I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
6
,
Decem
b
er
20
25
,
p
p
.
5
8
0
4
~
5
8
1
4
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
6
.
pp
5
8
0
4
-
5
8
1
4
5804
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
H
y
brid
a
r
tif
icia
l i
ntelligence
a
ppro
a
ch t
o
count
erfe
i
t
curre
n
cy
detec
tion
M
o
nther
T
a
ra
wneh
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
m
a
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
ma
t
i
o
n
T
e
c
h
n
o
l
o
g
y
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
s
,
T
a
f
i
l
a
Te
c
h
n
i
c
a
l
U
n
i
v
e
r
si
t
y
,
Ta
f
i
l
a
,
Jo
r
d
a
n
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
1
5
,
2
0
2
5
R
ev
is
ed
J
u
l 2
9
,
2
0
2
5
Acc
ep
ted
Sep
1
6
,
2
0
2
5
Th
e
u
se
o
f
p
h
y
sic
a
l
m
o
n
e
y
c
o
n
t
in
u
e
s,
p
o
si
n
g
o
n
g
o
in
g
c
h
a
ll
e
n
g
e
s
in
th
e
fo
rm
o
f
c
o
u
n
terfe
it
m
o
n
e
y
.
T
h
is
p
r
o
b
lem
n
o
t
o
n
l
y
p
o
se
s
a
th
re
a
t
to
e
c
o
n
o
m
ic
sta
b
il
it
y
b
u
t
a
lso
u
n
d
e
rm
in
e
s
c
o
n
fid
e
n
c
e
in
t
h
e
fin
a
n
c
ial
sy
ste
m
s
in
u
se
.
Trad
it
io
n
a
l
m
e
th
o
d
s
su
c
h
a
s
m
a
n
u
a
l
in
sp
e
c
ti
o
n
s
a
n
d
tes
ti
n
g
o
f
se
c
u
rit
y
fe
a
tu
re
s
h
a
v
e
b
e
c
o
m
e
in
e
ffe
c
ti
v
e
in
d
e
tec
ti
n
g
a
d
v
a
n
c
e
d
c
o
u
n
terfe
it
in
g
tec
h
n
iq
u
e
s
o
n
a
n
o
n
g
o
in
g
b
a
sis.
Th
is
stu
d
y
p
r
o
p
o
se
s
a
h
y
b
rid
m
o
d
e
l
th
a
t
h
a
rn
e
ss
e
s
th
e
p
o
we
r
o
f
a
rti
fici
a
l
in
tell
ig
e
n
c
e
,
c
o
m
b
i
n
in
g
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s
(CNN
s),
lo
n
g
s
h
o
rt
-
term
m
e
m
o
ry
n
e
two
r
k
s
(L
S
TM
s),
a
n
d
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
i
n
e
s
(S
VMs)
fo
r
c
o
u
n
terfe
it
d
e
tec
ti
o
n
.
Th
e
p
ro
p
o
se
d
m
o
d
e
l
lev
e
ra
g
e
s
t
h
e
d
iv
e
rse
stre
n
g
t
h
s
o
f
a
n
u
m
b
e
r
o
f
a
rti
ficia
l
i
n
telli
g
e
n
c
e
tec
h
n
iq
u
e
s,
c
o
m
b
i
n
in
g
th
e
a
b
il
it
y
to
d
e
tec
t
c
o
u
n
terfe
it
in
g
,
a
n
a
l
y
z
e
v
isu
a
l
a
sp
e
c
ts,
a
n
d
se
q
u
e
n
c
e
s
o
f
b
a
n
k
n
o
tes
.
Th
e
p
r
o
p
o
se
d
m
o
d
e
l
wa
s
tes
ted
u
sin
g
re
a
l
Jo
rd
a
n
ian
c
u
rre
n
c
y
se
ts
o
f
d
iffere
n
t
d
e
n
o
m
i
n
a
ti
o
n
s
a
n
d
d
a
tas
e
ts
g
e
n
e
ra
ted
u
sin
g
g
e
n
e
ra
ti
v
e
a
d
v
e
rsa
rial
n
e
two
r
k
s
(G
AN
s).
T
h
e
re
su
lt
s
sh
o
we
d
t
h
a
t
th
e
m
o
d
e
l
wa
s
a
b
le
to
d
e
tec
t
c
o
u
n
terfe
it
i
n
g
wit
h
h
ig
h
a
c
c
u
ra
c
y
o
f
9
8
.
6
%
.
a
n
d
m
in
ima
l
e
rr
o
rs
c
o
m
p
a
re
d
to
o
th
e
r
m
e
th
o
d
s.
Th
is
o
u
tstan
d
i
n
g
p
e
rfo
rm
a
n
c
e
d
e
m
o
n
stra
tes
t
h
e
b
e
n
e
fit
s
o
f
in
teg
ra
ti
n
g
a
rti
ficia
l
i
n
telli
g
e
n
c
e
(
AI
)
tec
h
n
o
l
o
g
ies
a
n
d
t
h
a
t
th
e
re
i
s
ro
o
m
fo
r
d
e
v
e
l
o
p
m
e
n
t
a
n
d
so
lu
ti
o
n
s
t
h
a
t
c
a
n
k
e
e
p
u
p
wit
h
a
d
v
a
n
c
e
d
c
o
u
n
t
e
rfe
it
in
g
stra
teg
ies
.
Th
e
stu
d
y
d
e
m
o
n
stra
tes
th
e
imp
o
rtan
c
e
o
f
in
teg
ra
ti
n
g
A
I
in
m
a
in
tain
i
n
g
th
e
in
te
g
rit
y
o
f
p
h
y
sic
a
l
c
u
rre
n
c
y
tran
sa
c
ti
o
n
s
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
in
tellig
en
ce
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
C
o
u
n
ter
f
eit
d
etec
tio
n
Fra
u
d
d
etec
tio
n
Gen
er
ativ
e
ad
v
e
r
s
ar
ial
n
etwo
r
k
L
o
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
n
etwo
r
k
s
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mo
n
th
er
T
a
r
awn
eh
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atio
n
T
e
ch
n
o
lo
g
y
,
Facu
lty
o
f
I
n
f
o
r
m
at
io
n
T
ec
h
n
o
lo
g
y
an
d
C
o
m
m
u
n
i
ca
tio
n
s
,
T
af
ila
T
ec
h
n
ical
Un
iv
er
s
ity
T
af
ila,
J
o
r
d
an
E
m
ail:
s
m
tar
awn
eh
@
ttu
.
ed
u
.
jo
1.
I
NT
RO
D
UCT
I
O
N
C
o
u
n
ter
f
eit
m
o
n
e
y
is
o
n
e
o
f
th
e
m
o
s
t
s
er
io
u
s
p
r
o
b
lem
s
f
a
cin
g
th
e
wo
r
ld
.
I
t
p
o
s
es
a
th
r
ea
t
to
th
e
s
ec
u
r
ity
an
d
s
tab
ilit
y
o
f
f
in
a
n
cial
o
r
g
an
izatio
n
s
,
wh
ich
lea
d
s
to
m
is
tr
u
s
t
in
f
in
an
cial
in
s
ti
tu
tio
n
s
an
d
ca
u
s
es
ec
o
n
o
m
ic
d
is
tr
ess
.
C
o
u
n
ter
f
e
it
m
o
n
ey
is
a
cu
r
r
en
cy
th
at
h
as
b
ee
n
is
s
u
ed
o
r
m
in
ted
o
u
ts
id
e
th
e
leg
al
f
r
am
ewo
r
k
an
d
with
o
u
t
g
o
v
e
r
n
m
en
t
s
u
p
er
v
is
io
n
,
an
d
th
e
p
u
r
p
o
s
e
o
f
th
o
s
e
wh
o
wo
r
k
with
it
is
to
d
ec
eiv
e
p
eo
p
le
a
n
d
b
u
ild
wea
lth
in
a
n
illeg
al
way
[
1
]
.
Desp
ite
b
e
in
g
illeg
al,
it
h
as
c
o
n
tin
u
e
d
t
o
d
e
v
elo
p
with
th
e
d
ev
elo
p
m
e
n
t o
f
tec
h
n
o
lo
g
y
a
n
d
h
as b
ec
o
m
e
a
f
r
a
u
d
u
len
t
m
e
an
s
o
f
m
ak
i
n
g
wea
lth
ea
s
ily
.
W
h
ile
th
e
wo
r
ld
is
g
r
a
d
u
ally
s
h
if
tin
g
to
war
d
s
d
i
g
ital
cu
r
r
e
n
cies
[
2
]
,
th
is
m
a
y
tak
e
a
lo
n
g
tim
e
an
d
p
h
y
s
ical
cu
r
r
en
cy
will
s
till
b
e
u
s
ed
in
d
aily
tr
an
s
ac
tio
n
s
,
es
p
ec
ially
s
in
ce
m
o
s
t
o
f
th
e
ec
o
n
o
m
y
is
b
ased
o
n
ca
s
h
.
Du
e
t
o
th
e
r
is
in
g
co
s
t
o
f
liv
in
g
an
d
th
e
n
ee
d
f
o
r
ca
s
h
,
c
o
u
n
ter
f
eit
cu
r
r
en
cies
h
av
e
b
ec
o
m
e
a
n
in
cr
ea
s
in
g
ly
tem
p
tin
g
a
n
d
ea
s
y
tar
g
et
f
o
r
c
o
u
n
ter
f
eiter
s
[
3
]
.
C
o
u
n
ter
f
eit
cu
r
r
e
n
cies
ca
n
n
o
lo
n
g
er
b
e
d
etec
ted
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
a
r
tifi
cia
l in
tellig
en
ce
a
p
p
r
o
a
c
h
to
co
u
n
terf
eit
cu
r
r
en
cy
d
etec
tio
n
(
Mo
n
th
er Ta
r
a
w
n
eh
)
5805
u
s
in
g
tr
ad
itio
n
al
m
eth
o
d
s
s
u
ch
as
m
an
u
al
in
s
p
ec
tio
n
,
u
ltra
v
io
let
ex
a
m
in
atio
n
s
,
a
n
d
ch
ec
k
in
g
s
ec
u
r
ity
f
ea
tu
r
es,
th
e
y
ar
e
f
ailed
to
ca
t
ch
s
o
p
h
is
ticated
f
r
au
d
wh
ic
h
l
ea
d
s
to
u
n
ex
p
ec
ted
n
u
m
b
er
o
f
f
alse
n
e
g
ativ
e
[
4
]
,
also
in
s
u
f
f
icien
t to
k
ee
p
u
p
with
th
e
m
o
d
er
n
tech
n
o
lo
g
ies th
at
co
u
n
ter
f
eiter
s
ar
e
cu
r
r
e
n
tly
u
s
in
g
.
T
h
er
e
is
a
n
ee
d
f
o
r
m
o
r
e
ac
c
u
r
ate
an
d
r
eliab
le
co
u
n
ter
f
eit
d
etec
tio
n
s
y
s
tem
s
.
Ar
tific
ial
i
n
tellig
en
ce
o
f
f
er
s
s
o
lu
tio
n
s
to
v
ar
io
u
s
asp
ec
ts
o
f
life
f
r
o
m
cu
ltu
r
al,
ec
o
n
o
m
ic,
in
d
u
s
tr
ial,
m
ed
ical,
an
d
o
th
er
s
[
5
]
–
[
9
]
a
n
d
ca
n
b
e
a
r
a
d
ical
s
o
lu
tio
n
to
t
h
is
ch
allen
g
e.
AI
an
d
d
ee
p
lear
n
in
g
tech
n
iq
u
es
s
u
ch
as
c
o
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs),
d
ee
p
n
e
u
r
al
n
etwo
r
k
s
(
DNNs)
,
an
d
g
en
er
ativ
e
ad
v
er
s
ar
ial
n
et
wo
r
k
s
(
GANs)
ca
n
d
is
tin
g
u
is
h
co
m
p
lex
p
atter
n
s
an
d
c
h
ar
ac
ter
is
tics
in
cu
r
r
e
n
cy
an
d
th
u
s
d
etec
t
c
o
u
n
te
r
f
eits
with
g
r
ea
ter
ac
cu
r
ac
y
an
d
ef
f
icien
c
y
th
a
n
t
r
ad
itio
n
al
m
eth
o
d
s
.
I
n
a
d
d
itio
n
,
AI
to
o
ls
h
a
v
e
th
e
ab
ilit
y
t
o
lear
n
o
v
e
r
tim
e
n
ew
f
r
au
d
tech
n
i
q
u
es
[
1
0
]
.
T
h
ese
to
o
ls
ar
e
d
esig
n
ed
b
ased
o
f
th
e
id
ea
o
f
s
elf
-
lear
n
in
g
o
v
e
r
tim
e
to
im
p
r
o
v
e
th
eir
ac
cu
r
ac
y
as
n
ew
d
ata
ap
p
ea
r
.
T
h
is
s
elf
-
lear
n
in
g
ab
ilit
y
m
ak
es
th
em
p
o
wer
f
u
l
ag
ain
s
t
n
ew
co
u
n
ter
f
eitin
g
tech
n
iq
u
es
with
o
u
t
th
e
n
ee
d
f
o
r
r
e
p
r
o
g
r
am
in
g
o
r
m
a
n
u
a
l
u
p
d
ate.
W
h
ile
AI
p
r
e
d
ictio
n
s
d
ep
en
d
o
n
d
ata
q
u
ality
,
th
er
e
is
in
cr
ea
s
in
g
co
n
f
id
en
ce
i
n
AI
'
s
ab
ilit
y
to
en
h
an
ce
ac
cu
r
ac
y
in
d
etec
tin
g
c
o
u
n
ter
f
eit
c
u
r
r
en
c
y
.
As
a
r
esu
lt,
th
er
e
is
in
v
estme
n
t
b
y
f
in
an
cial
f
ir
m
s
in
th
ese
t
o
o
ls
to
s
tr
en
g
th
e
n
th
eir
d
etec
t
io
n
s
y
s
tem
s
ag
ain
s
t
s
m
ar
t f
r
au
d
m
eth
o
d
s
[
1
1
]
.
Mo
s
t
r
ec
en
t
r
esear
ch
h
as
u
s
ed
AI
tech
n
i
q
u
es
in
d
ep
e
n
d
en
t
ly
f
o
r
f
o
r
g
er
y
d
etec
tio
n
.
Fo
r
ex
am
p
le,
C
NNs
ar
e
u
s
ed
to
ex
tr
ac
t
v
i
s
u
al
f
ea
tu
r
es,
w
h
ile
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
n
etwo
r
k
s
(
L
STM
s
)
ar
e
u
s
ed
to
an
aly
ze
s
eq
u
en
ce
s
.
GANs
ar
e
also
u
s
ed
to
au
g
m
en
t
d
ata
v
o
lu
m
e.
W
h
ile
we
ag
r
ee
th
at
th
ese
m
eth
o
d
s
ar
e
ef
f
ec
tiv
e,
th
ey
h
av
e
s
o
m
e
li
m
itatio
n
s
wh
en
u
s
ed
alo
n
e:
C
NNs
m
ay
f
ac
e
ch
allen
g
es
in
h
an
d
lin
g
u
n
s
ee
n
f
o
r
g
er
ies,
in
c
o
m
p
lete
s
eq
u
e
n
c
es
m
ay
m
is
lead
L
STM
s
,
an
d
GAN
-
b
ased
tr
ain
in
g
m
a
y
b
e
in
s
u
f
f
icien
t
if
th
e
d
ataset
is
b
iased
to
war
d
a
p
a
r
ticu
lar
class
.
Few
r
esear
ch
er
s
co
m
b
in
e
th
ese
tech
n
i
q
u
es
in
to
a
s
in
g
le
f
o
r
g
er
y
d
etec
tio
n
m
o
d
el.
