GOVERNMENT POLYTECHNIC COLLEGE

MEPPADI, WAYANAD

SEMINAR REPORT ON

SECURE ATM BY IMAGE PROCESSING

SUBMITTED BY

ATHUL KB
2101132763

DEPARTMENT OF COMPUTER ENGINEERING

2023-2024
 GOVERNMENT POLYTECHNIC COLLEGE MEPPADI,
WAYANAD

DEPARTMENT OF COMPUTER ENGINEERING

2023-2024
CERTIFICATE

Certified that seminar work entitled “SECURE ATM BY IMAGE PROCESSING”

is a bonafide work carried out in the 5 th semester by ATHUL KB (2101132763) in
partial fulfilment for the award of Diploma in Computer Engineering from Govt
Polytechnic College, Meppadi, Wayanad during the academic year 2023-2024.

Internal Guide: Head of Department:

Internal Examiner: External Examiner:

II
 ACKNOWLEDGEMENT

I take this opportunity to express my sincere gratitude to Mr. JOHNSON JOSEPH, Head of
Department, Computer Engineering, GPTC Meppadi for his valuable guidance and support. I am
indepted to our beloved principal Mrs. SWARNA C, Principal, GPTC Meppadi for his valuable
support and motivation. I am also thankful to Mrs. RAJEENA, Lecturer, Computer Engineering for
his valuable guidance and support. I thank all teaching and nonteaching staffs of Computer Science
and Engineering Department for their support and help. Last but not least I thank my parents and
friends who helped me in completion of my Seminar.

ATHUL KB

II
 ABSTRACT

With the rapid evolution of technology and the increasing instances of fraudulent activities
at Automated Teller Machines (ATMs), there arises a pressing need to fortify the security
measures surrounding these financial systems. This paper proposes a novel approach to
bolster ATM security by leveraging advanced image processing methodologies.The primary
objective is to develop a robust system that utilizes computer vision algorithms for real-time
surveillance and authentication at ATMs. The proposed system integrates a multi-layered
security framework, employing techniques such as facial recognition, object detection, and
anomaly detection.Through the implementation of image processing algorithms, the system
aims to authenticate users, detect unauthorized access or tampering, and prevent fraudulent
activities such as skimming or card trapping. Furthermore, the system is designed to raise
alerts in case of suspicious behavior or potential security threats, thereby enhancing the
overall security posture of ATMs.The study includes the design architecture,
implementation details, and evaluation of the proposed system using real-world scenarios
and datasets. Additionally, it discusses the potential challenges, ethical considerations, and
future research directions in the domain of securing ATMs through image
processing.Overall, this research presents a promising avenue for fortifying ATM security
by harnessing the power of image processing techniques, paving the way for more resilient
and safeguarded financial systems in the digital age.

4
IV
 CONTENTS

TITLE PAGE NO:

CHAPTER 1: INTRODUCTION 1

CHAPTER 2: ATM AND SECURITY THREATS 2

CHAPTER 3: LITERATURE REVIEW 3

CHAPTER 4: BIOMETRICS 4

CHAPTER 5: IMAGE PROCESSING 5

CHAPTER 6: METHODOLOGY 6

CHAPTER 7: FACIAL RECOGNITION 7

CHAPTER 8: IMPLEMENTATION 8

CHAPTER 9: ADVANTAGES 9

CHAPTER 10: DISADVANTAGES 10

CHAPTER 11: CONCLUSION 11

CHAPTER 12: REFERENCE 16

5
 CHAPTER 1
INTRODUCTION

The rise of technology has brought into force several types of tools that aspire at more customer
pleasure. ATM is a machine which has made money transactions effortless for customers. But it has
both advantages and disadvantages. Current ATMs make use of not more than an access card and PIN
for uniqueness confirmation. This exposes ATMs to a lot of fake attempts to use them by means of card
theft, PIN theft, stealing and hacking of customer’s account details. Using Face Recognition System in
ATMs can show the way to deal with such cases. Biometrics innovation permits persistence and
demonstrate of onece personality through physical qualities. It transforms your body into your secret
key. We talked about different biometric procedures. That is all are retina filter, finger examine, facial
output, hand check and so on there are two calculations have been configuration by taking biometric
systems to verification an ATM account holder or the record client, empowering a safe ATM by picture
handling. Biometrics is currently accessible in any resembled in different open and private segments
moreover. No more issues if passwords and I'd codes have been overlooked, biometrics is the
innovation that deals with it, making your body your secret key. Typically To make your mystery word
assurance and advancement controls logically exhaustive, the more problematic it will be for customers
to review their passwords. Unfortunately, to stop essential software engineer strikes on the framework,
serious mystery key rules are required. The current accessible age security issue is viewed as the
fundamental TCP/IP encryptions and different variables that are given by the utilizing system. Be that
as it may, there was a considerable lot of predictable distinguishing proof of each one separately, at that
point the recently created innovation is Biometrics, came into picture. Biometrics can be utilized to
evade the unapproved access to ATM, PDAs, home security frameworks, entryway locks, keen cards,
work area PC's workstations.

6
 CHAPTER 2
ATM AND SECURITY THREATS

Automated Teller Machines (ATMs) have revolutionized banking by providing convenient, round-the-
clock access to financial services. However, their widespread use and reliance on technology have
made them vulnerable to a range of security threats. Understanding these threats is crucial to
implementing effective security measures.

ATM Overview:
Functionality:
ATMs allow users to perform various financial transactions, including cash withdrawals, deposits,
balance inquiries, funds transfers, and more. They rely on a combination of hardware (card reader, cash
dispenser, PIN pad) and software systems to facilitate these transactions.

Components:
1. Card Reader: Reads data from the user's ATM card.
2. PIN Pad: Allows users to input their Personal Identification Number (PIN).
3. Cash Dispenser: Dispenses cash after authentication.
4. Network Connectivity: Establishes communication with the bank's servers for transaction
authorization.
Security Threats:
1. Skimming:
Skimming devices are illicitly installed on ATMs to capture card information. These devices can
overlay devices on the card reader or hidden cameras to record PINs, compromising user data.

2. Card Trapping:
Fraudsters use devices or sticky substances to trap the user's card inside the machine. They retrieve the
card later to gain unauthorized access or clone it.

3. Cash Trapping:
Similar to card trapping, cash trapping involves devices that prevent cash from being dispensed
properly. Criminals retrieve the cash once the user leaves the ATM.

4. PIN Theft:
Hidden cameras or fake PIN pads record PINs, allowing fraudsters to access accounts. Shoulder
surfing, where individuals observe PIN entry, is also a threat.

5. Malware and Skimming Software:

ATMs can be infected with malware or skimming software that captures card information or
manipulates transaction processes, leading to unauthorized access or fund theft.

.
 CHAPTER 3

LITERATURE REVIEW
The utilization is the no more issues if client overlooked there passwords and id codes
additionally, In truth of the biometrics is the innovation that deals with it. which is transforms
your body into the your passwords. The all the more requesting you make your own secret
key decision and development the guidelines done the more trouble to the clients can have in
recalling their passwords. However, severe and hard secret word rules are important to keep
away from simple programmer assaults on the system.

The essential downside with secret key is two folds. And furthermore They are assignable,
that can be recorded onto the paper and that can exchanged to somebody who shouldn't have
them. Furthermore, they will overlooked. As of late examination proposes that the overlooked
secret phrase will prize as much as US$ 340 for each occasion, the possibility and expenses of
course of action passwords are a central point. In the reality the significant and principle
requirement for additional dimension of security has offered ascend to field of
"BIOMETRICS" Biometric ATM authentication system based on fingerprint[1 ]. We can
refer to fingerprint scanning technology with the help of this paper.

Biometric technology used in a wide range of physical access and logical access applications
is most commonly used in finger scanning technology. The characteristics and patterns of all
fingerprints are unique. A normal fingerprint consists of spaces, lines. These lines are called
ridges, while valleys are called the spaces between the ridges. This provides a summary of the
techniques of recognition of the human or user face. And that applies a lot to the front faces,
there are advantages and disadvantages of each method are also given. The methods
considered are individual features and semantic networks, as well as the dynamic link
architecture used to verify the face of the person, hidden Markov model, matching
computational features, and matching disorder.

