I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
3
,
No
.
4
,
Dec
em
b
er
201
4
,
p
p
.
156
~
165
I
SS
N:
2252
-
8938
156
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
A
I
O
f
fline
S
ig
na
ture
Verific
a
tion
and
Forg
ery Dete
ctio
n Ba
sed o
n
Co
m
pu
ter Vis
io
n
a
nd F
u
zz
y
Lo
g
ic
G
a
uta
m
S.
P
ra
k
a
s
h
,
Sh
a
n
u
S
h
a
r
m
a
Co
m
p
u
ter S
c
ien
c
e
&
En
g
in
e
e
rin
g
De
p
a
rtm
e
n
t,
A
S
E
T
,
Am
it
y
Un
iv
e
rsit
y
,
No
id
a
,
Uttar
P
ra
d
e
sh
,
I
n
d
i
a
.
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
4
,
2
0
1
4
R
ev
i
s
ed
No
v
1
0
,
2
0
1
4
A
cc
ep
ted
No
v
2
1
,
2
0
1
4
A
u
to
m
a
ted
sig
n
a
tu
re
v
e
ri
f
ica
t
io
n
a
n
d
f
o
rg
e
r
y
d
e
te
c
ti
o
n
h
a
s
m
a
n
y
a
p
p
li
c
a
ti
o
n
s
in
t
h
e
f
ield
o
f
Ba
n
k
-
c
h
e
q
u
e
p
ro
c
e
ss
in
g
,
d
o
c
u
m
e
n
t
a
u
th
e
n
ti
c
a
ti
o
n
,
A
T
M
a
c
c
e
ss
e
tc.
Ha
n
d
w
rit
ten
sig
n
a
tu
re
s
h
a
v
e
p
r
o
v
e
d
to
b
e
im
p
o
rtan
t
in
a
u
t
h
e
n
ti
c
a
ti
n
g
a
p
e
rso
n
'
s
id
e
n
t
it
y
,
w
h
o
is
sig
n
in
g
th
e
d
o
c
u
m
e
n
t.
In
th
is
p
a
p
e
r
a
F
u
z
z
y
L
o
g
ic
a
n
d
A
rti
f
icia
l
Ne
u
ra
l
Ne
t
w
o
rk
Ba
s
e
d
Of
f
-
li
n
e
S
ig
n
a
tu
re
V
e
rif
ica
ti
o
n
a
n
d
F
o
rg
e
ry
De
tec
ti
o
n
S
y
ste
m
is
p
re
se
n
ted
.
A
s
th
e
re
a
re
u
n
iq
u
e
a
n
d
im
p
o
rtan
t
v
a
riatio
n
s
in
th
e
f
e
a
tu
re
e
le
m
e
n
ts
o
f
e
a
c
h
sig
n
a
tu
re
,
so
in
o
rd
e
r
t
o
m
a
tch
a
p
a
rti
c
u
lar
sig
n
a
tu
re
w
it
h
th
e
d
a
tab
a
se
,
th
e
stru
c
tu
ra
l
p
a
ra
m
e
ters
o
f
th
e
sig
n
a
tu
re
s
a
lo
n
g
w
it
h
th
e
lo
c
a
l
v
a
riatio
n
s
in
t
h
e
sig
n
a
tu
re
c
h
a
ra
c
teristics
a
re
u
se
d
.
T
h
e
s
e
c
h
a
ra
c
teristics
h
a
v
e
b
e
e
n
u
se
d
to
train
t
h
e
a
rti
f
icia
l
n
e
u
ra
l
n
e
tw
o
rk
.
T
h
e
s
y
ste
m
u
se
s
th
e
f
e
a
tu
re
s
e
x
trac
ted
f
ro
m
th
e
sig
n
a
tu
re
s
su
c
h
a
s
c
e
n
tro
i
d
,
h
e
ig
h
t
–
w
id
th
ra
ti
o
,
t
o
tal
a
re
a
,
I
st
a
n
d
II
nd
o
rd
e
r
d
e
riv
a
ti
v
e
s,
q
u
a
d
ra
n
t
a
re
a
s
e
tc.
Af
ter
th
e
v
e
ri
f
ica
ti
o
n
o
f
t
h
e
sig
n
a
tu
re
t
h
e
a
n
g
le f
e
a
tu
re
s are
u
se
d
in
f
u
z
z
y
l
o
g
ic b
a
se
d
sy
ste
m
f
o
r
f
o
rg
e
r
y
d
e
t
e
c
ti
o
n
.
K
ey
w
o
r
d
:
A
r
ti
f
icial
Ne
u
r
al
Net
w
o
r
k
AN
N
Fu
zz
y
L
o
g
ic
C
o
m
p
u
ter
Vis
io
n
Fo
r
g
er
y
d
etec
tio
n
Sig
n
at
u
r
e
v
er
i
f
icatio
n
Co
p
y
rig
h
t
©
2
0
1
4
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Gau
ta
m
S.
P
r
ak
as
h
,
C
o
m
p
u
ter
Scien
ce
&
E
n
g
i
n
ee
r
in
g
Dep
ar
t
m
e
n
t,
A
SET
,
Am
it
y
U
n
i
v
er
s
it
y
,
No
id
a,
Uttar
P
r
a
d
esh
,
I
n
d
ia
E
m
ai
l
:
G
au
t
am
s
p
r
a
k
ash
@
g
m
a
i
l
.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
Fo
r
g
er
y
is
a
p
r
o
ce
s
s
b
y
w
h
ic
h
,
id
en
tit
y
d
o
cu
m
e
n
ts
o
f
a
p
er
s
o
n
ar
e
co
p
ie
d
o
r
m
o
d
if
ied
b
y
s
u
c
h
a
p
er
s
o
n
w
h
o
is
n
o
t
a
u
t
h
o
r
ized
to
d
o
s
o
,
o
r
a
r
e
in
v
o
lv
ed
i
s
m
o
d
if
icatio
n
f
o
r
t
h
e
p
u
r
p
o
s
e
o
f
d
ec
eiv
i
n
g
th
o
s
e
w
h
o
v
ie
w
t
h
e
d
o
cu
m
e
n
t a
b
o
u
t t
h
e
i
d
en
tit
y
o
f
th
e
s
tat
u
s
b
ea
r
er
[
1
]
.
Sig
n
at
u
r
e,
f
r
o
m
th
e
L
ati
n
w
o
r
d
"
Sig
n
ar
e"
m
ea
n
i
n
g
"
Sig
n
"
is
a
s
ty
lized
h
an
d
w
r
itte
n
r
ep
r
esen
tatio
n
o
f
a
p
er
s
o
n
's
n
a
m
e
o
r
an
id
e
n
ti
f
icatio
n
m
ar
k
t
h
at
a
p
er
s
o
n
w
r
ite
s
o
n
d
o
cu
m
en
ts
/te
x
t
s
.
F
o
r
m
a
n
y
ce
n
t
u
r
ies,
s
ig
n
at
u
r
es
h
a
v
e
b
ee
n
u
s
ed
a
s
an
i
m
p
o
r
tan
t
ele
m
e
n
t
i
n
a
u
t
h
e
n
ticatio
n
o
f
an
y
p
er
s
o
n
's
id
e
n
t
it
y
,
w
h
o
is
s
i
g
n
i
n
g
th
e
d
o
cu
m
e
n
t
[
2
]
.
T
h
e
u
n
iq
u
e
ch
ar
ac
ter
is
tics
o
f
a
p
er
s
o
n
’
s
s
ig
n
at
u
r
e
r
ep
r
esen
t
th
e
p
er
s
o
n
'
s
id
e
n
tit
y
an
d
t
h
e
p
er
s
o
n
's
co
n
s
e
n
t
f
o
r
th
e
ter
m
s
o
f
th
e
d
o
cu
m
en
t
/tex
t.
T
h
e
f
ield
o
f
s
ig
n
at
u
r
e
au
t
h
en
ticatio
n
is
v
er
y
i
m
p
o
r
tan
t
an
d
h
e
n
ce
t
h
e
p
r
o
b
le
m
o
f
v
er
i
f
icatio
n
an
d
f
o
r
g
er
y
d
etec
tio
n
is
o
f
th
e
u
t
m
o
s
t
i
m
p
o
r
tan
ce
.
H
an
d
w
r
i
tten
s
t
y
lized
s
ig
n
at
u
r
es
v
ar
y
lar
g
el
y
f
r
o
m
p
er
s
o
n
to
p
er
s
o
n
.
T
h
e
y
d
if
f
er
in
th
eir
s
ize
s
an
d
s
h
ap
es,
an
d
th
e
v
ar
iatio
n
s
ar
e
s
o
m
u
c
h
,
t
h
at
f
o
r
a
h
u
m
a
n
b
ein
g
,
j
u
s
t
b
y
h
av
in
g
a
g
la
n
ce
at
t
h
e
s
i
g
n
at
u
r
e,
it
is
v
er
y
d
if
f
ic
u
lt
to
s
ep
ar
ate
o
u
t
a
g
en
u
i
n
e
s
i
g
n
atu
r
e
f
r
o
m
a
o
n
e
t
h
at
is
f
o
r
g
ed
.
A
n
au
to
m
atic
s
i
g
n
a
tu
r
e
v
er
if
ica
tio
n
s
y
s
te
m
ca
n
eit
h
er
b
e
o
n
lin
e
o
r
o
f
f
li
n
e.
I
n
an
o
n
lin
e
v
er
if
icatio
n
s
y
s
te
m
,
a
s
th
e
p
er
s
o
n
s
ig
n
s
t
h
e
d
o
cu
m
e
n
t/te
x
t,
t
h
e
p
er
s
o
n
's
s
i
g
n
at
u
r
es
ar
e
r
ec
o
r
d
ed
.
T
h
e
m
er
it
o
f
s
u
c
h
a
s
y
s
te
m
is
t
h
at,
a
p
er
s
o
n
's
d
y
n
a
m
ic
in
f
o
r
m
a
tio
n
c
h
ar
ac
te
r
is
tics
ca
n
al
s
o
b
e
ac
co
u
n
te
d
.
B
u
t
th
e
p
r
o
b
lem
w
it
h
s
u
c
h
a
s
y
s
te
m
is
t
h
at,
in
r
ea
lit
y
,
m
o
s
t
o
f
th
e
d
o
cu
m
en
t
s
ar
e
alr
ea
d
y
p
r
e
-
s
i
g
n
ed
[
3
]
.
Hen
ce
to
d
ea
l
w
it
h
s
u
c
h
s
it
u
atio
n
s
,
an
o
f
f
li
n
e
v
er
if
icatio
n
s
y
s
te
m
i
s
u
s
ed
,
w
h
ic
h
o
n
l
y
ac
co
u
n
t
s
f
o
r
th
e
s
tatic
f
ea
t
u
r
es
o
f
a
s
ig
n
at
u
r
e.
I
m
ag
e
P
r
o
ce
s
s
in
g
h
a
s
f
o
u
n
d
n
u
m
b
er
o
f
ap
p
licatio
n
s
in
t
h
e
f
ield
o
f
f
o
r
en
s
ic
ex
a
m
i
n
ati
o
n
.