T
h
er
e
f
o
r
e,
a
h
y
b
r
id
m
o
d
el
is
p
r
o
p
o
s
ed
to
f
i
ll th
is
g
ap
.
T
h
is
s
tu
d
y
p
r
o
p
o
s
es
a
h
y
b
r
id
d
ee
p
lear
n
i
n
g
m
o
d
el
f
o
r
ac
c
u
r
ate
co
u
n
ter
f
eit
cu
r
r
en
cy
d
ete
ctio
n
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
c
o
m
b
in
es
t
h
e
ad
v
an
tag
es o
f
c
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs),
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
n
etwo
r
k
s
(
L
STM
s
)
,
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
(
SVMs),
au
g
m
en
ted
b
y
g
en
e
r
ativ
e
ad
v
er
s
ar
ial
n
etwo
r
k
s
(
GANs)
.
T
h
is
m
o
d
el
aim
s
t
o
in
teg
r
ate
s
p
atial
an
d
s
eq
u
en
tial
lear
n
in
g
ca
p
ab
ilit
ies
with
s
y
n
th
etic
d
ata
g
en
er
atio
n
t
o
im
p
r
o
v
e
d
etec
tio
n
p
er
f
o
r
m
a
n
ce
.
T
h
is
r
esear
ch
aim
s
to
s
tu
d
y
a
n
d
ev
al
u
ate
ar
tific
ial
in
tellig
en
ce
tech
n
iq
u
es
f
o
r
co
u
n
ter
f
eit
c
u
r
r
en
cy
d
etec
tio
n
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
ad
d
r
ess
es
cu
r
r
en
t
g
ap
s
in
co
u
n
ter
f
eit
cu
r
r
en
c
y
d
etec
tio
n
b
y
im
p
r
o
v
in
g
g
en
er
aliza
tio
n
an
d
r
o
b
u
s
tn
ess
u
n
d
er
r
ea
lis
tic
co
n
d
iti
o
n
s
.
T
h
e
s
tu
d
y
also
ex
p
lo
r
es
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
o
f
th
e
m
o
d
el,
s
u
ch
as
b
an
k
s
an
d
AT
Ms,
an
d
d
is
cu
s
s
es
its
ab
ilit
y
to
d
etec
t
co
u
n
ter
f
eit
cu
r
r
en
cy
in
s
tan
tly
.
T
h
e
in
teg
r
atio
n
o
f
a
r
tific
ial
in
tellig
en
ce
en
h
a
n
ce
s
ac
cu
r
ate
c
o
u
n
ter
f
eit
cu
r
r
e
n
cy
d
etec
tio
n
an
d
d
ev
elo
p
s
a
m
o
r
e
co
n
s
is
ten
t
to
o
l
f
o
r
p
r
e
v
en
ti
n
g
f
r
au
d
an
d
s
ec
u
r
in
g
p
h
y
s
ic
al
ca
s
h
tr
an
s
ac
tio
n
s
,
th
er
eb
y
p
r
o
tectin
g
th
e
ec
o
n
o
m
y
.
T
h
e
r
est
o
f
th
e
p
ap
e
r
is
o
r
g
an
ized
in
to
liter
atu
r
e
r
ev
iew,
m
eth
o
d
o
l
o
g
y
,
r
esu
lt
an
d
d
is
cu
s
s
io
n
,
an
d
co
n
cl
u
s
io
n
at
th
e
en
d
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
C
o
u
n
ter
f
eitin
g
is
n
o
t
a
n
ew
is
s
u
e
in
th
e
ec
o
n
o
m
ic
wo
r
l
d
.
I
t
h
as
b
ee
n
ar
o
u
n
d
s
in
ce
th
e
b
e
g
in
n
in
g
o
f
cu
r
r
en
c
y
u
s
e
a
n
d
is
m
o
n
ito
r
e
d
an
d
d
ea
lt
with
b
y
tr
ad
itio
n
a
l
m
eth
o
d
s
s
u
c
h
as
m
an
u
al
in
s
p
ec
tio
n
,
u
ltra
v
io
let
s
ca
n
n
in
g
,
an
d
th
e
u
s
e
o
f
s
ec
u
r
ity
f
ea
tu
r
es
s
u
ch
as
wate
r
m
ar
k
s
an
d
h
o
lo
g
r
am
s
.
T
h
ese
m
eth
o
d
s
h
av
e
b
ee
n
ef
f
ec
tiv
e
in
d
etec
tin
g
c
o
u
n
t
er
f
eit
cu
r
r
en
cy
an
d
h
av
e
b
ee
n
wid
ely
ad
o
p
ted
[
1
2
]
.
Ho
wev
er
,
s
ev
e
r
al
d
is
ad
v
an
tag
es
h
av
e
em
er
g
e
d
t
h
at
h
av
e
r
ed
u
ce
d
th
eir
u
s
e,
s
u
ch
as
th
e
lev
el
o
f
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
.
Ma
n
u
al
in
s
p
ec
tio
n
s
tak
e
a
lo
n
g
tim
e,
wh
ile
r
ely
in
g
s
o
lely
o
n
th
e
ex
p
er
ien
ce
an
d
s
k
ill
o
f
th
e
au
d
it.
Ad
d
itio
n
ally
,
s
k
illed
co
u
n
ter
f
eiter
s
h
a
v
e
s
u
r
p
ass
ed
th
ese
m
eth
o
d
s
,
an
d
th
e
y
h
av
e
b
ec
o
m
e
n
o
o
b
s
tacle
to
th
em
.
E
ar
ly
n
eu
r
al
n
etwo
r
k
s
wer
e
s
i
m
p
le
in
th
eir
lear
n
in
g
ab
ilit
ies
d
u
e
to
th
e
lack
o
f
co
m
p
u
tatio
n
al
p
o
wer
an
d
d
ata
av
ailab
ilit
y
[
1
3
]
.
T
h
e
r
ev
o
lu
tio
n
i
n
co
m
p
u
tatio
n
al
p
o
wer
an
d
th
e
a
v
ailab
ilit
y
o
f
b
i
g
d
ata
h
av
e
o
p
en
ed
th
e
d
o
o
r
to
AI
ad
v
an
ce
m
e
n
ts
,
p
r
o
p
ellin
g
t
h
e
f
ield
in
to
a
n
ew
e
r
a
o
f
in
n
o
v
ati
o
n
a
n
d
ca
p
a
b
ilit
y
.
T
h
ese
d
ev
elo
p
m
e
n
ts
h
av
e
s
ig
n
if
ica
n
tly
ac
ce
ler
ated
AI
c
o
m
p
u
ta
tio
n
s
,
en
ab
lin
g
m
o
r
e
s
o
p
h
is
ticated
m
o
d
els
an
d
ap
p
licatio
n
s
[
1
4
]
.
Ad
v
an
ce
d
tech
n
o
lo
g
ies
lu
n
ch
t
h
e
p
o
wer
o
f
ar
tific
ial
in
tellig
en
ce
a
n
d
d
ee
p
lear
n
in
g
to
p
r
o
ce
s
s
v
ast
am
o
u
n
ts
o
f
d
at
a
an
d
lea
r
n
c
o
m
p
lex
p
atter
n
s
.
Ar
tific
ial
in
tellig
en
ce
in
tr
o
d
u
ce
s
in
n
o
v
ativ
e
ap
p
r
o
ac
h
es
in
v
ar
io
u
s
in
d
u
s
tr
ies
s
u
ch
as
h
ea
lth
ca
r
e
[
1
5
]
,
[
1
6
]
,
f
in
an
ce
[
1
7
]
,
[
1
8
]
,
ag
r
ic
u
ltu
r
e
[
1
9
]
,
s
ec
u
r
ity
an
d
au
to
n
o
m
o
u
s
s
y
s
tem
s
[
5
]
.
As
tech
n
o
lo
g
y
b
ec
o
m
es
m
o
r
e
p
o
wer
f
u
l,
a
n
d
AI
m
et
h
o
d
s
b
ec
o
m
e
m
o
r
e
ef
f
icien
t,
d
ee
p
lear
n
in
g
h
as
i
n
tr
o
d
u
ce
d
in
n
o
v
ativ
e
m
eth
o
d
s
f
o
r
d
etec
tin
g
co
u
n
ter
f
eit
cu
r
r
en
cy
[
2
0
]
.
T
h
ese
m
eth
o
d
s
im
p
r
o
v
e
d
etec
tio
n
ac
cu
r
ac
y
b
y
an
aly
zi
n
g
a
g
r
ea
ter
n
u
m
b
er
o
f
f
ea
t
u
r
es,
e
v
en
s
p
ec
tr
al
s
ig
n
atu
r
es
a
n
d
co
m
p
lex
p
atter
n
s
th
at
ca
n
n
o
t
b
e
s
ee
n
with
th
e
ey
es.
T
h
e
u
s
e
o
f
d
ee
p
lear
n
in
g
h
as
tr
an
s
f
o
r
m
ed
th
e
d
etec
tio
n
p
r
o
ce
s
s
to
ad
ap
tab
le,
f
ast an
d
ac
cu
r
ate
m
eth
o
d
o
lo
g
ies.
Deep
lear
n
in
g
is
a
s
u
b
s
ec
tio
n
o
f
m
ac
h
in
e
lear
n
in
g
.
Dee
p
lea
r
n
in
g
u
s
es
o
f
d
ee
p
n
eu
r
al
n
et
wo
r
k
s
th
at
co
n
s
is
t
o
f
m
u
ltip
le
lay
er
s
[
2
1
]
.
A
d
ee
p
n
eu
r
al
n
etwo
r
k
h
as
a
n
in
p
u
t
la
y
er
,
o
n
e
o
r
m
o
r
e
h
i
d
d
en
lay
er
s
,
a
n
d
a
n
o
u
tp
u
t
lay
er
.
E
ac
h
lay
er
c
o
n
ta
in
s
n
o
d
es
(
n
eu
r
o
n
s
)
th
at
ar
e
f
u
lly
co
n
n
ec
te
d
to
all
n
o
d
es
in
th
e
ad
jace
n
t
lay
er
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
0
4
-
5
8
1
4
5806
T
h
is
s
tr
u
ctu
r
e
h
elp
n
etwo
r
k
t
o
lear
n
co
m
p
lex
p
atter
n
s
th
r
o
u
g
h
s
eq
u
en
tial
la
y
er
s
,
p
r
ed
i
ct
o
u
tp
u
t
v
ar
ia
b
les
b
ased
o
n
th
e
lear
n
ed
r
ep
r
esen
t
atio
n
[
2
2
]
.
T
h
e
DNN
m
o
d
el
a
n
d
its
ap
p
licatio
n
s
ar
e
s
h
o
w
n
i
n
Fig
u
r
e
1
.
Fig
u
r
e
1
.
DNN
ar
ch
itectu
r
e
an
d
its
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
[
2
3
]
T
h
e
i
n
i
t
i
al
a
p
p
l
i
c
at
i
o
n
o
f
t
e
c
h
n
o
l
o
g
y
i
n
c
o
u
n
t
e
r
f
e
i
t
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
wa
s
r
o
o
t
e
d
i
n
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
a
n
d
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
t
e
c
h
n
iq
u
e
s
.
T
h
e
s
e
m
et
h
o
d
s
i
n
v
o
l
v
ed
s
c
a
n
n
i
n
g
c
u
r
r
e
n
c
y
n
o
t
e
s
t
o
i
d
e
n
t
i
f
y
m
i
s
s
i
n
g
o
r
a
l
t
e
r
e
d
f
e
a
t
u
r
es
,
s
u
c
h
as
c
o
l
o
r
,
s
i
z
e
,
li
n
e
s
,
a
n
d
te
x
t
u
r
e
s
.
B
y
an
a
l
y
z
i
n
g
t
h
e
s
e
p
h
y
s
i
c
a
l
at
t
r
i
b
u
t
e
s
[
2
4
]
,
[
2
5
]
,
t
h
e
s
e
s
y
s
te
m
s
a
i
m
e
d
t
o
d
e
t
e
ct
a
n
y
d
i
s
c
r
e
p
a
n
c
i
es
t
h
a
t
c
o
u
l
d
i
n
d
i
ca
t
e
a
c
o
u
n
t
e
r
f
e
i
t
n
o
t
e
.
A
l
t
h
o
u
g
h
e
f
f
e
c
t
i
v
e
t
o
s
o
m
e
e
x
t
e
n
t
,
t
h
es
e
m
e
t
h
o
d
s
w
e
r
e
l
im
i
t
e
d
i
n
t
h
e
i
r
a
b
i
l
it
y
t
o
a
d
a
p
t
t
o
m
o
r
e
s
o
p
h
is
t
ic
a
t
e
d
c
o
u
n
t
e
r
f
e
i
t
i
n
g
te
c
h
n
i
q
u
e
s
as
t
h
e
y
r
e
l
i
e
d
o
n
p
r
e
d
e
f
i
n
e
d
p
at
t
e
r
n
s
a
n
d
m
a
n
u
a
l
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
.
F
u
r
t
h
e
r
m
o
r
e
,
r
e
s
e
a
r
ch
e
r
s
m
e
r
g
e
d
t
h
e
s
e
t
e
c
h
n
i
q
u
e
s
wi
t
h
m
a
c
h
i
n
e
l
ea
r
n
i
n
g
[
2
6
]
,
[
2
7
]
t
o
i
m
p
r
o
v
e
t
h
e
i
r
p
e
r
f
o
r
m
a
n
c
e
a
n
d
i
n
c
r
e
a
s
e
a
c
c
u
r
a
c
y
.
B
y
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
,
t
h
e
s
y
s
t
e
m
s
c
o
u
l
d
l
e
a
r
n
f
r
o
m
d
a
t
a
a
n
d
a
u
t
o
m
a
t
i
c
a
l
l
y
d
e
t
e
c
t
m
o
r
e
c
o
m
p
l
e
x
p
a
t
t
e
r
n
s
,
r
e
d
u
c
i
n
g
h
u
m
a
n
e
r
r
o
r
a
n
d
i
n
c
r
e
a
s
i
n
g
t
h
e
d
et
e
c
ti
o
n
p
r
o
c
e
s
s
i
n
r
e
a
l
-
ti
m
e
.
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs)
ar
e
m
o
s
t
ef
f
ec
tiv
e
in
im
ag
e
r
ec
o
g
n
itio
n
d
u
e
to
its
ab
ilit
y
to
an
aly
ze
cu
r
r
en
cy
f
ea
tu
r
es.
I
t
is
a
m
u
ltil
ay
er
d
ee
p
lear
n
in
g
n
eu
r
al
n
etwo
r
k
;
th
ey
h
av
e
m
u
ltip
le
lay
er
s
th
at
ex
tr
ac
t
f
ea
tu
r
es
an
d
t
r
ain
cla
s
s
if
ier
s
.
C
NNs
u
tili
ze
h
ier
ar
c
h
ical
r
ep
r
esen
tatio
n
s
to
ex
tr
a
ct
k
ey
to
p
o
l
o
g
ical
f
ea
tu
r
es
an
d
f
ea
t
u
r
es
f
r
o
m
cu
r
r
en
cy
im
ag
es.
C
NNs
m
an
ip
u
l
ate
p
ix
els
th
r
o
u
g
h
v
ar
i
o
u
s
lay
er
s
,
th
is
will
ca
p
tu
r
e
f
ea
tu
r
es
f
r
o
m
co
m
p
lex
an
d
h
i
g
h
-
d
im
e
n
s
io
n
al
d
ata.
I
t
h
as
b
ee
n
im
p
lem
en
ted
b
y
r
esear
ch
er
s
to
d
etec
t
f
ak
e
in
d
if
f
er
en
t
cu
r
r
e
n
cy
s
u
ch
as
I
n
d
ian
[
2
8
]
,
[
2
9
]
,
E
u
r
o
[
3
0
]
,
J
o
r
d
an
ian
[
3
1
]
,
B
an
g
lad
esh
[
2
5
]
an
d
o
t
h
er
c
u
r
r
e
n
cy
[
3
2
]
–
[
3
4
]
.
