Volume III, Issue X, October 2014 IJLTEMAS ISSN 2278 - 2540

An Introduction to Biometrics: The Power of Security

Selva Priya G1, Anitha P2, Vinothini C3
1
PG Scholar, Department of CSE, Dr. N.G.P Institute of Technology
2
Assistant Professor , Department of CSE, Dr. N.G.P Institute of Technology
3
Assistant Professor, Department of CSE, Dr. N.G.P Institute of Technology

Abstract- Biometrics is the science of establishing security for A. Physiological vs. Behavioral
person identification. It refers to the automatic recognition of a
person based on his/her physiological or behavioral When describing this biometric technology, it is
characteristics. It is possible to confirm or establish an essential to distinguish between physiological and
individual’s identity based on the traditional methods
involving passwords and PIN numbers for its accuracy and
behavioral human characteristic.
case sensitiveness by using biometrics. Examples of such
applications include secure access to laptops, buildings, cellular A physiological characteristic is relatively a stable
phones, computer systems and ATMs. In the absence of strong human physical characteristic, such as a iris pattern,
personal recognition schemes, these systems are vulnerable to fingerprint, or blood vessel pattern on the back of the eye.
an impostor. In this paper, a brief overview for biometrics will This type of measurement is coherent and fixed without
be presented. significant duress. Alternatively, a behavioral characteristic
is a reflection of an individual psychological character,
Index terms Biometrics, security, identification, impostor. although physical peculiarity such as size and gender have a
major inspiration. Some of the examples of behavioral
I. INTRODUCTION peculiarity used to identify individuals include: Keystroke
dynamics, and speech identification and/or verification [2].

A biometric system is essentially a pattern-recognition

system that recognizes a person based on a feature
vector derived from a specific physiological or behavioral
B. Physical Biometrics

characteristic that the person possesses [1]. A biometric 1. Fingerprint: Analyzing fingertip blueprint
system based on a physiological characteristics is generally
more reliable than one which adopts behavioral 2. Facial recognition/face location: Measuring facial
characteristics, even if the latter may be more easy to physical appearance
combine within certain specific applications. Biometric
system which perform in two modes: verification or 3. Hand geometry: Measuring the of the hand outline
identification. Identification which comparing the acquired
biometric information against templates corresponding to all 4. Iris scan: Analyzing highlight of colored ring of the eye
users in the database. In verification, it involves comparison
with only those templates corresponding to the privilege 5. Retinal scan: Analyzing blood container in the eye
identity. This suggests that identification and verification
6. Vascular patterns: Analyzing vein blueprint
are two problems that should be deal with separately. A
biometric system consists of four basic components: 7. DNA: Analyzing genetic nature
 Sensor – It collects data and then converts the 8. Biometric data watermarking (which is really a method
information to a digital format.

rather than a physical aspect) is used to store/hide biometric
Signal processing algorithms – It perform quality information [3].
control activities and then develop the biometric


template. C. Behavioral Biometrics
Data Storage – It keeps information that new biometric
1. Speaker/voice recognition: Analyzing vocal deeds

templates will be compared to
Matching algorithm – It compares the new biometric
2. Signature/handwriting: Analyzing signature active

template to one or more templates in data storage
Decision Process – It uses the results from matching
3. Keystroke/patterning: Measuring the interval spacing of
component to make a system-level decision (either
typed words
automated or human-assisted)

www.ijltemas.in Page 12
Volume III, Issue X, October 2014 IJLTEMAS ISSN 2278 - 2540

II. TYPES OF BIOMETRIC SYSTEMS stringent performance requirements. Such systems are
probable to be more consistent due to the presence of
A. Unimodal biometric systems multiple, individual pieces of proof. These systems are able
to encounter the severe performance requirements imposed
There is a collection of problematic with biometric by various applications.
systems installed in real world applications which prove that
biometrics is not fully solved problem. A multimodal system could be, for case in point, a
mixture of face recognition, voice verification, fingerprint
Drawback of biometric systems using any distinct biometric verification and smart-card or any other combination of
characteristic [4]: biometrics. This superior structure takes benefit of the
proficiency of each individual biometric and can be used to
1) Noise in sensed data: Example is a fingerprint with a overcome some of the limitations of a single biometric.
panic. Earsplitting data can also result from accumulation of
dirt on a sensor or from ambient conditions. III. A COMPARISON OF VARIOUS BIOMETRICS

