See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/371567274

Biometric Authentication

Research · December 2022

CITATION READS

1 2,685

2 authors:

Uran Lajçi Elvir Misini

University of Prishtina University of Prishtina
8 PUBLICATIONS 1 CITATION 6 PUBLICATIONS 1 CITATION

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Uran Lajçi on 14 June 2023.

The user has requested enhancement of the downloaded file.

 Biometric Authentication
1st Elvir Misini 2nd Uran Lajçi
Department of Computer and Software Department of Computer and Software
Engineering Engineering
University of Prishtina University of Prishtina
Pristina, Kosovo Pristina, Kosovo
elvir.misini@student.uni-pr.edu uran.lajci@student.uni-pr.edu

Abstract — Biometric authentication is a way of identifying II. BIOMETRICS

and authenticating users. In this paper, we explain biometric
authentication and its use in identifying users. Biometric A. Definition and use
authentication uses unique physical or behavioral Initially, by biometrics we mean the branch that deals with
characteristics, such as fingerprints, patterns of a person's face, the use of the characteristics of the human body to identify a
iris patterns, and voice recognition to grant access to devices. person, e.g. if a certain crime is committed and there are
The paper discusses the advantages of biometric authentication
fingerprints, then the main suspect is the person to whom those
over traditional methods, such as passwords, PINs or smart
cards, as biometric traits are difficult to replicate or steal. The
fingerprints belong.
paper also examines the various types of biometric Fingerprints represent a unique human biometric that is
authentication systems and their relative strengths and being used a lot today in information security. In recent years
weaknesses. The paper concludes by emphasizing the potential it has been used a lot in smartphones, personal tablets, and also
of biometric authentication as a secure and convenient method laptops thus enabling the opening of the device. One of the
of identification in various applications. reasons why biometrics are used for such purposes is that,
apart from being significantly more secure than passwords or
Keywords—Biometric authentication, physical characteristics,
behavioral characteristics, identifying users, authenticating users
PINs, they also save more time, and, above all, they are not
something that can be forgotten since they are characteristics
I. INTRODUCTION of the body. But in addition to the advantages that biometrics
have, they also have some disadvantages, where if we take a
Every day technology grows and evolves, and with this very simple example of fingerprints, what happens if we lose
growth, we have an increase in technology users. Users and a finger (e.g. in an accident) or even suffer a fingerprint burn?
the technology that they use produce enormous collections of Well, the answer is very simple, we cannot open the phone in
data, and some of this data for whatever reason, we do not this case, assuming of course that the phone opens with the
want to make it public. Here comes the question of how we finger in which we have suffered an injury.
will secure this data one hundred percent. This question is
difficult to answer since it is theoretically not possible to be B. Specifications
one hundred percent safe. However, to secure information, Biometrics are divided into physiological and behavioral.
most of us have used and still use passwords, PINs, or smart Common biometrics that is used for authentication and
cards which represent what we know and what we have in identification are:
authentication systems. But we all know that some of these are
breakable for example passwords - people usually associate a Physiological
password with some personal event when they set a password • Fingerprint recognition
and some can be stolen - a smart card. Knowing that we cannot • Hand geometry
rely on passwords and classic forms of information security, • Iris recognition
then cybersecurity has decided to go one step further and • Facial recognition
cooperate with the human body and behavior also known as • Signature recognition
biometrics which represents what we are in authentication • Signature recognition
systems. When biometrics are mentioned, we immediately • Vein recognition
think of human eyes, faces, ears, palms of hands, and • Retina recognition
fingerprints. These are the parts of our body that we focus on • DNA recognition
since they are unique in different persons. What is also unique
is our behavior, like the way we walk, our voice, and our Behavioral
keystroke. It is already proven that every person has unique
fingerprints, unique IRIS, forms of ears, faces, voices, and • Typing rhythm
signatures. The uniqueness of these parts of our body and • Gait analysis
behavior is the reason why in the last decades the technology • Mouse dynamics
industry is trying to use biometric authentication, for example, • Keyboard dynamics
the smartphone industry has focused on opening the phone • Voice recognition
using fingerprints and face recognition. With all the
advantages that biometric authentication and identification This list is not exhaustive, and new biometric modalities
give for example, we do not need to remember long and are constantly being developed and researched.
complicated passwords and carry smart cards, it has also
An ideal biometric system would posses several key
disadvantages if we do not implement them in the right way
characteristics [1]:
or in the right place.
1. Universality: The biometric should be applicable to
all individuals, regardless of age, gender, race or
other demographic factors.

