlab. Grayscale images are also called monochromatic,
denoting the absence of any chromatic variation. The resize
level of the image is converted to gray scale image and then Fig- 4.3.1. Block diagram of the matlab processing
to matrix. From the matrix it checks the contrast, correlation,
entropy, homogeneity and energy levels of the three images. 5. EXPERIMENTAL RESULTS
These characteristics are essential to describe the quality of
the image. The figure shows the final prototype of BIOMETRIC
EXAMHALL AUTHENTICATION system. It consists of
4.2.3. SUPPLEMENTARY PROCESS: Arduino placed on board with fingerprint sensor placed
along with motor, relay and buzzer. It is cost effective and
The gray scale images are subjected to fusion process in beneficial.
matlab. Multimodal biometrics can be accomplished by
fusion of two or more images, in which the resulting fused
image is secured more. Fusion at feature extraction level
generates a homogeneous template for fingerprint, iris and
palm vein features. The feature extraction of three biometric
traits fused using feature level fusion and encrypted using
RSA and stored in a database for desired authentication and
verification. The newly captured biometric traits of the

© 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 6277
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 04 | Apr 2020 www.irjet.net p-ISSN: 2395-0072

AUTHENTICATION OF IRIS

The extracted iris image is compared with image stored in

database and result is shown if it is authenticated or not
authenticated.

Fig: 5.1. Prototype model

5.1. CODE COMPILATION

Fig: 5.2.3. Fusion image of authenticated

Fig -5.1.1. Code compilation

The code compilation is done using Arduino 1.8.8-Windows Fig: 5.2.4. Fusion image of unauthenticated
compiler and output is converted into microcontroller
readable language and fused into microcontroller. 5.3. FUSION PROCESS:

5.2. SIMULATION The iris and palmvein database are fused together into single
image through fusion.
The simulation is made to run using Matlab software.

Fig: 5.3.1 Fusion process using matlab

Fig: 5.2.1. Gray image conversion 6. FUTURE SCOPE

However, there are some limitations in the proposed

solution which can be addressed in the future
implementations. Hence, the following features are
recommended to be incorporated in the future versions.
 Integration of all these scanner units and making it
as a kit fixed at doorstep of all exam halls.
 complexity of design can be reduced and made
Fig: 5.2.2. Gray scale matrix compact.
 Fusing information extracted from the red, green
and blue components of an image might produce a
better fused feature vectors which contains richer
information than that in gray scale image.

© 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 6278
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 04 | Apr 2020 www.irjet.net p-ISSN: 2395-0072

7. CONCLUSION images,” IEEE Trans. on Information Forensics and Security,

vol. 7, no. 6, pp. 1717–1726, 2012.
The advent of fast-growing technologies is making users
have hi [8].A. Rattani and A. Ross, “Automatic adaptation of
fingerprint liveness detector to new spoof materials,” in
gh-security systems with options for electronic Proc. IEEE International Joint Conference on Biometrics,
identification. The identification based on biometric or 2014, pp. 1–8.
fingerprint authentication is the efficient and reliable
solution for stringent protection. Microcontroller-based [9].Maneesh Upmanyu, Anoop M. Namboodiri, Kannan
review hall authentication device based on fingerprint. This Srinathan, and C. V. Jawahar, “Blind Authentication: A
system is more accurate and faster than previous feature- Secure Crypto-Biometric Verification Protocol,” IEEE
extraction Iris technology is expanding in real time Transactions on Information forensics and security, vol. 5,
applications of security measures since it is stable, secure no. 2, June 2010, pp.225-268.
and authentic. Authentication of the palm vein is one of the
most accurate of authentications. This paper represents the [10].Ajay Kumar, Vivek Kanhangad, and David Zhang, “A
authentication system for contactless palm veins that takes New Framework for Adaptive Multimodal Biometrics
the design of the blood vessels as a personal identification. Management,” IEEE Transactions on Information forensics
By verifying all these parameters, if user is valid then allows and security, vol. 5, no. 1, March 2010p, p. 92-102.
attending the exam else not allowed.
[11].Abhishek Nagar, Karthik Nandakumar, Anil K. Jain, and
Dekun Hu, “Multibiometric Cryptosystems Based on
8. REFERENCES
Feature-Level Fusion,” IEEE Transactions on Information
[1]. Ruben Tolosana, Marta Gomez-Barrero, Christoph Busch, forensics and security, vol. 7, no. 1, February 2012, pp. 255–
Javier Ortega-Garcia, Fellow, “Biometric Presentation 268.
Attack Detection: Beyond the Visible Spectrum”, IEEE
[12].Koen Simoens, Julien Bringer, Herve Chabanne, and
Transactions on Information Forensics and Security,
Stefaan Seys, “ A Framework for Analyzing Template
10.11.19/TIFS.2019.2934867 (Main Paper).
Security and Privacy in Biometric Authentication
[2].J. Galbally, S. Marcel, and J. Fierrez, “Biometric Systems,” IEEE Transactions on Information forensics and
antispoofing methods: A survey in face recognition,” security, vol. 7, no. 109 – 116 2, April 2012, pp. 833-841.
IEEE Access, vol. 2, pp. 1530–1552, 2014.
[13].Q. Zhang, Y. Yin, D. Zhan and J. Peng, "A Novel Serial
[3].R.Tolosana, R. Vera-Rodriguez et al., s“Exploring Multimodal Biometrics Framework Based on
recurrent neural networks for on-line handwritten Semisupervised Learning Techniques," in IEEE Trans. on
signature biometrics,” IEEE Access, pp.1– 11, 2018. Info. Forensic and Sec., vol. 9, no. 10, pp. 1681-1694, Oct.
2014.
[4].A.Rattani, W.Scheirer, and A. Ross, “Open set fingerprint
spoof detection across novel fabrication materials,” IEEE [14]. B. E. Manjunathswamy, J. Thriveni, and K. R. Venugopal,
Trans. on Information Forensics and Security, vol. 10, no. 11, “Bimodal biometric verification mechanism using
pp. 2447–2460, 2015. fingerprint and face images(BBVMFF),” in Proc. IEEE 10th
Int. Conf. Ind. Inf. Syst. (ICIIS), pp. 372-377, 2015.
[5]. D. Menotti, G. Chiachia et al., “Deep representations for
iris, face, and fingerprint spoofing detection,” IEEE Trans. [15].Y. Lin, E. Y. Du, Z. Zhou, and N. L. Thomas, “An efficient
on Information Forensics and Security, vol. 10, no. 4, pp. parallel approach for Sclera vein recognition,” IEEE
864–879, 2015. Trans. Inf. Forensics Security, vol. 9, no. 2, pp. 147–157, Feb.
2014.
[6].A. Toosi, A. Bottino, S. Cumani, P. Negri, and P. L. Sottile,
“Feature fusion for fingerprint liveness detection: a
comparative study,” IEEE Access, vol. 5, pp. 23 695–23 709,
2017.

[7].F. Nicolo and N. A. Schmid, “Long range cross-spectral

face recognition: matching SWIR against visible light

© 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 6279