The Survey of Architecture of Multi-Modal (Fingerprint and Iris Recognition) Biometric Authentication System

Afshan Ashraf. Int. Journal of Engineering Research and Application www.ijera.com
ISSN : 2248-9622, Vol. 7, Issue 4, ( Part -3) April 2017, pp.16-22
www.ijera.com DOI: 10.9790/9622-0704031625 16 | P a g e
The Survey of Architecture of Multi-Modal (Fingerprint and Iris
Recognition) Biometric Authentication System
Afshan Ashraf1
, Isha Vats2
1
Department of Computer Science and Engineering, Chandigarh Engineering College, Landran, Mohali,
Punjab, India.
2
Department of Computer Science and Engineering, Chandigarh Engineering College, Landran, Mohali
, Punjab, India.
ABSTRACT
Biometrics based individual identification is observed as an effective technique for automatically knowing, with
a high confidence a person’s identity. Multi-modal biometric systems consolidate the evidence accessible by
multiple biometric sources and normally better recognition performance associate to system based on a single
biometric modality.Multi biometric systems are used to overcome this issue by providing multiple pieces of
indication of the same identity. This system provides effective fusion structure that combines information
provided by the multiple field experts based on decision-level and score-level fusion method, thereby increasing
the efficiency which is not conceivable in uni-modal system.Multi-modal biometrics can be attained through a
fusion of two or more images, where the subsequent fused image will be more protected. This paper discusses
various fusion techniques, architecture of multi-modal biometric authentication and working of biometric fusion
i.e. Iris and Fingerprint recognition that are used in multi-modal biometrics.
Keywords: Multi-modal Biometric Authentication, Iris and Fingerprint recognition, Fusion methods i.e. Score
Level Fusion and Decision Level Fusion.
I. INTRODUCTION
Biometric systems automatically
determine or confirm a person’s identity based on
his anatomical and behavioural characteristics like
fingerprint, palm print, vein pattern, face and
iris[1]. A method of recognizing or confirming the
identity of an individual person or subject that
depends on the physiological and behavioural
characteristics is biometric recognition. Multimodal
biometrics rise correctness by considering other
very specific biological traits to limit the no of
applicant for an identity. Multimodal biometric
systems use more than one physiological or
behavioural characteristic for enrolment,
verification and identification.
A multimodal biometric authentication,
which recognizes an individual person using
physiological and behavioural characteristics, such
as face, fingerprints, finger geometry, iris, retina,
vein and speech is one of the best attractive and
effective techniques. These methods are more
reliable and capable than knowledge-based[2] (e.g.
password) or token-based (e.g. Key) methods.
Since biometric features are hardly stolen and
forgotten. However, a single biometric feature
sometimes fails to be exact sufficient for
confirming the identity of a person. Through
combining multiple modalities enhanced
presentation reliability could be achieved. Due to
its promising requests as well as the theoretical
challenges, multimodal biometric has drawn more
and more attention in current years. Iris and
fingerprint multimodal biometrics are beneficial
due to the use of non-invasive and low-cost picture
acquisition. In this method we can easily acquire
iris and fingerprint images using two touch sensors
simultaneously. Existing studies in this approach
[3] serve holistic features for face representation
and results are shown with minorrecords set that
was reported. Multimodal system also provides
anti-spoofing measures by which it becomes
problematic for an intruder to spool multiple
biometric traits at the same time. However, an
integration scheme is required to fuse the data
presented by the individual modalities.
Multimodal biometric scheme has addressed some
issues related to uni-modal biometrics such as; [4]
(a) Non-universality or insufficient population
coverage (reduce failure to enrol rate which rises
population coverage).
(b) It becomes totally unmanageable for an
impostor to imitate several biometric traits of a
legitimately enrolled user separately.
(c) Multimodal-biometric systems offer climbing
evidence in solving the difficulty of noisy data
(illness affecting voice, scar affecting fingerprint).
RESEARCH ARTICLE OPEN ACCESS