C
NN
h
as
b
ee
n
co
m
b
in
ed
with
o
th
er
m
o
d
els.
R
esear
ch
er
s
i
m
p
lem
en
t
a
s
y
s
tem
th
at
co
m
b
in
e
C
NN
with
L
STM
n
etwo
r
k
s
th
at
ad
a
p
t
to
n
ew
co
u
n
ter
f
eitin
g
m
eth
o
d
s
as
th
ey
ap
p
ea
r
[
3
]
,
[
3
5
]
,
[
3
6
]
.
Ho
wev
er
,
th
is
in
teg
r
atio
n
h
as
s
o
m
e
lim
itatio
n
in
ter
m
o
f
co
m
p
lex
ity
,
tim
e
,
an
d
n
ee
d
m
o
r
e
tr
ain
in
g
.
Als
o
,
C
NN
co
m
b
in
e
d
with
SVM,
wh
er
e
C
NN
u
s
ed
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
f
r
o
m
cu
r
r
en
cy
im
ag
es f
o
llo
we
d
b
y
class
if
icatio
n
s
b
y
SVM
[
3
7
]
.
T
h
e
s
tu
d
y
s
h
o
ws
g
o
o
d
i
m
p
r
o
v
e
m
en
t
in
d
etec
tin
g
f
a
k
e
cu
r
r
en
c
y
an
d
p
r
o
v
es
th
at
th
e
ac
cu
r
ac
y
in
c
r
ea
s
ed
b
y
in
teg
r
atin
g
C
NN
with
o
th
er
m
o
d
els.
Gen
er
ativ
e
ad
v
e
r
s
ar
ial
n
etwo
r
k
s
(
GANs)
u
s
ed
to
g
e
n
er
ate
ar
tific
ial
im
ag
es
f
o
r
cu
r
r
en
c
y
,
w
h
ich
u
s
ed
to
tr
ain
d
ee
p
lear
n
in
g
m
o
d
els
to
im
p
r
o
v
e
f
a
k
e
d
etec
tio
n
[
3
8
]
.
I
t
allo
ws
AI
to
o
ls
to
lear
n
an
d
c
o
m
p
r
eh
en
d
d
if
f
er
en
ce
s
b
etwe
en
r
ea
l
a
n
d
f
ak
e
cu
r
r
e
n
cy
.
Als
o
,
h
elp
to
d
ig
est
ch
allen
g
es
o
f
lim
ited
av
ailab
ilit
y
o
f
co
u
n
ter
f
eit
s
am
p
les.
Ho
wev
er
,
th
is
m
ay
in
cr
ea
s
e
th
e
co
u
n
ter
f
eit
p
r
o
b
lem
an
d
p
o
s
e
eth
ical
an
d
leg
al
is
s
u
e
s
.
Ad
d
itio
n
ally
,
th
e
g
en
er
ate
d
im
ag
es c
o
u
ld
n
o
t b
e
th
at
r
ea
lis
tic
an
d
m
is
lead
th
e
r
esu
lts
ac
cu
r
ac
y
.
B
lo
ck
ch
ain
tech
n
o
lo
g
y
h
as
b
e
en
u
s
ed
to
d
etec
t
co
u
n
ter
f
eit
c
u
r
r
en
cies,
as
its
c
o
r
e
p
r
o
p
er
tie
s
s
u
ch
as
d
ec
en
tr
aliza
tio
n
an
d
im
m
u
tab
ilit
y
ca
n
b
e
u
s
ed
to
co
m
b
at
co
u
n
ter
f
eitin
g
an
d
in
cr
ea
s
e
c
o
n
f
id
en
ce
in
th
eir
v
er
if
icatio
n
.
B
lo
ck
ch
ai
n
tech
n
o
lo
g
y
ca
n
b
e
e
m
p
lo
y
e
d
to
r
ec
o
r
d
,
tr
ac
k
,
a
n
d
v
e
r
if
y
th
e
s
er
ial
n
u
m
b
er
o
f
cu
r
r
en
cies
in
r
ea
l
tim
e,
p
r
ev
e
n
tin
g
u
n
r
eg
is
ter
ed
f
u
n
d
s
f
r
o
m
en
ter
in
g
th
e
f
in
an
cial
s
y
s
tem
[
1
]
,
[
2
]
.
I
n
ad
d
itio
n
to
s
er
ial
n
u
m
b
er
s
,
cr
y
p
to
g
r
a
p
h
ic
tag
s
ca
n
b
e
p
lace
d
an
d
s
ec
u
r
e
d
atab
ases
ca
n
b
e
u
s
ed
to
au
to
m
ate
th
e
v
er
if
icatio
n
o
f
cu
r
r
en
cies
b
e
f
o
r
e
th
e
y
en
ter
th
e
f
in
a
n
cial
s
y
s
tem
u
s
in
g
b
lo
ck
c
h
ain
-
b
a
s
ed
s
m
ar
t
co
n
tr
ac
t
tech
n
o
lo
g
y
[
3
]
.
T
o
r
ed
u
ce
th
e
r
is
k
o
f
cu
r
r
en
c
y
co
u
n
ter
f
eit
in
g
,
th
e
co
n
ce
p
t
o
f
u
s
in
g
d
ig
ital
cu
r
r
en
cies
an
d
m
ak
in
g
th
em
a
n
alter
n
ativ
e
to
p
h
y
s
ical
cu
r
r
e
n
cies
em
er
g
e
d
,
an
d
b
lo
c
k
ch
ain
a
p
p
licatio
n
s
wer
e
em
p
lo
y
ed
to
cr
ea
te
d
ig
ital c
u
r
r
en
cies.
B
lo
ck
ch
ain
was e
n
h
an
ce
d
b
y
in
teg
r
atin
g
it with
ar
tific
ial
in
tellig
en
ce
to
an
aly
ze
d
at
a
an
d
p
r
e
d
ict
an
y
illeg
al
ac
tiv
ity
in
th
e
f
ield
o
f
cu
r
r
en
c
y
tr
ad
in
g
[
3
]
–
[
5
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
a
r
tifi
cia
l in
tellig
en
ce
a
p
p
r
o
a
c
h
to
co
u
n
terf
eit
cu
r
r
en
cy
d
etec
tio
n
(
Mo
n
th
er Ta
r
a
w
n
eh
)
5807
T
h
er
e
ar
e
s
tu
d
ies
th
at
h
av
e
u
s
ed
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
s
to
d
etec
t
co
u
n
ter
f
eitin
g
b
y
tr
ac
k
in
g
th
e
s
er
ial
n
u
m
b
er
s
o
n
cir
cu
latin
g
cu
r
r
en
c
y
[
3
9
]
.
T
h
ey
id
en
tif
y
a
n
y
ir
r
eg
u
la
r
ities
in
th
e
s
eq
u
en
ce
o
f
s
er
ial
n
u
m
b
er
s
an
d
h
av
e
p
r
o
v
en
ef
f
ec
tiv
e
in
id
en
tify
i
n
g
p
o
ten
tial
co
u
n
te
r
f
eitin
g
ac
tiv
ities
an
d
en
h
an
ci
n
g
th
e
s
ec
u
r
ity
o
f
cu
r
r
en
c
y
cir
cu
latio
n
.
T
o
p
o
wer
b
o
th
th
e
im
ag
e
r
ec
o
g
n
i
tio
n
ab
ilit
ies
o
f
C
NN
s
an
d
th
e
s
eq
u
en
tial
d
ata
p
r
o
ce
s
s
in
g
p
o
wer
s
o
f
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
s
(
R
NNs
)
,
a
h
y
b
r
i
d
m
o
d
el
h
as
b
ee
n
im
p
lem
en
ted
th
at
in
teg
r
ates CNN w
ith
R
N
N,
p
r
o
v
id
in
g
a
m
o
r
e
co
m
p
r
eh
en
s
iv
e
s
o
lu
tio
n
f
o
r
co
u
n
ter
f
eit
d
etec
tio
n
[
4
0
]
,
[
4
1
]
.
AI
is
a
cr
u
cial
elem
en
t
in
im
p
r
o
v
in
g
f
in
an
cial
s
ec
u
r
ity
b
y
p
r
o
v
id
in
g
to
o
ls
th
at
ca
n
d
etec
t
f
r
au
d
u
le
n
t
ac
tiv
ities
in
r
ea
l
-
tim
e.
AI
-
b
a
s
ed
s
y
s
tem
s
ar
e
u
s
ed
in
b
an
k
s
an
d
f
in
a
n
cial
s
er
v
ices
to
s
ec
u
r
e
tr
an
s
ac
tio
n
s
,
au
th
en
ticate
cu
r
r
en
c
y
,
an
d
s
to
p
th
e
cir
cu
latio
n
o
f
f
a
k
e
m
o
n
e
y
.
T
h
er
e
ar
e
in
cr
ea
s
in
g
n
u
m
b
e
r
o
f
s
tu
d
ies th
at
u
s
e
AI
in
f
in
an
ce
an
d
f
o
c
u
s
o
n
f
r
au
d
d
etec
tio
n
,
esp
ec
ially
co
u
n
ter
f
eit
cu
r
r
en
c
y
d
etec
tio
n
.
Dee
p
lear
n
in
g
ap
p
r
o
ac
h
es
p
r
o
v
id
e
a
s
tr
o
n
g
s
o
lu
tio
n
to
th
e
o
n
g
o
in
g
co
u
n
te
r
f
eiter
s
m
eth
o
d
s
.
Ho
we
v
er
,
o
n
g
o
in
g
ac
tiv
ities
o
f
co
u
n
ter
f
eits
tactics
r
eq
u
ir
e
c
o
n
tin
u
o
u
s
ef
f
o
r
t
to
d
ev
elo
p
n
ew
f
r
am
ewo
r
k
s
to
p
r
ev
e
n
t
f
r
au
d
a
n
d
in
cr
ea
s
e
ac
cu
r
ac
y
in
d
etec
tin
g
f
a
k
e
cu
r
r
en
cy
.
3.
M
E
T
H
O
D
T
o
d
ev
elo
p
a
m
o
d
el
to
d
etec
t
co
u
n
ter
f
eit
cu
r
r
en
c
y
r
eg
ar
d
less
o
f
its
ty
p
e,
it
is
n
ec
ess
ar
y
to
co
m
b
in
e
th
e
ca
p
ab
ilit
ies
o
f
d
if
f
er
en
t
AI
alg
o
r
ith
m
s
in
to
a
s
in
g
le
m
o
d
el.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
is
a
h
y
b
r
id
m
o
d
el
th
at
co
m
b
in
es
C
NNs,
L
STM
s
,
G
ANs,
an
d
SVMs.
Fig
u
r
e
2
s
h
o
ws
th
e
s
tep
s
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el.
T
h
is
m
o
d
el
im
p
r
o
v
es a
cc
u
r
ac
y
b
y
p
r
o
ce
s
s
in
g
all
asp
ec
ts
o
f
th
e
o
r
ig
in
al
a
n
d
s
eq
u
en
tial im
ag
es to
d
etec
t
co
u
n
ter
f
eitin
g
.
T
o
u
n
d
er
s
tan
d
th
e
m
u
ltip
le
asp
ec
ts
o
f
co
u
n
ter
f
eit
d
etec
tio
n
,
th
e
n
ew
m
o
d
el
was
ch
o
s
en
as
a
co
m
b
in
atio
n
o
f
C
NN,
L
STM
,
GAN,
an
d
SVM.
C
NNs
ar
e
b
etter
s
u
ited
f
o
r
an
al
y
zin
g
b
an
k
n
o
te
im
ag
es
d
u
e
t
o
th
eir
ab
ilit
y
to
lear
n
s
p
atial
p
atter
n
s
an
d
im
ag
e
f
ea
tu
r
es.
L
STM
s
h
av
e
p
r
o
v
en
th
ei
r
ab
ilit
y
to
m
o
d
el
s
eq
u
en
tial
d
ep
en
d
e
n
cies
an
d
i
d
en
tify
ir
r
e
g
u
lar
p
atter
n
s
in
n
u
m
b
e
r
s
eq
u
en
ce
s
,
m
ak
in
g
th
em
s
u
itab
le
f
o
r
p
r
o
ce
s
s
in
g
s
er
ial
n
u
m
b
er
s
.
Du
e
to
th
e
lac
k
o
f
leg
al
ten
d
er
f
o
r
c
o
u
n
ter
f
eit
cu
r
r
en
c
y
,
GANs
wer
e
u
s
ed
to
g
en
er
ate
r
ea
lis
tic
im
ag
es
o
f
co
u
n
ter
f
eit
cu
r
r
e
n
c
y
f
o
r
s
u
f
f
icien
t
tr
ain
i
n
g
.
SVMs,
o
n
th
e
o
th
er
h
a
n
d
,
h
a
v
e
th
e
p
o
ten
tial
to
h
an
d
le
h
ig
h
-
d
im
e
n
s
io
n
al
d
ata
a
n
d
p
r
o
v
id
e
r
o
b
u
s
t d
ec
is
io
n
b
o
u
n
d
ar
i
es.
Fig
u
r
e
2
s
h
o
ws
th
e
m
ain
co
m
p
o
n
e
n
ts
o
f
th
e
m
o
d
el
an
d
s
tep
b
y
s
tep
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
els.
I
t
in
clu
d
es:
a.
Data
s
et:
a
co
llect
io
n
o
f
im
ag
e
im
ag
es
o
f
J
o
r
d
an
ian
b
a
n
k
n
o
tes
(
1
,
5
,
1
0
,
2
0
,
5
0
)
f
r
o
m
Ka
g
g
le
.
T
h
e
d
ataset
co
n
tain
s
7
3
1
2
im
ag
es (
5
4
7
3
r
e
al
+
1
8
3
9
GAN
-
g
e
n
er
ated
)
.
b.
I
m
ag
e
p
r
e
p
r
o
ce
s
s
in
g
:
n
u
m
b
e
r
o
f
o
p
e
r
atio
n
s
ap
p
lied
o
n
t
h
e
im
ag
es
b
ef
o
r
e
p
r
o
ce
s
s
in
g
it:
r
esized
to
224
×
2
2
4
p
i
x
els,
co
n
tr
ast ad
ju
s
tm
en
t,
r
o
tatio
n
,
f
lip
p
in
g
,
s
ca
lin
g
,
an
d
cr
o
p
p
in
g
.
c.
Ser
ial
Nu
m
b
er
: e
x
tr
ac
t a
10
-
d
ig
it
s
eq
u
en
ce
n
u
m
b
er
s
as in
p
u
t
f
o
r
L
STM
.
d.
GNA:
to
g
et
g
o
o
d
tr
ain
in
g
a
n
d
h
a
v
e
d
iv
e
r
s
ity
in
th
e
d
ata
s
et,
DC
GAN
u
s
ed
to
g
en
er
ate
r
ea
lis
tic
f
ak
e
cu
r
r
en
c
y
im
ag
es.
e.
Featu
r
e
ex
tr
ac
tio
n
:
C
NN
u
s
ed
th
r
ee
co
n
v
o
lu
tio
n
al
b
l
o
ck
s
(
3
2
,
6
4
,
1
2
8
f
ilter
s
)
,
R
eL
U
ac
tiv
atio
n
s
,
an
d
m
a
x
p
o
o
lin
g
.
Featu
r
es f
latten
ed
an
d
p
r
o
ce
s
s
ed
v
ia
d
en
s
e
lay
er
s
with
d
r
o
p
o
u
t.
f.
