I
n
d
on
e
s
ian
Jou
r
n
al
o
f
E
lec
t
r
ica
l
E
n
gin
e
e
r
in
g
a
n
d
Com
p
u
t
e
r
S
c
ience
Vo
l
.
25
,
N
o
.
1
,
J
a
n
ua
r
y
2022
,
pp.
1
83
~
190
I
S
S
N:
2502
-
4752,
DO
I
:
10
.
11591/i
j
e
e
c
s
.
v
25
.i
1
.
pp
183
-
190
183
Jou
r
n
al
h
o
m
e
page
:
ht
tp:
//
ij
e
e
c
s
.
iaes
c
or
e
.
c
om
E
f
f
ic
ie
n
t
r
e
sam
p
li
n
g f
e
at
u
r
e
s an
d
c
on
v
ol
u
t
io
n
n
e
u
r
al
n
e
t
w
o
r
k
m
od
e
l
f
or
i
m
age
f
or
ge
r
y
d
e
t
e
c
t
io
n
M
an
j
u
n
at
h
a
S
1,
2
,
M
al
in
i
M
.
P
at
il
3
1
D
e
p
a
r
tm
e
nt
of
I
nf
o
r
ma
t
i
on
S
c
i
e
nc
e
a
nd
E
ngi
n
e
e
r
i
n
g
,
G
l
ob
a
l
A
c
a
de
m
y
of
T
e
c
hno
l
o
gy
,
B
e
nga
l
ur
u,
I
nd
i
a
2
D
e
p
a
r
tm
e
nt
of
C
S
E
,
J
.
S
.
S
A
c
a
d
e
m
y
of
T
e
c
hn
i
c
a
l
E
duc
a
t
i
on
B
e
n
ga
l
ur
u,
B
e
ng
a
l
u
r
u,
I
nd
i
a
3
D
e
p
a
r
tm
e
nt
of
I
nf
o
r
ma
t
i
on
S
c
i
e
nc
e
a
nd
E
ngi
n
e
e
r
i
n
g
,
J
.
S
.
S
A
c
a
d
e
m
y
of
T
e
c
hn
i
c
a
l
E
du
c
a
t
i
on,
B
e
nga
l
ur
u,
I
nd
i
a
Ar
t
ic
l
e
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
i
ve
d
F
e
b
4
,
2021
R
e
vi
s
e
d
Oc
t
31
,
2021
A
c
c
e
pt
e
d
De
c
1
,
2021
T
h
e
e
x
t
e
n
d
e
d
u
t
i
l
i
zat
i
o
n
o
f
p
i
c
t
u
r
e
-
en
h
a
n
c
i
n
g
o
r
m
an
i
p
u
l
at
i
n
g
t
o
o
l
s
h
as
l
ed
t
o
e
as
e
o
f
m
an
i
p
u
l
at
i
n
g
mu
l
t
i
med
i
a
d
at
a
w
h
i
c
h
i
n
c
l
u
d
e
s
d
i
g
i
t
al
i
m
a
g
e
s
.
T
h
e
s
e
m
a
n
i
p
u
l
at
i
o
n
s
w
i
l
l
d
i
s
t
u
rb
t
h
e
t
ru
t
h
fu
l
n
e
s
s
an
d
l
aw
f
u
l
n
e
s
s
o
f
i
m
a
g
e
s
,
re
s
u
l
t
i
n
g
i
n
mi
s
ap
p
re
h
e
n
s
i
o
n
,
an
d
mi
g
h
t
d
i
s
t
u
rb
s
o
c
i
a
l
s
e
cu
ri
t
y
.
T
h
e
i
m
ag
e
fo
ren
s
i
c
ap
p
ro
a
c
h
h
as
b
ee
n
em
p
l
o
y
ed
f
o
r
d
e
t
ec
t
i
n
g
w
h
et
h
e
r
o
r
n
o
t
a
n
i
m
ag
e
h
as
b
e
en
m
an
i
p
u
l
at
e
d
w
i
t
h
t
h
e
u
s
ag
e
o
f
p
o
s
i
t
i
v
e
at
t
ac
k
s
w
h
i
ch
i
n
c
l
u
d
e
s
s
p
l
i
c
i
n
g
,
an
d
c
o
p
y
-
mo
v
e.
T
h
i
s
p
ap
e
r
p
ro
v
i
d
e
s
a
co
m
p
e
t
en
t
t
am
p
e
r
i
n
g
de
t
ec
t
i
o
n
t
ech
n
i
q
u
e
u
s
i
n
g
r
e
s
am
p
l
i
n
g
f
e
at
u
r
e
s
an
d
co
n
v
o
l
u
t
i
o
n
n
e
u
ra
l
n
e
t
w
o
rk
(CN
N
)
.
In
t
h
i
s
m
o
d
e
l
ran
g
e
s
p
at
i
al
fi
l
t
e
ri
n
g
(
RSF
)
-
C
N
N
,
t
h
ro
u
g
h
o
u
t
p
re
p
ro
ce
s
s
i
n
g
t
h
e
i
m
a
g
e
i
s
d
i
v
i
d
ed
i
n
t
o
c
o
n
s
i
s
t
e
n
t
p
at
c
h
e
s
.
T
h
e
n
,
w
i
t
h
i
n
e
v
e
r
y
p
at
c
h
,
t
h
e
r
e
s
a
m
p
l
i
n
g
f
e
at
u
r
e
s
are
e
x
t
rac
t
ed
b
y
u
t
i
l
i
zi
n
g
aff
i
n
e
t
ran
s
fo
rm
at
i
o
n
a
n
d
t
h
e
L
ap
l
a
c
i
a
n
o
p
e
rat
o
r.
T
h
en
,
t
h
e
ex
t
rac
t
e
d
f
e
at
u
r
e
s
ar
e
acc
u
mu
l
at
e
d
fo
r
c
r
e
at
i
n
g
d
e
s
c
ri
p
t
o
rs
b
y
u
s
i
n
g
C
N
N
.
A
w
i
d
e
-
ra
n
g
i
n
g
a
n
al
y
s
i
s
i
s
p
e
rfo
r
med
fo
r
as
s
e
s
s
i
n
g
t
a
m
p
e
r
i
n
g
d
e
t
ec
t
i
o
n
an
d
t
am
p
e
r
e
d
r
eg
i
o
n
s
e
g
me
n
t
a
t
i
o
n
a
ccu
raci
e
s
o
f
p
ro
p
o
s
e
d
RSF
-
C
N
N
b
as
e
d
t
am
p
e
r
i
n
g
d
e
t
ec
t
i
o
n
p
ro
ce
d
u
r
e
s
c
o
n
s
i
d
e
r
i
n
g
v
ar
i
o
u
s
fal
s
i
fi
c
at
i
o
n
s
an
d
p
o
s
t
-
p
ro
c
e
s
s
i
n
g
at
t
ack
s
w
h
i
ch
i
n
cl
u
d
e
j
o
i
n
t
p
h
o
t
o
g
rap
h
i
c
e
x
p
e
rt
g
ro
u
p
(J
PE
G
)
c
o
m
p
re
s
s
i
o
n
,
s
c
al
i
n
g
,
ro
t
at
i
o
n
s
,
n
o
i
s
e
ad
d
i
t
i
o
n
s
,
an
d
mo
r
e
t
h
an
o
n
e
m
a
n
i
p
u
l
at
i
o
n
.
Fro
m
t
h
e
ac
h
i
e
v
ed
r
e
s
u
l
t
s
,
i
t
c
an
b
e
v
i
s
i
b
l
e
t
h
e
RSF
-
C
N
N
p
ri
m
a
ri
l
y
b
as
e
d
t
am
p
e
ri
n
g
d
e
t
ec
t
i
o
n
w
i
t
h
ad
e
q
u
at
el
y
h
i
g
h
e
r
a
cc
u
rat
e
n
e
s
s
t
h
an
e
x
i
s
t
i
n
g
t
a
m
p
e
r
i
n
g
d
e
t
ec
t
i
o
n
me
t
h
o
d
o
l
o
g
i
e
s
.
K
e
y
w
o
r
d
s
:
C
o
n
v
o
l
ut
i
o
n
n
e
ur
a
l
n
e
t
wo
r
k
De
e
p
l
e
a
r
ni
ng
H
y
b
r
i
d
a
tt
a
ck
I
m
a
ge
t
a
m
p
e
r
i
n
g
de
t
e
c
t
i
o
n
R
e
s
a
m
p
li
ng
f
e
a
t
ur
e
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
cen
s
e.
C
or
r
e
s
pon
din
g
A
u
th
or
:
M
a
nj
u
n
a
t
h
a
S
De
pa
r
t
m
e
n
t
o
f
I
nf
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
a
n
d
E
n
g
i
ne
e
r
in
g,
G
l
o
b
a
l
A
c
a
de
my
o
f
T
e
c
hn
o
l
o
g
y
B
e
n
ga
l
ur
u,
I
n
d
i
a
E
m
a
i
l
:
m
a
nj
u.
dv
g2020
@
g
m
a
il
.
c
o
m
1.
I
NT
RODU
C
T
I
ON
T
h
e
e
m
e
r
ge
nc
e
o
f
s
o
c
i
a
l
ne
t
wor
ks
i
n
o
ur
da
i
ly
li
f
e
s
pur
s
t
h
e
a
d
ve
n
t
o
f
v
a
r
i
o
us
d
i
g
i
t
a
l
im
a
ge
e
d
i
t
i
n
g
too
l
s
w
hi
c
h
l
e
a
d
s
t
o
b
e
li
e
f
i
s
s
ue
s
o
f
m
u
l
t
im
e
d
i
a
c
o
n
t
e
n
t
b
e
i
n
g
c
i
r
c
u
l
a
t
e
d
.
De
s
i
g
ni
ng
hi
g
h
-
qua
l
i
t
y
t
a
m
pe
r
i
ng
e
m
p
l
o
yi
ng
a
m
a
c
hi
ne
l
e
a
r
ni
ng
m
o
de
l
t
h
a
t
i
s
vi
s
u
a
l
ly
u
n
de
t
e
c
t
a
bl
e
t
h
r
o
ugh
t
h
e
h
u
m
a
n
e
y
e
[
1]
,
[
2]
i
s
we
ll
w
i
t
hi
n
r
e
a
c
h
o
f
t
h
e
us
e
r
t
h
r
o
ugh
t
h
e
f
o
l
l
o
w
i
ng
too
l
s
uc
h
a
s
F
a
c
e
A
pp
[
3]
,
A
do
b
e
S
e
n
s
i
e
i
[
4]
,
De
e
pP
h
o
to
e
d
i
t
o
r
[
5]
,
a
n
d
A
do
be
s
k
y
R
e
p
l
a
c
e
[
6]
.
T
h
us
,
c
l
a
s
s
i
f
yi
ng
a
n
im
a
ge
a
s
t
a
m
pe
r
e
d
w
i
t
h
o
r
n
o
t
i
s
b
e
c
o
mi
ng
a
n
e
x
t
r
e
m
e
ly
d
i
f
f
i
c
u
l
t
t
a
s
k.
I
n
ge
n
e
r
a
l
,
t
a
m
pe
r
i
ng
i
s
c
l
a
s
s
if
i
e
d
i
n
t
o
c
o
n
t
e
n
t
pr
e
s
e
r
vi
n
g
a
n
d
c
h
a
n
g
i
ng
[
7]
.
T
h
e
c
o
py
-
c
l
o
n
e
,
s
p
li
c
i
ng,
a
n
d
o
bj
e
c
t
e
x
c
l
us
i
o
n
a
r
e
w
ide
l
y
u
s
e
d
pr
im
a
r
y
a
t
t
a
c
ks
,
a
n
d
bl
ur
r
i
n
g,
c
o
m
pr
e
s
s
i
o
n
,
a
n
d
c
o
n
t
r
a
s
t
a
d
j
us
t
m
e
n
t
a
r
e
c
o
m
m
o
nly
u
s
e
d
s
e
c
o
n
d
a
r
y
a
t
t
a
c
ks
.