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

 2. Uniqueness: The biometric should be able to
distinguish one individual from all other, with a low
rate of false matches.
3. Permanence: The biometric should remain stable
over time, so that it can be used for identification and
authentication in the long-term.
4. Collectabillity: The biometric should be easly
collectable and measurable using non-intrusive
methods.
5. Performance: The biometric should have a high level Table 2. Strength comparison of biometrics and passwords [3]
of accuract and reliability, with low rastes of false
matches and false non-matches. Passwords are disliked by many people (users and experts)
6. Acceptability: The biometric should be socially and but have proven to be a strong contender. Biometric schemes
culturally acceptable, with a low level of perceived are not the only systems that have failed to supplant them.
instrusiveness or privacy invasion. None of the proposed alternative authentication mechanisms
7. Circumvention: The biometric should be difficult to can match the security evaluation statistics that traditional
circumvent or forge, ensuring the system’s security. passwords already provide. Despite countless efforts to
8. Interoperability: The biometric should be able to supplant them, their continued ubiquity speaks volumes about
work with a variety of different systems and devices, their credibility and accomplishments. In particular,
allowing for east intefration and use. passwords are not the problem: rather it is the way we interact
with them. Like them or loathe them, passwords are here to
It is worth noting that currently no single biometric system stay for the foreseeable future, where the objective is not
possesses all these characteristics perfectly, and it is also impenetrable security but reducing harm at an acceptable cost:
possible that in the future new biometric modalities will be on the other hand, biometrics has major potential for areas
developed and will be considered more ideal. such as law enforcement, identity management, airport
Bellow, we can see a table that compares different types security, counter-terrorism, child recovery, healthcare, etc. In
of biometrics. these and similar cases, biometrics achieve implicit
credibility. However, privacy will always remain the key
concern for these systems’ implementation, besides possibly
empowering inappropriate approaches of stigmatization and
exclusion (eg, racism or ageism) using ancillary biometric
information [3].
III. SECURITY
Table 1. Comparison of biometric technologies based on We know that despite the greater security of biometrics,
perceptions of three biometric experts [2] these are also falsifiable and someone else can be identified
instead of someone else. If we take some examples e.g.
In this table, we can see the differences between these Fingerprints can be easily taken from a glass that we drank
types of biometrics, where it can be seen that fingerprint, iris, water and identified as us, our face is also very easy to take,
and retinal scan have a high degree of uniqueness, while in where only a qualitative picture of us is enough to be
this case, faceID has similarities (so two faces can be very identified, the retina also by means of a picture can be
similar), we also have comparisons in other properties such as provided, so as we can see that even by these metrics we are
performance, acceptability, etc. down. In addition to fakes, we also have cases when the
C. Biometrics vs passwords system itself gives access to someone else, if we take an
example of facial identification on a fake phone, there are
Passwords have been in use since the legend of Ali Baba. cases that guarantee someone else access due to the similarity
However, we have long been told that passwords are of the face with the owner of the phone, but this happens very
outdated, inconvenient, and a security risk. Whenever news rarely, since biometrics have a property of updating, self-
on a password security breach circulates, some commentator, learning or evolution, thus guaranteeing access only to the
vendor, or expert comes forward and says ‘biometrics is the authorized person, because if there was no evolution, then
solution’. Historically, biometric solutions have always been only with the aging of the owner, the owner of the phone
launched with the claim of achieving the highest security, would not have the opportunity to open the device.
while traditional passwords were known for their
shortcomings. Unfortunately, security attackers do not
simply give up their self-enriching efforts and go home
defeated [3].
 speak, for example, which vowels we pronounce more or
similar characteristics.

