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
.
3
8
,
N
o
.
2
,
M
a
y
20
2
5
,
pp.
1
331
~
1
338
I
S
S
N:
2
502
-
4
7
52
,
DO
I
:
10
.
11591/i
j
e
e
cs
.v
3
8
.
i
2
.
pp
1
331
-
1
338
1331
Jou
r
n
al
h
o
m
e
page
:
ht
tp:
//
ij
e
e
cs
.
iaes
c
or
e
.
c
om
A
c
c
u
r
a
t
e
se
gm
e
n
t
at
io
n
of
f
r
u
it b
ase
d
on
d
e
e
p
l
e
ar
n
in
g
E
s
r
aa
Abu
E
l
s
ou
d
1
,
Om
ar
A
l
id
m
at
1
,
S
u
h
ail
a
Abu
owaid
a
2
,
E
s
r
aa
Al
h
e
n
awi
1
,
Naw
af
Al
s
h
d
aif
at
3
,
Ahm
ad
Abu
r
om
m
an
1
,
Hu
ah
Yong
C
h
an
4
1
D
e
pa
r
tm
e
nt
of
C
o
mput
e
r
S
c
i
e
n
c
e
,
F
a
c
ul
t
y
of
I
n
f
or
ma
ti
o
n
T
e
c
hno
l
o
g
y
,
Z
a
r
qa
U
ni
ve
r
s
it
y
,
Z
a
r
qa
, J
o
r
da
n
2
D
e
pa
r
tm
e
nt
of
C
o
mput
e
r
S
c
i
e
nc
e
,
F
a
c
u
lt
y
of
I
n
f
o
r
ma
ti
o
n
T
e
c
h
no
l
o
g
y
,
Al
-
A
lb
a
y
t
U
ni
ve
r
s
it
y
,
M
a
f
r
a
q,
J
or
da
n
3
F
a
c
ul
t
y
of
I
n
f
or
ma
ti
o
n
T
e
c
hn
o
l
o
g
y
,
A
ppl
i
e
d
S
c
i
e
n
c
e
P
r
i
v
a
t
e
U
ni
ve
r
s
it
y
,
A
mm
a
n,
J
o
r
da
n
4
S
c
h
oo
l
of
C
o
mpu
te
r
S
c
i
e
nc
e
s
,
U
ni
ve
r
s
it
i
S
a
in
s
M
a
la
y
s
ia
,
G
e
lu
go
r
,
M
a
la
y
s
ia
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
A
ug
10
,
202
4
R
e
vi
s
e
d
No
v
10
,
202
4
A
c
c
e
pt
e
d
N
o
v
24
,
202
4
I
n
t
h
e
l
as
t
few
y
ears
,
d
ee
p
l
e
arn
i
n
g
h
as
e
x
h
i
b
i
t
ed
i
t
s
e
ffi
c
a
cy
a
n
d
c
ap
aci
t
y
i
n
t
h
e
fi
el
d
o
f
co
m
p
u
t
e
r
v
i
s
i
o
n
o
w
i
n
g
t
o
i
t
s
e
x
ce
p
t
i
o
n
al
p
r
ec
i
s
i
o
n
an
d
w
i
d
e
s
p
re
a
d
a
cce
p
t
an
ce.
T
h
e
p
ri
m
ar
y
o
b
j
ec
t
i
v
e
o
f
t
h
i
s
s
t
u
d
y
i
s
t
o
i
n
v
e
s
t
i
g
at
e
an
i
m
p
ro
v
ed
ap
p
ro
ach
f
o
r
s
eg
m
en
t
at
i
o
n
i
n
t
h
e
c
o
n
t
e
x
t
o
f
v
ar
i
o
u
s
fr
u
i
t
c
at
eg
o
r
i
e
s
.
D
e
s
p
i
t
e
the
u
t
i
l
i
zat
i
o
n
of
d
ee
p
l
e
arn
i
n
g
,
the
c
u
rr
en
t
s
e
g
men
t
at
i
o
n
t
e
ch
n
i
q
u
e
s
f
o
r
v
ar
i
o
u
s
fru
i
t
i
t
em
s
e
x
h
i
b
i
t
s
u
b
p
ar
p
e
rf
o
r
m
an
ce.