I
m
a
g
e
p
r
o
ce
s
s
in
g
h
a
s
p
r
o
v
ed
to
b
e
v
er
y
ef
f
ec
ti
v
e
to
o
l
to
a
n
al
y
ze
t
h
o
u
s
a
n
d
s
o
f
s
ig
n
at
u
r
es
i
n
t
h
e
d
atab
ase,
an
d
ap
p
l
y
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Offlin
e
S
ig
n
a
tu
r
e
V
erif
ica
tio
n
a
n
d
F
o
r
g
ery
Dete
cti
o
n
B
a
s
ed
o
n
C
o
mp
u
ter V
is
io
n
…
(
Ga
u
ta
m
S
.
P
r
a
ka
s
h
)
157
tech
n
iq
u
es
f
o
r
d
etailed
an
al
y
s
is
s
u
ch
as
f
u
zz
y
lo
g
ic
an
d
ar
t
if
icial
n
e
u
r
al
n
et
w
o
r
k
to
d
ec
r
ea
s
e
th
e
a
m
o
u
n
t
o
f
ti
m
e,
an
d
i
n
cr
ea
s
e
t
h
e
ef
f
ec
ti
v
en
es
s
o
f
th
e
s
y
s
te
m
[
4
]
.
Fo
r
b
etter
u
n
d
er
s
ta
n
d
in
g
o
f
f
u
r
th
er
s
t
u
d
ies,
it
is
i
m
p
o
r
tan
t
t
o
b
e
ac
q
u
ain
ted
w
it
h
t
h
e
b
asi
c
co
m
m
o
n
co
n
ce
p
ts
s
u
c
h
as
co
m
p
u
ter
v
is
io
n
tec
h
n
o
lo
g
y
,
an
d
t
h
e
n
e
ed
f
o
r
au
to
m
ated
s
i
g
n
at
u
r
e
v
er
if
icatio
n
.
A
b
r
ie
f
ex
p
lan
atio
n
ab
o
u
t th
e
m
i
s
g
i
v
e
n
b
elo
w
.
1
.
1
.
Co
m
p
ute
r
Vis
io
n
T
ec
hn
o
lo
g
y
C
o
m
p
u
ter
Vis
io
n
T
ec
h
n
o
lo
g
y
is
u
s
ed
f
o
r
au
to
m
a
tin
g
t
h
e
v
is
i
o
n
p
er
ce
p
tio
n
p
r
o
ce
s
s
.
C
o
m
p
u
ter
v
is
io
n
is
a
f
ield
th
at
i
n
cl
u
d
es
m
e
th
o
d
s
f
o
r
ac
q
u
ir
in
g
,
p
r
o
ce
s
s
in
g
,
an
al
y
zi
n
g
,
an
d
u
n
d
er
s
ta
n
d
i
n
g
i
m
a
g
es
a
n
d
,
in
g
en
er
al,
h
i
g
h
-
d
i
m
en
s
io
n
a
l
d
ata
f
r
o
m
t
h
e
r
ea
l
w
o
r
ld
i
n
o
r
d
er
to
p
r
o
d
u
ce
n
u
m
er
ical
o
r
s
y
m
b
o
lic
in
f
o
r
m
a
tio
n
,
e.
g
.
,
in
th
e
f
o
r
m
s
o
f
d
ec
is
io
n
s
[
5
]
.
C
o
m
p
u
ter
v
is
io
n
co
v
er
s
t
h
e
co
r
e
tech
n
o
lo
g
y
o
f
au
to
m
a
ted
i
m
a
g
e
a
n
al
y
s
is
w
h
ic
h
is
u
s
ed
in
m
a
n
y
f
ield
s
[
6
]
.
A
s
a
s
cien
t
if
ic
d
is
cip
lin
e,
co
m
p
u
ter
v
is
io
n
is
co
n
ce
r
n
ed
w
ith
t
h
e
th
eo
r
y
b
eh
in
d
ar
ti
f
icia
l
s
y
s
te
m
s
th
at
ex
tr
ac
t
i
n
f
o
r
m
atio
n
f
r
o
m
i
m
ag
es.
T
h
e
i
m
ag
e
d
ata
ca
n
tak
e
m
an
y
f
o
r
m
s
,
s
u
ch
a
s
v
id
eo
s
eq
u
e
n
ce
s
,
v
ie
w
s
f
r
o
m
m
u
ltip
le
ca
m
er
a
s
,
o
r
m
u
l
ti
-
d
i
m
en
s
io
n
al
d
ata
f
r
o
m
a
m
e
d
ical
s
ca
n
n
er
.
As
a
tech
n
o
lo
g
ical
d
is
cip
li
n
e,
co
m
p
u
ter
v
is
io
n
s
ee
k
s
to
ap
p
ly
its
th
eo
r
ies
an
d
m
o
d
els
to
th
e
co
n
s
tr
u
ctio
n
o
f
co
m
p
u
ter
v
i
s
io
n
s
y
s
te
m
s
[
5
]
.
1
.
2
.
Nee
d o
f
a
uto
m
a
t
ed
Sig
n
a
t
ure
Ver
if
ica
t
io
n
Sig
n
at
u
r
e
v
er
if
ica
tio
n
is
v
er
y
i
m
p
o
r
tan
t
i
n
r
ea
lizin
g
tele
-
b
an
k
i
n
g
an
d
tele
-
n
et
w
o
r
k
i
n
g
s
y
s
te
m
s
,
w
h
er
e
s
ig
n
at
u
r
es
ca
n
b
e
u
s
ed
to
id
en
tify
a
n
d
au
th
e
n
tica
te
a
s
u
b
s
cr
ib
er
.
A
n
a
u
to
m
a
ted
v
er
if
icatio
n
p
r
o
ce
s
s
w
o
u
ld
en
ab
le
b
an
k
s
a
n
d
o
th
er
f
i
n
an
cia
l
in
s
tit
u
tio
n
s
to
s
ig
n
i
f
ican
tl
y
r
ed
u
ce
c
h
ec
k
a
n
d
m
o
n
e
y
o
r
d
er
f
o
r
g
er
ies,
w
h
ic
h
ac
co
u
n
t
f
o
r
a
lar
g
e
m
o
n
etar
y
lo
s
s
ea
c
h
y
ea
r
.
R
elia
b
le
s
ig
n
at
u
r
e
v
er
i
f
icat
io
n
ca
n
b
e
o
f
g
r
ea
t
h
elp
in
m
an
y
o
t
h
e
r
ap
p
licatio
n
ar
ea
s
s
u
c
h
as
la
w
e
n
f
o
r
ce
m
e
n
t,
in
d
u
s
tr
y
,
s
ec
u
r
it
y
co
n
tr
o
l
a
n
d
s
o
o
n
.
Han
d
w
r
it
ten
s
ig
n
at
u
r
es
ap
p
ea
r
o
n
m
an
y
t
y
p
es
o
f
d
o
cu
m
e
n
t
s
s
u
c
h
as
b
an
k
c
h
ec
k
s
an
d
cr
ed
it
s
lip
etc
[
7
]
[
1
2
]
.
T
h
e
lar
g
e
v
o
lu
m
e
o
f
s
u
c
h
d
o
cu
m
e
n
t
s
m
a
k
es
a
u
to
m
atic
s
ig
n
at
u
r
e
v
er
i
f
i
ca
tio
n
d
esira
b
le.
A
s
y
s
te
m
f
o
r
s
ig
n
at
u
r
e
v
er
if
ica
tio
n
r
eq
u
ir
es
h
i
g
h
r
elia
b
ilit
y
.
T
h
e
r
est
o
f
th
i
s
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
w
s
:
-
I
n
s
ec
tio
n
I
I
s
o
m
e
b
as
ic
tech
n
iq
u
e
s
o
f
s
ig
n
at
u
r
e
v
er
if
ica
tio
n
an
d
alr
ea
d
y
d
ev
e
lo
p
ed
s
y
s
te
m
s
ar
e
s
u
m
m
ar
ize
d
.
Sectio
n
I
I
I
d
escr
ib
es
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
,
s
ec
tio
n
I
V
p
r
esen
t
s
th
e
e
x
p
er
im
en
tal
r
esu
lts
a
n
d
s
ec
tio
n
V
c
o
n
clu
d
es t
h
e
p
ap
er
.
2.
RELATED
W
O
RK
T
h
e
p
r
o
b
lem
o
f
s
i
g
n
at
u
r
e
v
er
i
f
icatio
n
an
d
f
o
r
g
er
y
d
etec
tio
n
o
f
d
o
cu
m
e
n
t
s
h
a
s
lo
n
g
b
ee
n
an
ar
ea
o
f
in
ter
est
i
n
th
e
f
ield
o
f
i
m
ag
e
p
r
o
ce
s
s
in
g
.
Ma
n
y
s
tu
d
ie
s
h
a
v
e
b
ee
n
d
o
n
e
till
n
o
w
i
n
o
r
d
er
to
d
ev
elo
p
o
f
f
li
n
e
s
ig
n
at
u
r
e
v
er
i
f
icatio
n
s
y
s
te
m
s
u
s
i
n
g
co
m
p
u
ter
v
i
s
io
n
tec
h
n
o
lo
g
y
an
d
s
o
f
t
co
m
p
u
ti
n
g
tec
h
n
iq
u
es
[
7
]
.
Ma
n
y
r
esear
ch
er
s
ar
e
s
till
w
o
r
k
i
n
g
o
n
d
esig
n
,
d
ev
elo
p
m
e
n
t
an
d
i
m
p
le
m
en
ta
tio
n
o
f
a
n
au
to
m
atic
s
y
s
te
m
f
o
r
f
ast
a
n
d
m
u
c
h
m
o
r
e
e
f
f
ec
tiv
e
as
w
ell
a
s
r
eliab
le
s
i
g
n
at
u
r
e
v
er
i
f
icatio
n
s
y
s
te
m
.
So
m
e
alr
ea
d
y
d
e
v
el
o
p
ed
s
y
s
te
m
i
n
t
h
e
p
r
o
b
lem
ar
ea
ar
e
ex
p
lain
ed
b
elo
w
R
a
m
ee
z
W
aj
id
et
al.
[
8
]
h
av
e
ev
al
u
ated
th
e
p
er
f
o
r
m
an
ce
o
f
v
ar
io
u
s
clas
s
if
ier
s
f
o
r
o
f
f
l
in
e
s
ig
n
at
u
r
e
v
er
if
ica
tio
n
b
a
s
ed
u
p
o
n
t
h
e
lo
ca
l
b
in
ar
y
p
atter
n
s
(
L
B
P
)
f
ea
t
u
r
e
s
et.
T
h
e
y
h
a
v
e
p
er
f
o
r
m
ed
th
e
f
ea
t
u
r
e
v
ec
to
r
b
y
d
i
v
id
in
g
t
h
e
s
i
g
n
at
u
r
e
i
m
a
g
es
i
n
to
t
w
el
v
e
lo
ca
l
r
eg
io
n
s
an
d
f
o
r
m
i
n
g
a
co
d
e
m
atr
ix
b
y
t
h
eir
L
B
P
s
.