2) Intra-class variations: Biometric data acquired from an There are several human distinguishable traits that fit
individual during authentication may be very different from the definition of biometrics. In order to be used for
the data that was used to generate the template during distinguish a person, the human feature needs to be
registration. This distinction is typically caused by a user distinctive and no matter to change. Fingerprints, for
who is incorrectly interacting with the sensor. example, have been used for over one hundred years and,
therefore, are commonly well acknowledged as a
3) Distinctiveness: While a biometric trait is expected to recognition technology. Other technologies such as hand
vary significantly across persons, there may be large geometry, speaker and iris recognition, face are also
different class similarities in the feature sets used to commonly accepted. A biometric that would require giving
represent these traits. This drawback restricts the intolerance a blood sample for frequent personal verification would
provided by the biometric trait. probably not be very well prevalent. Performance
considerations are significant. No biometrics can warrant
4) Non-universality: While every user is expected to possess
one hundred percent correctness. A short-term beginning of
the biometric trait being acquired, in reality it is possible
the commonly used biometrics is given below:
that a group of users do not posses that particular biometric
DNA: Deoxyribo Nucleic Acid (DNA) is the one-
characteristic.
dimensional ultimate unique code for one’s individuality -
5) Spoof attacks: An individual may attempt to forge the except for the fact that identical twins have identical DNA
biometric trait. This is predominantly straightforward when blueprint. It is, however, recently used above all in the
signature and voice are used as an identifier. context of forensic applications for person identification. [8]
Three misgivings bound the utility of this biometrics for
other applications:
B. Multimodal biometric systems
(i) Contamination and sensitivity
Drawbacks of unimodal biometric systems can be
overcome by using multimodal biometric systems [5]. A (ii) Automatic real-time recognition issues
multimodal biometric system uses multiple applications to
(iii) Privacy issues
capture different types of biometrics. This allows the
combination of two or more types of biometric recognition Ear: It has been suggested that the shape of the ear and the
and verification systems in order to meet stringent structure of the cartilegenous tissue of the pinna are
performance requirements. Such systems are probable to be characteristic. The ear recognition methods are based on
more consistent due to the presence of multiple, individual matching the distance of salient points on the pinna from a
pieces of proof [6]. These systems are able to encounter the landmark site on the ear. The structure of an ear are not
severe performance requirements imposed by various anticipated to be very distinctive in establishing the identity
applications [7]. A multimodal system could be, for case in of an individual. [8]
point, a mixture of face recognition, voice verification,
fingerprint verification. Face Recognition: Different technologies can be used for
face recognition. One method consists on observing an
A multimodal biometric system uses multiple image of the face using an inexpensive camera (visible
applications to capture different types of biometrics. This spectrum). This method classically models key
allows the assimilation of two or more types of biometric characteristics from the central portion of a facial image
recognition and verification systems in order to meet

www.ijltemas.in Page 13
Volume III, Issue X, October 2014 IJLTEMAS ISSN 2278 - 2540