These approaches are analyzed in terms of the image processing facial representations they
used. Arunkumar, Vasanth Kumar, Naveenly King, Aravindan[1] have observed that the
growth in the electronic transactions has resulted in a greater demand for fast and accurate
user identification and authentication. Users have been largely depending on and trusting the
Automatic Teller Machine (ATM) to conveniently meeting their banking needs. The ATM
fraud has recently become more widespread. This system is used to avoid the ATM robberies
and wrong person misuses the ATM... A S Tolba [2] et al have proposed an overview of face
recognition and its applications. Divyarajsinh N. Parmar [3] remarks that, Face recognition
presents a challenging problem in the field of image analysis and computer vision. The
security of information is becoming very significant and difficult.

Face recognition is a biometric system used to identify or verify a person from a digital image.
The utilization is the no more issues if client overlooked there passwords and id codes
additionally, In truth of the biometrics is the innovation that deals with it. which is transforms
your body into the your passwords. The all the more requesting you make your own secret
key decision and development the guidelines done the more trouble to the clients can have in
recalling their passwords. However, severe and hard secret word rules are important to keep
away from simple programmer assaults on the system.
 CHAPTER 4
BIOMETRICS
Biometrics refers to the use of unique biological or behavioral characteristics to identify
individuals. It's an advanced technology that has gained immense popularity in various
sectors, primarily for authentication and identification purposes. Here's a detailed note on
biometrics:

Types of Biometric Characteristics:

1.Physiological Biometrics:

Fingerprint Recognition: Analyzes unique patterns on a person's fingertip.

Facial Recognition: Identifies individuals by analyzing facial features and structures.

Iris or Retina Recognition: Uses iris or retina patterns for identification.

Hand Geometry: Measures the shape and size of the hand for identification purposes.

DNA Matching: Compares genetic markers for identification (primarily forencic applic)..

2. Behavioral Biometrics:

Voice Recognition:Analyzes speech patterns, tone, and voice characteristics.

Keystroke Dynamics: Recognizes individuals by their typing patterns, including rhythm.

Gait Analysis: Identifies individuals by their walking style and patterns.

Working Principle:

Biometric systems capture and analyze these unique characteristics, converting them into
digital data. This data is then stored in a secure database or compared against previously
stored records to verify or authenticate an individual's identity.

Advantages of Biometrics:

1.Uniqueness: Biometric traits are highly distinctive to each individual, making them difficult
to forge or replicate.

2.Convenience: Eliminates the need to remember passwords or carry physical identification

documents.

3Accuracy:Advanced algorithms ensure precise identification, reducing false positives

4.Enhanced Security: Provides a robust layer of security in various applications, including
access control, banking, and government systems.

Applications of Biometrics:

1. Access Control and Security: Used in workplaces, government facilities, and homes to
control access to premises or systems.

2. Border Control and Immigration: Verifies identities at border crossings and airports to
enhance security and streamline processes.

3. Financial Services: Used in banking and payment systems to authenticate transactions and
prevent fraud.

4. Healthcare: Ensures secure access to medical records and controls access to sensitive areas
within healthcare facilities.

5. Law Enforcement and Forensics: Assists in criminal investigations by matching biometric

data with databases of known individuals.

Diagram
 CHAPTER 5
IMAGE PROCESSING

Image processing refers to the manipulation, analysis, and interpretation of digital images using
various algorithms and techniques. It involves altering an image in order to enhance its quality, extract
useful information, or perform certain tasks for specific applications.

Basic Operations in Image Processing:

1. Image Acquisition: Obtaining images through various devices such as cameras, scanners, or
sensors.
2. Image Enhancement: Improving the quality of an image by adjusting contrast, brightness, or
sharpness.
3. Image Restoration: Removing noise or distortions caused during image acquisition or
transmission.
4. Image Compression: Reducing the size of an image for efficient storage and transmission
without significant loss of quality.
5. Image Segmentation: Dividing an image into meaningful parts or regions to facilitate analysis.
6. Feature Extraction: Identifying and extracting important features or patterns from images.
7. Object Recognition: Recognizing and identifying specific objects or patterns within an image.

Image processing is any form of signal processing for which our inputs is an image,such as
photographs or frames of video and our
Output can be either an image or a set of characteristics or parameters related to the image .It is
defined as the use of computerized algorithms for the analysis of two dimensional images with
respect to an application.
Image processing is divided into three types:
Descretization and representation:
Converting visual information into a descrete form:suitable for computer processing to solve
storage space as well as time requirement in subsequent processing

Processing:
Improving image quality by filtering:compressing data to save storage and channel capacity during
transmission.
Analysis:
Extracting image features :qualifying shapes,interpretation and recognition
Image Processing Techniques:
Advanced image processing techniques such as facial recognition, object detection, and anomaly
detection can be used to enhance ATM security. These techniques can detect and prevent
fraudulent activities, ensuring the safety of ATM transactions.
 CHAPTER 6
METHODOLOGY
The two classes of biometric strategies are:- Physiology based procedures that measure the
physiological qualities of an individual. These incorporate unique mark confirmation, iris
examination, facial investigation, hand geometry vein designs, ear recognition, smell
discovery, and DNA design examination. The social procedures which measure the conduct of
an individual. These incorporate written by hand signature check and discourse examination.
No more issues if passwords and I d codes have been overlooked, biometrics is the innovation
that deals with it, making your body your secret word. Typically. To make your secret key
determination and development runs progressively thorough, the more troublesome it will be
for clients to recall their passwords. Sadly, To stop basic programmer assaults on the system,
severe secret key standards are required. Fig. shows the types of biometrics

1) Finger print scan:

The basic intention of fingerprint-recognition is to carry on with authentication using
fingerprint-impressions. This is mostly done by minutiae-features of fingerprint images.
And the general flowchart is shown in Fig.1.
2)Iris Scan
The iris has colored streaks and lines radiating from the eye's pupil. After DNA, the Iris provides the
most complete biometric data. The iris has information that is more unique than any other organ in the
body. The iris scan is safe than the fingerprint. ordinary camera can take a picture of users iris. Use
users iris picture can use for one code of that user. and that code user can use after for using the system.
In the iris scan capture a 240 points are scan. The iris is the hued ring of muscle that opens and closes
the student of the eye like a camera screen. The shaded example of our irises is resolved hereditarily
when we're in the belly however not full fledged until we're matured around two. It originates from a
color called melanin— more melanin gives you browner eyes and less creates bluer eyes. In spite of the
fact that we talk about individuals having "blue eyes," "green eyes," "darker eyes," or whatever, in all
actuality the shading and example of individuals' eyes is incredibly mind boggling and totally
interesting: the examples of one individual's two eyes are very unique in relation to one another and
even hereditarily indistinguishable twins have distinctive iris design.

3) Face Scan
FRS is an application that mechanically identifies a person from a digital image or a video outline from
a video source. One of the processes to do this method is by matching chosen facial features from a
facial database and the image. Like all biometrics arrangements, face acknowledgment innovation
measures and matches the interesting attributes for the reasons for distinguishing proof or validation.
Regularly utilizing an advanced or associated camera, facial acknowledgment programming can
identify faces in pictures, evaluate their highlights, and afterward coordinate them against put away
layouts in a database. Face examining biometric tech is unbelievably adaptable and this is reflected in
its wide scope of potential applications.
4)Optical scanning
System can take the scan of user palm. then this print can register in the database. when user can use
this technique again then scan the hand then match with database scan. If they are match then user can
use the system. this type is not much secure. any other fraud user can use this. this fraud user can take
users fingerprint easily and use it.
5) Capacitor scanning
Capacitor sensor use in this scanning. breadth if this sensor is 1mm(micro meter) or less than
1mm.using this feature the capacitor can take print of users fingerprint.in that the use scan only limited
points. When this process is start then some capacitor are activate and some are not activated. Then
take and print of users fingerprint. Capacitor can search only 45 point and catch it.
 CHAPTER 7
FACIAL RECOGNITION

Facial recognition technology can be used to identify criminals who attempt to use stolen cards or
commit other crimes at ATMs. The system captures images of the person's face and compares them to

a database of known criminals. If a match is found, an alert is sent to the bank' security team. Facial
recognition is a way of identifying a human face through technology known as biometrics, oftentimes mapping
facial features from a photograph or video and then comparing the information with a database of known faces to
find a match.

Like all biometrics arrangements, face acknowledgment innovation measures and matches the interesting
attributes for the reasons for distinguishing proof or validation. Regularly utilizing an advanced or associated
camera, facial acknowledgment programming can identify faces in pictures, evaluate their highlights, and
afterward coordinate them against put away layouts in a database. Face examining biometric tech is unbelievably
adaptable and this is reflected in its wide scope of potential applications. Face biometrics can possibly be
incorporated anyplace you can locate an advanced. camera. Law requirement offices the world over use
biometric programming to examine faces in CCTV film, just as to recognize people of enthusiasm for the field.
Outskirt control arrangements use face acknowledgment to check the personalities of explorers. It even has
purchaser applications. We are additionally observing face biometrics in the advanced world, with Facebook,
Shutter Stock, and other social stages that look to sort out amazing measures of rich picture information by
distinguishing the general population caught in them. Facial acknowledgment doesn't simply manage hard
characters, yet in addition can assemble statistic information on groups. This has made face biometrics
arrangements much looked for after in the retail showcasing industry. Human facial structure is an individual
trademark. Facial acknowledgment biometrics utilizes this reality to recognize and verify people. Human minds
have common capacity to recollect and recognize various countenances. We distinguish and confirm individuals
just by perceiving their face once a day. We perceive our family, companions, partners, and pets principally by
their facial structure. Facial acknowledgment framework can recognize individuals by handling their advanced
pictures if their facial acknowledgment personality has been pre-built up. The framework exploits advanced
pictures or still casings from a video source, which are taken through the facial acknowledgment calculation.
This calculation extricates information out of facial qualities like position and state of eyes, nose, cheekbones
and jaw. It can likewise gauge remove between these qualities and mapped information is put away in a database
pictures in a second, contrast them and what is put away in the database and delivered results
Facial recognition technology has found applications across various industries and sectors due to its
ability to identify and verify individuals based on their facial features. Here are some notable
applications:

1. Security and Surveillance:

Access Control:
 Physical Access: Used in offices, airports, and high-security facilities to control entry based on
facial recognition.
 Device Authentication:Securely unlocks smartphones, laptops, and other devices using facial
recognition.

Public Safety:
 Law Enforcement:Assists in identifying suspects, locating missing persons, and monitoring
public spaces for potential threats.
 Border Control: Enhances border security by verifying travelers' identities at immigration
checkpoints.

2. Banking and Finance:

 ATM Authentication: Allows secure access to ATM services by verifying account holders'
identities through facial recognition.
 Transaction Security: Verifies customers during financial transactions for enhanced security and
fraud prevention.

3. Retail and Marketing:

 Personalized Shopping Experience: Analyzes facial expressions to gauge customer satisfaction

and preferences in retail stores.
 Targeted Advertising: Tailors ads based on demographic information gathered through facial
recognition in digital signage.

4. Healthcare:

 Patient Identification:Ensures accurate patient identification in hospitals and clinics, reducing

medical errors and enhancing records management.
 Monitoring Health Conditions: Analyzes facial cues for early detection of health issues like
pain, stress, or neurological disorders.

5. Automotive Industry:

 Vehicle Access and Security: Allows entry to vehicles and provides personalized settings based
on recognized driver profiles.
 Driver Monitoring: Monitors driver attentiveness and alerts in case of drowsiness or distraction.

6. Education:

 Campus Security: Controls access to school premises and enhances safety measures by
identifying students and staff.
 Attendance Monitoring:Automates attendance tracking in classrooms or educational
institutions.
7. Entertainment and Personal Devices:
 Social Media and Photo Tagging: Recognizes individuals in photos and assists in social media
tagging.
 Gaming: Provides interactive experiences by incorporating facial recognition into gaming
consoles and applications.

8. Humanitarian Aid and Identity Management:

 Aid Distribution: Helps in managing aid distribution by accurately identifying recipients,

reducing fraud and ensuring fair distribution.
 Identity Verification: Verifies identities for government services, voting systems, and refugee
camps.

Facial recognition technology, while offering numerous benefits, also raises concerns about privacy,
data security, and potential misuse. Striking a balance between utility and safeguarding individual
rights remains a significant consideration in its widespread adoption.
 CHAPTER 8
IMPLEMENTATION

ATM security can be enhanced by using image processing:

• Initially the customer’s picture is taken when the account is opened and the user is allowed to set
nonverified transaction limits.
• At ATM, access card and PIN are used to per verify the user.
• User’s snap is taken and an attempt is made to match it to the record image.
• If the match procedure becomes successful, allow the transaction.
• If the match is unsuccessful, limit the available transactions.
• When a match is complete with the PIN but not the imagery, the bank could limit the transactions in
a way contracted by the user when the account was opened, and could store the photograph of the user
for later examination by the bank official. In the case of using credit card at ATMs, confirmation
system would not presently be feasible without revamping the entire credit card issuing industry.
 CHAPTER 9
ADVANTAGES

As technology continues to evolve, the integration of image processing and facial recognition in
securing Automated Teller Machines (ATMs) has emerged as a cutting-edge solution. This
convergence not only enhances security measures but also introduces several advantages that
redefine the landscape of financial transactions.

 Enhanced Authentication:
• Unparalleled Accuracy: Facial recognition technology provides a highly accurate means
of authentication, reducing the risk of unauthorized access. This ensures that only
authorized individuals can perform transactions, significantly enhancing the security of
ATM operations.
 Increased Security Against Skimming and Card Fraud:
• Mitigation of Skimming: Facial recognition, when coupled with image processing, adds
an extra layer of security against skimming devices. Even if a fraudster possesses a
copied card, they would still need to pass the facial recognition check, making
unauthorized transactions more challenging.
 Reduced Incidents of Identity Theft:
• Unique Biometric Identifier: Facial features are unique to each individual, providing a
robust biometric identifier. This reduces the likelihood of identity theft compared to
traditional methods relying solely on cards and PINs.

 Real-time Monitoring and Alerts:

• Immediate Response to Suspicious Activity: Image processing enables real-time
monitoring of ATM users. Facial recognition algorithms can identify and alert security
personnel to suspicious behavior, such as multiple unsuccessful attempts or the presence
of unauthorized individuals.

 Prevention of Card Trapping and Cash Trapping:

• Immediate Intervention: Image processing can detect anomalies in the card insertion
and cash dispensing processes. In the case of card or cash trapping attempts, the system
can trigger alerts and temporarily disable the ATM, preventing successful fraud.

 User Convenience and Speed:

• Faster Transactions:Facial recognition streamlines the authentication process,
eliminating the need for users to input PINs or insert cards. This not only enhances user
convenience but also expedites transaction times, contributing to a more efficient
banking experience.

 Adaptability to Multimodal Security:

• Integration with Other Biometrics: Facial recognition can seamlessly integrate with
other biometric technologies, such as fingerprint or iris recognition, creating a
multimodal security system that is more resilient against fraudulent attempts.

 Deterrent for Criminal Activities:

• Visible Security Measures:The implementation of facial recognition technology acts as a
visible deterrent to potential criminals. The awareness of advanced security measures
can discourage fraudulent activities around ATMs, contributing to overall crime
prevention.
 CHAPTER 10
DISADVANTAGES

While image processing in secure ATMs offers enhanced security by verifying user identity through
techniques like facial or fingerprint recognition, it does have some disadvantages:

1. Cost: Implementing advanced image processing technologies can be expensive, impacting the
overall cost of ATMs. Maintenance and upgrades further contribute to ongoing expenses.

