Sei sulla pagina 1di 16

Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.

2010, July 5th-9th, Ota, Nigeria Vol 1

Enhanced Persons Authentication: A Multimodal


Biometrics Approach
A. F.Thompson1, B. K. Alese1, O. S. Adewale1 and O. S.Falaki1
1
Department of Computer Science, Federal University of Technology, Akure.
P.M.B. 704, Akure.
ronkeniran@yahoo.com

Abstract. Biometrics is a new trend of information security technology in todays Computer


network society. Determining persons authentication is becoming critical in our vastly
interconnected information society. Most biometric systems that are presently in use, typically use a
single biometric trait to establish identity, they are known as unimodal biometric systems. There
exist some limitations with these systems. The later could be solved efficiently, using multimodal
biometrics. In the light of this, some of the limitations are discussed and a Face, Fingerprint and ear
verification algorithm based on Classifier scheme with decision-level fusion is presented.

Keywords: authentication, unimodal biometrics, multimodal biometrics, decision-level


fusion model

1 Introduction
The dependence of people on computers has increased tremendously in recent years.
Many applications rely heavily on the effective operations of their computer systems and
networks. Thus, the need to authenticate ourselves to machines is ever increasing in
todays networked society [42]; it is necessary to close the air gap between man [2, 3].
[52] described authentication main techniques as Knowledge-based systems, token-based
systems, and biometrics-based systems. A paper [9] asserted that, knowledge-based
schemes are predominantly used for authentication while [4, 68] described the traditional
methods are using password or cryptography to protect information. These techniques are
faced with several challenges on confidentiality, integrity and authentication while in
storage or in transit. Thus, Information security has become a more and more pressing
issue [65; 79, 4, 1; 58].

Biometrics, have witnessed a fast development in the last decades with applications
ranging from accessing ones computer to obtaining visa for international travel [6]. The
author [76] described Biometrics technology as relying on the body as password for
human recognition purposes to provide better security, increased efficiency and improved
service [16, 23, 53, 44]. As the technology becomes more economically viable,
technically perfected and widely deployed, biometrics could become the passwords and
PINs of the twenty-first century [40, 41]. In concrete terms, [60] affirmed the influence of
this technology; it translates into about $1 billion worth of computer systems that include
biometric devices which were estimated to be installed worldwide during 1997. [15, 76]
described categories and application areas such as Point-of-Sale (POS) verification,

167
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

automotive, health care. Law Enforcement, Prison Management, Military & Border
Control & Immigration Checks, National Identity Card & Voter Registration.

Authors in [54, 79, 52, 45] as well, define biometric as the most secure and
convenient authentication tool that relies on the automated identification or verification of
an individual. It cannot be borrowed, stolen, or forgotten and forging one is practically
impossible. Biometrics measure individual's unique physical (physiological characteristics
refer to inherited traits that are formed in the early embryonic stages of human
development) or behavioural characteristics (not inherited, but learned) to recognize or
authenticate their identity. Biometric systems can be categorized into two.

(i) Unimodal Biometric Systems


Unimodal biometric system uses only a single biometric characteristic. This system is
usually more cost-efficient than a multimodal biometric system. However, it may not
always be applicable in a given domain because of unacceptable performance and
inability to operate on a large user population (34). Examples are: face recognition: [72];
fingerprint matching: [28]; Hand Geometry: [63]; Palm Prints: [25]; Dental: [27]; On-line
Signature: [29].

(ii) Multimodal Biometric Systems


Multimodal Biometric system conbines multiple sources of biometric traits. This can be
accomplished by fusing multiple traits of an individual, or multiple feature extraction and
matching algorithms operating on the same biometric. These multimodal systems [62, 38,
35], can improve the matching accuracy of a biometric system while increasing
population coverage and deterring spoof attacks [39].

Some of the earliest multimodal biometric systems utilized face and voice features
[13, 11]; Iris and Face [64]; Fingerprint and speaker [70]; Face and Fingerprint [34]; Face
and Hand [71]; face and Fingerprint [12]; face and ear [67]; Scores fusion [5]. Physically
uncorrelated traits (e.g., fingerprint and iris) are expected to result in better improvement
in performance than correlated traits (e.g., voice and lip movement).

Persons authentication is associating identity with an individual. [43], categorized


persons identity into two fundamentally distinct types of problems with different inherent
complexities:

(i) Verification (authentication) confirms or denys a persons claimed identity (Am I


who I claim to be? or "Is this person whom he/she claims to be?") Verification is a
one-to-one comparison of the biometric sample with the reference template on file;
this is illustrated in [34].
(ii) recognition (more popularly known as identification) Who am I? or "Who is the
person?" attempts to establish a persons identity - either from a set of already known
identities (closed identification problem) or otherwise (open identification problem).

168
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

Identification makes a one-to-many comparison [34] to determine a user's identity. It


checks a biometric sample against all the reference templates on file. If any of the
templates on file match the biometric sample, there is a good probability that the
positive personal identification of such individual has been identified.
2 Unimodal Biometrics Limitations
Most biometrics systems deployed in the real world are unimodal. It relies on a single
biometric characteristic which could be hacked by the modern buglars. There are some
challenges posed against unimodal biometrics system. Among such are:
Unacceptable error rate due to deformation or aging, noise in the signals produced,
intra-class variation. [61, 43]. Noisy data can result in an individual being incorrectly
labeled as an impostor thereby increasing the False Reject Rate (FRR) of the system. The
biometric system may not be able to acquire meaningful biometric data from a subset of
individuals resulting in a failureto-enroll (FTE) error, this is known as non-universality.
For example, a fingerprint system may fail to image the friction ridge structure of some
individuals due to the poor quality of their fingerprints. Similarly, an iris recognition
system may be unable to obtain the iris information of a subject with long eyelashes,
drooping eyelids or certain pathological conditions of the eye.

Tuning the feature extraction and matching modules cannot continuously improve the
matching performance of a unimodal biometric system.[42]. There is an implicit upper
bound on the number of distinguishable patterns (i.e., the number of distinct biometric
feature sets) that can be represented using a template. The capacity of a template is
constrained by the variations observed in the feature set of each subject (i.e., intra-class
variations) and the variations between feature sets of different subjects (i.e., inter-class
variations).

Spoof attacks: Behavioral traits such as voice [26] and signature [32]; [29] are
vulnerable to spoof attacks by an impostor attempting to mimic the traits corresponding to
legitimately enrolled subjects. Physical traits such as fingerprints can also be spoofed by
inscribing ridge-like structures on synthetic material such as gelatine and play-doh [50,
56]. Targeted spoof attacks can undermine the security afforded by the biometric system
and, consequently, mitigate its benefits [59]. See figure 1.

Figure 1: Modern buglars Biometrics Threats (http://scgwww.epfl.ch/)

169
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

Similarly, to consider a pocketbook example, the worlds financial community has


long been concerned about growing problems of ATM fraud and unauthorized account
access, estimated to cost $400 million a year [31]. Passwords or PINs may be forgotten or
guessed by the imposters.

Credit card fraud is estimated at $2 billion per year. The financial services industry
believes that a significant percentage of these losses could be eliminated by biometric
scanning. Critics give too much credit to biometrics alleged ability to erode anonymity.
Multimodal will secure the financial communities and e commerce.

Issues highlighted by [76] on Biometric Centralization, Biometric Balkanization and


The Big Brother concern could be offered solution by deploying multimodal biometrics.
The Big Brother concern is using a biometric feature against an individual, would be
verified at the fusion decision level, thus, proving otherwise 1984-like worst-case scenario
offered by Professor Clarke that Any high-integrity identifier [such as biometric
scanning] would make all human behaviour become transparent would caution citizens
to defy committing crime.

From the above scenarios, some of the limitations of a unimodal biometric system
can be addressed by designing a system that consolidates multiple sources of biometric
information.

3 Background: Voronoi diagram and Delaunay Tessellation


A commonly used term in computational geometry is the Voronoi diagram and Delaunay
Tessellation. A generalized Voronoi diagram (GVD) for a set of objects in space is the set
of generalized Voronoi regions.
Input:A setof points (sites)

Output:A partitioning of the plane into regions of equal nearest neighbors as shown
below. So, for each unimodal biometrics, the feature extraction would be done based on
the VDs. [66].

Figure 2a : Voronoi diagram

170
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

(1)
it follows that

(2)

Figure 2b: Voronoi Diagrams

A generalized Delaunay tessellation (triangulation in 2d) is the dual of the generalized


Voronoi diagram obtained by joining all pairs of sites whose Voronoi regions share a
common Voronoi edge according to some specific rule.

Figure 3: a) Voronoi Diagram in 2D b) VD and DT in Power metric

3.1 Generalized Distance metrics for


1/ p
Voronoi Diagrams
p
d
d ( x, p) xi pi
General Lp distance i1


(3)
Manhattan d (4)
d ( x, p ) xi p i
i 1
Supremum d (x, p) max xi pi (5)
i 1..d

Manalanobis d ( x, p) ( x p )Cov ( D ) 1 ( x p ) (6)

171
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

3.2 Properties of Voronoi Diagram and Delaunay Triangulation [30]

1. The vertex of generalized Voronoi diagram is a center of a sphere inscribed


between d +1 Voronoi sites (spheres).
2. The inscribed sphere is empty, i.e. it does not contain any other sites.
3. One of the facets of the generalized Voronoi region Vor(P) defines a nearest-
neighbor of P.
4. A Voronoi region Vor(P) is unlimited if and only if site P belongs to the convex
hull of S.
5. The sphere inscribed between the sites comprising a simplex of generalized
Delaunay tessellation is an empty sphere.
6. The power Delaunay tessellation of a set of spheres S is a tetrahedrization.

3.3 DT for matching


Delaunay Triangulation can be used for Matching: For each Delaunay triangle, the length
of three edges, the three angles and the ridge numbers between each edge are recorded to
construct a 9 dimensional local vector to find the best-matched local structure in two
fingerprints

Figure 4 :DT for matching [30]

4 Modalities and Database


The generic biometric technology or combinations of technologies use these basic
procedures [34]. The proposed method processes each face, ear and fingerprint data set
through a common pipeline of algorithms. This technique represents each respective
biometrics classes and is purely geometrical. Registrations of each set and transmission
process are adopted to acquire its shape. A 2-dimension image is extracted and a feature
extraction matching is done and the template stored. The steps that each face, ear or
fingerprint dataset goes through are:

(1) Preprocessing : Raw data preprocessed and segmented to eliminate sensor


specific problems
(2) Signal Processing: Stable and distinctive features are extracted from the received
signal
(3) Analysis: Storage of templates derived and decision, which accepts or
rejects based upon the system policy and the scores received from the signal
processing system.

172
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

In practice, fusion of two or more independent biometrics has proven powerful for
increasing the performance of biometric systems (ibid).

4.1 Database
A database containing face, ear and fingerprints of 50 individuals is proposed for this
study. The test subjects would be between 17 and 55 years of age, female and male ratio
would also be considered. A two-phase data collection regarded as enrollment and
experimental stages form the database for this study. Fingerprint database contains the
right index finger, for each test subject. The face is taken with frontal view using digital
camera of high resolution. All the test data from real test subjects adopts [33] method
which is different from the database method used in [61, 74]. Although, independent
assumed database modalities are valid.

4.1.1 Fingerprint method


Fingerprints are probably the most widely used biometric modality. They are well-known
and they have a long history in forensics [8]. Fingerprints are universal to all people (with
exceptions to injured or disability) unique and stable throughout a persons life. The
fingerprints are generally collected with proper sensors, but typically there is a fraction of
people, for whom the capture of fingerprints with a certain sensor is very difficult or even
impossible [43]. Forensic fingerprint recognition systems have shown excellent
recognition performance. Most state-of-the-art fingerprint recognition systems follow a
minutiae-based approach [48]. The minutiae are the bifurcations and endpoints of the
ridge lines.

4.1.2 Ear method

The Ear is segmented using the method proposed in [78], which is based on the landmark
selection. The rationale for using landmarks is that the position of the landmark is stable
over time for a particular ear. The searched landmarks are the Triangular Fossa and the
Incisure Intertragica (Fig. 5). Since the dimensions of the segmented images can vary,
thus, re-sizing is done. In order to improve the quality of the input image a preprocessing
[47] method is adopted using Matlab 7.0

Fig. 5. Landmarks: Triangular Fossa and Incisure Intertragica [47]

173
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

The preprocessing method proposed in [81] and stated in [47] is the following: let P(x, y)
be the pixel value in the range [0, 255] at the coordinate (x, y), while m and v be the
image mean and variance, respectively. The normalized image is computed by using the
operation below:

(7)

(8)

Where, mt and vt are the pre-set values for mean and variance for the normalized image.

4.1.3 Face method

Face recognition methods can be classified into three groups [69, 80]: local feature-based
methods, holistic or appearance-based methods and mixed methods. We employ the local
feature-based methods. Local features such as the eyes, the nose and the mouth are first
extracted and their locations (geometrical and/or appearance) are used for recognition.
Examples are structural matching, elastic and bunch graph matching. Structural matching
is used in some of the earliest methods [21] where relations between the feature points are
used as descriptors. In elastic bunch graph matching as introduced by [75] first a sparse
grid is overlaid on the facial image and its nodes are adjusted to a set of fiducial points.

4.2 Multimodal Biometrics Algorithm

We propose Voronoi Diagram to extract all the features of each of the biometrics. The
methodology is making its way to the core methods of biometrics, such as fingerprint
identification, iris and retina matching, face analysis, ear geometry and others [14]. The
methods are using Voronoi diagram to partition the area of a studies image and compute
some important features (such as areas of Voronoi region, boundary simplification etc.)
and compare with similarly obtained characteristics of other biometric data.

4.2.1 Production of Voronoi Diagrams Methods

A 2-dimensional tessellation, known as a Voronoi diagram, may be produced from any


planar region by the following method [55]:

1. A Finite collection of seed points is produced. These points may either be generated
by a stochastic process, or acquired from a preexisting data source.

174
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

2. To each seed point is associated a planar sub-region known as a cell. Each cell
consists of all points which are closer to the given seed point than any other seed
point in the collection. This tessellation is uniquely determined by the seed points.

Our work considers the Voronoi diagram obtained from the minutiae of a given right
index finger, which we refer to as a minutiae Voronoi diagram; right ear and face frontal
view using the Qhull algorithm [57].

4.2.2 Proposed Algorithm for Multimodal Biometrics (Using real dataset


templates)

For Multimodal Biometrics, MB = 1 to 3


Check if Image {face, ear or fingerprint}
Extract Features at the Minutiae level using Voronoi Diagrams. (a form of a non
reversible compression. This is to exploit effective topology of Delaunay
Triangulations.) using [57]
Do minutiae-based matching using triangular matching. (Find the alignment between
the templates and the input minutiae sets that result in the maximum number of
minutiae pairings.)
Compute the Delaunay triangulation of minutiae sets Q and P.
Use triangle edge to compare index. To compare two edges, Length, 1, 2.
Find thresholds of the multimodal units. (compute using [51] approach)
Integrate in orthogonal modes (i.e., statistically independent)
While ALL minutiae-based matching is done (This is Segmentation, known as the
process of isolating foreground from background)

Accessibility i.e. easy to image using electronic sensors or camera and Acceptability i.e.
people do not object to having this measurement taken on them serve as the pivot of the
proposed multimodal biometric features. They also are robust, distinctive and available. In
reality due to the variation in the captured signals (noise, etc.) errors are made by the
decision module. Thus,. Failure to Enroll (FTE), the most important quality measures of a
biometric system. (FTE measures the availability of a biometric modality.). Failure to
Acquire (FTA), in some cases a usable biometric sample cannot be presented or acquired
for reasons of availability or for reasons of technical incapability. (FTA includes FTE.).
False Match Rate (FMR) and other metrics for performance, evaluation and quality would
be evaluated using [9] performance, accuracy and quality metrics.

4.3 Decision-level fusion

In literature, the following methods proposed for decision level fusion include AND and
OR rules [22], majority voting and weighted voting [46, 61), Bayesian decision fusion,
the Dempster-Shafer theory of evidence [77] and behavior knowledge space [37].

175
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

In this work,we propose the Bayesian decision fusion scheme which relies on
transforming the discrete decision labels output by the individual matchers into
continuous probability values. Confusion matrix will be generated for each matcher by
applying the matcher to a training set D. Let M denote the number of possible decisions
(also known as class labels or simply classes in the pattern recognition literature) in a
biometric system. Also, let 1, 2, . . . M indicate the classes associated with each of
these decisions. Let CMj be the M M confusion matrix for the jth matcher. The (k, r)th
element of the matrix CMj (denoted as cmj k,r) = the number of instances in the training
data set where a pattern whose true class label is k is assigned to the class r by the jth
matcher. Let the total number of data instances in D be N and the number of elements that
belong to class k be Nk. The Bayes decision fusion technique chooses that class which
has the largest value of discriminant function calculated. Assume conditional
independence between the different matchers; the decision rule is known as naive Bayes
rule and P (c|k) is computed as

P (c|k) = P (c1, . . . , cR|k) =


(9)

The accuracy of the naive Bayes decision fusion rule has been found to be fairly
robust even when the matchers are not independent [24].

Table 1: Practical Evidence: Multimodal Biometrics Fusion and Related Works

Author Method Result

Ross and Scores Fusion of face, The fusion result was significantly better than any
Jain [61] fingerprint and hand geometry single modality, e.g. with fixed FAR of 0.03% the
FRR for best single modality, the fingerprint was
more than 25% while for the combination of the
three modalities it was less than 2%.

Wang et Score fusion of Face and iris The experiment shows that combination of the
al. [74] biometric scores face and iris recognition gives more robust
performance than each of them separately.

Chang et Image fusion of face and ear Their experiments show 90.9% rank-one
al. [17] images, not the scores. recognition rate for multimodal biometrics while
face alone produces 70.5% and ear alone 71.6%.
The training set contained 197 subjects.

176
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

Ben- compared support vector In their case, support vector machines and
Yacoub et machines, multilayer Bayesian classifier performed best. [61] focused
al. [10] perceptron, C4.5 decision tree, on matching score level, and in their study simple
Fisher_s linear discriminant, SUM rule was the best classifier
and Bayesian classifier for
fusion of face and speech
data.

5 Conclusion and Further Work


Security techniques should be constantly appraised. This is to sidetrack the hackers that
devise new ways of bridging securities perpetually. Consequently, this paper has brought
to limelight, some of the limitations of single biometrics; and proposed a Multimodal
system which alleviates some of the problems imposed by single-biometric systems. The
proposed Biometric fusion of face, fingerprint and ear is at the decision level. Further
work, which is on-going, is to implement the proposed algorithm using Java programming
Language and evaluate its performance vis--vis Face, Fingerprint and Ear unimodal
biometrics system.

References
1. Adetunmbi O.A. (2007), Intrusion Detection Using Machine Learning TechniquesPh.D.
Thesis, Federal University of Technology, Akure, Nigeria.
2. Adewale, O S and Falaki S O (2001): Electronic Commerce Promises, Threats, Trust and
Payment Systems. Proceedings of the 17th National Computer Association of Nigeria
Conference at Benin-City, June 2001, 12, pp. 84 103
3. Alese B.K. and Falaki S.O. (2004b), Security Computer and Computer Networks against
Intruders, Expertus, Journal of Sustainable Development, 2(1): 12-24.
4. Alese B.K., (2004a) Design of Public Key Crypto System using Elliptic Curve Ph.D.
Thesis, Federal University of Technology, Akure, Nigeria.
5. Alsaade, F., Ariyaeeinia, A.M., Malegaonkar, A.S., Pillay, S.G. (2009), Qualitative fusion of
normalised scores in multimodal biometrics, PRL(30), No. 5, 1 April 2009, pp. 564-569.
6. Anil K. J., Patrick J. Flynn and Arun Ross (2008), Hand book of Biometrics, Springer.
7. Barthel M.(1995), Banks Eyeball Sci-Fi Style Identification for ATMs, American Banker,
Sept. 22, 1995.
8. Beavan, C., 2001. Fingerprints. Fourth Estate, London.
9. Ben Schouten and Bart Jacobs, (2009), Biometrics and their use in e-passports, Image and
Vision Computing 27 (2009) 305312

177
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

10. Ben-Yacoub S., Abdeljaoued Y, and Mayoraz e. (1999). Fusion of Face and Speech data
for Person Identity Verification. IEEE Transactions on Neural Networks, 10(5):10651075,
September 1999.
11. Bigun E. S., Bigun J., Duc B., and Fischer S.. (1997), Expert Conciliation for Multimodal
Person Authentication Systems using Bayesian Statistics. In First International Conference
on Audio- and Video-based Biometric Person Authentication (AVBPA), pages 291300,
Crans-Montana, Switzerland, March 1997.
12. Bouchaffra, D., Amira, A., Structural hidden Markov models for biometrics: Fusion of face
and fingerprint, PR(41), No. 3, March 2008, pp. 852-867.
13. Brunelli R. and Falavigna D.(1995), Person Identification Using Multiple Cues. IEEE
Transactions on Pattern Analysis and Machine Intelligence, 17(10):955966, October 1995.
14. Burge, M., Burger, W., 2000. Ear biometrics in computer vision. Proc.Internat. Conf. Pattern
Recognition 2, 822826
15. Cavoukian Ann and Stoianov Alex, (2007), Biometric Encryption: A Positive-Sum
Technology That Achieves Strong Authentication, Security, and Privacy In Privacy by
Design Book available at www.ipc.on.ca. Accessed on December 06, 2009.
16. Chandrasekaran R. (1997), Brave New Whorl: ID Systems Using the Human Body Are
Here, But Privacy Issues Persist, Washington Post, March 30, 1997.
17. Chang, K., Bowyer, K., Sarkar, S., Victor, B., (2003), Comparison and combination of ear
and face biometrics in appearance based biometrics. IEEE Trans. Pattern Anal. Machine
Intell. 25 (9), 11601165.
18. Chatzis V., Bors A. G., and Pitas I. (1999). Multimodal Decision-level Fusion for Person
Authentication. IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and
Humans, 29(6):674681, November 1999.
19. Chibelushi C. C., Mason J. S. D., and Deravi F.. (1997), Feature-level Data Fusion for
Bimodal Person Recognition. In Proceedings of the Sixth International Conference on Image
Processing and Its Applications, volume 1, pages 399403, Dublin, Ireland.
20. Comparison and combination of ear and face images in appearance-based biometrics,
PAMI(25), No. 9, September 2003, pp. 1160-1165.
21. Cox I.J., Ghosn J., Yianilos P.N., (1996), Feature-based face recognition using
mixturedistance, in: Proc. Int. Conf. Comput. Vis. Patt. Recogn., 1996, pp. 209216.
22. Daugman J. (2000), Combining Multiple Biometrics. Available at
http://www.cl.cam.ac.uk/users/jgd1000/combine/combine.html.
23. Davis A. (1997), The Body as Password, Wired, July 1997.
24. Domingos P. and Pazzani M.(1997), On the Optimality of the Simple Bayesian Classifier
under Zero-One Loss. Machine Learning, 29(2-3):103130.
25. Duta N., Jain A. K., and Mardia Kanti V., "Matching of Palmprint ", (2002) Pattern
Recognition Letters, vol. 23, Number 4, pp. 477-485..
26. Eriksson A. and Wretling P.. How Flexible is the Human Voice? A Case Study of Mimicry.
In Proceedings of the European Conference on Speech Technology, pages 10431046,
Rhodes, 1997.
27. Fahmy G., Nassar D., Haj-Said E., Chen H., Nomir O., Zhou J., Howell R., Ammar H. H.,
Abdel-Mottaleb M. and Jain A. K.,(2004) " Towards an Automated Dental Identification
System (ADIS)", Proc. of the International Conference on Biometric Authentication (ICBA),
Hong Kong, July.
28. Feng J., Yoon S., and Jain A. K., (2009) "Latent Fingerprint Matching: Fusion of Rolled and
Plain Fingerprints", Proc. International Conference on Biometrics (ICB), June.

178
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

29. Fierrez -Aguilar J., Krawczyk S., Ortega-Garcia J. and Jain A. K.,(2005) " Fusion of local and
regional approaches for on-line signature verification", Proc. International Workshop on
Biometric Recognition Systems (IWBRS), pp. 188-196, Beijing, China..
30. Gavrilova Marina L., (2009), Geometric Algorithms in Biometrics: Theory and Recent
Developments, Lecture Notes (GEOIDE), BT Laboratory Dept of Computer Science,
University of Calgary, Calgary, AB, Canada, T2N1N4 . Accessed April, 2009.
31. Hall J. (1995), For New ATM, the Eyes Have It, Trenton Times, September 19, 1995.
32. Harrison W. R. (1981), Suspect Documents, their Scientific Examination. Nelson-Hall
Publishers, 1981.
33. Heikki a., Elana V., Mikko L., Satu-Marja M. and Johannes P., (2006), Pattern Recognition
Letters, 325-334
34. Hong L. and Jain A., (1998) Integrating faces and fingerprints for personal identification, In
Proceedings 3d Asian Conference on Computer Vision, pp. 16-23 Hong Kong, China.
35. Hong L., Jain A. K., and Pankanti S.. (1999), Can Multibiometrics Improve Performance?
In Proceedings of IEEE Workshop on Automatic Identification Advanced Technologies
(AutoID), pages 5964, New Jersey, USA.
36. Hong, L.and Jain, A.K. (1998), Integrating Faces and Fingerprints for Personal Identification,
PAMI(20), No. 12, December 1998, pp. 1295-1307.
37. Huang Y. S. and Suen C. Y. (1995), Method of Combining Multiple Experts for the
Recognition of Unconstrained Handwritten Numerals. IEEE Transactions onPattern Analysis
and Machine Intelligence, 17(1):9094.
38. Jain A. K. and Ross A. (2004), Multibiometric Systems. Communications of the ACM,
Special Issue on Multimodal Interfaces, 47(1):3440.
39. Jain A. K. and Ross A., (2003), "Information Fusion in Biometrics", Pattern Recognition
letters, Vol. 24, pp. 2115-2125, September 2003.
40. Jain A. K. and Ross A., (2005) "Biometrics", in The Enyclopedia of Cryptography and
Security, (Henk C. A. van Tilborg, Ed.), Springer, August 2005.
41. Jain A. K., Ross A. and Pankanti S., (2006)" Biometrics: A Tool for Information Security",
IEEE Transactions on Information Forensics and Security Vol. 1, No. 2, pp. 125-143, June
2006.
42. Jain Anil, Bolle Ruud And Pankanti Sharath (2002), Introduction To Biometrics In
Biometrics Personal Identification In Networked Society, The Kluwer International Series In
Engineering And Computer Science, Kluwer Academic Publishers, New York.
43. Jain, A.K., Ross, Arun, Prabhakar, Salil, (2004a), . An introduction to biometric recognition.
IEEE Trans. Circ. Systems Video Technol. 14 (1), 421.
44. James F. (1997), Body Scans Could Make ID Process Truly Personal, Chicago Tribune,
June 4, 1997
45. Jonathan Coupal (2009), Biometric Security ITX Corp. available on WWW.ITX.NET
accessed on December 6, 2009.
46. Lam L. and Suen C. Y.. (1997), Application of Majority Voting to Pattern Recognition: An
Analysis of its Behavior and Performance. IEEE Transactions on Systems,Man, and
Cybernetics, Part A: Systems and Humans, 27(5):553568.
47. Loris N. and Alessandra L., (2007), A Multi-Matcher for Ear Authentication, Pattern
Recognition Letters 2219-2226.
48. Maltoni D., Maio D., Jain A.K., Prabhakar S., Handbook of Fingerprint Recognition, Springer
Verlag, 2003.
49. Manjunath B.S., Chellappa R., von der Marsburg C., (1992), A feature based approach to face
recognition, in: Proc. IEEE Conf. Comput. Vis. Patt. Recogn., 1992, pp. 373378.

179
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

50. Matsumoto T., Matsumoto H., Yamada K., and Hoshino S. (2002), Impact of Artificial
Gummy Fingers on Fingerprint Systems. In Optical Security and Counterfeit Deterrence
Techniques IV, Proceedings of SPIE, volume 4677, pages 275289, San Jose, USA.
51. Md. Mahburbur R., Md. Rshedul I., Nazmul I.B., Bulbul A. and Md Aminul I. (2007),
Personal Identification Using Ear Biometrics, Int. Journal of The Computer, Internet and
Management, 15, #2 pp 18.
52. Michael Down and Richard Sands, (2009), Risk and Control of Biometric Technologies: A
Security, Audit and Control Primer, The IT Governance Institute publication. Available on
www.isaca.org/biometric Accessed on 06 December, 2009.
53. Nanavati R. (1998), Presentation on Top Ten Trends in Biornetrics at CardTech/SecurTech
Conference, in Washington, D.C, April 27, 1998
54. Newham E. (1995), The Biometric Report. SJB Services, New York,
55. Okabe, Boots, Sugihara, and Chiu. 2000. Spatial Tessellation. West Sussex: Wiley.
56. Putte T. and Keuning J. (2002), Biometrical Fingerprint Recognition: Dont Get Your
Fingers Burned. In Proceedings of IFIP TC8/WG8.8 Fourth Working Conference on Smart
Card Research and Advanced Applications, pages 289303.
57. Qhull algorithm.. The Geometry Center, http://www.geom.uiuc.edu/. Accessed 16/4/10.
58. Qijun Zhao (2009),Advanced Information Security Technology: Watermarking and
Biometrics, ACM-HK Student Research and Day, 2009; Biometrics Research Centre,
Department of Computing, The Hong Kong Polytechnic University, 2766-7902, 00852,
csqjzhao@comp.polyu.edu.hk
59. Ratha N. K., Connell J. H., and Bolle R. M.. (2001) An Analysis of Minutiae Matching
Strength. In Proceedings of Third International Conference on Audio- and Video-Based
Biometric Person Authentication (AVBPA), pages 223228, Halmstad, Sweden.
60. Roethenbaugh G. (1997), Biometrics: A Global Perspective, in BiometriCon 97
Conference Proceedings, March 12-14, Arlington, VA..
61. Ross A. , Jain K., A., (2003), Learning user-specific parameters in a multibiometric system.
In: Proc. Internat. Conf. on Image Processing, Rochester, New York.
62. Ross A., Nandakumar K., and Jain A. K.. (2006), Handbook of Multibiometrics. Springer,
New York, USA, 1st edition.
63. Ross, Arun; "A Prototype Hand Geometry-based Verification System," M.S. Project Report,
1999, biometrics.cse.msu.edu/RossHand_MS99.pdf
64. Son, B.J., Lee, Y.B. (2006), The Fusion of Two User-friendly Biometric Modalities: Iris and
Face, IEICE(E89-D), No. 1, January 2006, pp. 372-376.
65. Stallings, W. (1999) Cryptography and Network Security Principles and Practice (second ed.).
Prentice-Hall Inc., Englewood Cliffs, NJ.
66. Subhas C. Nandy, (2007), Voronoi Diagram, Advanced Computing and Microelectronics
Unit Indian Statistical Institute Kolkata 700108, Lecture Notes.
67. Theoharis, T., Passalis, G., Toderici, G., Kakadiaris, I.A. (2008), Unified 3D face and ear
recognition using wavelets on geometry images, PR(41), No. 3, March 2008, pp. 796-804.
68. Thompson, A.F. (2005) Encryption /Decryption of Speech, M Tech Thesis, FUTA, Nigeria,
2005.
69. Tistarelli M. BioSecure consortium, (Ed.), Report on state of the art face recognition,
Deliverable 7.2.2, 2005. Available from: <http://www.biosecure.info/>.
70. Toh, K.A., Yau, W.Y. (2005), Fingerprint and speaker verification decisions fusion using a
functional link network, SMC-C(35), No. 3, August 2005, pp. 357-370.

180
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

71. Tsalakanidou, F., Malassiotis, S., Strintzis, M.G. (2007), A 3D face and hand biometric
system for robust user-friendly authentication, PRL(28), No. 16, December 2007, pp. 2238-
2249.
72. Unsang Park, Yiying Tong, and Anil K. Jain, (2008) "Face Recognition with Temporal
Invariance: A 3D Aging Model", 8th IEEE Int'l Conference on Automatic Face and Gesture
Recognition, Amsterdam, Netherlands, September.
73. Verlinde P. and Cholet G. (1999), Comparing Decision Fusion Paradigms using k-NN based
Classifiers, Decision Trees and Logistic Regression in a Multi-modal Identity Verification
Application. In Proceedings of Second International Conference on Audio- and Video-Based
Biometric Person Authentication (AVBPA), pages 188193, Washington D.C., USA..
74. Wang, Y., Tan, T., Jain, A.K., (2003), Combining face and iris biometrics for identity
verification. In: Proc. AVBPA_2003.
75. Wiskott L., Fellous J.M., Kruger N., von der Malsburg C., (1997) Face recognition by elastic
bunch graph matching, IEEE Transactions on Pattern Analysis and Machine Intelligence 19
(1997) 775779.
76. Woodward J. D. (1997), Biometric Scanning, Law & Policy: Identifying the Concerns;
Drafting the Biometric Blueprint, University of Pittsburgh Law Review, Fall 1997.
77. Xu L., Krzyzak A., and Suen C. Y. (1992), Methods for Combining Multiple Classifiers and
their Applications to Handwriting Recognition. IEEE Transactions on Systems, Man, and
Cybernetics, 22(3):418435.
78. Yan, P., Bowyer, K.W., (2005a). Empirical evaluation of advanced ear biometrics. Empirical
Evaluation Methods in Computer Vision (EEMCV 2005) (June), San Diego.
79. Zhang, D. 2000. Automated Biometrics: Technologies and Systems. Kluwer and Academic
Publishers, USA.
80. Zhao W., Chellappa R., Phillips P.J., (2003), A. Rosenfeld, Face recognition: a literature
survey, ACM Computing Surveys 35 (4) (2003) 399458.

181
Proceedings of the International Conference on Software Engineering and Intelligent Systems SEIS 2010.
2010, July 5th-9th, Ota, Nigeria Vol 1

182

Potrebbero piacerti anche