A SURVEY ON MULTIMODAL BIOMETRIC SYSTEM

Abstract

Multi-biometrics is an exciting and interesting research topic. It is used to recognizing individuals for security purposes; to increase security levels. The recent research trends toward next biometrics generation in real-time applications. Also, integration of biometrics solves some of unimodal system limitations. However, design and evaluation of such systems raises many issues and trade-offs. A state of the art survey of multi-biometrics benefits, limitations, integration strategies, and fusion levels are discussed in this paper.

Full text

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
A SURVEY ON MULTIMODAL
BIOMETRIC SYSTEM
POOJA B S ARJUN B C
PG Student Asst. Prof. and HOD
Department of CS & E Department of IS & E
Rajeev Institute of Technology Rajeev Institute of Technology
Hassan, Karnataka, India Hassan, Karnataka, India
pooja.bs187@gmail.com bc.arjun@gmail.com
ABSTRACT
Multi-biometrics is an exciting and interesting research topic. It is used to recognizing individuals for
security purposes; to increase security levels. The recent research trends toward next biometrics generation in
real-time applications. Also, integration of biometrics solves some of unimodal system limitations. However,
design and evaluation of such systems raises many issues and trade-offs. A state of the art survey of multi-
biometrics benefits, limitations, integration strategies, and fusion levels are discussed in this paper.
Keywords—Biometrics; Unimodal biometric systems; Multimodal biometric systems; Fusion levels;
Recognition methods; Design and Implementation.
I. INTRODUCTION
Authentication (identifying an individual using security system) of users is an essential but, difficult
accurate and secured practical authentication technology. Traditional techniques for user authentication could be
categorized as: (1) Token based techniques (i.e. key cards and smart cards) and (2) Knowledge-based techniques
include text-based and picture based passwords (often mix of username and password).
Due to vulnerabilities in above methods (It could be easily transgressed or lost or forgotten); Traditional
techniques are considered to be not reliable or secure, and are not presently sufficient in some security
application zones. The primary advantage of biometrics over these methods is that it cannot be misplaced,
forgotten or stolen. Also, it is very difficult to spoof biometric traits. Due to greater accuracy and higher
robustness of biometric recognition, biometric solutions become popular and preferred methods to analyze
human characteristics for security - authentication and identification - purposes. It could not be duplicated or
counterfeited and misused.
Practically, the use of biometrics information is the most secure method. Consequently, it is now needed in
many fields such as surveillance systems, security systems, and physical buildings. Other applications of
biometrics systems include: access control (access to computer networks), forensic investigations, verification
and authentication, e-commerce, online banking, border control, parenthood determination, medical records
management, welfare disbursement and security monitoring. Biometrics applications increased dramatically in
functionality in many more fields.
In the most general definition, "Biometric Technologies" is defined as an automated methods of verifying
and/or recognizing the identity of a living individual based on two categories: (1) Physiological Biometrics
include (Facial, Hand and Hand vein infrared thermogram, Odor, Ear, Hand and Finger geometry, Fingerprint,
Face, Retina, Iris, Palm print, Voice, and DNA), and (2) Behavioral Biometrics like (Gait, Keystroke, Signature)
which measure the human actions. Also, human electrocardiogram (ECG) signal is considered one of Biometric
features used in individual recognition and authentication.