The
p
ro
p
o
s
e
d
e
n
h
an
ced
m
u
l
t
i
p
l
e
fru
i
t
s
e
g
men
t
at
i
o
n
a
l
g
o
ri
t
h
m
h
as
t
h
e
fo
l
l
o
w
i
n
g
m
ai
n
s
t
e
p
s
:
1
)
m
o
d
i
f
y
i
n
g
t
h
e
s
i
z
e
of
the
fi
l
t
e
r
,
2)
t
he
p
ro
ces
s
of
o
p
t
i
m
i
z
i
n
g
the
Re
s
N
e
t
-
101
b
l
o
ck
i
n
v
o
l
v
e
s
s
el
ec
t
i
n
g
t
h
e
mo
s
t
s
u
i
t
ab
l
e
c
o
u
n
t
o
f
r
e
p
e
t
i
t
i
o
n
s
.
T
h
e
mu
l
t
i
p
l
e
fru
i
t
d
at
as
e
t
i
s
s
p
l
i
t
8
0
%
i
n
t
h
e
t
rai
n
i
n
g
s
t
ag
e
a
n
d
2
0
%
i
n
t
h
e
t
e
s
t
i
n
g
s
t
ag
e
.
T
h
e
s
e
i
m
ag
e
s
w
e
r
e
u
t
i
l
i
z
ed
t
o
t
rai
n
a
d
ee
p
l
e
arn
i
n
g
(D
L
)
b
as
e
d
al
g
o
r
i
t
h
m
,
w
h
i
c
h
ai
m
s
to
i
d
e
n
t
i
f
y
m
u
l
t
i
p
l
e
fru
i
t
i
t
em
s
w
i
t
h
i
n
i
m
ag
e
s
acc
u
rat
e
l
y
.
T
h
e
p
ro
p
o
s
ed
al
g
o
r
i
t
h
m
h
as
a
l
o
w
er
t
rai
n
i
n
g
t
i
me
c
o
m
p
ar
e
d
t
o
t
h
e
o
t
h
e
r
al
g
o
r
i
t
h
m
s
.
T
h
e
t
h
r
e
s
h
o
l
d
s
e
x
h
i
b
i
t
g
r
e
at
e
r
v
al
u
e
s
co
m
p
ared
to
t
h
e
t
h
re
s
h
o
l
d
s
of
s
t
at
e
-
of
-
the
-
art
al
g
o
r
i
t
h
m
s
.
K
e
y
w
o
r
d
s
:
De
e
p
l
e
a
r
ni
n
g
M
a
s
k
R
-
C
NN
M
u
l
t
i
p
l
e
f
r
u
i
t
R
e
g
i
o
n
pr
o
po
s
a
l
n
e
t
wor
k
S
e
g
m
e
n
t
a
t
i
o
n
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
:
S
uh
a
il
a
Ab
uo
wa
i
da
De
pa
r
t
m
e
n
t
o
f
C
o
m
put
e
r
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
I
nf
o
r
m
a
t
i
o
n
T
e
c
hn
o
l
o
g
y
,
Al
-
Al
ba
y
t
U
ni
ve
r
s
i
t
y
M
a
f
r
a
q,
J
o
r
da
n
E
m
a
i
l
:
s
u
hil
a
@
a
a
b
u.
e
du.
m
y
1.
I
NT
RODU
C
T
I
ON
F
r
ui
t
i
m
a
g
e
s
e
gm
e
n
ta
ti
on
i
s
a
s
i
gn
i
f
i
c
a
n
t
a
n
d
c
om
pl
e
x
p
r
ob
l
e
m
wi
t
h
i
n
th
e
f
i
e
l
d
o
f
c
o
m
p
u
te
r
v
i
s
i
o
n
[
1
]
.
T
hi
s
t
a
s
k
i
s
o
f
i
m
po
r
t
a
n
c
e
a
s
i
t
di
r
e
c
t
l
y
im
pa
c
t
s
a
n
d
f
a
c
il
i
t
a
t
e
s
d
i
e
t
a
r
y
a
s
s
e
s
s
m
e
n
t
,
a
s
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
s
e
r
v
e
s
a
s
a
f
u
n
da
m
e
n
t
a
l
s
t
e
p
i
n
t
h
e
s
e
pr
o
c
e
s
s
e
s
.
