Final year paper draft - Copy

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

Next-Gen Biometric ATM Protection and Monitoring System

Dr. Kavitha Jayaram Namitha B M Rajath Prabhu Shree Raksha R


Associate Professor, Dept. of Dept. of Computer Science and Dept. of Computer Science and Dept. of Computer Science and
Computer Science and Engineering Engineeering Engineeering Engineeering
BNM Institute of Technology BNM Institute of Technology BNM Institute of Technology BNM Institute of Technology
Bengaluru, India Bengaluru, India Bengaluru, India Bengaluru, India
kavithajayaram@bnmit.in namithaboggaram@gmail.com rajathbprabhu@gmail.com raksha4643@gmail.com

Abstract-- The rise in electronic ATMs have come a high-value target for hackers and
transactions has increased the demand for stealers. ATMs are vulnerable to hacking, fraud, thievery,
quick and accurate user identification and security breaches. To descry similar anomalous
and authentication. Biometric authentication in ATMs behaviour, a stable system is needed. The maturity of
has several advantages. Biometric authentication makes ATMs now have surveillance cameras installed. still, this
use of fingerprints and facial recognition. The use will not help ATM thieveries. Some ATM theft occurs due
of password-based PINs is a disadvantage of to a lack of security guards. As a result, there's a critical need
the current ATM authentication method. Because for an effective system that detects suspicious exertion and
PINs are easily tracked and misused. Our proposed prevents thievery by arresting the miscreant before they
system is designed to provide greater protection for flee. As a result, public safety is assured, crime is reduced,
ATMs in order to achieve security and resolve these and serious tragedies are avoided.
criminal activities. In this case, along with PINs,
The proposed model in this paper employs biometric
OTP is generated at random and sent via IoT. The
authentication, which includes fingerprint and face
project's goal is to completely eliminate the use of
recognition authentication and OTP generation and
ATM cards. After biometric and OTP pin
verification. The disadvantage of the current ATM
authentication, the customer will be able to
authentication pattern is the use of PIN as a password. In this
proceed with the transaction. If three
case, along with the PINs, a randomly generated OTP is sent
consecutive incorrect attempts are made, the account
via IoT services. In the event of a robbery, the sensor
will be blocked. This initiative also addresses the
detects the vibrations on the ATM, the surrounding area is
prevention of ATM fraud. If any suspicious activity is
alerted by a buzzer, and the door to the ATM room is
detected by the vibration sensor, the ATM doors are
immediately closed. A fainting gas is also pumped inside the
closed, the fainting gas is released, and
ATM to render the thief unconscious.
the surrounding area is alerted. This will
catch the perpetrator in the act and prevent the II. Related Works
fraud from occurring.
Keywords— IoT; Biometric authentication; The dramatic increase in ATM usage motivates
microcontroller; face recognition; fingerprint sensor; researchers to investigate ATM development, security, and
OTP (One Time Password); ATM. enhanced facilities. Samina Anjum, Achal Sontakke, Aditya
Tapase and Asfiya Saba Sheikh replaced ATM cards with
I. Introduction fingerprints and user ID of all the users were stored in the
database. Fingerprint scanner was used to acquire the
In 1966, the first Automated Teller Machine( ATM) was individual's fingerprint [1]. C.Karthik, V Praveen Kumar,
used in London. numerous theoretical and practical ATM P.Bhargavi, P Nagendra Babu used the face recognition
studies have been conducted around the world. This is due to technique to authorize users. The system compares the
the expansion of ATM options in order to achieve a cashless individuals' face input to the face details of the users in the
frugality. The ATM is a type of electronic database and determines whether or not a match is found. By
telecommunications device that allows for fiscal deals comparing the data of the individual given with all of the
similar as cash recessions, cash deposits, finances transfers, other individuals in the database, the system generates a
and mileagepayments. ATM fraud has come a global ranked list of matches. After the verification, an OTP will be
problem that has soared in recent times. ATM fraud affects generated to proceed with the transaction [2]. Gyanprakash
both guests and bank drivers. Singh and Pragati Goel, used fingerprint identification,
They have been dealing with criminals and security threats which sped up the transaction while also increasing security.
to the existing card system. The current ATM system relies on The proposed system uses a biometric fingerprint to replace
PINs and ATM cards for authentication, which has several the existing ATM transaction card system. False Rejection
disadvantages. Criminals use techniques such as ATM Rate (FRR), False Acceptance Rate (FAR), and Average
skimming, cash trapping, shoulder surfing, and card trapping Matching Time (AMT) tests on the application illustrate the
to steal ATM cards and their information. Some customers adequacy and applicability for ATM user verification and
use their phone numbers or birth dates as PINs, which are authentication [3]. Face recognition technology analyzed the
easily guessable by fraudsters or hacked by cybercriminals. unique shape, pattern, and positioning of the facial features.
ATM card and PIN problems can be solved using biometric
authentication. This is due to the fact that the biological
details are unique and cannot be replicated by others.
The facial image of the user was stored in the database
at the time of registration. So, if any user wanted to
withdraw an amount from their account, then that user had
to scan their face at the camera present at the ATM.
This system used HOG-SVM algorithm for face
recognition [4]. First, the user was asked for the account
number and ATM PIN, after which the user received an
OTP on his/her registered mobile number. After logging
into the OTP, the user was asked to withdraw or
deposit. A GSM Modem was used to send the OTP
to the user[5]. Pratiksha Shetiya, Meryl
Mascarenhas and Mrunal Deshmukh offered a new
solution for security by analyzing the random pattern of
the iris. The iris recognition system automatically (a)
recognized the identity of a person from a new eye
image by comparing it to the human iris patterns stored
in an iris template database[6]. Sumanth C M,
developed a secure unintelligible PIN authentication
protocol for ATM terminals using personal mobile
devices (SPAQ)[7]. SPAQ was a mobile application that
allowed a user to scan a QR code from the screen of a point-
of-service terminal and connected to the bank’s SPAQ server
to obtain secure one-time-use PIN templates[7].
(b)
III. Working
FIGURE 1: (a) BLOCK DIAGRAM OF ATM
We have posited a new concept that enhances the overall SOFTWARE AND (b) SECURITY SYSTEM
experience, usability, security, and convenience of the HARDWARE
transaction at the ATM. First, the face and fingerprint of the