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
Depending on the application context, biometric systems may operate in two modes: verification mode and
identification mode. Through verification mode, the system verifies the identity by comparing the enrolled
biometric trait by a stored biometric template in the system (1:1). This mode is used for positive recognition,
and it aims to prevent the multiple individuals from using the same identity. In the identification mode, the
enrolled sample is then compared with existing templates in a – central – database (1: M). A database search is
crucial and needed. The identification mode is critical in negative recognition applications, which aims to
prevent a single user from using multiple identities. Negative identification is also known as screening.
Obviously, verification is less computationally expensive and more robust compared with identification. On the
other hand, the latter is more convenient and less obtrusive.
Multi-biometric systems distinguished over traditional unibiometric systems as it addresses the issue of
non-universality and noisy data. Multi-biometric systems can facilitate the indexing of large-scale biometric
databases. Also, it becomes not easy for an impostor to spoof all the biometric traits of an authorized enrolled
person. Generally, it is much more vital to fraudulent technologies because it is more difficult to forge multiple
biometric characteristics. Multi-biometric recognition systems also have benefits in the continuous monitoring
of an individual in situations or tracking him when a single trait is not sufficient in use. These systems continue
to operate even if parts of biometric sources become unavailable of a failed (i.e. sensor malfunction, software
malfunction, or deliberate user manipulation); it may view as a fault tolerant system. For these benefits,
multimodal expected to provide higher accuracy rate.
II. BIOMETRICS OVERVIEW
A biometric system to be practical and reliable should meet the specified requirements/characteristics:
 Universality (availability): Each person should have the characteristic. Availability is measured by the
"failure to enroll" rate.
 Distinctiveness: It declares that any two persons should sufficiently have different characteristic. It is
measured by the False Match Rate (FMR), also known as "Type (II) error".
 Permanence (robustness): The characteristic should be stable (with respect to the matching features)
over a period of time. This means the stability over age. Robustness is measured by the False Non-
Match Rate (FNMR), also known as "Type (I) error".
 Collectability (accessible): The characteristic can be measured quantitatively and easy to image using
electronic sensors. Accessibility can be quantified by the "throughput rate" of the system.
 Performance: It means to achieve recognition accuracy, speed, and the resources required to the
application.
 Acceptability: The particular user population and the public, in general, should have no (strong)
objections to the measuring/collection of the biometric characteristic. Acceptability is measured by
polling the device users.
 Resistance to Circumvention: Tests and proofs how the system resists fraudulent methods easily.
Accordingly, each one could be used in authentication and/or identification applications. Predicting the
"false acceptance" and "false rejection" rates, system throughput, user acceptance, and cost savings for
operational systems from test data, is a surprisingly difficult task.
Consequently, it is impossible to state that a single biometric characteristic is "best" for all applications,
populations, technologies and administration policies.
Table 1: Comparison of Biometric Characteristics
Biometric
Characteri
stic
Universal
ity
Distinctiven
ess
Permane
nce
Collectabil
ity
Performa
nce
Acceptabil
ity
Circumvent
ion
Hand Vein
M
M
M
M
M
M
L
Gait
M
L
L
H
L
H
M

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
Keystroke
L
L
L
M
L
M
M
Odor
H
H
H
L
L
M
L
Ear
M
M
H
M
M
H
M
Hand
Geometry
M
M
M
H
M
M
M
Fingerprin
t
M
H
H
M
H
M
M
Face
H
L
M
H
L
H
H
Retina
H
H
M
L
H
L
L
Iris
H
H
H
M
H
L
L
Palm Print
M
H
H
M
H
M
M
Voice
M
L
L
M
L
H
H
Signature
L
L
L
H
L
H
H
DNA
H
H
H
L
H
L
L
(H: High, M: Medium, and L: Low)
III. UNIMODAL BIOMETRICS LIMITATIONS
Any single modal biometric has limitations. For example, iris recognition suffers from some problems like
camera distance, eyelids and eyelashes occlusion, lenses, and reflections. Face changes overages and unstable,
and twins may have similar face features. Also, fake faces from mobiles as example, and masks used to attack
the system. Fingerprint may have some cuts, burns, and small injuries temporary or permanent. Moreover, fake
fingers made from gelatin and/or silicon has ability to attack the fingerprint-based recognition system. Cold
leads to voice problems and tape recordings may be used to hack the system. The fingerprint of DNA needs
several hours to be obtained. Besides, DNA includes sensitive information related to genetic of individuals and
the test is quite expensive to perform. Hand geometry is not distinctive enough to be applied to a large
population. Thus, it is not suitable for purpose of identification. Gait is sensitive to body weight and not stable;
it is not used for large population and not reliable enough. Signature is not universal and changes with time.
Offline ones are forgery while, Online signature cannot applied for documents verification (i.e. Government
documents and bank cheques). None of above traits alone can ensure perfect recognition performance.
Nevertheless, the biometric system (either an 'identification' system or a 'verification' system) can also be
attacked by the outsider or unauthorized person at various points. Combining multiple modalities is a good idea
to decrease these conditions.
The unimodal biometric rely on the evident single source of information for authentication (e.g., single
fingerprint, face). Single modal biometric traits may not achieve the desired performance requirements; as they
have plenty of error rates. These systems have to contend with a variety of problems such as:
 Noise in sensed data: Defective or improperly maintained sensors (i.e. accumulation of dirt on a
fingerprint sensor) could produce deformed and noisy data. For instance, a cold has effects on the
voice, wearing glasses alters iris recognition performance, variations in light or illumination in face
sensed,etc
 Distinctiveness (Intra-class variations and Inter-class similarities): Biometric trait is expected to be
varied significantly across two persons. Intra-class variations occur when a user interacts with the