M
o
r
e
o
v
e
r
,
t
h
e
pr
o
c
e
s
s
o
f
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
pr
e
s
e
n
t
s
s
e
v
e
r
a
l
c
h
a
ll
e
n
g
e
s
,
m
o
s
t
l
y
due
to
t
h
e
pr
e
s
e
n
c
e
o
f
m
u
l
t
ip
l
e
f
r
u
i
t
i
t
e
m
s
t
h
a
t
h
a
v
e
s
im
il
a
r
s
h
a
pe
s
a
n
d
a
pp
e
a
r
a
n
c
e
s
.
C
o
n
s
e
que
n
t
l
y
,
t
hi
s
s
im
il
a
r
i
t
y
hi
nde
r
s
t
h
e
a
c
c
ur
a
t
e
di
f
f
e
r
e
n
t
i
a
t
i
o
n
o
f
f
r
u
i
t
s
o
l
e
ly
b
a
s
e
d
o
n
t
h
e
i
r
vi
s
ua
l
r
e
pr
e
s
e
n
t
a
t
i
o
ns
.
F
ur
t
h
e
r
m
o
r
e
,
s
e
v
e
r
a
l
f
r
u
i
t
i
t
e
m
s
e
xhi
b
i
t
a
d
i
v
e
r
s
e
r
a
n
ge
o
f
s
ha
pe
s
,
c
o
l
o
r
s
,
a
n
d
s
i
z
e
s
,
h
e
n
c
e
po
s
i
n
g
a
gr
e
a
t
e
r
di
f
f
i
c
u
l
t
y
in
t
h
e
c
l
a
s
s
if
i
c
a
t
i
o
n
of
f
r
u
i
t
t
y
pe
s
.
T
h
e
pr
o
c
e
s
s
of
s
e
pa
r
a
t
i
n
g
i
r
r
e
gu
l
a
r
ly
s
h
a
pe
d
f
r
u
i
t
i
t
e
m
s
in
a
m
u
l
t
i
p
l
e
-
f
r
u
i
t
i
m
a
ge
b
e
c
o
m
e
s
c
ha
l
l
e
n
g
i
ng,
pa
r
t
i
c
u
l
a
r
l
y
w
h
e
n
t
h
e
r
e
is
an
o
c
c
l
us
i
o
n
pr
e
s
e
n
t
.
M
o
r
e
o
v
e
r
,
t
h
e
pr
e
s
e
n
c
e
of
d
i
v
e
r
s
e
qua
n
t
i
t
i
e
s
of
f
r
u
i
t
de
p
i
c
t
e
d
in
t
h
e
i
m
a
ge
is
a
s
i
g
nif
i
c
a
n
t
de
t
e
r
m
in
a
n
t
t
h
a
t
i
n
f
l
u
e
n
c
e
s
t
h
e
a
ppr
o
a
c
h
e
m
p
l
o
y
e
d
f
o
r
f
r
u
i
t
r
e
c
o
gni
t
i
o
n
.
T
hi
s
f
a
c
t
or
i
s
vi
t
a
l
i
n
t
h
e
a
d
v
a
n
c
e
m
e
n
t
o
f
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
He
n
c
e
,
t
h
e
de
ve
l
o
p
m
e
n
t
o
f
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
po
s
e
s
a
s
i
g
nif
i
c
a
n
t
c
h
a
ll
e
n
ge
.
Nu
m
e
r
o
us
a
l
go
r
i
t
hm
s
h
a
v
e
pr
o
-
p
o
s
e
d
di
s
t
i
n
c
t
m
e
t
h
o
do
l
o
g
i
e
s
f
o
r
a
ddr
e
s
s
i
ng
t
h
e
t
a
s
k
of
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
[
2]
.
Ho
we
v
e
r
,
t
h
e
m
a
j
o
r
i
t
y
of
s
t
ud
i
e
s
h
a
ve
f
o
c
us
e
d
on
t
h
e
i
de
n
t
i
f
i
c
a
t
i
o
n
of
m
u
l
t
i
p
l
e
f
r
u
i
t
i
t
e
m
s
b
a
s
e
d
on
a
s
i
n
g
le
im
a
ge
[
3]
.
