T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
2
,
A
p
r
il
2
0
2
1
,
p
p
.
4
3
8
~
4
4
3
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
:
1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i
2
.
1
6
1
3
4
438
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Tra
nsla
ting cunei
form sy
mbo
ls usi
ng
artificial n
eur
a
l net
wo
rk
Arw
a
H
a
m
ed
Sa
lih
H
a
m
da
ny
,
Ra
id Ra
f
i O
m
a
r
Al
-
Nima
,
L
ub
a
b H
.
Alba
k
Tec
h
n
ica
l
En
g
in
e
e
ri
n
g
C
o
ll
e
g
e
o
f
M
o
su
l
,
No
r
th
e
rn
Tec
h
n
ica
l
Un
iv
e
rsity
,
Ira
q
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
r
2
4
,
2
0
2
0
R
ev
is
ed
Ap
r
1
5
,
2
0
2
0
Acc
ep
ted
Ap
r
3
0
,
2
0
2
0
Cu
n
e
ifo
rm
lan
g
u
a
g
e
is
a
n
o
l
d
la
n
g
u
a
g
e
th
a
t
wa
s
in
v
e
n
ted
b
y
th
e
p
e
o
p
le
o
f
S
u
m
e
rian
n
a
ti
o
n
.
It
is
a
n
e
ss
e
n
ti
a
l
lan
g
u
a
g
e
fo
r
m
a
n
y
a
rc
h
e
o
l
o
g
ists.
Esp
e
c
ially
wh
o
a
re
in
tere
ste
d
in
stu
d
y
i
n
g
a
n
d
i
n
v
e
sti
g
a
ti
n
g
t
h
e
o
ld
n
a
ti
o
n
s
o
f
Ira
q
.
De
a
li
n
g
wit
h
t
h
is
t
y
p
e
o
f
lan
g
u
a
g
e
u
su
a
l
ly
re
q
u
ires
s
p
e
c
ialist
to
t
ra
n
sla
te
it
s
sy
m
b
o
ls,
wh
ich
a
re
b
a
sic
a
ll
y
fo
r
m
s
o
f
n
a
il
sh
a
p
e
s.
Th
is
st
u
d
y
p
re
se
n
ts
a
n
e
w
a
p
p
ro
a
c
h
to
tra
n
sla
te
th
e
c
u
n
e
ifo
rm
writi
n
g
b
y
e
m
p
lo
y
in
g
a
rti
fi
c
ial
n
e
u
ra
l
n
e
two
rk
(AN
N)
tec
h
n
i
q
u
e
.
Eff
e
c
ti
v
e
ly
,
m
u
lt
i
-
la
y
e
r
p
e
r
c
e
p
tro
n
(
M
LP
)
n
e
u
ra
l
n
e
two
rk
h
a
s
b
e
e
n
a
d
a
p
ted
fo
r
tr
a
n
sla
ti
n
g
t
h
e
S
u
m
e
rian
c
u
n
e
if
o
r
m
sy
m
b
o
l
ima
g
e
s
to
th
e
ir
c
o
rre
sp
o
n
d
i
n
g
En
g
li
sh
letters
.
Th
is
w
o
rk
h
a
s
b
e
e
n
s
u
c
c
e
ss
fu
ll
y
e
sta
b
li
sh
e
d
a
n
d
it
a
t
tain
e
d
1
0
0
%
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
C
u
n
eif
o
r
m
s
y
m
bo
ls
Mu
lti
-
lay
er
p
er
ce
p
tr
o
n
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
:
Ar
wa
Ham
ed
Salih
Ham
d
an
y
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
No
r
th
er
n
T
ec
h
n
ical
Un
iv
e
r
s
ity
Mo
s
u
l 4
1
0
0
2
,
I
r
aq
E
m
ail: a
r
wah
am
id
7
8
@
n
tu
.
e
d
u
.
iq
1.
I
NT
RO
D
UCT
I
O
N
C
u
n
eif
o
r
m
la
n
g
u
a
g
e
is
o
n
e
o
f
th
e
o
ld
est
lan
g
u
ag
es
i
n
th
e
wo
r
ld
.
I
t
was
s
tar
ed
in
3
0
0
0
B
.
C
.
,
wh
er
e
it
was
in
v
en
ted
b
y
a
n
o
ld
civ
iliz
atio
n
in
I
r
aq
.
T
h
is
civ
ilizatio
n
was
k
n
o
wn
as
Su
m
er
.
C
u
n
eif
o
r
m
s
y
m
b
o
ls
wer
e
th
e
Su
m
er
ian
wr
itin
g
s
ty
le.
T
h
ey
wer
e
ef
f
ec
tiv
ely
u
s
ed
to
r
e
p
o
r
t
ev
en
ts
,
ac
tio
n
s
an
d
o
th
e
r
in
f
o
r
m
atio
n
th
at
wer
e
p
r
ev
io
u
s
ly
h
ap
p
e
n
ed
[
1
]
.
I
n
its
f
ir
s
t p
icto
g
r
ap
h
ic
s
tag
es,
it wa
s
lar
g
ely
co
n
s
is
ted
o
f
r
e
b
u
s
wr
itin
g
o
f
n
o
u
n
s
.
B
y
2
5
0
0
B
.
C
.
,
th
e
s
cr
ib
es
p
lace
d
th
e
cu
n
eif
o
r
m
s
ig
n
s
in
t
o
co
r
r
e
ct
o
r
d
er
s
.
T
h
en
,
th
e
ea
r
lies
t
tex
ts
wer
e
b
ein
g
m
o
r
e
s
tr
u
ctu
r
ed
[
2
]
.
C
u
n
eif
o
r
m
wr
itin
g
s
y
s
tem
is
s
u
b
jecte
d
to
m
a
n
y
s
tag
es
o
f
d
e
v
elo
p
m
en
t
to
f
ac
ilit
ate
its
ch
ar
ac
ter
is
ti
cs
ab
o
u
t
th
e
s
h
ap
e
o
f
s
y
m
b
o
ls
an
d
n
u
m
b
er
s
th
at
r
ep
r
esen
t
th
e
d
ev
elo
p
m
en
t
s
tate
o
f
o
ld
Su
m
er
ian
s
cr
ip
lan
g
u
ag
e
to
B
ab
y
lo
n
ia
n
a
n
d
Ass
y
r
ian
cu
n
eif
o
r
m
lan
g
u
ag
es.
At
th
e
b
eg
i
n
n
in
g
o
f
th
e
1
9
t
h
ce
n
tu
r
y
,
th
o
u
s
an
d
s
o
f
cu
n
eif
o
r
m
tab
lets
wer
e
d
is
co
v
er
e
d
in
I
r
aq
.
T
h
e
y
r
ep
r
esen
t
v
ar
io
u
s
Ass
y
r
ian
an
d
B
ab
y
lo
n
ian
wr
itin
g
s
.
T
o
d
a
y
m
a
n
y
cu
n
eif
o
r
m
tab
lets
ex
is
t
in
m
an
y
m
u
s
eu
m
s
.
No
ticea
b
ly
,
t
h
e
p
r
o
ce
s
s
o
f
tr
an
s
latin
g
th
e
cu
n
eif
o
r
m
s
y
m
b
o
ls
r
eq
u
ir
es
ex
p
er
ie
n
ce
an
d
tim
e
.
Ho
wev
er
,
th
e
n
ee
d
o
f
in
f
o
r
m
atio
n
tech
n
o
lo
g
y
is
r
eq
u
ir
e
d
to
a
d
d
r
ess
th
ese
p
r
o
b
lem
s
[
3
]
.
T
h
e
aim
o
f
th
is
s
tu
d
y
is
tr
an
s
latin
g
th
e
cu
n
eif
o
r
m
s
y
m
b
o
ls
o
f
Su
m
er
ian
w
r
itin
g
in
to
E
n
g
lis
h
letter
s
.
T
h
e
ANN
is
em
p
lo
y
ed
to
p
r
o
v
i
d
e
in
tellig
en
t
tr
a
n
s
latin
g
b
etwe
en
th
e
two
la
n
g
u
ag
es.