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
sensor incorrectly (e.g., incorrect facial pose). Also, characteristics of the individuals are formed with
the large inter-class similarity (overlap) in the feature sets of multiple users.
 Non-universality: The non-ability of the biometric to acquire meaningful biometric data from a group
of users due to the poor quality and consistency of the acquired biometric data as a result to error or a
fault in the sensor. For example, many of population (about 4%) may have scars or cuts in fingerprints.
As a result, a fingerprint biometric system, may extract incorrect minutiae features from them. Also,
user-sensor interaction is adjustment incorrectly. Of course, this may give undesired matching result.
 Spoof attacks: A fake traits or biometrics of the authorized user are enrolled and saved in the template
database; an imposter person may attempt to spoof these sensed data when the traits are used. As in,
artificial fingers/fingerprint can be used to spoof the verification system. This type of attack is
common when using behavioral characteristics.
On behave of above problems, unimodal biometric systems suffer other drawbacks like: insufficient
population coverage, lack of individuality, lack of invariant representation, and susceptibility to circumvention.
These problems lead to higher False Reject Rate (FRR) and False Accept Rate (FAR).
IV. MULTI-BIOMETRICS AS A SOLUTION
Biometric fusion has a history of more than 30 years. More than one biometric combined to investigate high
performance multi-biometric recognition system. Multi-biometrics has addressed some issues related to
unimodal this make it has some benefits over unimodal biometrics such as recognition accuracy, privacy, and
biometric data enrollment.
 Recognition Accuracy: Its accuracy is better as compared to the unimodal biometric system. The multi-
biometric system is expected to be more accuracy and reliability due to the multiple, biometric traits
independency and difficult to forge all of them. As the combination of each of the biometric identifiers
offers some additional evidence about the authenticity of an identity claim, one can have more
confidence in the result. For example, two persons may have the similar signature patterns, in which
case, the signature verification system will produce large FAR for that system. Addition of face
recognition system with the signature verification system may solve the problem and reduce the FAR.
Experiments have shown that the accuracy of multimodality can reach near 100% in identification.
 Privacy: Multimodal biometric systems increase resistance to certain type of vulnerabilities. It prevents
from stolen the templates of biometric system as at the time it stores the two characteristics of
biometric system in the database. For example, it would be more challenge for attacker to spoof many
different biometric identifiers. Further, when two or more modalities are used for authentication, it
leads to become not easy to spoof the biometric system.
 Biometric Data Enrollment: Multimodal biometric systems can address the problem of non-
universality. In case of unavailability or poor quality of a particular biometric data, other biometric
identifier of the multimodal biometric system can be used to capture data. For example, a face
biometric identifier can be used in a multimodal system (involves fingerprint of general labors with lots
of scars in the hand). It makes better system operation. Multi-biometric system also addresses the
problem of noisy data effectively (i.e. illness affecting voice, scar affecting fingerprint). They allow
indexing or filtering of large biometric databases, and are robust to noise. Thus, it provides universal
coverage and improves matching accuracy.
A. Multimodal Categories
Multi-biometric systems have two basic categories: synchronous and asynchronous. In synchronous, two or
more biometrics combined within a single authorization process. On the other hand, asynchronous system uses
two biometric technologies in sequence (one after the other). Multimodal biometric systems can operate in three
different modes:
 Serial Mode (cascade mode): Each modality is examined before the next modality is investigated. The
overall recognition duration can be decreased, as the total number of possible identities - before using
the next modality - could be reduced