T
h
e
s
e
g
m
e
n
t
a
t
i
o
n
of
f
r
u
i
t
i
t
e
m
s
wa
s
n
ot
po
s
s
i
bl
e
due
to
t
h
e
c
h
a
ll
e
n
g
e
s
t
h
a
t
a
r
e
a
s
s
o
c
i
a
t
e
d
w
i
t
h
t
hi
s
s
pe
c
i
f
i
c
t
a
s
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
8
,
N
o
.
2
,
M
a
y
20
2
5
:
1
331
-
1
338
1332
T
h
e
o
bj
e
c
t
i
ve
o
f
t
h
e
pa
pe
r
i
s
to
e
s
t
a
bl
i
s
h
a
t
h
e
o
r
e
t
i
c
a
l
b
u
il
d
f
o
r
t
h
e
a
i
m
o
f
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
to
s
o
l
v
e
t
h
e
i
s
s
u
e
s
m
e
n
t
i
o
n
e
d
a
b
o
v
e
.
He
n
c
e
,
a
s
pe
c
i
a
l
a
l
go
r
i
t
hm
r
e
f
e
r
r
e
d
to
a
s
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
ti
o
n
wa
s
i
n
t
r
o
duc
e
d,
c
o
m
pr
i
s
i
ng
t
w
o
s
t
a
ge
s
.
T
h
e
f
i
r
s
t
s
t
a
g
e
i
m
pr
o
v
e
s
f
r
u
i
t
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
by
im
pr
o
vi
ng
R
e
s
t
Ne
t
-
101,
r
e
s
u
l
t
i
n
g
i
n
a
n
i
m
pr
o
v
e
d
R
e
s
Ne
t
-
101
b
a
c
kb
o
n
e
to
s
o
l
v
e
l
o
w
de
t
e
c
t
i
o
n
r
a
t
e
s
.
T
h
e
n
e
x
t
s
t
e
p
i
nvo
l
v
e
s
t
h
e
im
p
l
e
m
e
n
t
a
t
i
o
n
o
f
t
h
e
r
e
g
i
o
n
pr
o
p
o
s
a
l
n
e
t
wo
r
k
(
R
P
N)
to
s
i
m
p
li
f
y
t
h
e
l
o
c
a
l
i
z
a
t
i
o
n
o
f
d
i
f
f
e
r
e
n
t
f
r
u
i
t
i
t
e
m
s
.
T
h
e
s
ugge
s
t
e
d
a
l
go
r
i
t
hm
f
o
c
us
e
s
o
n
t
h
e
t
a
s
k
o
f
mu
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
T
h
e
e
nh
a
n
c
e
d
m
u
l
t
ip
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
t
e
c
h
ni
que
e
m
p
l
o
y
s
t
h
e
r
e
g
i
o
n
o
f
i
n
t
e
r
e
s
t
(
R
oI
)
,
f
o
r
s
e
gm
e
n
t
a
t
i
o
n
,
to
e
x
t
r
a
c
t
f
e
a
t
ur
e
s
s
p
e
c
i
f
i
c
to
e
a
c
h
i
ns
t
a
n
c
e
.
T
h
e
e
v
a
l
ua
t
i
o
n
o
f
t
h
e
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
c
o
ns
i
s
t
s
o
f
t
w
o
m
e
a
s
ur
e
m
e
n
t
s
:
t
h
e
a
v
e
r
a
ge
o
v
e
r
uni
o
n
s
e
c
t
i
o
n
o
v
e
r
I
o
U
(
A
P
)
wi
t
h
va
r
i
o
us
t
h
r
e
s
h
o
l
ds
a
n
d
t
h
e
dur
a
t
i
o
n
o
f
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g.
T
h
e
m
e
a
s
ur
e
m
e
n
t
s
we
r
e
f
ur
t
h
e
r
c
o
m
pa
r
e
d
w
i
t
h
t
h
e
a
l
go
r
i
t
hm
s
M
a
s
k
R
-
C
NN
[
4]
,
YO
L
A
C
T
[
5]
,
a
n
d
C
A
S
C
A
DE
R
-
C
NN
[
6
]
.