L
STM
:
L
STM
s
eq
u
en
ce
m
o
d
el
with
6
4
u
n
its
em
p
lo
y
e
d
to
an
aly
ze
s
er
ial
n
u
m
b
er
s
b
e
f
o
r
e
R
eL
U
-
ac
tiv
ated
d
en
s
e
lay
er
.
g.
Fu
s
io
n
an
d
C
lass
if
icatio
n
:
lin
ea
r
SVM
(
C
=1
,
to
ler
an
ce
=
1
e
-
3
)
u
s
ed
to
class
if
y
th
e
o
u
tp
u
t
o
f
in
teg
r
ated
C
NN
an
d
L
STM
.
h.
Mo
d
el
tr
ain
in
g
:
Ad
a
m
o
p
tim
i
ze
r
u
s
ed
to
tr
ain
th
e
m
o
d
el
(
0
.
5
d
r
o
p
o
u
t
f
o
r
C
NN,
0
.
2
f
o
r
L
STM
)
,
b
atch
s
ize
o
f
3
2
,
an
d
5
0
e
p
o
ch
s
.
E
a
r
ly
s
to
p
p
in
g
u
s
ed
to
a
v
o
id
o
v
er
f
itti
n
g
.
i.
Per
f
o
r
m
an
ce
e
v
alu
atio
n
:
th
e
p
r
o
p
o
s
ed
m
o
d
el
e
v
alu
ated
u
s
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
al
l,
F1
-
s
co
r
e,
an
d
AUC.
T
h
e
ab
o
v
e
s
tep
s
o
f
th
e
r
esear
ch
p
r
o
ce
d
u
r
e
ar
e
ex
p
lain
ed
in
m
o
r
e
d
etail
in
s
ep
ar
ate
co
m
in
g
s
u
b
s
ec
tio
n
s
.
3
.
1
.
Da
t
a
s
et
T
h
e
d
ataset
u
s
ed
i
n
th
is
s
tu
d
y
d
o
wn
lo
ad
e
d
f
r
o
m
Ka
g
g
le
[
4
2
]
is
a
co
llectio
n
o
f
b
alan
ce
d
h
i
g
h
-
q
u
ality
im
ag
es
f
ea
tu
r
in
g
b
an
k
n
o
tes
J
o
r
d
an
ian
Din
a
r
(
J
D)
:
1
,
5
,
1
0
,
2
0
,
an
d
5
0
J
Ds.
I
t
in
clu
d
es
n
e
w
ed
itio
n
s
o
f
1
,
2
0
,
an
d
5
0
J
Ds
b
an
k
n
o
tes.
T
h
is
d
ataset
co
n
tain
s
5
4
7
3
im
ag
es
a
n
d
d
esig
n
e
d
f
o
r
th
e
d
ev
elo
p
m
en
t
an
d
e
v
alu
atio
n
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els
f
o
r
au
t
o
m
atic
cu
r
r
e
n
cy
r
ec
o
g
n
itio
n
a
n
d
d
e
n
o
m
in
atio
n
class
if
icatio
n
task
s
.
T
ab
le
1
s
h
o
ws
th
e
n
u
m
b
er
o
f
s
am
p
les
f
o
r
ea
ch
J
o
r
d
an
ian
Din
ar
(
J
D)
d
en
o
m
in
atio
n
,
a
n
d
Fig
u
r
e
3
s
h
o
ws
g
en
u
in
e
a
n
d
c
o
u
n
ter
f
eit
1
J
D.
Ser
i
al
n
u
m
b
e
r
s
r
ec
o
r
d
ed
f
r
o
m
b
a
n
k
n
o
t
es
s
ets
t
o
d
ete
ct
i
r
r
e
g
u
la
r
s
e
q
u
e
n
c
e
a
n
d
r
e
c
o
g
n
i
ze
c
o
u
n
te
r
f
eit
cu
r
r
e
n
c
y
t
h
r
o
u
g
h
i
r
r
e
g
u
la
r
iti
e
s
in
ci
r
c
u
la
ti
o
n
p
att
er
n
s
.
Als
o
,
GANs
we
r
e
u
s
ed
to
c
r
e
ate
s
y
n
t
h
eti
c
c
o
u
n
te
r
f
ei
t
b
a
n
k
n
o
t
es
to
en
h
a
n
ce
m
o
d
el
’
s
a
b
il
it
y
t
o
d
et
ec
t
f
o
r
g
e
r
i
es.
W
e
m
u
s
t
b
e
aw
ar
e
th
at
t
h
e
s
a
m
p
l
es
g
e
n
e
r
a
te
d
GANs
m
a
y
b
e
u
s
e
d
b
y
m
ali
ci
o
u
s
in
d
i
v
i
d
u
als
to
p
r
o
d
u
c
e
c
o
u
n
te
r
f
ei
t
cu
r
r
e
n
c
y
t
h
a
t
is
c
lo
s
e
t
o
t
h
e
r
e
al
c
u
r
r
en
c
y
,
s
o
t
h
is
m
u
s
t
b
e
c
o
n
t
r
o
lle
d
.
T
h
e
r
e
f
o
r
e,
in
t
h
is
r
esea
r
c
h
,
we
a
d
h
e
r
e
t
o
e
th
i
ca
l
g
u
id
eli
n
es
r
e
g
a
r
d
i
n
g
t
h
e
u
s
e
o
f
GANs
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
0
4
-
5
8
1
4
5808
Fig
u
r
e
3
.
wo
r
k
f
lo
w
o
f
th
e
h
y
b
r
id
co
u
n
ter
f
eit
d
etec
tio
n
m
o
d
e
l
T
ab
le
1
.
Nu
m
b
er
o
f
s
am
p
les f
o
r
ea
ch
J
o
r
d
an
ian
Din
a
r
(
J
D)
d
en
o
m
in
atio
n
N
o
t
e
s
(
JD
)
N
u
mb
e
r
s
50
1
1
7
8
20
1
2
2
5
10
8
6
7
5
1
0
8
1
1
1
2
3
Co
un
t
er
f
eit
G
enuin
e
Fig
u
r
e
3
.
Sam
p
le
o
f
g
en
u
in
e
(
l
ef
t)
an
d
c
o
u
n
ter
f
eit
(
r
ig
h
t)
1
J
D
b
an
k
n
o
te
3
.
2
.
P
re
pro
ce
s
s
ing
T
o
o
b
tain
h
ig
h
-
q
u
ality
im
a
g
es
an
d
ex
tr
ac
t
f
ea
tu
r
es,
th
e
b
an
k
n
o
te
im
ag
es
wer
e
s
tan
d
ar
d
ize
d
in
ter
m
s
o
f
s
ize
(
2
2
4
,
2
2
4
,
3
)
,
r
eso
lu
tio
n
,
an
d
lig
h
tin
g
co
n
d
itio
n
s
to
b
e
u
s
ed
b
y
d
ee
p
lear
n
i
n
g
m
o
d
els.
T
h
e
d
ataset
co
n
tain
s
im
ag
es
o
f
J
o
r
d
an
ian
cu
r
r
en
cy
with
d
if
f
er
en
t
b
ac
k
g
r
o
u
n
d
s
an
d
co
n
d
itio
n
s
to
ad
d
co
m
p
lex
ity
to
th
e
d
ata
an
d
s
im
u
late
r
ea
l
co
n
d
itio
n
s
.
I
t
in
clu
d
es
im
ag
es
o
f
f
o
l
d
ed
o
r
p
ar
tially
h
id
d
e
n
b
an
k
n
o
tes.
T
h
is
is
v
er
y
im
p
o
r
tan
t
to
im
p
r
o
v
e
th
e
m
o
d
el's
ab
ilit
y
to
d
etec
t
co
u
n
ter
f
e
it
cu
r
r
en
cy
in
v
ar
io
u
s
co
n
d
itio
n
s
.
T
h
is
was
d
o
n
e
u
s
in
g
n
o
is
e
r
ed
u
ctio
n
,
c
o
n
tr
ast
ad
ju
s
tm
en
t,
an
d
g
r
a
y
s
ca
le
tr
an
s
f
o
r
m
atio
n
a
p
p
licatio
n
s
.
Pre
p
r
o
ce
s
s
in
g
was
co
n
d
u
cted
to
en
s
u
r
e
th
e
q
u
ality
o
f
th
e
b
an
k
n
o
te
im
ag
es
as
well
a
s
t
o
m
ee
t
th
e
r
eq
u
ir
em
e
n
ts
o
f
th
e
d
ee
p
lear
n
in
g
m
o
d
el.
Prio
r
ity
was
g
iv
en
to
in
ter
p
r
etin
g
co
n
tr
ast
b
ec
au
s
e
it
r
ev
ea
ls
ess
en
tial
v
is
u
al
f
ea
tu
r
es
f
o
r
d
i
s
tin
g
u
is
h
in
g
co
u
n
ter
f
eitin
g
,
s
u
ch
as
ed
g
es
an
d
p
atter
n
s
.
T
h
is
is
d
u
e
to
th
e
lack
o
f
lig
h
tin
g
,
an
d
th
is
m
o
d
if
ica
tio
n
m
ad
e
it
p
o
s
s
ib
le
to
id
en
t
if
y
th
em
d
u
r
in
g
f
ea
tu
r
e
e
x
tr
a
ctio
n
in
th
e
C
NN.
So
m
e
im
ag
es
m
ay
co
n
tain
u
n
wan
ted
ar
tifa
cts
th
at
h
in
d
er
th
e
m
o
d
el
d
u
r
in
g
th
e
tr
ain
in
g
p
r
o
ce
s
s
.
T
h
er
ef
o
r
e,
n
o
is
e
r
ed
u
ctio
n
was
em
p
lo
y
e
d
to
en
h
a
n
ce
th
e
clar
ity
o
f
t
h
e
im
ag
es,
an
d
t
h
e
m
o
d
el
t
h
en
co
n
ce
n
t
r
ates
o
n
s
ig
n
if
ican
t
f
ea
tu
r
es
in
s
tead
o
f
n
o
is
e.
Oth
er
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es
wer
e
em
p
lo
y
ed
to
im
p
r
o
v
e
th
e
g
en
er
aliza
b
ilit
y
an
d
r
elia
b
ilit
y
o
f
th
e
m
o
d
el
u
n
d
er
d
if
f
er
e
n
t
co
n
d
itio
n
s
,
s
u
ch
as
g
r
a
y
s
ca
le
tr
an
s
f
o
r
m
atio
n
,
r
o
tatio
n
,
g
r
ad
ien
t,
a
n
d
r
e
f
lectio
n
.
Dif
f
er
e
n
t
v
iews
wer
e
cr
ea
ted
to
f
ac
ilit
ate
th
e
m
o
d
ellin
g
o
f
th
e
m
o
d
el
f
r
o
m
d
if
f
er
en
t
an
g
les
a
n
d
p
o
s
itio
n
s
o
f
th
e
cu
r
r
en
cy
.
Pre
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
th
at
i
n
cr
ea
s
e
co
m
p
lex
ity
an
d
r
eq
u
ir
e
a
co
m
p
u
tatio
n
al
p
r
o
b
lem
wer
e
av
o
id
ed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
a
r
tifi
cia
l in
tellig
en
ce
a
p
p
r
o
a
c
h
to
co
u
n
terf
eit
cu
r
r
en
cy
d
etec
tio
n
(
Mo
n
th
er Ta
r
a
w
n
eh
)
5809
T
o
en
s
u
r
e
th
at
t
h
e
m
o
d
el
w
o
r
k
s
r
ea
lis
tically
an
d
p
r
e
v
en
t
o
v
er
f
itti
n
g
,
s
o
m
e
tech
n
iq
u
e
s
s
u
ch
as
r
o
tatio
n
,
s
ca
lin
g
,
clip
p
in
g
s
,
a
n
d
f
lip
p
in
g
wer
e
ap
p
lied
to
th
e
im
ag
es,
th
is
cr
ea
te
n
ew
i
m
ag
es
with
d
if
f
er
en
t
s
itu
atio
n
s
an
d
en
s
u
r
e
th
at
th
e
tr
ain
in
g
is
m
o
r
e
th
an
m
em
o
r
izin
g
s
p
ec
if
ic
f
ea
tu
r
es.
T
h
e
n
u
m
b
er
o
f
im
ag
es
in
cr
ea
s
e
f
r
o
m
5
4
7
3
to
7
3
1
2
.
As
f
o
r
th
e
s
er
ial
n
u
m
b
er
,
it
was
clea
n
ed
an
d
f
o
r
m
atted
to
e
n
s
u
r
e
co
n
s
is
ten
t
in
p
u
t
f
o
r
an
aly
s
is
b
y
th
e
L
STM
m
o
d
el.
T
o
m
in
im
ize
th
e
r
is
k
o
f
o
v
er
f
itti
n
g
,
d
r
o
p
o
u
t
lay
er
s
wer
e
ad
d
ed
to
th
e
m
o
d
el.
Fu
r
th
e
r
m
o
r
e
,
co
m
p
lex
m
o
d
els
wer
e
p
e
n
alize
d
,
an
d
th
e
s
im
p
lest
s
o
lu
tio
n
s
wer
e
tak
en
.
T
o
cr
ea
te
a
b
alan
ce
d
d
ataset,
an
d
to
en
s
u
r
e
th
at
th
er
e
is
n
o
o
v
er
f
itti
n
g
f
o
r
ce
r
tain
ty
p
es
o
f
d
ata,
r
ea
l
s
am
p
les
wer
e
co
m
b
in
ed
with
th
o
s
e
p
r
o
d
u
c
ed
b
y
GANs.
T
h
is
en
s
u
r
es
t
h
at
th
e
m
o
d
el'
s
ab
ilit
y
is
en
h
an
ce
d
b
y
u
s
in
g
th
e
s
am
p
les p
r
o
d
u
ce
d
b
y
GANsz.
3
.
3
.
I
m
ple
m
ent
a
t
io
n
T
h
e
b
a
n
k
n
o
tes
im
ag
es
will
b
e
lo
ad
ed
in
to
C
NN
m
o
d
el
to
ex
tr
ac
t
f
ea
tu
r
es
to
ca
p
tu
r
e
all
f
e
atu
r
es
th
at
ar
e
im
p
o
r
ta
n
t
f
o
r
C
o
u
n
ter
f
eit
d
etec
tio
n
.
T
h
e
C
NN
in
teg
r
a
ted
with
L
STM
to
an
aly
ze
th
e
p
atter
n
o
f
s
er
ial
n
u
m
b
er
s
an
d
id
e
n
tify
ir
r
e
g
u
la
r
p
atter
n
s
.
T
h
e
n
th
e
e
x
tr
ac
ted
f
ea
tu
r
es
class
if
ied
b
y
SVM
as
r
ea
l
o
r
c
o
u
n
ter
f
eit.
GANs
u
s
ed
to
cr
ea
te
r
ea
lis
tic
f
ak
e
im
ag
es,
wh
ich
wer
e
u
s
ed
to
au
g
m
e
n
t
th
e
tr
ain
i
n
g
d
atas
et.
T
h
e
C
NN
m
o
d
el
is
im
p
lem
en
ted
with
f
o
u
r
m
ai
n
lay
er
s
:
a.
C
o
n
v
o
lu
tio
n
al
l
ay
e
r
s
:
T
h
e
m
o
d
el
s
tar
ts
with
a
s
er
ies
o
f
co
n
v
o
lu
ti
o
n
al
lay
er
s
d
esig
n
e
d
to
ca
p
tu
r
e
lo
ca
l
f
ea
tu
r
es
s
u
ch
as
ed
g
es,
tex
tu
r
e
s
,
s
h
ap
es,
an
d
p
atter
n
s
in
th
e
i
n
p
u
t
im
ag
es.
