embedded system for signature verification - Massbunkz.com

wispsyndicateΑσφάλεια

23 Φεβ 2014 (πριν από 3 χρόνια και 5 μήνες)

178 εμφανίσεις

1






APPLICATION BASED EM
BEDDED SYSTEM FOR SIGNATURE
VER
IFICATION



ABSTRACT:


This signature verification involves data acquisition, feature extraction, feature selection,
decision making and performance evaluation. It is a process where the user uses a normal

pen to
create his/her signature that produces measurements such as pen location, speed and pressure
.The verification system, the signature to verify is compared to the prototypes of the genuine
signature at disposal similarities measure. Signature verifi
cation are becoming increasing popular
for application ranging from access control to restricted areas to fraud preventation in financial
transactions online signature verification algorithm that achieves high performance. The
implementation accuracy, fusi
on strategies are promising.

Signature verification is of particular impo
rtance as it is the only widely accepted method for
endorsing financial transactions. Signature verification is a way to determine the validity and
authority of documents. In addition, it could also be applied to security systems to prevent
critical data o
r information from being modified or stolen. For that purpose we want to
implement signature verification embedded system with high performance









2









INTRODUCTION:



Signature verification is a great
importan
ce in
electronic documentation. It’s
applications may be include banking, credit
card authorization, and personal
identification. Visual examination is the
most popular approach for signature
verification, yet ther
e are an occasion for the
examiner’s to make mistakes or lower the
threshold of acceptance. For instance, in a
commercial transaction, comparing the
signature with a previously written
signature, such as the signature on the back
of a credit card, is more
or less a challenge
to the clerks. As a result, large financial
losses may occur. It is clear that signature
verification through visual examination can
only judge the authentication roughly.
Human eyes can hardly analyze the detailed
writing features.

Aut
omatic signature verification can be
classified into two categories: on
-
line
signature verification and off
-
line signature
verification. An off
-
line signature
verification system uses the camera or the
scanner to capture the image of the signature
and the
uses the shape of signature for
verification .The shape related features
include X
-
Y coordinates, height to

width
ratio, direction histogram, and curvature.
The main application of off
-
line signature
verification systems is in automatic
verification of si
gnature found on bank
cheques and documents. AN on
-
line
signature verification system uses a digitizer
for data acquisition purpose that generates
the dynamic features such as velocity,
acceleration and pressure. More over total
signature time, RMS speed,
and
instantaneous velocity at sampling points is
also consider as dynamic features it is
always harder for the forger to imitate both
the dynamic features and the shape of a
signature than just imitating the shape alone.

Since many years of researches in s
ignature
verification field, several approaches and
practices have been suggested by researches.
3


II. AIM AND OBJECTIVE

Verification based on static signature

having
a following drawback

-
Difficult to obtain dynamic information
from a static signature.

-
People can modify the signature.

-
performance is low

Depending on the acquisition process,
automatic signature verification systems
are divided into tw
o categories :online and
offline. In early offline cases, signatures are
captured until the writing process is
finished, thus only static images are
available .Recently more researches focus
on the online signature verification, where
signatures are acqui
red during the writing
process with a special instrument, such as a
digital tablet. In fact dynamic information is
available in the case of online signature
verification ,such as velocity acceleration
and pressure which is more difficult than the
static sh
ape of signature .There for online
signature verification can usually achieve
better performance than

the offline instance
.Signatur
e verification system in which the
sample signature which is already stored in
database. The dynamic signature
simultaneous
ly compare with store d sample
signature of human being. While comparing
data acquisition, feature extraction, feature
selection, decision making and performance
evaluated using best algorithm improve the
performance and find the acceptance or
rejection ra
te.


III.LITERATURE REVIEW

A.Existing System and its drawbacks


The signature verification problem has two
distinctive perspectives:

(1
) Verification

and (
2
) recognition
. In
verification, features of

a test signature are
contrasted with the features of a limited set
of signatures of the class whose identity is
claimed ,whereas,in recognition ,the
presence of an identity of a test signature in
the data base is ascertained.Recognition
involves feature m
atching stage that extends
to entire data base.