T
h
e
f
a
c
to
r
s
c
o
n
tr
i
b
ut
i
n
g
to
t
h
e
v
a
r
i
a
t
i
o
ns
o
b
s
e
r
v
e
d
in
t
h
e
o
u
t
c
o
m
e
s
pr
o
duc
e
d
by
t
h
e
s
e
f
o
ur
a
l
go
r
i
t
hm
s
h
a
v
e
b
e
e
n
i
d
e
n
t
i
f
i
e
d.
T
h
e
pa
pe
r
c
on
c
l
ude
s
w
i
t
h
a
s
u
m
m
a
r
y
of
t
h
e
r
e
s
u
l
t
s
.
2.
RE
L
AT
E
D
WORK
T
hi
s
s
e
c
t
i
o
n
d
i
s
c
u
s
s
e
s
a
c
o
m
pr
e
h
e
n
s
i
ve
e
x
a
mi
na
t
i
o
n
o
f
t
h
e
a
l
go
r
i
t
hm
s
de
s
c
r
i
be
d
i
n
pr
i
o
r
w
or
k,
f
o
-
c
us
i
ng
o
n
t
h
e
i
r
us
e
i
n
i
de
n
t
i
f
yi
ng,
de
t
e
c
t
i
n
g,
a
n
d
s
e
g
m
e
n
t
i
n
g
f
r
u
i
t
s
.
T
h
e
s
e
a
l
go
r
i
t
hm
s
e
n
c
o
m
pa
s
s
a
dv
a
n
c
e
d
de
e
p
-
l
e
a
r
ni
ng
a
ppr
o
a
c
h
e
s
.
De
e
p
l
e
a
r
ni
ng
h
a
s
de
m
o
n
s
t
r
a
t
e
d
i
t
s
e
f
f
e
c
t
i
v
e
n
e
s
s
i
n
c
o
m
put
e
r
vi
s
i
o
n
[
7
]
-
[
11]
.
I
t
h
a
s
b
e
e
n
us
e
d
t
o
de
t
e
c
t
f
r
u
i
t
b
a
s
e
d
o
n
i
m
a
ge
s
,
d
i
s
t
i
n
gu
i
s
he
d
by
i
t
s
hi
g
h
a
c
c
ur
a
c
y
,
a
s
pr
e
s
e
n
t
e
d
i
n
t
hi
s
s
e
c
t
i
o
n
.
I
n
t
hi
s
r
e
s
pe
c
t
,
c
o
n
v
o
l
ut
i
o
n
a
l
ne
ur
a
l
n
e
t
wor
k
(
C
NN
)
i
s
c
o
n
s
i
de
r
e
d
a
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
a
l
go
r
i
t
hm
i
n
de
e
p
l
e
a
r
ni
ng
i
m
a
ge
s
e
g
m
e
n
t
a
t
i
o
n
.
F
ur
t
h
e
r
m
o
r
e
,
[
12]
u
s
e
d
C
NN
to
s
e
g
m
e
n
t
f
r
u
i
t
i
m
a
ge
t
a
s
ks
.
C
NN
de
m
o
n
s
t
r
a
t
e
d
t
h
a
t
i
t
wa
s
m
o
r
e
a
c
c
ur
a
t
e
a
n
d
c
o
ul
d
m
a
na
ge
l
a
r
ge
da
t
a
s
e
t
s
t
h
a
n
t
r
a
d
i
t
i
o
n
a
l
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
go
r
i
t
hm
s
.
T
he
a
l
go
r
i
t
hm
b
a
s
e
d
o
n
de
e
p
l
e
a
r
ni
ng
wa
s
a
pp
l
i
e
d
to
s
e
g
m
e
n
t
f
r
u
i
t
us
i
ng
c
o
m
bi
ne
d
s
e
l
e
c
t
i
ve
s
e
a
r
c
h
,
C
NN
,
a
n
d
Gr
a
b
C
ut
a
l
go
r
i
t
hm
s
[
13]
.
T
h
e
m
a
i
n
i
s
s
u
e
i
n
t
h
e
a
l
go
r
i
t
hm
wa
s
a
n
o
v
e
r
l
a
pp
i
ng
e
r
r
o
r
,
e
s
pe
c
i
a
ll
y
i
n
t
he
e
dge
s
.