Af
te
r
th
e
in
tr
o
d
u
ctio
n
,
th
e
r
em
ain
in
g
s
ec
tio
n
s
will
b
e
o
r
g
an
ize
d
as
f
o
llo
ws:
s
ec
tio
n
2
r
ev
iews
p
r
io
r
wo
r
k
,
s
ec
tio
n
3
illu
s
tr
ates
th
e
m
eth
o
d
o
lo
g
y
o
f
th
is
wo
r
k
,
s
ec
tio
n
4
d
is
cu
s
s
es
th
e
p
r
ac
tical
r
esu
lts
an
d
s
ec
tio
n
5
co
n
clu
d
es
th
e
p
ap
er
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Tr
a
n
s
la
tin
g
cu
n
eifo
r
m
s
ymb
o
ls
u
s
in
g
a
r
tifi
cia
l n
eu
r
a
l n
etw
o
r
k
(
A
r
w
a
Ha
med
S
a
lih
Ha
md
a
n
y
)
439
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
I
n
th
e
liter
atu
r
e,
v
er
y
f
ew
wo
r
k
s
wer
e
co
n
s
id
er
ed
tr
an
s
l
atin
g
cu
n
eif
o
r
m
s
y
m
b
o
ls
u
s
in
g
m
o
d
e
r
n
an
aly
s
is
m
eth
o
d
s
s
u
ch
as
ar
tifi
cial
in
tellig
en
ce
(
AI
)
tech
n
i
q
u
es.
I
n
2
0
0
0
,
Su
laim
an
ex
p
lain
e
d
th
e
Su
m
er
ian
a
n
d
Aca
d
ian
wr
itin
g
s
ty
le
ac
co
r
d
in
g
to
th
e
o
b
tain
e
d
ex
p
e
r
tis
e
[
1
]
.
I
t
s
ee
m
s
th
at
m
an
u
al
s
ty
le
was
u
s
ed
f
o
r
tr
a
n
s
latin
g
th
e
Su
m
er
ian
an
d
Aca
d
ian
w
r
itin
g
s
to
Ar
ab
ic
lan
g
u
a
g
e.
I
n
2
0
0
7
,
Po
s
tg
ate
ed
ited
a
g
r
o
u
p
o
f
i
n
f
o
r
m
atio
n
r
eg
ar
d
in
g
I
r
a
q
i
lan
g
u
a
g
es.
Su
m
er
lan
g
u
a
g
e
was
o
n
e
o
f
th
is
in
f
o
r
m
atio
n
.
Usef
u
l
illu
s
tr
atio
n
s
wer
e
p
r
esen
ted
f
o
r
Su
m
er
ian
wr
itin
g
s
u
ch
as sy
n
t
ax
,
p
h
o
n
o
lo
g
y
,
lex
ical
ca
teg
o
r
ies,
n
o
m
in
als,
ad
jectiv
es,
p
r
o
n
o
u
n
s
an
d
v
e
r
b
s
[
3
]
.
Ag
ain
,
th
e
lan
g
u
ag
e
was m
an
u
ally
an
aly
s
ed
b
ased
o
n
t
h
e
o
b
tain
ed
ex
p
er
tis
e.
I
n
2
0
1
0
,
Yu
s
h
u
h
ig
h
lig
h
te
d
h
o
w
th
e
in
v
en
tio
n
o
f
wr
i
tin
g
was
co
n
s
id
er
e
d
b
y
Su
m
er
ian
[
4
]
.
I
t
ap
p
ea
r
s
th
at
m
an
u
al
tr
an
s
latin
g
was
also
u
tili
ze
d
in
th
is
s
tu
d
y
.
I
n
2
0
1
7
,
Ak
tas
an
d
Asu
r
o
g
lu
p
r
o
p
o
s
ed
a
s
tu
d
y
f
o
r
r
ea
d
in
g
cu
n
eif
o
r
m
s
ig
n
s
b
y
ex
p
lo
itin
g
co
m
p
u
ter
tech
n
i
q
u
es.
B
asically
,
th
e
cu
n
eif
o
r
m
s
ig
n
s
o
f
Hittit
e
w
r
itin
g
we
r
e
u
s
ed
.
Fu
r
th
er
m
o
r
e
,
d
ata
m
in
in
g
o
f
clu
s
ter
in
g
an
d
class
if
icatio
n
alg
o
r
ith
m
s
wer
e
em
p
lo
y
ed
[
5
]
.
O
b
v
io
u
s
ly
,
Su
m
e
r
ian
wr
itin
g
s
ty
le
d
id
n
o
t c
o
n
s
id
er
i
n
th
is
wo
r
k
.
I
n
2
0
1
9
,
Saeid
et
a
l
.
,
em
p
lo
y
ed
th
e
s
u
p
p
o
r
t v
ec
t
o
r
m
ac
h
in
e
(
SVM)
f
o
r
r
ec
o
g
n
izin
g
t
h
e
c
u
n
eif
o
r
m
letter
s
.
I
m
ag
e
p
r
o
ce
s
s
in
g
s
tep
s
wer
e
im
p
lem
e
n
ted
b
ef
o
r
e
th
e
SVM
[
2
]
.
T
h
is
s
tu
d
y
co
n
ce
n
tr
ated
o
n
r
ec
o
g
n
i
zin
g
(
n
o
t tr
an
s
latin
g
)
th
e
cu
n
ei
f
o
r
m
letter
s
.
I
n
th
e
s
am
e
y
ea
r
,
B
o
r
n
et
a
l.
,
illu
s
tr
ated
an
attem
p
t
o
f
u
tili
zi
n
g
m
eth
o
d
s
f
r
o
m
ca
lcu
latio
n
al
lin
g
u
is
tic
s
to
an
aly
s
e
s
cr
ip
ts
o
f
u
n
d
ec
ip
h
er
ed
p
r
o
to
-
E
lam
ite.
Hier
ar
ch
ical
clu
s
ter
in
g
,
n
-
g
r
a
m
f
r
e
q
u
en
cies
an
d
laten
t
d
ir
ich
let
allo
ca
tio
n
(
L
DA)
to
p
ic
m
o
d
els
wer
e
em
p
lo
y
ed
.
R
esu
lts
wer
e
ac
h
iev
ed
b
y
r
ev
ea
lin
g
p
r
ev
io
u
s
ly
-
u
n
o
b
s
er
v
ed
r
elatio
n
s
h
ip
s
o
f
s
ig
n
s
an
d
m
a
n
u
al
d
e
cip
h
er
in
g
[
6
]
.
Her
e,
clu
s
ter
in
g
d
if
f
er
en
t
s
ig
n
letter
s
wer
e
p
r
o
v
id
e
d
.
I
t
ca
n
b
e
in
v
esti
g
ated
th
at
th
er
e
was
n
o
co
n
s
id
er
atio
n
o
n
tr
a
n
s
latin
g
th
e
Su
m
er
ian
wr
itin
g
s
y
m
b
o
ls
to
E
n
g
lis
h
letter
s
b
y
u
s
in
g
th
e
ANN
tech
n
iq
u
e
i
n
p
r
io
r
wo
r
k
.
T
h
is
p
ap
e
r
will
a
d
d
r
ess
th
is
g
ap
a
n
d
p
r
o
v
id
e
an
im
p
o
r
tan
t c
o
n
tr
ib
u
t
io
n
in
th
is
m
atter
.
3.
T
H
E
P
RO
P
O
SE
D
M
E
T
H
O
D
I
n
th
is
s
tu
d
y
,
an
a
r
tific
ial
in
tellig
en
ce
(
AI
)
tech
n
iq
u
e
o
f
m
u
lti
-
la
y
e
r
p
er
ce
p
tr
o
n
(
MLP
)
n
eu
r
al
n
et
w
o
r
k
h
as
b
ee
n
ad
ap
ted
f
o
r
tr
an
s
latin
g
th
e
im
ag
es o
f
Su
m
er
ian
cu
n
eif
o
r
m
s
y
m
b
o
ls
in
to
E
n
g
lis
h
letter
s
.