extracting these features from the captured image(s) that do applications. In addition, an individual's jewelry (e.g., rings)
not change over time while avoiding superficial features or limitations in dexterity (e.g., from arthritis), may pose
such as facial appearance or hair. Major charity event of further challenges in extracting the correct hand geometry
facial recognition are that it is non-intrusive, hands-free, information. [11]
provides for continuous authentication and is accepted by
majority users. Acceptance sample sizes(e.g., 5 face Iris: As far as is known, every human iris is measurably
samples) may range from 1 KB-2KB/sample). Smaller distinctive. It is fairly easy to detect in a video picture, does
template sizes are also used (e.g., less than 100 bytes). [9] not wear out, and is secluded from the external environment
by the cornea (which in turn has its own cleaning
Fingerprints: Fingerprints are essential. By 1998, mechanism). The iris blueprint contains a large amount of
fingerprint recognition products reported for 78% of the casualness and emerge to have many times the number of
total sales of biometric skill. These products look at the degrees of freedom of a fingerprint. It is molded between
abrasion ranges that cover the fingertips and organize the third and eighth month of gestation, and (like the
patterns, such as branches and end points of the ridges. fingerprint pattern) is phenotypic in that there appears to be
Some also look at the apertures in the skin of the ridges. limited genetic influence; the mechanisms that form it
Fingerprint recognition gadget for desktop and laptop access appear to be tumultuous. So the blueprint are different even
are widely available from many different vendors at a low for identical twins (and for the two eyes of a single
cost. [10] The relatively small size allows the sensor to be individual), and they materialize to be stable throughout life.
integrated in other devices (e.g., mice, keyboards). This [8]
biometric technology uses the pattern of friction ridges and
valleys on an person's fingertips. These outline are Retinal Scanning: This method of personal authentication
considered distinctive to a specific person. The same fingers uses the vascular patterns of the retina of the eye. [9] In
of equal twins will also differ. A user does not need to type healthy individuals, the vascular pattern in the retina does
passwords - instead, only a touch to a fingerprint device not change over the course of an person’s life. The blueprint
provides almost instant access (typically less than 1 sec.). A are scanned using a low-intensity (e.g., near-infrared) light
representative enrollment identifier may include two finger source. It need the user to look into a device and focus on a
samples (e.g., 1 KB) although smaller finger samples are given point. The image acquisition involves coordination of
also used. One of the trials of fingerprint expertise is the subject, entails contact with the eye fragments. [10]
individuals that have poorly defined (or tenuous) ridges in
their fingerprints. [9] Signature Verification: The way a person signs her name
is known to be a characteristic of that person. Monogram of
Gait: Gait is the peculiar way one walks and is a complex some people vary materially: even successive impressions
spatial-temporal biometric. Gait is not supposed to be very of their signature are substantially different. [17] It is based
typical, but is amply discriminatory to allow verification in on measuring energetic signature features such as velocity,
some low-security applications. posture is a behavioral gravity and direction used when a person signs a standard,
biometric and may not wait invariant, unusually over a long recorded pattern (e.g., autograph). One spotlight for this
period of time, due to vicissitudes in body weight, chief method has been e-business applications. [9]
grievances involving joints or brain, or due to
intoxication.[11] Voice Recognition: Voice recognition or speaker
recognition is the problem of identifying a speaker from a
Hand and Finger Geometry: Hand recognition has been short utterance. [8] This biometric technology uses the
available for over twenty years. To realize personal acoustic features of speech that have been found to differ
certification, a system may measure physical characteristics between person. These acoustic methods reflect both
of the fingers or the hand such as length, width, thickness framework (e.g., size and shape of the throat and mouth)
and outward area of the hand. These procedures of personal and learned behavioral patterns (e.g., voice pitch, speaking
certification are well established. [9] Some systems require style). [9] A disadvantage of voice-based recognition is that
a very small biometric sample (e.g., 9 bytes). Hand speech features are sensitive to a number of factors such as
geometry can frequently be found in physical access control contextual blast. Speaker recognition is most suitable in
for commercial and residential purposes, for time and phone-based applications but the voice signal over phone is
turnout systems, and for common personal certification typically degraded in quality by the microphone and the
communication channel. [10]

www.ijltemas.in Page 14
Volume III, Issue X, October 2014 IJLTEMAS ISSN 2278 - 2540
IV. COMPARISON BETWEEN VARIOUS BIOMETRICS

Biometrics Advantages Disadvantages

Facial recognition It does not require any co-operation of Face recognition isn’t perfect and face
the test subject to do any work. challenges to perform under certain
conditions.
This performs massive identification
which usually other biometric system Not much effective for low resolution
can’t perform [13] images [13].
Iris High accuracy and High recognition Iris scanners tend to be more expensive in
process speed [13]. comparison with additional biometrics [12].

Scalability along with the speed is Iris recognition is vulnerable to inadequate

significantly advantageous [14]. image quality.
Finger print A fingerprint pattern has individually Cuts, marks transform fingerprints which
distinctive composition and often has negatively effect on
characteristic remains the same with performance [13].
time [13].

It is easy to use along with the high Finger prints aren’t private. The finger
verification process speed and accuracy prints are for life time and there is no way to
[12]. get back to a secure situation [15].
Finger Vein Vein patterns need only low image Expensive: The actual technologies
resolution. seriously aren’t cheap enough for bulk
deployment.