Figure no.1: Biometric Authentication [5]
II. STEPS OF MULTI-MODAL BIOMETRIC
SYSTEM
The simple requirements of any biometric system
are Input device, Biometric software and Databank.
A. Input Device: An input device like scanner,
writing pad etc. are used to control the input which
is then used by the software part.
B. Biometric Software: Software to process the
input and changes it into digital form, extracts the
features and compares the result[6]. In terms of
correctness, the performance of a biometric system
totally depends on the quality of the software.
C. Database: A databankis used to store the
mathematical data which is further used for the
comparison. Features extracted from the input
samples are kept instead of input samples as the
samples take more space and putting away features
saves time for processing samples again to extract
the features. The presentation of the software is
most important part of a biometric system as the
accuracy of the system should be contingent on
quality of the software.
III. ARCHITECTURE OF MULTI-
MODAL BIOMETRIC SYSTEM
The two main processes involved are enrolment or
registration, confirmation and identification. Fig. 2
shows the architecture of a general biometric
system.
A. Registration Process: In order to identify an
individual, it is necessary to store the
individual’s features in a databank, which are
extracted from the reliable samples of the
biometric trait either look over or recorded
using input device like writing pads[7]. These
features are then matched with the features
extracted from the traits of the individual
necessary to be identified.
Figure no.2: Architecture of a general biometric
system[7]
B. In order to extract features, the input model is
pre-processed and feature extraction algorithm
is applied on these pre-processed examples to
form feature vectors. Instead of input models
these feature vectors are stored in the databank
as input samples take more space on secondary
memory than mathematical files and features
are computed just once which protect a lot of
processing time. A classifier is trained by these
feature vectors which then classifies the
unknown input sample.
C. Verification: In the verification manner, the
system validates a person’s identity through
associating the captured biometric data with his
own biometric template kept in the system
database. In such a system, an individual who
needs to be recognized claims an identity,
mainly via PIN (Personal Identification
number), a user name or a smart card and the
system store uses a one to one comparison to
define whether the claim is true or not [8].
D. Identification: In identification mode the
system recognizes a genuine user by searching
the templates provided by the user in the

databank for a match. Therefore, the system
conducts a one to several comparisons to
identify an individual entity or fails to identify
if the subject is not enrolled in the system
database[9].
IV. LEVELS OF MULTI-MODAL
BIOMETRIC FUSION SYSTEM
Biometric Fusion takes place at various levels i.e
Sensor, decision and score etc. described below:
[10]
Figure no. 3: Levels of Biometric Fusion
A. Sensor level: This fusion strategy requires the
raw data to be acquired from multiple sensors
which can further be processed and integrated to
make new data from which features can be
extracted. Sensor level fusion can be done only if
multiple cues of the similar biometric are obtained
from multiple compatible sensors.
B. Feature level: The feature set is mined from
many sources of information and is further
concatenated into a joint feature vector. This novel
high dimensional feature vector represents an
individual. In case of feature level fusion some
reduction method must be used in order to select
only useful features [11].
C. Match score level: Match score is a measure
between input biometric and template biometric
feature vectors. Based on the similarity of feature
vector and the template, every subsystem calculates
its own match score value. These individual scores
are finally joined to obtain a total score, which is
then passed to the decision module, after which
recognition is performed.
D.Rank level: Rank level fusion is normally
adopted for the identification of the person rather
than verification. Thus, fusion entails consolidating
the ranks associated with an identity and
determining a new rank that would aid in
establishing the final decision.
E. Decision level: In a multi-modal biometric
system, fusion is carried out at this level when only
the decision output is available. Here, a separate
authentication decision is computed for every
biometric trait which is then combined that results
in a final output. Different strategies are available
to group the distinct decisions of individual
modality to a final authentication decision. Fusion
at this stage is regarded to be definite as far as other
fusion levels are considered because very less
information is available[12].
V. RELATED WORK
Satrajit Mukherjee et.al,2014[13] defined that
theNovel adaptive weight and supporter based
function mapping the matching scores from
dissimilar biometric causes into a single merged
matching score to be used by a classifier for further
decision making. Differential Growth has been
working to regulate these tuneable parameters with
the independent being the minimization of the
covering area of the occurrence distributions of
open and imposter scores in the fused score space,
which are projected by Gaussian kernel density
method to achieve higher level of accuracy.
Samarth Bharadwaj et.al, 2014 [14] the paper
present theReview of the features, strengths, and
boundaries of existing quality evaluation technique
in fingerprint, iris, and face biometric are also
obtainable. lastly, a courier set of quality metrics
from these three modalities are evaluate on a
multimodal database consisting of 2D images, to
appreciate their performance with deference to
match score obtained from the state of the art
recognition systems. The study of the characteristic
function of excellence and match scores show that
a cautious selection of admiring set of superiority
metrics can provide more advantage to various
applications of biometric excellence.Vincenzo
Cont et.al,2013 [15] In this section fingerprint and
iris based uni-modal and multimodal confirmation
systems will be describe, analyse and evaluate. To
conclude, a proto typed embedded multimodal
biometric sensor will be sketch. Software and
hardware proto-types have been checked against
common and broadly used databases.Sambit
Bakshi et.al., 2012 [16] in this articleachieved
classification operation on the detected key points.
Each set of the key points of the query image was
exposed to nearest national match with respective
set of key points of the database image. Hence
there were two notches generated by the matching
of two classes. This paper also recommends a
accurate monotonic function on these dual scores to
produce a single score such that the final score rate
gives rise to better disjunction between unaffected
and imposter scores than conservative SIFT.
Vincenzo Conti et.al. 2012 [17]
presentedthecontract with modern computing
systems safety issues, focusing on biometric
founded asymmetric keys generation procedure.
Conservative PKI systems were based on
private/public keys produced through RSA or
similar algorithms. The present solution embeds
biometric information on the private or public keys
generation process. In addition the conforming
private key depends on physical or interactive

