Biometric authentication

Saleem Ahmad1
Ordonez Bonilla Juan Esteban2
1 Óbudai Egyetem, Bánki Donát Gépész és Biztonságtechnikai Mérnöki Kar, MSc V. évfolyam. Budapest, Hungary,
ahmada.saleem@hotmail.com
1 Óbudai Egyetem, Bánki Donát Gépész és Biztonságtechnikai Mérnöki Kar, MSc V. évfolyam. Budapest, Hungary

juesteordonez@stud.uni-obuda.hu

Abstract
Biometrics is defined as the science of identifying individuals supported by physical or behavioural attributes of the indi-
vidual like fingerprints, palmprint, face, voice, iris & gait. The continuously growing necessity for robust human recogni-
tion techniques in sensitive applications such as secure access control, international border crossing, and enforcement,
biometrics emerged itself as a crucial feasible technology that can be integrated into a huge-scale identity management
system. Biometric systems can operate under the premise that many of the physical or behavioural characteristics of the
human being which are always distinctive from one individual to another, and that they can be reliably acquired via appro-
priately designed sensors and even represented in a numerical format that lends to automated procedure within the context
of identity management. Hence, these systems could also be viewed as pattern recognition engines which will be incorpo-
rated in diverse markets. Every part of the human body is exclusive and if we will steadily capture the required area or
trait, the efficient biometric validation system can be built. With the advancement in technology sensors for a variety of
biometrics are obtainable. For instance, the fingerprint, handwritten signatures, palmprint, face, voice as conventional bi-
ometric traits, mainly due to advanced sensors where we can have a 3D face and hand geometry, finger-knuckle print, ear
geometry, facial thermogram, gait, DNA, retinal scan as emerging traits and so on.

Keywords: Recognition, biometrics, trait, identify.

− Permanence: Measures how well a biometric resist

1. Biometrics aging and other variance over time.
− Collectability: simple acquisition for measurement
In computer science, particularly biometrics is used − Performance: Accuracy, speed, and robustness of
as thanks to identity access management and access technology used.
control. it's also will not identify people in groups − Acceptability: Degree of approval of a technology.
that are under observation.
− Circumvention: simple use of a substitute.
Physiological are related to the shape of the body. By using biometrics we might able to identify an
Examples include hand geometry, fingerprint, DNA, identity supported by who you are, rather than by
palmprint, face recognition, iris recognition, which what you have, such as an ID card, or what you
has taken the place of the retina, and odor. remember, such as a password.
In some applications, biometrics are used to
Behavioral are related to the behavior of human. Ex- supplement ID cards and passwords thereby
amples include typing rhythm, gait, and voice. Some imparting a further level of security. Such an
researchers have coined the term behavior metrics for appointment is usually called a dual-factor
this class of biometrics. authentication scheme. [1]
Speaking and voice feature is also a physiological
trait because each person has a special vocal tract, but 1.1. Brief History of Biometrics
voice recognition has supported the tactic of the way
Using biometrics date was back over a thousand
an individual speaks, commonly classified as behav-
years. In East Asia, potters placed their finger-prints
ioral. We can highlight, it's possible to know if a per-
on their products as an early form of brand identity.
son's feature is often used for biometrics in terms of
In Egypt‟s Nile Valley, traders were formally
the subsequent strictures:
identified based on physical characteristics such as
− Universality: Everyone has the characteristic. height, eye color, and complexion. This information
− Uniqueness: Indicates how well the biometric sep- helped identify trusted trad-ers whom tradesmen had
arates individuals from another. successfully transacted business within the past.
2 Authors – Biometric authentication. (2020) Ahmad Saleem – Juan Ordonez

Biometrics was employed as a billboard, modern A greater variety of fingerprint scanning devices are
technology has been around since the earlies 1970s, available than for any other biometric. Nowadays
when the primary commercially available device was can be easily reached almost everyone and its cur-
delivered to the market. one among the primary rently added on mobile phones as a standard feature.
commercial applications was used in 1972 when a [3]
Wall Street company, Shearson Hamil, installed
Identimat, a finger-measurement device that served 2.1.2. Hand Geometry
as a timekeeping and monitoring application. Since
this 1972 deployment, bio-metrics has improved Hand geometry involves measuring and analyzing
tremendously in ease of use and diversity of the hand’s shape. This biometric mechanism offers a
applications. The advancement of biometrics has good balance of performance characteristics and is
been driven by the increased computing power at relatively easy to apply. It is useful where there are
lower costs, better algorithms, and cheaper storage several users or where users shall access the system
mechanisms available today. [2] infrequently and even perhaps less strict in their ap-
proach to the system. Hand scanners use this infor-
2. Biometric traits mation to determine the length, width, thickness, and
curvature itself of your hand and fingers. It digitalizes
The biometric traits are defined as actual character- that information into a data template, which is used
istics or entities which are commonly used to iden- for matching/recognition. Since hands and fingers are
tify and classify humans. These characteristics are less distinctive than fingerprints or irises, some peo-
referred to as traits, indicators, identifiers, or mo- ple are less likely to be part of this system. However,
dalities. As quoted earlier there are two classes of many people's hands change over time and are vul-
biometric traits as physiological and behavioral, nerable to aging, injuries, changes in weight, or some
they are explained as follows. exceptional cases as arthritis. Leading on the trend to
combine different systems, this might be a good ex-
2.1. Physiological Biometric Traits ample since is possible to combine and share features
within the hand shape.
Based on the physical characteristics of the human
body. Every physical feature that can be measured 2.1.3. Hand Vein Geometry
can be included in this category. [2]
A person's veins are a completely unique feature.
2.1.1. Fingerprints Even twins do not have identical veins, and for in-
stance, an individual's veins differ between their left
A fingerprint looks at the patterns found on a finger- and right sides. Veins are not normally visible
tip. There are a variety of approaches to fingerprint through the skin, making them hard to counterfeit or
verification. Some emulate the traditional police tamper with. The shape also changes as little as indi-
method of matching minutiae; others use straight vidual ages.
pattern-matching devices; and still, others are a bit
more unique, including things like moiré fringe pat-
terns and ultrasonic. Some verification approaches
can detect when a live finger is pre-scented; some
can not. Figure 1. shows features of fingerprints on
a human’s hand and a typical fingerprint captured
through an optical fingerprint scanner.