Single biometric systems may not be
always applicable because of unacceptable
performanceand inability to operate on a
lerge user populat
ion.Multiple biometric
systems
can over

c
ome these limitations
.Five scenarios of the multiple biometric

system have been proposed in existing
system that is ,multi sensor system,multi
-
model system, multi
-
unit system muklti
impression system,and multi
-
matcher
system.Among of them,the multi matc
her
system which uses multiple representation
and matching algorithm for the same input
biometric signal is the most cost effective
way to improve the performance of
biometric system.







4


IV.PROPOSED PLAN OF WORK


Signature verification embedded system
in
which the original sample signature in
normal condition which is stored in
database. Global feature can be calculated.



Various features of sign can be stores
in

database
.



Permanent memory (data base)
which stores features of sample
signature.



Temporary memory stores features
features of user’s sign which is to be
comparing with simple signature.



The users signature simultaneously
compare with stored sample
signature of human being



While comparing the data
acquisition, feature extraction,
fe
ature selection, decision making
and the performance evaluated.



After the verification of the various
features of user signature and
original sample signature of user in
normal condition which is stored in
data base. The embedded system will
find the acc
eptance rate and rejection
rate of signature.



The proposed embedded system will
used to improve the accuracy of
signature verification as compare to
existing signature verification
system.

SYSTEM ARCHITECTURE:

It will be consist of three essential
components as shown if figure.

INPUTUNIT
: User can sign and use a
application control unit for verification,
decision making and evaluation

DISPLAY UNIT
:
It is used to display
the result






The online signature verification system
including dynamic features extraction task
consists of the following modules. In the
proposed system the following
preprocessing are used.


(1)

Data acqu
isition module
.

(2)

Preproc
essing.

(3)

Dynamic feature extraction.


(4)

Verification
.









5



V.RESEARCH METHODOLOGY
TO BE EMPLOYED

Progress according to the proposed plan of
work in that we devide the signature
verification embedded system into four
working module as per the requirement as
follows.

(A)Selection of
hardware and its
implementation.

(B)Selection of software and its
implementation.

(C)Hardware/Software co design.

(D)Result.

(E)Actual working project

(A)Selection of hardware and its
implementation

For creating a working module of hardware

first we collect the information of the entire
hardware component that we require for
input unit,output unit,control unit we decide
to collect the requirement specifications
according to the unit required for embedded
system.In processer selection is depe
nd on
the optimal cost,system performance,
memory space required and testability.

(B)Selection of software and its
implementation

Symultaneously in second phase according
to our embedded system it is required
software for verification,decision making
and e
valuation of signature as per the
hardware implementation and capitability
of that hardware we would select the
software.The algorithm require for
verification,decision making suitable
algorithm is to be used according to the
software is selected and usin
g suitable
algorithm.

(C)Hardware software co design

After creating the hardware and software
selection and its implementation ,we have
tointerface the both the unit that means it is
a hardware software integration phase of the
devolope cycle must have sp
ecial tools and
methods to manage the complexcity.he

process of the integrating embedded
software is an exercise in debugging and
discovery is the software really understood
the hardware specification.

(D)Result

It is the process of developing the result
after making verification with the original
data .While making feature extraction of
6


signature and making decision the following
process are used.

(E)Actual working project

Finally after collecting the sample sign
ature
in the database we get actual working
project for verification of sample
signature.We find authentication of user
.


APPLICATION:



ATM SYSTEM



BANK



DOOR ACCESS CONTROL









































7




CONCLUSION
:


We describe the signature verification
involves data acquisition feature extraction,
feature selection, decision making and
performance evaluation.
Signature
verification are becoming increasing popular
for application ranging from access control
to restri
cted areas to fraud preventation in
financial transactions online signature
verification algorithm that achieves high
performance. The implementation accuracy,
fusion strategies are promising. Signature
verification is of particular importance as it
is the

only widely accepted method for
endorsing financial transactions. Signature
verification is a way to determine the
validity and authority of documents.









































8





REFERENCES
:

1.

WWW.MASSBUNKZ.COM


2.

K.toh,w.Yag and X.Jiang”multiple
Biometrics and decision fusion”

3.

J.yi,c.lee and J.kim.Online signature
verification using temporal shift
estimated by phase of gabor
filter.IEEE

4.

D.S.Guru and H.N.Prakash,”
symbolic
Representation of On
-
line
signature”.

5.

M.K.Khan, M.A.U.Khan “online
signature verification”.