biometric features and it can be produced when it
was needed. Initial from fingerprint acquisition, the
biometric identifier were extracted, cyphered, and
stored in tamper resilient smartcard to overcome
the security difficulties of centralized databases.
Biometric information is then used for user
verification and for public/private keys generation.
VI. FINGERPRINT AND IRIS
RECOGNITION SYSTEM
A. Fingerprint Recognition
A serious step in automatic fingerprint
matching is to automatically and reliably extract
minutiae features from the input fingerprint picture.
However, the performance of a minutiae extraction
algorithm relies heavily on the quality of the input
fingerprint picture. In order to ensure that the
performance of an automatic fingerprint
identification/verification scheme would be robust
with respect to the quality of the fingerprint
imageries, it would be essential to incorporate a
fingerprint improvement algorithm in the minutiae
extraction module.[18]
Figure no. 4: Original Image[18]
Procedure of Fingerprint Recognition System
(Steps):
- Normalization: Normalization allows
standardizing the distorted levels of variation in
the gray scale values amongst ridges and
valleys. Histogram equalization, as
normalization method, is a process to enhance
the contrast of a picture by transforming its
intensity values.
- Segmentation: In general, only a Region of
Interest is useful to be recognized for each
fingerprint picture. The image area without
effective ridges and furrows is first discarded
since it only holds background data.
- Minutiae Extraction:The endings and
bifurcations of the fingerprint images are
known as the minutiae.
Figure no. 5(i) Normalized Image (ii) Segmented
and (iii) Minutiae Features[18]
B. Iris Recognition
Iris recognition is a method of biometric
authentication that uses pattern appreciation
techniques based on high-resolution picture of the
edges of an individual's eyes.[19]
Figure no. 6: Iris Recognition[19]
Process of Iris recognition:
- Iris Segmentation: This involves first
employing Canny Edge Detection to create an
edge map.
- Iris Localization: In the work, in order to
increase the complete speed of the system,
circle detection algorithm is used.
- Iris Normalization: After successfully
extracting the iris part from the eye picture, in
order to allow comparisons between
dissimilar irises, transform the extracted iris
area so that it has a fixed dimension, and
hence, removing the dimensional
inconsistencies between eye pictures due to
the stretching of the iris caused by the pupil
dilation from varying levels of illumination.
- Iris Feature Extraction: This is the most key
part of an iris recognition system and
determines the system’s performance to a
great extent. Iris recognition produces the
exact result by extracting features of the input
picture and matching these features with
known designs in the feature database.