T
h
e
k
ey
id
e
a
o
f
o
u
r
p
r
o
p
o
s
ed
ap
p
r
o
a
ch
is
to
co
llect
an
y
cu
n
eif
o
r
m
s
y
m
b
o
l
as
im
ag
e
an
d
p
r
o
d
u
c
e
an
in
d
icato
r
f
o
r
its
co
r
r
esp
o
n
d
in
g
E
n
g
lis
h
letter
.
Acc
o
r
d
in
g
ly
,
E
n
g
lis
h
letter
s
ca
n
in
tellig
en
tly
b
e
g
en
er
ate
d
f
r
o
m
cu
n
eif
o
r
m
s
y
m
b
o
l
im
ag
es.
ANN
o
f
m
u
ltip
le
o
u
tp
u
ts
,
as
in
[
7
-
1
6
]
,
h
as
b
ee
n
f
o
u
n
d
to
b
e
u
s
ef
u
l
in
o
u
r
ca
s
e.
Fig
u
r
e
1
illu
s
tr
ates
th
e
g
en
er
al
f
o
r
m
o
f
o
u
r
s
u
g
g
ested
in
tellig
en
t a
p
p
r
o
ac
h
.
Fig
u
r
e
1
.
T
h
e
g
e
n
e
r
al
f
o
r
m
o
f
o
u
r
s
u
g
g
es
te
d
in
tel
li
g
e
n
t
a
p
p
r
o
ac
h
Prin
cip
ally
,
th
e
ML
P
is
co
n
s
is
ted
o
f
in
p
u
t
lay
er
I
,
h
id
d
e
n
lay
er
H
an
d
o
u
tp
u
t
la
y
er
O
.
Fu
r
th
er
m
o
r
e,
it
in
v
o
lv
es
d
if
f
er
e
n
t
co
n
n
ec
tio
n
s
o
f
weig
h
ts
.
1
r
ep
r
esen
ts
th
e
f
ir
s
t
co
n
n
ec
tio
n
weig
h
ts
to
H
lay
er
an
d
2
r
ep
r
esen
ts
th
e
s
ec
o
n
d
co
n
n
ec
tio
n
weig
h
ts
to
O
la
y
er
.
T
o
u
tili
ze
th
e
ML
P,
two
s
tag
es
ar
e
r
eq
u
ir
ed
:
tr
ain
i
n
g
s
tag
e
an
d
test
in
g
s
tag
e.
T
h
e
v
alu
es
o
f
1
an
d
2
will st
ar
t a
s
s
m
all
in
itia
l r
an
d
o
m
s
at
th
e
b
e
g
in
n
in
g
o
f
t
h
e
tr
ain
in
g
s
tag
e.
On
th
e
o
t
h
er
h
an
d
,
th
ei
r
f
in
al
v
alu
es
at
th
e
en
d
o
f
th
e
tr
ain
in
g
s
tag
e
will
b
e
s
to
r
e
d
.
T
h
e
f
in
al
weig
h
t
v
alu
es
will
b
e
ex
p
lo
ited
in
th
e
test
in
g
s
tag
e.
T
h
e
ML
P
tr
ain
in
g
s
tag
e
co
n
tain
s
th
r
ee
m
ain
s
te
p
s
:
f
ee
d
f
o
r
war
d
in
g
in
p
u
ts
to
o
u
tp
u
ts
(
1
-
4
)
,
b
ac
k
p
r
o
p
ag
atin
g
er
r
o
r
s
(
5
-
1
1
)
an
d
u
p
d
atin
g
weig
h
ts
an
d
b
iases
(
1
2
-
1
5
)
.
T
h
e
f
o
llo
win
g
eq
u
atio
n
s
d
escr
ib
e
th
e
ess
en
tial M
L
P o
p
er
atio
n
s
:
=
0
1
+
∑
1
=
1
(
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
2
,
Ap
r
il
2
0
2
1
:
43
8
-
44
3
440
wh
er
e:
is
an
in
p
u
t
v
alu
e
t
o
th
e
h
id
d
en
lay
e
r
,
Y
is
th
e
in
d
ex
o
f
h
id
d
e
n
n
o
d
es,
0
1
is
a
co
n
n
ec
tio
n
weig
h
t
b
etwe
en
th
e
f
ir
s
t
b
ias
n
o
d
e
B1
an
d
h
id
d
en
lay
er
,
Q
is
th
e
n
u
m
b
er
o
f
h
id
d
e
n
n
o
d
es,
is
an
in
p
u
t
v
alu
e
o
f
th
e
in
p
u
t
lay
er
,
X
is
th
e
in
d
ex
o
f
in
p
u
t
n
o
d
es,
an
d
1
is
a
co
n
n
ec
tio
n
weig
h
t
b
etwe
en
th
e
in
p
u
t
an
d
h
id
d
en
lay
er
s
.
=
(
)
(
2
)
wh
er
e:
is
an
o
u
t
p
u
t v
alu
e
f
r
o
m
th
e
h
id
d
en
lay
er
.
=
0
2
+
∑
2
=
1
(
3
)
wh
er
e:
is
an
in
p
u
t
v
al
u
e
to
th
e
o
u
tp
u
t
lay
er
,
Z
is
th
e
in
d
e
x
o
f
o
u
tp
u
t
n
o
d
es,
0
2
is
a
co
n
n
ec
tio
n
weig
h
t
b
etwe
en
th
e
s
ec
o
n
d
b
ias
n
o
d
e
B2
an
d
o
u
tp
u
t
lay
er
,
R
is
th
e
n
u
m
b
er
o
f
o
u
tp
u
t
n
o
d
es,
an
d
2
is
a
co
n
n
ec
tio
n
weig
h
t b
etwe
en
th
e
h
id
d
en
an
d
o
u
tp
u
t la
y
er
s
.
=
(
)
(
4
)
wh
er
e:
is
an
o
u
t
p
u
t v
alu
e
f
r
o
m
th
e
o
u
t
p
u
t la
y
er
.
=
(
−
)
′
(
)
(
5
)
wh
er
e:
is
an
o
u
t
p
u
t e
r
r
o
r
v
al
u
e
an
d
is
a
d
eter
m
in
ed
tar
g
et
v
alu
e.
2
=
(
6
)
wh
er
e:
is
a
lear
n
in
g
r
ate
v
alu
e
.
0
2
=
(
7
)
=
0
2
+
∑
2
=
1
(
8
)
wh
er
e:
is
an
in
p
u
t e
r
r
o
r
v
alu
e
to
th
e
h
id
d
en
lay
er
.
=
′
(
)
(
9
)
wh
er
e:
is
an
o
u
t
p
u
t e
r
r
o
r
v
al
u
e
f
r
o
m
t
h
e
h
id
d
en
lay
er
.
1
=
(
1
0
)
0
1
=
(
1
1
)
2
(
)
=
2
(
)
+
2
(
1
2
)
1
(
)
=
1
(
)
+
1
(
1
3
)
0
2
(
)
=
0
2
(
)
+
0
2
(
1
4
)
0
1
(
)
=
0
1
(
)
+
0
1
(
1
5
)
C
o
n
s
eq
u
en
tly
,
th
e
ML
P
test
in
g
s
tag
e
ca
n
b
e
ca
r
r
ied
o
u
t.
I
t
h
a
s
o
n
ly
o
n
e
m
ain
s
tep
(
1
-
4
)
.
As
m
en
tio
n
ed
,
th
e
f
in
al
weig
h
t
v
alu
es
th
at
ar
e
o
b
tain
ed
f
r
o
m
th
e
tr
ain
in
g
s
tag
e
will
b
e
ex
p
lo
ited
in
th
is
s
tag
e
[
1
7
]
.
I
n
th
is
p
ap
er
,
th
e
n
u
m
b
er
o
f
in
p
u
t
n
o
d
es in
th
e
I
lay
er
is
u
s
ed
as
P
=2
5
0
0
,
s
o
,
th
is
lay
er
ca
n
ac
ce
p
t
all
th
e
p
ix
el
v
alu
es
o
f
a
cu
n
eif
o
r
m
s
y
m
b
o
l
im
ag
e.