A
n
e
w
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
wa
s
pr
e
s
e
n
t
e
d:
a
f
u
l
ly
a
ut
o
m
a
t
e
d
n
o
v
e
l
f
r
u
i
t
i
n
t
a
ke
s
e
g
m
e
n
t
a
t
i
o
n
s
y
s
t
e
m
f
o
r
s
e
m
a
n
t
i
c
s
e
g
m
e
n
t
a
t
i
o
n
us
i
n
g
F
C
N
[
14]
.
T
h
e
b
e
ne
f
i
t
o
f
t
h
i
s
a
l
go
r
i
t
hm
i
s
e
nh
a
n
c
e
d
r
e
l
i
a
bil
i
t
y
i
n
t
e
r
m
s
o
f
s
e
g
m
e
n
t
a
t
i
o
n
t
y
pe
s
.
On
t
h
e
ot
h
e
r
h
a
n
d,
t
hi
s
a
l
g
o
r
i
t
hm
c
o
u
l
d
n
o
t
h
a
n
d
l
e
t
h
e
s
a
m
e
i
ns
t
a
n
c
e
o
f
th
e
f
r
u
i
t
s
e
g
m
e
n
t
.
A
s
e
m
a
n
t
i
c
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
wa
s
a
pp
l
i
e
d
to
f
i
nd
t
h
e
por
t
i
o
n
o
f
t
h
e
f
r
u
i
t
[
15]
-
[
17]
.
T
h
e
a
l
go
r
i
t
hm
c
o
m
bi
ne
d
F
C
N
a
n
d
f
a
s
t
R
-
C
NN
,
whi
c
h
de
c
r
e
a
s
e
d
t
h
e
t
i
m
e
s
p
e
e
d
i
n
t
h
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
T
h
e
M
a
s
k
R
-
C
NN
wa
s
us
e
d
i
n
f
r
u
i
t
i
m
a
ge
s
e
g
m
e
n
t
a
t
i
o
n
[
18]
-
[
24]
.
T
h
e
pr
o
p
o
s
e
d
a
l
go
r
i
t
hm
s
s
h
o
we
d
e
n
c
o
ur
a
gi
n
g
im
pr
o
v
e
m
e
n
t
s
i
n
t
h
e
qua
l
i
t
y
o
f
f
r
u
i
t
i
m
a
ge
s
e
g
m
e
n
t
a
t
i
o
n
t
h
a
n
t
r
a
di
t
i
o
n
a
l
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
g
o
r
i
t
hm
s
.
Ho
we
v
e
r
,
t
h
e
a
l
go
r
i
t
hm
s
r
e
l
i
e
d
o
n
M
a
s
k
R
-
C
NN
.
T
hi
s
l
o
s
t
c
e
r
t
a
i
n
f
e
a
t
ur
e
s
a
t
t
h
e
i
n
s
t
a
n
c
e
s
e
g
m
e
n
t
a
ti
o
n
l
e
v
e
l
a
n
d
r
e
qu
i
r
e
d
a
l
e
n
gt
hy
t
r
a
i
ni
ng
pe
r
i
o
d.
T
hi
s
s
t
ud
y
i
s
c
o
n
c
e
r
n
e
d
w
i
t
h
t
h
e
s
e
g
m
e
n
t
a
t
i
o
n
o
f
f
r
u
i
t
im
a
ge
s
.
Va
r
i
o
us
a
ppr
o
xi
m
a
t
i
o
n
a
l
go
r
i
t
hm
s
h
a
v
e
b
e
e
n
inv
e
s
t
i
g
a
t
e
d,
a
n
d
r
e
s
e
a
r
c
h
e
r
s
us
u
a
l
ly
wo
r
k
o
n
th
e
m
a
i
n
pr
o
bl
e
m
:
f
r
u
i
t
i
m
a
ge
s
e
g
m
e
n
t
a
t
i
o
n
.