Ho
we
v
e
r
,
t
h
e
pr
i
m
a
r
y
a
tt
a
c
k
w
i
ll
r
e
s
u
l
t
i
n
a
s
e
m
a
n
t
i
c
a
l
il
l
us
t
r
a
t
i
o
n
o
f
a
n
i
m
a
ge
.
T
h
us
,
t
hi
s
wo
r
k
f
o
c
us
e
s
o
n
de
t
e
c
t
i
n
g
t
h
e
p
r
i
m
a
r
y
a
t
t
a
c
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
25
,
N
o
.
1
,
J
a
n
ua
r
y
20
22
:
183
-
190
184
R
e
c
e
n
t
l
y
,
s
e
v
e
r
a
l
m
e
t
h
o
ds
h
a
ve
b
e
e
n
pr
e
s
e
n
t
e
d
f
o
r
de
t
e
c
t
i
n
g
pr
i
m
a
r
y
t
a
m
pe
r
i
ng
a
t
t
a
c
ks
[
8]
-
[
10
]
.
Ho
we
v
e
r
,
t
h
e
s
e
m
e
t
h
o
ds
h
a
v
e
f
o
c
us
e
d
o
n
de
t
e
c
t
i
n
g
w
h
e
t
h
e
r
a
n
i
m
a
ge
i
s
t
a
m
pe
r
e
d
o
r
n
ot
a
n
d
f
a
il
e
d
t
o
l
o
c
a
li
z
e
t
a
m
p
e
r
e
d
r
e
g
i
o
n
w
i
t
hi
n
a
t
a
m
pe
r
e
d
im
a
ge
[
11]
,
[
12]
.
I
n
[
13]
,
[
14]
a
bl
e
t
o
l
o
c
a
l
i
z
e
t
h
e
p
a
t
c
h
e
s
t
h
a
t
a
r
e
t
a
m
pe
r
e
d
w
i
t
h
t
h
r
o
ugh
t
h
e
e
m
p
l
o
y
m
e
n
t
o
f
f
r
e
que
n
c
y
do
m
a
i
n
[
15
]
,
[
16]
.
S
udi
a
t
m
i
k
a
e
t
al
.
[
17]
us
e
d
n
o
i
s
y
i
n
f
o
r
m
a
t
i
o
n
[
18]
o
f
c
o
m
pr
e
s
s
e
d
i
m
a
ge
f
o
r
e
s
t
a
bli
s
hi
ng
w
he
t
h
e
r
t
h
e
im
a
ge
h
a
s
t
a
m
pe
r
e
d
o
r
n
o
t.
I
n
r
e
c
e
n
t
t
i
m
e
s
de
e
p
l
e
a
r
ni
ng
t
e
c
hni
que
s
h
a
v
e
a
c
hi
e
v
e
d
a
ve
r
y
go
o
d
r
e
s
ul
t
i
n
c
o
m
put
e
r
vi
s
i
o
n
a
pp
li
c
a
t
i
o
ns
[
19
]
-
[
21]
i
nc
l
ud
i
ng
im
a
ge
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
a
n
d
c
l
a
s
s
if
ica
t
i
o
n
[
22]
,
[
2
3]
.
T
h
e
a
ut
o
e
n
c
o
de
r
[
24]
a
n
d
c
o
nv
o
l
ut
i
o
n
n
e
ur
a
l
ne
t
wor
k
(
C
NN
)
[
22
]
,
[
23]
h
a
v
e
b
e
e
n
w
i
de
ly
us
e
d
t
o
de
t
e
c
t
pr
i
m
a
r
y
a
t
t
a
c
ks
s
uc
h
a
s
s
p
li
c
i
ng
[
24]
,
[
25]
a
n
d
c
o
p
y
-
m
o
v
e
[
26]
.
H
o
we
v
e
r
,
t
h
e
s
e
m
o
de
l
f
a
il
s
to
p
r
o
vi
de
go
o
d
a
c
c
ur
a
c
y
whe
n
i
m
a
g
e
un
d
e
r
go
e
s
a
hy
br
i
d
t
r
a
n
s
f
o
r
m
a
t
i
o
n
a
s
t
h
e
y
a
r
e
de
s
i
g
ne
d
to
de
t
e
c
t
e
i
t
h
e
r
s
p
li
c
i
ng
o
r
c
o
py
-
m
o
v
e
.
F
ur
t
h
e
r
,
t
h
e
t
r
a
di
t
i
o
na
l
f
u
ll
y
-
c
o
n
n
e
c
t
e
d
C
NN
-
b
a
s
e
d
f
r
a
m
e
wo
r
k
[
27]
f
a
il
s
t
o
ge
n
e
r
a
l
i
z
e
d
if
f
e
r
e
n
t
n
o
i
s
e
i
n
du
c
e
d
t
h
r
o
ugh
d
i
f
f
e
r
e
n
t
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
e
t
h
o
ds
;
t
h
us
,
poo
r
t
a
m
pe
r
i
ng
r
e
g
i
o
n
l
o
c
a
l
i
z
a
t
i
o
n
o
ut
c
o
m
e
i
s
a
c
hi
e
v
e
d.
I
n
a
ddr
e
s
s
i
n
g
s
uc
h
i
s
s
ue
s
i
n
t
hi
s
w
o
r
k
pr
e
s
e
n
t
e
d
a
n
im
pr
o
v
e
d
t
a
m
p
e
r
i
n
g
de
t
e
c
t
i
o
n
m
o
de
l
e
m
p
l
o
yi
ng
im
pr
o
v
e
d
pr
e
pr
o
c
e
s
s
i
ng,
f
e
a
t
ur
e
a
ggr
e
ga
t
i
o
n
,
a
n
d
C
NN
a
r
c
hi
t
e
c
t
ur
e
[
11]
.
B
un
k
e
t
al
.
[
11]
s
h
o
we
d,
i
m
a
ge
t
a
m
pe
r
i
ng
i
nduc
e
s
n
o
i
s
e
b
e
c
a
us
e
o
f
pe
r
i
o
d
i
c
i
n
t
e
r
po
l
a
t
i
o
n
a
m
o
n
g
a
d
j
a
c
e
n
t
p
i
x
e
l
s
w
hi
c
h
c
a
n
b
e
u
nde
r
s
too
d
t
h
r
o
ugh
r
e
s
a
m
p
li
ng
f
e
a
t
ur
e
s
[
13
]
,
[
28
]
.
He
r
e
a
f
f
i
ne
t
r
a
n
s
f
o
r
m
a
t
i
o
n
a
n
d
L
a
p
l
a
c
i
a
n
f
u
nc
t
i
o
n
i
s
us
e
d
f
o
r
e
x
t
r
a
c
t
i
n
g
r
e
s
a
m
p
li
ng
f
e
a
t
ur
e
s
a
n
d
de
s
c
r
i
pt
o
r
i
s
b
u
il
t
t
h
r
o
ugh
C
NN
.
T
h
e
s
i
g
nif
i
c
a
n
c
e
o
f
r
a
n
ge
s
pa
t
i
a
l
f
i
l
t
e
r
i
n
g
(
R
S
F
)
-
t
a
m
pe
r
d
e
t
e
c
t
i
o
n
(
TD
)
i
s
de
s
c
r
i
be
d
n
e
x
t
.
T
h
e
pa
pe
r
pr
e
s
e
n
t
e
d
a
C
NN
-
b
a
s
e
d
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
m
e
t
h
o
d
by
l
e
a
r
ni
ng
R
e
s
a
m
p
l
i
ng
F
e
a
t
ur
e
s
.
T
h
e
R
S
F
-
T
D
c
a
n
e
x
t
r
a
c
t
us
e
f
u
l
f
e
a
t
ur
e
s
a
m
o
n
g
a
d
j
a
c
e
n
t
p
i
x
e
l
s
o
f
b
o
t
h
h
o
r
i
z
o
n
t
a
l
a
n
d
v
e
r
t
i
c
a
l
d
i
r
e
c
t
i
o
n
s
w
i
t
h
b
e
t
t
e
r
a
c
c
ur
a
c
y
.
T
he
R
S
F
-
T
D
c
a
n
b
e
us
e
d
f
o
r
de
t
e
c
t
i
n
g
ta
m
pe
r
e
d
im
a
ge
t
h
a
t
h
a
s
u
n
de
r
go
n
e
m
u
l
t
i
p
l
e
t
a
m
pe
r
i
ng.
T
h
e
R
S
F
-
T
D
a
c
hi
e
v
e
s
v
e
r
y
go
o
d
p
r
e
c
i
s
i
o
n
,
F
1
-
s
c
o
r
e
,
a
n
d
r
e
c
a
l
l
pe
r
f
o
r
m
a
n
c
e
i
n
c
o
m
pa
r
i
s
o
n
w
i
t
h
t
h
e
r
e
c
e
n
t
ta
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
e
t
h
o
d.
T
h
e
pa
pe
r
i
s
a
r
r
a
n
ge
d
a
s
f
o
l
l
o
ws
:
t
h
e
pr
o
po
s
e
d
r
e
s
a
m
p
li
ng
f
e
a
t
ur
e
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
e
t
h
o
d
t
h
r
o
ugh
C
NN
i
s
c
o
nf
e
r
r
e
d
i
n
s
e
c
t
i
o
n
2.
T
h
e
o
ve
r
a
l
l
o
ut
c
o
m
e
a
c
hi
e
v
e
d
us
i
ng
t
h
e
R
S
F
-
T
D
m
e
t
ho
d
o
v
e
r
d
i
f
f
e
r
e
n
t
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
o
de
l
a
r
e
g
i
ve
n
.
T
h
e
l
a
s
t
s
e
c
t
i
o
n
d
i
s
c
us
s
e
s
t
h
e
s
i
g
ni
f
i
c
a
nc
e
o
f
R
S
F
-
T
D
a
n
d
a
l
s
o
di
s
c
us
s
e
s
t
h
e
f
ut
ur
e
di
r
e
c
t
i
o
n
o
f
r
e
s
e
a
r
c
h
wo
r
k
.
2.
E
F
F
I
CI
E
NT
RSF
AN
D
CN
N
M
ODE
L
F
OR
I
M
AGE
F
ORGE
RY
DE
T
E
CT
I
ON
He
r
e
t
h
e
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
t
h
r
o
ugh
r
e
s
a
m
p
li
ng
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
a
n
d
C
NN
d
e
s
c
r
i
pt
or
i
s
pr
e
s
e
n
t
e
d.
F
o
r
de
t
e
c
t
i
n
g
t
a
m
pe
r
i
ng
a
n
d
s
e
g
m
e
n
t
i
ng
t
a
m
pe
r
e
d
r
e
g
i
o
n
e
f
f
i
c
i
e
n
t
l
y
t
h
e
f
o
l
l
o
w
i
ng
d
e
s
i
g
n
i
s
pr
e
s
e
n
t
e
d
i
n
F
i
g
ur
e
1.
T
hi
s
R
S
F
-
T
D
a
r
c
hi
t
e
c
t
ur
e
i
s
h
a
vi
ng
s
i
x
s
t
e
ps
.
F
i
r
s
t
,
t
h
e
i
m
a
ge
i
s
s
e
g
m
e
n
t
e
d
i
n
t
o
d
i
f
f
e
r
e
n
t
pa
t
c
h
e
s
.
T
h
e
n
,
t
h
e
f
e
a
t
ur
e
i
s
e
x
t
r
a
c
t
e
d
us
i
n
g
a
s
c
a
l
e
-
i
nv
a
r
i
a
n
t
de
s
c
r
i
pt
o
r
f
o
r
e
s
t
a
bl
i
s
hi
ng
t
h
e
dup
l
i
c
a
t
e
d
r
e
gi
o
n
e
v
e
n
u
n
de
r
t
h
e
s
m
a
ll
a
n
d
s
m
o
o
t
h
r
e
gi
o
n
.
F
i
gur
e
1.
M
e
t
h
o
d
o
l
o
gy
o
f
pr
o
p
o
s
e
d
R
S
F
-
C
NN
M
ode
l
f
o
r
t
a
m
p
e
r
i
n
g
de
t
e
c
t
i
o
n
2
.