T
h
e
au
th
o
r
s
h
a
v
e
i
n
v
esti
g
ated
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
s
e
v
e
n
clas
s
i
f
i
er
s
o
n
T
h
e
FUM
-
P
er
s
ian
Ha
n
d
w
r
itte
n
Si
g
n
at
u
r
e
Data
b
ase
(
FUM
-
P
HSDB
)
co
m
p
r
i
s
i
n
g
o
f
2
0
cla
s
s
es
o
f
g
e
n
u
i
n
e
a
n
d
f
o
r
g
ed
s
i
g
n
at
u
r
es
o
f
d
ep
th
2
0
an
d
1
0
r
esp
ec
tiv
el
y
.
T
h
e
clas
s
i
f
ier
s
c
o
n
s
id
er
ed
b
y
t
h
e
m
ar
e
S
u
p
p
o
r
t
v
ec
to
r
Ma
ch
i
n
es
(
SVM)
,
L
e
ast
Sq
u
ar
es
-
S
u
p
p
o
r
t
Vec
to
r
Ma
ch
i
n
es
(
L
S
-
SVM)
,
Dis
ta
n
ce
L
ik
el
ih
o
o
d
r
atio
T
est
(
D
L
R
T
)
,
A
r
tific
ial
Neu
r
a
l
Net
w
o
r
k
(
A
NN)
,
Fis
h
er
's
L
i
n
ea
r
Dis
cr
i
m
i
n
an
t
(
FL
D)
,
L
o
g
is
tic
s
Dis
cr
i
m
i
n
an
t
an
d
Naiv
e
B
ay
e
s
.
T
h
eir
ex
p
er
im
e
n
tal
f
in
d
i
n
g
s
d
ep
ict
th
at
L
S
-
S
VM
p
er
f
o
r
m
s
th
e
b
est
a
m
o
n
g
t
h
e
s
e
v
en
cla
s
s
i
f
ier
s
,
ac
h
ie
v
in
g
t
h
e
E
q
u
al
E
r
r
o
r
R
ate
(
E
E
R
)
o
f
1
3
%.
Mu
h
a
m
m
ad
I
m
r
an
Ma
li
k
et
al.
[
9
]
h
av
e
ev
al
u
ated
th
e
i
m
p
ac
t
o
f
t
w
o
s
tate
o
f
th
e
ar
t
o
f
f
li
n
e
s
ig
n
a
tu
r
e
v
er
if
ica
tio
n
s
y
s
te
m
s
w
h
ic
h
ar
e
b
ased
o
n
lo
ca
l
an
d
g
lo
b
al
f
e
atu
r
es
r
esp
ec
ti
v
el
y
.
T
h
e
a
u
t
h
o
r
s
h
a
v
e
in
v
es
tig
a
te
th
e
p
er
f
o
r
m
a
n
ce
o
f
a
u
to
m
a
t
ed
s
y
s
te
m
s
o
n
d
is
g
u
is
ed
s
i
g
n
atu
r
e
s
.
T
h
e
s
y
s
te
m
s
w
er
e
ev
alu
a
ted
u
p
o
n
t
h
e
p
u
b
licall
y
av
ai
lab
le
d
ataset
s
f
r
o
m
s
i
g
n
at
u
r
e
v
er
i
f
icatio
n
c
o
m
p
eti
tio
n
.
T
h
e
I
C
D
AR
2
0
0
9
Of
f
li
n
e
S
ig
n
at
u
r
e
Ver
if
icatio
n
C
o
m
p
et
i
tio
n
d
ata
s
et
an
d
t
h
e
I
C
FH
R
2
0
1
0
4
N
Sig
n
C
o
m
p
d
atasets
w
er
e
co
n
s
id
er
ed
.
T
h
e
o
f
f
li
n
e
s
ig
n
at
u
r
e
v
er
i
f
icat
io
n
s
y
s
te
m
s
co
n
s
id
er
ed
f
o
r
ev
al
u
atio
n
w
er
e
L
o
ca
l
Featu
r
es
co
m
b
i
n
ed
w
it
h
Gau
s
s
ia
n
Mix
t
u
r
e
Mo
d
els
(
GM
Ms)
a
n
d
Glo
b
al
Featu
r
es
co
m
b
in
ed
w
it
h
k
-
Nea
r
e
s
t
Nei
g
h
b
o
u
r
(
k
NN)
.
I
n
t
h
eir
ex
p
er
i
m
e
n
ts
it
w
a
s
o
b
s
er
v
ed
th
at
g
lo
b
al
f
ea
t
u
r
es
ar
e
ca
p
ab
le
o
f
p
r
o
v
id
in
g
g
o
o
d
r
esu
lts
i
f
o
n
l
y
a
d
etec
tio
n
o
f
g
en
u
i
n
e
an
d
f
o
r
g
ed
s
ig
n
at
u
r
es
is
n
ee
d
ed
.
L
o
ca
l
f
ea
t
u
r
es
ar
e
m
u
c
h
b
etter
s
u
ited
to
s
o
lv
e
th
e
f
o
r
en
s
ic
s
ig
n
at
u
r
e
v
er
if
ica
tio
n
ca
s
e
s
w
h
e
n
d
is
g
u
i
s
ed
s
ig
n
at
u
r
es a
r
e
also
in
v
o
lv
e
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
3
,
No
.
4
,
Dec
em
b
er
20
1
4
:
1
5
6
–
1
6
5
158
J
u
an
H
u
et
al.
[
1
0
]
h
a
v
e
p
r
esen
ted
a
n
o
f
f
lin
e
s
i
g
n
atu
r
e
v
er
if
icatio
n
s
y
s
te
m
u
s
i
n
g
th
r
e
e
d
if
f
er
en
t
p
s
eu
d
o
-
d
y
n
a
m
ic
f
ea
tu
r
es,
t
wo
d
if
f
er
en
t
cla
s
s
i
f
ier
tr
ai
n
i
n
g
ap
p
r
o
ac
h
es
an
d
t
w
o
d
atase
ts
.
T
h
r
ee
s
ep
ar
ate
p
s
eu
d
o
-
dy
n
a
m
ic
f
ea
tu
r
e
s
b
ase
d
o
n
g
r
a
y
le
v
el:
L
o
ca
l
B
in
ar
y
P
atter
n
(
L
B
P
)
,
Gr
ay
L
ev
el
C
o
-
o
cc
u
r
r
en
ce
Ma
tr
i
x
(
GL
C
M)
an
d
His
to
g
r
a
m
Or
ien
ted
Gr
ad
ien
ts
(
HOG)
h
av
e
b
ee
n
u
s
ed
.
T
h
e
cla
s
s
i
f
icatio
n
i
s
p
er
f
o
r
m
ed
u
s
i
n
g
t
h
e
w
r
iter
d
ep
en
d
en
t
S
u
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
(
S
VM
s
)
cla
s
s
i
f
ie
r
an
d
Glo
b
al
R
ea
l
A
d
ab
o
o
s
t
m
eth
o
d
.
I
n
t
h
eir
ex
p
er
i
m
e
n
ts
,
th
e
r
es
u
lt
s
o
f
th
e
E
q
u
al
E
r
r
o
r
R
ate
(
E
E
R
)
o
f
s
k
illed
f
o
r
g
er
y
te
s
t
u
s
in
g
th
e
w
r
iter
-
d
ep
en
d
e
n
t
ap
p
r
o
ac
h
o
b
tain
ed
w
er
e
1
1
.
7
3
%
f
o
r
L
B
P
,
1
1
.
5
4
%
f
o
r
GL
C
M
an
d
9
.
8
3
%
f
o
r
HOG.
T
h
e
co
m
b
i
n
atio
n
o
f
t
h
e
th
r
ee
r
es
u
lted
in
E
E
R
o
f
7
.
6
6
%.
T
h
e
r
esu
lts
o
f
E
E
R
o
f
s
k
illed
f
o
r
g
er
y
te
s
t
u
s
i
n
g
th
e
w
r
iter
-
i
n
d
ep
en
d
en
t
ap
p
r
o
ac
h
o
b
tain
ed
w
er
e
1
3
.
0
9
%
f
o
r
L
B
P
,
1
9
.
3
3
%
f
o
r
GL
C
M,
1
3
.
1
8
%
f
o
r
HOG
a
n
d
co
m
b
in
atio
n
o
f
al
l
t
h
r
ee
r
esu
lted
in
E
E
R
o
f
9
.
9
4
%.
Md
.
I
q
b
al
Qu
r
aish
i
et
al
[
1
]
h
av
e
p
r
o
p
o
s
ed
in
th
eir
p
ap
er
an
A
r
ti
f
icia
l
Neu
r
al
Net
w
o
r
k
ap
p
r
o
ac
h
w
h
ic
h
i
m
p
le
m
e
n
ts
a
n
Au
to
m
a
ted
Sig
n
a
tu
r
e
Ver
i
f
icatio
n
a
n
d
Au
t
h
en
tica
tio
n
s
y
s
te
m
.
T
h
eir
m
et
h
o
d
co
m
p
r
is
e
s
o
f
v
ar
io
u
s
tr
a
n
s
f
o
r
m
atio
n
tech
n
iq
u
e
s
f
r
o
m
t
h
e
s
p
atial
as
w
ell
as
f
r
eq
u
en
c
y
d
o
m
ain
.
I
t
also
i
m
p
le
m
e
n
t
s
th
e
u
s
e
o
f
R
ip
let
-
I
I
tr
an
s
f
o
r
m
atio
n
to
ex
tr
ac
t
t
h
e
r
eg
io
n
o
f
i
n
ter
es
t.
T
o
en
h
an
ce
t
h
e
i
m
ag
e,
f
u
r
th
er
it
i
m
p
le
m
e
n
ts
th
e
u
s
e
o
f
L
o
g
P
o
lar
T
r
an
s
f
o
r
m
ati
o
n
.
T
h
ey
h
a
v
e
i
m
p
le
m
e
n
ted
a
Feed
Fo
r
w
ar
d
B
ac
k
P
r
o
p
ag
atio
n
Neu
r
al
Ne
t
w
o
r
k
f
o
r
th
e
v
er
i
f
icatio
n
an
d
au
t
h
e
n
ticatio
n
.
T
h
e
y
h
a
v
e
co
n
s
id
er
ed
3
0
n
eu
r
o
n
s
in
t
h
e
h
id
d
en
la
y
er
o
f
th
e
A
NN
T
h
e
s
y
s
te
m
p
r
o
p
o
s
ed
b
y
th
e
au
t
h
o
r
s
,
h
as
th
e
ac
cu
r
ac
y
o
f
9
6
.
1
5
%,
w
it
h
th
e
f
o
r
g
er
y
d
etec
tio
n
r
ate
o
f
9
2
%.
T
h
e
Fals
e
A
cc
ep
tan
ce
R
a
te
(
FAR
)
is
f
o
u
n
d
to
b
e
5
.
2
8
%,
an
d
Fals
e
R
ej
ec
tio
n
R
ate
(
FR
R
)
o
f
2
.
5
6
%.