T
h
ese
lay
er
s
ar
e
r
esp
o
n
s
ib
le
f
o
r
lear
n
in
g
lo
w
-
lev
el
an
d
h
ig
h
-
lev
el
im
ag
e
f
ea
tu
r
es
th
at
ar
e
ess
en
tial
f
o
r
d
etec
tin
g
in
tr
icate
d
if
f
er
en
ce
s
b
etwe
en
r
ea
l a
n
d
co
u
n
ter
f
eit
b
an
k
n
o
tes.
b.
Ma
x
p
o
o
lin
g
lay
er
s
: A
f
ter
ea
c
h
co
n
v
o
lu
tio
n
al
lay
er
,
a
m
ax
p
o
o
lin
g
lay
er
r
ed
u
ce
s
th
e
s
p
atia
l d
im
en
s
io
n
s
o
f
th
e
ex
tr
ac
ted
f
ea
tu
r
e
m
ap
s
.
T
h
is
d
o
wn
s
am
p
lin
g
allo
ws th
e
m
o
d
el
to
f
o
cu
s
o
n
th
e
m
o
s
t im
p
o
r
tan
t f
ea
tu
r
es.
T
h
e
co
m
p
u
tatio
n
al
co
m
p
lex
ity
was
r
ed
u
ce
d
,
wh
ile
m
ai
n
tain
in
g
th
e
m
o
s
t
im
p
o
r
ta
n
t
f
ea
tu
r
es
f
r
o
m
th
e
im
ag
e.
c.
Flatten
l
ay
er
:
On
ce
t
h
e
m
o
d
el
h
as
s
elec
ted
an
d
r
e
f
in
ed
th
e
i
m
p
o
r
tan
t
v
is
u
al
f
ea
tu
r
es
u
s
in
g
its
s
p
ec
ialized
lay
er
s
,
it
n
ee
d
s
to
s
im
p
lify
th
is
in
f
o
r
m
atio
n
f
o
r
f
u
r
th
er
p
r
o
c
ess
in
g
.
T
h
is
i
s
d
o
n
e
b
y
f
latten
in
g
th
e
co
m
p
lex
2
D
f
ea
tu
r
e
m
ap
s
in
to
a
s
im
p
l
e
s
tr
aig
h
t
lin
e
—
a
1
D
v
ec
to
r
.
T
h
is
s
tep
is
cr
u
cial
b
ec
a
u
s
e
it
tr
an
s
f
o
r
m
s
t
h
e
p
r
o
ce
s
s
ed
im
ag
e
d
ata
i
n
to
a
s
im
p
ler
f
o
r
m
t
h
at
th
e
n
ex
t
p
ar
t
o
f
th
e
n
etwo
r
k
,
wh
ic
h
d
o
es
th
e
ac
t
u
al
class
if
icatio
n
,
ca
n
ea
s
ily
wo
r
k
with
d.
Af
ter
tr
an
s
f
o
r
m
in
g
th
e
c
o
m
p
l
ex
d
ata
i
n
to
a
s
in
g
le,
s
tr
aig
h
t
lin
e
(
th
e
f
latten
ed
f
ea
tu
r
e
v
e
cto
r
)
,
it
e
n
ter
s
a
d
en
s
e
lay
er
p
ac
k
ed
with
5
1
2
t
in
y
p
r
o
ce
s
s
in
g
s
tatio
n
s
.
R
eL
U
ac
tiv
atio
n
f
u
n
ctio
n
h
elp
to
f
o
cu
s
o
n
th
e
m
o
s
t
u
s
ef
u
l f
ea
tu
r
es a
m
o
n
g
all
f
ea
t
u
r
es
.
Nex
t,
th
is
p
r
ep
ar
ed
d
ata
m
o
v
e
s
to
an
o
th
er
d
en
s
e
lay
er
,
wh
ic
h
f
o
cu
s
ed
s
o
lely
o
n
m
a
k
in
g
o
n
e
cr
u
cial
d
ec
is
io
n
:
is
th
e
b
an
k
n
o
te
r
ea
l
o
r
co
u
n
ter
f
eit?
I
t
d
o
es
th
is
th
r
o
u
g
h
a
s
in
g
le
u
n
it
eq
u
ip
p
ed
with
a
s
ig
m
o
id
ac
tiv
atio
n
f
u
n
ctio
n
,
wh
ich
s
q
u
ee
ze
s
th
e
in
co
m
in
g
d
ata
in
t
o
a
r
an
g
e
b
etwe
en
0
an
d
1
.
T
h
is
m
eth
o
d
allo
ws
th
e
m
o
d
el
to
m
ak
e
s
en
s
e
o
f
b
a
n
k
n
o
te
im
ag
es
in
a
s
o
p
h
is
ticated
way
,
u
ltima
tely
p
r
o
v
i
d
in
g
a
p
r
o
b
a
b
ilit
y
th
at
ac
ts
m
u
ch
lik
e
a
c
o
n
f
id
e
n
ce
s
co
r
e
o
n
wh
eth
er
th
e
b
a
n
k
n
o
te
is
g
e
n
u
in
e
o
r
f
ak
e
.
C
NN
in
teg
r
ated
with
L
STM
(
n
etwo
r
k
is
u
s
ed
to
a
n
aly
ze
b
an
k
n
o
te
s
er
ial
n
u
m
b
er
s
o
f
l
en
g
th
1
-
1
0
.
T
h
e
s
er
ial
n
u
m
b
er
s
o
n
th
e
b
a
n
k
n
o
tes
ar
e
p
r
o
ce
s
s
ed
u
s
in
g
a
n
L
STM
lay
er
with
6
4
m
o
d
u
l
es.
T
h
e
in
p
u
t
s
h
ap
e
p
ar
am
eter
is
f
ed
to
th
e
L
ST
M
lay
er
to
d
eter
m
in
e
th
e
s
tr
u
ctu
r
e
o
f
th
e
in
p
u
t
s
eq
u
en
ce
.
T
o
ca
p
tu
r
e
n
o
n
lin
ea
r
r
elatio
n
s
h
ip
s
,
a
f
u
lly
c
o
n
n
ec
te
d
d
en
s
e
lay
er
co
n
tain
in
g
1
2
8
a
n
d
R
eL
U
ac
tiv
atio
n
u
n
its
wer
e
ad
d
ed
.
T
h
e
h
y
b
r
id
m
o
d
el
was
tr
ain
ed
with
a
s
u
p
er
v
is
ed
lear
n
in
g
ap
p
r
o
ac
h
.
T
h
e
C
NN
an
d
L
STM
n
etwo
r
k
s
o
p
tim
ized
u
s
in
g
Ad
am
o
p
tim
izer
,
wh
ile
SVM
u
s
ed
lin
ea
r
k
er
n
el
f
o
r
f
in
al
class
if
icatio
n
.
A
co
m
b
in
atio
n
o
f
d
r
o
p
o
u
t a
n
d
ea
r
ly
s
to
p
p
in
g
tec
h
n
iq
u
es wa
s
em
p
lo
y
e
d
to
a
v
o
i
d
o
v
er
f
itti
n
g
an
d
en
s
u
r
e
e
f
f
ec
t
iv
e
tr
ain
in
g
.
I
n
o
r
d
er
t
o
g
et
t
h
e
b
est
p
e
r
f
o
r
m
an
ce
o
f
th
e
h
y
b
r
id
m
o
d
el,
th
e
h
y
p
er
p
a
r
am
eter
s
o
f
th
e
alg
o
r
ith
m
s
in
clu
d
ed
in
th
e
m
o
d
el
wer
e
m
o
d
if
ied
.
Fo
r
th
e
C
NN,
d
if
f
e
r
en
t
f
ilter
s
izes
an
d
n
u
m
b
e
r
s
wer
e
test
ed
an
d
th
e
b
est
p
er
f
o
r
m
an
ce
was
ac
h
iev
ed
u
s
in
g
3
x
3
,
3
2
an
d
6
4
f
ilter
s
in
th
e
f
ir
s
t
two
lay
er
s
,
f
o
llo
wed
b
y
1
2
8
in
th
e
d
ee
p
er
lay
er
s
.
R
eL
U
ac
tiv
atio
n
is
u
s
ed
d
u
e
to
its
ab
ilit
y
to
h
an
d
le
n
o
n
li
n
ea
r
ity
,
an
d
r
ed
u
c
e
s
p
atial
d
im
en
s
io
n
s
with
o
u
t
lo
s
in
g
im
p
o
r
tan
t
f
ea
t
u
r
es,
m
ax
p
o
o
lin
g
is
u
s
ed
with
a
p
o
o
l
s
ize
o
f
2
x
2
.
Als
o
,
g
r
id
s
ea
r
ch
d
eter
m
in
ed
an
o
p
tim
al
lear
n
in
g
r
ate
o
f
0
.
0
0
0
1
,
an
d
a
d
r
o
p
o
u
t
r
ate
o
f
0
.
5
in
th
e
f
u
lly
co
n
n
ec
ted
la
y
er
s
,
wh
ich
av
o
i
d
s
o
v
er
f
itti
n
g
.
Fo
r
L
STM
,
th
e
b
e
s
t
n
u
m
b
er
o
f
h
id
d
en
u
n
its
wa
s
6
4
a
n
d
1
0
len
g
th
u
n
its
wer
e
co
n
ca
ten
ate
d
to
f
it
th
e
s
tr
u
ctu
r
e
o
f
th
e
s
eq
u
e
n
ce
d
ata.
A
d
r
o
p
o
u
t
r
ate
o
f
0
.
2
was
u
s
ed
to
av
o
id
th
e
o
v
er
f
itti
n
g
p
r
o
b
lem
.
T
o
e
n
s
u
r
e
ad
ap
tiv
e
an
d
ef
f
icie
n
t
lear
n
in
g
,
Ad
am
o
p
tim
izer
was
u
s
ed
with
a
lear
n
in
g
r
ate
o
f
0
.
0
0
1
.
T
o
o
b
tain
ef
f
icien
t
a
n
d
ac
c
u
r
a
t
e
c
l
a
s
s
i
f
i
ca
t
i
o
n
,
a
l
i
n
e
a
r
S
V
M
k
e
r
n
e
l
w
a
s
u
s
e
d
.
T
h
e
o
p
t
im
a
l
v
a
lu
e
o
f
t
h
e
r
eg
u
l
ar
i
z
a
t
io
n
p
a
r
am
e
t
e
r
(
C
)
w
a
s
1
w
i
t
h
a
t
o
l
e
r
an
c
e
o
f
1
e
-
3
t
o
e
n
s
u
r
e
f
a
s
t
c
o
n
v
er
g
en
c
e
an
d
b
a
la
n
c
ed
a
cc
u
r
a
c
y
.
Ma
n
y
e
x
p
er
i
m
en
t
s
w
e
r
e
c
o
n
d
u
c
t
ed
t
o
ch
o
o
s
e
t
h
e
s
e
h
y
p
er
p
ar
a
m
e
t
e
r
s
t
o
e
n
s
u
r
e
th
e
ef
f
i
c
i
en
t
p
e
r
f
o
r
m
a
n
c
e
o
f
th
e
h
y
b
r
i
d
m
o
d
e
l.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
0
4
-
5
8
1
4
5810
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
A
v
ar
iety
o
f
m
etr
ics u
tili
ze
d
to
ev
alu
ate
t
h
e
p
e
r
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el,
in
clu
d
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e
,
an
d
th
e
co
n
f
u
s
io
n
m
atr
ix
.
T
o
en
s
u
r
e
th
e
r
eliab
ilit
y
o
f
th
e
m
o
d
el
,
it
was
tr
ain
e
d
a
n
d
ass
es
s
ed
o
n
b
o
th
g
en
u
in
e
d
ata
(
J
o
r
d
an
ia
n
b
a
n
k
n
o
tes)
an
d
GAN
-
g
en
er
ated
c
o
u
n
ter
f
eit
d
ata
,
s
ev
er
al
co
n
d
itio
n
s
co
n
s
id
er
ed
f
o
r
r
ec
o
g
n
izin
g
g
e
n
u
in
e
an
d
co
u
n
te
r
f
eit
b
an
k
n
o
t
es
to
p
r
o
v
id
e
a
co
m
p
r
e
h
en
s
iv
e
en
v
ir
o
n
m
en
t
f
o
r
ev
alu
atin
g
m
o
d
el
p
er
f
o
r
m
an
ce
o
n
d
iv
er
s
e
in
p
u
ts
.
T
h
e
im
p
lem
en
tatio
n
was
im
p
l
em
en
ted
in
C
o
lab
en
v
ir
o
n
m
e
n
t
u
s
in
g
Py
th
o
n
lib
r
ar
ies.
I
n
th
i
s
r
esear
ch
,
a
h
y
b
r
id
s
y
s
tem
was
im
p
lem
en
ter
b
y
in
te
g
r
atin
g
C
NN
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
,
L
STM
f
o
r
s
eq
u
en
tial
p
atter
n
r
ec
o
g
n
itio
n
(
cu
r
r
e
n
cy
s
er
ial
n
u
m
b
er
s
)
,
a
n
d
SVM
f
o
r
class
if
icatio
n
o
f
cu
r
r
en
c
y
as
f
ak
e
o
r
g
en
u
i
n
e.
T
h
is
m
eth
o
d
en
a
b
les
th
e
m
o
d
el
to
an
aly
ze
v
is
u
al
f
ea
tu
r
es
f
r
o
m
cu
r
r
en
c
y
an
d
th
e
p
atter
n
s
o
f
s
eq
u
en
tial
s
er
ial
n
u
m
b
er
s
,
im
p
r
o
v
in
g
its
ac
cu
r
ac
y
.
T
o
p
r
o
v
id
e
s
u
f
f
icien
t
d
at
a
f
o
r
m
o
d
el
lear
n
in
g
in
th
e
tr
ain
in
g
an
d
ac
cu
r
ate
ev
alu
atio
n
o
n
h
id
d
en
d
ata,
th
e
d
ataset
was
s
p
lit
in
to
8
0
%
f
o
r
tr
ain
in
g
an
d
2
0
%
f
o
r
test
in
g
.
T
h
e
r
esu
lts
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
(
C
NN
-
L
ST
M
-
SVM)
wer
e
v
er
y
r
em
a
r
k
a
b
le,
r
ea
ch
in
g
an
ac
cu
r
ac
y
o
f
9
8
.
6
%,
wh
ich
p
r
o
v
es
th
e
m
o
d
el'
s
ab
ilit
y
to
ac
cu
r
ately
d
is
tin
g
u
is
h
b
etwe
en
co
u
n
te
r
f
eit
an
d
g
e
n
u
in
e
c
u
r
r
en
c
y
in
m
o
s
t
s
itu
atio
n
s
.
T
h
e
co
m
p
u
tatio
n
al
e
f
f
icien
cy
a
n
d
r
eso
u
r
ce
r
eq
u
ir
em
en
ts
f
o
r
t
h
e
h
y
b
r
id
m
o
d
el
wer
e
test
ed
an
d
co
m
p
ar
ed
to
in
d
iv
id
u
al
r
eq
u
i
r
em
en
ts
f
o
r
C
NN,
L
STM
an
d
SVM.
Desp
ite
th
e
h
i
g
h
er
r
eso
u
r
ce
r
eq
u
ir
e
m
en
ts
ex
p
ec
ted
f
o
r
th
e
m
o
d
el
in
teg
r
atin
g
C
NN,
L
STM
,
an
d
SVM,
th
e
h
y
b
r
i
d
m
o
d
el
d
em
o
n
s
tr
ated
s
tr
o
n
g
p
er
f
o
r
m
an
ce
an
d
ac
cu
r
ac
y
in
t
h
e
r
esu
lts
,
r
esu
ltin
g
in
its
co
m
p
u
tatio
n
al
co
s
t.