1.
P
r
e
p
r
oc
e
s
s
in
g
an
d
r
e
s
a
m
p
l
in
g
f
e
at
u
r
e
d
e
t
e
c
t
ion
an
d
e
x
t
r
ac
t
io
n
Ge
n
e
r
a
ll
y
,
t
h
e
t
a
m
pe
r
e
d
i
m
a
ge
s
w
il
l
h
a
v
e
s
i
g
ni
f
i
c
a
n
t
i
m
pa
c
t
s
o
n
t
h
e
s
t
a
t
i
s
t
i
c
a
l
pr
o
pe
r
t
i
e
s
a
l
o
n
g
t
he
e
dge
s
.
S
i
mi
l
a
r
t
o
m
e
t
h
o
do
l
o
gi
e
s
pr
e
s
e
n
t
e
d
i
n
[
29]
i
n
t
hi
s
wo
r
k
t
h
e
r
e
s
a
m
p
l
i
ng
f
e
a
t
ur
e
a
r
e
e
x
t
r
a
c
t
e
d
t
h
r
o
ugh
a
f
f
i
ne
t
r
a
n
s
f
o
r
m
a
t
i
o
n
a
n
d
L
a
p
l
a
c
i
a
n
f
u
nc
t
i
o
n
.
H
e
r
e
t
h
e
i
m
a
ge
i
s
d
i
vi
de
d
i
n
t
o
n
o
n
-
o
v
e
r
l
a
pp
i
ng
w
i
t
h
p
a
t
c
h
s
i
z
e
s
e
t
to
64.
T
h
e
d
i
m
e
n
s
i
o
n
o
f
t
h
e
pa
t
c
h
w
i
ll
be
64*64
f
o
r
c
o
n
s
i
de
r
i
ng
a
n
im
a
ge
s
i
z
e
o
f
512*512.
K
e
e
p
i
n
g
t
h
e
s
i
z
e
to
512*512
s
i
g
ni
f
i
c
a
n
t
l
y
r
e
duc
e
t
h
e
c
o
m
put
a
ti
o
n
t
i
m
e
i
n
d
e
t
e
c
t
i
n
g
a
n
d
l
o
c
a
li
z
i
ng
t
a
m
pe
r
i
ng
r
e
g
i
o
n
.
T
o
pr
e
d
i
c
t
t
h
e
l
i
ne
a
r
e
r
r
o
r
L
a
p
l
a
c
i
a
n
f
u
nc
t
i
o
n
i
s
ut
il
i
z
e
d
[
13]
.
Af
f
i
ne
t
r
a
n
s
f
o
r
m
a
t
i
o
n
m
a
t
r
i
c
e
s
a
r
e
us
e
d
f
o
r
c
o
l
l
e
c
t
i
n
g
t
h
e
e
r
r
o
r
s
c
o
n
s
i
de
r
i
ng
v
a
r
i
o
us
d
i
r
e
c
t
i
o
n
s
a
n
d
a
n
ge
l
s
.
F
i
na
ll
y
,
F
o
ur
i
e
r
t
r
a
n
s
f
o
r
m
a
t
i
o
n
i
s
us
e
d
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
I
S
S
N:
2502
-
4752
E
ff
icie
nt
r
e
s
ampling
f
e
atur
e
s
and
c
onv
olut
ion
ne
ur
al
ne
t
w
or
k
mode
l
f
or
…
.
(
M
anjunat
ha
S
.)
185
e
x
t
r
a
c
t
i
n
g
R
S
F
c
o
n
s
i
de
r
i
ng
d
i
ve
r
s
e
t
a
m
pe
r
i
ng
t
r
a
n
s
f
o
r
m
a
t
i
o
n
.
T
h
e
e
xi
s
t
i
n
g
m
e
t
h
o
d
pe
r
f
o
r
m
s
l
o
c
a
li
z
a
t
i
o
n
a
t
p
i
x
e
l
l
e
v
e
l
;
h
o
we
v
e
r
,
t
h
e
R
S
F
-
T
D
pe
r
f
o
r
m
s
l
o
c
a
li
z
a
t
i
o
n
a
t
t
h
e
p
i
x
e
l
l
e
v
e
l
.
He
r
e
t
h
e
pa
t
c
h
s
i
z
e
i
s
s
e
t
to
32*32
a
n
d
bl
o
c
k
s
i
z
e
i
s
ke
pt
to
8*8
f
o
r
e
x
t
r
a
c
t
i
n
g
l
o
c
a
l
R
S
F
w
i
t
h
l
e
s
s
r
e
du
n
da
n
c
y
.
O
n
c
e
R
S
F
i
s
e
x
t
r
a
c
t
e
d
t
h
e
y
a
r
e
t
r
a
i
n
e
d
u
s
i
ng
C
NN
;
a
c
hi
e
vi
ng
hi
g
he
r
a
c
c
ur
a
c
y
d
e
pe
n
d
s
o
n
t
h
e
o
r
de
r
o
f
f
e
a
t
ur
e
a
ggr
e
ga
t
i
o
n
.
T
h
e
e
xi
s
t
i
n
g
m
o
de
l
a
r
r
a
n
ge
s
t
h
e
f
e
a
t
ur
e
s
i
n
t
h
e
h
o
r
i
z
o
n
t
a
l
o
r
v
e
r
t
i
c
a
l
d
i
r
e
c
t
i
o
n;
t
h
u
s
,
poo
r
c
or
r
e
l
a
t
i
o
n
a
m
o
n
g
d
i
f
f
e
r
e
n
t
d
i
r
e
c
t
i
o
n
s
s
i
g
ni
f
i
c
a
n
t
l
y
im
p
a
c
t
s
a
c
c
ur
a
c
y
.
T
o
a
ddr
e
s
s
t
h
e
a
f
o
r
e
m
e
n
t
i
o
n
e
d
i
s
s
ue
s
i
n
t
hi
s
s
pa
c
e
-
f
i
ll
i
n
g
c
ur
v
e
[
30
]
-
[
33
]
i
s
us
e
d.
T
he
s
pa
c
i
ng
f
il
l
i
ng
i
s
v
e
r
y
e
f
f
e
c
t
i
v
e
i
n
c
o
nv
e
r
t
i
n
g
t
h
e
m
u
l
t
i
-
d
im
e
n
s
i
o
n
pr
o
ble
m
s
to
a
s
i
n
g
l
e
d
im
e
n
s
i
o
n
[
3
4
]
-
[
3
7
].
2.
2.
Aggr
e
gat
ion
of
f
e
at
u
r
e
s
an
d
d
e
c
is
ion
u
s
in
g
CN
N
f
o
r
t
am
p
e
r
in
g
d
e
t
e
c
t
ion
T
h
e
pr
e
vi
o
us
s
e
c
t
i
o
n
e
x
t
r
a
c
t
s
e
f
f
i
c
i
e
n
t
R
S
F
f
e
a
t
ur
e
s
;
t
h
e
s
e
f
e
a
t
ur
e
s
m
us
t
b
e
c
u
m
u
l
a
t
e
d
to
b
u
i
l
d
a
n
e
f
f
i
c
i
e
n
t
l
e
a
r
ni
ng
m
e
c
h
a
ni
s
m
f
o
r
t
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
t
h
e
i
m
a
ge
i
s
un
d
e
r
go
n
e
hy
br
i
d
p
r
i
m
a
r
y
a
tt
a
c
ks
.
D
i
f
f
e
r
e
n
t
a
ggr
e
ga
t
i
o
n
m
e
t
h
o
ds
a
r
e
m
o
de
l
e
d
i
n
(
1)
to
(
4)
[
38]
.
T
h
e
(
1)
de
f
i
ne
s
t
h
e
mi
n
im
u
m
a
g
gr
e
ga
t
i
o
n
m
o
de
l
,
↓
=
m
in
=
1
,
…
,
(
1)
T
h
e
(
2)
de
f
i
ne
s
t
h
e
m
a
xi
m
u
m
a
ggr
e
ga
t
i
o
n
f
u
n
c
t
i
o
n
↑
=
m
a
x
=
1
,
…
,
(
2)
T
h
e
(
3)
de
f
i
ne
s
t
h
e
m
e
a
n
a
ggr
e
ga
t
i
o
n
f
u
nc
t
i
o
n
→
=
1
∑
=
1
(
3)
T
h
e
(
4)
s
qua
r
e
d
m
e
a
n
a
ggr
e
ga
t
i
o
n
f
u
n
c
t
i
o
n
←
=
1
∑
2
=
1
(
4)
I
n
(
1)
to
(
4
)
,
t
h
e
pa
r
a
m
e
t
e
r
=
[
,
1
,
…
,
,
]
de
f
i
ne
s
t
h
e
a
m
o
un
t
o
f
f
e
a
t
ur
e
e
x
t
r
a
c
t
e
d
i
n
t
h
e
ℎ
pa
t
c
h
a
n
d
de
f
i
ne
s
t
h
e
m
a
xim
u
m
pa
t
c
h
s
i
z
e
us
e
d.
I
n
t
hi
s
wo
r
k
we
us
e
d
d
i
f
f
e
r
e
n
t
poo
l
i
ng
m
e
t
h
o
ds
;
t
h
e
n
f
e
a
t
ur
e
s
a
r
e
a
ggr
e
ga
t
e
d
f
o
r
e
l
im
i
na
t
i
n
g
s
pa
t
i
a
l
de
pe
n
de
nc
y
.
L
e
t
de
f
i
ne
s
C
NN
pa
r
a
m
e
t
e
r
,
de
f
i
ne
s
l
o
s
s
f
u
n
c
t
i
o
n
o
f
C
NN
s
t
r
uc
t
ur
e
s
ue
d,
a
n
d
de
f
i
ne
s
c
u
mu
l
a
t
e
d
f
e
a
t
ur
e
s
.
T
h
e
n
,
t
h
e
gr
a
d
i
e
n
t
o
f
c
o
n
c
e
r
ni
ng
i
s
de
f
i
ne
d
a
s
f
o
l
l
o
ws
:
=
∑
,
=
1
,
(
5)
w
i
t
h
,
,
=
{
,
∙
,
↑
(
)
m
a
x
p
o
o
l
i
ng
,
∙
,
↓
(
)
m
in
p
o
o
l
in
g
1
∑
,
=
1
a
v
e
r
a
g
e
p
o
o
l
in
g
1
∑
2
,
,
=
1
a
v
g
.
s
q
r
p
o
o
l
in
g
(
6)
Us
i
n
g
(
6)
we
c
a
n
s
t
a
t
e
t
h
a
t
,
=
1
pr
o
vi
de
d
=
;
o
t
h
e
r
w
i
s
e
,
=
0
.
T
h
e
pa
r
a
m
e
t
e
r
s
↓
(
)
a
n
d
↑
(
)
de
f
i
ne
t
h
e
f
e
a
t
ur
e
v
e
c
t
or
wi
t
h
t
h
e
s
m
a
l
l
e
s
t
a
n
d
l
a
r
g
e
s
t
ℎ
c
o
m
po
n
e
n
t
.
Us
i
n
g
a
ll
po
o
l
i
n
g
t
o
ge
t
h
e
r
a
s
s
ur
e
s
t
h
a
t
gr
a
d
i
e
n
t
we
i
g
h
t
s
c
a
n
b
e
o
pt
i
m
i
z
e
d
m
o
r
e
e
f
f
i
c
i
e
nt
l
y
;
t
h
us
,
a
i
d
i
ng
i
n
a
c
hi
e
vi
ng
be
tt
e
r
t
a
m
pe
r
i
ng
d
e
t
e
c
t
i
o
n
a
c
c
ur
a
c
y
.
F
i
na
ll
y
,
t
h
e
o
p
t
i
m
i
z
e
d
R
S
F
a
r
e
a
ggr
e
ga
t
e
d
f
o
r
b
ui
l
d
i
ng
de
s
c
r
i
pt
o
r
us
i
n
g
t
w
o
-
l
a
y
e
r
f
u
l
ly
c
o
n
n
e
c
t
e
d
C
NN
[
39]
;
t
hi
s
a
s
s
ur
e
s
hi
g
he
r
a
c
c
ur
a
c
y
w
i
t
h
m
i
n
im
a
l
c
o
m
put
a
t
i
o
n
t
i
m
e
.