customer should be received and uploaded to the database. Users can access any bank account, but it must
To achieve interaction with the customer, a user interface is be linked to their ID. Customers can perform
created using Python GUI. The whole transaction will be ATM functions such as account details, withdrawal,
shown through the user interface. The image of the face and deposit, and money transfer here. They can also view
fingerprint of the customer is coded into digital data and then their previous transaction with that account. Following
it gets stored in the database. While registering biometric a successful proceeding, the user can choose to continue
details, the customer will be provided with a bank account or exit. This system will be created using Python and
and balance. They also want to provide a 4-digit PIN while hardware peripherals to provide a low-cost ATM
registering their account to the administrator of the system. This system also allows for access to multiple
corresponding bank. bank accounts.
When a customer wishes to use an ATM, he or she The ATM security system protects the system
must choose if the user is authorized or unauthorized user. If from robbery. It also helps the Cops to catch the
authorized, user inserts the RFID card and enters the PIN. Robber without any problems. When anyone tries to
Then the user must place their finger on the fingerprint damage ATM or break into ATM, the vibration sensor
module. Following acceptance, they should show their face detects the vibration. The vibration sensor triggers an
to the camera in front of them. Both biometric information is input to the microcontroller. Then microcontroller
converted into digital code. This digital data is now triggers a buzzer and red LED to alert the surrounding.
compared to a list of registered customers' biometric data. If A servo motor will act as an ATM door controller.
the data matches, the user must select the bank for the
transaction process. If unauthorized, user inserts the RFID IV. Biometric Authentication
card and enters the PIN. The system will send a confirmation
message to the corresponding account holder's mobile phone A. Face recognition
via IoT, as illustrated in Figure 1(a). If the account holder's The first step for many facial-related applications, such
confirms the transaction, then an OTP is generated and sent as face recognition or authentication, is generally
to the account holder, the account holder forwards it to the face detection. Face recognition involves the automated
user, which is then verified by the system. If the provided identification of individuals by analyzing facial features and
data does not match, the system will not allow the process to patterns. The process typically begins with face detection,
continue. After entering the OTP, the system validates it where algorithms locate and extract facial regions from an
against the corresponding data. After three failed attempts, image or video. Feature extraction follows, capturing key
the account will be blocked. If all three authentication facial attributes like the arrangement of eyes, nose, and
matches then the customer asks to select a Bank from the mouth.
given option. After this selection customer allows to make
the transaction in the respective Bank account.
These features are then transformed into a V. Result & Discussion
unique template representing the individual's face.
During recognition, a live face is compared to Thus, the hardware and the software
stored templates in a database using matching setups are linked together and processed for User
algorithms. The degree of similarity determines authentication and security protocols. Tight security is
the likelihood of a match, allowing the system maintained while the transaction process, as well as
to identify the person. Face recognition is physical security, is also provided for the ATM
widely used for various applications, including environment, which leads to safe and secure ATM
security, access control, and personalized transactions.
technology, capitalizing on the distinctiveness and The User interfaces is developed using the Python
perceptual ease of facial characteristics. GUI library called tkinter.
1. Face Detection: Detecting faces in the picture is The integration capabilities of tkinter allow it to
the very first task we do. We will now get the precise interact harmoniously with other ATM Surveillance System
coordinates of the face for further analysis. modules, such as the backend handling biometric
authentication, OTP generation, and database management,
2. Feature Extraction: The technique involves ensuring a cohesive and efficient user experience. Overall,
selecting and transforming raw data into a reduced and the role of tkinter in this project is to build a user-friendly
informative representation, facilitating efficient analysis and interface that simplifies user interactions while seamlessly
pattern recognition. integrating with the system's core functionalities for optimal
3. Comparing faces: Now that we have facial performance and usability.
patterns for every face in our database, we can compare them.
We recognize the face if the pattern produced is similar to or VI. Conclusion
similar to any other patterns. As the result of fraudulent activities increased in
B. Fingerprint Recognition the only card-based system, this proposed ATM
system uses biometric and OTP for authorization. The
Fingerprint verification operates by capturing and biometric authorizations include both fingerprint
analyzing unique patterns present in an individual's and face recognition, in which the fingerprint of every
fingerprints for secure identification. The process begins with individual is uniquely identified. The OTP is sent to the
the acquisition of a fingerprint image, typically through corresponding account holder's mobile phone through
optical or capacitive sensors. Ridge and valley patterns, IOT Services. This combined system provides secure
minutiae points (ridge endings, bifurcations), and other ATM cash transactions. The vibration detection
distinctive features are extracted from the fingerprint. These system detects the suspicious behaviour of the person
features are then encoded into a template, a condensed digital by sensors. This system helps the Cops to catch the
representation. During verification, a live fingerprint is Robber without any problems.
similarly processed, and its template is compared with a
stored template in the database. Matching algorithms assess The proposed system is a prototype of an ATM
the similarities between the live and stored templates, system in which money withdrawal facilities can be
determining the degree of correlation. If the match surpasses added to be implemented at the ATM. In addition to
a predefined threshold, the verification is successful, this, the release of fainting gas will also be shown by a
confirming the individual's identity. Fingerprint verification prototype model. By this means of three successive
is renowned for its accuracy and reliability in personal identity verification and security system, we can
authentication due to the uniqueness and permanence of provide safety and security to both customers and
fingerprint patterns. bank's money.