Figure 2. Hand veins data obtaining

In order to gather vein recognition information,

Figure 1.Fingerprint scanner technology
simply place your finger, wrist, palm, or the back of
your hand on or near the scanner and a camera will
Authors – Biometric authentication. (2020) Ahmad Saleem – Juan Ordonez 3

cluster a digital picture using near-infrared light. The the proposed system, the inherent finger knuckle
hemoglobin in your blood absorbs the sunlight, so print patterns can be captured and hence the unique
veins appear black within the picture. The system features of FKP can be better exploited.
creates a reference template that supported the shape,
form, and spot of the vein structure. [3] 2.1.6. Face

2.1.4. Palmprints Face recognition analyzes facial characteristics. It re-

quires a digital camera to develop a facial image of
The palms also contain a pattern of ridges and valleys the user for authentication. Because facial scanning
much like the fingerprints. The palm’s area is much needs an extra peripheral not customarily included
larger than the area of a finger and. Hence, palmprints with basic PCs, it is more of a niche market for net-
are expected to be even more distinctive than finger- work authentication. Security cameras present in a
prints. Though palmprint scanners need to capture a variety of public places facial recognition is a viable
large area, they are bulkier and costlier than the fin- option for biometric identification. Face biometrics is
gerprint sensors. Human palms also contain addi- relatively less accurate but requires low user co-op-
tional distinctive features such as principal lines and eration. With the advancement in technology varia-
crinkles which will be captured even with a lower tions such as facial thermogram (temperature distri-
resolution scanner, which would be cheaper. So, bution on a face) and 3D face, mapping is also being
when using a high-resolution palmprint scanner, all implemented. Figure 4 (a) and (b) shows these bio-
the features of the palm such as hand geometry, ridge metrics. Facial thermography refers to the pattern of
and valley features, wrinkles, and principal lines may facial heat mainly caused by the distinctive flow of
be combined to build a highly accurate biometric sys- blood which is under the skin. The modern IR cam-
tem. [3] eras capture this heat to produce a thermal pattern.
Due the vein and tissue structure of an individual’s
face are idiosyncratic, the infrared image is also dis-
tinctive. The thermal data is analyzed to yield bodily
information, which is converted once again to a tem-
plate and used for identification. [5]

Figure 3. Typical Palmprint showing Principle lines,

Ridges and wrinkles

2.1.5. Finger-knuckle Print

Figure 4. Advanced Face Biometrics (a) 3D Face
The FKP refers to the image pattern of the outer sur-
Mapping (b) Facial Thermogram
face around the phalangeal joint of one’s finger,
which is created by bending slightly the finger-
knuckle. FKP contains a piece of rich texture infor-
mation formed by wrinkles present on the finger back
surface. This biometric trait is relatively new and yet
to be much explored. The FKP capture device is de-
veloped by Hongkong polytechnic university, the
proposed system captures the image around the fin-
ger knuckle area of a finger directly, which largely
simplifies the following pre-processing steps. In the
meantime, with specific design, the size of the imag-
ing system can be greatly reduced, which improves
much of its applicability. Subsequently, the finger-
knuckle will be slightly bent when being imaged in
4 Authors – Biometric authentication. (2020) Ahmad Saleem – Juan Ordonez