T
h
e
n
u
m
b
e
r
o
f
h
id
d
en
n
o
d
es is
u
s
ed
as
Q
=2
5
5
,
th
is
n
u
m
b
er
h
as
b
ee
n
ac
h
iev
ed
ac
co
r
d
in
g
to
a
s
u
g
g
ested
m
eth
o
d
in
[
1
8
]
.
T
h
e
n
u
m
b
er
o
f
o
u
t
p
u
t
n
o
d
es
is
u
tili
ze
d
as
R
=2
6
,
wh
er
e
th
is
n
u
m
b
e
r
is
eq
u
al
to
t
h
e
n
u
m
b
er
o
f
E
n
g
lis
h
alp
h
ab
ets.
B
y
th
is
ca
s
e,
it
is
f
ea
s
ib
le
to
tr
an
s
lat
Su
m
er
ia
n
cu
n
eif
o
r
m
s
y
m
b
o
ls
to
th
eir
co
r
r
esp
o
n
d
in
g
E
n
g
lis
h
letter
s
.
4.
P
RACTI
CAL
I
M
P
L
E
M
E
N
T
AT
I
O
NS A
N
D
DIS
CUSS
I
O
NS
Fo
r
p
r
ac
tical
im
p
lem
e
n
tatio
n
s
,
Su
m
er
ian
cu
n
eif
o
r
m
d
ataset
was f
ir
s
tly
r
eq
u
ir
ed
.
Af
ter
in
v
esti
g
atio
n
s
,
a
u
s
ef
u
l
d
ataset
f
r
o
m
[
1
9
]
h
a
s
b
ee
n
f
o
u
n
d
an
d
em
p
lo
y
ed
.
I
t
in
clu
d
es
th
e
m
ea
n
in
g
s
o
f
Su
m
er
ian
s
y
m
b
o
ls
in
E
n
g
lis
h
,
s
ee
Fig
u
r
e
2
.
Hen
ce
,
cu
n
if
o
r
m
s
y
m
b
o
l
im
ag
es
ar
e
ca
r
ef
u
lly
e
x
tr
ac
ted
.
T
h
e
n
,
ea
c
h
s
y
m
b
o
l
h
as
b
ee
n
r
esized
to
5
0
50
p
ix
els.
T
h
e
r
ea
s
o
n
o
f
u
s
in
g
a
f
ix
e
d
r
esize
is
to
es
tab
lis
h
f
ea
s
ib
le
ad
ap
t
atio
n
b
etwe
en
an
y
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Tr
a
n
s
la
tin
g
cu
n
eifo
r
m
s
ymb
o
ls
u
s
in
g
a
r
tifi
cia
l n
eu
r
a
l n
etw
o
r
k
(
A
r
w
a
Ha
med
S
a
lih
Ha
md
a
n
y
)
441
ap
p
lied
s
y
m
b
o
l
im
a
g
e
an
d
th
e
in
p
u
t
n
o
d
es
o
f
th
e
p
r
o
p
o
s
ed
ML
P.
No
w,
b
ec
a
u
s
e
o
f
th
e
li
m
itatio
n
s
o
f
a
v
ailab
le
d
ata,
im
ag
e
au
g
m
en
tatio
n
s
a
r
e
ex
p
lo
ited
to
p
r
o
v
id
e
a
b
ig
n
u
m
b
er
o
f
tr
ain
i
n
g
in
f
o
r
m
atio
n
.
As
m
en
tio
n
e
d
in
[
2
0
-
2
2
]
,
au
g
m
en
tatio
n
s
tr
at
eg
ies
o
f
r
o
tatio
n
s
an
d
tr
an
s
latio
n
s
ca
n
b
e
u
tili
ze
d
.
Var
io
u
s
tr
ain
in
g
im
ag
es
h
av
e
b
ee
n
estab
lis
h
ed
b
y
ap
p
ly
i
n
g
d
if
f
er
en
t
r
o
tatio
n
a
n
d
tr
an
s
lati
o
n
p
r
o
ce
s
s
es.
T
ab
le
1
s
h
o
ws
ex
am
p
les
o
f
t
h
e
ap
p
lied
o
p
er
atio
n
s
to
Su
m
er
ian
cu
n
eif
o
r
m
s
y
m
b
o
l
im
ag
es.
T
h
at
is
,
im
ag
es
o
f
cu
n
ei
f
o
r
m
s
y
m
b
o
ls
h
av
e
b
ee
n
an
aly
s
ed
b
y
em
p
l
o
y
in
g
m
u
ltip
le
o
p
er
atio
n
s
o
f
r
e
s
izin
g
,
r
o
tatio
n
s
an
d
tr
an
s
latio
n
s
.
Dif
f
er
en
t
r
o
tatio
n
s
an
d
tr
an
s
latio
n
s
h
av
e
b
ee
n
a
p
p
lied
to
ea
ch
cu
n
eif
o
r
m
s
y
m
b
o
l
im
ag
e.
T
o
tal
o
f
7
8
0
s
y
m
b
o
l im
ag
es
h
av
e
b
ee
n
estab
lis
h
ed
f
o
r
d
if
f
er
e
n
t r
o
tati
o
n
an
g
les,
3
9
0
s
y
m
b
o
l
im
ag
es
r
o
tated
to
th
e
r
i
g
h
t
d
ir
ec
tio
n
an
d
3
9
0
s
y
m
b
o
l
i
m
ag
es
r
o
tated
t
o
th
e
lef
t
d
ir
ec
tio
n
.
L
ik
ewise,
3
3
8
s
y
m
b
o
l
im
ag
es
h
av
e
b
ee
n
estab
lis
h
ed
f
o
r
d
if
f
er
en
t
tr
a
n
s
latio
n
d
ir
ec
tio
n
s
(
tr
an
s
latio
n
s
to
th
e
to
p
,
b
o
tto
m
,
r
ig
h
t
an
d
lef
t)
.
T
h
ese
au
g
m
en
tatio
n
im
ag
es h
av
e
b
e
en
u
s
ed
in
th
e
tr
ain
in
g
s
tag
e.
Fig
u
r
e
2
.
T
h
e
m
ea
n
i
n
g
s
o
f
S
u
m
e
r
ia
n
c
u
n
ei
f
o
r
m
s
y
m
b
o
ls
i
n
E
n
g
l
is
h
as
s
h
o
wn
i
n
[
1
9
]
T
a
b
le
1
.
E
x
am
p
l
es
o
f
t
h
e
ap
p
l
i
ed
o
p
er
ati
o
n
s
to
S
u
m
er
ia
n
c
u
n
eif
o
r
m
s
y
m
b
o
l
im
a
g
es
O
p
e
r
a
t
i
o
n
S
y
mb
o
l
1
S
y
mb
o
l
2
S
y
mb
o
l
3
S
y
mb
o
l
4
S
y
y
m
b
o
l
5
O
r
g
i
n
s
R
e
si
z
i
n
g
R
o
t
a
t
i
o
n
Tr
a
n
s
l
a
t
i
o
n
s
ML
P
n
etw
o
r
k
t
r
ai
n
i
n
g
p
a
r
a
m
e
ter
s
h
a
v
e
b
e
en
s
et
as
f
o
l
lo
ws:
m
i
n
im
u
m
tr
ai
n
i
n
g
er
r
o
r
=
0
.
0
0
1
,
tr
an
s
f
e
r
f
u
n
cti
o
n
in
H
la
y
e
r
=
t
a
n
s
ig
m
o
i
d
,
t
r
an
s
f
e
r
f
u
n
c
ti
o
n
in
O
la
y
e
r
=
p
u
r
e
li
n
ea
r
a
n
d
t
r
ai
n
i
n
g
t
y
p
e
=
s
ca
le
d
c
o
n
j
u
g
ate
g
r
a
d
ie
n
t
(
SC
G
)
.