C. One-Time Password(OTP) Authentication


References
The OTP generation process is based on a secure algorithm
within the ATM system that generates a unique code based [1] Samina Anjum, Achal Sontakke, Aditya Tapase,
on various factors such as transaction details, timestamp, and Asfiya Saba Sheikh, Huzaif Sheikh, "Fingerprint
user identity. This code is quickly delivered to the registered Based ATM Systems". International Research
mobile number via SMS or other secure communication Journal of Modernization in Engineering
channels, ensuring its confidentiality and integrity. Technology and Science, Vol.5, Issue.4, April 2023.
To complete the transaction, the user must enter the received [2] C.Karthik, V Praveen Kumar, P.Bhargavi, P Nagendra
OTP into the ATM interface, thereby validating their Babu. "A Protected and Improved ATM Security System
ownership of the registered mobile device and confirming the exploitation Image Capture and SMTP
authenticity of the transaction request. The OTP acts as a one- Protocol". Industrial Engineering
time code that is only valid for a short period of time, Journal, Vol.50, Issue.7, July 2021.
typically a few minutes, enhancing the transaction's security [3] Gyanprakash Singh and Pragati Goel, "ATM
by ensuring its time-bound validity. Using Fingerprint", International Journal of Research
Publication and Reviews, Vol.3, no.7, July 2022.
[4] Dr. Y.V. RAM KUMAR, Kotikalapudi Satya Syamala
Kameswari, Pattabhi Keerthana, Koppanati
Sri Harshavardhan, Boosa Pavan Vinay, Kusampudi
Sri Venkata Abhishikth, "An Efficient Application to
Provide Security to The Banks Using Face
Recognition Using Open CV", Journal of Science and
Technology, Vol.8, Issue.4, April 2023.
[5] Chandana P, Kushi Sangani and A. Rengarajan, "Study
on ATM with OTP for a safe and smart future",
International Journal for Research in Applied Science
and Engineering Technology, Vol.10, Issue.2, Feb 2022.
[6] Pratiksha Shetiya, Meryl Mascarenhas and Mrunal
Deshmukh, "ATM Security System using Iris
Recognition by Image processing", International Journal
of Engineering Research and technology, Vol.9,
Issue.7, July 2020.
[7] Sumanth C M, "Securing ATM Transactions Using QR
Code based Secure PIN Authentication",
International Journal of Scientific Research in
Computer Science Engineering and Information
Technology, Vol.5, Issue.3, June 2019

You might also like