2.
3
.
T
r
ain
in
g
of
CN
N
T
h
e
R
S
F
-
T
D
w
i
t
h
o
ut
t
h
e
n
e
e
d
f
o
r
a
ny
f
i
xe
d
n
e
t
wo
r
k
c
a
n
b
e
t
r
a
i
ne
d
e
n
d
-
to
-
e
n
d
by
us
i
ng
t
h
e
wh
o
l
e
im
a
ge
.
T
he
n
,
t
h
e
de
c
i
s
i
o
n
i
s
t
a
ke
n
t
o
c
l
a
s
s
if
y
im
a
ge
i
s
t
a
m
pe
r
e
d
o
r
n
o
t.
He
r
e
l
o
s
s
f
u
n
c
t
i
o
ns
a
r
e
b
a
c
k
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
25
,
N
o
.
1
,
J
a
n
ua
r
y
20
22
:
183
-
190
186
pr
o
pa
ga
t
e
d
wi
t
hi
n
t
h
e
n
e
t
wo
r
k
o
f
r
e
s
pe
c
t
i
ve
pa
t
c
h
e
s
;
t
hi
s
h
e
l
p
s
i
n
o
b
t
a
i
ni
ng
b
e
t
t
e
r
-
c
o
r
r
e
l
a
t
e
d
f
e
a
t
ur
e
s
i
n
a
n
a
da
pt
i
v
e
n
a
t
ur
e
.
T
h
us
,
a
s
s
ur
e
s
R
S
F
-
TD
a
c
hi
e
v
e
s
be
tt
e
r
de
t
e
c
t
i
o
n
a
c
c
ur
a
c
y
w
i
t
h
l
e
s
s
m
i
s
c
l
a
s
s
i
f
i
c
a
t
i
o
n.
3.
RE
S
UL
T
S
A
ND
DI
S
CU
S
S
I
ONS
He
r
e
e
x
p
e
r
i
m
e
n
t
i
s
c
o
n
duc
t
e
d
to
s
t
udy
t
h
e
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
o
ut
c
o
m
e
o
f
R
S
F
-
T
D
a
n
d
s
t
a
n
da
r
d
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
e
t
h
o
ds
.
T
h
e
e
x
pe
r
i
m
e
n
t
i
s
c
o
n
duc
t
e
d
us
i
n
g
w
i
ndo
w
10
o
pe
r
a
t
i
n
g
s
y
s
t
e
m
w
i
t
h
I
n
t
e
l
qua
d
pr
o
c
e
s
s
o
r
a
n
d
16
GB
R
a
m
.
T
h
e
R
S
F
-
T
D
h
a
s
b
e
e
n
i
m
p
l
e
m
e
n
t
e
d
t
h
r
o
ugh
M
a
t
l
a
b
,
C
+
+
,
a
n
d
t
he
An
a
c
o
n
d
a
p
y
t
h
o
n
3
f
r
a
m
e
wo
r
ks
.
T
h
e
e
x
p
e
r
i
m
e
n
t
i
s
do
ne
us
i
ng
t
wo
s
t
a
n
da
r
d
da
t
a
s
e
t
s
s
uc
h
a
s
m
e
d
i
a
i
n
t
e
gr
a
t
i
o
n
a
n
d
c
o
m
m
u
ni
c
a
t
i
o
n
c
e
n
t
e
r
(
M
I
C
C
)
a
n
d
D0
da
t
a
s
e
t
b
e
c
a
us
e
i
t
ha
s
u
n
de
r
go
n
e
a
hy
b
r
i
d
t
r
a
n
s
f
o
r
m
a
t
i
o
n
a
t
t
a
c
k.
M
or
e
de
t
a
i
l
o
f
t
h
e
da
t
a
s
e
t
us
e
d
f
o
r
t
h
e
e
x
pe
r
i
m
e
n
t
i
s
g
i
ve
n
i
n
T
a
bl
e
1.
T
h
e
m
e
t
r
i
c
u
s
e
d
f
o
r
e
v
a
l
ua
t
i
n
g
t
h
e
pe
r
f
o
r
m
a
n
c
e
o
f
R
S
F
-
T
D
a
n
d
t
h
e
e
xi
s
t
i
n
g
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
m
o
de
l
[
40
]
,
[
41
]
i
s
r
e
c
a
ll
,
F1
-
s
c
o
r
e
,
a
n
d
f
a
l
s
e
-
po
s
i
t
i
v
e
r
a
t
e
.
T
a
bl
e
1.
Da
t
a
s
e
t
de
s
c
r
i
pt
i
o
n
D
a
t
a
s
e
t
I
m
a
ge
s
i
z
e
S
c
a
li
n
g
a
nd
r
ot
a
t
i
on
C
ompr
e
s
s
i
on
M
I
C
C
600
Y
e
s
No
D0
50
ye
s
Y
e
s
3.
1.
E
val
u
at
ion
of
p
e
r
f
o
r
m
an
c
e
on
M
I
CC
d
at
as
e
t
T
h
e
M
I
C
C
i
s
c
o
m
po
s
e
d
o
f
300
ge
n
u
i
ne
im
a
ge
s
a
n
d
300
i
m
a
ge
s
un
d
e
r
go
a
hy
b
r
i
d
t
r
a
n
s
f
o
r
m
a
t
i
o
n
a
tt
a
c
k.
T
h
e
r
e
f
o
r
e
,
a
tot
a
l
o
f
600
i
m
a
ge
s
a
r
e
a
v
a
il
a
bl
e
.
O
n
a
v
e
r
a
ge
1.
2%
o
f
pa
t
c
h
e
s
a
r
e
t
a
m
pe
r
e
d
w
i
t
h
.
T
h
us
,
i
t
i
s
e
x
t
r
e
m
e
ly
d
if
f
i
c
u
l
t
i
n
de
t
e
c
t
i
n
g
t
a
m
pe
r
i
n
g.
T
h
e
F
i
gur
e
2(
a
)
s
h
o
ws
t
h
e
o
r
i
g
i
na
l
im
a
ge
,
F
i
g
ur
e
2
(
b
)
s
h
o
ws
r
e
s
pe
c
t
i
ve
s
e
g
m
e
n
t
a
t
i
o
n
o
ut
c
o
m
e
us
i
ng
R
S
F
-
T
D
r
e
s
pe
c
t
i
v
e
ly
.
I
n
F
i
gur
e
2,
t
h
e
t
a
m
pe
r
i
ng
r
e
g
i
o
n
s
e
g
m
e
n
t
a
t
i
o
n
o
u
t
c
o
m
e
a
c
hi
e
v
e
d
u
s
i
ng
R
S
F
-
T
D
i
s
s
h
o
wn
.
S
im
il
a
r
l
y
,
T
h
e
F
i
g
ur
e
3(
a
)
s
h
o
ws
t
h
e
o
r
i
g
i
na
l
i
m
a
ge
,
F
i
g
ur
e
3(
b
)
s
h
o
ws
r
e
s
pe
c
t
i
ve
gr
o
un
d
tr
u
t
h
o
f
t
a
m
pe
r
e
d
r
e
gi
o
n
,
s
e
g
m
e
n
t
a
t
i
o
n
o
ut
c
o
m
e
a
c
hi
e
v
e
d
u
s
i
ng
e
xi
s
t
i
n
g
a
n
d
R
S
F
-
T
D
m
o
de
l
i
s
s
h
o
wn
i
n
F
i
g
ur
e
3(
c
)
a
n
d
F
i
g
ur
e
3(
d)
,
r
e
s
pe
c
t
i
ve
ly
.
F
i
gur
e
3
s
h
o
ws
t
h
e
s
e
g
men
t
a
t
i
o
n
o
u
t
c
o
m
e
a
c
hi
e
v
e
d
us
i
ng
R
S
F
-
T
D
a
n
d
t
h
e
e
xi
s
t
i
n
g
s
e
g
m
e
n
t
a
t
i
o
n
m
o
de
l
[
26]
.
T
a
bl
e
2,
s
h
o
ws
t
h
e
r
e
c
a
l
l
,
f
a
l
s
e
po
s
i
t
i
v
e
r
a
t
e
(
F
P
R
)
,
a
n
d
F
1
-
S
c
o
r
e
pe
r
f
o
r
m
a
n
c
e
a
c
hi
e
v
e
d
us
i
ng
R
S
F
-
T
D
a
n
d
t
h
e
e
xi
s
t
i
n
g
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
o
de
l
[
40]
.
F
r
o
m
t
h
e
r
e
s
u
l
t
,
we
c
a
n
s
t
a
t
e
t
h
e
R
S
F
-
T
D
a
r
c
hi
v
e
s
h
a
v
e
m
uc
h
b
e
t
t
e
r
de
t
e
c
t
i
o
n
a
c
c
ur
a
c
ies
u
n
de
r
hy
br
i
d
a
tt
a
c
k
i
n
c
o
m
p
a
r
i
s
o
n
w
i
t
h
e
xi
s
t
i
n
g
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
m
o
de
l
s
.
(
a
)
(
b
)
F
i
gur
e
2.
T
h
e
r
e
s
ul
t
s
o
f
t
h
e
pr
o
p
o
s
e
d
R
S
F
-
C
NN
m
o
de
l
;
(
a
)
o
r
i
g
i
na
l
im
a
g
e
a
n
d
(
b
)
s
e
g
m
e
n
t
a
t
i
o
n
o
ut
c
o
m
e
us
i
n
g
R
S
F
-
TD
3.
2.
E
val
u
at
ion
of
p
e
r
f
o
r
m
an
c
e
on
D0
d
at
as
e
t
I
n
t
hi
s
s
e
c
t
i
o
n
pe
r
f
o
r
m
a
n
c
e
o
f
R
S
F
-
T
D
a
n
d
t
h
e
e
xi
s
t
i
n
g
t
a
m
pe
r
i
ng,
de
t
e
c
t
i
o
n
m
e
t
h
o
ds
a
r
e
s
t
udi
e
d
us
i
n
g
t
h
e
D0
da
t
a
s
e
t.
T
h
e
da
t
a
s
e
t
un
de
r
g
o
e
s
hy
b
r
i
d
t
r
a
n
s
f
o
r
m
a
t
i
o
n
s
uc
h
a
s
s
c
a
li
ng
a
n
d
r
ot
a
t
i
o
n
w
i
t
h
J
P
E
G
c
o
m
pr
e
s
s
i
o
n
.
T
h
u
s
,
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
m
a
k
e
s
ve
r
y
c
ha
l
l
e
ng
i
ng.
T
h
e
F
i
gu
r
e
4
(
a
)
s
h
o
ws
t
h
e
or
i
g
i
na
l
i
m
a
ge
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
I
S
S
N:
2502
-
4752
E
ff
icie
nt
r
e
s
ampling
f
e
atur
e
s
and
c
onv
olut
ion
ne
ur
al
ne
t
w
or
k
mode
l
f
or
…
.
(
M
anjunat
ha
S
.)
187
F
i
gu
r
e
4
(
b
)
s
h
o
w
s
r
e
s
pe
c
t
i
v
e
g
r
oun
d
tr
u
t
h
o
f
tam
p
e
r
e
d
r
e
gi
o
n
,
f
e
a
tu
r
e
e
x
tr
a
c
t
i
o
n
a
n
d
s
e
gm
e
n
ta
t
i
o
n
ou
tc
o
m
e
a
c
hi
e
v
e
d
us
i
n
g
e
x
i
s
t
i
n
g
a
n
d
R
S
F
-
T
D
m
o
de
l
i
s
s
h
o
wn
i
n
F
i
gu
r
e
4
(
c
)
,
a
n
d
F
i
gu
r
e
4
(
d
)
,
r
e
s
pe
c
t
i
v
e
l
y
.