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
 Parallel Mode: Sensed/captured data from multiple modalities are used in concurrent way to perform
recognition. Then the results are combined to make final decision.
 Hierarchical Mode: Individual classifiers are combined in a hierarchy -tree like- structure. This mode is
preferred when a large number of classifiers are expected.
B. Multi-Biometrics Integration Scenarios
Recognition systems using multiple biometric traits are designed to operate in one of the integration
scenarios as below:
1) Multi-sensor systems: The information of the same biometric obtained from different sensors are combined
for all. For example, complementary information corresponding to fingerprints can be acquired using different
types of sensors (like optical and capacitive sensors). Information obtained is then integrated using sensor level
fusion technique.
2) Multi-modal systems: More than one biometric trait is used for user identification. For example, the
information obtained using face and voice features or other can be integrated to establish the identity of the user.
This can be more costly; because it requires multiple sensors with each sensor sensing different biometric
characteristics. But, the improvement in performance is substantial.
3) Multi-instance systems: Multiple instances of a single biometric trait are captured. For example, images of
the left and right irises can be used for iris recognition. Also, fingerprints from two or more fingers of a person
may be combined or one image each of the same person may be combined. If a single senor is used to acquire
these images in a sequential manner, the system can be made really cost effective, as it does not require multiple
sensors. Moreover, it does not incorporate additional feature extraction and matching modules.
4) Multi-sample systems: Multiple samples of a same biometric trait are used for the enrollment and recognition.
For example, along with the frontal face, the left and right profiles are also captured. Multiple impression of the
same finger and multiple samples of a voice can be combined. Multiple samples may overcome poor
performance. But, it requires multiple copies of sensors, or the user may wait a longer period of time to be
sensed or a combination of both.
5) Multi-algorithm systems: Multiple different approaches to feature extraction and matching algorithms are
applied to a single biometric trait. Final decision obtained if any of the matching fusion technique can be applied
on the results obtained using different matching algorithms. These systems are more economical as no extra
device is required to capture the data. But, these are more complex because of application of different
algorithms.
6) Hybrid systems: It is a system which integrates more than one of the above mentioned multi-biometric
systems. For example, two face recognition algorithms can be combined with two fingerprint recognition
algorithms. Such a system will be multi-modal and multi-algorithmic system. Moreover, if multiple sensors are
used to obtain these images, then it will be multi-sensory, and if multiple instance of the finger is used, it will be
multi-instance system also.
Both of hybrid systems and multi-modal systems can be desired by using multiple modalities.
However, the rest can be achieved with the only help of even single modality. The different types of multi-
biometric are shown in figure (1).
C. Limitation of Multi-biometrics System
Some lacks are still found such as noise in the biometrics like scratches in the fingerprint and lens mark
in iris, this will lead to increase the (FRR). Moreover, the accuracy of the multi-biometric enrollment and multi-
biometric identification need to be improved. In multi-biometrics, failure of one biometrics will make the whole
system to fail. In addition, multimodal biometric systems may be more expensive and complicated due to the
requirement of additional hardware and matching algorithms, and there is a greater demand for computational
poser and storage. Recent research has revealed that multi-biometric systems can increase the security level as a
means to enhance network security to people who are encouraged to use biometric systems in this field.
However, it need more efforts and research to face some types of attacks such as: spoof attack, replay attack,
substitution attack, Trojan horse attack, transmission attack, template database attack, and decision attack. Next
section will list the performance metrics that distinguish between the multi-biometrics techniques.

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
Figure 1: The different types of multi-biometric system
V. QUALITY PERFORMANCE AND METRICS
Various quality performance metrics measure the performance of any biometric authentication techniques.
It helps comparing systems and motivating the progress. The most common performance metrics of biometric
systems are described below:
 False Accept Rate (FAR) or (False Match Rate (FMR)): Mistaking the biometric measurements from
two different persons to appear as if they are from the same person due to large inter-user similarity. It
measures the percent of invalid matches. The FAR is defined as in (1):
FAR% = TFaccept / TFsubmit * 100-----------------------------1
Where, TFaccept is total number of forgeries accepted and TFsubmit is total number of forgeries submitted to the
system test. In a good authentication system this rate must be low.
 False Reject Rate (FRR) or (False Non-Match Rate (FNMR)): Mistaking two biometric measurements
from the same person to appear that they are from two different persons due to large intra-class
variations. It measures the percent of valid inputs being rejected. The FRR is defined as in (2):
FRR% = TGreject / TGsubmit * 100------------------------------2
Where TGreject is the total number of genuine test pattern rejected, and TGsubmit is total number of genuine
test submitted to the system. This must be low to achieve good Performance. The average of the FRR and FAR
is called the Average Error Rate (AER). Genuine Acceptance Rate (GAR) sometimes used, which is the
percentage of the likelihood that a genuine individual is recognized as a match. GAR of a valid user can be
obtained by equation (3).
GAR % = 1 - FRR% -----------------------------------------3
 Equal Error Rate (EER): For a simple empirical measure, it is used to summarize the performance of a
biometric system that is defined at the point where False Reject Rate (FRR) and False Accept Rate
(FAR) are equal. System with the lower EER is the more accurate and precise. The EER is also called
the type (III) error.
 Failure to Capture (FTC): denotes the percentage of times the biometric device fails to automatically
capture a biometric characteristic when presented correctly. This usually happens when system deals
with a signal of insufficient quality.
 Failure to Enroll Rate (FER or FTE): denotes the percentage of times users cannot enroll in the
recognition system. Data input is considered invalid due to poor quality.

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
 Template Capacity: It is the maximum number of sets of data which can be input in to the system.
Usually, the above performance metrics are expressed using different graphs such as Receiver Operating
Characteristic (ROC), Score Histogram (SH), and Cumulative Match Characteristic (CMC).
 Receiver Operating Characteristic (ROC) curve: There is a trade-off between FAR and FRR in every
biometric system. In fact, both of them are functions of the system threshold (t); if it is declined to
make the system achieves higher tolerance to input variations and noise, then FAR increases. On the
other hand, if it is raised to make the system more secure, then FRR increases accordingly. The ROC
plot is obtained by graphing the values of FAR against FRR, at various operating points (thresholds) on
a linear or logarithmic or semi-logarithmic curve. Detection Error Trade off (DET) is a common
variation, which is obtained via normal deviate scales on both axes. This graph is more linear that
illuminates the differences for higher performances.
 Cumulative Match Characteristic (CMC) curve: is used in biometric identification to summarize the
identification rate at different rank values [8]. Score Histogram (SH): plots the frequency of the scores
for matches and non-matches over the match score range. These metrics are needed to differentiate
between each level fusion and method considered for the multi-biometrics as a solution. Categorization
of different levels of fusion will be discussed in next section.
VI. LEVELS OF FUSION IN MULTIMODAL BIOMETRICS
Multimodal biometric fusion combines the distinguished aspect from different biometric features to support
the advantages and reduce the drawbacks of the individual aspects. The fundamental issue of information fusion
is to determine the type of information that should be fused and the selection of method for fusion. The goal of
fusion is to devise an appropriate function that can optimally combine the information rendered by the biometric
subsystems.
In multimodal biometrics, the fusion scheme can be classified as sensor level, feature level, match score
level, rank level, and decision level. The process can be subdivided into two main categories: prior-to-matching
fusion and after matching fusion. Figure (2), shows these fusion levels possibilities at each module. The hybrid
one is mixing two or more from these level fusions.
Figure 2: Prior-to-matching and after matching fusion levels related to biometric system modules
A. Prior to Matching Fusion
Fusion in this category integrates evidences before matching. This can be classified into two different
categories as follows:

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
1) Sensor level fusion
Principles- A new biometric data generated by merging the raw data obtained from multiple sources.
Then, trait can be extracted. A single sensor or different compatible sensors like fingerprint, iris scanner, etc.,
represents the samples of the single biometric trait sensed. This level of fusion is also known as data level fusion
or image level fusion (for image based biometrics).
2) Feature level fusion
Principles- The correlated feature sets extracted from different biometric channels (modalities) can be
fused by using specific fusion algorithm forming a composite feature set, passed to the matching module. This is
done after normalization, transformation and reduction schemes. The goal of feature normalization is to modify
the location (mean) and the scale (variance) of the feature value via a transform function in order to map them
into a common domain. (e.g. Min-max normalization, Median normalization...etc.). Transformation or Feature
Selection is algorithm use to reduce the dimensionality of the feature set. (e. g. Sequential forward selection,
Sequential backward selection, Principal Component Analysis (PCA), etc.).
B. After Matching Fusion
Prior to matching fusions sometime don’t involve multiple modalities. Also, the fusion of data set is
more complex, and it is not good to ignore any data. After matching fusion integrates evidences of after
matching module. This can be classified into three different categories:
1) Matching score level fusion
Principles- Individually, Extracted feature vectors (generated separately for each modality) is compared
with the templates enrolled in the database for each biometric trait in order to generate the match scores. Output
set of match scores are fused to create composite matching score (single scalar score). This fusion technique is
also known as confidence level or measurement level fusion. Density, transformation, and classifier based score
fusion are different methods to achieve this fusion level.
The matching scores cannot be used or combined directly; because these scores are from different modalities
and based on different scaling methods. Score normalization is required, by converting the scores into common
similar domain or scale. This can be carried out with different methods- piecewise linear normalization - new
normalization technique. Their experiments used palm print and facial features.
2) Rank level fusion
Principles- In this new fusion approach, each classifier associates a rank with each enrolled trait to the
system (a higher rank indicating a good match). It consolidates multiple unimodal biometric matcher outputs,
and determining a new rank that would help in estimating the final decision. Generally, the rank level fusion is
adopted for the identification rather than verification. Here, the working procedures are: first, generate a rank of
identities sorted with all modalities. Second, by help of any method of fusion, the ranking for each individual
available for different modalities fused. Finally, the identity with the lowest score is the correct identified one.
3) Decision level fusion
Principles- The final decision - in multimodal biometric systems - is formed from obtaining
individually separate decision of different biometric modalities using different techniques include behavior
knowledge space, majority voting, weighted voting, AND rule, and OR rule. Decision level fusion is also named
abstract level fusion; because it is used when there is access to only decisions from individual.
Majority voting approach is the mostly used for decision level fusion. The input sample with agreed in majority
of matchers is given the identity. AND/OR rules are rarely used; because they combine two different matchers,
so this sometimes degrade of performance of the system. AND combination improves the FAR while, OR
combination improves the FRR. The main advantage of the majority voting method is that it does not require
prior knowledge about the matcher, and it requires no training for final decision making too.
C. Hybrid Level Fusion
Tri-level fusion scenarios (different fusion in different levels of the system) can be investigated to make the
system faster and significantly reduce the error rate. The fusion of level increased the performance. In 2007, C.
Lupu fused fingerprint, voice and iris. Next year 2008, S. Asha combined fingerprint with mouse dynamics. In