These systems use AI and ML in order to learn our

behaviors continuously since they can change over time, so
they are not like a password that has a certain value, and they
are also not like physical characteristics that have a certain
pattern that does not change. Also, a nice advantage of these
systems is that we are continuously in the process of
authentication, so even after authentication where we have
gained access somewhere, we may be denied access later,
because e.g. we pass the authentication with facial
Figure 1. Forgery of fingerprints [4] recognition on the phone, and someone else can freely use my
phone since there is only one authentication at the beginning,
The biometric system must verify liveness, otherwise, the
while through behaviors if someone else picks up our phone,
system can be fooled with duplicate biometric features.
based on their behaviors the system understands that is not
Sometimes the vitality test can be replaced by a
guard(human), however, it is debatable whether a guard can authorized to use that device. But of course, despite the
protect against more advanced biometric forgeries (like a thin advantages that this type of authentication has, it also has its
layer of silicon on fingerprints) [5]. own disadvantages, because in the event of an accident, our
walk is not the same, nor are our behaviors, also in the event
A. Applications of an illness, authentication through speech will not work. ,
Some of the applications where biometrics are used are or even in case of the influence of any narcotic substance or
[6]: alcohol, the authentication will also not work.
 Computer/Network Security
General Scenario: Human gesture/gait has been used as a
 Online transactions modern method to identify/authenticate individuals' identities
 Security of the physical area [7]. Due to the increased demand for employing an efficient
 Bank(ATM) technology that provides a high level of security, gait
 VOTING recognition received significant attention from various
 Prisons security research communities. It is an emerging technology
that showed an acceptable performance in video surveillance.
Then taking into account the digitization of almost every Fig.3, shows the basic mechanism of the gait recognition
service that we have developed in the traditional way system.
(physically waiting for an order for a service), now most of
them are done online, therefore transactions are also requiring
a biometric authentication to verify identity, or Banks are
deciding even fingerprint identifiers in ATMs or facial
recognition, or a very targeted application would be electronic
voting, which is going more and more every day around such
a vote, also in prisons by identifying visitors who enter.
B. Behavioral authentication
So far we have only talked about biometrics such as
fingerprints, facial recognition, retinal scanning, and so on,
but these are physical characteristics of the human body that
already exist. Another biometric is the authentication through
the behavior of the individual, which means that we do not
need to perform the authentication by taking any action, such
as giving a password or scanning the finger but only want to
access somewhere that we have performed the authentication.
Although it seems a bit strange at first, it turns out to be very Figure 2. Block diagram of gait recognition system
successful. Even this method of authentication has already
Identifying/recognizing a person based on his/her style of
come into use, such as in institutions with high security. This walking is one of the behavioral biometrics techniques since
authentication can be developed in several ways, some of it depends on the analysis of body structures or body
which we will mention below, e.g. the way an individual movements. As a general idea, limb movements in any living
walks can be a type of password that can be detected through organism refer to the ability of walking which is also known
the sensor of smartphones, knowing that smartphones have among human beings as human locomotion. In order to
this type of sensor that is used to track our activity during the provide a unique identity, an extraction of body movements
day, another authentication can be even the way we type on can be obtained from a repeated style of walking (gait cycle)
our computer keyboard, i.e. the location of our fingers, the that comes from balanced cooperation among a human legs,
speed of writing, etc. Another authentication is done by body, and arms.
moving the cursor, i.e. the way we move it, also the way we
grab the phone and hold it, another way can be the way we
C. Ear Recognition what he/she has written so far. He/she has to look at the
Another biometric authentication technique is conducted computer monitor to see the signature. This is a considerable
based on the recognition of the unique shape and appearance drawback for many (inexperienced) users. Some special pens
of the human being's ear. Naturally, a person is born with a work like normal pens, they have ink cartridges inside and
visual shape of his/her ears. However, the human ear is not can be used to write with them on paper. The accuracy of the
subject to change while a person’s growth and even aging. It signature dynamics biometric systems is not high, the
has dependable stability which increases its level of security crossover rate published by manufacturers is around 2%, but
as a proposed method for the security according to our own experience, the accuracy is much worse
identification/verification of individuals [8]. [10].