T
h
e
co
m
b
in
atio
n
o
f
C
NN,
L
STM
,
an
d
SVM
s
u
b
s
tan
tially
en
h
an
ce
d
th
e
m
o
d
el'
s
ca
p
ab
i
lity
to
d
ea
l
with
b
o
th
s
p
atial
an
d
s
eq
u
en
ti
al
d
ata.
T
ab
le
2
p
r
o
v
id
es
a
co
m
p
ar
is
o
n
b
etwe
en
th
e
p
r
o
p
o
s
ed
m
o
d
el
an
d
o
th
er
m
o
d
els test
ed
o
n
th
e
s
am
e
d
ataset,
h
ig
h
lig
h
tin
g
t
h
e
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
o
f
th
e
C
NN
-
L
STM
-
SVM
ap
p
r
o
ac
h
.
T
ab
le
2
.
C
o
m
p
ar
is
o
n
b
etwe
en
th
e
p
r
o
p
o
s
ed
m
o
d
el
an
d
o
th
er
m
o
d
els
M
o
d
e
l
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
S
c
o
r
e
A
c
c
u
r
a
c
y
C
N
N
9
5
.
4
0
%
9
5
.
2
0
%
9
5
.
3
0
%
9
5
.
0
0
%
C
N
N
-
LSTM
9
6
.
5
0
%
9
6
.
3
0
%
9
6
.
4
0
%
9
6
.
2
0
%
C
N
N
-
GAN
9
5
.
0
0
%
9
4
.
5
0
%
9
4
.
7
0
%
9
4
.
8
0
%
C
N
N
-
LSTM
-
S
V
M
9
8
.
8
0
%
9
8
.
3
0
%
9
8
.
5
0
%
9
8
.
6
0
%
C
N
N
-
C
A
M
9
6
.
6
0
%
9
6
.
4
0
%
9
6
.
5
0
%
9
6
.
0
0
%
C
N
N
-
G
R
U
9
5
.
8
0
%
9
5
.
6
0
%
9
5
.
7
0
%
9
5
.
7
0
%
C
N
N
-
R
N
N
9
4
.
5
0
%
9
4
.
0
0
%
9
4
.
3
0
%
9
4
.
2
0
%
C
N
N
-
T
r
a
n
sf
o
r
mer
9
7
.
3
0
%
9
7
.
0
0
%
9
7
.
1
0
%
9
7
.
1
0
%
T
h
e
r
esu
lt
s
h
o
ws
th
e
ef
f
ec
tiv
en
ess
o
f
co
m
b
in
in
g
th
e
class
if
icatio
n
ac
cu
r
ac
y
o
f
SVM
a
n
d
f
ea
tu
r
e
r
ec
o
g
n
itio
n
o
f
C
NN.
T
h
e
h
i
g
h
ac
cu
r
ac
y
o
f
th
e
m
o
d
els,
clo
s
e
to
1
0
0
%,
i
n
d
icate
s
th
a
t
th
er
e
is
r
o
o
m
f
o
r
im
p
r
o
v
em
e
n
t
o
f
th
e
s
y
s
tem
an
d
f
u
r
th
er
im
p
r
o
v
em
e
n
t
o
f
ac
cu
r
ac
y
.
Ho
wev
e
r
,
th
is
d
ep
e
n
d
s
o
n
th
e
s
ize
o
f
th
e
av
ailab
le
d
ata
an
d
its
d
iv
er
s
ity
.
As
well
as
th
e
q
u
ality
o
f
th
e
im
ag
es
av
ailab
le
in
d
if
f
er
en
t
co
n
d
itio
n
s
an
d
an
g
les.
I
n
ad
d
itio
n
to
r
ea
ch
in
g
th
e
ap
p
r
o
p
r
iate
s
ettin
g
s
f
o
r
th
e
m
o
d
el.
All
th
is
m
ay
g
u
a
r
an
tee
r
esu
lts
with
ex
ce
llen
t a
cc
u
r
ac
y
in
d
etec
tin
g
f
o
r
g
er
y
.
Fig
u
r
e
4
s
h
o
ws
th
e
R
O
C
cu
r
v
e
o
f
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
h
y
b
r
id
m
o
d
el.
T
h
e
ca
lcu
lated
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC)
o
f
0
.
9
2
s
h
o
ws
th
at
th
e
m
o
d
el
h
as
th
e
ab
ilit
y
to
d
is
tin
g
u
is
h
b
e
twee
n
g
en
u
in
e
an
d
co
u
n
ter
f
eit
m
o
n
e
y
.
T
h
is
r
esu
lt
co
n
f
ir
m
s
th
e
r
esu
lts
lis
ted
in
T
a
b
le
2
th
at
th
e
p
r
o
p
o
s
ed
h
y
b
r
id
m
o
d
el
o
u
tp
er
f
o
r
m
s
o
t
h
er
m
eth
o
d
s
i
n
m
o
s
t
m
etr
ics
in
clu
d
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
s
co
r
e.
T
h
e
tr
ain
in
g
an
d
v
alid
atio
n
lo
s
e
d
ep
icted
i
n
Fig
u
r
e
5
s
h
o
ws
th
e
ef
f
icien
cy
o
f
th
e
m
o
d
el
with
lo
w
o
v
e
r
f
itti
n
g
.
W
e
n
o
tice
th
at
th
e
tr
ain
in
g
lo
s
s
d
ec
r
ea
s
es
f
r
o
m
a
b
o
u
t
0
.
6
to
less
th
an
0
.
2
,
wh
ich
is
r
elate
d
to
th
e
t
r
ain
i
n
g
ac
cu
r
ac
y
as
th
e
m
o
d
el
b
ec
o
m
es
m
o
r
e
co
n
f
id
e
n
t
in
its
p
r
ed
ictio
n
s
.
T
h
e
v
alid
atio
n
lo
s
s
d
ec
r
ea
s
es
f
r
o
m
0
.
6
to
ab
o
u
t
0
.
2
5
.
T
h
is
co
n
f
ir
m
s
th
at
th
e
m
o
d
el'
s
ab
i
lity
to
g
en
er
alize
im
p
r
o
v
es
with
ea
ch
ep
o
ch
.
T
h
e
co
n
v
er
g
e
n
ce
o
f
th
e
tr
ain
in
g
an
d
v
alid
atio
n
lo
s
s
v
alu
es to
war
d
s
th
e
f
o
llo
win
g
ep
o
c
h
s
p
r
o
v
es th
at
th
e
m
o
d
el
is
n
o
t o
v
e
r
f
itti
n
g
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
d
em
o
n
s
tr
ated
g
o
o
d
p
er
f
o
r
m
an
ce
in
d
is
tin
g
u
is
h
in
g
b
etwe
en
co
u
n
t
er
f
eit
an
d
g
en
u
in
e
b
a
n
k
n
o
tes.
T
h
e
co
m
b
in
atio
n
o
f
C
NN,
L
STM
an
d
SVM
AI
tech
n
iq
u
es h
ad
a
s
ig
n
if
ican
t im
p
ac
t o
n
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
m
o
d
el.
T
h
is
is
ex
p
ec
ted
as
th
e
m
o
d
el
co
m
b
in
es
th
e
f
ea
tu
r
es
o
f
C
NN
in
d
is
tin
g
u
is
h
in
g
b
etwe
en
f
in
e
v
is
u
al
f
ea
tu
r
es
t
h
at
ar
e
im
p
o
r
tan
t
in
d
is
tin
g
u
is
h
in
g
b
etwe
en
c
o
u
n
ter
f
eit
an
d
g
en
u
in
e
b
an
k
n
o
tes
an
d
th
e
f
ea
tu
r
es
o
f
L
STM
in
d
ea
lin
g
with
s
eq
u
en
tial
p
atter
n
s
in
s
er
ial
n
u
m
b
er
s
,
g
iv
in
g
t
h
e
ab
ilit
y
to
tr
ac
k
b
an
k
n
o
tes
an
d
id
en
tify
t
h
e
cu
l
p
r
it.
W
h
en
th
e
m
o
d
el
was
test
ed
o
n
co
u
n
ter
f
eit
b
a
n
k
n
o
tes
cr
ea
ted
b
y
GAN,
its
p
er
f
o
r
m
an
ce
was
s
atis
f
ac
to
r
y
with
an
ac
cu
r
ac
y
o
f
9
8
.
6
%.
T
h
e
m
o
d
el
d
em
o
n
s
tr
ated
a
s
lig
h
t
in
cr
ea
s
e
in
f
alse
n
eg
ativ
es
d
u
e
to
th
e
p
r
esen
ce
o
f
s
o
m
e
co
u
n
ter
f
eit
b
an
k
n
o
t
es
th
at
ar
e
clo
s
e
to
r
ea
l
b
an
k
n
o
tes
an
d
ca
n
n
o
t
b
e
d
is
tin
g
u
is
h
ed
b
y
v
is
u
al
in
s
p
ec
tio
n
.
Als
o
,
f
alse
p
o
s
itiv
es
wh
er
e
g
en
u
in
e
n
o
tes
co
u
l
d
b
e
class
if
ied
as
co
u
n
ter
f
eit.
T
h
is
b
r
in
g
s
a
b
o
u
t
th
e
n
ee
d
f
o
r
r
esear
ch
an
d
d
e
v
elo
p
m
en
t
to
im
p
r
o
v
e
d
etec
tio
n
ac
c
u
r
ac
y
.
B
o
th
Fals
e
p
o
s
itiv
e
an
d
f
alse
n
eg
ati
v
e
p
o
s
e
a
ch
allen
g
e
f
o
r
th
e
f
in
an
cial
s
y
s
tem
s
an
d
allo
w
co
u
n
ter
f
eit
cu
r
r
en
cy
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
a
r
tifi
cia
l in
tellig
en
ce
a
p
p
r
o
a
c
h
to
co
u
n
terf
eit
cu
r
r
en
cy
d
etec
tio
n
(
Mo
n
th
er Ta
r
a
w
n
eh
)
5811
b
e
u
s
ed
.
T
h
is
will
r
e
d
u
ce
th
e
t
r
u
s
t
in
th
e
d
etec
tio
n
s
y
s
tem
an
d
r
etu
r
n
u
s
to
th
e
m
an
u
al
v
er
if
icatio
n
p
r
o
ce
s
s
es.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el,
wh
ich
in
teg
r
ates
C
NN,
L
STM
an
d
SVM,
s
ig
n
if
ican
tly
r
ed
u
c
es
f
alse
n
eg
ativ
es
b
y
p
er
f
o
r
m
in
g
a
co
m
p
r
eh
e
n
s
iv
e
an
aly
s
is
o
f
v
is
u
al
an
d
s
eq
u
e
n
tial
f
ea
tu
r
es.
T
h
e
AUC
s
co
r
e
d
em
o
n
s
tr
ates
th
e
m
o
d
el'
s
ab
ilit
y
to
ef
f
ec
tiv
ely
b
alan
ce
tr
ad
e
-
o
f
f
s
.
Fig
u
r
e
4
.
R
OC
cu
r
v
e
o
f
th
e
p
r
o
p
o
s
ed
h
y
b
r
id
m
o
d
el
Fig
u
r
e
5
.
T
r
ain
in
g
a
n
d
v
alid
atio
n
ac
cu
r
ac
y
an
d
lo
s
s
o
f
th
e
p
r
o
p
o
s
ed
h
y
b
r
id
m
o
d
el
o
v
er
1
0
ep
o
ch
s
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
s
h
o
ws
b
etter
ac
c
u
r
ac
y
an
d
ca
n
b
e
co
n
s
id
er
ed
ap
p
licab
le
to
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
s
u
ch
as
AT
Ms,
b
an
k
n
o
te
co
u
n
tin
g
m
ac
h
in
es
an
d
s
ales
p
o
in
ts
.
Ho
wev
er
,
r
ely
i
n
g
s
o
lely
o
n
im
a
g
e
q
u
ality
an
d
s
er
ial
n
u
m
b
er
s
p
o
s
es
s
o
m
e
lim
itat
io
n
s
wh
en
wo
r
k
in
g
in
u
n
co
n
tr
o
lled
en
v
i
r
o
n
m
en
ts
,
s
u
ch
as
d
am
ag
ed
b
an
k
n
o
tes.
W
h
ile
th
e
u
s
e
o
f
GANs
r
ed
u
ce
s
r
elian
ce
o
n
lar
g
e
d
atasets
,
th
eir
u
s
e
also
ca
r
r
ies
s
o
m
e
r
is
k
s
.
T
h
er
ef
o
r
e,
e
x
p
lain
ab
le
ar
tific
ial
in
tellig
en
ce
(
XAI
)
s
h
o
u
ld
b
e
em
p
lo
y
ed
t
o
h
el
p
in
v
esti
g
ate
an
o
m
alies.
T
h
is
m
o
d
el
is
s
ca
lab
le
to
in
clu
d
e
o
th
er
alg
o
r
ith
m
s
to
co
m
b
at
co
u
n
ter
f
eitin
g
wo
r
ld
wid
e.
Als
o
,
it
n
ee
d
s
to
b
e
in
teg
r
ated
with
a
b
a
n
k
n
o
te
s
ca
n
n
in
g
s
y
s
tem
an
d
a
ce
n
tr
al
d
atab
ase
f
o
r
f
u
r
th
er
u
p
d
atin
g
an
d
tr
ain
in
g
th
e
m
o
d
e
l
an
d
en
a
b
lin
g
it
to
d
etec
t
co
u
n
ter
f
eit
cu
r
r
e
n
cy
i
n
r
ea
l
tim
e
b
ef
o
r
e
ac
ce
p
tin
g
d
ep
o
s
it
o
r
d
is
p
en
s
in
g
ca
s
h
.
T
h
e
f
in
d
in
g
s
o
f
t
h
is
s
tu
d
y
d
em
o
n
s
tr
ate
th
e
a
b
ilit
y
o
f
Hy
b
r
id
m
o
d
el
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
o
v
er
s
in
g
le
m
eth
o
d
s
.
T
h
is
r
esear
ch
co
n
tr
i
b
u
tes to
th
e
d
ig
ital secu
r
ity
in
th
e
f
in
an
c
e
s
y
s
tem
s
.
5.
CO
NCLU
SI
O
N
C
o
u
n
ter
f
eit
cu
r
r
e
n
cy
h
as
b
ec
o
m
e
a
p
r
o
b
lem
all
o
v
e
r
th
e
wo
r
ld
,
as
tr
ad
itio
n
al
m
eth
o
d
s
h
a
v
e
b
ec
o
m
e
in
ac
cu
r
ate
in
d
etec
tin
g
co
u
n
t
er
f
eit
b
an
k
n
o
tes,
s
o
tech
n
o
lo
g
y
h
as
b
ee
n
em
p
lo
y
ed
i
n
th
i
s
r
eg
ar
d
,
esp
ec
ially
m
ac
h
in
e
lear
n
in
g
an
d
ar
tific
ia
l
in
tellig
en
ce
m
eth
o
d
s
.
T
h
is
s
tu
d
y
p
r
o
p
o
s
es
an
e
f
f
ec
tiv
e
a
r
tific
ial
in
tellig
en
ce
-
b
ased
m
eth
o
d
f
o
r
d
etec
tin
g
c
o
u
n
ter
f
eit
cu
r
r
en
cy
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
0
4
-
5
8
1
4
5812
T
h
e
n
ew
p
r
o
p
o
s
ed
h
y
b
r
i
d
m
o
d
el
u
tili
ze
s
m
o
r
e
th
an
o
n
e
ar
ti
f
icial
in
tellig
en
ce
tech
n
iq
u
e
to
u
tili
ze
th
e
s
tr
en
g
th
s
o
f
ea
ch
a
n
d
f
in
d
a
co
m
p
r
eh
en
s
iv
e
s
o
lu
tio
n
to
co
u
n
ter
f
eit
c
u
r
r
e
n
cy
.