I
n
F
i
gu
r
e
4
,
t
h
e
s
e
g
m
e
n
t
a
t
i
o
n
o
u
t
c
o
m
e
o
f
t
h
e
t
a
m
p
e
r
e
d
r
e
g
i
o
n
us
i
ng
R
S
F
-
T
D
i
s
s
h
o
w
n
.
T
h
e
de
t
e
c
t
i
o
n
a
c
c
ur
a
c
y
us
i
ng
R
S
F
-
T
D
a
n
d
t
h
e
e
xi
s
t
i
n
g
m
o
de
l
o
n
D0
da
t
a
s
e
t
i
s
s
h
o
w
n
i
n
T
a
bl
e
3.
T
h
r
o
ugh
t
he
o
ut
c
o
m
e
a
c
hi
e
v
e
d
we
c
a
n
s
t
a
t
e
t
h
a
t
R
S
F
-
T
D
pe
r
f
o
r
m
s
s
i
g
nif
i
c
a
n
t
l
y
b
e
t
t
e
r
t
h
a
n
t
h
e
e
xi
s
t
i
n
g
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
o
de
l
i
n
t
e
r
m
s
o
f
F1
-
s
c
o
r
e
,
F
P
R
,
pr
e
c
i
s
i
o
n
,
a
n
d
r
e
c
a
ll
.
(
a
)
(
b
)
(
c
)
(
d)
F
i
gur
e
3.
T
a
m
pe
r
i
n
g
r
e
g
i
o
n
s
e
g
m
e
n
t
a
t
i
o
n
o
u
t
c
o
m
e
us
i
n
g
R
S
F
-
T
D
a
n
d
e
xi
s
t
i
n
g
t
a
m
pe
r
i
n
g
r
e
g
i
o
n
s
e
g
m
e
n
t
a
t
i
o
n
m
o
de
l
;
(
a
)
o
r
i
g
i
na
l
im
a
g
e
,
(
b
)
gr
o
un
d
t
r
u
t
h
,
(
c
)
e
xi
s
t
i
n
g
m
o
de
l
[
26]
,
a
n
d
(
d)
R
S
F
-
TD
T
a
bl
e
2.
R
S
F
-
T
D
a
n
d
e
xi
s
t
i
n
g
m
e
t
h
o
d
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
a
c
c
ur
a
c
y
f
o
r
M
I
C
C
da
t
a
s
e
t
M
ode
l
R
e
c
a
ll
/T
P
R
F
P
R
F1
-
S
c
or
e
R
a
ju
a
nd
N
a
i
r
2018
[
40]
89.
14
-
92.
6
R
S
F
-
TD
97.
5
1.
4
97.
7
(
a
)
(b
)
(c
)
(d
)
F
i
gur
e
4
.
S
e
gm
e
n
t
a
t
i
o
n
o
u
t
c
o
m
e
o
f
R
S
F
-
T
D
m
o
de
l
;
(
a
)
i
n
put
i
m
a
ge
,
(
b
)
gr
o
un
d
tr
u
t
h
,
(
c
)
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
,
a
n
d
(
d)
s
e
g
m
e
n
t
a
t
i
o
n
o
u
t
c
o
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
25
,
N
o
.
1
,
J
a
n
ua
r
y
20
22
:
183
-
190
188
T
a
bl
e
3.
R
S
F
-
T
D
a
n
d
e
xi
s
t
i
n
g
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
m
o
de
l
f
o
r
D0
da
t
a
s
e
t
M
ode
l
R
e
c
a
ll
P
r
e
c
i
s
i
on
F
P
R
F1
H
ua
n
g
a
nd
C
i
ou
[
41]
84.
88
92.
81
3.
39
88.
67
R
S
F
-
C
N
N
98.
08
98.
84
1.
68
99.
28
3.
3.
Com
p
u
t
at
io
n
c
om
p
l
e
x
it
y
p
e
r
f
o
r
m
an
c
e
e
valu
at
ion
He
r
e
e
x
pe
r
i
m
e
n
t
i
s
do
n
e
t
o
s
t
udy
t
h
e
c
o
m
put
a
ti
o
n
c
o
m
p
l
e
xi
t
y
o
f
d
i
f
f
e
r
e
n
t
t
a
m
p
e
r
i
n
g
de
t
e
c
t
i
o
n
m
e
t
h
o
ds
.
T
h
e
c
o
m
put
a
t
i
o
n
c
o
m
p
l
e
xi
t
y
i
s
m
e
a
s
ur
e
d
i
n
t
e
r
m
s
o
f
t
h
e
a
m
o
un
t
o
f
t
i
m
e
t
a
ke
n
t
o
de
t
e
c
t
t
a
m
pe
r
i
ng
r
e
g
i
o
n
s
.
F
i
gur
e
5
s
h
o
ws
t
h
e
c
o
m
put
a
t
i
o
n
t
i
m
e
t
a
ke
n
t
o
pr
o
c
e
s
s
M
I
C
C
-
600
u
s
i
n
g
t
h
e
de
e
p
l
e
a
r
nin
g
-
b
a
s
e
d
m
e
t
h
o
d
[
42]
,
t
h
e
hi
e
r
a
r
c
hi
c
a
l
m
e
t
h
o
d,
a
n
d
R
S
F
-
T
D.
S
i
mi
l
a
r
ly
,
F
i
gur
e
6
s
h
o
ws
t
h
e
c
o
m
put
a
t
i
o
n
t
i
me
t
a
ke
n
to
pr
o
c
e
s
s
M
I
C
C
-
2000
us
i
n
g
t
h
e
de
e
p
l
e
a
r
ni
ng
-
ba
s
e
d
m
e
t
h
o
d
[
42]
a
n
d
R
S
F
-
T
D.
F
i
gur
e
7
s
ho
ws
t
h
e
c
o
m
put
a
t
i
o
n
t
i
m
e
t
a
ke
n
t
o
pr
o
c
e
s
s
t
h
e
C
o
M
o
F
o
D
da
t
a
s
e
t
us
i
n
g
t
he
DB
S
C
A
N
-
b
a
s
e
d
m
e
t
h
o
d,
a
n
d
R
S
F
-
T
D.
F
r
o
m
F
i
gur
e
s
5
t
o
7
we
c
a
n
s
e
e
t
h
e
R
S
F
-
T
D
s
i
g
nif
i
c
a
n
t
l
y
r
e
duc
e
s
c
o
m
put
a
t
i
o
n
t
i
m
e
i
n
c
o
m
pa
r
i
s
o
n
w
i
t
h
t
h
e
de
e
p
l
e
a
r
ni
ng
-
ba
s
e
d
m
e
t
h
o
d,
hi
e
r
a
r
c
hi
c
a
l
m
e
t
h
o
d,
a
n
d
DB
S
C
AN
-
b
a
s
e
d
m
e
t
h
o
d.
F
i
gur
e
5.
C
o
m
put
a
t
i
o
n
t
i
m
e
f
o
r
M
I
C
C
-
600
F
i
gur
e
6.
C
o
m
put
a
t
i
o
n
t
i
m
e
f
o
r
M
I
C
C
-
2000
F
i
gur
e
7.
C
o
m
put
a
t
i
o
n
t
i
m
e
f
o
r
C
o
M
o
F
o
D
4.
CONC
L
USI
ON
He
r
e
t
h
e
pa
pe
r
e
m
p
h
a
s
i
z
e
d
u
s
i
ng
t
h
e
r
e
s
a
m
p
li
ng
f
e
a
t
ur
e
t
h
r
o
ugh
e
f
f
e
c
t
i
v
e
pr
e
pr
o
c
e
s
s
i
ng
a
n
d
C
NN
m
o
de
l
.
T
h
e
R
S
F
-
T
D
b
r
i
ngs
i
n
go
o
d
tr
a
de
o
f
f
s
b
e
t
we
e
n
a
c
hi
e
vi
ng
hi
g
he
r
de
t
e
c
t
i
o
n
a
c
c
ur
a
c
i
e
s
w
i
t
h
mi
n
im
a
l
c
o
m
put
a
t
i
o
n
t
i
m
e
.
F
ur
t
h
e
r
,
a
bl
e
to
a
c
hi
e
v
e
hi
g
he
r
de
t
e
c
t
i
o
n
a
c
c
ur
a
c
i
e
s
w
i
t
h
b
e
t
t
e
r
s
e
g
m
e
n
t
a
t
i
o
n
o
ut
c
o
m
e
s
i
n
c
o
m
pa
r
i
s
o
n
w
i
t
h
t
h
e
s
t
a
n
da
r
d
t
a
m
pe
r
i
ng
de
t
e
c
t
i
o
n
m
o
de
l
.
T
h
e
r
e
s
u
l
t
pr
o
v
e
s
t
h
e
R
S
F
-
T
D
c
a
n
e
x
t
r
a
c
t
hi
g
hly
c
o
r
r
e
l
a
t
e
d
f
e
a
t
ur
e
s
a
n
d
e
li
mi
na
t
e
s
pa
t
i
a
l
de
pe
n
d
e
n
c
i
e
s
.
T
h
e
e
x
p
e
r
i
m
e
n
t
c
o
n
duc
t
e
d
o
n
t
wo
da
t
a
s
e
t
s
w
i
t
h
hy
br
i
d
t
a
m
p
e
r
i
n
g
a
t
t
a
c
k
t
r
a
n
s
f
o
r
m
a
t
i
o
n
s
h
o
ws
t
h
e
R
S
F
-
T
D
a
c
hi
e
ve
s
m
uc
h
b
e
t
t
e
r
a
c
c
ur
a
c
i
e
s
i
n
c
o
m
pa
r
i
s
o
n
w
i
t
h
t
h
e
r
e
c
e
n
t
t
a
m
pe
r
i
n
g
de
t
e
c
t
i
o
n
m
o
de
l
w
hi
c
h
is
m
e
a
s
ur
e
d
t
h
r
o
ugh
t
h
e
f
o
l
l
o
w
i
ng
m
e
t
r
i
c
s
s
uc
h
a
s
F
1
-
s
c
o
r
e
,
F
P
R
,
T
P
R
,
pr
e
c
i
s
i
o
n
,
a
n
d
r
e
c
a
ll
.
De
s
p
i
t
e
a
c
hi
e
vin
g
s
upe
r
i
o
r
de
t
e
c
t
i
o
n
a
c
c
ur
a
c
i
e
s
t
h
e
m
o
de
l
c
a
n
be
f
ur
t
h
e
r
im
pr
o
v
e
d
i
n
t
h
e
f
ut
ur
e
by
e
li
mi
na
t
i
n
g
o
ut
l
i
e
r
s
t
h
r
o
ugh
e
f
f
e
c
t
i
v
e
de
s
i
g
n
o
f
C
N
N
f
r
a
m
e
wo
r
k
t
h
a
t
i
s
r
o
b
us
t
a
ga
i
ns
t
n
o
i
s
e
.
F
ur
t
h
e
r
,
v
a
l
i
da
t
e
t
h
e
m
o
de
l
us
i
ng
m
o
r
e
di
v
e
r
s
e
da
t
a
s
e
t
s
.
RE
F
E
R
E
NC
E
S
[1
]
F
.
L
ua
n,
S
.
P
a
r
i
s
,
E
.
S
he
c
ht
m
a
n,
a
nd
K
.
B
a
l
a
,
“
D
e
e
p
P
hot
o
S
ty
l
e
T
r
a
ns
f
e
r
,
”
2017
I
E
E
E
C
onf
e
r
e
n
c
e
on
C
om
put
e
r
V
is
io
n
and
P
a
tt
e
r
n R
e
c
ogni
ti
on (
C
V
P
R
)
,
2017
,
pp
.
6997
-
7005
,
doi
:
10.
110
9/
C
V
P
R
.
2017
.
740
.
[2
]
N
.
P
.
J
ogl
e
ka
r
a
nd
P
.
N
.
