NEWSLETTER

NEWSLETTER
Volume 22, 2017
Content
Two New Vice President s
IEEE Biometrics Council welcomes two new Vice Presidents in
the Executive Committee. Dr. Sébastien Marcel and Prof. Pong C.
Yuen will be Vice Presidents of Conferences and Technical
Activities, respectively. The Council thank outgoing Vice
Presidents Prof. Sudeep Sarkar and Prof. Ioannis Kakadiaris for
their services and contributions.
- New Vice Presidents
- IEEE Trans. on Biometrics
- Biometrics News
- Biometrics Interview:
Kenneth D. Gantt
- Spotlight
- Research Paper Summary
- Conference Report:
BIOSIG2016
Prof. Yuen is a Professor and the Head of the Department of
Computer Science. His research interests include video
surveillance, human face recognition, and biometric security and
privacy. Dr. Yuen serves as a Hong Kong Research Grant Council
Engineering Panel Member. He is an Editorial Board Member of
Pattern Recognition and an Associate Editor of IEEE T-IFS and
SPIE Journal of Electronic Imaging. He was the Program Co-Chair
of the IEEE Fifth International Conference on Biometrics: Theory,
Applications and Systems in 2012.
LOREM IPSUM DOLOR SIT AMET
Dr. Sébastien Marcel is a Senior Research Scientist with the Idiap
Research Institute, where he leads the Biometrics Group and
conducts research on multimodal biometrics, including face
recognition, speaker recognition, vein recognition, and spoofing
and anti-spoofing. He is also the Director of the Swiss Center for
Biometrics Research and Testing. He serves as an Associate
Editor of the IEEE T-IFS. He is Coeditor of the upcoming
Handbook on Biometric Anti-Spoofing (Springer), and an Area
Editor of the Encyclopedia of Biometrics? Second Edition.
- Calender

I EEE Transactions on Biometrics, Behavior, and I dentity Science
Proposal on IEEE Transact ions on Biomet rics, Behavior, and Ident it y Science
Biometrics and Identity Science is a core opportunity area with a large social impact which is becoming a
major component in several large-scale identity management projects. In US, Europe, Asia, Africa, and
Australasia, there are several active program funding biometrics-related activities. For example, OBIM,
India?a UIDAI (Aadhaar) project, FBI?s NGI, and other nation-wide identity management programs have
shown the usefulness of biometrics for identity management.
IEEE Biometrics Council has submitted a proposal to IEEE for starting a new Transactions, tentatively named
as "IEEE Transactions on Biometrics, Behavior, and Identity Science". The proposed IEEE Transactions on
Biometrics, Behavior, and Identity Science would be a major source of dissemination of scientific
innovations and progress in this domain. A broad scope would also ensure a widespread reach to academic,
technical, engineering, and legal/law enforcement communities investing in biometrics. The proposed topics
include,
- Management of human identities using traditional behavioral and physiological characteristics
(biometric modalities) including Face, Fingerprint, Iris, Periocular, Palmprints, Gait, Ear, and Voice.
- Emerging Biometrics including Social and Behavioral Biometrics, Gesture based Biometrics,
Keystroke, Vascular, Knuckle, and Soft Information/Biometrics such as Quality, Age, Gender, Kinship,
Social Context, and Biographic Analysis
- Pattern Classification and Machine Learning, including Deep Learning, Algorithms for Biometrics,
Behavior and Identity Science
- Multibiometrics including Fusion Approaches
- Person Re-identification
- Novel Biometric Sensors
- Biometrics Anti-spoofing, Template Protection, Template Update, and Data Protection
- Adversarial Attacks on Biometrics Systems
- Detection, Tracking, Identification, and De-identification in Surveillance
- Novel (challenging) Biometrics Databases, Protocol and Benchmarking, Large Scale Evaluation,
Confidence Interval Estimation, and Performance Modeling and Prediction
- Mobile Biometrics, Active Authentication, Mobile Device Security, and Device Identification
- Forensic Analysis in Identity and Activity
- Biometrics and Human Behavior Analysis, Human Activity Understanding, and Behavior Modeling
- Modeling the Interplay of Behavioral Modalities with Applications in Biometrics and Security
- Applications of Biometrics and Identity in Privacy Preserving Computing, User Centric Security,
Usable Privacy and Security, Social and Criminal Network Inference, Privacy Protection and
Enhancement
- Applications of Biometrics and Identity in Modeling Deception, Deceptive Intent, and Predictive
Analytics
- Applications of Biometrics in Law Enforcement and Cybercrime, Finance, Smart-cards, Border
Control, Healthcare, Civil Registry, Access Control, Securing Documents, Fraud Detection, Cloud, and
Entertainment.
The IEEE Biometrics Council request the research community to fill the online form to show their interest in
the proposal Transactions: ht t ps:/ / t inyurl.com / km yqhl4
Deadline to submit the online form: March 31, 2017
If you have any feedback on this proposal, please send an email to Dr. Mayank Vatsa, Vice President
(Publications) at mayank@iiitd.ac.in

Biometrics News
Propriet ary Fingerprint Templat e Evaluat ion (PFTII)
The NIST Proprietary Fingerprint Template Test Phase II is part of an ongoing program to measure the
performance of fingerprint matching software by utilizing vendor proprietary fingerprint templates. The
99.98% true match rate, combined with the related equal error rates achieved by Innovatrics?s algorithm,
are the best results ever recorded during PFT evaluations.
?This unique combination of industry-leading accuracy and top-ranked speed underlines our belief that
Innovatrics?s algorithm is the most suitable solution for practical implementation and use with large-scale
databases in Automatic Fingerprint Identification System (AFIS) solutions,? said Jan Lunter, CEO of
Innovatrics. The test assesses the accuracy of end-stage matchers ? the computationally expensive
algorithms used in the last stages of one-to-many AFIS searches.
More: https://www.nist.gov/itl/iad/image-group/proprietary-fingerprint-template-evaluation-pftii
St udy looks deeper int o face recognit ion
Researchers at Carnegie Mellon University have released a new study that seeks to explain how our brains
process facial images so quickly and accurately. The new study, published in the Dec. 26, 2016 issue of the
Proceedings of the National Academy of Sciences (PNAS), examined how humans use their brains by using
magnetoencephalography (MEG) scans.
The team used highly sophisticated brain imaging tools and computational methods to measure the
real-time brain processes that convert the appearance of a face into the recognition of an individual. "Our
results provide a step toward understanding the stages of information processing that begin when an image
of a face first enters a person's eye and unfold over the next few hundred milliseconds, until the person is
able to recognize the identity of the face," said Mark D. Vida, a postdoctoral research fellow in the Dietrich
College of Humanities and Social Sciences' Department of Psychology and Center for the Neural Basis of
Cognition (CNBC).
More: http://www.planetbiometrics.com/article-details/i/5367/
Liveness Det ect ion Compet it ion Arrives at Crit ical Time for Iris Scanning
With the technology thus poised to reach a huge number of consumers, the issue of liveness detection in
iris scanning is perhaps more important than ever. A coalition of universities from around the world has
once again convened to test liveness detection in iris scanning technologies, and is inviting academic and
corporate organizations to take part in the competition. Liveness Detection Competition Arrives at Critical
Time for Iris ScanningCalled LivDet-Iris: Liveness Detection-Iris Competition 2017, it has been organized by
Clarkson University, Notre Dame University, West Virginia University, and Warsaw University of Technology.
It?s a sort of sequel event to LivDet-Iris 2015, and arrives at a time when iris scanning is emerging as a major
modality. LivDet-Iris 2017 is taking place as part of the International Joint Conference on Biometrics (ICJB
2017), which will run from October 1st to 4th in Denver, Colorado.
Read more: http://iris2017.livdet.org
Mult ifact or aut hent icat ion sees 40% growt h in 2016
The number of firms using multifactor authentication either for employees or customers leaped this year,
according to a new research report. A study by SecureAuth Corp has found that multi-factor authentication
jumped by more than 40% year-over-year in 2016. Of the organisations surveyed, more than half (51%)
were using MFA across the organisation, while 38% have implemented it in some areas. Meanwhile, next
year more than 30% of organizations are looking to expand or implement MFA in the next 12 months.
?Using a second-factor can be a deterrent but is no longer enough against attacks, and organizations must
evolve their methods to safeguard critical points of access such as Single Sign-On (SSO) portals and the
VPN,? said Keith Graham, CTO of SecureAuth. ?By implementing adaptive access authentication,
organizations can both eliminate that threat vector and provide an outstanding user experience. The latest
advances in adaptive authentication include transparent techniques, such as device recognition,
geo-location, the use of threat services, and even behavioral biometrics.?
Read more: http://www.planetbiometrics.com/article-details/i/5354/

Biometrics I nterview
Quest ion: US-VISIT was a
program started after 9/11. How
did US-VISIT evolve to OBIM
today? What is the biggest impact
that OBIM has had on U.S.
national security so far?
Kennet h: The US-VISIT Program
was created in 2003 to develop
an entry/exit system for the
United States, and to serve as a
biometric service provider to the
immigration components of DHS,
also created in 2003, through the
operation of the Automated
Biometric Identification System
(IDENT), the central DHS
repository for matching, storing,
and sharing biometric and
associated biographic
information. Over time, as the
use of biometric identification
was expanded, US-VISIT?s mission
grew beyond its original
mandate.
In March 2013, Congress
recognized OBIM as the
Department?s designated
provider of enterprise level
biometric identity services to
customers within DHS, other
stakeholders in the U.S.
Government, and partner
nations. OBIM is different than
US-VISIT in that we now operate
with a clear view of our role as a
service provider supporting
stakeholders on the front lines of
homeland security. There are
some legacy US-VISIT functions
that no longer reside within
OBIM, such as biographic
entry/exit and overstay analysis.
These functions were transferred
to U.S. Customs and Border
Protection (CBP) and U.S.
Immigration and Customs
Enforcement (ICE), respectively.
The biggest impact that OBIM
has had on U.S. national security
stems from improvements in
interoperability and information
sharing that directly support four
key Homeland Security Missions:
Prevent Terrorism and Enhance
Securit y; Secure and Manage
Our Borders; Enforce and
Administ er Our Immigrat ion
Laws; and St rengt hen Nat ional
Preparedness and Resilience.
IDENT now stores approximately
204.7 million unique identities. It
processes on average more than
280,000 biometric transactions a
day, more than 130,000 of which
come from CBP as daily U.S. entry
transactions. Automated
interoperability between IDENT
and the Federal Bureau of
Investigation?s (FBI) Next
Generation Identification (NGI)
biometric repository amplifies the
effectiveness of the two systems.
OBIM also conducts searches and
transactions for the U.S.
Department of State, and
continues to work toward
automated interoperability with
the U.S. Department of Defense?s
Automated Biometric
Identification System (ABIS). In
September 2014, OBIM and the
FBI?s Terrorist Screening Center
(TSC) established a direct system
link between IDENT and the TSC?s
Terrorist Screening Database,
enabling faster, more efficient
sharing of biometric information
for the identification of known or
suspected terrorists (KST). Finally,
OBIM shares information with
Australia, Canada, New Zealand,
and the United Kingdom through
the Five Country Conference.
These collaborations vastly
increase the value of OBIM?s
biometric identity services.
Quest ion: What effort does
OBIM make to ensure privacy of
those individuals who get
enrolled in IDENT?
Kennet h: Privacy plays an
integral part of OBIM?s biometric
services from conception through
the planning, development, and
execution of its systems and
processes. The organization has a
dedicated privacy team to ensure
that privacy protections are built
into its programs and systems to
make the safeguarding of
personal information a priority.
Personal information is used only
for the purposes for which it is
collected, unless specifically
authorized or mandated by law.
OBIM monitors its systems and
has security practices in place to
protect the privacy of those
whose data are collected, and to
ensure the integrity of that data.
Kennet h D. Gant t is the
Deputy Director of the Office
of Biometric Identity
Management (OBIM) within
the National Protection and
Programs Directorate (NPPD
of the Department of
Homeland Security (DHS).
OBIM is the lead entity for
biometric identity
management services. Mr.
Gantt?s responsibilities
include working with OBIM?s
customers, stakeholders, and
DHS mission partners to
expand the program?s
services to ensure a safe,
secure, and resilient
homeland.

Biometrics I nterview
All sharing of personal
information is authorized by law
and governed, as appropriate, by
information sharing agreements
that ensure compliance with
privacy requirements. OBIM also
conducts privacy impact
assessments, which provide a
transparent view of the personal
information maintained and how
that information is used, shared,
and stored.
All OBIM personnel and users of
the program?s data are
responsible and accountable for
treating personally identifiable
information in accordance with
the Fair Information Practice
Principles (FIPPS).
Quest ion: Could you list three
desired advancements in the
biometrics technologies that
would allow for OBIM?s biometric
identification services to operate
more efficiently and effectively?
Kennet h: OBIM works each day
with its customers and other
stakeholders to pursue faster,
more efficient, cost effective, and
accurate biometric identity
services. Three technological
advancements in this realm
would include the ability to utilize
multiple hardware and software
solutions; offer advanced
reporting; and make use of
different storage and retention
needs, as we don?t need all of our
records in an active mode. More
generally, a fourth advancement
would involve greater use of
multimodal biometrics, including
iris and face. While IDENT
continues to perform well, it is
more than two decades old, and
OBIM is planning for its
replacement with a new and
more robust system, the
Homeland Advanced Recognition
Technology (HART). HART
promises a more flexible and
scalable architecture that will
more efficiently support core DHS
missions and operations with
multimodal capabilities, more
efficient system processing,
increased capacity and storage,
and web portal access and web
services.
A good example of a recent
technological advancement that
had a significant impact on
national security was as follows:
"At U.S. air port s of ent ry, t he
t ime needed for CBP t o send a
request on an individual t o
IDENT and NGI and receive a
response used t o t ake bet ween
24 and 72 hours, meaning t hat
t he individual had left t he
airport by t he t ime t he
response arrived. A program
called Rapid Response
compressed t his t urnaround
t ime t o 15 seconds, allowing
CBP t o det ermine t he
admissibilit y of individuals in
real t ime. Over t he last 12
mont hs, IDENT has processed
1.42 million Rapid Response
searches. "
Quest ion: One of the major
vulnerabilities of a biometric
system is presentation of a spoof
or an altered biometrics sample.
What is OBIM?s position on this
issue?
Kennet h: OBIM watches for
those individuals who try to
impede law enforcement or
obtain unwarranted privileges
and benefits by spoofing
biometric systems. While IDENT
returns an automated answer for
the vast majority of biometric
queries, about 2,000 sets of
fingerprints a day must be
resolved by fingerprint examiners
in OBIM?s Biometric Support
Center (BSC). These highly trained
professionals provide OBIM?s
best defense against spoofing.
Analyzing prints on
high-definition work stations, BSC
experts can identify the markers
of altered biometrics, as well as
areas that have not been altered.
The BSC furnishes assured
biometric identification services
24 hours a day, including 10-print
comparisons and verification,
unknown deceased
identifications, latent print
comparisons, and enrollment of
biometric records from law
enforcement and intelligence
agencies.
CALL FOR ARTICLES
The editor ial boar d of IEEE
Biometr ics Council New sletter
invites shor t ar ticles r elated to
- Recent new s r elated to
biometr ics
- Job-posting including
PhD/PostDoc positions,
- Biometr ics Database
- Open Sour ce Code
Please send the shor t ar ticles to:
hugomcp@di.ubi.pt

Spotlight
In this spotlight we would like to point out the relationship between two cool research topics, namely biometric
recognition and deep learning, and between them and industry latest trends.
It is quite usual that successful research products reach the industry application after that consolidated results
testified their effectiveness. Biometrics and deep learning are not an exception. In fact, leaving out fingerprint
recognition that has a long story of application in both forensics and security, the other traits have a much more
recent presence in real-world applications. The first mention of face recognition out of the fields of experimental
psychology and cognitive neurosciences dates back to ?60, but entails a semi-automatic (forensic) process supported
by human experts. In 1988, Kirby and Sirovich applied principle component analysis to the face recognition problem
and following this work, in 1991, Turk and Pentland proposed Eigenfaces. However, the approach was successful with
a small dataset and in extremely controlled conditions. In order to have the first commercial off-the-shelf (COTS) face
recognition systems, though still limited to well controlled scenarios, we have to wait until 1997. Joseph Atick, a
pioneer of facial recognition, helped developing the first facial recognition system in the world and founded the
American company Visionics, one of the first companies in the field, in 1994. In August 1997, Visionics announced the
first large scale shipping of its FaceIt(R) face recognition software engine to application developers and system
integrators. The engine was encapsulated into a powerful Software Developer Kit to enable developers to quickly
incorporate facial recognition capabilities into their applications, even beyond the forensics context. Afterwards, in
November 1998, Visionics Corporation announced that its FaceIt(R) face recognition software was the first software of
its kind to be used in a Closed Circuit Television (CCTV) control room application. The new CCTV-based anti-crime
system called 'Mandrake' was installed in the borough of Newham, East London, in the United Kingdom. In March
2000 two early direct competitors, Viisage Technology, Inc. and Biometrica Systems, Inc. announced the fiftieth order
and installation of its full surveillance suite of Visual Casino loss-reduction systems, at Mirage Resort, Las Vegas.
Visual Casino, a computer application with its facial recognition technology, was used by surveillance operators to
identify cheaters and other casino undesirables, as well as casino VIPs. After that new competitors entered the
market, being Cognitec one of e most renowned nowadays.
Another biometric trait that has reached the market is iris, with a similar history from the first significant work
presented by J. Daugman in 1993, except for the fact that it often requires special equipment. . In 1995, the first
commercial product was distributed on the market by Iriscan, Inc., that become Iridian Technologies, Inc. in 2000
(that in 2006 was acquired by Viisage Technologies, Inc.). In 2005, the broad patent covering iris recognition expired,
so that other companies could develop their own algorithms for iris recognition. The patent on the IrisCodes® has
expired too in 2011. Going to deep learning, it gave a second life to neural networks, and is gaining increasing
attention in industry too. Notwithstanding its definition as a ?a new area of Machine Learning research?
(http://deeplearning.net) many researchers consider the Neocognitron introduced by Kunihiko Fukushima in 1980 as
one of the first ancestors of deep learning architectures, in particular of those deriving from artificial neural networks
(ANN). However, it is only after 2007 that the work by Yoshua Bengio renewed the research community interest on
related topics. ?The large-scale impact of Deep Learning in Big Tech Companies began around 2010 with speech
recognition. ? It took around 30 years to become mainstream. Computers were not powerful enough and companies
didn?t have such a large amount of data.? (http://tinyurl.com/lnv77uo). At present, Skymind offers to enterprise deep
learning for big data, e.g., company fraud detection in finance, recommender systems in e-commerce, and anomaly
detection for hardware makers. But, as noticed in December 2014 by Forbes (http://tinyurl.com/kwswv7p) big
American companies like Google, IBM, Microsoft, Facebook and China?s Baidu are well-represented in applications
providing the core technologies for applications addressing a radical change in enterprise and industry.
Last but not least, the use of deep learning as a strategy for biometric extends from research to industry and to social
networks. DeepIris by CASIA (Chinese Academy of Science) is one of the latest proposals for iris recognition. But in the
?real world? Google already uses deep learning for voice recognition on Android phones, and Facebook uses DeepFace
to identify friends in users?photographs. To this aim, the latter exploits ?a face representation from a nine-layer deep
neural network. This deep network involves more than 120 million parameters using several locally connected layers
without weight sharing, rather than the standard convolutional layers. Thus we trained it on the largest facial dataset
to-date, an identity labeled dataset of four million facial images belonging to more than 4,000 identities.?
(http://tinyurl.com/kco8e7q)
What is the key for this successful marriage? According to many observers, the availability of big amounts of data is a
possible explanation of recent achievements. From one side, it might represent a possible stressing factor for
machine learning methods. However, the increased computing power and storage easily compensate this. On the
other side, the main the positive aspect of this wealth of data is the possibility to carry out a huge training and a
fine-tuning of a large number parameters relevant to classification, that was impossible before.

Paper Summary
Overview:
Cardiovascular Biom et rics: Com bining Mechanical and Elect rical Signals (* )
Ikenna Odinaka a , Joseph A. O'Sullivan b , Erik J. Sirevaag c , John W. Rohrbaugh c
ikenna.odinaka@duke.edu
a - Duke University, b- Washington University in Saint Louis
c - Washington University School of Medicine
Two biometric traits that have been investigated for their potential use in human recognition are the
electrocardiogram (ECG) and the laser Doppler vibrometry (LDV) signal. Although the ECG and LDV signals are both
measures of cardiovascular activity, the differences between them are substantial and important, and reflect
fundamentally different aspects of cardiovascular performance. Whereas the ECG is an electrical signal, which is
generated in the myocardium and is conducted by volume to the surface electrodes, the LDV signal (recorded here
from the neck, overlying the carotid artery) derives from mechanical displacements of the skin overlying the carotid
artery. Previous studies have shown that biometric systems based on the ECG or LDV signal can achieve single digit
authentication performance, in terms of equal error rate (EER). In particular, EERs between 5 and 7% have been
reported, when training and testing data are obtained from different sessions [1], [2]. In this work, we propose
combining ECG and LDV signals for human recognition. In addition, we employ standard fusion techniques for
aggregating the information provided by each biometric system, to achieve a substantial increase in recognition
performance. We show that a reduction by at least half in the EER can be obtained from this fusion. Moreover, we
report on the methods we have found to work best for authentication and identification. The principal goal of this
effort is thus to assess authentication and identification performance using combined ECG and LDV signals. We
pursued this objective using a number of well-established and accepted fusion methods.
A multibiometric system is one that uses more than one biometric trait (multimodal), biometric classifier
(multi-algorithm), instance of the same biometric trait (multi-instance), sensor for extracting the same biometric trait
(multi-sensor), or sample of the same biometric trait (multi-sample) for biometric recognition [3]. A multi-instance
system applies to biometric traits that do not occur singly; e.g., ten fingers, two irises, and multiple ECG channels. A
multi-sample system combines multiple samples or different portions of the same biometric trait; e.g. the front and
side of a face, segments of the same finger, multiple ECG heartbeats. For physiologically related signals, which change
over time, we can combine signals from more than one session, the so called multi-session system [1]. In this work,
we present the performance gains that result from a multimodal, multi-session, multi-sample system involving the
ECG and LDV signals. The individual systems that constitute a multibiometric system can be combined at multiple
levels, ranging from the raw signal to decision processes [4]. In this work, we compare the authentication and
identification performance of the cardiovascular multibiometric system, when fusion is performed at the feature,
score, rank, and decision levels of the biometric pipeline. In addition, we analyze the performance gains of the
cardiovascular multibiometric system over the individual ECG and LDV biometric systems. The subtle differences
between ECG and LDV signals may not be readily appreciated, and it may not be directly obvious that combining the
ECG and LDV signals will lead to improved performance. For example, based on Figure 1, the use of temporal fiducial
features may decrease the performance of the joint biometric system, since the timing information of the two signals
are highly correlated. However, the waveform features of the two signals are very different, suggesting that
amplitude-based, angle-based, area-based fiducial features, and non-fiducial features are better able to capture
uncorrelated information from the ECG and LDV signals. We adopted the short-time Fourier transform approach to
feature extraction. This is by no means the only way the features can be gotten from each modality for later fusion.
See the paper by Odinaka et al. [2] for a review of the feature extraction methods that have been proposed for ECG
biometrics.
Informat ion Fusion:
1) Feature-level Fusion: We concatenate the features vectors from the different modalities to obtain a fused vector. A
classifier is then employed on the fused vectors.
2) Score-level Fusion: We explored the three major approaches to score-level fusion: score normalization, density
estimation, and classifier-based fusion [3], [4]. Examples of score normalization techniques include: Min-Max, Z-Score
(Mean-Std), Median-MAD (median absolute deviation).
(* ) Appeared in IEEE Transactions on Information Forensics and Security, 10(1):16-27. January, 2015.

Paper Summary
3) Rank-level Fusion: After computing the match scores of each test signal, each biometric matcher (ECG or LDV)
outputs a rank for each of the classes (enrolled individuals). We used techniques including the highest rank, Borda
count, and the weighted Borda count (logistic regression) [5], to obtain the fused rank.
4) Decision-level Fusion: The decision made by a biometric matcher can be an accept or reject, in authentication
mode, or an identity or reject, in identification mode. Example techniques we utilized to fuse the decisions from both
biometric systems, include, ?AND? and ?OR? rules [6].
Fig. 1: A 5s segment of a subject?s simultaneously recorded ECG and LDV velocity waveforms.
Result s:
Using the score-level fusion techniques gave the best multibiometric performance improvement for both
authentication and identification modes of operation. The authentication performance, as measured by EERs,
improved from ~6% for each of the unimodal ECG and LDV biometric systems, to ~2% for the multibiometric system.
Moreover, in the identification mode, the rank-1 accuracy improves from ~80% for each unimodal biometric system,
to ~92% for the multibiometric system.
Conclusions:
We performed a large scale cardiovascular biometric experiment on 258 individuals from three sessions. This
represents a significant extension of the results of uni-variate modeling [1], [2] to a multidimensional case. We
performed a comprehensive study of the performance of the multibiometric system at different levels of the
biometric pipeline, using standard fusion techniques. In one measure of performance, namely EERs, we showed that
performance improves from about 5.3% to about 2.1%. Based on our analysis, we made recommendations on the
methods that we found to work best in the authentication and identification modes of operation.
References:
[1] M. Chen, J. A. O?Sullivan, N. Singla, E. J. Sirevaag, S. D. Kristjansson, P.-H. Lai, A. D. Kaplan, and J. W. Rohrbaugh,
?Laser Doppler vibrometry measures of physiological function: evaluation of biometric capabilities,? TIFS, vol. 5, no. 3,
pp. 449?460, 2010.
[2] I. Odinaka, P.-H. Lai, A. D. Kaplan, J. A. O?Sullivan, E. J. Sirevaag, and J. W. Rohrbaugh, ?ECG biometric recognition: A
comparative analysis,? TIFS, vol. 7, no. 6, pp. 1812?1824, Dec. 2012. [3] A. K. Jain and A. Ross, ?Multibiometric systems,?
Commun. ACM, vol. 47, pp. 34?40, Jan. 2004.
[4] A. Ross, K. Nandakumar, and A. K. Jain, Handbook of Multibiometrics. Springer-Verlag New York, Inc., 2006.
[5] T. K. Ho, J. Hull, and S. Srihari, ?Decision combination in multiple classifier systems,? PAMI, vol. 16, no. 1, pp. 66?75,
Jan 1994.
[6] J. Daugman, ?Combining Multiple Biometrics.? http://www.cl.cam.ac.uk/_jgd1000/combine/combine.html, 2000.

Conference Report - BI OSI G 2016
The 15th edition of the International Conference of
the Biometrics Special Interest Group (BIOSIG) took
place at Fraunhofer IGD in Darmstadt, Germany
from September 21 to 23 and attracted registered
more than 100 participants. This year participants
even travelled all the way from Japan, Korea,
Argentina, Uruquay and the US to Darmstadt in
order to join the BIOSIG community.
The program was composed of scientific research
contributions on the one hand and reports about
large-scale applications on the other hand. The
opening keynote talk was given by Richard Rinkens
(EC, DG Home), who presented the new proposal
for the Entry Exit System (EES), adopted by the
European Commission after an intense period of
analysis and real-world testing (eu-LISA has tested
13 entries - sea, rail, air - with different options).
The new proposal is based on multi-modal
biometrics combining 4-fingerprints and a good
quality facial-image taken live. In particular, new
procedures for EU citizens, 3rd country nationals
with visa and 3rd country without visa are
considered. One of the main objectives is the
identification of overstayers and to enable
subsequent evidence based visa policy making. To
tackle these issues, the EES will substitute manual
procedures with electronic stamps in an electronic
database. In addition, a new link has been made
between EES and VIS, and a common AFIS should
be deployed for the three systems. Future features
include a webservice to check credentials and
authorized time left, which will be the first
consultable service for an EU system of this kind
through internet. Finally, it was highlighted that the
EES will seriously reinforce internal security and
fight against terrorism.
Accepted conference contributions included 16
presentations and covering soft biometrics for face
or iris, fingerprint, vein, face, iris, voice, gait and
keystroke recognition. Other challenges addressed
included facial forensics use cases, voice activity
segmentation, 3D gloves for contactless verification,
new biometric template protection schemes or the
generation of synthetic fingerprint alteration
database to allow the further development of
algorithms to detect them. Another relevant topic
covered in the conference was Presentation Attack
Detection (PAD) for characteristics such as face and
iris. The poster session with 25 contributions was a
good mix of research results from academic and
industrial research labs and visitors did spend a
long time in the poster exhibition before the start
of the social event with the traditional late summer
barbeque ? providing lots of opportunities for
networking.
On the last day of the conference, Arun Ross
addressed the issue of personal privacy in
biometric systems. In particular, he considered
several questions such as (a) Can additional
information about an individual be automatically
gleaned from biometric data? (b) Can biometrics be
used to surreptitiously track an individual? (c) Can
multiple biometric databases be linked to develop a
more complete profile of an individual? (d) Who
owns the biometric data collected from an
individual, and how long should the biometric data
be retained in an identity management system? He
started the keynote with a definition stemming
from 1890: ?privacy is the right to be left alone?.
However, in contemporary society it is easy for
someone to take a picture of you and find out who
you are by harnessing the power of automated face
recognition and cloud computing. In the first part
of the talk, he discussed methods by which
biographical (e.g., gender from fingerprints),
anatomical (e.g., crypts in iris), sensorial (e.g., type
of device used to image a certain fingerprint), and
environmental (e.g., intensity of ambient lighting
during iris image acquisition) information can be
gleaned from biometric data.

Conference Report - BI OSI G 2016
While the resultant information can be used to
improve person recognition performance in the
framework of ?soft? biometrics or forensics, the
extraction of such information can be deemed to
violate privacy. He gave the example where face or
iris data can be surreptitiously used to divulge
pathological information about an individual.
In the second part of the talk, he discussed ways by
which privacy can be accorded to stored biometric
data. The proposed methods rely on the principles
of visual cryptography and signal mixing to
generate biometric templates that can suppress
some of the additional information about a person
that is resident in the biometric data. He reported
experimental results discussing the efficacy of
these methods. The talk concluded by pointing out
that privacy enhancing technologies can be
judiciously used by biometric systems to ensure
that the benefits of biometrics are not undermined
by privacy concerns.
The last conference day concluded with a keynote
by Davide Maltoni addressing the issue of
large-scale verification within projects such as
UIDAI, BMS, ePassports or SIS-II: predicting
accuracy of biometric systems is a very difficult
problem. Statistical modelling techniques often
require to make assumptions on data distributions
that we cannot validate in practice. A different
approach is running fingerprint recognition
algorithms on large datasets of synthetic data.
However, we have to face two problems: i) ensure
that synthetic data well approximate real data; ii)
running huge amount of fingerprint comparisons in
a reasonable time. In particular, the SFinGe
software developed at the University of Bologna
within the EU FIDELITY project can mitigate this lack
of large databases, where not only one but several
samples of a given synthetic fingerprint can be
generated modelling distortion. For the empirical
evaluation and comparison with real data,
verification was run using Minutiae Cilynder Codes
(MCC), which model spatial and directional
contributions. The software has been optimized for
GPUs, needing only 5.6 ms for a single identification
on 250K subjects (44.6 millions of comparisons per
seconds). The empirical evaluation on synthetic data
yielded predictions similar to those found in a real
case study: IUDAI. As future actions an analysis with
NFIQ2.0 will be considered and scaling up to a
database modelling the world population (7 billion),
which will need about 3 months for generation and
a few weeks for identification.
As in previous BIOSIG conferences participants of
the conference themselves voted for the best paper
and the best poster that was presented at the
conference. The winner of the BIOSIG 2016 best
paper award is Sunpreet S. Arora (Michigan State
University) for his presentation ?3D Whole Hand
Target: Evaluating Slap and Contactless Fingerprint
Readers?, which convinced the majority of the
participants.
While the poster session showed many impressive
research results that stimulated long discussions,
there was one contribution, which was chosen by
the participants as best and it received thus the
BIOSIG 2016 best poster award. It was the poster of
Yoshinori Koda (NEC Corporation) with the title:
?Advances in Capturing Child Fingerprints: A High
Resolution CMOS Image Sensor with SLDR Method?.
The BIOSIG conference was preceded by the 3rd
EAB Research Project Conference and was further
co-located with two satellite workshops:
The meeting of the TeleTrusT Biometric Working
Group and joint meeting of the ethical committee of
the European Association of Biometrics.
The 2016 BIOSIG conference was jointly organized
by the Competence Center for Applied Security
Technology (CAST) and the special interest group
BIOSIG of the Gesellschaft für Informatik e.V. (GI).
The conference was technically co-sponsored by
IEEE Biometric Council and the papers will be added
to IEEE Xplore.

Calender
Call for Proposal for IEEE Conference on Biomet rics Theory, Applicat ion
and Syst ems (BTAS), 2018
Deadline: May 31, 2017
BTAS is the flagship conference for IEEE Biometrics Council. Each proposal
should include information on (but not limited to):
- Organizing and technical committees, at least general and program
chairs
- Verified venue and dates (without possible conflict with other major
conferences in closely related fields). Please provide venue contact info.
The venue is not restricted to be in Washington, DC
- Verified accommodation arrangements;
- Publicity and plans for growing the conference.
- Timetable with critical milestones;
- Budget estimates using the standard budget template available at:
http://www.ieee.org/documents/financial_reporting_tool.xls
- Possible external sponsors.
Further guidelines on IEEE conference organization can be found at
http://www.ieee.org/conferences_events/conferences/organizers/index.html
Please send proposals to Sebastien Marcel, Vice-President for Conferences, IEEE
Biometrics Council at marcel@idiap.ch
The proposals will be vetted by the IEEE Biometrics Council Conference
Committee and result announced through IEEE Biometrics mailings and the
website.
May 30 - Jun 3, 2017
IEEE FG2017
http://w w w.fg2017.or g/
Sep 20 - 22, 2017
BIOSIG 2017
www.biosig.de/biosig2017
Oct 1 - 4, 2017
IEEE/IAPR IJCB2017
http://w w w.ijcb2017.or g/
16t h Int ernat ional Conference of t he
Biomet rics Special Int erest Group
(BIOSIG 2017)
Sep 20 ? 22, 2017
Darmstadt, Germany
www.biosig.de/biosig2017
Important Dates:
- Full Paper Submission deadline: 30 May
- Notification of acceptance date: 15 Jul
- Final submission deadline: 15 Aug
2017 IEEE/IAPR Inter nat i onal Joi nt
Confer ence on Bi om et r i cs (IJCB 17)
The Inter national Joint Confer ence on
Biometr ics (IJCB 2017) combines tw o major
biometr ics r esear ch annual confer ences, the
Biometr ics Theor y, Applications and Systems
(BTAS) confer ence and the Inter national
Confer ence on Biometr ics (ICB). IJCB 17 w ill be
held fr om October 1 till October 4, 2017 in
Denver , CO. IJCB 2017 solicits submissions
r elated to biometr ics. Call of paper s is available
at: http://w w w.ijcb2017.or g/
Paper submission deadline is Apr il 15, 2017.
Paper submission instr uctions:
http://w w w.ijcb2017.or g/ijcb2017/author s.php
Paper submission site:
https://cmt3.r esear ch.micr osoft.com/IJCB2017/
SI BWild 2017: Image and Vision Comput ing (IVC) Special Issue
on Biomet rics in t he Wild
Bir Bhanu, Abdenour Hadid, Qiang Ji, Mark Nixon, Vitomir Struc
Paper Submission Deadline: Jun 30, 2017
Publication Date: April 2018
http://tinyurl.com/kp5dxxg
SI VSB2017: Pat t ern Recognit ion Let t ers (PRL) Special Issue on
Video Surveillance-orient ed Biomet rics
Changxing Ding, Kaiqi Huang, Vishal M. Patel, Brian C. Lovell
Paper Submission Deadline: March 30, 2017
http://tinyurl.com/l7mhhwr
SI UER 2017: IET Biomet rics Special issue on Unconst rained Ear
Recognit ion
Peter Peer, Vitomir Struc
Paper Submission Deadline: Sep 1, 2017
Publication Date: May, 2018
http://tinyurl.com/n8svouv