T
h
e
au
th
o
r
s
h
av
e
co
m
p
ar
ed
th
e
ir
s
y
s
te
m
w
it
h
o
th
er
e
x
is
t
in
g
s
y
s
te
m
a
n
d
h
av
e
f
o
u
n
d
th
at
t
h
eir
p
r
o
p
o
s
ed
s
y
s
te
m
h
as
b
etter
p
er
f
o
r
m
a
n
ce
a
s
co
m
p
ar
ed
t
o
o
th
er
s
.
T
h
e
d
r
a
w
b
ac
k
is
th
at
t
h
e
te
s
t
n
ee
d
s
to
b
e
tr
ain
ed
b
e
f
o
r
e
th
e
i
m
p
le
m
en
ta
tio
n
,
w
h
i
c
h
is
ti
m
e
co
n
s
u
m
in
g
.
T
h
er
e
ca
n
b
e
f
u
r
th
er
i
m
p
r
o
v
e
m
e
n
t
i
s
t
h
e
s
y
s
te
m
w
i
th
b
etter
p
er
f
o
r
m
a
n
ce
r
ates.
Oth
m
an
o
-
k
h
ali
f
a
et
al.
[
2
]
h
a
s
r
ev
ie
w
ed
o
f
f
li
n
e
s
i
g
n
at
u
r
e
v
er
if
icatio
n
s
ch
e
m
e
s
in
th
eir
p
a
p
er
.
T
h
ey
h
av
e
co
n
s
id
er
ed
th
e
A
r
ti
f
icia
l
Neu
r
al
Net
w
o
r
k
T
ec
h
n
iq
u
e
,
an
d
h
av
e
co
m
p
ar
ed
v
ar
io
u
s
o
f
f
lin
e
s
i
g
n
at
u
r
e
v
er
if
ica
tio
n
ap
p
r
o
ac
h
es
an
d
th
eir
is
s
u
es.
Fo
r
t
h
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
o
f
th
e
d
ata
ac
q
u
ir
ed
th
e
h
a
v
e
u
s
ed
tec
h
n
iq
u
e
s
s
u
c
h
as
B
ac
k
g
r
o
u
n
d
E
li
m
in
a
t
io
n
,
No
is
e
R
ed
u
ctio
n
,
T
h
in
n
i
n
g
a
n
d
W
id
th
No
r
m
a
lizatio
n
.
Fo
r
th
e
p
u
r
p
o
s
e
o
f
f
ea
t
u
r
e
ex
tr
ac
tio
n
,
t
h
e
y
h
a
v
e
co
n
s
id
er
ed
th
e
g
lo
b
al,
g
eo
m
e
tr
ic,
tex
t
u
r
e,
m
as
k
a
n
d
g
r
id
f
ea
tu
r
es.
T
h
e
y
h
av
e
ex
p
lain
ed
h
o
w
t
h
e
A
NN
ap
p
r
o
ac
h
w
o
r
k
s
i
n
t
h
e
s
ig
n
at
u
r
e
Ver
if
icatio
n
a
n
d
w
h
at
s
tep
s
ar
e
in
v
o
lv
ed
.
T
h
e
au
th
o
r
s
h
a
v
e
also
p
o
in
ted
o
u
t
th
at
t
h
e
m
a
in
co
n
ce
r
n
o
f
t
h
e
s
i
g
n
at
u
r
e
v
er
i
f
icat
io
n
s
y
s
te
m
i
s
to
p
r
o
v
id
e
th
e
h
ig
h
s
ec
u
r
it
y
to
ac
ce
s
s
a
n
y
co
n
f
id
e
n
tial t
h
i
n
g
s
th
o
s
e
ar
e
h
ig
h
l
y
r
estricte
d
.
3
.
P
RO
P
O
SE
D
M
E
T
H
O
DO
L
O
G
Y
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
aim
s
at
d
ev
elo
p
in
g
a
u
to
m
a
tic
o
f
f
lin
e
s
i
g
n
at
u
r
e
v
er
i
f
icatio
n
a
n
d
f
o
r
g
e
r
y
d
etec
tio
n
s
y
s
te
m
.
Fi
g
.
1
s
h
o
w
s
t
h
e
al
g
o
r
ith
m
t
h
at
i
s
u
s
ed
in
o
r
d
er
to
b
u
ild
th
e
a
u
to
m
ated
s
ig
n
at
u
r
e
v
er
if
ica
tio
n
a
n
d
f
o
r
g
er
y
d
etec
t
io
n
s
y
s
te
m
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
t
e
m
h
a
s
b
ee
n
d
iv
id
ed
in
to
t
w
o
p
ar
ts
n
a
m
el
y
:
[
1
]
T
r
ain
in
g
[
2
]
T
esti
n
g
3
.
1
.
T
ra
ini
ng
P
ha
s
e:
I
n
th
e
tr
ain
in
g
p
ar
t o
f
th
e
s
y
s
te
m
,
th
e
f
o
llo
w
in
g
s
tep
s
ar
e
p
er
f
o
r
m
ed
:
3
.
1
.
1
.
I
m
a
g
e
Da
t
a
ba
s
e
Cre
a
t
io
n:
T
h
e
i
m
ag
e
s
ar
e
co
llected
f
o
r
tr
ain
in
g
an
d
ar
e
s
to
r
ed
in
a
d
atab
ase.
T
h
e
i
m
a
g
es
ar
e
co
llected
b
y
s
ca
n
n
in
g
t
h
e
m
f
r
o
m
a
p
h
y
s
ical
p
ap
er
s
o
u
r
ce
.
T
h
e
d
atab
ase
u
s
ed
i
s
a
s
elf
-
cr
ea
ted
d
atab
ase
w
h
ic
h
co
n
tain
s
s
ig
n
atu
r
es
o
f
t
h
r
ee
d
if
f
er
e
n
t
p
eo
p
le.
T
h
e
d
atab
ase
co
n
s
is
t
s
o
f
f
i
f
tee
n
s
ig
n
at
u
r
es
b
elo
n
g
i
n
g
to
ea
c
h
p
er
s
o
n
,
an
d
s
u
m
m
i
n
g
u
p
to
b
e
f
o
r
t
y
-
f
iv
e
s
ig
n
at
u
r
es
i
n
to
tal.
Mo
r
e
s
ig
n
atu
r
es
ca
n
b
e
ad
d
ed
to
th
e
d
atab
ase
ea
s
il
y
a
n
d
also
th
e
n
u
m
b
er
o
f
s
ig
n
at
u
r
es p
er
p
er
s
o
n
ca
n
also
b
e
in
cr
ea
s
ed
o
r
d
ec
r
ea
s
ed
.
3
.
1
.
2
.
P
re
-
P
ro
ce
s
s
ing
:
I
n
th
is
s
tep
,
ea
ch
o
f
th
e
s
ca
n
n
ed
s
i
g
n
at
u
r
e
g
o
es
t
h
r
o
u
g
h
a
s
er
ies
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
s
w
h
ich
i
n
cl
u
d
e
th
e
f
o
llo
w
in
g
[
1
5
]
:
[
1
]
I
m
ag
e
R
esizi
n
g
: T
h
e
i
m
a
g
e
is
r
esized
to
a
p
r
e
-
d
ef
in
ed
s
ize
o
f
1
2
8
x
1
2
8
p
ix
els.
[
2
]
B
in
ar
izatio
n
:
A
f
ter
r
esizi
n
g
t
h
e
i
m
a
g
e,
t
h
e
i
m
a
g
e
i
s
b
i
n
ar
ized
,
i.e
.
it
i
s
co
n
v
er
ted
to
b
lack
a
n
d
w
h
ite
[
1
4
]
.
[
3
]
T
h
in
n
i
n
g
:
Af
ter
th
e
p
r
o
ce
s
s
o
f
b
in
ar
izatio
n
,
t
h
e
i
m
a
g
e
g
o
es t
h
r
o
u
g
h
th
e
p
r
o
ce
s
s
o
f
t
h
i
n
n
in
g
,
i
.
e.
th
e
th
ic
k
n
e
s
s
o
f
th
e
s
tr
o
k
es
o
f
t
h
e
s
ig
n
at
u
r
e
is
t
h
i
n
n
ed
d
o
w
n
t
o
a
s
in
g
le
p
ix
el.
I
t
i
s
d
o
n
e
is
o
r
d
er
t
o
ex
clu
d
e
th
e
v
ar
iatio
n
s
in
t
h
ick
n
es
s
o
f
s
i
g
n
atu
r
e
w
h
ic
h
m
a
y
o
cc
u
r
d
u
e
to
t
h
e
u
s
e
o
f
d
if
f
er
en
t
t
y
p
es
o
f
p
en
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Offlin
e
S
ig
n
a
tu
r
e
V
erif
ica
tio
n
a
n
d
F
o
r
g
ery
Dete
cti
o
n
B
a
s
ed
o
n
C
o
mp
u
ter V
is
io
n
…
(
Ga
u
ta
m
S
.
P
r
a
ka
s
h
)
159
[
4
]
R
o
tatio
n
:
T
h
e
i
m
a
g
e
i
s
t
h
en
r
o
tated
o
n
t
h
e
b
a
s
is
o
f
t
h
e
lo
wer
m
o
s
t
p
i
x
els.
W
h
en
a
p
er
s
o
n
s
i
g
n
s
a
d
o
cu
m
en
t,
d
ep
en
d
in
g
o
n
t
h
e
w
r
iti
n
g
s
t
y
le
o
f
t
h
e
p
er
s
o
n
,
t
h
er
e
is
a
ce
r
tain
a
n
g
le
to
t
h
e
s
i
g
n
at
u
r
e
i
n
w
h
ic
h
it is
d
o
n
e.
T
h
is
p
r
o
ce
s
s
s
tr
aig
h
te
n
s
o
u
t t
h
e
s
i
g
n
at
u
r
e.
[
5
]
C
r
o
p
p
in
g
o
f
th
e
i
m
ag
e
:
Af
ter
th
e
i
m
a
g
e
is
r
o
tated
,
th
e
ex
ce
s
s
ar
ea
ar
o
u
n
d
th
e
s
ig
n
atu
r
e
is
r
e
m
o
v
ed
an
d
th
e
i
m
ag
e
i
s
cr
o
p
p
ed
to
th
e
o
u
ter
m
o
s
t p
ix
el
s
i
n
f
o
u
r
d
ir
ec
tio
n
s
,
i.e
.
to
p
,
b
o
tto
m
,
lef
t a
n
d
r
ig
h
t.
Fig
u
r
e
1
.
P
r
o
p
o
s
ed
A
lg
o
r
ith
m
3
.
1
.
3
.
F
ea
t
ure
E
x
t
ra
ct
io
n:
A
f
ter
th
e
i
m
a
g
e
h
as
g
o
n
e
t
h
r
o
u
g
h
th
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
,
v
ar
io
u
s
f
ea
t
u
r
es
ar
e
ex
tr
ac
ted
f
r
o
m
t
h
e
i
m
ag
e.
T
h
e
ex
tr
ac
ted
f
ea
t
u
r
es o
u
t o
f
ea
c
h
i
m
a
g
e
ar
e
th
e
n
s
to
r
ed
in
a
M
AT
L
A
B
f
ile.