C
ha
tu
r
,
“
A
C
ompr
e
s
s
i
ve
S
ur
ve
y
on
A
c
t
i
ve
a
nd
P
a
s
s
i
ve
M
e
th
ods
f
or
I
m
a
ge
F
or
ge
r
y
D
e
t
e
c
t
io
n,
”
I
nt
e
r
nat
io
nal
J
our
nal
of
E
ngi
ne
e
r
in
g and C
om
put
e
r
Sc
i
e
nc
e
,
v
o
l
.
4
,
no
.
1
,
pp
.
10187
-
10190
,
2015
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
I
S
S
N:
2502
-
4752
E
ff
icie
nt
r
e
s
ampling
f
e
atur
e
s
and
c
onv
olut
ion
ne
ur
al
ne
t
w
or
k
mode
l
f
or
…
.
(
M
anjunat
ha
S
.)
189
[3
]
J
.
B
unk
e
t
al
.
,
“
D
e
te
c
t
i
on
a
nd
L
oc
a
li
z
a
t
i
on
of
I
m
a
g
e
F
or
ge
r
ie
s
U
s
i
n
g
R
e
s
a
mp
li
n
g
F
e
a
tu
r
e
s
a
nd
D
e
e
p
L
e
a
r
n
i
n
g
,
”
2017
I
E
E
E
C
onf
e
r
e
nc
e
on
C
om
put
e
r
V
is
io
n
and
P
at
te
r
n
R
e
c
ogni
ti
on
W
or
k
s
hops
(
C
V
P
R
W
)
,
2017,
pp.
1881
-
1889,
do
i
:
10.
1109/C
V
P
R
W
.
2017
.
235
.
[4
]
V
.
S
c
he
t
i
n
ge
r
,
M
.
M
.
O
l
i
ve
i
r
a
,
R
.
D
a
S
i
l
va
,
a
nd
T
.
J
.
C
a
r
va
lh
o,
“
H
uma
n
s
a
r
e
e
a
s
il
y
f
oo
l
e
d
b
y
d
i
g
i
t
a
l
i
m
a
ge
s
,
”
C
om
put
e
r
s
&
G
r
aphi
c
s
,
v
o
l
.
68
,
pp
.
142
-
151
,
2015
,
doi
:
10.
1016/j
.
c
a
g
.
2017.
0
8.
010
.
[5
]
R
.
M
.
J
os
e
ph
a
nd
A
.
S
.
C
hi
th
r
a
,
“
L
i
t
e
r
a
tu
r
e
s
ur
ve
y
on
i
m
a
ge
ma
n
i
pu
l
a
t
i
on
de
t
e
c
t
i
on
,
”
I
nt
e
r
nat
io
nal
R
e
s
e
ar
c
h
J
our
na
l
of
E
ngi
ne
e
r
in
g and T
e
c
hnol
ogy
(
I
R
J
E
T
)
,
v
o
l
.
2
,
no
.
04
,
pp
.
740
-
744,
2015
.
[6
]
M.
T
.
S
ha
s
h
i
dh
a
r
a
nd
K
.
B
.
R
a
me
s
h,
“
A
nove
l
f
r
a
m
e
w
or
k
f
o
r
opt
i
m
i
z
e
d
d
i
gi
t
a
l
f
o
r
e
n
s
i
c
f
or
m
i
t
i
ga
t
i
n
g
c
o
mp
l
e
x
i
m
a
ge
a
tt
a
c
ks
,
”
I
nt
e
r
nat
io
nal
J
our
nal
of
E
le
c
tr
ic
al
and
C
om
put
e
r
E
ngi
n
e
e
r
in
g
(
I
J
E
C
E
)
,
v
o
l
.
10
,
no.
5,
pp
.
5198
-
5207
2020,
do
i
:
10.
11591/i
j
e
c
e
.
v
10
i
5.
pp5198
-
5207.
[7
]
H
.
L
i
,
W
.
L
uo,
X
.
Q
i
u,
a
nd
J
.
H
ua
n
g
,
“
I
m
a
ge
F
or
g
e
r
y
L
o
c
a
li
z
a
t
i
on
vi
a
I
nt
e
g
r
a
t
i
n
g
T
a
mp
e
r
i
n
g
P
os
s
i
b
il
i
t
y
M
a
p
s
,
”
i
n
I
E
E
E
T
r
ans
ac
ti
ons
on I
n
f
or
m
at
io
n F
or
e
ns
ic
s
and Se
c
ur
it
y
,
v
o
l
.
12
,
n
o.
5
,
pp
.
1240
-
125
2
,
M
a
y
2017,
doi
:
10.
1109/
T
I
F
S
.
2017
.
26568
23
.
[8
]
A
.
P
our
k
a
s
h
a
n
i
,
A
.
S
ha
hba
hr
a
m
i
,
a
nd
A
.
A
k
ou
s
h
i
d
e
h
,
“
C
op
y
-
move
f
or
ge
r
y
d
e
t
e
c
t
i
on
us
i
n
g
c
on
v
o
l
ut
i
on
a
l
n
e
ur
a
l
n
e
t
w
or
k
a
n
d
K
-
me
a
n
c
l
u
s
t
e
r
i
n
g
,
”
I
nt
e
r
nat
io
nal
J
our
nal
o
f
E
le
c
tr
ic
al
and
C
o
m
put
e
r
E
ngi
ne
e
r
in
g
(
I
J
E
C
E
)
,
vo
l.
11,
no.
3
,
pp
.
2604
-
2612
,
2
021,
do
i
:
10.
11591/i
je
c
e
.
v
11
i
3.
pp2604
-
2612.
[9
]
J
.
H
.
B
a
ppy
,
A
.
K
.
R
oy
-
C
how
dhu
r
y
,
J
.
B
unk
,
L
.
N
a
ta
r
a
j
,
a
nd
B
.
S
.
M
a
nj
una
th
,
“
E
x
p
l
o
i
t
i
n
g
S
pa
t
i
a
l
S
tr
uc
tu
r
e
f
or
L
o
c
a
li
z
i
n
g
M
a
n
i
pu
l
a
t
e
d
I
ma
ge
R
e
gi
ons
,
”
2017
I
E
E
E
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on
C
om
put
e
r
V
is
io
n
(
I
C
C
V
)
,
2017
,
pp
.
4980
-
4989
,
do
i
:
10.
1109/I
C
C
V
.
2017
.
532
.
[1
0
]
L
.
B
ondi
,
S
.
L
a
me
r
i
,
D
.
G
üe
r
a
,
P
.
B
e
s
ta
gi
n
i
,
E
.
J
.
D
e
l
p,
a
nd
S
.
T
uba
r
o,
“
T
a
mp
e
r
i
n
g
D
e
t
e
c
t
i
on
a
nd
L
o
c
a
li
z
a
t
i
on
T
hr
ough
C
l
us
t
e
r
i
n
g
of
C
a
me
r
a
-
B
a
s
e
d
C
N
N
F
e
a
tu
r
e
s
,
”
2017
I
E
E
E
C
onf
e
r
e
nc
e
on
C
om
put
e
r
V
is
io
n
and
P
at
te
r
n
R
e
c
ogni
ti
on
W
or
k
s
h
ops
(
C
V
P
R
W
)
,
2017
,
pp
.
1855
-
1864
,
doi
:
10.
1109/C
V
P
R
W
.
2017
.
2
32
.
[1
1
]
R
.
T
ha
k
ur
a
nd
R
.
R
ohi
ll
a
,
“
R
e
c
e
nt
a
d
v
a
n
c
e
s
i
n
d
i
g
i
t
a
l
i
ma
ge
m
a
n
i
pu
l
a
t
i
on
de
te
c
t
i
on
t
e
c
hn
i
qu
e
s
:
A
br
i
e
f
r
e
vi
e
w
,
”
F
or
e
ns
ic
Sc
ie
nc
e
I
nt
e
r
nat
io
nal
,
v
o
l
.
312
,
2020
,
doi
:
10.
1016/j
.
f
or
s
c
ii
nt
.
2020.
110311
.
[1
2
]
Y
.
L
i
u,
Q
.
G
ua
n,
X
.
Z
ha
o,
a
nd
Y
.
C
a
o
,
“
I
ma
ge
f
o
r
ge
r
y
l
o
c
a
li
z
a
t
i
on
ba
s
e
d
on
mu
l
t
i
-
s
c
a
l
e
c
on
v
o
l
ut
i
on
a
l
n
e
ur
a
l
n
e
t
w
or
ks
,
”
I
H
&M
M
Se
c
'
18:
P
r
oc
e
e
di
ngs
of
th
e
6t
h
A
C
M
W
or
k
s
hop
on
I
nf
or
m
at
io
n
H
id
in
g
and
M
ul
ti
m
e
di
a
Se
c
ur
it
y
,
201
8
,
pp
.
85
-
90
,
do
i
:
10.
1145/3206004
.
3206010
.
[1
3
]
B
.
M
a
hd
i
a
n
a
nd
S
.
S
a
i
c
,
“
D
e
te
c
t
i
on
of
c
op
y
-
mo
ve
f
o
r
g
e
r
y
u
s
in
g
a
m
e
th
od
b
a
s
e
d
on
b
l
u
r
mom
e
nt
i
n
v
a
r
i
a
nt
s
,
”
F
or
e
ns
ic
s
c
ie
nc
e
in
te
r
nat
io
nal
,
v
o
l
.
171
,
no
.
2
,
pp
.
180
-
189
,
2007
,
doi
:
10.
1016/j
.
f
or
s
c
ii
nt
.
2006.
11
.
002
.
[1
4
]
W
.
W
a
n
g
,
J
.
D
ong
,
a
nd
T
.
T
a
n,
“
E
x
p
l
o
r
i
n
g
D
C
T
C
oe
f
f
i
c
i
e
n
t
Q
ua
nt
i
z
a
t
i
on
E
f
f
e
c
t
s
f
or
L
o
c
a
l
T
a
mp
e
r
i
n
g
D
e
te
c
t
i
on,
”
i
n
I
E
E
E
T
r
ans
ac
ti
ons
on I
n
f
or
m
at
io
n F
or
e
ns
ic
s
and Se
c
ur
it
y
,
v
o
l
.
9
,
no
.
10
,
pp
.
1653
-
1666
,
O
c
t.
2014
,
doi
:
10.
1109/T
I
F
S
.
2014
.
234547
9
.
[1
5
]
I.
-
C
.
C
ha
ng
,
J
.
C
.
Y
u,
a
nd
C
.
-
C
.
C
ha
ng
,
“
A
f
or
ge
r
y
de
t
e
c
t
i
on
a
l
gor
i
th
m
f
or
e
xe
mp
l
a
r
-
b
a
s
e
d
i
np
a
i
nt
i
n
g
i
m
a
g
e
s
u
s
i
n
g
mu
l
t
i
-
r
e
gi
o
n
r
e
l
a
t
i
on
,
”
I
m
age
and V
is
io
n C
om
put
in
g
,
v
o
l
.
31
,
no
.
1
,
pp
.
57
-
7
1,
2013
,
doi
:
10.
1016/j
.
i
ma
v
i
s
.
2012.
09
.
002
.
[1
6
]
A
.
G
upt
a
,
“
N
e
w
C
opy
M
ove
F
or
ge
r
y
D
e
te
c
t
i
onT
e
c
hn
i
qu
e
U
s
i
n
g
A
d
a
pt
i
ve
O
ve
r
-
s
e
g
m
e
nt
a
t
i
on
a
nd
F
e
a
tu
r
e
P
o
i
nt
M
a
tc
h
i
n
g
,
”
B
ul
le
ti
n of
E
le
c
tr
ic
al
E
ngi
ne
e
r
in
g and I
nf
o
r
m
at
ic
s
(
B
E
E
I
)
,
v
o
l
.
7,
n
o.
3
,
pp
.
345
-
349
,
2018
,
doi
:
10.
11591/e
e
i
.
v
7
i
3.
754.
[1
7
]
I
.
B
.
K
.
S
udi
a
tm
i
ka
,
F
.