Table 1: Comparison between Techniques in Fingerprint and Iris Recognition System
AUTHOR FUSION DATA SET
Muhammad et al.[18] SCORE Face, Finger vein
Mohammed et al.[19] SCORE Face, Speech
Dhanashree et al.[20] FEATURE Palmprint,Palm vein
Rattan et.al[21] FEATURE Face, Fingerprint
Bhagat et al[22] FEATURE Palmvein,face
Feifei et al[23] SCOPE Fingerprint, vein
Krishneswari et al[24] FEATURE Fingerprint, palm print
Nazmeen et al[25] DECISION Face, ear
Nageshkumar et al[26] SCORE Palmprint,Face
Krzyszof et al[27] DECISION Face, Speech
Mohamed et al[28] DECISION Fingerprint, iris
Lin hong et al[29] DECISION Palmprint,face
Gayatri et al[30] FEATURE Face,Palmprint
Mitil et al[31] FEATURE Palm print, fingerprint
Jegadeesan et al[32] FEATURE Fingerprint, iris
VII. EXISTING APPROACH
A. The rule-based: Fusion method includes a
variety of simple rules for joining of
multimodal information. These include
statistical rule-depend methods like linear
weighted fusion (sum and product), MAX,
MIN, AND, OR, mainly voting. There are
custom-defined rules that are constructed for
the particular application perspective. The
rule-based patterns generally perform well if
the quality of temporal alignment between
dissimilar modalities is good.[20]
B. Classification-based fusion methods: This
category of techniques includes a range of
classification methods that have been used to
categorize the multimodal observation into one
of the pre-defined classes. The techniques in
this category are the(SVM) support vector
machine, Bayesian inference, Dumpster–
Shafer theory, active Bayesian networks,
neural networks and maximum entropy model.
We can further classify these methods as
generative and discriminative models from the
device learning perspective. For example,
Bayesian inference and dynamic Bayesian
links are generative models, while support
vector machine and neural networks are
discriminative models.
C. Estimation-based fusion methods: The
estimation classification includes the Kalman
filter, extended Kalman filter and particle filter
fusion methods. These techniques have been
primarily used to better estimate the state of a
moving object based on multimodal data. For
example, for the duty of object tracking,
multiple modalities such as audio and video
are fused to estimate the position of the object.
D. Scale Invariant Feature Transformation: The
SIFT approach, for picture highlight era, takes
a picture and changes it into an "expansive
meeting of neighbourhood highlight vectors"
Each of these highlight vectors is invariant to
any scaling, revolution or interpretation of the
picture. This methodology offers numerous
highlights with neuron reactions in primate
vision. To help the extraction of these
highlights the SIFT calculation applies a 4
stage separating methodology:
- Scale Space
- Localization
- Assignment and Orientation [33]
- Key point Descriptor
E. RSA Algorithm:The RSA algorithm is based on
the assumption that integer factorization is a
difficult problem. This means that given a
large value n, it is problematic to find the
prime factors that make up n. It is most
popular asymmetric key algorithm[34].
VIII. CONCLUSION
Biometric features are unique to each individual
and remain unaltered throughout a person’s
lifetime. These features made biometrics a
promising solution to the society. In this paper, a
robust multimodal biometric recognition structure
integrating iris and fingerprint is studied. Fusion of
two biometric traits is carried out at the match
score level. The domain of multi biometrics is a
new and exciting area of information science study
which is directed towards understanding of traits
and methods for accurate and reliable personal data
representation for subsequent decision making and
matching. In the current years there is a significant
increase in study activity directed at understanding
all aspects of biometric data system representation
and utilization for decision-making support, for use