Fo
llo
w
i
n
g
u
n
iq
u
e
f
ea
t
u
r
es a
r
e
ex
tr
ac
ted
f
r
o
m
ea
c
h
th
e
i
m
a
g
e
s
:
[
1
]
Heig
h
t
-
W
id
th
R
at
io
: A
f
ter
t
h
e
i
m
ag
e
i
s
cr
o
p
p
ed
,
h
eig
h
t
-
w
id
t
h
r
atio
o
f
th
e
s
ig
n
at
u
r
e
is
ca
lc
u
lated
.
[
2
]
C
en
tr
o
id
o
f
Si
g
n
at
u
r
e
:
T
h
e
ce
n
tr
o
id
o
r
th
e
b
ar
y
ce
n
tr
e
o
f
th
e
i
m
a
g
e
i
s
ca
lc
u
lated
.
T
h
e
ce
n
tr
o
id
g
iv
e
s
th
e
ce
n
tr
al
p
o
in
t
o
f
t
h
e
s
ig
n
at
u
r
e
w
h
ic
h
is
a
u
n
iq
u
e
s
i
g
n
at
u
r
e
ch
ar
ac
ter
is
tic.
T
h
e
s
ig
n
atu
r
e
i
s
b
r
o
k
en
d
o
w
n
v
er
ticall
y
i
n
to
tw
o
h
a
lv
e
s
,
an
d
th
e
ce
n
tr
o
id
o
f
th
e
ea
ch
h
al
f
is
ca
lc
u
lated
.
[
3
]
First
Der
iv
at
iv
e
s
:
T
h
e
f
ir
s
t
d
er
iv
ati
v
es
o
f
t
h
e
i
m
a
g
e
m
atr
i
x
ar
e
ca
lcu
lated
r
o
w
w
i
s
e
as
w
ell
as
co
lu
m
n
w
i
s
e.
[
4
]
Seco
n
d
Der
iv
a
tiv
e
s
:
Af
ter
t
h
e
ca
lcu
lat
io
n
o
f
f
ir
s
t
d
er
iv
ati
v
e
s
,
th
e
s
ec
o
n
d
d
er
iv
ati
v
es
o
f
th
e
i
m
a
g
e
m
atr
i
x
ar
e
ca
lcu
lated
b
o
th
r
o
w
an
d
co
lu
m
n
w
i
s
e.
[
5
]
Qu
ad
r
an
t
A
r
ea
s
:
T
h
e
i
m
a
g
e
is
b
r
o
k
en
d
o
w
n
i
n
to
f
o
u
r
q
u
ad
r
an
ts
,
an
d
t
h
en
t
h
e
ar
ea
o
f
th
e
s
ig
n
at
u
r
e
p
ix
els
i
n
ea
c
h
q
u
ad
r
an
t
i
s
ca
l
cu
lated
.
T
h
is
ar
ea
is
t
h
e
ar
e
a
o
f
s
tr
o
k
e
s
o
f
t
h
e
s
ig
n
at
u
r
e
in
t
h
at
p
ar
ticu
lar
q
u
ad
r
an
t a
n
d
d
o
es n
o
t in
clu
d
e
t
h
e
ar
ea
o
f
th
e
b
ac
k
g
r
o
u
n
d
.
[
6
]
C
OM
Ma
tr
ix
:
C
OM
Ma
tr
i
x
o
r
C
o
-
Occ
u
r
r
en
ce
Ma
tr
ix
r
ef
er
s
to
th
e
d
is
tr
ib
u
tio
n
o
f
t
h
e
co
-
o
cc
u
r
r
in
g
v
alu
e
s
at
a
g
iv
e
n
o
f
f
s
et.
I
t
is
u
s
ed
to
m
ea
s
u
r
e
th
e
te
x
t
u
r
e
o
n
th
e
i
m
a
g
e.
W
h
at
is
d
o
es
is
,
as
o
u
r
i
m
a
g
e
is
i
n
b
lack
an
d
w
h
ite
a
f
ter
th
e
p
r
e
-
p
r
o
ce
s
s
,
t
h
at
m
ea
n
s
t
h
e
i
m
a
g
e
m
atr
i
x
h
as
v
al
u
e
s
eith
er
0
o
r
1
.
I
t
lo
o
k
s
f
o
r
p
atter
n
d
is
tr
ib
u
tio
n
o
f
t
h
ese
v
a
lu
e
s
an
d
lo
o
k
s
w
h
er
e
th
e
p
atter
n
s
0
0
,
0
1
,
1
1
an
d
1
0
o
cc
u
r
.
T
h
e
co
-
o
cc
u
r
r
en
ce
m
atr
ix
is
al
s
o
ca
lcu
lated
f
o
r
th
e
s
i
g
n
atu
r
e.
[
7
]
E
d
g
e
P
o
in
t
C
a
lcu
la
tio
n
:
T
h
e
n
u
m
b
er
o
f
ed
g
e
p
o
in
ts
in
th
e
s
ig
n
at
u
r
e
ar
e
ca
lc
u
lated
w
h
ic
h
g
iv
e
s
a
d
is
tin
ct
c
h
ar
ac
ter
is
tic
ab
o
u
t t
h
e
s
ig
n
at
u
r
e.
[
8
]
Ho
r
izo
n
tal
an
d
Ver
t
ical
Hi
s
to
g
r
a
m
:
E
ac
h
r
o
w
a
n
d
ea
ch
co
lu
m
n
o
f
t
h
e
s
i
g
n
at
u
r
e
i
s
g
o
n
e
th
r
o
u
g
h
an
d
th
e
n
u
m
b
er
o
f
b
lack
p
ix
el
s
is
ca
lcu
lated
.
T
h
e
r
o
w
an
d
t
h
e
co
lu
m
n
w
it
h
th
e
m
a
x
i
m
u
m
n
u
m
b
er
o
f
b
lack
p
ix
e
ls
i
s
r
ec
o
r
d
ed
an
d
u
s
ed
as
a
f
ea
t
u
r
e.
All
t
h
ese
f
ea
t
u
r
es
g
iv
e
o
u
t
u
n
iq
u
e
ch
ar
a
cter
is
tics
ab
o
u
t th
e
s
i
g
n
atu
r
e
a
n
d
ar
e
u
s
ed
f
o
r
class
if
icatio
n
o
f
t
h
e
s
ig
n
atu
r
e
s
.
3
.
1
.
4
.
G
ener
a
t
e
T
ra
ini
ng
F
ea
t
ure
Set
:
I
n
th
is
s
tep
,
o
n
ce
all
th
e
f
ea
t
u
r
es
ca
lcu
la
ted
is
s
av
ed
,
th
e
n
th
e
r
eq
u
ir
ed
o
u
tp
u
t
is
g
en
er
a
ted
o
n
th
e
b
asis
o
f
w
h
ich
t
h
e
Ne
u
r
al
Net
w
o
r
k
is
tr
ain
ed
.
T
h
e
v
ales
ass
ig
n
ed
to
th
e
tr
ai
n
in
g
i
m
a
g
es
c
an
b
e
eit
h
er
0
o
r
1
.
T
h
ese
v
alu
e
s
alo
n
g
w
it
h
th
e
v
alu
es o
f
t
h
e
f
ea
t
u
r
es a
r
e
u
s
ed
t
o
tr
ain
th
e
A
NN.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
3
,
No
.
4
,
Dec
em
b
er
20
1
4
:
1
5
6
–
1
6
5
160
3
.
1
.
5
.
T
ra
ini
ng
Usi
ng
ANN
:
On
ce
th
e
f
ea
tu
r
e
v
a
lu
e
s
an
d
o
u
tp
u
t
v
alu
e
s
o
f
th
e
i
m
ag
e
s
ar
e
d
ec
id
ed
,
th
en
th
e
n
e
u
r
al
n
et
wo
r
k
ca
n
b
e
tr
ain
ed
u
s
in
g
n
eu
r
al
n
et
w
o
r
k
t
o
o
l b
o
x
as sh
o
w
n
in
Fig
2
.
.
T
h
e
co
m
m
a
n
d
to
s
tar
t
n
eu
r
al
n
et
w
o
r
k
i
s
“
n
n
s
tar
t”
i
n
MA
T
L
A
B
w
h
ich
t
h
en
o
p
en
s
t
h
e
to
o
lb
o
x
.
Af
ter
th
e
tr
ain
in
g
o
f
s
y
s
te
m
u
s
in
g
A
NN,
t
h
e
tr
ain
ed
ar
tif
icial
n
eu
r
al
n
et
w
o
r
k
i
s
o
b
ta
in
ed
[
1
3
]
.
Fig
u
r
e
2
.
Net
w
o
r
k
T
o
o
l B
o
x
3
.
1
.
T
esting
P
ha
s
e:
T
h
is
p
h
ase
is
u
s
ed
d
u
r
i
n
g
t
h
e
r
u
n
ti
m
e
i
m
p
le
m
en
tatio
n
o
f
t
h
e
s
y
s
te
m
.
I
t c
o
n
s
is
t
s
o
f
f
o
llo
w
in
g
s
tep
s
.
3
.
2
.
1
.
B
r
o
w
s
e
I
m
a
g
e:
Un
li
k
e
t
h
e
tr
ai
n
in
g
p
ar
t
w
h
er
e
th
e
i
m
a
g
es
ar
e
a
u
to
m
a
ticall
y
r
ea
d
f
r
o
m
th
e
tr
ain
i
n
g
d
ata
b
ase,
in
t
h
e
test
i
n
g
p
ar
t,
th
e
i
m
a
g
e
is
m
an
u
all
y
s
elec
ted
f
r
o
m
t
h
e
test
i
n
g
d
atab
ase.
3
.
2
.
2
.
P
re
-
P
r
o
ce
s
s
:
I
n
th
e
test
i
n
g
p
ar
t,
th
e
i
m
a
g
e
s
elec
ted
g
o
es
th
r
o
u
g
h
th
e
s
a
m
e
p
r
e
-
p
r
o
ce
s
s
s
tep
s
as
i
n
t
h
e
tr
ain
i
n
g
p
ar
t.
3
.
2
.
3
.
F
ea
t
ure
E
x
t
ra
ct
io
n:
I
n
test
i
n
g
p
ar
t
t
h
e
f
ea
t
u
r
es
o
f
t
h
e
s
elec
ted
i
m
a
g
e
ar
e
ca
lc
u
lated
a
n
d
s
to
r
ed
.
T
h
ese
v
alu
es
ar
e
t
h
e
n
later
u
s
ed
f
o
r
t
h
e
clas
s
if
icatio
n
s
tep
.
T
h
e
f
ea
t
u
r
es
ex
tr
ac
ted
in
t
h
e
test
i
n
g
p
h
ase
ar
e
t
h
e
s
a
m
e
as
t
h
at
o
f
t
h
e
tr
ain
i
n
g
p
h
a
s
e
an
d
f
o
llo
w
t
h
e
s
a
m
e
p
r
o
ce
s
s
.
3
.
2
.
4
.