R
a
hma
n,
T
r
i
s
no,
a
nd
S
uy
ot
o
,
“
I
m
a
ge
f
or
ge
r
y
d
e
t
e
c
t
i
on
us
i
n
g
e
r
r
o
r
l
e
ve
l
a
na
l
ys
i
s
a
nd
d
e
e
p
l
e
a
r
n
i
n
g
,
”
T
E
L
K
O
M
N
I
K
A
(
T
e
le
c
om
m
uni
c
at
io
n,
C
om
put
in
g,
E
le
c
tr
oni
c
s
and
C
ont
r
ol
)
,
v
o
l
.
17,
n
o.
2
,
pp
.
653
-
659
,
2019
,
do
i
:
10.
12928/T
E
L
K
O
M
N
I
K
A
.
v
17
i
2.
8976.
[1
8
]
J
.
L
ong
,
E
.
S
he
l
h
a
m
e
r
,
a
nd
T
.
D
a
r
r
e
ll
,
“
F
u
ll
y
c
on
v
o
l
ut
i
on
a
l
ne
t
w
or
ks
f
or
s
e
m
a
nt
i
c
s
e
g
m
e
nt
a
t
i
on,
”
2015
I
E
E
E
C
onf
e
r
e
nc
e
on
C
om
put
e
r
V
is
io
n and P
at
te
r
n R
e
c
ogni
ti
on (
C
V
P
R
)
,
2015
,
pp
.
3
431
-
3440,
doi
:
10.
1109/
C
V
P
R
.
2015
.
7298965
.
[1
9
]
B
.
Z
hou
,
A
.
L
a
p
e
dr
i
z
a
,
J
.
X
i
a
o,
A
.
T
or
r
a
l
b
a
,
a
nd
A
.
O
l
i
va
,
“
L
e
a
r
n
i
n
g
d
e
e
p f
e
a
tu
r
e
s
f
or
s
c
e
n
e
r
e
c
o
g
n
i
t
i
on u
s
i
n
g
a
p
l
a
c
e
s
d
a
t
a
ba
s
e
,
”
I
n A
dv
anc
e
s
i
n ne
ur
al
i
n
f
or
m
at
io
n pr
o
c
e
s
s
in
g s
y
s
te
m
s
,
2014
.
[2
0
]
B
.
B
a
ya
r
a
nd
M
.
C
.
S
ta
mm
,
“
A
de
e
p
l
e
a
r
n
i
n
g
a
ppr
o
a
c
h
to
u
n
i
ve
r
s
a
l
i
m
a
g
e
ma
n
i
pu
l
a
t
i
on
de
te
c
t
i
on
u
s
i
n
g
a
n
e
w
c
on
v
o
l
u
t
io
na
l
l
a
ye
r
,
”
I
n
P
r
oc
e
e
di
ngs
o
f
th
e
4t
h
A
C
M
W
or
k
s
hop
on
I
n
f
o
r
m
at
io
n
H
id
in
g
and
M
ul
ti
m
e
di
a
Se
c
ur
it
y
,
2016
,
pp
5
-
10
,
do
i
:
10.
1145/2909827
.
2930786
.
[2
1
]
Y
.
R
a
o
a
nd
J
.
N
i
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
a
ppr
o
a
c
h
to
d
e
t
e
c
t
i
o
n
of
s
p
li
c
i
n
g
a
nd
c
op
y
-
mo
v
e
f
or
ge
r
i
e
s
i
n
i
m
a
g
e
s
,
”
2016
I
E
E
E
I
nt
e
r
nat
io
nal
W
or
k
s
hop on
I
nf
or
m
at
io
n F
or
e
ns
ic
s
and Se
c
u
r
it
y
(
W
I
F
S
)
,
2016
,
pp
.
1
-
6
,
doi
:
10.
1109/W
I
F
S
.
2016
.
7823911
.
[2
2
]
Y
.
Z
ha
n
g
,
J
.
G
oh
,
L
.
L
.
W
i
n,
a
nd
V
.
L
.
T
h
i
n
g
,
“
I
m
a
ge
r
e
g
i
on
f
or
ge
r
y
d
e
t
e
c
t
i
on:
A
de
e
p
l
e
a
r
n
i
n
g
a
ppr
o
a
c
h
,
”
I
n
P
r
oc
e
e
di
ngs
o
f
th
e
Si
ngapor
e
C
y
be
r
-
Se
c
ur
it
y
C
on
f
e
r
e
nc
e
(
S
G
-
C
R
C
)
,
2016
,
v
o
l
.
14
,
pp
.
1
-
11
,
doi
:
10.
3233/978
-
1
-
61499
-
617
-
0
-
1
.
[2
3
]
V
.
T
.
M
a
nu
a
nd
B
.
M
.
M
e
ht
r
e
,
“
V
i
s
u
a
l
a
r
t
i
f
a
c
t
s
b
a
s
e
d
im
a
ge
s
p
li
c
i
n
g
de
te
c
t
i
on
i
n
un
c
ompr
e
s
s
e
d
i
m
a
ge
s
,
”
2015
I
E
E
E
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on
C
om
put
e
r
G
r
aphi
c
s
,
V
is
io
n
and
I
nf
or
m
at
io
n
Se
c
ur
i
ty
(
C
G
V
I
S)
,
2015,
pp
.
145
-
150,
do
i
:
10.
1109/C
G
V
I
S
.
2015
.
7449911
.
[2
4
]
J
.
L
i
,
X
.
L
i
,
B
.
Y
a
n
g
,
a
nd
X
.
S
un
,
“
S
e
gme
nt
a
t
i
on
-
B
a
s
e
d
I
m
a
ge
C
opy
-
M
o
v
e
F
or
ge
r
y D
e
te
c
t
i
on
S
c
he
me
,
”
i
n
I
E
E
E
T
r
ans
ac
ti
ons
on
I
nf
or
m
at
io
n F
or
e
ns
ic
s
and Se
c
ur
it
y
,
v
o
l
.
10
,
no
.
3
,
pp
.
507
-
518
,
M
a
r
c
h
2015
,
do
i
:
10.
1109/
T
I
F
S
.
2014
.
2381872
.
[2
5
]
V
.
B
a
dr
i
n
a
r
a
ya
n
a
n,
A
.
K
e
nda
l
l
,
a
nd
R
.
C
i
po
ll
a
,
“
S
e
g
N
e
t:
A
D
e
e
p
C
onv
o
l
ut
i
on
a
l
E
n
c
od
e
r
-
D
e
c
od
e
r
A
r
c
h
i
t
e
c
tu
r
e
f
or
I
ma
ge
S
e
g
m
e
nt
a
t
i
on,
”
i
n
I
E
E
E
T
r
ans
ac
ti
ons
on
P
at
te
r
n
A
nal
y
s
is
and
M
ac
hi
ne
I
nt
e
ll
ig
e
nc
e
,
v
o
l
.
39,
no
.
12
,
pp
.
2481
-
2495,
D
e
c
.
20
17,
do
i
:
10.
1109/T
P
A
M
I
.
2016.
2644615
.
[2
6
]
S.
-
J
.
R
y
u
a
nd
H
.
-
K
.
L
e
e
,
“
E
s
t
i
m
a
t
i
on
of
li
n
e
a
r
tr
a
ns
f
or
m
a
t
i
on
by
a
n
a
l
yz
i
n
g
th
e
pe
r
i
od
i
c
i
t
y
of
i
n
te
r
po
l
a
t
i
on
,
”
P
at
te
r
n
R
e
c
ogni
ti
on
L
e
tt
e
r
s
,
v
o
l
.
36
,
pp
.
89
-
99,
2014
,
doi
:
10.
1016/j
.
pa
tr
e
c
.
2013.
09
.028
.
[2
7
]
B
.
M
a
hdi
a
n
a
nd
S
.
S
a
i
c
,
“
B
li
nd
A
ut
he
n
t
i
c
a
t
i
on
U
s
i
n
g
P
e
r
i
od
i
c
P
r
ope
r
t
i
e
s
of
I
nt
e
r
po
l
a
t
i
on,
”
i
n
I
E
E
E
T
r
ans
ac
ti
ons
on
I
nf
or
m
at
io
n
F
or
e
ns
ic
s
and Se
c
ur
it
y
,
v
o
l
.
3
,
no
.
3
,
pp
.
529
-
538
, S
e
pt
.
2008
, do
i
:
10.
1109/T
I
F
S
.
2004
.
92460
3
.
[2
8
]
D
.
V
oor
h
i
e
s
,
“
S
pa
c
e
-
f
il
li
n
g
c
ur
ve
s
a
nd
p
a
c
e
-
f
il
li
n
g
c
ur
ve
s
a
nd
a
m
e
a
s
u
r
e
of
c
oh
e
r
e
n
c
e
,
”
G
r
aphi
c
s
G
e
m
s
I
I
,
pp.
26
-
30
,
1991
,
do
i
:
10.
1016/B
978
-
0
-
08
-
050754
-
5.
50018
-
9
.
[2
9
]
B
.
M
oon
,
H
.
V
.
J
a
g
a
d
i
s
h,
C
.
F
a
l
out
s
o
s
,
a
nd J
.
H
.
S
a
l
t
z
,
“
A
na
l
ys
i
s
of
t
h
e
c
l
us
t
e
r
i
n
g
pr
op
e
r
t
i
e
s
of
t
h
e
H
il
b
e
r
t
s
pa
c
e
-
f
il
li
n
g
c
ur
ve
,
”
i
n
I
E
E
E
T
r
ans
ac
ti
ons
on K
now
le
dge
and Data E
ngi
ne
e
r
in
g
,
v
o
l
.
13,
no
.
1
,
pp
.
124
-
141
,
J
a
n.
-
F
e
b.
2001
,
doi
:
10.
1109/69
.
908985
.
[3
0
]
T
.
M
.
M
oha
mm
e
d
e
t
al
.
,
“
B
oos
t
i
n
g
i
m
a
g
e
f
or
ge
r
y
d
e
te
c
t
i
on
us
i
n
g
r
e
s
a
mp
li
n
g
d
e
t
e
c
t
i
on
a
nd
c
op
y
-
mo
ve
a
n
a
l
ys
i
s
,
”
Soc
ie
ty
f
or
I
m
agi
ng Sc
ie
nc
e
and T
e
c
hnol
ogy
,
v
o
l
.
7
,
pp
.
118
-
1
-
118
-
7
,
2018
,
d
oi
:
10.
2352/I
S
S
N
.
2470
-
1173
.
2018
.
07
.
M
W
S
F
-
118
.
[3
1
]
B
.
B
a
y
a
r
a
nd
M
.
C
.
S
ta
mm
,
“
O
n
th
e
r
obu
s
tn
e
s
s
of
c
on
s
tr
a
i
n
e
d
c
on
v
o
l
ut
i
on
a
l
n
e
ur
a
l
n
e
t
w
or
ks
to
J
P
E
G
pos
t
-
c
ompr
e
s
s
i
on
f
or
i
m
a
ge
r
e
s
a
mp
li
n
g
d
e
t
e
c
t
i
on,
”
2017
I
E
E
E
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on
A
c
ous
ti
c
s
,
Spe
e
c
h
and
Si
gnal
P
r
oc
e
s
s
in
g
(
I
C
A
SSP
)
,
2017,
pp.
2152
-
2156
,
doi
:
10.
1109/I
C
A
S
S
P
.
2017
.
7952537
.
[3
2
]
L
.
J
i
a
o
a
nd
J
.
Z
ha
o,
“
A
S
ur
ve
y
on
th
e
N
e
w
G
e
n
e
r
a
t
i
on
of
D
e
e
p
L
e
a
r
n
i
n
g
i
n
I
m
a
ge
P
r
oc
e
s
s
i
n
g
,
”
i
n
I
E
E
E
A
c
c
e
s
s
,
v
o
l
.
7,
p
p
.
172231
-
172263,
2019
,
doi
:
10.
1109/A
C
C
E
S
S
.
2019
.
2956508
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2502
-
4752
I
n
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
25
,
N
o
.
1
,
J
a
n
ua
r
y
20
22
:
183
-
190
190
[3
3
]
Z
.
J
.
B
a
r
a
d
a
nd
M
.
M
.