Larger Size: The existence of CC camera

The level of accuracy from vein makes the system larger than a fingerprint
recognition systems is quite outstanding. scanner.
Voice recognition This technique helps those people who May hacked with prerecorded voice
have difficulty of using their hands. messages.

It does not require any training for users.

Possesses primary amount of time for
adjustment with each user's voice

V. CONCLUSION VI. FUTURE WORK

This paper presents a shorter introduction on Biometrics technology is used in a number of ways and in
numerous biometric techniques undertaking the comparison different fields of our daily lives. In future we mainly focus
examination regarding widely used biometric identifiers and on facial expression recognition technology. Using this
also the identification strategies. As this is a new technology technology, we can easily identify a person in a crowd and
for most of the peoples since it has simply been by so we can verify their identity. We can furthermore make
implemented in public areas for short time period. It use of this technology to detect previously identified
provides benefits that may improve our lives in such a way terrorists, criminals or scammers in society. It may help us
by increasing security and efficiency, decreasing scams and to reduce the criminal offense in the world. But as we can
reducing password administrator cost. Despite the fact the see from the above given comparison between different
biometrics security systems still have many issues like data biometrics, it is clear that face recognition faces a
privacy, physical privacy, and spiritual arguments. challenging problem in the field of accuracy, efficiency,
speed cost and security. So we need to work upon it to make

www.ijltemas.in Page 15
Volume III, Issue X, October 2014 IJLTEMAS ISSN 2278 - 2540
it more effective.

REFERENCES

[1] S. Prabhakar, S. Pankanti, A. K. Jain, "Biometric Recognition: Security

and Privacy Concerns", IEEE Security & Privacy, March/April 2003, pp.
33-42.
[2] Arpita Gopal, Chandrani Singh, e-World : Emerging Trends in
Information Technology, Excel Publication, New Delhi (2009).
[3] John Woodward, Nicholas M. Orlans, Peter T. Higgins, Biometrics,
Tata McGraw Hill.
[4]A. K. Jain, A. Ross, S. Prabhakar, "An Introduction to Biometric
Recognition", IEEE Trans. on Circuits and Systems for Video Technology,
Vol. 14, No. 1, pp 4-19, January 2004.
[5] L. Hong, A. K. Jain, S. Pankanti, "Can multibiometrics improve
performance?," in Proc. AutoID'99, Summit, NJ, October 1999, pp. 59-64.
[6] L. I. Kuncheva, C. J. Whitaker, C. A. Shipp, R. P. W. Duin, "Is
independence good for combining classifiers?," in Proc. Int. Conf. Pattern
Recognition (ICPR), Vol. 2, Barcelona, Spain, 2001, pp. 168-171.
[7] L. Hong, A. K. Jain, "Integrating faces and fingerprints for personal
identification," IEEE Trans. Pattern Analysis Machine Intell., Vol. 20, pp.
1295-1307, December 1998.
[8] Anil K. Jain, Arun Ross and Salil Prabhakar: “An Introduction to
Biometric Recognition” IEEE Transactions on Circuits and Systems for
Video Technology, Special Issue on Image- and Video-Based Biometrics,
Vol. 14, No. 1, January 2004.
[9] Fernando L. Podio: “Personal Authentication Through Biometric
Technologies”.
[10] Ross Anderson's: “Chapter 13th Biometrics of Security Engineering”.
[11] Anil K. Jain, Arun Ross and Salil Prabhakar: “An Introduction to
Biometric Recognition” IEEE Transactions on Circuits and Systems for
Video Technology, Special Issue on Image- and Video-Based Biometrics,
Vol. 14, No. 1, January 2004.
[12] Anil K. Jain, Arun Ross and Salil Prabhakar (2004), “An Introduction
to Biometric Recognition.”.
[13] Alina Klokova, “Comparison of Various Biometric Methods”.
[14] Joseph N. Pato and Lynette I. Millett, Editors; Whither Biometrics
Committee; National Research Council (2010), “Biometric Recognition:
Challenges and Opportunities”.
[15] David Weiss (2009), “Fingerprint Biome.

www.ijltemas.in Page 16