D. Retina Geometry Technology F. DNA Analysis

It is based on the blood vessel pattern in the retina of the This method of verification is mostly used in criminal
eye as the blood vessels at the back of the eye have a unique cases. DNA of the user in the form of blood, tissue, hair, and
pattern, from eye to eye and person to person (figure 4). The nails is collected for confirmation. DNA analysis takes time.
retina is not directly visible and so a coherent infrared light DNA also is a unique characteristic but a hair or nail can be
source is necessary to illuminate the retina. The infrared stolen [9].
energy is absorbed faster by blood vessels in the retina than
by the surrounding tissue. The image of the retina blood First of all, DNA differs from standard biometrics in
vessel pattern is then analyzed. several ways:
 DNA matching does not employ templates or
feature extraction but rather represents the
comparison of actual samples.
 DNA requires a tangible physical sample as opposed
to an impression, image, or recording.
 DNA matching is not done in real-time, and
currently, not all stages of comparison are
automated.

Regardless of these basic differences, DNA is a type of

biometric inasmuch as it is the use of a physiological
characteristic to verify or determine identity. DNA testing is a
technology with a high degree of accuracy however, the
Figure 3. Image of retina [9]
possibility of sampling contamination and degradation will
pose an impact on the accuracy of the method [10].
Retina scans require that the person removes their glasses, IV. RECENT ADVANCES IN THE USE OF BIOMETRIC
places their eye close to the scanner, stares at a specific point,
TECHNOLOGY IN DEVICES
remain still, and focus on a specified location for
approximately 10 to 15 seconds while the scan is completed. Recent advances in the use of biometric technology
A retinal scan involves the use of a low-intensity coherent [11] in devices include:
light source, which is projected onto the retina to illuminate  Smartphones: Biometric sensors such as fingerprint
the blood vessels which are then photographed and analyzed. scanners, facial recognition cameras, and iris
A coupler is used to read the blood vessel patterns. A retina scanners are becoming increasingly common in
scan cannot be faked as it is currently impossible to forge a smartphones. These sensors are used to unlock the
human retina. Furthermore, the retina of a deceased person device and authenticate mobile payments.
decays too rapidly to be used to deceive a retinal scan. A  Laptops and tablets: Laptops and tablets are also
retinal scan has an error rate of 1 in 10,000,000, compared to
adopting biometric technology, such as fingerprint
fingerprint identification errors sometimes as high as 1 in 500
scanners, facial recognition cameras, and iris
[9].
scanners, to enhance security and convenience for
E. Signature dynamics users.
The signature dynamics recognition is based on the  Smart home devices: Smart home devices, such as
dynamics of making the signature, rather than a direct door locks and security cameras, are using biometric
comparison of the signature itself afterward. The dynamics technology to enable secure and convenient access
are measured as a means of the pressure, direction, for authorized users.
acceleration, and length of the strokes, dynamics number of  Internet of Things (IoT) devices: Biometric
strokes, and their duration. The most obvious and important technology is being integrated into IoT devices to
advantage of this is that a fraudster cannot glean any improve security and personalization. For example,
information on how to write the signature by simply looking biometric sensors in smartwatches can be used to
at one that has been previously written. Special pens are able authenticate mobile payments and track fitness.
to capture movements in all 3 dimensions. Tablets have two  Automotive: Biometric technology is being used to
significant disadvantages. First, the resulting digitalized enhance the security of connected cars. Some
signature looks different from the usual user signature. And carmakers have started to use fingerprints and facial
input devices second, while signing the user does not see recognition technology to unlock and start the car,