G
os
w
a
m
i
,
“
I
ma
ge
F
or
ge
r
y
D
e
t
e
c
t
i
on
us
i
n
g
D
e
e
p
L
e
a
r
n
i
n
g
:
A
S
ur
ve
y
,
”
2020
6t
h
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on A
dv
anc
e
d C
om
put
in
g and C
om
m
uni
c
at
io
n Sy
s
te
m
s
(
I
C
A
C
C
S)
,
2020
,
pp
.
571
-
576,
doi
:
10.
1109/I
C
A
C
C
S
48705
.
2020
.
9074
408
.
[3
4
]
A
.
K
uz
ne
ts
o
v
,
“
D
i
g
i
t
a
l
i
m
a
ge
f
o
r
g
e
r
y
d
e
t
e
c
t
i
on
us
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
ppr
o
a
c
h
,
”
J
our
nal
of
P
hy
s
ic
s
:
C
onf
e
r
e
nc
e
S
e
r
ie
s
,
v
o
l
.
13
68,
no.
3
,
2019
,
doi
:
10.
1088/1742
-
6596/1368/
3/
032028.
[3
5
]
R
.
H
ua
ng
,
F
.
F
a
ng
,
H
.
H
.
N
g
uy
e
n,
J
.
Y
a
m
a
gi
s
h
i
,
a
nd
I
.
E
c
h
iz
e
n,
“
A
M
e
th
od
f
or
I
d
e
nt
i
f
y
i
n
g
O
r
i
g
i
n
of
D
i
g
i
t
a
l
I
m
a
ge
s
U
s
in
g
a
C
onv
o
l
ut
i
on
a
l
N
e
ur
a
l
N
e
tw
o
r
k
,
”
2020
A
s
ia
-
P
ac
if
ic
Si
gnal
and
I
n
f
or
m
at
io
n
P
r
oc
e
s
s
in
g
A
s
s
o
c
ia
ti
on
A
nnual
Sum
m
it
and
C
onf
e
r
e
nc
e
(
A
P
SI
P
A
A
SC
)
,
2020
,
pp
.
1293
-
1299
.
[3
6
]
A
.
F
l
e
nn
e
r
,
L
.
P
e
te
r
s
on,
J
.
B
unk
,
T
.
M
.
M
oha
mm
e
d,
L
.
N
a
ta
r
a
j,
a
nd
B
.
S
.
M
a
nj
una
th
,
“
R
e
s
a
mp
li
n
g
F
or
g
e
r
y
D
e
te
c
t
i
on
U
s
i
n
g
D
e
e
p
L
e
a
r
n
i
n
g
a
nd
A
-
C
ont
r
a
r
i
o
A
n
a
l
ys
i
s
,
”
Soc
ie
ty
f
or
I
m
agi
ng
Sc
ie
nc
e
and
T
e
c
hnol
ogy
,
v
o
l
.
7,
pp.
212
-
1
-
212
-
7,
2018
,
do
i
:
10.
2352/I
S
S
N
.
2470
-
1173
.
2018
.
07
.
M
W
S
F
-
212
.
[3
7
]
F
.
M
a
r
r
a
,
D
.
G
r
a
g
n
a
n
i
e
ll
o,
L
.
V
e
r
do
li
va
,
a
nd
G
.
P
og
gi
,
“
A
F
u
l
l
-
I
m
a
g
e
F
u
ll
-
R
e
s
o
l
ut
i
on
E
nd
-
to
-
E
nd
-
T
r
a
i
n
a
b
l
e
C
N
N
F
r
a
me
w
or
k
f
or
I
m
a
ge
F
or
g
e
r
y
D
e
t
e
c
t
i
on,
”
i
n
I
E
E
E
A
c
c
e
s
s
,
v
o
l
.
8
,
pp
.
1334
88
-
133502,
2020
,
doi
:
10.
1109/A
C
C
E
S
S
.
2020
.
3009877
.
[3
8
]
C
.
P
un
,
X
.
Y
ua
n,
a
nd
X
.
B
i
,
“
I
m
a
ge
F
or
g
e
r
y
D
e
t
e
c
t
i
on
U
s
i
n
g
A
d
a
pt
i
ve
O
ve
r
s
e
g
m
e
nt
a
t
i
on
a
nd
F
e
a
tu
r
e
P
o
i
nt
M
a
tc
h
i
n
g
,
”
i
n
I
E
E
E
T
r
ans
ac
ti
ons
on I
n
f
or
m
at
io
n F
or
e
ns
ic
s
and Se
c
ur
it
y
,
v
o
l
.
10
,
n
o.
8
,
pp
.
1705
-
1716
,
A
ug
.
2015
,
doi
:
10.
1109/T
I
F
S
.
2015
.
24232
61
.
[3
9
]
A
.
K
uz
n
e
ts
o
v
a
nd
V
.
M
ya
s
n
i
ko
v
,
“
A
n
e
w
c
op
y
-
mo
v
e
f
or
ge
r
y
de
t
e
c
t
i
on
a
l
gor
i
th
m
us
i
n
g
i
m
a
ge
pr
e
pr
o
c
e
s
s
i
n
g
pr
o
c
e
du
r
e
,
”
P
r
oc
e
di
a E
ngi
ne
e
r
in
g
,
v
o
l
.
201
,
pp
.
436
-
444,
2017
,
doi
:
10.
101
6/
j.
pr
oe
ng
.
2017.
09
.
671
.
[4
0
]
P
.
M
.
R
a
ju
a
nd
M
.
S
.
N
a
i
r
,
“
C
op
y
-
mo
ve
f
or
ge
r
y
d
e
t
e
c
t
i
on
u
s
i
n
g
b
i
n
a
r
y
d
i
s
c
r
i
m
i
na
n
t
f
e
a
tu
r
e
s
,
”
J
our
nal
of
K
in
g
Saud
U
ni
v
e
r
s
i
ty
-
C
om
put
e
r
and I
n
f
or
m
at
io
n Sc
ie
n
c
e
s
,
2018
,
doi
:
10.
1016/j
.
jk
s
u
c
i
.
2018.
11.
004
.
[4
1
]
H
-
Y
.
H
ua
n
g a
nd
A
-
J
.
C
i
ou,
“
C
op
y
-
mo
ve
f
o
r
g
e
r
y
d
e
t
e
c
t
i
o
n
f
or
im
a
ge
f
o
r
e
n
s
i
c
s
us
i
n
g
t
h
e
s
upe
r
p
i
xe
l
s
e
g
me
n
ta
t
i
on
a
nd t
h
e
H
e
lm
e
r
t
tr
a
n
s
f
or
ma
t
i
on,
”
E
U
R
A
S
I
P
J
our
nal
on I
m
age
and V
id
e
o
P
r
oc
e
s
s
in
g
,
v
o
l
.
68
,
2019
,
doi
:
10.
1186/s
13640
-
019
-
0469
-
9.
[4
2
]
M
.
B
i
l
a
l
,
H
.
A
.
H
a
b
i
b,
Z
.
M
e
hmood
,
T
.
S
a
ba
,
a
nd
M
.
R
a
s
h
i
d
“
S
i
n
gl
e
a
nd
M
u
l
t
i
p
l
e
C
op
y
–
M
o
v
e
F
or
g
e
r
y
D
e
t
e
c
t
i
on
a
nd
L
oc
a
li
z
a
t
i
on
i
n
D
i
g
i
t
a
l
I
m
a
ge
s
B
a
s
e
d
on
th
e
S
pa
r
s
e
l
y
E
n
c
od
e
d
D
i
s
t
i
n
c
t
i
ve
F
e
a
tu
r
e
s
a
nd
D
B
S
C
A
N
C
l
us
t
e
r
i
n
g
,
”
A
r
abi
an
jo
ur
nal
f
or
s
c
ie
nc
e
and e
ngi
ne
e
r
in
g,
v
o
l
.
45,
pp
.
2975
-
2992
,
2019
,
doi
:
10.
1007/s
13369
-
019
-
04238
-
2.
BI
OG
RA
P
HI
E
S
OF
AU
T
HO
RS
M
a
n
j
un
a
t
h
a
S
i
s
a
Re
s
e
ar
c
h
S
ch
o
l
ar
at
J
SSA
T
E
Res
e
arch
C
e
n
t
r
e
,
D
e
p
t
.
o
f
CSE
,
J
SSA
T
E
,
affi
l
i
at
e
d
t
o
V
T
U
Bel
ag
a
v
i
.
H
e
co
m
p
l
e
t
ed
M.
T
e
ch
Co
m
p
u
t
e
r
S
c
i
en
ce
an
d
E
n
g
i
n
ee
r
i
n
g
fro
m
N
M
A
M
I
T
N
i
t
t
e
Man
g
al
o
r
e
,
affi
l
i
at
e
d
t
o
V
T
U
Be
l
a
g
av
i
,
K
ar
n
at
ak
a.
N
o
w
h
e
i
s
w
o
rk
i
n
g
as
an
A
s
s
o
c
i
at
e
Pro
f
e
s
s
o
r
D
e
p
t
.
O
f
I
SE
,
G
l
o
b
al
A
c
ad
emy
o
f
T
ec
h
n
o
l
o
g
y
,
Be
n
g
al
u
ru
.
H
e
c
an
b
e
co
n
t
a
c
t
e
d
at
em
a
i
l
:
m
a
n
j
u
n
at
h
s
@
g
at
.
ac
.
i
n
.
D
r.
M
a
l
i
ni
M
.
P
a
ti
l
i
s
p
re
s
e
n
t
l
y
w
o
r
k
i
n
g
as
an
A
s
s
o
ci
at
e
Pro
f
e
s
s
o
r
i
n
t
h
e
D
e
p
art
men
t
o
f
In
fo
r
m
at
i
o
n
S
ci
e
n
ce
an
d
E
n
g
i
n
ee
r
i
n
g
at
J
.
S.
S.
A
c
a
d
emy
o
f
T
ec
h
n
i
c
al
E
d
u
c
at
i
o
n
,
Ban
g
al
o
r
e
,
K
arn
at
ak
a,
In
d
i
a.
S
h
e
r
ece
i
v
ed
h
e
r
P
h
.
D
.
D
eg
r
ee
fr
o
m
Bh
arat
h
i
ar
U
n
i
v
e
rs
i
t
y
i
n
t
h
e
y
e
ar
2
0
1
5
.
H
e
r
r
e
s
e
arch
i
n
t
e
r
e
s
t
s
are
b
i
g
d
at
a
an
a
l
y
t
i
c
s
,
b
i
o
i
n
fo
r
m
at
i
c
s
,
cl
o
u
d
c
o
m
p
u
t
i
n
g
,
i
m
a
g
e
p
ro
ce
s
s
i
n
g
.
Sh
e
h
as
p
u
b
l
i
s
h
e
d
m
o
r
e
t
h
an
5
0
re
s
e
ar
ch
p
ap
e
rs
i
n
re
p
u
t
e
d
p
ee
r
-
re
v
i
ew
ed
i
n
t
e
r
n
at
i
o
n
al
j
o
u
rn
a
l
s
.
Sh
e
i
s
a
mem
b
e
r
o
f
IE
E
E
,
IE
I
,
I
ST
E
,
CS
I
.
S
h
e
h
as
at
t
e
n
d
ed
an
d
p
re
s
e
n
t
e
d
p
ap
e
rs
i
n
m
a
n
y
i
n
t
e
r
n
at
i
o
n
al
co
n
f
e
r
e
n
ce
s
i
n
In
d
i
a
an
d
A
b
ro
a
d
.
Pr
e
s
en
t
l
y
s
h
e
i
s
g
u
i
d
i
n
g
3
re
s
e
ar
c
h
s
ch
o
l
ars
an
d
o
n
e
re
s
e
ar
c
h
s
ch
o
l
ar
h
as
c
o
m
p
l
e
t
ed
t
h
e
P
h
.
D
.
H
e
c
a
n
b
e
c
o
n
t
ac
t
e
d
at
em
a
i
l
:
d
r
m
al
i
n
i
m
p
at
i
l
@
g
m
a
i
l
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.