T
h
e
t
r
ai
n
i
n
g
c
u
r
v
e
d
u
r
i
n
g
t
h
e
t
r
ai
n
i
n
g
s
t
a
g
e
is
g
iv
en
i
n
Fi
g
u
r
e
3
.
I
t
c
a
n
b
e
o
b
s
e
r
v
ed
f
r
o
m
t
h
is
f
i
g
u
r
e
t
h
at
t
h
e
t
r
ai
n
i
n
g
c
u
r
v
e
i
s
s
m
o
o
t
h
l
y
d
e
cli
n
e
d
t
o
w
ar
d
a
m
i
n
im
u
m
e
r
r
o
r
v
al
u
e
o
f
0
.
0
0
0
9
9
9
9
.
T
h
is
c
a
n
b
e
co
n
s
i
d
e
r
e
d
as
a
n
i
n
d
ic
at
o
r
t
o
t
h
e
s
u
cc
ess
f
u
l
n
ess
o
f
t
h
e
tr
ai
n
i
n
g
s
ta
g
e
.
F
o
r
t
h
e
t
esti
n
g
s
t
ag
e,
s
e
r
ies
o
f
u
n
au
g
m
e
n
t
ed
Su
m
er
ia
n
cu
n
ei
f
o
r
m
s
y
m
b
o
ls
ca
n
in
tel
li
g
e
n
tl
y
b
e
t
r
a
n
s
l
ate
d
f
r
o
m
i
m
a
g
es
t
o
E
n
g
lis
h
l
ett
er
s
.
T
a
b
le
2
s
h
o
ws
v
a
r
i
o
u
s
E
n
g
lis
h
te
x
ts
th
at
c
a
n
s
u
cc
ess
f
u
l
ly
b
e
tr
an
s
lat
e
d
f
r
o
m
c
u
n
e
if
o
r
m
s
y
m
b
o
ls
b
y
u
s
in
g
o
u
r
s
u
g
g
es
te
d
ML
P
a
p
p
r
o
a
ch
.
T
h
is
t
ab
le
d
em
o
n
s
t
r
a
tes
e
x
a
m
p
les
o
f
E
n
g
lis
h
te
x
ts
t
h
a
t
c
a
n
b
e
a
c
q
u
ir
ed
f
r
o
m
S
u
m
e
r
ia
n
s
y
m
b
o
ls
i
f
th
e
p
r
o
p
o
s
e
d
M
L
P
m
et
h
o
d
is
u
s
ed
.
I
t
is
a
p
le
asu
r
e
t
o
y
iel
d
t
h
at
o
u
r
p
r
o
p
o
s
e
d
a
p
p
r
o
ac
h
h
as
b
e
n
ch
m
a
r
k
e
d
a
s
u
c
ce
s
s
f
u
l
ac
c
u
r
a
cy
o
f
1
0
0
%
.
T
h
is
c
an
p
r
o
v
i
d
e
ess
e
n
t
i
al
ad
v
a
n
t
a
g
es
o
f
q
u
ic
k
l
y
tr
an
s
lat
in
g
th
e
S
u
m
e
r
ia
n
cu
n
eif
o
r
m
w
r
iti
n
g
,
r
e
d
u
ci
n
g
t
h
e
e
f
f
o
r
ts
o
f
i
n
t
er
p
r
e
tin
g
s
u
c
h
in
te
r
est
in
g
s
t
y
l
e
a
n
d
p
r
ese
n
t
in
g
t
h
is
w
r
iti
n
g
t
o
p
u
b
li
c
w
h
er
e
a
n
y
p
e
r
s
o
n
ca
n
u
n
d
e
r
s
t
an
d
i
ts
s
y
m
b
o
ls
.
T
h
e
p
r
o
p
o
s
e
d
n
e
u
r
al
n
e
tw
o
r
k
a
p
p
r
o
a
c
h
ca
n
b
e
s
ig
n
i
f
i
ed
wit
h
o
t
h
e
r
i
n
f
o
r
m
ati
o
n
tec
h
n
o
lo
g
y
a
n
d
A
I
m
o
d
els
as
i
n
[
2
3
-
3
3
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
2
,
Ap
r
il
2
0
2
1
:
43
8
-
44
3
442
Fig
u
r
e
3
.
T
h
e
t
r
ai
n
i
n
g
c
u
r
v
e
d
u
r
in
g
t
h
e
t
r
ai
n
i
n
g
s
ta
g
e
T
a
b
le
2
.
V
ar
i
o
u
s
E
n
g
lis
h
te
x
ts
th
a
t
ca
n
s
u
cc
ess
f
u
ll
y
b
e
tr
a
n
s
la
ted
f
r
o
m
S
u
m
e
r
ia
n
c
u
n
ei
f
o
r
m
s
y
m
b
o
ls
b
y
u
s
i
n
g
o
u
r
s
u
g
g
este
d
ML
P
a
p
p
r
o
a
c
h
S
u
m
e
r
i
a
n
C
u
n
e
i
f
o
r
m
S
y
m
b
o
l
s
E
n
g
l
i
s
h
Le
t
t
e
r
s
S
U
M
E
R
I
A
N
C
U
N
EI
F
O
R
M
W
R
I
TI
N
G
S
Y
M
B
O
LS
M
O
S
O
P
O
T
A
M
I
A
C
I
V
I
LI
ZA
TI
O
N
N
O
R
TH
E
R
N
TE
C
H
N
I
C
A
L
U
N
I
V
E
R
S
I
TY
TE
C
H
N
I
C
A
L
EN
G
I
N
E
E
R
I
N
G
C
O
L
LE
G
E
M
O
D
E
R
N
S
C
I
EN
TI
F
I
C
R
ES
EA
R
C
H
ES
5.
CO
NCLU
SI
O
N
I
n
f
o
r
m
ati
o
n
te
c
h
n
o
l
o
g
y
a
n
d
A
I
a
r
e
r
ec
en
tl
y
o
cc
u
p
y
i
n
g
s
i
g
n
if
ica
n
t
p
o
s
it
io
n
s
in
d
i
f
f
e
r
e
n
t
tas
k
s
.
T
h
is
is
n
o
t
o
n
l
y
t
h
e
m
a
tte
r
o
f
m
o
d
er
n
s
ci
en
ce
s
an
d
a
p
p
li
ca
ti
o
n
s
.
I
n
f
ac
t
,
t
h
e
y
c
a
n
b
e
a
p
p
lie
d
t
o
s
o
l
v
e
a
n
cie
n
t
is
s
u
es
d
u
r
i
n
g
t
h
e
b
asis
ci
v
il
iz
ati
o
n
s
o
f
h
u
m
a
n
it
y
.
W
r
iti
n
g
b
y
S
u
m
e
r
i
an
c
u
n
ei
f
o
r
m
s
y
m
b
o
ls
is
o
n
e
o
f
th
e
an
ci
en
t
s
t
y
l
es
th
a
t
is
wo
r
t
h
t
o
b
e
c
o
n
s
i
d
e
r
e
d
.
I
n
t
h
is
s
tu
d
y
,
c
u
n
ei
f
o
r
m
s
y
m
b
o
l
i
m
a
g
es
o
f
Su
m
e
r
i
an
w
r
iti
n
g
we
r
e
t
r
an
s
lat
ed
t
o
E
n
g
l
is
h
let
te
r
s
.
T
o
r
ea
c
h
t
h
is
g
o
al
,
an
AI
te
ch
n
i
q
u
e
o
f
M
L
P
n
etw
o
r
k
w
as
e
f
f
i
cie
n
tl
y
em
p
l
o
y
ed
.
I
t
is
d
eli
g
h
te
d
to
y
ie
ld
t
h
at
o
u
r
s
u
g
g
este
d
a
p
p
r
o
ac
h
h
as
b
e
n
ch
m
a
r
k
e
d
a
v
e
r
y
h
ig
h
a
cc
u
r
a
cy
o
f
1
0
0
%.