through public and security services, and for
understanding the difficult processes behind
biometric matching and recognition. Uni-modal has
some disadvantages so we can go for multi-modal
biometric system. In order to increase the
complexity to the im-poster and to the improve the
accuracy, we can go for Multi-modal biometrics.
In Future Scope, Future works could go in the
direction of using Genetic algorithm or ICA in
hybridization with BFO. Independent Component
Analysis (ICA) is a computational method to get
hidden values of random variables. ICA is basically
designed for multivariate data.
REFERENCES
[1] Patil, Savitri B. "A Study of Biometric,
Multimodal Biometric Systems: Fusion
Techniques, Applications and
Challenges." IJCST 3, no. 1 (2012): 524-
526.
[2] Rodrigues, Ricardo N., Lee Luan Ling, and
VenuGovindaraju. "Robustness of
multimodal biometric fusion methods
against spoof attacks." Journal of Visual
Languages & Computing 20, no. 3 (2009):
169-179.
[3] Nageshkumar, M., P. K. Mahesh, and MN
ShanmukhaSwamy. "An efficient secure
multimodal biometric fusion using palmprint
and face image." arXiv preprint
arXiv:0909.2373 (2009).
[4] Gawande, Ujwalla, AnushreeSapre, Apurva
Jain, SanchitaBhriegu, and Shruti Sharma.
"Fingerprint-Iris Fusion Based Multimodal
Biometric System Using Single Hamming
Distance Matcher." International Journal of
Engineering Inventions e-ISSN: 2278-7461.
[5] Sree, SR Soruba, and Dr N. Radha. "A
survey on fusion techniques for multimodal
biometric identification." International
Journal of Innovative Research in Computer
and Communication Engineering 2, no. 12
(2014).
[6] Nair, S. Anu H. "Analysis of Image Fusion
Techniques for fingerprint Palmprint
Multimodal Biometric
System." International Journal of
engineering Research and Applications 1,
no. 5: 77-83.
[7] Kaur, Dapinder, and GaganpreetKaur.
"Level of fusion in multimodal biometrics: a
review." International Journal of Advanced
Research in Computer Science and Software
Engineering 3, no. 2 (2013): 242-246.
[8] W. Yunhong, T. Tan, & A. K. Jain,
Combining Face and Iris Biometrics for
Identity Verification, Proceedings of Fourth
International Conference on AVBPA,
Guildford, UK, 2003, 805-813.
[9] S. C. Dass, K. Nandakumar, & A. K. Jain, A
Principled Approach to Score Level Fusion
in Multimodal Biometric Systems, Proc. of
Audio- and Video-based Biometric Person
Authentication (AVBPA), Rye Brook, NY,
2005
[10] Phalguni Gupta, AjitaRattani,
HunnyMehrotra, Anil Kumar Kaushik,
Multimodal Biometrics System for Efficient
Human Recognition, Indian Institute of
Technology Kanpur, India – 208016.
[11] SangramBana, Dr.DavinderKaur,
Fingerprint Recognition using Image
Segmentation, (IJAEST)
INTERNATIONAL JOURNAL OF
ADVANCED ENGINEERING SCIENCES
AND TECHNOLOGIES Vol No. 5, Issue
No. 1, 012 – 023.
[12] Masek, L. (2003), ―Recognition of Human
Iris Patterns ForBiometric identification‖,
Thesis Report, The University ofWestern
Australia.
[13] Ghoualmi, Lamis, SalimChikhi, and
AmerDraa. "A SIFT-Based Feature Level
Fusion of Iris and Ear
Biometrics." Multimodal Pattern
Recognition of Social Signals in Human-
Computer-Interaction. Springer International
Publishing, 2015. 102-112.
[14] Bharadwaj, Samarth, MayankVatsa, and
Richa Singh. "Biometric quality: a review of
fingerprint, iris, and face." EURASIP
Journal on Image and Video
Processing 2014.1 (2014): 1-28.
[15] Conti, Vincenzo, et al. "Fingerprint and Iris
Based Authentication in Inter-cooperative
Emerging e-Infrastructures." Internet of
Things and Inter-cooperative Computational
Technologies for Collective Intelligence.
Springer Berlin Heidelberg, 2013. 433-462.
[16] Bakshi, Sambit, et al. "Score level fusion of
SIFT and SURF for iris."Devices, Circuits
and Systems (ICDCS), 2012 International
Conference on. IEEE, 2012.
[17] Conti, Vincenzo, Salvatore Vitabile, and
FilippoSorbello. "Fingerprint traits and rsa
algorithm fusion technique." Complex,
Intelligent and Software Intensive Systems
(CISIS), 2012 Sixth International
Conference on. IEEE, 2012.
[18] Muhammed Imran Razzak, RubiyahYuosf
and Marzuki Khalid, “Multimodal face and
finger veins biometric authentication”,
ScientificResearch and Essays, Vol.5,
No.17, pp. 2529-2534, 2010.