 and to personalize settings such as seat and mirror with Windows 10 and Windows 11, whether they're
positions. external or integrated into laptops or USB keyboards.
 Banking: Biometric technology is being  Iris Recognition. This type of biometric recognition
increasingly used in the banking sector, such as uses cameras to perform scans of your iris. HoloLens
fingerprint and facial recognition to authenticate 2 is the first Microsoft device to introduce an Iris
transactions, along with other biometric scanner. These iris scanners are the same across all
identification methods such as voice and iris HoloLens 2 devices.
recognition. V. RECENT DEVELOPMENTS IN BIOMETRIC AUTHENTICATION
Overall, the use of biometric technology in devices is Recent developments in biometric authentication
becoming more prevalent as it improves the security and user include:
experience, and the technology is becoming more affordable  Face recognition: Advancements in deep learning
and accessible. and artificial intelligence have led to significant
improvements in facial recognition technology. This
has made it possible to identify individuals from a
distance, in poor lighting, or with masks on [15].
 Behavioral biometrics: Behavioral biometrics is the
process of identifying individuals based on their
unique behavior, such as keystroke dynamics,
mouse movement, and gait analysis. These methods
are becoming increasingly popular as they can be
used to authenticate users without the need for
Table 3. Biometrics in commercial devices [12] physical contact [16].
 Multi-modal biometrics: Multi-modal biometrics
The iPhone 13 includes the latest version of Apple's Face involves the use of multiple biometric traits, such as
ID technology, which uses facial recognition to unlock the fingerprint and facial recognition, to identify an
device and authenticate user actions. This technology uses a individual. This approach is more secure than using
TrueDepth camera system that projects and analyzes over a single biometric trait and can also improve the user
30,000 invisible dots to create a precise depth map of the user's
experience [17].
face, which is then used to authenticate the user. The Face ID
system is designed to be fast, secure, and easy to use, and it  Biometric liveness detection: Liveness detection is
can be used to unlock the device, make purchases, and access used to ensure that the biometric data being used is
password-protected apps. from a live person and not a fake or spoofed
biometric [18].
Additionally, the iPhone 13 also includes Touch ID, which  Fusion of biometrics with other authentication
is a fingerprint recognition system that allows users to unlock methods: Biometric authentication is increasingly
their devices and authenticate purchases by placing their being combined with other authentication methods,
fingers on the home button. The Touch ID system is also fast,
such as knowledge-based authentication (KBA), to
secure, and easy to use, and it provides an alternative means
create more secure and convenient authentication
of authentication for users who prefer not to use facial
recognition. systems [19].
 Biometric on Edge: Biometric authentication on
It's important to note that, while iPhone 13 includes both Face edge devices such as smartphones, tablets, and
ID and Touch ID, the biometric features may vary by region, laptops is becoming more common. The use of
carrier, or model of the device. [13] edge-based biometrics reduces the need for a
Windows Hello for Business replaces passwords with centralized database and helps to protect users'
strong two-factor authentication on devices. This privacy [20].
authentication consists of a type of user credential that is tied
to a device and uses a biometric or PIN [14].
VI. FUTURE WORK
 Facial recognition. This type of biometric
Considering that authentication based on biometrics
recognition uses special cameras that see in IR light,
which allows them to reliably tell the difference is developing every day and more, of course, it will advance
between a photograph or scan and a living person. even more in the future, thus making authentication and the