T
h
is
m
ay
m
a
k
e
S
u
m
e
r
ia
n
cu
n
e
if
o
r
m
s
y
m
b
o
ls
t
o
ea
s
il
y
a
n
d
ac
c
u
r
at
el
y
b
e
u
n
d
e
r
s
t
o
o
d
b
y
e
x
p
lo
r
e
r
s
a
n
d
r
es
ea
r
c
h
e
r
s
.
RE
F
E
R
E
NC
E
S
[1
]
A.
S
u
lai
m
a
n
,
“
C
u
n
e
if
o
rm
W
ri
ti
n
g
,”
Da
r
A
l
-
Ku
t
u
b
f
o
r
P
r
i
n
t
i
n
g
a
n
d
P
u
b
l
is
h
i
n
g
-
M
o
s
u
l
,
2
0
0
0
.
[2
]
Ali
A
d
e
l
S
a
e
id
,
e
t
a
l
.
,
“
C
u
n
e
if
o
rm
s
y
m
b
o
ls
re
c
o
g
n
it
i
o
n
b
y
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
(
S
V
M
)
,
”
J
o
u
r
n
a
l
o
f
A
L
-
Q
a
d
is
iy
a
h
fo
r
c
o
m
p
u
ter
sc
ie
n
c
e
a
n
d
m
a
t
h
e
m
a
t
ics
,
v
o
l
.
1
1
,
n
o.
1
,
Ja
n
u
a
r
y
201
8
.
[3
]
Jo
h
n
Ni
c
h
o
las
P
o
st
g
a
te
,
“
La
n
g
u
a
g
e
s
o
f
I
ra
q
,
a
n
c
ie
n
t
a
n
d
m
o
d
e
r
n
,”
B
rit
is
h
S
c
h
o
o
l
o
f
Arc
h
a
e
o
l
o
g
y
in
Ir
a
q
,
2
0
0
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Tr
a
n
s
la
tin
g
cu
n
eifo
r
m
s
ymb
o
ls
u
s
in
g
a
r
tifi
cia
l n
eu
r
a
l n
etw
o
r
k
(
A
r
w
a
Ha
med
S
a
lih
Ha
md
a
n
y
)
443
[4
]
G
o
n
g
Y
u
s
h
u
,
“
Th
e
s
u
m
e
r
ia
n
a
c
c
o
u
n
t
o
f
t
h
e
i
n
v
e
n
t
i
o
n
o
f
w
ri
ti
n
g
-
a
n
e
w i
n
t
e
r
p
re
ta
ti
o
n
,
”
E
lse
v
ier
,
Pr
o
c
e
d
i
a
-
S
o
c
ia
l
a
n
d
Be
h
a
v
i
o
r
a
l
S
c
ie
n
c
e
s
,
v
o
l
.
2
,
n
o
.
5
,
p
p
.
7
4
4
6
-
7
4
5
3
,
2
0
1
0
.
[5
]
A.
Zi
y
a
A
k
tas
a
n
d
T
u
n
c
As
u
r
o
g
l
u
,
“
C
o
m
p
u
ter
ize
d
h
i
t
ti
te
c
u
n
e
if
o
r
m
s
i
g
n
re
c
o
g
n
it
i
o
n
a
n
d
d
a
t
a
m
i
n
in
g
a
p
p
l
ica
t
i
o
n
e
x
a
m
p
les
,
”
E
u
r
o
p
e
a
n
I
n
ter
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
c
ie
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
Oc
t
o
b
e
r
2017.
[6
]
L
o
g
a
n
B
o
r
n
,
e
t
a
l
.
,
“
S
i
g
n
c
l
u
s
ter
i
n
g
a
n
d
t
o
p
ic
e
x
tra
c
t
i
o
n
i
n
p
r
o
t
o
-
e
l
a
m
it
e
,”
Pr
o
c
.
o
f
t
h
e
3
rd
J
o
in
t
S
I
G
HU
M
W
o
rk
s
h
o
p
o
n
C
o
m
p
u
t
a
t
io
n
a
l
L
i
n
g
u
is
ti
c
s
f
o
r
Cu
lt
u
r
a
l
He
r
i
ta
g
e
,
S
o
c
i
a
l
S
c
ie
n
c
e
s,
H
u
m
a
n
i
t
ies
a
n
d
L
i
ter
a
t
u
re
,
J
a
n
u
a
r
y
2
0
1
9
.
[7
]
R.
R
.
A
l
-
N
ima
,
e
t
a
l
.
,
“
A
n
e
w
a
p
p
ro
a
c
h
t
o
p
re
d
ict
i
n
g
p
h
y
s
ica
l
b
i
o
m
e
tr
ics
f
r
o
m
b
e
h
a
v
i
o
u
r
a
l
b
i
o
m
e
t
rics
,”
I
n
ter
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
te
r,
I
n
f
o
rm
a
ti
o
n
,
S
y
ste
ms
a
n
d
C
o
n
tr
o
l
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l.
8
,
n
o
.
1
1
,
p
p
.
2
0
0
1
-
2
0
0
6
,
2
0
1
4
.
[8
]
R.
R
.
O
.
A
l
-
Nim
a
,
e
t
a
l
.
,
“
Ef
fic
ie
n
t
fi
n
g
e
r
s
e
g
m
e
n
ta
ti
o
n
r
o
b
u
s
t
t
o
h
a
n
d
a
l
i
g
n
m
e
n
t
i
n
ima
g
i
n
g
w
it
h
a
p
p
l
i
c
a
ti
o
n
t
o
h
u
m
a
n
v
e
r
if
ica
ti
o
n
,”
5
th
IE
E
E
I
n
ter
n
a
ti
o
n
a
l
W
o
rk
s
h
o
p
o
n
B
i
o
me
tric
s
a
n
d
F
o
re
n
s
ics
,
A
p
r
il
2
0
1
7
.
[9
]
R.
R.
O
.
A
l
-
Nim
a
,
“
S
i
g
n
a
l
p
r
o
c
e
ss
i
n
g
a
n
d
m
a
c
h
i
n
e
le
a
r
n
in
g
te
c
h
n
i
q
u
e
s
f
o
r
h
u
m
a
n
v
e
ri
fic
a
t
i
o
n
b
a
se
d
o
n
f
i
n
g
e
r
te
x
t
u
re
s
,
”
P
h
D
t
h
e
si
s,
S
c
h
o
o
l
o
f
E
n
g
i
n
e
e
r
i
n
g
,
Ne
w
c
a
st
le
U
n
iv
e
rs
it
y
,
2
0
1
7
.
[1
0
]
M
.
T
.
Al
-
Ka
lt
a
k
c
h
i
,
e
t
a
l.
,
“
F
i
n
g
e
r
t
e
x
t
u
r
e
v
e
r
i
fica
ti
o
n
s
y
s
tem
s
b
a
se
d
o
n
m
u
l
ti
p
le s
p
e
c
t
r
u
m
li
g
h
ti
n
g
se
n
s
o
r
s
w
i
th
f
o
u
r
fu
si
o
n
le
v
e
ls
,
"
Ir
a
q
i
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
ti
o
n
&
C
o
m
mu
n
ic
a
ti
o
n
s
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
,
n
o
.
3
,
p
p
.
1
-
1
6
,
2
0
1
7
.
[1
1
]
R.
R
.
O
.
Al
-
N
ima
,
e
t
a
l
.
,
“
P
e
rs
o
n
a
l
v
e
ri
fic
a
t
i
o
n
b
a
se
d
o
n
m
u
lt
i
-
s
p
e
c
tra
l
fi
n
g
e
r
te
x
tu
re
l
i
g
h
t
i
n
g
ima
g
e
s
,
”
I
ET
S
i
g
n
a
l
Pr
o
c
e
ss
i
n
g
,
v
o
l.
1
2
,
no
.
9
,
p
p
.
1
1
5
4
-
1164
,
2018.
[1
2
]
R.
R
.
O
.
Al
-
N
ima
,
e
t
a
l
.