G
ener
a
t
e
T
esting
F
ea
t
ure
Set
:
Af
ter
t
h
e
f
ea
t
u
r
es
ar
e
ca
lc
u
lat
ed
,
th
e
y
ar
e
s
to
r
ed
an
d
ar
e
u
s
e
d
to
g
en
er
ate
th
e
te
s
ti
n
g
f
ea
t
u
r
e
s
et.
T
h
is
f
ea
t
u
r
e
s
et
is
t
h
en
f
ed
in
to
th
e
tr
ain
ed
ANN
s
y
s
te
m
.
3
.
2
.
5
.
Sig
na
t
ure
I
dentif
ica
t
io
n Usin
g
T
ra
i
ned Ar
t
if
icia
l N
eura
l N
et
w
o
rk
:
T
h
e
test
in
g
f
ea
t
u
r
e
s
et
g
e
n
er
ated
in
t
h
e
te
s
ti
n
g
p
ar
t
is
f
ed
to
th
e
tr
ai
n
ed
ar
ti
f
icial
n
e
u
r
al
n
et
w
o
r
k
s
y
s
te
m
t
h
at
w
a
s
o
b
tain
ed
in
t
h
e
tr
ain
i
n
g
p
h
a
s
e,
s
elec
ted
s
ig
n
at
u
r
e
is
t
h
en
cla
s
s
i
f
ied
[
1
3
]
.
B
a
s
ed
o
n
th
e
cla
s
s
i
f
icatio
n
d
o
n
e
b
y
t
h
e
s
y
s
te
m
,
t
h
e
s
ig
n
at
u
r
e
b
elo
n
g
s
to
w
h
ic
h
p
er
s
o
n
is
id
e
n
ti
f
ied
an
d
is
d
is
p
la
y
ed
o
n
s
cr
ee
n
as a
m
e
s
s
a
g
e
w
i
n
d
o
w
.
3
.
2
.
6
.
F
o
r
g
er
y
Det
ec
t
i
o
n:
An
g
le
f
ea
tu
r
es
h
a
v
e
b
ee
n
u
s
e
d
f
o
r
th
e
p
u
r
p
o
s
e
o
f
d
etec
tio
n
o
f
f
o
r
g
er
y
.
T
h
e
f
o
llo
w
i
n
g
a
l
g
o
r
ith
m
i
s
u
s
ed
f
o
r
th
e
p
u
r
p
o
s
e
o
f
f
o
r
g
er
y
d
etec
tio
n
:
[
1
]
T
ak
e
th
e
b
o
tto
m
le
f
t
m
o
s
t p
o
in
t a
s
r
ef
er
e
n
ce
p
o
in
t
f
o
r
th
e
an
g
le
ca
lcu
latio
n
.
[
2
]
C
alcu
late
an
g
le
f
o
r
ea
ch
p
ix
el
o
f
th
e
s
i
g
n
at
u
r
e
f
r
o
m
t
h
is
r
e
f
er
en
ce
p
o
in
t.
[
3
]
Div
id
e
th
e
s
e
an
g
le
s
in
to
1
8
ca
teg
o
r
ies (
0
-
5
,
5
-
1
0
,
.
.
.
.
.
,
8
5
-
9
0
)
.
[
4
]
Fin
d
t
h
e
av
er
ag
e
o
f
an
g
les i
n
t
h
ese
ca
te
g
o
r
ies,
th
u
s
g
i
v
in
g
u
s
w
it
h
ei
g
h
tee
n
f
ea
t
u
r
es p
er
s
ig
n
atu
r
e.
[
5
]
R
ep
ea
t th
ese
s
tep
s
f
o
r
1
0
s
am
p
les f
o
r
ea
ch
p
er
s
o
n
.
[
6
]
Fro
m
th
e
ab
o
v
e
d
ata,
a
f
ea
t
u
r
e
v
ec
to
r
w
il
l b
e
cr
ea
ted
,
w
h
ic
h
w
il
l h
o
ld
th
e
m
i
n
i
m
u
m
,
m
ax
i
m
u
m
a
n
d
av
er
ag
e
a
n
g
l
e
f
o
r
ea
ch
ca
teg
o
r
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Offlin
e
S
ig
n
a
tu
r
e
V
erif
ica
tio
n
a
n
d
F
o
r
g
ery
Dete
cti
o
n
B
a
s
ed
o
n
C
o
mp
u
ter V
is
io
n
…
(
Ga
u
ta
m
S
.
P
r
a
ka
s
h
)
161
T
en
s
am
p
le
s
f
o
r
ea
ch
p
er
s
o
n
ar
e
tak
en
.
An
g
les
f
r
o
m
t
h
ese
s
a
m
p
les
ar
e
ca
lcu
lated
f
r
o
m
th
e
r
ef
er
en
ce
p
o
in
t.
T
h
e
an
g
les
s
o
o
b
tain
ed
ar
e
th
en
d
iv
id
ed
in
to
ca
te
g
o
r
ies
o
f
in
ter
v
al
o
f
5
.
T
h
e
ca
teg
o
r
ies
in
to
w
h
ic
h
t
h
e
an
g
le
s
ar
e
d
iv
id
ed
ar
e
0
-
5
,
5
-
1
0
,
1
0
-
1
5
,
.
.
.
.
till
8
5
-
9
0
.
T
h
is
p
r
o
v
id
es
u
s
w
it
h
t
h
e
to
tal
o
f
1
8
ca
teg
o
r
ies.
T
h
e
av
er
ag
e
o
f
an
g
les
o
f
ea
c
h
ca
teg
o
r
y
is
tak
e
n
p
r
o
v
id
in
g
u
s
w
it
h
to
tal
o
f
eig
h
tee
n
f
ea
t
u
r
es
p
er
s
ig
n
at
u
r
e.
T
h
is
d
ata
is
t
h
en
u
s
ed
to
cr
ea
te
a
f
ea
tu
r
e
v
ec
to
r
w
h
ic
h
h
o
ld
s
t
h
e
m
a
x
i
m
u
m
,
m
i
n
i
m
u
m
a
n
d
t
h
e
av
er
ag
e
a
n
g
les
f
o
r
ea
ch
ca
teg
o
r
y
.
T
ab
le
1
.
A
n
g
le
f
ea
t
u
r
e
v
ec
to
r
f
o
r
1
0
s
am
p
le
s
o
f
a
p
er
s
o
n
.
A
n
g
l
e
R
a
n
g
e
M
i
n
A
v
g
M
a
x
0
--
5
0
.
0
2
2
1
7
5
4
2
9
0
.
0
2
5
6
5
6
4
0
1
0
.
0
2
7
0
0
6
0
6
4
5
--
10
0
.
0
8
2
0
1
7
7
1
6
0
.
0
8
4
6
9
9
6
1
2
0
.
0
8
6
9
0
8
5
3
5
10
--
15
0
.
1
3
1
4
5
7
7
7
8
0
.
1
3
6
6
4
5
8
9
1
0
.
1
4
2
6
0
1
2
9
2
15
--
20
0
.
1
9
3
4
1
7
4
5
5
0
.
2
0
2
6
2
4
0
2
5
0
.
2
0
6
7
4
5
7
9
5
20
--
25
0
.
2
5
1
1
9
1
8
3
5
0
.
2
5
5
5
9
5
5
2
6
0
.
2
6
3
3
7
5
0
2
8
25
--
30
0
.
3
0
3
0
6
1
4
4
2
0
.
3
1
3
6
3
3
1
4
7
0
.
3
2
0
9
9
7
0
4
5
30
--
35
0
.
3
6
5
3
9
4
1
2
2
0
.
3
7
1
4
6
3
1
3
2
0
.
3
7
9
3
7
5
0
2
8
35
--
40
0
.
4
2
1
9
9
9
9
2
5
0
.
4
2
6
3
3
5
8
7
7
0
.
4
3
1
0
9
4
6
8
9
40
--
45
0
.
4
6
8
6
0
4
9
1
8
0
.
4
7
7
4
1
6
6
4
4
0
.
4
8
4
9
8
5
6
3
2
45
--
50
0
.
5
3
1
2
2
2
8
0
2
0
.
5
3
8
8
8
4
6
5
1
0
.
5
4
7
9
1
5
0
2
5
50
--
55
0
.
5
8
3
4
2
2
1
6
5
0
.
5
9
6
4
0
0
7
2
1
0
.
6
0
5
5
5
6
0
8
3
55
--
60
0
.
6
4
4
6
2
8
0
1
3
0
.
6
5
6
2
7
1
0
8
9
0
.
6
6
5
7
7
3
4
3
6
60
--
65
0
.
7
0
4
0
5
6
2
7
2
0
.
7
1
1
3
3
0
6
5
7
0
.
7
1
6
5
2
9
6
1
5
65
--
70
0
.
7
5
5
0
2
8
7
4
6
0
.
7
6
8
2
2
4
5
3
0
.
7
7
8
9
2
6
8
6
7
70
--
75
0
.
8
1
2
6
5
9
7
2
6
0
.
8
2
4
1
2
6
9
3
4
0
.
8
3
0
1
9
3
5
8
1
75
--
80
0
.
8
6
8
1
4
7
0
5
7
0
.
8
7
8
3
9
5
9
7
8
0
.
8
8
9
4
5
7
3
1
3
80
--
85
0
.
9
2
0
4
3
6
0
2
5
0
.
9
3
7
9
6
0
0
8
1
0
.
9
6
5
9
2
3
4
7
1
85
--
90
0
0
.
8
8
8
7
9
7
4
9
2
1
Fig
u
r
e
3
.
GUI
f
o
r
Fu
zz
y
T
o
o
lb
o
x
Fig
u
r
e
4
.
C
r
ea
tio
n
o
f
A
n
g
le
f
ea
tu
r
es a
s
a
n
I
n
p
u
t
Var
iab
le
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
3
,
No
.
4
,
Dec
em
b
er
20
1
4
:
1
5
6
–
1
6
5
162
On
ce
th
e
f
ea
t
u
r
e
v
ec
to
r
f
i
le
is
cr
ea
ted
,
w
e
u
s
e
t
h
i
s
as a
n
i
n
p
u
t f
o
r
th
e
F
u
zz
y
T
o
o
lb
o
x
.
T
o
o
p
en
th
e
f
u
zz
y
to
o
lb
o
x
,
w
e
w
r
i
te
“f
u
zz
y
”
i
n
t
h
e
M
AT
L
A
B
co
m
m
a
n
d
w
i
n
d
o
w
[
1
1
]
.
Fig
u
r
e
3
s
h
o
w
s
t
h
e
i
n
ter
f
ac
e
o
f
th
e
F
u
z
z
y
T
o
o
lb
o
x
.
On
ce
th
e
to
o
lb
o
x
h
as
o
p
en
ed
,
w
e
u
s
e
t
h
e
Ma
m
d
a
n
i
F
u
zz
y
Mo
d
el
an
d
ad
d
th
e
an
g
le
f
ea
t
u
r
es
o
f
th
e
s
i
g
n
at
u
r
e
as v
ar
iab
les.
Fi
g
4
s
h
o
w
s
th
e
a
d
d
itio
n
o
f
n
e
w
v
ar
iab
les to
th
e
to
o
lb
o
x
.
