Sensor Interoperability and Fusion in Signature Verification A Case Study Using Tablet PC Fernando Alonso-Fernandez Julian Fierrez-Aguilar and Javier Ortega-Garcia

2025-05-03 0 0 9.33MB 8 页 10玖币
侵权投诉
Sensor
Interoperability
and
Fusion
in
Signature
Verification: A
Case
Study
Using
Tablet
PC
Fernando
Alonso-Fernandez,
Julian
Fierrez-Aguilar,
and
Javier
Ortega-Garcia
Biometrics
Research
Lab.
-
ATVS,
EPS,
Universidad
Autonoma
de
Madrid
Campus
de
Cantoblanco
-
C/
Francisco
Tomas
y
Valiente
11, 28049
Madrid,
Spain
{fernando.alonso,julian.fierrez,javier.ortega}©uam.es
Abstract.
Several
works
related
to
information
fusion
for
signature
ver-
ification
have
been
presented.
However,
few
works
have
focused
on
sen-
sor
fusion
and
sensor
interoperability.
In
this
paper,
these
two
topics
are
evaluated
for
signature
verification
using
two
different
commercial
Tablet
PCs.
An
enrolment
strategy
using
signatures
from
the
two
Tablet
PCs
is also
proposed.
Authentication
performance
experiments
are
reported
by
using
a
database
with
over
3000
signatures.
1
Introduction
Personal
authentication
in
our
networked
society
is
becoming
a
crucial
issue [1
].
In
this
environment,
automatic
signature
verification
has
been
intensely
studied
due
to
its
social
and
legal
acceptance
[2, 3
].
Furthermore,
the
increasing
use
of
portable
devices
capable
of
capturing
signature
signals (i.e.
Tablet
PCs,
mobile
telephones,
etc.)
is
resulting
in
a growing
demand
of
signature-based
authenti-
cation
applications.
Several works
related
to
information
fusion for
signature
verification
have
been
presented
[4-8
].
However, few works
have
focused
on
sensor
fusion
and
sensor
interoperability
[9
].
In
this
work, we
evaluate
sensor
interoperability
and
fusion
using
the
ATVS
Tablet
PC
signature
verification
system
[10
].
An
enrol-
ment
strategy
using
signatures
from
the
two
Tablet
PCs
is also
proposed.
The
rest
of
the
paper
is
organized
as follows.
The
sensor
interoperability
and
fusion
topics
are
briefly
addressed
in
Sects. 2
and
3, respectively.
Experiments
and
results
are
described
in
Sect. 4. Conclusions
are
finally
drawn
in
Sect. 5.
2
Sensor
Interoperability
In
biometrics,
sensor
interoperability
can
be
defined as
the
capability
of
a recog-
nition
system
to
adapt
to
the
data
obtained
from different sensors. Sensor in-
teroperability
has
received
limited
attention
in
the
literature
[
11
].
Biometrics
systems
are
usually
designed
and
trained
to
work
with
data
acquired
using
a
unique
sensor. As a
result,
changing
the
sensor
affects
the
performance
of
the
verification
system.
Martin
et
al. [12]
reported
a significant difference
in
per-
formance
when
different
microphones
are
used
during
the
training
and
testing
Sensor Interoperability and Fusion in Signature Verification
181
Input
Signature Pre-
Processing
~~
I DECISION I
.....
THRESHOLD _
~•7•_
Fig.
1.
System model
for
person authentication based on written signature
phases
of
a
speaker
recognition
system.
Ross
et
al. [11]
studied
the
effect
of
matching
fingerprints
acquired
with
two
different
fingerprint
sensors,
resulting
in
an
important
drop
of
performance.
The
sensor
interoperability
problem
is also
being
addressed
by
standardiza-
tion
bodies. To
standardize
the
content,
meaning,
and
representation
of
bio-
metric
data
formats,
ISO
and
IEC
have
established
the
Sub-Committee
37
of
the
Joint
Technical
Committee
1
JTC
1/SC
37 [13
].
Different
Working
Groups
constitute
this
Sub-Committee,
including
those
related
to
biometric
technical
interfaces
and
data
exchange
formats.
3
Fusion
of
Sensors
Multibiometric
systems
refer
to
biometric
systems
based
on
the
combination
of
a
number
of
instances,
sensors,
representations,
units
and/
or
traits
[14
].
Sev-
eral
approaches
for
combining
the
information
provided
by
these
sources
have
been
proposed
in
the
literature
[15, 16
].
However, fusion
of
sensors
has
not
been
extensively
analyzed.
Chang
et
al. [17]
studied
the
effect
of
combining
2D
and
3D
images
acquired
with
two different
cameras
for face recognition. Marcialis
et
al. [18]
reported
experiments
fusing
the
information
provided
by
two different
fingerprint
sensors. No
previous
work
on
sensor
fusion for
signature
verification
has
been
found
in
the
literature.
4
Experiments
4.1
On-Line
Signature
Verification
System
The
ATVS
Tablet
PC
signature
verification
system
is
used
in
this
work
[10
].
This
system
represents
signatures
as a
set
of
14
discrete-time
functions, includ-
ing
coordinate
trajectories,
pressure
and
various
dynamic
properties.
Given
an
enrolment
set
of
signatures
of
a client, a
left-to-right
Hidden
Markov
Model
(HMM) is
estimated
and
used
for
characterizing
the
client identity.
This
HMM
is
used
to
compute
the
similarity
matching
score
between
a given
test
signature
and
a
claimed
identity.
For
more
details, we refer
the
reader
to
[19, 20
].
In
Fig. 1,
the
overall
system
model
is
depicted.
4.2
Sensors
In
this
work,
the
following
Tablet
PCs
have
been
used: i)
Hewlett-Packard
TCll00
with
Intel
Pentium
Mobile 1.1
Ghz
processor
and
512
Mb
RAM,
see
摘要：

SensorInteroperabilityandFusioninSignatureVerification:ACaseStudyUsingTabletPCFernandoAlonso-Fernandez,JulianFierrez-Aguilar,andJavierOrtega-GarciaBiometricsResearchLab.-ATVS,EPS,UniversidadAutonomadeMadridCampusdeCantoblanco-C/FranciscoTomasyValiente11,28049Madrid,Spain{fernando.alonso,julian.fierr...

展开>> 收起<<
Sensor Interoperability and Fusion in Signature Verification A Case Study Using Tablet PC Fernando Alonso-Fernandez Julian Fierrez-Aguilar and Javier Ortega-Garcia.pdf

共8页,预览2页

还剩页未读， 继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

相关推荐

更多
立即下载
分类：图书资源 价格：10玖币 属性：8 页 大小：9.33MB 格式：PDF 时间：2025-05-03

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务

作者详情

相关内容

更多

热门标签

人际关系 动力学连接体 力的合成 配电装置 高考理综 全宋诗作者索引 公务员考试
/ 8
评分 收藏
分享
立即下载
客服
关注