,
“
R
o
b
u
s
t
fe
a
t
u
re
e
x
trac
ti
o
n
a
n
d
sa
l
v
a
g
e
s
c
h
e
m
e
s
f
o
r
fi
n
g
e
r
te
x
tu
re
-
b
a
se
d
b
io
m
e
t
ric
s
,”
IE
T
Bi
o
me
tr
ics
,
v
o
l
.
6
,
n
o
.
2
,
p
p
.
4
3
-
5
2
,
2
0
1
7
.
[1
3
]
R.
R
.
O
.
A
l
-
N
ima
,
e
t
a
l
.
,
“
A
n
o
v
e
l
b
i
o
m
e
tr
ic
a
p
p
r
o
a
c
h
t
o
g
e
n
e
ra
te
R
OC c
u
r
v
e
fr
o
m
t
h
e
p
r
o
b
a
b
il
is
ti
c
n
e
u
r
a
l
n
e
t
w
o
r
k
,”
2
4
t
h
I
E
EE
S
i
g
n
a
l
Pr
o
c
e
ss
i
n
g
a
n
d
Co
mm
u
n
ic
a
ti
o
n
Ap
p
li
c
a
t
i
o
n
Co
n
f
e
re
n
c
e
(
S
IU
),
2
0
1
6
.
[1
4
]
R.
A
l
-
Nim
a
,
e
t
a
l
.
,
“
H
u
m
a
n
a
u
t
h
e
n
ti
c
a
t
io
n
wi
t
h
f
i
n
g
e
r
te
x
t
u
re
s
b
a
se
d
o
n
ima
g
e
f
e
a
t
u
re
e
n
h
a
n
c
e
m
e
n
t
,
”
2
n
d
IE
T
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
I
n
tel
l
ig
e
n
t
S
i
g
n
a
l
Pr
o
c
e
s
si
n
g
(
I
S
P
)
,
2
0
1
5
.
[1
5
]
Ra
i
d
R
.
O.
A
l
-
N
ima
,
e
t
a
l
.
,
“
F
i
n
g
e
r
te
x
t
u
re
b
i
o
m
e
tr
ic
c
h
a
ra
c
ter
is
ti
c
:
a
s
u
r
v
e
y
,
”
a
rX
iv
p
re
p
r
i
n
t
a
r
Xi
v
:
2
0
0
6
.
0
4
1
9
3
,
2
0
2
0
.
[1
6
]
R.
R
.
A
l
-
Nim
a
,
“
H
u
m
a
n
a
u
t
h
e
n
t
ica
t
i
o
n
w
it
h
e
a
r
p
r
i
n
t
fo
r
se
c
u
re
t
e
le
p
h
o
n
e
s
y
s
tem
,”
Ir
a
q
i
J
o
u
r
n
a
l
o
f
C
o
m
p
u
te
rs
,
Co
mm
u
n
ic
a
ti
o
n
s
,
C
o
n
tr
o
l
a
n
d
S
y
s
tem
s
E
n
g
i
n
e
e
r
i
n
g
IJ
C
CC
E
,
v
o
l
.
1
2
,
n
o
.
2
,
J
u
n
e
2012.
[1
7
]
L.
F
a
u
se
tt
,
“
F
u
n
d
a
m
e
n
ta
ls
o
f
n
e
u
r
a
l
n
e
tw
o
r
k
s:
a
rc
h
i
tec
tu
re
s
,
a
l
g
o
ri
t
h
m
s
,
a
n
d
a
p
p
li
c
a
t
i
o
n
s
,”
Pre
n
ti
c
e
-
Ha
ll
,
I
n
c
.
,
1
9
9
4
.
[1
8
]
Ka
ts
u
n
a
ri
S
h
i
b
a
t
a
a
n
d
Yu
s
u
k
e
I
k
e
d
a
,
“
Ef
fe
c
t
o
f
n
u
m
b
e
r
o
f
h
i
d
d
e
n
n
e
u
r
o
n
s
o
n
lea
r
n
i
n
g
i
n
lar
g
e
-
sc
a
le
la
y
e
re
d
n
e
u
r
a
l
n
e
tw
o
r
k
s
,
”
2
0
0
9
ICC
A
S
-
S
IC
E
,
2
0
0
9
.
[1
9
]
Ab
d
Al
-
Ra
h
m
a
n
Ab
d
Al
-
Latif
Al
-
Na
m
ir,
“
Ho
w t
h
e
writi
n
g
is d
e
v
e
lo
p
e
d
i
n
h
ist
o
ry
?
,
”
Z
e
d
n
i
Ne
tw
o
r
k
fo
r
E
d
u
c
a
t
i
o
n
to
C
o
m
p
re
h
e
n
d
t
h
e
F
u
t
u
re
,
2
0
2
0
.
[O
n
l
i
n
e
]
.
A
v
a
il
a
b
le:
h
t
t
p
:/
/ze
d
n
i.
c
o
m
/
%
D
8
%
A
F
%
D
8
%
B
1
%
D
8
%
A7
%
D
8
%
B
3%
D8
%
A
7
%
D
8
%
AA
-
%
D
8
%
B
9
%
D
8
%
A
7
%
D
9
%
8
5
/
%
D
8
%
AF
%
D
8
%
B
1
%
D
8
%
A7
%
D
8
%
B
3
%
D
8
%
A7
%
D
8
%
A
A
-
%
D
8
%
B
9
%
D8
%
B
1
%
D
8
%
A
8
%
D9
%
8
A
%
D
9
%
9
1
%
D
8
%
A
9
/
%
D
9
%
8
3
%
D
9
%
8
A
%
D
9
%
8
1
-
%
D
8
%
AA
%
D
8
%
B
7
%
D9
%
8
8
%
D
8
%
B
1
%
D
8
%
AA
-
%
D
8
%
A
7
%
D
9
%
8
4
%
D
9
%
8
3
%
D
8
%
AA
%
D
8
%
A
7
%
D
8
%
A
8
%
D
8
%
A
9
-
%
D8
%
B
9
%
D8
%
A
8
%
D
8
%
B
1
-
%
D
8
%
A
7
%
D
9
%
8
4
%
D
8
%
AA
%
D
8
%
A7
%
D
8
%
B1
%
D
9
%
8
A%
D
8
%
AE
%
D8
%
9
F
_
1
/
?
f
b
c
li
d
=
IwA
R0
u
J_
7
S
tV
8
h
w
S
Hx
g
fn
Z
1
m
g
lZ
Cc
2
x
J
6
m
m
f
q
_
C
rL
u
C
1
Y
0
OF
M
-
h
Vz
p
KX
t
M
OI
4
[2
0
]
E.
D
.
C
u
b
u
k
,
e
t
a
l
.
,
“
A
u
t
o
A
u
g
m
e
n
t:
lea
r
n
i
n
g
a
u
g
m
e
n
ta
ti
o
n
s
tra
te
g
i
e
s
f
r
o
m
d
a
ta
,
”
P
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
E
EE
C
o
n
fer
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
isi
o
n
a
n
d
P
a
t
ter
n
Rec
o
g
n
i
ti
o
n
,
2019.
[2
1
]
S
h
a
n
g
S
h
a
n
g
,
e
t
a
l.
,
“
A
u
t
o
m
a
ti
c
z
e
b
r
a
f
is
h
e
g
g
p
h
e
n
o
t
y
p
e
re
c
o
g
n
it
i
o
n
f
r
o
m
b
r
ig
h
t
-
fi
e
l
d
m
icr
o
sc
o
p
ic
i
m
a
g
e
s
u
s
i
n
g
d
e
e
p
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
r
k
,”
A
p
p
l
ie
d
S
c
i
e
n
c
e
s
,
v
o
l.
9
,
n
o
.
1
6
,
p
p
.
1
-
12
,
A
u
g
u
st
2
0
1
9
.
[2
2
]
Y.
L
u
,
“
F
o
o
d
im
a
g
e
r
e
c
o
g
n
i
ti
o
n
b
y
u
s
i
n
g
c
o
n
v
o
l
u
t
io
n
a
l
n
e
u
ra
l
n
e
t
wo
r
k
s
(c
n
n
s)
,
”
a
rX
iv
p
re
p
ri
n
t
a
r
Xi
v
:
1
6
1
2
.