2.1.7. Iris be provided on regard to the type of features which

depend on human behavior. [4]
On the other hand, iris analysis involves infinitive
features found in the coloured ring of tissue that sur- 2.2.1. Handwritten Signature
rounds the pupil (Figure 5.).
A human handwriting style is unique from person to
person. initially glance, using handwriting to identify
people might not appear to be a good Consequently,
many people can learn to copy other people's hand-
writing with a touch time and practice. It looks like it
would be easy to urge a replica of someone's signa-
ture or the specified password and learn to forge it.
Though, biometric systems don't just check out how
you shape each letter; they analyze the act of writing.
Basically, they examine the pressure you employ and
therefore the speed and rhythm with which you write.
They as well, record the sequence in which you form
letters like whether you add dots and crosses as you
go or after you finish the word. [6]
Figure 5. Captured iris Image - Anatomy of Human
Eye Signature verification analyzes the way a user signs.
Signing features such as speed, velocity, and pressure
are as important because the finished signature’s
static shape. Figure 6 (a) and (b) show such signature
The iris scanning certainly represents the less intru- scanners. A signature having different pressure levels
sive of the eye-related biometrics, uses a fairly con- is shown in Figure 6 (c). Signature verification enjoys
ventional camera element, and requires no close con- a synergy with existing processes that other biomet-
tact between the user and the reader, enough fair dis- rics don't. People are wont to signatures as a means
tance to catch the iris features (i.e. shape, color, size). of transaction-related identity verification, and most
Besides, it has the potential for higher than aver-age would see nothing unusual in extending this to en-
prototype matching performance. Iris scanners are compass biometrics. Signature verification devices
becoming common as the high-security applications are reasonably accurate operational and noticeably
because people's eyes are so unique (the chance of lend themselves to applications where a signature is
mistaking one iris code for an additional is 1 in 1 an accepted identifier. The main application domain
x1078). Even simple scanners allow at least 210 of handwritten signature is in banking and e-com-
points of reference for comparison, as against 60 or merce & document authentication.
70 points in fingerprints. The iris may be a visible but
sheltered structure, and it doesn't usually change over
time. Often, people's eyes also remain unchanged af-
ter eye surgery, and blind people can use iris scanners
as long as their eyes have irises eye’s element. Con-
tact lenses and glasses do not interfere or origin im-
precise readings. [4] Figure 6.(a) and (b) Dynamic Signature Capturing
Devices (c) Dynamic Signature showing Different
The biometric traits listed and described above are Pressure Levels by Different Colors
considered the most valuable and representative.
However, besides these significant biometrics traits,
3. Keystroke Dynamics
some other complicated contention systems are still
under development as ear geometry, DNA, and a still
hard one, the unique body odor. Unlike signature verification (and most every other
biometric) keystroke dynamics are somewhat unique
in that they do not require special sensor equipment
2.2. Behavioral Biometric Traits
beyond a regular keyboard. The keystroke dynamics
are captured entirely by software, so the technique
The behavioral biometrics deals with a certain act
performed by a human. This includes speaking, writ-
ing, walking etc. As follows, some small details will
Authors – Biometric authentication. (2020) Ahmad Saleem – Juan Ordonez 5

can be applied to any system that accepts and pro- https://www.theseus.fi/bitstream/han-

cesses keyboard input events. dle/10024/44684/Babich_Aleksandra.pdf.

Keystroke dynamics can be used for single authenti- [3] D. Bhattacharyya, R. Ranjan, F. A. a, and
cation events or continuous monitoring. Continuous M. Choi, “Biometric Authentication : A
monitoring is not normally done in commercial bio- Review,” Int. J. Serv. Sci. Technol., vol. 2,
metrics products; however, it has been proposed as a no. 3, pp. 13–28, 2009.
legitimate and reasonable means to help prevent un-
authorized use of unattended terminals. Keystroke
[4] A. K. Jain, K. Nandakumar, and A. Ross,
monitoring is the unsophisticated, yet surprisingly
“50 years of biometric research: Accom-
easy way to achieve logging of every key pressed by
plishments, challenges, and opportunities,”
a user. While many have argued the potential for
Pattern Recognit. Lett., vol. 79, pp. 80–
abuse far outweighs any legitimate use of the process,
105, 2016, doi:
keystroke monitoring is sometimes used to provide
10.1016/j.patrec.2015.12.013.
auditing and security information that may be re-
quired in certain sensitive environments. Such sys-
tems should be marked with banners and notices to [5] R. H. C. Work, “(12) United States Patent,”
inform users that their use is monitored. The most vol. 1, no. 12, 2001.
natural application for keystroke dynamics is to
“harden” passwords. The keystroke dynamics of each [6] K. Kang, J. Park, S. Hyoung, and J. Y. Oh,
user are used to augment existing passwords by re- “( 12 ) United States Patent,” vol. 2, no. 12,
quiring that the password be entered in a manner con- 2014.
sistent with the intended user. A related use of key-
stroke dynamics is for human typing detection. That
is, keystroke patterns can be used to help determine [7] W. Luckhardt and S. Diego, “( 12 ) United
the difference between man and machine (live human States Patent ( 76 ) Inventor : CA George,” vol. 2,
typing vs. scripted programs). Other biometric traits no. 12, 2012.
based on human behavior are Gait, Voice, and Facial
Expressions, etc. Table 1.1 shows a comparison be-
tween key biometric traits. [7]
Table 1.Biometrics mechanisms comparison

4. References

[1] C. S. Forensics, “History of Fingerprints,”

2015. http://www.crimescene-foren-
sics.com/Crime_Scene_Forensics/His-
tory_of_Fingerprints.html.

[2] A. Babich, “Biometric Authentication .

Types of biometric identifiers,” pp. 1–56,
2012, [Online]. Available: