Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
NITTTR, Chandigarh EDIT -2015 176
An Approach to Speech and Iris based
Multimodal Biometric System
1
SakshiSahore, 2
TanviSood
1
M.Tech Student, 2
Assistant Professor
1,2
ECE Department, Chandigarh Engineering College, Ladran, Mohali
1
sakshisahore@gmail.com, 2
cecm.ece.ts@gmail.com
Abstract—Biometrics is the science and technology of
human identification and verification through the use of
feature set extracted from the biological data of the
individual to be recognized. Unimodal and Multimodal
systems are the two modal systems which have been
developed so far. Unimodal biometric systems use a
single biometric trait but they face limitations in the
system performance due to the presence of noise in data,
interclass variations and spoof attacks. These problems
can be resolved by using multimodal biometrics which
rely on more than one biometric information to produce
better recognition results. This paper presents an
overview of the multimodal biometrics, various fusion
levels used in them and suggests the use of iris and
speech using score level fusion for a multimodal
biometric system.
Keywords—Biometric, unimodal, multimodal,
recognition, score level fusion
I. INTRODUCTION
With the recent advancement in technology and
development of electrically interconnected society, there is
an essential requirement of accurate personal
authentication system to handle various person
authentication issues in daily life. There are several
authentication systems that we use on daily basis such as
personal identification number (PIN), smartcards and
passwords. These systems are possession based and
knowledge based and can easily be misplaced, forgotten or
forged [1]. To overcome these difficulties, biometric
systems for authentication are introduced. Biometrics is a
robustious approach for the recognition of a person [2].
Biometrics verify the identity of the subject based on a
feature set extracted from the subject’s biological
characteristics.Biometric characteristics are of two types:
 Physiological: The characteristics related to the body
of a person are called physiological characteristics.
Fingerprints, face, iris, palm geometry, DNA are the
examples of the physiological characteristics. These
characteristics do not change over time.
 Behavioral: The characteristics related to the behavior
of a person are called behavioral characteristic. Voice,
gait, signature and keystroke are the examples of
behavioral characteristics. These are variant in nature.
A biometric system consists of two modes that are
enrollment mode and authentication mode. In enrollment
mode, the biometric data of the subject is taken and
processed for feature extraction. These features are used
for the generation of template of that subject in which all
the feature variations are captured and stored in a database.
During authentication mode, the features from the subject
to be identified are computed and then compared with the
stored template in the database. If the features match, the
subject is recognized. Figure 1 shows a typical biometric
system.
Biometric based person recognition system [3]
II. MODAL SYSTEM
A. Unimodal Biometrics
A unimodal biometric system uses a single source of
biometric information to generate the recognition result.
Most of the deployed real world applications in biometrics
are unimodal, that is, they use a single biometric trait for
authentication such as a biometric system based on
fingerprints [4]. While unimodal biometric systems have
successfully been installed in various applications, but
unimodal biometrics is still not fully solved problem [5].
These systems a variety of issues like
 Noisy data – The input biometric data might be noisy
or the biometric sensors might be susceptible to noise
which may lead to inaccurate matching and hence
false rejection.
 Intra-class variations – This occurs when the biometric
data acquired from an individual during verification is
not identical to the data stored in the template during
enrollment. This occurs due to incorrect interaction of
the individual with the sensor.
 Non-universality – Sometimes it is possible that
certain individuals may not provide a particular
biometric causing failure to enroll (FTE).
 Spoof attack –Unimodal biometrics are susceptible to
spoof attacks where an imposter may attempt to fake
Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
NITTTR, Chandigarh EDIT -2015 176
An Approach to Speech and Iris based
Multimodal Biometric System
1
SakshiSahore, 2
TanviSood
1
M.Tech Student, 2
Assistant Professor
1,2
ECE Department, Chandigarh Engineering College, Ladran, Mohali
1
sakshisahore@gmail.com, 2
cecm.ece.ts@gmail.com
Abstract—Biometrics is the science and technology of
human identification and verification through the use of
feature set extracted from the biological data of the
individual to be recognized. Unimodal and Multimodal
systems are the two modal systems which have been
developed so far. Unimodal biometric systems use a
single biometric trait but they face limitations in the
system performance due to the presence of noise in data,
interclass variations and spoof attacks. These problems
can be resolved by using multimodal biometrics which
rely on more than one biometric information to produce
better recognition results. This paper presents an
overview of the multimodal biometrics, various fusion
levels used in them and suggests the use of iris and
speech using score level fusion for a multimodal
biometric system.
Keywords—Biometric, unimodal, multimodal,
recognition, score level fusion
I. INTRODUCTION
With the recent advancement in technology and
development of electrically interconnected society, there is
an essential requirement of accurate personal
authentication system to handle various person
authentication issues in daily life. There are several
authentication systems that we use on daily basis such as
personal identification number (PIN), smartcards and
passwords. These systems are possession based and
knowledge based and can easily be misplaced, forgotten or
forged [1]. To overcome these difficulties, biometric
systems for authentication are introduced. Biometrics is a
robustious approach for the recognition of a person [2].
Biometrics verify the identity of the subject based on a
feature set extracted from the subject’s biological
characteristics.Biometric characteristics are of two types:
 Physiological: The characteristics related to the body
of a person are called physiological characteristics.
Fingerprints, face, iris, palm geometry, DNA are the
examples of the physiological characteristics. These
characteristics do not change over time.
 Behavioral: The characteristics related to the behavior
of a person are called behavioral characteristic. Voice,
gait, signature and keystroke are the examples of
behavioral characteristics. These are variant in nature.
A biometric system consists of two modes that are
enrollment mode and authentication mode. In enrollment
mode, the biometric data of the subject is taken and
processed for feature extraction. These features are used
for the generation of template of that subject in which all
the feature variations are captured and stored in a database.
During authentication mode, the features from the subject
to be identified are computed and then compared with the
stored template in the database. If the features match, the
subject is recognized. Figure 1 shows a typical biometric
system.
Biometric based person recognition system [3]
II. MODAL SYSTEM
A. Unimodal Biometrics
A unimodal biometric system uses a single source of
biometric information to generate the recognition result.
Most of the deployed real world applications in biometrics
are unimodal, that is, they use a single biometric trait for
authentication such as a biometric system based on
fingerprints [4]. While unimodal biometric systems have
successfully been installed in various applications, but
unimodal biometrics is still not fully solved problem [5].
These systems a variety of issues like
 Noisy data – The input biometric data might be noisy
or the biometric sensors might be susceptible to noise
which may lead to inaccurate matching and hence
false rejection.
 Intra-class variations – This occurs when the biometric
data acquired from an individual during verification is
not identical to the data stored in the template during
enrollment. This occurs due to incorrect interaction of
the individual with the sensor.
 Non-universality – Sometimes it is possible that
certain individuals may not provide a particular
biometric causing failure to enroll (FTE).
 Spoof attack –Unimodal biometrics are susceptible to
spoof attacks where an imposter may attempt to fake
Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
NITTTR, Chandigarh EDIT -2015 176
An Approach to Speech and Iris based
Multimodal Biometric System
1
SakshiSahore, 2
TanviSood
1
M.Tech Student, 2
Assistant Professor
1,2
ECE Department, Chandigarh Engineering College, Ladran, Mohali
1
sakshisahore@gmail.com, 2
cecm.ece.ts@gmail.com
Abstract—Biometrics is the science and technology of
human identification and verification through the use of
feature set extracted from the biological data of the
individual to be recognized. Unimodal and Multimodal
systems are the two modal systems which have been
developed so far. Unimodal biometric systems use a
single biometric trait but they face limitations in the
system performance due to the presence of noise in data,
interclass variations and spoof attacks. These problems
can be resolved by using multimodal biometrics which
rely on more than one biometric information to produce
better recognition results. This paper presents an
overview of the multimodal biometrics, various fusion
levels used in them and suggests the use of iris and
speech using score level fusion for a multimodal
biometric system.
Keywords—Biometric, unimodal, multimodal,
recognition, score level fusion
I. INTRODUCTION
With the recent advancement in technology and
development of electrically interconnected society, there is
an essential requirement of accurate personal
authentication system to handle various person
authentication issues in daily life. There are several
authentication systems that we use on daily basis such as
personal identification number (PIN), smartcards and
passwords. These systems are possession based and
knowledge based and can easily be misplaced, forgotten or
forged [1]. To overcome these difficulties, biometric
systems for authentication are introduced. Biometrics is a
robustious approach for the recognition of a person [2].
Biometrics verify the identity of the subject based on a
feature set extracted from the subject’s biological
characteristics.Biometric characteristics are of two types:
 Physiological: The characteristics related to the body
of a person are called physiological characteristics.
Fingerprints, face, iris, palm geometry, DNA are the
examples of the physiological characteristics. These
characteristics do not change over time.
 Behavioral: The characteristics related to the behavior
of a person are called behavioral characteristic. Voice,
gait, signature and keystroke are the examples of
behavioral characteristics. These are variant in nature.
A biometric system consists of two modes that are
enrollment mode and authentication mode. In enrollment
mode, the biometric data of the subject is taken and
processed for feature extraction. These features are used
for the generation of template of that subject in which all
the feature variations are captured and stored in a database.
During authentication mode, the features from the subject
to be identified are computed and then compared with the
stored template in the database. If the features match, the
subject is recognized. Figure 1 shows a typical biometric
system.
Biometric based person recognition system [3]
II. MODAL SYSTEM
A. Unimodal Biometrics
A unimodal biometric system uses a single source of
biometric information to generate the recognition result.
Most of the deployed real world applications in biometrics
are unimodal, that is, they use a single biometric trait for
authentication such as a biometric system based on
fingerprints [4]. While unimodal biometric systems have
successfully been installed in various applications, but
unimodal biometrics is still not fully solved problem [5].
These systems a variety of issues like
 Noisy data – The input biometric data might be noisy
or the biometric sensors might be susceptible to noise
which may lead to inaccurate matching and hence
false rejection.
 Intra-class variations – This occurs when the biometric
data acquired from an individual during verification is
not identical to the data stored in the template during
enrollment. This occurs due to incorrect interaction of
the individual with the sensor.
 Non-universality – Sometimes it is possible that
certain individuals may not provide a particular
biometric causing failure to enroll (FTE).
 Spoof attack –Unimodal biometrics are susceptible to
spoof attacks where an imposter may attempt to fake
Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
177 NITTTR, Chandigarh EDIT-2015
the biometric trait of an enrolled user in order to
bypass the system.
The problems imposed in the unimodal biometrics limit the
accuracy and system performance.
B. Multimodal Biometrics
A multimodal biometric system uses more than one
source of biological data to generate the recognition result,
for example a multimodal biometric system using iris and
ear [6]. Multimodal biometrics overcome the shortcomings
of unimodal biometrics [7] and is more reliable because of
the use of more than one biometric trait and hence more
pieces of information.
Multimodal scenarios: A multimodal biometric system
can be designed to work in one of the following scnarios.
 Multiple sensors: Highlight all author and
affiliation lines.The informationof the same biometric can
be acquiredby different sensors [8].The different samples
are thenprocessed by the same algorithm and the resultsare
fused to get the resultant algorithm.
 Multiple instances: The biometric information is
extracted from the multiple instances of the same biometric
[9].
 Multiple algorithms: More than one
approach/algorithm is used for feature extraction or
classification of the same biometric to improve the system
performance [10].
 Multiple biometric:Evidence from the multiple
biometric characteristics is taken [11].
 Multiple samples:Multiple samples are acquired
from the same biometric by a single sensor and processed
by the same algorithm to obtain the recognition results
[12].
Multimodal Biometric System [13]
Fusion Levels in Multimodal Biometrics: While
designing a multimodal system, different fusion strategies
can be used to integrate the biometric data.
Fusion Levels in Multimodal Biometrics
 Sensor level fusion:Highlight all author and
affiliation lines. Sensor level fusion is done in a system
system using multiple sensors or in a system using a single
sensor at multiple instances. In this, the biometric data
obtained by the sensor is combined.
 Feature level fusion:Feature level fusion is done
by extracting the features of different biometric sources
individually and then combining those features into a
single feature vector [14].
 Score level fusion:Score level fusion is performed
by individually processing (sensing and extracting
features) different biometric sources and finding their
match scores. These scores are then combined to make
classification [15].
 Decision level fusion:After each biometric source
is processed and recognition decision is made for each
biometric data, fusion is executed at the decision level
[16].
III. SCORE LEVEL FUSION
Score level fusion is the most popular and common
approach in the multimodal biometrics system due its
simple procedure. Matching scores contain rich
information about the input pattern. Each classifiers
provides a matching scores and scores of different
classifiers are combined to produce the final score.
A. Fusion algorithms
When different matching scores of different biometrics are
acquired, their fusion is done. For the fusion of the
matching scores, different algorithms can be applied.
These algorithms include product rule, sum rule, max rule
and min rule.
Consider ( ⃗) as the output of individual classifiers,
as a feature vector to ith classifier, R as the number of
different classifiers and be the output. The different rules
can be applied as
BIOMETRIC
FUSION
BEFORE
MATCHIG
AFTER
MATCHING
SCORE
LEVEL
FEATURE
LEVEL
SCORE
LEVEL
DECISIO
N LEVEL
Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
177 NITTTR, Chandigarh EDIT-2015
the biometric trait of an enrolled user in order to
bypass the system.
The problems imposed in the unimodal biometrics limit the
accuracy and system performance.
B. Multimodal Biometrics
A multimodal biometric system uses more than one
source of biological data to generate the recognition result,
for example a multimodal biometric system using iris and
ear [6]. Multimodal biometrics overcome the shortcomings
of unimodal biometrics [7] and is more reliable because of
the use of more than one biometric trait and hence more
pieces of information.
Multimodal scenarios: A multimodal biometric system
can be designed to work in one of the following scnarios.
 Multiple sensors: Highlight all author and
affiliation lines.The informationof the same biometric can
be acquiredby different sensors [8].The different samples
are thenprocessed by the same algorithm and the resultsare
fused to get the resultant algorithm.
 Multiple instances: The biometric information is
extracted from the multiple instances of the same biometric
[9].
 Multiple algorithms: More than one
approach/algorithm is used for feature extraction or
classification of the same biometric to improve the system
performance [10].
 Multiple biometric:Evidence from the multiple
biometric characteristics is taken [11].
 Multiple samples:Multiple samples are acquired
from the same biometric by a single sensor and processed
by the same algorithm to obtain the recognition results
[12].
Multimodal Biometric System [13]
Fusion Levels in Multimodal Biometrics: While
designing a multimodal system, different fusion strategies
can be used to integrate the biometric data.
Fusion Levels in Multimodal Biometrics
 Sensor level fusion:Highlight all author and
affiliation lines. Sensor level fusion is done in a system
system using multiple sensors or in a system using a single
sensor at multiple instances. In this, the biometric data
obtained by the sensor is combined.
 Feature level fusion:Feature level fusion is done
by extracting the features of different biometric sources
individually and then combining those features into a
single feature vector [14].
 Score level fusion:Score level fusion is performed
by individually processing (sensing and extracting
features) different biometric sources and finding their
match scores. These scores are then combined to make
classification [15].
 Decision level fusion:After each biometric source
is processed and recognition decision is made for each
biometric data, fusion is executed at the decision level
[16].
III. SCORE LEVEL FUSION
Score level fusion is the most popular and common
approach in the multimodal biometrics system due its
simple procedure. Matching scores contain rich
information about the input pattern. Each classifiers
provides a matching scores and scores of different
classifiers are combined to produce the final score.
A. Fusion algorithms
When different matching scores of different biometrics are
acquired, their fusion is done. For the fusion of the
matching scores, different algorithms can be applied.
These algorithms include product rule, sum rule, max rule
and min rule.
Consider ( ⃗) as the output of individual classifiers,
as a feature vector to ith classifier, R as the number of
different classifiers and be the output. The different rules
can be applied as
BIOMETRIC
FUSION
BEFORE
MATCHIG
AFTER
MATCHING
SCORE
LEVEL
FEATURE
LEVEL
SCORE
LEVEL
DECISIO
N LEVEL
Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
177 NITTTR, Chandigarh EDIT-2015
the biometric trait of an enrolled user in order to
bypass the system.
The problems imposed in the unimodal biometrics limit the
accuracy and system performance.
B. Multimodal Biometrics
A multimodal biometric system uses more than one
source of biological data to generate the recognition result,
for example a multimodal biometric system using iris and
ear [6]. Multimodal biometrics overcome the shortcomings
of unimodal biometrics [7] and is more reliable because of
the use of more than one biometric trait and hence more
pieces of information.
Multimodal scenarios: A multimodal biometric system
can be designed to work in one of the following scnarios.
 Multiple sensors: Highlight all author and
affiliation lines.The informationof the same biometric can
be acquiredby different sensors [8].The different samples
are thenprocessed by the same algorithm and the resultsare
fused to get the resultant algorithm.
 Multiple instances: The biometric information is
extracted from the multiple instances of the same biometric
[9].
 Multiple algorithms: More than one
approach/algorithm is used for feature extraction or
classification of the same biometric to improve the system
performance [10].
 Multiple biometric:Evidence from the multiple
biometric characteristics is taken [11].
 Multiple samples:Multiple samples are acquired
from the same biometric by a single sensor and processed
by the same algorithm to obtain the recognition results
[12].
Multimodal Biometric System [13]
Fusion Levels in Multimodal Biometrics: While
designing a multimodal system, different fusion strategies
can be used to integrate the biometric data.
Fusion Levels in Multimodal Biometrics
 Sensor level fusion:Highlight all author and
affiliation lines. Sensor level fusion is done in a system
system using multiple sensors or in a system using a single
sensor at multiple instances. In this, the biometric data
obtained by the sensor is combined.
 Feature level fusion:Feature level fusion is done
by extracting the features of different biometric sources
individually and then combining those features into a
single feature vector [14].
 Score level fusion:Score level fusion is performed
by individually processing (sensing and extracting
features) different biometric sources and finding their
match scores. These scores are then combined to make
classification [15].
 Decision level fusion:After each biometric source
is processed and recognition decision is made for each
biometric data, fusion is executed at the decision level
[16].
III. SCORE LEVEL FUSION
Score level fusion is the most popular and common
approach in the multimodal biometrics system due its
simple procedure. Matching scores contain rich
information about the input pattern. Each classifiers
provides a matching scores and scores of different
classifiers are combined to produce the final score.
A. Fusion algorithms
When different matching scores of different biometrics are
acquired, their fusion is done. For the fusion of the
matching scores, different algorithms can be applied.
These algorithms include product rule, sum rule, max rule
and min rule.
Consider ( ⃗) as the output of individual classifiers,
as a feature vector to ith classifier, R as the number of
different classifiers and be the output. The different rules
can be applied as
BIOMETRIC
FUSION
BEFORE
MATCHIG
AFTER
MATCHING
SCORE
LEVEL
FEATURE
LEVEL
SCORE
LEVEL
DECISIO
N LEVEL
Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
NITTTR, Chandigarh EDIT -2015 178
Score Level Fusion in Multimodal Biometric System
Product rule:Different biometric traits of an individual
(such as iris, fingerprints and ear) are mutually
independent and the product rule is applied based on this.
= ( ⃗)
Sum rule: The sum rule takes the scores of the
individual classifiers to simply calculate their sum.
= ( ⃗)
Max rule:The max rule approximates the output by the
maximum value of the scores.
= max ( ⃗)
Min rule:The max rule approximates the output by the
minimum value of the scores.
= min ( ⃗)
B. Normalization
Normalization is done after determining the matching
scores from different biometrics. Score normalization is
essential because the matching scores of different
biometrics are obtained from different algorithms and
hence may not have the same underlying properties, that is,
they may be of different nature and scale. Normalization
changes the scale of the different scores and brings them to
a common domain. After normalization, the scores are
combined. The most common normalization algorithms
used are
If S = ( , , ,… ,……. ) is a vector of M scores, then
the normalization score, will be
Min max normalization:
=
µ( )
( ) ( )
Where max(s) = maximum value of raw score and
min(s) = minimum value of raw score
Z-score normalization
=
µ( )
( )
Where µ(s) = mean deviation of set of score vectors
and σ(s) = standard deviation of the set of score vectors
Median-MAD normalization
=
Where MAD = median (| − |)
Tanh normalization
= 0.5 ℎ 0.01
− µ( )
( )
+ 1
Where µ( )= mean deviation calculated from scores
and ( ) = standard deviation calculated from scores.
IV. MULIIMODAL BIOMETRIC USING IRIS AND
SPEECH
As mentioned earlier, there are two types of biometric
characteristics in human beings, physiological
characteristics and behavioral characteristics. With the
selection of appropriate modals and fusion scheme,
optimal results can be achieved. There are several
inspirations to choose iris and speech for a multimodal
biometric system. Iris is a physiological trait while speech
is a behavioral trait. These two biometrics can be
combined to form an effective multimodal biometric
system. Iris recognition requires small high quality
cameras for operating and processes the output in 1 to 2
seconds. Iris patterns carry astonishing amount of
information and remain unchanged throughout the
individual’s lifetime. Iris recognition suffers no problem
with eyeglasses and contact lenses. It is hence, one of the
most stable and precise personal identification biometric
which gives excellent recognition performance [13] [17].
Voice recognition system is an emerging biometric
technology. Voice is usually considered as a behavioral
characteristic but it is actually a combination of both
physiological and behavioral characteristics. The
physiological part of the voice remains invariant while the
behavioral part changes over time depending on the age,
medication and emotional state of an individual [18].
Voice recognition biometric system is typically cheap with
the requirement of a microphone. It has high user
preference and the processing speed of 5 seconds with high
efficiency [18].
MATCH
SCORE
MATCHIN
G
MATCH
SCORE
FEATURE
VECTOR
MATCHIN
G
FUSION
TOTAL
SCORE
DECISI
ON
TEMPLA
TE
TEMPLA
TE
SENSOR
DATA
SENSO
R
DATA
FEATURE
EXTRACTI
ON
FEATURE
EXTRACTI
ON
FEATURE
VECTOR
Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426
179 NITTTR, Chandigarh EDIT-2015
V. CONCLUSION AND FUTURE SCOPE
The performance of the unimodal biometric recognition
systems suffer from several limitations that can be
overcome by the use of multimodal biometrics.
Multimodal biometrics combine the information obtained
from the different sources through the use of an effective
fusion scheme. Multimodal biometric systems work in
different scenarios and different fusion levels. The
performance of a multimodal biometric system can be
improved through the selection of appropriate fusion
scheme. In this paper, the modality of iris and speechare
suggested with their score level fusion due to its simple
procedure and rich information.
REFERENCES
[1] Arun Ross and Anil Jain, “Information Fusion in Biometrics”,
Pattern Recognition letters, vol. 2, issue 13, pp. 2115-2125, 2003
[2] Suma Swany and K. V. Ramakrishnan, “An efficient speech
recognition system”, Computer Science and Engeneering : An
international journal (CSEIT), vol. 3, no. 4, Aug 2013
[3] R. Frischholz, U. Dieckmann,“BiolD: A multimodal biometric
identification system”, Computer, Vol. 33,No. 2, pp. 64-68,2000
[4] Sravya V., Radha Krishna Murthy, RavindraBabuKallam, Srujana
B., “A survey on fingerprint biometric system”, International
Journal of Advanced Research in Computer Science and Software
Engineering, vol. 2, issue 4, April 2012
[5] SahilPrabhakar, SharathPankanti, Anil K. Jain, “Biometric
recognition: Security and privacy concerns”, Security and Privacy,
IEEE, vol. 1, issue 2, pp. 33-42, 2003
[6] M. Fatima Naddheen, S. Poornima, “Fusion in Multimodal
Biometric using Iris and Ear”, Proceedings of IEEE Conference on
Information and Communication Technologies, pp. 83-87, 2013
[7] KomalSondhi, YogeshBansal, “Concept of Unimodal and
Multimodal Biometric systems”, International Journal of Advanced
Research in Computer Science and Software Engineering, vol. 4,
issue 6, 2014
[8] ThirimachosBourlai, Nathan Kalka, Arun Ross, BojanCukic,
Lawrence Hornak, “ Cross Spectral Face Verification in the Short
Wave Infrared (SWIR) Band”, Proc. of International Conference on
Pattern Recognition, IEEE, pp. 1343-1347, 2010
[9] DzatiAthiarRamli, Nurue Hayat Che Rani, KhairulAnuarIshak,
“Performances of Weighted Sum Rule Fusion scheme in Multi-
instance and Multimodal Biometric system”, World Applied
Science Journal, vol. 12, no. 11, pp. 2160-2167, 2011
[10] Vaidehi V., Teena Mary Tressa, NareshBabu N. T., AnnisFathima
A., Balamurali P., Girish Chandra M., “Multi Algorithmic Face
Authentication System”, Proceedings of the International Multi-
conference of Engineers and Computer Scientists, vol. 1, pp.485-
490, 2011
[11] GandhimathiAmirthalingam, Radhamani G., “A Multimodal
Approach for Face and Ear Biometric System”, International Journal
of Computer Science Issues (IJCSI), vol.10, issue 5, no. 2, pp. 234-
240, 2013
[12] Xi Cheng, Sergey Tulvakov, VenGovindaraju, “Combination of
Multiple Samples Utilizing Identification Modal in Biometric
System”, International Conference on Biometric Compendium,
IEEE, pp. 1-5, 2011
[13] Anil K. Jain, Arun Ross, SahilPrabhakar, “An Introduction to
Biometric Recognition”, IEEE Transactions on Circuits and
Systems for Video Technology, vol. 14, no. 1, pp. 4-20, 2004
[14] Vincenzo Conti, Carmelo Militello, FlippoSorbello, “A Frequency
Based Approach for Feature Fusionin Fingerprint and Iris
Multimodal Biometric Identification Systems”, IEEE Transactions
on Systems, Man and Cybernetics, vol. 40, no. 4, pp. 384-395, 2010
[15] Sayed Hassan Sadeghzadeh, MortezaAmirsheibani,
AnsehDaneshArasteh, “Fingerprint and Speech Fusion: A
Multimodal Biometric System”, International Journal of Electronics
Communication and Computer Technology (IJECCT), vol. 4, no. 2,
pp. 570-576, 2011
[16] Kihal N., Chitroub S., Meunier J., “Fusion of Iris and Palmprint for
Multimodal Biometric Authentication,” 4th
International Conference
on Image Processing theory, tools and applications (IPTA), 2014
[17] A.K. Jain, A. Ross, S. Pankanti, “ Biometrics, a tool for
information”, IEEE Transactions on Information Forensics and
security, vol. 1, issue 2, pp. 125-143, 2006
[18] DwijenRudrapal, Smita Das, S. Debbarama, N. Kar, N. Debbarama,
“Voice recognition and authenticationas a proficient biometric tool
and its application in online exam for PH people”, International
Jouranal of Computer Applications, vol. 39, no. 12, pp. 6-12, 2012
[19] Suma Swamy and K. V. Ramakrishnan, “An Efficient Speech
Recognition System”, Computer Science and Engineering: An
International Journal (CSEIJ), vol. 3, no. 4, pp. 21-27, Aug 2013.

An Approach to Speech and Iris based Multimodal Biometric System

  • 1.
    Int. Journal ofElectrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 NITTTR, Chandigarh EDIT -2015 176 An Approach to Speech and Iris based Multimodal Biometric System 1 SakshiSahore, 2 TanviSood 1 M.Tech Student, 2 Assistant Professor 1,2 ECE Department, Chandigarh Engineering College, Ladran, Mohali 1 sakshisahore@gmail.com, 2 cecm.ece.ts@gmail.com Abstract—Biometrics is the science and technology of human identification and verification through the use of feature set extracted from the biological data of the individual to be recognized. Unimodal and Multimodal systems are the two modal systems which have been developed so far. Unimodal biometric systems use a single biometric trait but they face limitations in the system performance due to the presence of noise in data, interclass variations and spoof attacks. These problems can be resolved by using multimodal biometrics which rely on more than one biometric information to produce better recognition results. This paper presents an overview of the multimodal biometrics, various fusion levels used in them and suggests the use of iris and speech using score level fusion for a multimodal biometric system. Keywords—Biometric, unimodal, multimodal, recognition, score level fusion I. INTRODUCTION With the recent advancement in technology and development of electrically interconnected society, there is an essential requirement of accurate personal authentication system to handle various person authentication issues in daily life. There are several authentication systems that we use on daily basis such as personal identification number (PIN), smartcards and passwords. These systems are possession based and knowledge based and can easily be misplaced, forgotten or forged [1]. To overcome these difficulties, biometric systems for authentication are introduced. Biometrics is a robustious approach for the recognition of a person [2]. Biometrics verify the identity of the subject based on a feature set extracted from the subject’s biological characteristics.Biometric characteristics are of two types:  Physiological: The characteristics related to the body of a person are called physiological characteristics. Fingerprints, face, iris, palm geometry, DNA are the examples of the physiological characteristics. These characteristics do not change over time.  Behavioral: The characteristics related to the behavior of a person are called behavioral characteristic. Voice, gait, signature and keystroke are the examples of behavioral characteristics. These are variant in nature. A biometric system consists of two modes that are enrollment mode and authentication mode. In enrollment mode, the biometric data of the subject is taken and processed for feature extraction. These features are used for the generation of template of that subject in which all the feature variations are captured and stored in a database. During authentication mode, the features from the subject to be identified are computed and then compared with the stored template in the database. If the features match, the subject is recognized. Figure 1 shows a typical biometric system. Biometric based person recognition system [3] II. MODAL SYSTEM A. Unimodal Biometrics A unimodal biometric system uses a single source of biometric information to generate the recognition result. Most of the deployed real world applications in biometrics are unimodal, that is, they use a single biometric trait for authentication such as a biometric system based on fingerprints [4]. While unimodal biometric systems have successfully been installed in various applications, but unimodal biometrics is still not fully solved problem [5]. These systems a variety of issues like  Noisy data – The input biometric data might be noisy or the biometric sensors might be susceptible to noise which may lead to inaccurate matching and hence false rejection.  Intra-class variations – This occurs when the biometric data acquired from an individual during verification is not identical to the data stored in the template during enrollment. This occurs due to incorrect interaction of the individual with the sensor.  Non-universality – Sometimes it is possible that certain individuals may not provide a particular biometric causing failure to enroll (FTE).  Spoof attack –Unimodal biometrics are susceptible to spoof attacks where an imposter may attempt to fake Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 NITTTR, Chandigarh EDIT -2015 176 An Approach to Speech and Iris based Multimodal Biometric System 1 SakshiSahore, 2 TanviSood 1 M.Tech Student, 2 Assistant Professor 1,2 ECE Department, Chandigarh Engineering College, Ladran, Mohali 1 sakshisahore@gmail.com, 2 cecm.ece.ts@gmail.com Abstract—Biometrics is the science and technology of human identification and verification through the use of feature set extracted from the biological data of the individual to be recognized. Unimodal and Multimodal systems are the two modal systems which have been developed so far. Unimodal biometric systems use a single biometric trait but they face limitations in the system performance due to the presence of noise in data, interclass variations and spoof attacks. These problems can be resolved by using multimodal biometrics which rely on more than one biometric information to produce better recognition results. This paper presents an overview of the multimodal biometrics, various fusion levels used in them and suggests the use of iris and speech using score level fusion for a multimodal biometric system. Keywords—Biometric, unimodal, multimodal, recognition, score level fusion I. INTRODUCTION With the recent advancement in technology and development of electrically interconnected society, there is an essential requirement of accurate personal authentication system to handle various person authentication issues in daily life. There are several authentication systems that we use on daily basis such as personal identification number (PIN), smartcards and passwords. These systems are possession based and knowledge based and can easily be misplaced, forgotten or forged [1]. To overcome these difficulties, biometric systems for authentication are introduced. Biometrics is a robustious approach for the recognition of a person [2]. Biometrics verify the identity of the subject based on a feature set extracted from the subject’s biological characteristics.Biometric characteristics are of two types:  Physiological: The characteristics related to the body of a person are called physiological characteristics. Fingerprints, face, iris, palm geometry, DNA are the examples of the physiological characteristics. These characteristics do not change over time.  Behavioral: The characteristics related to the behavior of a person are called behavioral characteristic. Voice, gait, signature and keystroke are the examples of behavioral characteristics. These are variant in nature. A biometric system consists of two modes that are enrollment mode and authentication mode. In enrollment mode, the biometric data of the subject is taken and processed for feature extraction. These features are used for the generation of template of that subject in which all the feature variations are captured and stored in a database. During authentication mode, the features from the subject to be identified are computed and then compared with the stored template in the database. If the features match, the subject is recognized. Figure 1 shows a typical biometric system. Biometric based person recognition system [3] II. MODAL SYSTEM A. Unimodal Biometrics A unimodal biometric system uses a single source of biometric information to generate the recognition result. Most of the deployed real world applications in biometrics are unimodal, that is, they use a single biometric trait for authentication such as a biometric system based on fingerprints [4]. While unimodal biometric systems have successfully been installed in various applications, but unimodal biometrics is still not fully solved problem [5]. These systems a variety of issues like  Noisy data – The input biometric data might be noisy or the biometric sensors might be susceptible to noise which may lead to inaccurate matching and hence false rejection.  Intra-class variations – This occurs when the biometric data acquired from an individual during verification is not identical to the data stored in the template during enrollment. This occurs due to incorrect interaction of the individual with the sensor.  Non-universality – Sometimes it is possible that certain individuals may not provide a particular biometric causing failure to enroll (FTE).  Spoof attack –Unimodal biometrics are susceptible to spoof attacks where an imposter may attempt to fake Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 NITTTR, Chandigarh EDIT -2015 176 An Approach to Speech and Iris based Multimodal Biometric System 1 SakshiSahore, 2 TanviSood 1 M.Tech Student, 2 Assistant Professor 1,2 ECE Department, Chandigarh Engineering College, Ladran, Mohali 1 sakshisahore@gmail.com, 2 cecm.ece.ts@gmail.com Abstract—Biometrics is the science and technology of human identification and verification through the use of feature set extracted from the biological data of the individual to be recognized. Unimodal and Multimodal systems are the two modal systems which have been developed so far. Unimodal biometric systems use a single biometric trait but they face limitations in the system performance due to the presence of noise in data, interclass variations and spoof attacks. These problems can be resolved by using multimodal biometrics which rely on more than one biometric information to produce better recognition results. This paper presents an overview of the multimodal biometrics, various fusion levels used in them and suggests the use of iris and speech using score level fusion for a multimodal biometric system. Keywords—Biometric, unimodal, multimodal, recognition, score level fusion I. INTRODUCTION With the recent advancement in technology and development of electrically interconnected society, there is an essential requirement of accurate personal authentication system to handle various person authentication issues in daily life. There are several authentication systems that we use on daily basis such as personal identification number (PIN), smartcards and passwords. These systems are possession based and knowledge based and can easily be misplaced, forgotten or forged [1]. To overcome these difficulties, biometric systems for authentication are introduced. Biometrics is a robustious approach for the recognition of a person [2]. Biometrics verify the identity of the subject based on a feature set extracted from the subject’s biological characteristics.Biometric characteristics are of two types:  Physiological: The characteristics related to the body of a person are called physiological characteristics. Fingerprints, face, iris, palm geometry, DNA are the examples of the physiological characteristics. These characteristics do not change over time.  Behavioral: The characteristics related to the behavior of a person are called behavioral characteristic. Voice, gait, signature and keystroke are the examples of behavioral characteristics. These are variant in nature. A biometric system consists of two modes that are enrollment mode and authentication mode. In enrollment mode, the biometric data of the subject is taken and processed for feature extraction. These features are used for the generation of template of that subject in which all the feature variations are captured and stored in a database. During authentication mode, the features from the subject to be identified are computed and then compared with the stored template in the database. If the features match, the subject is recognized. Figure 1 shows a typical biometric system. Biometric based person recognition system [3] II. MODAL SYSTEM A. Unimodal Biometrics A unimodal biometric system uses a single source of biometric information to generate the recognition result. Most of the deployed real world applications in biometrics are unimodal, that is, they use a single biometric trait for authentication such as a biometric system based on fingerprints [4]. While unimodal biometric systems have successfully been installed in various applications, but unimodal biometrics is still not fully solved problem [5]. These systems a variety of issues like  Noisy data – The input biometric data might be noisy or the biometric sensors might be susceptible to noise which may lead to inaccurate matching and hence false rejection.  Intra-class variations – This occurs when the biometric data acquired from an individual during verification is not identical to the data stored in the template during enrollment. This occurs due to incorrect interaction of the individual with the sensor.  Non-universality – Sometimes it is possible that certain individuals may not provide a particular biometric causing failure to enroll (FTE).  Spoof attack –Unimodal biometrics are susceptible to spoof attacks where an imposter may attempt to fake
  • 2.
    Int. Journal ofElectrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 177 NITTTR, Chandigarh EDIT-2015 the biometric trait of an enrolled user in order to bypass the system. The problems imposed in the unimodal biometrics limit the accuracy and system performance. B. Multimodal Biometrics A multimodal biometric system uses more than one source of biological data to generate the recognition result, for example a multimodal biometric system using iris and ear [6]. Multimodal biometrics overcome the shortcomings of unimodal biometrics [7] and is more reliable because of the use of more than one biometric trait and hence more pieces of information. Multimodal scenarios: A multimodal biometric system can be designed to work in one of the following scnarios.  Multiple sensors: Highlight all author and affiliation lines.The informationof the same biometric can be acquiredby different sensors [8].The different samples are thenprocessed by the same algorithm and the resultsare fused to get the resultant algorithm.  Multiple instances: The biometric information is extracted from the multiple instances of the same biometric [9].  Multiple algorithms: More than one approach/algorithm is used for feature extraction or classification of the same biometric to improve the system performance [10].  Multiple biometric:Evidence from the multiple biometric characteristics is taken [11].  Multiple samples:Multiple samples are acquired from the same biometric by a single sensor and processed by the same algorithm to obtain the recognition results [12]. Multimodal Biometric System [13] Fusion Levels in Multimodal Biometrics: While designing a multimodal system, different fusion strategies can be used to integrate the biometric data. Fusion Levels in Multimodal Biometrics  Sensor level fusion:Highlight all author and affiliation lines. Sensor level fusion is done in a system system using multiple sensors or in a system using a single sensor at multiple instances. In this, the biometric data obtained by the sensor is combined.  Feature level fusion:Feature level fusion is done by extracting the features of different biometric sources individually and then combining those features into a single feature vector [14].  Score level fusion:Score level fusion is performed by individually processing (sensing and extracting features) different biometric sources and finding their match scores. These scores are then combined to make classification [15].  Decision level fusion:After each biometric source is processed and recognition decision is made for each biometric data, fusion is executed at the decision level [16]. III. SCORE LEVEL FUSION Score level fusion is the most popular and common approach in the multimodal biometrics system due its simple procedure. Matching scores contain rich information about the input pattern. Each classifiers provides a matching scores and scores of different classifiers are combined to produce the final score. A. Fusion algorithms When different matching scores of different biometrics are acquired, their fusion is done. For the fusion of the matching scores, different algorithms can be applied. These algorithms include product rule, sum rule, max rule and min rule. Consider ( ⃗) as the output of individual classifiers, as a feature vector to ith classifier, R as the number of different classifiers and be the output. The different rules can be applied as BIOMETRIC FUSION BEFORE MATCHIG AFTER MATCHING SCORE LEVEL FEATURE LEVEL SCORE LEVEL DECISIO N LEVEL Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 177 NITTTR, Chandigarh EDIT-2015 the biometric trait of an enrolled user in order to bypass the system. The problems imposed in the unimodal biometrics limit the accuracy and system performance. B. Multimodal Biometrics A multimodal biometric system uses more than one source of biological data to generate the recognition result, for example a multimodal biometric system using iris and ear [6]. Multimodal biometrics overcome the shortcomings of unimodal biometrics [7] and is more reliable because of the use of more than one biometric trait and hence more pieces of information. Multimodal scenarios: A multimodal biometric system can be designed to work in one of the following scnarios.  Multiple sensors: Highlight all author and affiliation lines.The informationof the same biometric can be acquiredby different sensors [8].The different samples are thenprocessed by the same algorithm and the resultsare fused to get the resultant algorithm.  Multiple instances: The biometric information is extracted from the multiple instances of the same biometric [9].  Multiple algorithms: More than one approach/algorithm is used for feature extraction or classification of the same biometric to improve the system performance [10].  Multiple biometric:Evidence from the multiple biometric characteristics is taken [11].  Multiple samples:Multiple samples are acquired from the same biometric by a single sensor and processed by the same algorithm to obtain the recognition results [12]. Multimodal Biometric System [13] Fusion Levels in Multimodal Biometrics: While designing a multimodal system, different fusion strategies can be used to integrate the biometric data. Fusion Levels in Multimodal Biometrics  Sensor level fusion:Highlight all author and affiliation lines. Sensor level fusion is done in a system system using multiple sensors or in a system using a single sensor at multiple instances. In this, the biometric data obtained by the sensor is combined.  Feature level fusion:Feature level fusion is done by extracting the features of different biometric sources individually and then combining those features into a single feature vector [14].  Score level fusion:Score level fusion is performed by individually processing (sensing and extracting features) different biometric sources and finding their match scores. These scores are then combined to make classification [15].  Decision level fusion:After each biometric source is processed and recognition decision is made for each biometric data, fusion is executed at the decision level [16]. III. SCORE LEVEL FUSION Score level fusion is the most popular and common approach in the multimodal biometrics system due its simple procedure. Matching scores contain rich information about the input pattern. Each classifiers provides a matching scores and scores of different classifiers are combined to produce the final score. A. Fusion algorithms When different matching scores of different biometrics are acquired, their fusion is done. For the fusion of the matching scores, different algorithms can be applied. These algorithms include product rule, sum rule, max rule and min rule. Consider ( ⃗) as the output of individual classifiers, as a feature vector to ith classifier, R as the number of different classifiers and be the output. The different rules can be applied as BIOMETRIC FUSION BEFORE MATCHIG AFTER MATCHING SCORE LEVEL FEATURE LEVEL SCORE LEVEL DECISIO N LEVEL Int. Journal of Electrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 177 NITTTR, Chandigarh EDIT-2015 the biometric trait of an enrolled user in order to bypass the system. The problems imposed in the unimodal biometrics limit the accuracy and system performance. B. Multimodal Biometrics A multimodal biometric system uses more than one source of biological data to generate the recognition result, for example a multimodal biometric system using iris and ear [6]. Multimodal biometrics overcome the shortcomings of unimodal biometrics [7] and is more reliable because of the use of more than one biometric trait and hence more pieces of information. Multimodal scenarios: A multimodal biometric system can be designed to work in one of the following scnarios.  Multiple sensors: Highlight all author and affiliation lines.The informationof the same biometric can be acquiredby different sensors [8].The different samples are thenprocessed by the same algorithm and the resultsare fused to get the resultant algorithm.  Multiple instances: The biometric information is extracted from the multiple instances of the same biometric [9].  Multiple algorithms: More than one approach/algorithm is used for feature extraction or classification of the same biometric to improve the system performance [10].  Multiple biometric:Evidence from the multiple biometric characteristics is taken [11].  Multiple samples:Multiple samples are acquired from the same biometric by a single sensor and processed by the same algorithm to obtain the recognition results [12]. Multimodal Biometric System [13] Fusion Levels in Multimodal Biometrics: While designing a multimodal system, different fusion strategies can be used to integrate the biometric data. Fusion Levels in Multimodal Biometrics  Sensor level fusion:Highlight all author and affiliation lines. Sensor level fusion is done in a system system using multiple sensors or in a system using a single sensor at multiple instances. In this, the biometric data obtained by the sensor is combined.  Feature level fusion:Feature level fusion is done by extracting the features of different biometric sources individually and then combining those features into a single feature vector [14].  Score level fusion:Score level fusion is performed by individually processing (sensing and extracting features) different biometric sources and finding their match scores. These scores are then combined to make classification [15].  Decision level fusion:After each biometric source is processed and recognition decision is made for each biometric data, fusion is executed at the decision level [16]. III. SCORE LEVEL FUSION Score level fusion is the most popular and common approach in the multimodal biometrics system due its simple procedure. Matching scores contain rich information about the input pattern. Each classifiers provides a matching scores and scores of different classifiers are combined to produce the final score. A. Fusion algorithms When different matching scores of different biometrics are acquired, their fusion is done. For the fusion of the matching scores, different algorithms can be applied. These algorithms include product rule, sum rule, max rule and min rule. Consider ( ⃗) as the output of individual classifiers, as a feature vector to ith classifier, R as the number of different classifiers and be the output. The different rules can be applied as BIOMETRIC FUSION BEFORE MATCHIG AFTER MATCHING SCORE LEVEL FEATURE LEVEL SCORE LEVEL DECISIO N LEVEL
  • 3.