0
0
9
8
3
,
2
0
1
6
.
[2
3
]
Lu
b
a
b
H.
Alb
a
k
,
e
t
a
l
.
,
“
De
sig
n
se
c
u
rit
y
sy
ste
m
b
a
se
d
o
n
Ard
u
in
o
,”
T
ES
T
En
g
in
e
e
rin
g
&
M
a
n
a
g
e
me
n
t
,
T
h
e
M
a
tt
i
n
g
ley
Pu
b
li
sh
i
n
g
Co
.
,
In
c
.
,
v
o
l.
8
2
,
M
a
rc
h
2
0
2
0
.
[2
4
]
Arw
a
Ha
m
id
S
a
li
h
Ha
m
d
a
n
y
,
e
t
a
l
.
,
“
Wi
re
les
s
wa
it
e
r
ro
b
o
t
,
”
T
ES
T
En
g
in
e
e
rin
g
&
M
a
n
a
g
e
me
n
t,
T
h
e
M
a
tt
in
g
le
y
Pu
b
li
s
h
in
g
C
o
.
,
I
n
c
.
,
v
o
l.
8
1
,
M
a
r
c
h
2
0
1
9
.
[2
5
]
M
u
sa
b
T.
S
.
Al
-
Ka
lt
a
k
c
h
i,
e
t
a
l
.
,
“
Th
o
ro
u
g
h
e
v
a
l
u
a
ti
o
n
o
f
TIM
IT
d
a
tab
a
se
sp
e
a
k
e
r
id
e
n
ti
fica
ti
o
n
p
e
r
fo
rm
a
n
c
e
u
n
d
e
r
n
o
ise
wi
th
a
n
d
wit
h
o
u
t
th
e
G
.
7
1
2
t
y
p
e
h
a
n
d
se
t
,”
S
p
ri
n
g
e
r,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
S
p
e
e
c
h
T
e
c
h
n
o
lo
g
y
,
v
o
l
.
2
2
,
n
o
.
3
,
p
p
.
8
5
1
-
8
6
3
,
2
0
1
9
.
[2
6
]
M
o
a
tas
e
m
Ya
se
e
n
Al
-
Rid
h
a
,
e
t
a
l
.
,
“
Ad
a
p
ti
v
e
n
e
u
r
o
-
f
u
z
z
y
in
fe
re
n
c
e
sy
ste
m
fo
r
c
o
n
tro
ll
in
g
a
ste
a
m
v
a
lv
e
,”
2
0
1
9
IEE
E
9
t
h
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
y
ste
m E
n
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
(ICS
E
T
)
,
2
0
1
9
.
[2
7
]
R
.
R
.
O
.
Al
-
N
ima
,
e
t
a
l
.
,
“
R
o
a
d
tr
a
c
k
i
n
g
u
si
n
g
d
e
e
p
r
e
i
n
f
o
rc
e
m
e
n
t
l
e
a
r
n
i
n
g
f
o
r
se
lf
-
d
ri
v
i
n
g
c
a
r
a
p
p
li
c
a
ti
o
n
s
,
”
Pr
o
g
re
ss
in
Co
m
p
u
ter
Rec
o
g
n
i
ti
o
n
S
y
s
tem
s
,
CO
R
ES
2
0
1
9
,
A
d
v
a
n
c
e
s
i
n
I
n
tel
l
i
g
e
n
t
S
y
s
tem
s
a
n
d
C
o
m
p
u
ti
n
g
,
Ja
n
u
a
r
y
2
0
2
0
.
[2
8
]
R.
R
.
Om
a
r
,
e
t
a
l
.
,
“
De
e
p
f
i
n
g
e
r
t
e
x
t
u
re
lea
r
n
i
n
g
f
o
r
v
e
ri
f
y
i
n
g
p
e
o
p
le
,
”
IE
T
B
i
o
me
tr
ics
,
v
o
l.
8
,
n
o
.
1
,
p
p
.
1
-
1
0
,
2
0
1
9
.
[2
9
]
Ra
i
d
Ra
f
i
Om
a
r
A
l
-
N
ima
,
e
t
a
l
.
,
“
Re
g
e
n
e
ra
t
i
n
g
fa
c
e
ima
g
e
s
f
ro
m
m
u
lt
i
-
s
p
e
c
t
ra
l
p
a
lm
ima
g
e
s
u
si
n
g
m
u
lt
i
p
le
f
u
s
i
o
n
m
e
t
h
o
d
s
,
”
T
E
L
KO
M
NIK
A
T
e
lec
o
mm
u
n
ic
a
ti
o
n
C
o
m
p
u
t
in
g
E
lec
tr
o
n
i
c
s
a
n
d
C
o
n
tr
o
l
,
v
o
l
.
1
7
,
n
o
.
6
,
p
p
.
3
1
1
0
-
3
1
1
9
,
De
c
e
m
b
e
r
2
0
1
9
.
[3
0
]
R.
R.
O.
A
l
-
N
ima
,
e
t
a
l
.
,
“
S
e
g
m
e
n
t
i
n
g
fi
n
g
e
r
i
n
n
e
r
s
u
rfa
c
e
f
o
r
t
h
e
p
u
r
p
o
se
o
f
h
u
m
a
n
re
c
o
g
n
i
t
i
o
n
,”
2
n
d
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
E
n
g
i
n
e
e
r
in
g
T
e
c
h
n
o
l
o
g
y
a
n
d
i
ts
A
p
p
li
c
a
ti
o
n
s
(II
CE
T
A)
,
2
0
1
9
.
[3
1
]
Ra
i
d
R
a
f
i
Om
a
r
Al
-
N
ima
,
e
t
a
l
.
,
“
Us
i
n
g
h
a
n
d
-
d
o
rsa
l
ima
g
e
s
t
o
re
p
ro
d
u
c
e
fa
c
e
ima
g
e
s
b
y
a
p
p
l
y
i
n
g
b
a
c
k
p
ro
p
a
g
a
t
i
o
n
a
n
d
c
a
sc
a
d
e
-
f
o
rwa
r
d
n
e
u
r
a
l
n
e
tw
o
rk
s
,
”
2
n
d
I
n
te
r
n
a
t
i
o
n
a
l
Co
n
fer
e
n
c
e
o
n
El
e
c
t
ric
a
l,
C
o
mm
u
n
ic
a
t
i
o
n
,
C
o
m
p
u
ter
,
P
o
we
r
a
n
d
C
o
n
tr
o
l
E
n
g
i
n
e
e
r
in
g
(IC
EC
C
PC
E
1
9
)
,
F
e
b
r
u
a
r
y
2019.
[3
2
]
M
.
A
.
M
.
A
b
d
u
l
la
h
,
e
t
a
l
.
,
“
Cr
o
s
s
-
s
p
e
c
tra
l
Ir
is
M
a
tc
h
i
n
g
f
o
r
S
u
r
v
e
il
la
n
c
e
A
p
p
l
ica
ti
o
n
s
,”
S
p
r
i
n
g
e
r
,
S
u
rv
e
il
l
a
n
c
e
i
n
Act
i
o
n
T
e
c
h
n
o
l
o
g
ies
f
o
r
Ci
v
i
li
a
n
,
M
il
it
a
r
y
a
n
d
Cy
b
e
r
S
u
rv
e
i
ll
a
n
c
e
,
p
p
.
1
0
5
-
1
2
5
,
Ja
n
u
a
r
y
2
0
1
8
.
[3
3
]
M
.
R
.
K
h
a
li
l,
e
t
a
l
.
,
“
P
e
rs
o
n
a
l
i
d
e
n
ti
fica
t
io
n
w
it
h
i
ri
s
p
a
tt
e
r
n
s
,”
AL
-
R
a
fi
d
a
i
n
J
o
u
rn
a
l
o
f
C
o
m
p
u
te
r
S
c
i
e
n
c
e
s
a
n
d
M
a
t
h
e
m
a
t
ics
,
v
o
l
.
6
,
n
o
.
1
,
2
0
0
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.