I
t
m
e
r
g
es
C
NNs,
L
STM
s
,
GANs,
an
d
SVMs.
T
h
e
C
N
N
-
L
STM
-
SVM
m
o
d
el
d
em
o
n
s
tr
ated
o
u
ts
tan
d
in
g
p
er
f
o
r
m
an
ce
,
ac
h
iev
in
g
an
im
p
r
ess
iv
e
9
8
% a
cc
u
r
ac
y
in
d
etec
tin
g
co
u
n
ter
f
eit
cu
r
r
en
c
y
c
o
m
p
ar
ed
t
o
p
r
ev
i
o
u
s
m
o
d
els.
C
NN
-
b
ased
m
o
d
els
ac
h
iev
ed
ac
cu
r
ac
y
r
ates
r
an
g
in
g
f
r
o
m
9
3
%
to
9
5
%,
w
h
ile
C
NN
-
L
STM
m
eth
o
d
s
ac
h
iev
ed
s
lig
h
tly
h
i
g
h
e
r
ac
cu
r
ac
y
r
ates
o
f
9
6
.
2
%.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
co
m
b
in
es
th
e
s
tr
en
g
th
s
o
f
f
ea
tu
r
e
ex
tr
ac
tio
n
(
C
NN)
,
s
eq
u
en
ce
lear
n
in
g
(
L
STM
)
,
class
if
icatio
n
r
o
b
u
s
tn
ess
(
SVM)
,
an
d
GA
N
-
b
ased
b
o
o
s
tin
g
.
Als
o
,
it
is
ca
p
ab
le
o
f
wo
r
k
in
g
in
d
if
f
er
e
n
t
s
itu
atio
n
s
,
g
r
an
t
ed
b
y
s
y
n
t
h
etic
d
ata
g
en
e
r
ated
b
y
g
en
er
ativ
e
a
d
v
er
s
ar
ial
n
etwo
r
k
s
with
its
s
ca
lab
ilit
y
.
R
esear
ch
an
d
d
e
v
e
lo
p
m
en
t
s
h
o
u
ld
c
o
n
tin
u
e
to
h
an
d
le
th
e
g
r
o
wth
o
f
co
u
n
ter
f
e
itin
g
m
eth
o
d
s
,
with
a
cr
itical
n
ee
d
to
c
o
n
tin
u
e
d
ev
elo
p
ar
tific
ial
in
tellig
en
ce
m
o
d
els an
d
tr
ain
in
g
m
eth
o
d
o
lo
g
ies.
T
h
er
e
is
a
n
ee
d
to
co
n
tin
u
e
d
e
v
elo
p
in
g
co
u
n
ter
f
eit
d
etec
tio
n
tech
n
o
l
o
g
ies.
T
h
e
m
ain
f
o
cu
s
s
h
o
u
ld
b
e
o
n
in
teg
r
atin
g
AI
alg
o
r
ith
m
s
to
im
p
r
o
v
e
d
etec
tio
n
m
et
h
o
d
s
.
T
h
e
m
o
d
el
s
h
o
u
ld
also
b
e
test
ed
o
n
d
if
f
er
en
t
cu
r
r
en
cies
f
r
o
m
d
if
f
er
en
t
c
o
u
n
tr
ies
in
o
r
d
er
to
p
a
r
ticip
ate
in
th
e
f
ig
h
t
ag
ain
s
t
co
u
n
ter
f
eit
m
o
n
ey
o
n
a
la
r
g
er
s
ca
le.
T
h
e
r
o
b
u
s
tn
ess
an
d
r
eliab
ilit
y
o
f
th
e
s
y
s
tem
ca
n
also
b
e
im
p
r
o
v
ed
b
y
in
co
r
p
o
r
atin
g
ad
v
a
n
ce
d
tr
an
s
f
o
r
m
er
m
o
d
els
an
d
r
ea
l
-
t
im
e
d
etec
tio
n
.
T
h
is
s
tu
d
y
p
a
v
es
th
e
way
f
o
r
in
n
o
v
atio
n
in
f
in
an
cial
s
ec
u
r
ity
,
s
u
p
p
o
r
tin
g
ec
o
n
o
m
ic
s
tab
ilit
y
an
d
en
h
a
n
cin
g
co
n
f
id
en
ce
in
g
lo
b
al
f
i
n
an
cial
s
y
s
tem
s
.
T
h
e
r
esu
lts
s
h
o
wed
th
e
ef
f
ec
tiv
en
ess
o
f
th
is
h
y
b
r
id
m
o
d
el,
a
n
d
with
p
r
o
p
er
s
ettin
g
s
an
d
in
c
r
ea
s
in
g
d
ata
an
d
d
iv
er
s
ity
,
th
e
m
o
d
el
will
b
e
m
o
r
e
p
o
wer
f
u
l a
n
d
ac
cu
r
at
e
in
d
etec
tin
g
c
o
u
n
ter
f
eit
cu
r
r
e
n
cies.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
d
ec
lar
e
th
at
t
h
er
e
is
n
o
f
u
n
d
in
g
i
n
v
o
lv
e
d
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Mo
n
th
er
T
a
r
awn
eh
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
d
ec
lar
e
th
at
th
e
r
e
is
n
o
co
n
f
lict o
f
in
ter
est r
eg
ar
d
in
g
th
e
p
u
b
licatio
n
o
f
th
is
p
a
p
er
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
u
s
ed
d
ataset
is
av
ailab
le
o
n
Kag
g
le
an
d
ca
n
b
e
ac
ce
s
s
ed
at
:
h
ttp
s
:
//w
w
w
.
ka
g
g
le.
co
m/d
a
ta
s
ets/
h
a
la
b
ilb
esei/jo
r
d
a
n
ia
n
b
a
n
kn
o
teso
ld
-
a
n
d
-
n
ew
-
ed
itio
n
T
h
e
d
ata
co
n
tain
s
im
ag
es
o
f
v
ar
io
u
s
J
o
r
d
an
ia
n
cu
r
r
en
cies,
w
h
ich
wer
e
u
s
ed
f
o
r
th
e
tr
ain
in
g
an
d
e
v
alu
atio
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
S
a
i
d
a
,
M
.
S
.
A
.
B
a
i
g
,
C
.
S
r
e
e
d
h
a
r
,
M
.
R
a
mes
h
,
R
.
S
p
o
o
r
t
i
,
a
n
d
K
.
K
a
v
e
r
i
,
“
D
e
t
e
c
t
i
o
n
o
f
c
o
u
n
t
e
r
f
e
i
t
c
u
r
r
e
n
c
y
,
”
i
n
Em
e
rg
i
n
g
T
ren
d
s
i
n
I
o
T
a
n
d
C
o
m
p
u
t
i
n
g
T
e
c
h
n
o
l
o
g
i
e
s
,
C
R
C
P
r
e
ss,
2
0
2
4
,
p
p
.
6
–
1
1
.
[
2
]
S
.
K
.
P
a
n
d
a
,
A
.
S
a
t
h
y
a
,
a
n
d
S
.
D
a
s,
“
B
i
t
c
o
i
n
:
B
e
g
i
n
n
i
n
g
o
f
t
h
e
c
r
y
p
t
o
c
u
r
r
e
n
c
y
e
r
a
,
”
i
n
Re
c
e
n
t
A
d
v
a
n
c
e
s
i
n
B
l
o
c
k
c
h
a
i
n
T
e
c
h
n
o
l
o
g
y
:
R
e
a
l
-
W
o
rl
d
Ap
p
l
i
c
a
t
i
o
n
s
,
S
p
r
i
n
g
e
r
,
2
0
2
3
,
p
p
.
2
5
–
5
8
.
[
3
]
M
.
S
W
E
ETY
a
n
d
M
.
R
.
S
A
LM
A
N
,
“
A
I
d
r
i
v
e
n
g
a
m
e
t
h
e
o
r
y
o
p
t
i
m
i
z
e
d
g
e
n
e
r
a
t
i
v
e
C
N
N
-
LSTM
met
h
o
d
f
o
r
f
a
k
e
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
,
”
J
o
u
rn
a
l
o
f
T
h
e
o
re
t
i
c
a
l
a
n
d
A
p
p
l
i
e
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
0
2
,
2
0
2
4
.
[
4
]
A
.
T.
G
.
Ta
p
e
h
a
n
d
M
.
Z.
N
a
ser,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
i
n
st
r
u
c
t
u
r
a
l
e
n
g
i
n
e
e
r
i
n
g
:
a
sci
e
n
t
o
m
e
t
r
i
c
s
r
e
v
i
e
w
o
f
t
r
e
n
d
s
a
n
d
b
e
s
t
p
r
a
c
t
i
c
e
s,”
Ar
c
h
i
v
e
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
Me
t
h
o
d
s
i
n
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
3
0
,
n
o
.
1
,
p
p
.
1
1
5
–
1
5
9
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
8
3
1
-
0
2
2
-
0
9
7
9
3
-
w.
[
5
]
M
.
T
a
r
a
w
n
e
h
,
F
.
A
l
Z
y
o
u
d
,
a
n
d
Y
.
S
h
a
r
r
a
b
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
t
r
a
f
f
i
c
a
n
a
l
y
si
s
f
r
a
mew
o
r
k
f
o
r
smar
t
c
i
t
i
e
s,
”
i
n
S
c
i
e
n
c
e
a
n
d
I
n
f
o
rm
a
t
i
o
n
C
o
n
f
e
re
n
c
e
,
2
0
2
3
,
p
p
.
6
9
9
–
7
1
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
a
r
tifi
cia
l in
tellig
en
ce
a
p
p
r
o
a
c
h
to
co
u
n
terf
eit
cu
r
r
en
cy
d
etec
tio
n
(
Mo
n
th
er Ta
r
a
w
n
eh
)
5813
[
6
]
M
.
Ta
r
a
w
n
e
h
,
H
.
A
b
d
A
l
w
a
h
e
d
,
a
n
d
F
.
A
l
Z
y
o
u
d
,
“
I
n
n
o
v
a
t
i
n
g
p
r
o
j
e
c
t
ma
n
a
g
e
m
e
n
t
:
A
I
a
p
p
l
i
c
a
t
i
o
n
s
f
o
r
s
u
c
c
e
ss
p
r
e
d
i
c
t
i
o
n
a
n
d
r
e
so
u
r
c
e
o
p
t
i
mi
z
a
t
i
o
n
,
”
i
n
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
A
d
v
a
n
c
e
s
i
n
C
o
m
p
u
t
i
n
g
Re
s
e
a
r
c
h
,
2
0
2
4
,
p
p
.
3
8
2
–
3
9
1
.
[
7
]
O
.
Ta
r
a
w
n
e
h
,
M
.
Ta
r
a
w
n
e
h
,
Y
.
S
h
a
r
r
a
b
,
a
n
d
M
.
H
u
s
n
i
,
“
M
u
sh
r
o
o
m
c
l
a
s
si
f
i
c
a
t
i
o
n
u
s
i
n
g
m
a
c
h
i
n
e
-
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,”
i
n
AI
P
C
o
n
f
e
re
n
c
e
Pr
o
c
e
e
d
i
n
g
s
,
2
0
2
3
,
p
.
0
3
0
0
0
3
.
[
8
]
L.
B
o
r
d
o
n
i
a
n
d
F
.
M
e
l
e
,
Ar
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
f
o
r c
u
l
t
u
ra
l
h
e
r
i
t
a
g
e
.
C
a
mb
r
i
d
g
e
S
c
h
o
l
a
r
s Pu
b
l
i
s
h
i
n
g
,
2
0
1
6
.
[
9
]
Y
.
Q
i
n
,
Z
.
X
u
,
X
.
W
a
n
g
,
a
n
d
M
.
S
k
a
r
e
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
a
n
d
e
c
o
n
o
mi
c
d
e
v
e
l
o
p
me
n
t
:
A
n
e
v
o
l
u
t
i
o
n
a
r
y
i
n
v
e
s
t
i
g
a
t
i
o
n
a
n
d
sy
st
e
ma
t
i
c
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
t
h
e
K
n
o
w
l
e
d
g
e
E
c
o
n
o
m
y
,
v
o
l
.
1
5
,
p
p
.
1
7
3
6
–
1
7
7
0
,
2
0
2
4
.
[
1
0
]
U
.
S
a
d
y
k
,
R
.
B
a
i
mu
k
a
sh
e
v
,
a
n
d
C
.
Tu
r
a
n
,
“
S
t
a
t
e
-
of
-
t
h
e
-
a
r
t
r
e
v
i
e
w
o
f
d
e
e
p
l
e
a
r
n
i
n
g
me
t
h
o
d
s
i
n
f
a
k
e
b
a
n
k
n
o
t
e
r
e
c
o
g
n
i
t
i
o
n
p
r
o
b
l
e
m,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
\
& A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
5
,
2
0
2
4
.
[
1
1
]
D
.
G
u
p
t
a
,
N
.
K
.
M
i
r
y
a
l
a
,
a
n
d
A
.
S
r
i
v
a
st
a
v
a
,
“
Le
v
e
r
a
g
i
n
g
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
f
o
r
c
o
u
n
t
e
r
i
n
g
f
i
n
a
n
c
i
a
l
c
r
i
m
e
s,”
J
o
u
r
n
a
l
I
D
,
v
o
l
.
2
1
5
7
,
p
.
1
7
8
,
2
0
2
3
.
[
1
2
]
O
.
Y
.
R
o
d
i
o
n
o
v
a
a
n
d
A
.
P
o
mera
n
t
s
e
v
,
“
N
I
R
-
b
a
se
d
a
p
p
r
o
a
c
h
t
o
c
o
u
n
t
e
r
f
e
i
t
-
d
r
u
g
d
e
t
e
c
t
i
o
n
,
”
T
rAC
T
re
n
d
s
i
n
A
n
a
l
y
t
i
c
a
l
C
h
e
m
i
s
t
ry
,
v
o
l
.
2
9
,
p
p
.
7
9
5
–
8
0
3
,
2
0
1
0
.
[
1
3
]
M
.
R
.
M
e
i
r
e
l
e
s
,
P
.
E.
A
l
me
i
d
a
,
a
n
d
M
.
G
.
S
i
mõ
e
s,
“
A
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
f
o
r
i
n
d
u
st
r
i
a
l
a
p
p
l
i
c
a
b
i
l
i
t
y
o
f
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s,”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
I
n
d
u
st
r
i
a
l
E
l
e
c
t
r
o
n
i
c
s
,
v
o
l
.
5
0
,
p
p
.
5
8
5
–
6
0
1
,
2
0
0
3
.
[
1
4
]
S
.
Z
h
u
,
T
.
Y
u
,
T.
X
u
,
H
.
C
h
e
n
,
S
.
D
u
st
d
a
r
,
a
n
d
S
.
G
i
g
a
n
,
“
I
n
t
e
l
l
i
g
e
n
t
c
o
mp
u
t
i
n
g
:
t
h
e
l
a
t
e
s
t
a
d
v
a
n
c
e
s,
c
h
a
l
l
e
n
g
e
s,
a
n
d
f
u
t
u
r
e
,
”
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
2
,
p
.
6
,
2
0
2
3
.
[
1
5
]
O
.
T
a
r
a
w
n
e
h
,
M
.
O
t
a
i
r
,
M
.
H
u
sn
i
,
H
.
Y
.
A
b
u
a
d
d
o
u
s
,
M
.
T
a
r
a
w
n
e
h
,
a
n
d
M
.
A
.
A
l
m
o
ma
n
i
,
“
B
r
e
a
s
t
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
d
e
c
i
si
o
n
t
r
e
e
a
l
g
o
r
i
t
h
ms,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
3
,
n
o
.
4
,
p
p
.
6
7
6
–
6
8
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
2
.
0
1
3
0
4
7
8
.