On
ce
w
e
h
a
v
e
ad
d
ed
th
e
v
ar
i
ab
le
t
o
th
e
f
u
zz
y
to
o
lb
o
x
,
w
e
d
ef
in
e
th
e
m
e
m
b
er
s
h
ip
f
u
n
cti
o
n
to
ea
c
h
v
ar
iab
le
ac
co
r
d
in
g
to
th
e
v
ar
iab
le
v
alu
e
s
.
Fig
u
r
e
5
s
h
o
w
s
th
e
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
d
e
f
in
ed
f
o
r
a
v
ar
iab
le.
Fig
u
r
e
6
.
Fu
zz
y
R
u
le
Gen
er
ati
o
n
Af
ter
t
h
e
g
e
n
er
atio
n
o
f
th
e
r
u
les,
th
e
f
u
zz
y
m
o
d
el
is
s
a
v
ed
w
i
th
a
n
a
m
e
o
f
“
s
i
g
n
atu
r
e.
f
is
”.
T
h
i
s
co
m
p
lete
s
o
u
r
f
u
zz
y
tr
ain
i
n
g
p
h
ase.
Fo
r
th
e
test
i
n
g
p
h
a
s
e,
a
n
i
n
p
u
t
i
m
ag
e
is
r
ea
d
an
d
i
ts
a
n
g
le
f
ea
t
u
r
es
ar
e
e
x
tr
ac
ted
.
T
h
en
t
h
e
tr
ai
n
ed
f
u
zz
y
m
o
d
el
is
r
ea
d
an
d
is
ev
alu
ated
w
ith
t
h
e
f
ea
t
u
r
es
ex
tr
ac
ted
f
r
o
m
t
h
e
test
i
m
ag
e
f
ile.
T
h
e
v
al
u
e
g
e
n
er
ated
f
r
o
m
t
h
e
test
i
n
g
is
u
s
ed
to
clas
s
if
y
t
h
e
s
i
g
n
a
tu
r
e
as Fo
r
g
ed
o
r
Gen
u
in
e.
T
h
e
v
ar
io
u
s
test
i
n
g
r
esu
lts
o
f
t
h
e
alg
o
r
it
h
m
f
o
r
th
e
te
s
t i
m
ag
e
ar
e
p
r
esen
ted
in
th
e
n
ex
t
s
ec
t
io
n
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
s
y
s
te
m
i
s
d
ev
elo
p
ed
w
ith
a
u
s
er
f
r
ien
d
l
y
GUI
.
T
h
e
r
esu
lts
a
t
v
ar
io
u
s
s
tag
e
s
o
f
t
h
e
a
p
p
licatio
n
ar
e
d
is
cu
s
s
ed
as b
elo
w
:
-
4
.
1
.
I
m
a
g
e
Select
io
n
T
h
e
i
m
ag
e
s
o
f
t
w
o
p
er
s
o
n
s
a
r
e
u
s
ed
f
o
r
te
s
ti
n
g
t
h
e
s
y
s
te
m
.
T
o
tal
5
g
en
u
in
e
an
d
f
o
r
g
ed
i
m
a
g
es
o
f
ea
ch
p
er
s
o
n
ar
e
u
s
ed
f
o
r
test
i
n
g
.
T
h
e
te
s
t i
m
a
g
e
i
s
s
elec
ted
u
s
in
g
th
e
s
elec
t
i
m
ag
e
b
u
t
to
n
i
n
th
e
GUI
a
s
s
h
o
w
n
in
Fi
g
u
r
e
.
7
.
Fig
u
r
e
7
.
GUI
f
o
r
T
esti
n
g
P
h
ase.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Offlin
e
S
ig
n
a
tu
r
e
V
erif
ica
tio
n
a
n
d
F
o
r
g
ery
Dete
cti
o
n
B
a
s
ed
o
n
C
o
mp
u
ter V
is
io
n
…
(
Ga
u
ta
m
S
.
P
r
a
ka
s
h
)
163
4
.
2
.
P
re
-
pro
ce
s
s
ing
T
h
e
s
elec
ted
i
m
a
g
e
w
i
ll
u
n
d
e
r
g
o
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
s
a
s
d
is
c
u
s
s
ed
i
n
s
ec
tio
n
I
I
I
.
T
h
e
o
u
tp
u
t
o
f
th
ese
s
tep
s
ca
n
b
e
s
h
o
w
n
u
s
i
n
g
th
e
R
e
s
izi
n
g
,
B
i
n
ar
izatio
n
,
T
h
in
n
i
n
g
a
n
d
C
r
o
p
p
in
g
b
u
tt
o
n
o
f
G
UI
.
On
e
o
f
th
ese
s
tep
s
is
s
h
o
w
n
in
F
ig
u
r
e
8
.
Fig
u
r
e
8
.
I
m
a
g
e
o
b
tain
ed
af
ter
C
r
o
p
p
in
g
4
.
3
.
Sig
na
t
ure
I
dentif
ica
t
io
n
W
h
en
th
e
“
I
d
en
ti
f
icatio
n
”
b
u
tt
o
n
i
n
t
h
e
GUI
i
s
clic
k
ed
,
t
h
e
a
lr
ea
d
y
d
is
c
u
s
s
ed
s
tep
s
o
f
test
i
n
g
p
h
a
s
ed
ar
e
d
o
n
e
in
th
e
b
ac
k
g
r
o
u
n
d
,
a
n
d
th
e
r
es
u
lt
is
s
h
o
w
ed
to
th
e
u
s
er
.
T
h
e
r
esu
lt
is
a
p
o
p
u
p
m
ess
a
g
e
as
s
h
o
w
n
i
n
Fig
u
r
e
.
9
th
at
d
is
p
la
y
s
th
e
n
a
m
e
o
f
t
h
e
p
er
s
o
n
to
w
h
o
m
th
e
s
ig
n
at
u
r
e
b
elo
n
g
s
.
Fig
u
r
e
9
.
Selecte
d
I
m
a
g
e
b
elo
n
g
s
to
Gau
ta
m
P
r
ak
as
h
4
.
4
.
F
o
rg
er
y
Det
ec
t
io
n
On
ce
t
h
e
s
ig
n
at
u
r
e
is
id
en
ti
f
ie
d
th
at
to
w
h
o
m
it
b
elo
n
g
s
,
th
e
n
e
x
t
s
tep
is
to
d
etec
t
f
o
r
g
er
y
.
W
h
en
w
e
click
o
n
t
h
e
“
Dete
c
t
Fo
r
g
er
y
”
b
u
tto
n
in
t
h
e
GUI
,
a
m
e
s
s
a
g
e
i
s
d
is
p
la
y
ed
in
a
p
o
p
u
p
w
i
n
d
o
w
s
ta
tin
g
th
e
s
ig
n
at
u
r
e
is
“
Ge
n
u
i
n
e”
o
r
“
Fo
r
g
ed
”.
Fig
1
0
s
h
o
w
s
th
at
t
h
e
s
e
lecte
d
i
m
ag
e
i
s
Ge
n
u
i
n
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
3
,
No
.
4
,
Dec
em
b
er
20
1
4
:
1
5
6
–
1
6
5
164
Fig
u
r
e
10
.
Selecte
d
Sig
n
a
tu
r
e
is
Gen
u
i
n
e
An
o
th
er
r
es
u
lt i
s
s
h
o
w
n
i
n
Fi
g
u
r
e
.
1
1
f
o
r
th
e
f
o
r
g
ed
i
m
a
g
e
o
f
s
ec
o
n
d
p
er
s
o
n
.
Fig
u
r
e
11
.
Selecte
d
Sig
n
a
tu
r
e
is
Fo
r
g
ed
5.
CO
NCLU
SI
O
N
T
h
e
d
escr
ib
ed
s
y
s
te
m
f
o
r
Au
t
o
m
a
tic
Si
g
n
at
u
r
e
Ver
if
icatio
n
an
d
Fo
r
g
er
y
Dete
ctio
n
h
av
e
n
u
m
er
o
u
s
ap
p
licatio
n
s
i
n
v
ar
io
u
s
f
ield
s
l
ik
e
B
an
k
-
C
h
eq
u
e
p
r
o
ce
s
s
i
n
g
,
A
T
M
ac
ce
s
s
,
Do
cu
m
e
n
t
Au
t
h
en
ticatio
n
s
e
tc
a
n
d
ca
n
b
e
u
s
ed
f
o
r
th
e
p
u
r
p
o
s
e
o
f
au
th
e
n
ticat
in
g
t
h
e
s
i
g
n
at
u
r
e.
Fu
r
t
h
er
,
t
h
e
v
ar
iat
io
n
i
n
p
er
s
o
n
alit
y
o
f
s
i
g
n
at
u
r
es,
b
ec
au
s
e
o
f
ag
e,
s
ic
k
n
es
s
,
g
eo
g
r
ap
h
ic
lo
ca
tio
n
an
d
e
m
o
tio
n
al
s
tate
o
f
t
h
e
p
er
s
o
n
a
ctu
ates t
h
e
p
r
o
b
le
m
.
An
o
t
h
er
p
r
o
b
lem
ass
o
ciate
d
w
it
h
o
f
f
li
n
e
s
ig
n
at
u
r
e
v
er
if
ica
tio
n
i
s
th
at,
f
o
r
s
ec
u
r
it
y
r
ea
s
o
n
s
,
it i
s
n
o
t
v
er
y
ea
s
y
t
o
m
a
k
e
a
s
i
g
n
atu
r
e
d
ataset
o
f
r
ea
l d
o
cu
m
en
ts
s
u
c
h
as b
an
k
i
n
g
d
o
cu
m
e
n
t
s
.
T
h
ese
p
r
o
b
lem
s
ca
n
b
e
co
n
s
id
er
ed
f
o
r
i
m
p
r
o
v
in
g
th
e
s
y
s
te
m
RE
F
E
R
E
NC
E
S
[1
]
M
d
.
Iq
b
a
l
Qu
ra
ish
i
,
A
rin
d
a
m
D
a
s
a
n
d
S
a
ik
a
t
Ro
y
(2
0
1
3
),
"
A
No
v
e
l
S
ig
n
a
tu
re
Ver
if
ica
ti
o
n
a
n
d
A
u
t
h
e
n
ti
c
a
ti
o
n
S
y
ste
m Usin
g
Ima
g
e
T
ra
n
sfo
rm
a
t
io
n
a
n
d
Arti
fi
c
ia
l
Ne
u
ra
l
Ne
twr
o
k
"
,
Na
ru
la
In
stit
u
te o
f
T
e
c
h
n
o
lo
g
y
,
Ko
lk
a
ta.
[2
]
Ot
h
m
a
n
o
-
k
h
a
li
f
a
,
M
d
.
K
h
o
rsh
e
d
A
la
m
a
n
d
A
ish
a
Ha
ss
a
n
A
b
d
a
ll
a
(2
0
1
3
),
"
An
Ev
a
lu
a
ti
o
n
o
n
Offl
i
n
e
S
i
g
n
a
tu
re
Ver
if
ica
ti
o
n
u
si
n
g
Arti
fi
c
ia
l
Ne
u
ra
l
Ne
two
rk
Ap
p
r
o
a
c
h
"
,
I
n
tern
a
t
io
n
a
l
Co
n
f
e
re
n
c
e
o
n
Co
m
p
u
ti
n
g
,
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ic E
n
g
in
e
e
rin
g
(ICCEE
E
).