Several vendors are shipping external cameras that security of sensitive information even more secure.
incorporate this technology, and major laptop
manufacturers are incorporating it into their devices, Some of the future trends for biometric authentication
as well. include the following [21]:
 Fingerprint recognition. This type of biometric  Physical Identity Verification: The expansion of AI-
recognition uses a capacitive fingerprint sensor to driven biometrics, particularly facial and behavioral
scan your fingerprint. Fingerprint readers have been biometrics (gait, voice, and accent recognition), will
available for Windows computers for years, but the fuel new forms of real-time biometric identity
current generation of sensors is more reliable and less verification through on-premises cameras.
error-prone. Most existing fingerprint readers work
  Advanced Biometric Authentication: While there biometrics in the industry, hackers also began to find new
are reliable forms of biometrics and biometric ways of attacking and breaking them, where in fact sometimes
passwords in the market now, new technologies are it seems very simple to break them. Biometrics were also
focusing on drawing even more advanced biometric categorized into two categories, such as physical biometrics
markers from the body, each providing another and non-physical (behavioral) biometrics, as one of the safe
hard-to-fake marker to use for secure authentication. and unique physical biometrics that provides authentication
These can include odor recognition, heartbeat only for an individual was fingerprinted, but even this began
to be forged by stealing from us the knees simply from a glass
pattern recognition, hand geometry, and DNA
of water we drank. Then also comes into play the scanning of
signature reading.
the retina, the face, then from the non-physical biometrics,
 Identity Proofing: One of the strengths of biometric walking, the way of speaking, behavior, etc. But with
authentication is the assumption that the user must biometrics, a new problem arose, since people often hate being
be physically present to provide biometric data. tracked in any way, so why give their fingerprints, face, or any
Even stronger measures, like live identity physical characteristic of their body or behavior? But on the
verification or AI-driven video identity verification, other hand, of course, these features are more secure than
layer additional “liveness” testing into the processes passwords. In this paper, we have tried to explain what
to thwart emerging forms of identity fraud. biometrics are, tell about their types, which are more suitable
 Continuous Authentication: Authentication usually for use, which require fewer resources, how safe they are, how
happens once, during login, or multiple times based they can be stolen or forged, and how they can fail. their
on user access to different resources. Continuous functionality, their application, and the future of this type of
authentication uses behavioral patterns or other authentication.
markers to maintain authentication to guarantee From the knowledge and study we have developed, we
continued user verification over time periodically. think that no matter how secure biometrics are, they also have
their weaknesses and sometimes their non-functionality,
Regarding the issues where biometric authentication will where attackers will always use them, but of course, our duty
be applied, it is likely that the use of this technology will is to let's make this process as difficult as possible, therefore
increase initially in banks, electronic voting in countries for the most secure authentication and the security of the
where such voting is allowed, and high-security institutions private data of an individual, business, or institution, of
where only persons must enter authorized, it can also be course, we must have authentication with two factors or multi-
applied to educational or similar institutions (driver's factor where these factors can there are combinations of
license), in cases of testing, and also not to forget the airports, passwords and biometrics or combinations of different
since only airports use such technologies. biometrics. In this way, it will be a much more secure
authentication in the future.