[
1
6
]
M
.
T
a
r
a
w
n
e
h
a
n
d
O
.
E
mb
a
r
a
k
,
“
H
y
b
r
i
d
a
p
p
r
o
a
c
h
f
o
r
h
e
a
r
t
d
i
s
e
a
se
p
r
e
d
i
c
t
i
o
n
u
si
n
g
d
a
t
a
mi
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
L
e
c
t
u
r
e
N
o
t
e
s
o
n
D
a
t
a
En
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
T
e
c
h
n
o
l
o
g
i
e
s
,
v
o
l
.
2
9
,
p
p
.
4
4
7
–
4
5
4
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
030
-
1
2
8
3
9
-
5
_
4
1
.
[
1
7
]
S
.
Ta
t
i
n
e
n
i
a
n
d
A
.
M
u
st
y
a
l
a
,
“
E
n
h
a
n
c
i
n
g
f
i
n
a
n
c
i
a
l
s
e
c
u
r
i
t
y
:
d
a
t
a
sc
i
e
n
c
e
’
s
r
o
l
e
i
n
r
i
s
k
ma
n
a
g
e
m
e
n
t
a
n
d
f
r
a
u
d
d
e
t
e
c
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
m
e
n
t
s i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
,
p
p
.
2
5
8
3
–
8
6
2
8
,
2
0
2
4
.
[
1
8
]
S
.
K
o
e
h
l
e
r
,
N
.
D
h
a
m
e
l
i
y
a
,
B
.
P
a
t
e
l
,
S
.
K
u
mar
,
a
n
d
R
.
A
n
u
ma
n
d
l
a
,
“
A
I
-
e
n
h
a
n
c
e
d
c
r
y
p
t
o
c
u
r
r
e
n
c
y
t
r
a
d
i
n
g
a
l
g
o
r
i
t
h
m
f
o
r
o
p
t
i
mal
i
n
v
e
s
t
me
n
t
st
r
a
t
e
g
i
e
s,
”
As
i
a
n
Ac
c
o
u
n
t
i
n
g
a
n
d
Au
d
i
t
i
n
g
A
d
v
a
n
c
e
m
e
n
t
,
v
o
l
.
9
,
n
o
.
1
,
2
0
1
8
.
[
1
9
]
M
a
n
a
s
W
a
k
c
h
a
u
r
e
,
B
.
K
.
P
a
t
l
e
,
a
n
d
A
.
K
.
M
a
h
i
n
d
r
a
k
a
r
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
A
I
t
e
c
h
n
i
q
u
e
s
a
n
d
r
o
b
o
t
i
c
s
i
n
a
g
r
i
c
u
l
t
u
r
e
:
A
r
e
v
i
e
w
,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
i
n
t
h
e
L
i
f
e
S
c
i
e
n
c
e
s
,
v
o
l
.
1
0
0
0
5
7
,
2
0
2
3
.
[
2
0
]
M
.
La
a
v
a
n
y
a
a
n
d
V
.
V
i
j
a
y
a
r
a
g
h
a
v
a
n
,
“
R
e
a
l
t
i
me
f
a
k
e
c
u
r
r
e
n
c
y
n
o
t
e
d
e
t
e
c
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
En
g
i
n
e
e
ri
n
g
a
n
d
A
d
v
a
n
c
e
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
9
,
n
o
.
1
s
5
,
p
p
.
9
5
–
9
8
,
2
0
1
9
,
d
o
i
:
1
0
.
3
5
9
4
0
/
i
j
e
a
t
.
a
1
0
0
7
.
1
2
9
1
s
5
1
9
.
[
2
1
]
Y
.
Le
C
u
n
,
Y
.
B
e
n
g
i
o
,
a
n
d
G
.
H
i
n
t
o
n
,
“
D
e
e
p
l
e
a
r
n
i
n
g
,
”
N
a
t
u
re
,
v
o
l
.
5
2
1
,
p
p
.
4
3
6
–
4
4
4
,
2
0
1
5
.
[
2
2
]
K
.
A
l
i
,
M
.
A
l
z
a
i
d
i
,
D
.
A
l
-
F
r
a
i
h
a
t
,
a
n
d
A
.
M
.
E
l
a
m
i
r
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
:
b
e
n
e
f
i
t
s,
a
p
p
l
i
c
a
t
i
o
n
,
e
t
h
i
c
a
l
i
ss
u
e
s,
a
n
d
o
r
g
a
n
i
z
a
t
i
o
n
a
l
r
e
s
p
o
n
s
e
s,”
2
0
2
3
,
p
p
.
6
8
5
–
7
0
2
.
[
2
3
]
D
.
A
l
-
F
r
a
i
h
a
t
,
Y
.
S
h
a
r
r
a
b
,
F
.
A
l
z
y
o
u
d
,
A
.
Q
a
h
mas
h
,
M
.
Ta
r
a
w
n
e
h
,
a
n
d
A
.
M
a
a
i
t
a
,
“
S
p
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
u
t
i
l
i
z
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
:
A
sy
st
e
ma
t
i
c
r
e
v
i
e
w
o
f
t
h
e
l
a
t
e
st
d
e
v
e
l
o
p
m
e
n
t
s,
”
H
u
m
a
n
-
c
e
n
t
ri
c
C
o
m
p
u
t
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
1
4
,
2
0
2
4
,
d
o
i
:
1
0
.
2
2
9
6
7
/
H
C
I
S
.
2
0
2
4
.
1
4
.
0
1
5
.
[
2
4
]
M
.
M
.
A
l
i
m
u
s
h
w
a
n
,
A
.
M
o
h
a
i
m
i
n
,
R
.
I
sl
a
m
,
a
n
d
S
.
C
h
o
w
d
h
u
r
y
,
“
F
a
k
e
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
u
s
i
n
g
i
m
a
g
e
p
r
o
c
e
ssi
n
g
me
t
h
o
d
,
”
i
n
I
O
P
C
o
n
f
e
re
n
c
e
S
e
r
i
e
s:
M
a
t
e
ri
a
l
s
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
2
0
1
6
,
p
.
0
5
2
0
4
7
.
[
2
5
]
Z.
A
h
m
e
d
,
S
.
Y
a
smi
n
,
M
.
N
.
I
sl
a
m,
a
n
d
R
.
U
.
A
h
me
d
,
“
I
mag
e
p
r
o
c
e
ss
i
n
g
b
a
s
e
d
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
o
f
B
a
n
g
l
a
d
e
s
h
i
b
a
n
k
n
o
t
e
s,
”
i
n
T
h
e
8
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
o
f
t
w
a
re
,
K
n
o
w
l
e
d
g
e
,
I
n
f
o
rm
a
t
i
o
n
Ma
n
a
g
e
m
e
n
t
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
(
S
K
I
MA
2
0
1
4
)
,
2
0
1
4
,
p
p
.
1
–
8.
[
2
6
]
S
.
C
.
D
.
B
a
n
d
u
,
M
.
K
a
k
i
l
e
t
i
,
S
.
S
.
J.
S
o
l
o
m
a
n
,
a
n
d
N
.
B
a
y
d
e
t
i
,
“
I
n
d
i
a
n
f
a
k
e
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
u
s
i
n
g
i
ma
g
e
p
r
o
c
e
ssi
n
g
a
n
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
p
p
.
1
–
1
4
,
2
0
2
4
.
[
2
7
]
A
.
P
a
t
h
a
k
,
A
.
C
h
a
k
r
a
b
o
r
t
y
,
M
.
R
a
h
a
man
,
T
.
S
.
R
a
f
a
,
a
n
d
U
.
N
a
y
e
ma
,
“
E
n
h
a
n
c
e
d
c
o
u
n
t
e
r
f
e
i
t
d
e
t
e
c
t
i
o
n
o
f
B
a
n
g
l
a
d
e
sh
c
u
r
r
e
n
c
y
t
h
r
o
u
g
h
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s:
a
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
n
o
v
a
t
i
v
e
R
e
se
a
rc
h
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
\
&
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
2
,
p
p
.
1
0
–
2
0
,
2
0
2
4
.
[
2
8
]
R
.
S
u
m
a
l
a
t
h
a
,
B
.
J.
R
e
d
d
y
,
a
n
d
T.
V
.
R
.
R
e
d
d
y
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
f
a
k
e
I
n
d
i
a
n
c
u
r
r
e
n
c
y
u
s
i
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
2
2
6
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
i
n
g
Me
t
h
o
d
o
l
o
g
i
e
s
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
(
I
C
C
M
C
)
,
2
0
2
2
,
p
p
.
1
6
1
9
–
1
6
2
3
.
[
2
9
]
J.
D
’
c
r
u
z
,
M
.
J
o
se
,
M
.
E
l
d
h
o
se
,
a
n
d
B
.
J
o
se
,
“
F
a
k
e
I
n
d
i
a
n
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
ri
n
g
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
5
,
p
p
.
2
1
4
3
–
2
4
5
5
,
2
0
2
1
.
[
3
0
]
C
.
G
.
P
a
c
h
ó
n
,
D
.
M
.
B
a
l
l
e
s
t
e
r
o
s
,
a
n
d
D
.
R
e
n
z
a
,
“
F
a
k
e
b
a
n
k
n
o
t
e
r
e
c
o
g
n
i
t
i
o
n
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
1
,
p
.
1
2
8
1
,
2
0
2
1
.
[
3
1
]
A
.
N
a
sa
y
r
e
h
,
A
.
S
.
Jara
d
a
t
,
H
.
G
h
a
r
a
i
b
e
h
,
W
.
D
a
w
a
g
h
r
e
h
,
R
.
M
.
A
l
M
a
m
l
o
o
k
,
a
n
d
Y
.
A
l
q
u
d
a
h
,
“
Jo
r
d
a
n
i
a
n
b
a
n
k
n
o
t
e
d
a
t
a
r
e
c
o
g
n
i
t
i
o
n
:
A
C
N
N
-
b
a
se
d
a
p
p
r
o
a
c
h
w
i
t
h
a
t
t
e
n
t
i
o
n
mec
h
a
n
i
sm,”
J
o
u
rn
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
-
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
6
,
p
.
1
0
2
0
3
8
,
2
0
2
4
.
[
3
2
]
M
.
M
e
l
a
k
u
,
“
E
t
h
i
o
p
i
a
n
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
a
n
d
c
o
u
n
t
e
r
f
e
i
t
v
e
r
i
f
i
c
a
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
M
a
s
t
e
r
T
h
e
s
i
s,
S
t
.
M
a
r
y
’
s Un
i
v
e
r
s
i
t
y
,
2
0
2
3
.
[
3
3
]
S
.
A
.
N
a
s
e
e
m
,
A
.
R
e
h
m
a
n
,
S
.
Z.
U
d
d
i
n
,
B
.
K
h
a
n
,
Z
.
M
e
h
m
o
o
d
,
a
n
d
M
.
U
.
N
i
sa,
“
C
o
u
n
t
e
r
f
e
i
t
r
e
c
o
g
n
i
t
i
o
n
o
f
P
a
k
i
s
t
a
n
i
c
u
r
r
e
n
c
y
,
”
K
I
ET J
o
u
rn
a
l
o
f
C
o
m
p
u
t
i
n
g
a
n
d
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
6
,
p
p
.
1
2
3
–
1
4
7
,
2
0
2
3
.
[
3
4
]
G
.
B
e
h
e
r
y
,
H
.
El
-
H
a
d
i
d
i
,
A
.
El
-
H
a
r
b
y
,
a
n
d
T
.
A
.
U
sam
a
d
,
“
I
n
t
e
l
l
i
g
e
n
t
Li
b
y
a
n
b
a
n
k
n
o
t
e
r
e
c
o
g
n
i
t
i
o
n
sy
s
t
e
m,”
I
n
t
e
rn
a
t
i
o
n
a
l
Re
se
a
rc
h
J
o
u
r
n
a
l
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
8
,
p
p
.
1
6
6
8
–
1
6
8
0
,
2
0
2
1
.
[
3
5
]
M
.
N
.
H
a
s
a
n
,
R
.
N
.
To
ma,
A
.
-
A
.
N
a
h
i
d
,
M
.
M
.
I
sl
a
m
,
a
n
d
J.
-
M
.
K
i
m,
“
El
e
c
t
r
i
c
i
t
y
t
h
e
f
t
d
e
t
e
c
t
i
o
n
i
n
smar
t
g
r
i
d
s
y
st
e
ms:
A
C
N
N
-
LSTM
b
a
se
d
a
p
p
r
o
a
c
h
,
”
E
n
e
r
g
i
e
s
,
v
o
l
.
1
2
,
p
.
3
3
1
0
,
2
0
1
9
.
[
3
6
]
F
.
A
n
t
o
n
i
u
s,
J
.
R
a
m
u
,
P
.
S
a
si
k
a
l
a
,
J.
S
e
k
h
a
r
,
a
n
d
S
.
S
.
C
.
M
a
r
y
,
“
D
e
e
p
C
y
b
e
r
D
e
t
e
c
t
:
H
y
b
r
i
d
A
I
f
o
r
c
o
u
n
t
e
r
f
e
i
t
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
w
i
t
h
G
A
N
-
C
N
N
-
R
N
N
u
si
n
g
A
f
r
i
c
a
n
B
u
f
f
a
l
o
o
p
t
i
mi
z
a
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
4
,
2
0
2
3
.
[
3
7
]
V
.
K
u
k
r
e
j
a
,
S
.
M
e
h
t
a
,
S
.
G
u
p
t
a
,
a
n
d
A
.
G
a
r
g
,
“
F
i
n
a
n
c
i
a
l
f
o
r
e
si
g
h
t
:
p
r
e
d
i
c
t
i
v
e
p
o
w
e
r
o
f
C
N
N
-
S
V
M
i
n
f
a
k
e
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
a
n
a
l
y
si
s
,
”
i
n
2
0
2
4
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
f
o
r E
m
e
rg
i
n
g
T
e
c
h
n
o
l
o
g
y
(
I
N
C
ET)
,
2
0
2
4
,
p
p
.
1
–
6.
[
3
8
]
T.
A
l
i
,
S
.
Ja
n
,
A
.
A
l
k
h
o
d
r
e
,
M
.
N
a
u
m
a
n
,
M
.
A
mi
n
,
a
n
d
M
.
S
.
S
i
d
d
i
q
u
i
,
“
D
e
e
p
M
o
n
e
y
:
c
o
u
n
t
e
r
f
e
i
t
mo
n
e
y
d
e
t
e
c
t
i
o
n
u
s
i
n
g
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s
,
”
Pe
e
rJ
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
5
,
p
.
e
2
1
6
,
2
0
1
9
.
[
3
9
]
S
.
K
o
r
e
,
D
.
M
i
s
h
r
a
,
I
.
B
h
i
o
g
a
d
e
,
a
n
d
D
.
Ji
t
u
r
i
,
“
F
a
k
e
c
u
r
r
e
n
c
y
d
e
t
e
c
t
i
o
n
u
si
n
g
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
s
(
R
N
N
)
,
”
J
o
u
r
n
a
l
F
o
r
Ba
si
c
S
c
i
e
n
c
e
s
,
v
o
l
.
2
3
,
n
o
.
6
,
2
0
2
3
.
[
4
0
]
J.
A
.
N
a
si
r
,
O
.
S
.
K
h
a
n
,
a
n
d
I
.
V
a
r
l
a
m
i
s,
“
F
a
k
e
n
e
w
s
d
e
t
e
c
t
i
o
n
:
A
h
y
b
r
i
d
C
N
N
-
R
N
N
b
a
se
d
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
Ma
n
a
g
e
m
e
n
t
D
a
t
a
I
n
si
g
h
t
s
,
v
o
l
.
1
,
n
o
.
1
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
j
i
me
i
.
2
0
2
0
.
1
0
0
0
0
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.