National Conference on Emerging Trends and Advances in Information Technology
Department of CS & E, AIT, Chikkamagalur, January 29th & 30th, 2016
©2015 ISRASE www.israse.com/digital ISRASE eXplore Digital Library
2011, Parallel Feature Extraction with the help of SIFT, SIMD, and HMA techniques was used by Anukul
Chandra Panda to fuse multiple iris. Next in 2013, Gandhimathi Amirthalingam, and Radhamani. G. used fuzzy
vault to implement multimodal system based on Face and ear traits.
VII. DESIGN AND IMPLEMENTATION OF MULTI-BIOMETRICS RECOGNITION
TRADE-OFFS
Generally, any biometric recognition system architecture is related to software-based techniques and
hardware-based techniques. The obstacle here is to satisfy all challenges requirement such as: user friendly, fast
(i.e. the system must identify individuals in real time), low cost, high performance, less intrusive, fraud prevents
and high fake detection rate. Briefly, design issues in multi-biometrics include:
 Choosing the biometric modalities and number of traits (defining and estimation of each modality
reliability is still open research issue).
 Choosing the best samples for a particular biometric.
 Fusion level and fusion methodology.
 Fusion scenario and common strategy.
 Learning weights of individual biometric for users.
 Cost versus performance and accuracy versus reliability trade-offs.
 Verification and/or identification system for application.
 Expert features selection difficulties.
In order to optimize the multi-biometric recognition benefits, the issues of system design firstly should be
understood better; so the more effective design methodology and system architecture can be developed. For
instance, to decide whether combining multiple biometrics or combining multiple samples of the same trait is
better, to achieve economic system. In addition, privacy issues should be considered, and compromising
between accuracy and coverage.
VIII. CONCLUSION
Multi-biometrics topic has attracted more interest in recent research. It is used to identify individuals based
on their physiological and behavioral characteristics for security purposes. Overview of biometrics showed that
it is impossible to find the best single biometric suitable for all applications, populations, technologies and
administration policies. Also, integration of biometric modalities can solve unimodal system limitations to
achieve higher performance.
REFERENCES
[1]. (IJACSA) International Journal of Advanced Computer Science and Applications Vol. 6, No. 6, 2015
[2]. F. Karray, J. A. Saleh, M. N. Arab, and M. Alemzadeh, "Multi Modal Biometric Systems: A State of the Art Survey," Pattern
Analysis and Machine Intelligence Laboratory, University of Waterloo, Waterloo, Canada, no. 2007.
[3]. Shweta Malhotra & Dr. Chander Kant ―A Novel approach for Securing Biometric Template ‖ IJARCSSE , Vol. 3 , Issue 5 ,pp
397- 403, May 2013.
[4]. Debnath bhattacharya ― Biometric Authentiction :A Review ‖ Vol. 2 , Issue 3 , Sep 7 2009 , pp 13-26.
[5]. Harpreet Saini & Kanwal Garg ―Comparitive Analysis of Various Biometric Techniques for database Security ‖ IJSR
(International journal of Science and Research ) Vol. 2 Issue 4, pp150-153, April 2013.
[6]. Sulochana Sonkamble & Dr. Ravindra Thool ―Survey of Biometric Recognition Systems And Their Applications‖ Journal of
Theoretical And Applied Information Technology (JATIT) , pp 45- 51, 2005