I
n
t
h
e
l
i
t
e
r
a
t
ur
e
,
f
r
u
i
t
i
m
a
g
e
s
e
g
m
e
n
t
a
t
i
o
n
s
t
i
ll
i
nc
l
ude
s
s
e
ve
r
a
l
we
a
k
-
n
e
s
s
e
s
,
s
uc
h
a
s
n
o
t
b
e
i
ng
f
l
e
xi
b
l
e
whe
n
de
a
l
i
ng
wi
t
h
d
i
ve
r
s
e
s
h
a
p
e
s
,
c
o
l
o
r
s
,
a
n
d
s
i
z
e
s
a
n
d
d
i
f
f
icu
l
t
y
in
s
e
pa
r
a
t
i
n
g
t
h
o
s
e
t
y
pe
s
h
a
vi
ng
i
r
r
e
gu
l
a
r
s
h
a
pe
s
,
e
s
pe
c
i
a
ll
y
w
h
e
n
t
h
e
r
e
i
s
a
n
o
c
c
l
us
i
o
n
i
n
t
h
e
f
r
u
i
t
i
m
a
g
e
.
T
h
e
i
ns
t
a
n
c
e
s
e
g
m
e
n
t
a
t
i
o
n
f
o
r
m
u
l
t
i
p
l
e
f
o
o
ds
s
u
f
f
e
r
s
f
r
o
m
po
or
pe
r
f
o
r
m
a
n
c
e
de
s
p
i
t
e
t
h
e
us
e
o
f
de
e
p
l
e
a
r
ni
ng
due
t
o
t
h
e
a
do
p
t
i
o
n
o
f
R
e
s
Ne
t
-
101
a
s
a
ba
c
kb
o
ne
f
o
r
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
.
T
h
e
R
e
s
Ne
t
-
101
pr
o
bl
e
m
s
s
t
i
ll
e
xi
s
t
,
s
uc
h
a
s
de
t
e
r
m
i
n
i
ng
a
s
u
i
t
a
bl
e
n
u
m
be
r
o
f
R
e
s
N
e
t
-
101
bl
o
c
ks
,
a
n
d
s
m
a
ll
f
e
a
t
ur
e
s
b
e
c
o
m
e
s
m
a
ll
e
r
o
r
v
a
ni
s
h
dur
i
ng
do
wn
s
a
m
p
li
ng.
An
e
nh
a
n
c
e
d
i
ns
t
a
nc
e
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
f
o
r
m
u
l
t
i
p
l
e
t
y
p
e
s
o
f
f
r
u
i
t
c
a
n
b
e
de
s
i
g
n
e
d
a
n
d
e
nh
a
nc
e
d.
T
hi
s
pr
o
p
o
s
e
d
a
l
go
r
i
t
hm
e
nh
a
n
c
e
d
t
h
e
R
e
s
Ne
t
-
101
b
a
c
kb
o
n
e
f
o
r
b
e
tt
e
r
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
by
f
i
nd
i
n
g
t
h
e
s
u
i
t
a
bl
e
n
u
m
be
r
o
f
R
e
s
Ne
t
-
101
bl
o
c
ks
a
n
d
c
o
n
ne
c
t
i
n
g
t
h
e
m
w
i
t
h
a
n
a
dd
i
t
i
o
n
a
l
c
o
n
v
o
l
ut
i
o
n
a
l
l
a
y
e
r
3.
T
HE
DA
T
ASE
T
T
h
e
s
uc
c
e
s
s
f
u
l
e
x
e
c
ut
i
o
n
o
f
s
e
g
m
e
n
t
a
t
i
o
n
t
a
s
ks
u
t
i
li
z
i
ng
de
e
p
l
e
a
r
ni
ng
a
l
go
r
i
t
hm
s
r
e
qu
i
r
e
s
a
c
c
e
s
s
to
l
a
r
ge
da
t
a
s
e
t
s
a
n
d
s
i
g
nif
i
c
a
n
t
c
o
m
put
a
t
i
o
n
a
l
r
e
s
o
ur
c
e
s
.
T
h
e
pr
o
bl
e
m
of
c
o
m
put
a
t
i
o
n
a
l
po
we
r
h
a
s
b
e
e
n
s
uc
c
e
s
s
-
f
u
l
ly
s
o
l
v
e
d
by
t
h
e
ut
i
li
z
a
t
i
o
n
of
gr
a
phi
c
s
pr
o
c
e
s
s
i
n
g
uni
t
s
(
GPUs
)
.