[19] Mohammed soltane,
NoureddineDoghmane,”Face and speech
based multimodal biometric authentication”,
International journal of advancescience and
technology, Vol 21, No.8, pp 41-46, 2010.
[20] Dhanashreevaidhya, sheetalpawar, “Feature
level fusion of palmprint and palm vein for
personal authentication based on Entrophy
technique”,International Journal on
Electronics and communication Technology,
Vol.5, Issue spl-1, 2014.
[21] A.Rattani, D.R.Kishu, M.Bicego,” Feature
level fusion of face and fingerprint
Biometrics”, Biometrics: Theory,
applications and systems, FirstIEEE
International conference, 2007.
[22] S.F.Bahgat, S.Ghoniemy, M.Alotabi,”
Proposed Multimodal palm-veins- face
biometric Authentication”, International
journal of advancedcomputer science and
applications, vol-4, No.6, 2013.
[23] Feifei cui, Gongping yang, “Score level
fusion of fingerprint and finger vein
Recognition”, Journal of Computer
Information’s systems: 16, 5723-5731, 2011.
[24] Krishneswari K, Arumugam S,”
Multimodal Biometrics using feature
fusion”, Journal of computer science
8(3):431-435, 2012.
[25] Nazmeenbibiboodoo, R.K.Subramanian,
“Robust multi-biometric recognition using
face and ear images”, IJCSIS-International
journal ofcomputer science and information
security, Vol.6, No.2, 2009.
[26] Nageshkumar, Mahesh.PK, Shanmuka
swami M.N, “A Efficient Multimodal
biometric fusion using palmprint and a face
image”, InternationalJournal of computer
science, Vol.2, Issue 3, 2009.
[27] Krzyszof, Jonas Richard, PlamenProdanov,
AndrzejnDrygajlo, “Reliability- Based
decision fusion in multimodal biometric
verificationsystems”, EURASIP Journal on
advances in signal processing, Article ID
86572, 9 Pages, 2007.
[28] Mohamad Abdolahi, Majid Mohamadi,
Mehdi Jafari,” Multimodal biometric system
fusion using fingerprint and iris with fuzzy
logic”,International Journal of soft
computing and engineering, Vol.2, Issue-6,
2013.
[29] Lin Hong, Anil Jain, “Integrating faces and
fingerprints for personal identification for
personal identification”, IEEE Transactions
on patternanalysis and machine intelligence,
Vol.20, No.12, 2008.
[30] Gayathriumakantbokade, ashokM.sapkal,
“Feature level fusion of palm and face for
secure recognition”, International Journal of
Computerand Electrical Engineering, Vol.4,
No.2, 2012.
[31] Mitul D. Dhameliya, JitendraP.Chaudri,” A
multimodal biometric recognition system
based on fusion of palmprint and
fingerprint”,International journal of
engineering trends, Vol.4, Issue 5, 2013.
[32] A.Jagadessan, Dr.K.Duraisamy,”Secured
Cryptographic Key Generation from
multimodal biometrics: Feature level fusion
of fingerprint andIris”, International Journal
of computer science and information
security, Vol.7, No.2, 2010.
[33] Bicego, Manuele, Andrea Lagorio, Enrico
Grosso, and Massimo Tistarelli. "On the use
of SIFT features for face authentication."
In Computer Vision and Pattern Recognition
Workshop, 2006. CVPRW'06. Conference
on, pp. 35-35. IEEE, 2006.
[34] Bicego, Manuele, Andrea Lagorio, Enrico
Grosso, and Massimo Tistarelli. "On the use
of SIFT features for face authentication."
In Computer Vision and Pattern Recognition
Workshop, 2006. CVPRW'06. Conference
on, pp. 35-35. IEEE, 2006.

The Survey of Architecture of Multi-Modal (Fingerprint and Iris Recognition) Biometric Authentication System

More Related Content

What's hot (19)

Similar to The Survey of Architecture of Multi-Modal (Fingerprint and Iris Recognition) Biometric Authentication System (20)

Recently uploaded (20)

The Survey of Architecture of Multi-Modal (Fingerprint and Iris Recognition) Biometric Authentication System