2. Complexity: Sophisticated image processing systems might introduce complexity in user

interactions, potentially leading to longer transaction times or user confusion.

3. Technical Issues:Glitches in image processing, such as misreads or errors in facial/fingerprint

recognition, could prevent users from accessing their accounts, causing inconvenience.

4. Privacy Concerns: Storing and processing biometric data raises privacy concerns. There’s
always a risk of data breaches or misuse, compromising users’ sensitive information.

5. Accessibility: Some users might face difficulties using biometric authentication due to
physical disabilities or technical limitations, creating barriers to accessing ATM services.

6. Reliability: Environmental factors like lighting conditions or the condition of the fingerprint
reader or camera may affect the reliability of image-based authentication, leading to
authentication failures.

Balancing enhanced security with user convenience and ensuring robustness against technical issues
and privacy concerns remains a challenge in implementing image processing in secure ATMs.
 CHAPTER 11
CONCLUSION

In conclusion, while image processing in secure ATMs offers heightened security through biometric
authentication and identity verification, its implementation presents challenges such as cost, technical
complexities, privacy concerns, accessibility issues, and reliability issues. Striking a balance between
security enhancements and addressing these drawbacks is crucial to ensure the effectiveness and
user-friendliness of secure ATM systems utilizing image processing technologies. Continued
advancements and careful considerations in implementation are essential to maximize security while
minimizing inconvenience and risks for users.
Access card / PIN provide insufficient ATM security. The integration of facial verification into the ATM
process stands as a powerful measure against fraudulent transactions. Present-day ATMs possess the
capability to conduct this verification locally, requiring only a software update for implementation. Moreover,
our team has pioneered a rapid fingerprint enhancement algorithm designed to dynamically enhance the
clarity of ridge and furrow structures within fingerprints. Leveraging estimated local ridge orientation and
frequency, this algorithm significantly bolsters the fingerprint recognition process. Fingerprint readers,
extensively used by banks for ATM authorization, are increasingly prevalent in grocery stores, where they
facilitate automatic recognition of registered customers for billing purposes. The enhancement algorithm
operates using minutiae goodness indices, significantly elevating input fingerprint verification accuracy.
Furthermore, the algorithm adeptly identifies and excludes unrecoverable corrupted fingerprint regions from
further processing, contributing to the overall efficacy of the fingerprint-based verification system.
 CHAPTER 12
REFERENCE

[1] Aru“ATM Security using Face Recognition”, International

Journal of Current Engineering And Scientific
Research- Volume- 5, Issue-4, 2018
.
[2] A S Tolba, A.H. El-Baz, and A.A. El-Harby, “Face
Recognition: A Literature Review”, International
Journal of Signal Processing, Volume 2 Number 2,
2014.

[3] Divyarajsinh N.Parmar1, Brijesh B. Mehta2, “Face

Recognition Methods & Applications”, International
Journal of Computer Applications in Technology,
January 2014.

[4] KresimirDelac, Sonja Grgic&MislavGrgic, “Image

Compression in Face Recognition” - A Literature
Survey, October 2008.

[5] Mourad Moussa, MahaHmila& Ali Douik, “A Novel

Face Recognition Approach Based on Genetic Algorithm
Optimization”, https://doi.org/10.24846/v27i1y201813.

[6] Priyanka, Research Scholar, Department of Computer

Science and Engineering, “A Study on Facial Feature
Extraction and Facial Recognition Approaches”,
International Journal of Computer Science and Mobile
Computing, May 2015.

[7] T.Suganya, T.Nithya,C.Sunita, B.Meena& Preethi,

“Securing ATM by Image Processing”–
FacialRecognition Authentication, International Journal
of Scientific Research Engineering & Technology

[8] J. G. Daughman, "How iris recognition works," IEEE

Transactions onCircuits and Systems for Video
Technology, vol. 14, no. 1, 2004nkumar, Vasanth Kumar, Naveenly King, Aravindan,
.