    Int. Journal ofElectrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 NITTTR, Chandigarh EDIT -2015 178 Score Level Fusion in Multimodal Biometric System Product rule:Different biometric traits of an individual (such as iris, fingerprints and ear) are mutually independent and the product rule is applied based on this. = ( ⃗) Sum rule: The sum rule takes the scores of the individual classifiers to simply calculate their sum. = ( ⃗) Max rule:The max rule approximates the output by the maximum value of the scores. = max ( ⃗) Min rule:The max rule approximates the output by the minimum value of the scores. = min ( ⃗) B. Normalization Normalization is done after determining the matching scores from different biometrics. Score normalization is essential because the matching scores of different biometrics are obtained from different algorithms and hence may not have the same underlying properties, that is, they may be of different nature and scale. Normalization changes the scale of the different scores and brings them to a common domain. After normalization, the scores are combined. The most common normalization algorithms used are If S = ( , , ,… ,……. ) is a vector of M scores, then the normalization score, will be Min max normalization: = µ( ) ( ) ( ) Where max(s) = maximum value of raw score and min(s) = minimum value of raw score Z-score normalization = µ( ) ( ) Where µ(s) = mean deviation of set of score vectors and σ(s) = standard deviation of the set of score vectors Median-MAD normalization = Where MAD = median (| − |) Tanh normalization = 0.5 ℎ 0.01 − µ( ) ( ) + 1 Where µ( )= mean deviation calculated from scores and ( ) = standard deviation calculated from scores. IV. MULIIMODAL BIOMETRIC USING IRIS AND SPEECH As mentioned earlier, there are two types of biometric characteristics in human beings, physiological characteristics and behavioral characteristics. With the selection of appropriate modals and fusion scheme, optimal results can be achieved. There are several inspirations to choose iris and speech for a multimodal biometric system. Iris is a physiological trait while speech is a behavioral trait. These two biometrics can be combined to form an effective multimodal biometric system. Iris recognition requires small high quality cameras for operating and processes the output in 1 to 2 seconds. Iris patterns carry astonishing amount of information and remain unchanged throughout the individual’s lifetime. Iris recognition suffers no problem with eyeglasses and contact lenses. It is hence, one of the most stable and precise personal identification biometric which gives excellent recognition performance [13] [17]. Voice recognition system is an emerging biometric technology. Voice is usually considered as a behavioral characteristic but it is actually a combination of both physiological and behavioral characteristics. The physiological part of the voice remains invariant while the behavioral part changes over time depending on the age, medication and emotional state of an individual [18]. Voice recognition biometric system is typically cheap with the requirement of a microphone. It has high user preference and the processing speed of 5 seconds with high efficiency [18]. MATCH SCORE MATCHIN G MATCH SCORE FEATURE VECTOR MATCHIN G FUSION TOTAL SCORE DECISI ON TEMPLA TE TEMPLA TE SENSOR DATA SENSO R DATA FEATURE EXTRACTI ON FEATURE EXTRACTI ON FEATURE VECTOR
  • 4.
    Int. Journal ofElectrical & Electronics Engg. Vol. 2, Spl. Issue 1 (2015) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 179 NITTTR, Chandigarh EDIT-2015 V. CONCLUSION AND FUTURE SCOPE The performance of the unimodal biometric recognition systems suffer from several limitations that can be overcome by the use of multimodal biometrics. Multimodal biometrics combine the information obtained from the different sources through the use of an effective fusion scheme. Multimodal biometric systems work in different scenarios and different fusion levels. The performance of a multimodal biometric system can be improved through the selection of appropriate fusion scheme. In this paper, the modality of iris and speechare suggested with their score level fusion due to its simple procedure and rich information. REFERENCES [1] Arun Ross and Anil Jain, “Information Fusion in Biometrics”, Pattern Recognition letters, vol. 2, issue 13, pp. 2115-2125, 2003 [2] Suma Swany and K. V. Ramakrishnan, “An efficient speech recognition system”, Computer Science and Engeneering : An international journal (CSEIT), vol. 3, no. 4, Aug 2013 [3] R. Frischholz, U. Dieckmann,“BiolD: A multimodal biometric identification system”, Computer, Vol. 33,No. 2, pp. 64-68,2000 [4] Sravya V., Radha Krishna Murthy, RavindraBabuKallam, Srujana B., “A survey on fingerprint biometric system”, International Journal of Advanced Research in Computer Science and Software Engineering, vol. 2, issue 4, April 2012 [5] SahilPrabhakar, SharathPankanti, Anil K. Jain, “Biometric recognition: Security and privacy concerns”, Security and Privacy, IEEE, vol. 1, issue 2, pp. 33-42, 2003 [6] M. Fatima Naddheen, S. Poornima, “Fusion in Multimodal Biometric using Iris and Ear”, Proceedings of IEEE Conference on Information and Communication Technologies, pp. 83-87, 2013 [7] KomalSondhi, YogeshBansal, “Concept of Unimodal and Multimodal Biometric systems”, International Journal of Advanced Research in Computer Science and Software Engineering, vol. 4, issue 6, 2014 [8] ThirimachosBourlai, Nathan Kalka, Arun Ross, BojanCukic, Lawrence Hornak, “ Cross Spectral Face Verification in the Short Wave Infrared (SWIR) Band”, Proc. of International Conference on Pattern Recognition, IEEE, pp. 1343-1347, 2010 [9] DzatiAthiarRamli, Nurue Hayat Che Rani, KhairulAnuarIshak, “Performances of Weighted Sum Rule Fusion scheme in Multi- instance and Multimodal Biometric system”, World Applied Science Journal, vol. 12, no. 11, pp. 2160-2167, 2011 [10] Vaidehi V., Teena Mary Tressa, NareshBabu N. T., AnnisFathima A., Balamurali P., Girish Chandra M., “Multi Algorithmic Face Authentication System”, Proceedings of the International Multi- conference of Engineers and Computer Scientists, vol. 1, pp.485- 490, 2011 [11] GandhimathiAmirthalingam, Radhamani G., “A Multimodal Approach for Face and Ear Biometric System”, International Journal of Computer Science Issues (IJCSI), vol.10, issue 5, no. 2, pp. 234- 240, 2013 [12] Xi Cheng, Sergey Tulvakov, VenGovindaraju, “Combination of Multiple Samples Utilizing Identification Modal in Biometric System”, International Conference on Biometric Compendium, IEEE, pp. 1-5, 2011 [13] Anil K. Jain, Arun Ross, SahilPrabhakar, “An Introduction to Biometric Recognition”, IEEE Transactions on Circuits and Systems for Video Technology, vol. 14, no. 1, pp. 4-20, 2004 [14] Vincenzo Conti, Carmelo Militello, FlippoSorbello, “A Frequency Based Approach for Feature Fusionin Fingerprint and Iris Multimodal Biometric Identification Systems”, IEEE Transactions on Systems, Man and Cybernetics, vol. 40, no. 4, pp. 384-395, 2010 [15] Sayed Hassan Sadeghzadeh, MortezaAmirsheibani, AnsehDaneshArasteh, “Fingerprint and Speech Fusion: A Multimodal Biometric System”, International Journal of Electronics Communication and Computer Technology (IJECCT), vol. 4, no. 2, pp. 570-576, 2011 [16] Kihal N., Chitroub S., Meunier J., “Fusion of Iris and Palmprint for Multimodal Biometric Authentication,” 4th International Conference on Image Processing theory, tools and applications (IPTA), 2014 [17] A.K. Jain, A. Ross, S. Pankanti, “ Biometrics, a tool for information”, IEEE Transactions on Information Forensics and security, vol. 1, issue 2, pp. 125-143, 2006 [18] DwijenRudrapal, Smita Das, S. Debbarama, N. Kar, N. Debbarama, “Voice recognition and authenticationas a proficient biometric tool and its application in online exam for PH people”, International Jouranal of Computer Applications, vol. 39, no. 12, pp. 6-12, 2012 [19] Suma Swamy and K. V. Ramakrishnan, “An Efficient Speech Recognition System”, Computer Science and Engineering: An International Journal (CSEIJ), vol. 3, no. 4, pp. 21-27, Aug 2013.