[3
]
Ra
m
e
e
z
W
a
ji
d
a
n
d
A
ti
f
Bin
M
a
n
so
o
r,
"
Clas
sif
ier
P
e
rf
o
rm
a
n
c
e
Ev
a
lu
a
ti
o
n
F
o
r
Off
li
n
e
S
ig
n
a
tu
re
V
e
rif
ica
ti
o
n
Us
in
g
L
o
c
a
l
Bin
a
r
y
P
a
tt
e
rn
s"
,
In
st
it
u
te
o
f
Av
io
n
ics
&
A
e
ro
n
a
u
ti
c
s,
A
ir
U
n
iv
e
rsity
,
Isla
m
a
b
a
d
,
P
a
k
istan
.
[4
]
Mu
h
a
m
m
a
d
I
m
r
a
n
M
a
li
k
,
M
a
rc
u
s
L
i
w
ic
k
i
a
n
d
A
n
d
re
a
s
De
n
g
e
l,
"
Ev
a
lu
a
ti
o
n
o
f
L
o
c
a
l
a
n
d
G
lo
b
a
l
F
e
a
tu
re
s
f
o
r
Off
li
n
e
S
ig
n
a
tu
re
V
e
rif
ica
ti
o
n
"
,
G
e
r
m
a
n
Re
se
a
r
c
h
Ce
n
ter f
o
r
A
I
(
DFKI
Gm
b
H).
[5
]
Ju
a
n
Hu
a
n
d
Yo
u
b
i
n
Ch
e
n
(2
0
1
3
),
"
Offl
in
e
S
ig
n
a
t
u
re
Ver
if
ica
t
io
n
Us
in
g
Re
a
l
A
d
a
b
o
o
st
Cl
a
ss
if
ier
Co
mb
in
a
ti
o
n
o
f
Pse
u
d
o
-
d
y
n
a
mic
Fea
tu
re
s"
,
1
2
th
In
tern
a
ti
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Do
c
u
m
e
n
t
A
n
a
l
y
sis &
Re
c
o
g
n
it
io
n
.
[6
]
V
a
ib
h
a
v
S
h
a
h
,
Um
a
n
g
S
a
n
g
h
a
v
i,
Ud
it
S
h
a
h
,
"
Off
-
li
n
e
S
ig
n
a
tu
re
V
e
rif
ica
ti
o
n
Us
in
g
Cu
rv
e
F
it
ti
n
g
A
l
g
o
rit
h
m
w
it
h
Ne
u
ra
l
Ne
tw
o
rk
s
"
,
D
w
a
r
k
a
d
a
s J.
S
a
n
g
h
v
i
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
M
u
m
b
a
i.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Offlin
e
S
ig
n
a
tu
r
e
V
erif
ica
tio
n
a
n
d
F
o
r
g
ery
Dete
cti
o
n
B
a
s
ed
o
n
C
o
mp
u
ter V
is
io
n
…
(
Ga
u
ta
m
S
.
P
r
a
ka
s
h
)
165
[7
]
M
.
Na
siri,
S
.
Ba
y
a
ti
a
n
d
F
.
S
a
f
i,
"
A
F
u
z
z
y
A
p
p
ro
a
c
h
f
o
r
t
h
e
A
u
to
m
a
ti
c
O
ff
-
li
n
e
S
ig
n
a
tu
re
V
e
rif
ica
ti
o
n
P
ro
b
lem
Ba
se
o
n
G
e
o
m
e
tri
c
F
e
a
tu
re
s
"
,
A
z
a
d
Un
iv
e
rsity
,
Ira
n
.
[8
]
S
u
ra
b
h
i
G
a
rh
a
w
a
l
a
n
d
Ne
e
ra
j
S
h
u
k
la
(2
0
1
3
)
,
"
A
S
tu
d
y
o
n
Ha
n
d
w
rit
ten
S
ig
n
a
tu
re
V
e
rif
ica
ti
o
n
A
p
p
ro
a
c
h
e
s"
,
In
ter
n
a
t
io
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
Res
e
a
rc
h
in
Co
mp
u
ter
E
n
g
i
n
e
e
rin
g
&
T
e
c
h
n
o
lo
g
y
(
IJ
AR
CET
),
V
o
l
u
m
e
2
,
Iss
u
e
8
,
A
u
g
u
st 2
0
1
3
.
[9
]
L
B.
M
a
h
a
n
ta,
A
lp
a
n
a
De
k
a
(2
0
1
3
),
"
A
S
tu
d
y
o
n
Ha
n
d
w
rit
te
n
S
ig
n
a
tu
re
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
fo
r
Co
mp
u
ter
Ap
p
li
c
a
ti
o
n
s (
0
9
7
5
-
8
8
8
7
)
,
Vo
l
u
m
e
7
9
-
N
o
.
2
,
Oc
to
b
e
r
2
0
1
3
.
[1
0
]
P
ra
d
e
e
p
Ku
m
a
r,
S
h
e
k
h
a
r
S
in
g
h
,
A
sh
w
a
n
i
G
a
rg
a
n
d
Nish
a
n
t
P
ra
b
h
a
t
(2
0
1
3
),
"
Ha
n
d
W
rit
ten
S
ig
n
a
tu
re
Re
c
o
g
n
it
i
o
n
&
V
e
rif
ica
ti
o
n
u
si
n
g
Ne
u
ra
l
Ne
tw
o
rk
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
Res
e
a
rc
h
in
Co
mp
u
te
r
S
c
ien
c
e
a
n
d
S
o
ft
w
a
re
En
g
in
e
e
rin
g
,
Vo
lu
m
e
3
,
Iss
u
e
3
,
M
a
rc
h
2
0
1
3
[1
1
]
Ish
it
a
S
h
a
rm
a
,
S
a
k
sh
i
G
o
y
a
l
a
n
d
S
h
a
n
u
S
h
a
rm
a
,
"
S
ig
n
L
a
n
g
u
a
g
e
Re
c
o
g
n
it
io
n
S
y
ste
m
f
o
r
De
a
f
a
n
d
Du
m
b
P
e
o
p
le"
,
In
ter
n
a
t
io
n
a
l
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
rin
g
Res
e
a
rc
h
&
T
e
c
h
n
o
lo
g
y
(
IJ
ER
T
)
IS
S
N:
2
2
7
8
-
0
1
8
1
,
Vo
l
2
,
Iss
u
e
4
,
A
p
ril
-
2
0
1
3
,
p
p
.
3
8
2
-
3
8
7
.
[1
2
]
R.
P
lam
o
n
d
o
n
a
n
d
S
.
N.
S
r
ih
a
ri,
"
On
li
n
e
a
n
d
Of
f
li
n
e
Ha
n
d
w
rit
in
g
Re
c
o
g
n
it
io
n
:
A
Co
m
p
re
h
e
n
siv
e
S
u
rv
e
y
"
,
IEE
E
T
ra
n
.
on
P
a
tt
e
rn
A
n
a
lys
is
a
n
d
M
a
c
h
in
e
I
n
telli
g
e
n
c
e
,
v
o
l.
2
2
n
o
.
1
,
p
p
.
6
3
-
8
4
,
Ja
n
.
2
0
0
0
.
[1
3
]
M
.
Blu
m
e
n
ste
in
.
S
.
A
rm
a
n
d
.
a
n
d
M
u
t
h
u
k
k
u
m
a
ra
sa
m
y
,
“
Off
-
li
n
e
S
ig
n
a
tu
re
Ver
if
ica
ti
o
n
u
sin
g
th
e
En
h
a
n
c
e
d
M
o
d
if
ied
Dire
c
ti
o
n
Fea
t
u
re
a
n
d
Ne
u
ra
l
b
a
se
d
Cla
ss
if
ica
ti
o
n
,
”
I
n
t
e
rn
a
ti
o
n
a
l
J
o
in
t
C
o
n
fer
e
n
c
e
o
n
N
e
u
ra
l
Ne
two
rk
s,
2
0
0
6
.
[1
4
]
L
a
l
Ch
a
n
d
ra
,
P
u
ja
L
a
l,
Ra
ju
G
u
p
ta,
A
ru
n
T
a
y
a
l,
Din
e
sh
G
a
n
o
tra:
Im
p
ro
v
e
d
a
d
a
p
ti
v
e
b
in
a
riza
ti
o
n
tec
h
n
iq
u
e
f
o
r
d
o
c
u
m
e
n
t
im
a
g
e
a
n
a
l
y
sis.
V
IS
A
P
P
(1
)
2
0
0
7
:
3
1
7
-
3
2
1
.
[
1
5
]
V
e
d
P
ra
k
a
sh
Ag
n
ih
o
tri
,
“
Off
li
n
e
Ha
n
d
w
ri
tt
e
n
De
v
a
n
a
g
a
ri
S
c
rip
t
Re
c
o
g
n
it
io
n
”
,
I.
J
.
In
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
y
a
n
d
Co
mp
u
ter
S
c
ien
c
e
,
2
0
1
2
,
8
,
3
7
-
42
B
I
B
L
I
O
G
R
AP
H
Y
O
F
AUT
H
O
RS
M
r.
G
a
u
ta
m
S
.
P
ra
k
a
sh
h
a
s
d
o
n
e
h
is
B.
tec
h
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
f
o
rm
CS
E
De
p
a
rtme
n
t,
Am
it
y
S
c
h
o
o
l
o
f
En
g
in
e
e
rin
g
&
T
e
c
h
n
o
lo
g
y
,
Am
it
y
Un
iv
e
rsit
y
,
No
id
a
in
2
0
1
4
.
His
re
se
a
rc
h
a
re
a
in
c
lu
d
e
s Dig
it
a
l
Im
a
g
e
P
ro
c
e
ss
in
g
&
Co
m
p
u
ter Visi
o
n
.
M
s.
S
h
a
n
u
S
h
a
rm
a
re
c
e
i
v
e
d
h
e
r
M
.
T
e
c
h
De
g
re
e
in
In
telli
g
e
n
t
S
y
st
e
m
s
f
ro
m
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
De
p
a
rt
m
e
n
t,
In
d
ian
I
n
stit
u
te
o
f
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
A
ll
a
h
a
b
a
d
in
2
0
1
0
.
S
h
e
is
c
u
rre
n
tl
y
w
o
rk
in
g
a
s an
A
s
sista
n
t
P
ro
f
e
ss
o
r
in
C
S
E
De
p
a
rtm
e
n
t,
Am
it
y
Un
iv
e
rsit
y
,
No
id
a
,
Uttar
P
ra
d
e
sh
,
I
n
d
ia.
H
e
r
Re
se
a
rc
h
a
re
a
in
c
lu
d
e
s Dig
it
a
l
Im
a
g
e
p
ro
c
e
ss
in
g
a
n
d
Co
n
te
n
t
b
a
se
d
im
a
g
e
re
tri
e
v
a
l.
Evaluation Warning : The document was created with Spire.PDF for Python.