VII. CONCLUSIONS REFERENCES

The security of personal or sensitive information has a
[1] J.D. Ross, A. Jain, and R. Bolle, "The ideal biometric trait," IEEE
much earlier history than today, or better we can say in the last Transactions on Systems, Man, and Cybernetics, Part C: Applications
two centuries, where often even short but more important and Reviews, vol. 36, no. 3, pp. 303-316, May 2006.
information has given another direction to the history of [2] A. Jain, L. Hong, and S. Pankanti, "Biometric identification,"
humanity and wars differently. This is also the main reason Commun. ACM, vol. 43, no. 2, pp. 90-98, 2000.
that today one of the main focuses of technology is that only [3] K. Siddique, Z. Akhtar, and Y. Kim, "Biometrics vs passwords: a
authorized persons have access to this information and that no modern version of the tortoise and the hare," Computer Fraud &
one else has the opportunity to provide access in any other Security, pp. 13-17, 2017.
way. Therefore, through authentication, we must identify [4] M. Espinoza, Ch. Champod, and P. Margot, "Vulnerabilities of
ourselves as the one who has access to that information, as we fingerprint reader to fake fingerprints attacks," 2011.
know that until recently and even now, authentication was [5] V. Matyas and Z. Riha, "Security of biometric authentication systems,"
2010.
initially done with a password, but the password presented a
[6] V. Pujari, R. Patil, and Sh. Sutar, "Research paper on biometrics
problem to society since people usually create the password security," 2021.
they enter something related to them, for example,
[7] I. Alsaadi, "Study on most popular behavioral biometrics, advantages,
NameSuraname1980, something that is very predictable from disadvantages and recent applications: a review," 2021.
the attackers to find that password, another problem is that [8] I. Alsaadi, "Physiological biometric authentication systems,
often out of fear of forgetting the password they also write it advantages, disadvantages, and future development: a review,"
on a piece of paper, thus exposing the password to a second International Journal of Scientific & Technology Research, vol. 4,
person. Although passwords have recently started to become 2015.
more complex and also two-factor authentication, reducing [9] K. Gupta, "Review paper on biometric authentication," International
the possibility of attacks by hackers, for example, a typical Journal of Engineering Research & Technology (IJERT) VIMPACT –
2017 (Vol. 5, Issue 23).
password today contains at least 8 characters, one uppercase
[10] S. Bhable, S. Kayte, R. Maher, J. Kayte, and C. Kayte, "DNA
letter, one lowercase letter, a number, and at least one symbol. biometrics," 2015.
To avoid passwords and any type of password attacks or even
[11] N.K. Ratha, J.H. Connell, and R.M. Bolle, "Recent advances in
the monotony of writing them, we have reached biometrics as biometrics: a survey," IEEE Transactions on Systems, Man, and
a much safer form of authentication, where every day we are Cybernetics, Part C: Applications and Reviews, vol. 36, no. 3, pp. 303-
avoiding passwords or PINs by replacing them with one of the 316, May 2006.
biometrics that is being offered to us. Well, is this also 100% [12] S.K. Sharma, K.N. Plataniotis, and A.N. Venestsanopoulos, "Recent
safe, of course not! Immediately with the introduction of developments in biometric on edge: a review," IEEE Transactions on
 Systems, Man, and Cybernetics: Systems, vol. 46, no. 1, pp. 3-17, [18] Xing Liu, Qiang Ji, and Anil K. Jain, "Recent Developments in
2016. Biometric Liveness Detection: A Review," IEEE Transactions on
[13] [Online]. Available: https://support.apple.com/en- Information Forensics and Security, vol. 15, no. 12, pp. 3196-3211,
us/HT208108#:~:text=The%20technology%20that%20enables%20Fa 2020.
ce,infrared%20image%20of%20your%20face. [19] Ahmed Bouridane, David Zhang, and Farzin Deravi, "Recent
[14] [Online]. Available: https://learn.microsoft.com/en- Developments in Fusion of Biometrics with Other Authentication
us/windows/security/identity-protection/hello-for-business/hello- Methods: A Review," International Journal of Computer Science and
overview Biometrics, vol. 8, no. 4, pp. 1-20, 2020.
[15] X. Liu, J. Wu, and Y. Wang, "Recent advances in face recognition: a [20] S. Sharma, A. K. Jain, and R. Ross, "Recent Developments in
survey," IEEE Transactions on Emerging Topics in Computational Biometric on Edge: A Review," IEEE Transactions on Information
Intelligence, vol. 4, no. 2, pp. 168-186, 2020. Forensics and Security, vol. 16, no. 10, pp. 2791-2807, 2021.
[16] A. Ross and N.K. Ratha, "Recent advances in behavioral biometrics: a [21] M. Mannschreck, “What Is the Future of Biometric Technology?
survey," IEEE Transactions on Information Forensics and Security, Trends to Watch”, https://www.1kosmos.com/biometric-
vol. 11, no. 6, pp. 1460-14 authentication/future-of-
biometrics/#:~:text=What%20is%20the%20future%20of%20biometri
[17] K. W. Bowyer, K. I. Chang, P. Yan, and P. J. Flynn, "Multi-Modal
cs%3F,or%20using%20money%20management%20systems, 2022.
Biometrics: An Overview," IEEE Transactions on Information
Forensics and Security, vol. 2, no. 3, pp. 297-314, 2007.

View publication stats