H
o
we
v
e
r
,
t
h
e
r
e
e
xi
s
t
s
a
l
i
mi
t
e
d
nu
m
b
e
r
o
f
da
t
a
b
a
s
e
s
t
h
a
t
c
o
n
t
a
i
n
m
e
t
i
c
u
l
o
us
ly
a
nn
o
t
a
t
e
d
o
pe
n
-
s
o
ur
c
e
da
t
a
,
pa
r
t
i
c
u
l
a
r
ly
i
n
t
h
e
do
m
a
i
n
o
f
s
e
g
men
t
a
t
i
o
n
,
w
i
t
h
a
s
p
e
c
i
f
i
c
f
o
c
u
s
o
n
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
A
s
a
r
e
s
u
l
t
,
a
da
t
a
s
e
t
wa
s
ge
n
e
r
a
t
e
d
t
h
a
t
i
nc
l
ude
s
s
e
v
e
r
a
l
f
r
u
i
t
i
t
e
m
s
t
o
h
e
l
p
w
i
t
h
t
h
e
t
a
s
k
o
f
m
u
l
t
ip
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
I
n
a
dd
i
t
i
o
n
,
t
h
e
tr
a
i
ni
ng
a
l
g
o
r
i
t
h
m
i
n
cl
u
d
ed
t
h
e
u
t
i
l
i
za
t
i
o
n
o
f
t
h
e
C
O
C
O
d
at
as
et
[
2
5
]
,
[
2
6
]
,
w
h
i
ch
b
o
t
h
i
n
cl
u
d
ed
f
o
o
d
-
r
el
at
ed
co
n
t
en
t
.
T
h
e
i
m
a
ge
s
us
e
d
i
n
t
hi
s
r
e
s
e
a
r
c
h
s
h
o
ws
v
a
r
i
o
us
l
e
v
e
l
s
o
f
r
e
s
o
l
u
t
i
o
n
.
S
e
v
e
r
a
l
c
h
a
ll
e
n
g
e
s
a
r
i
s
e
wh
e
n
s
e
l
e
c
t
i
n
g
i
m
a
ge
s
t
h
a
t
e
n
c
o
m
pa
s
s
a
r
a
n
ge
of
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
.
T
h
e
m
u
l
t
i
p
l
e
f
r
u
i
t
da
t
a
s
e
t
c
o
n
t
a
i
ns
27
d
i
s
t
i
n
c
t
c
l
a
s
s
e
s
.
He
n
c
e
,
t
h
e
da
t
a
s
e
t
c
o
m
pr
i
s
i
ng
s
e
v
e
r
a
l
f
r
u
i
t
i
m
a
ge
s
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:
2
5
0
2
-
4
7
52
A
c
c
ur
ate
s
e
gme
ntat
ion
of
f
r
uit
bas
e
d
on
de
e
p
lear
ning
(
E
s
r
aa
A
bu
E
ls
oud
)
1333
e
n
c
o
m
pa
s
s
e
s
a
tot
a
l
o
f
27
,
000
i
m
a
g
e
s
.
T
h
e
m
u
l
t
i
p
l
e
f
r
u
i
t
da
t
a
s
e
t
i
s
s
p
li
t
80%
in
t
h
e
tr
a
i
ni
n
g
s
t
a
ge
a
nd
20%
in
t
h
e
t
e
s
t
i
n
g
s
t
a
ge
.
4.
M
E
T
HO
D
T
h
e
pa
pe
r
pr
o
vi
de
s
s
i
g
ni
f
i
c
a
n
t
i
n
s
i
g
h
t
s
i
n
t
o
t
h
e
s
e
gm
e
n
t
a
t
i
o
n
o
f
m
u
l
t
i
-
f
o
o
d.
I
t
f
o
c
us
e
s
o
n
i
de
n
t
i
f
yi
ng
t
a
s
ks
r
e
l
a
t
e
d
to
m
u
l
t
i
p
l
e
f
r
u
i
t
i
t
e
m
s
,
a
s
we
ll
a
s
l
o
c
a
l
i
z
i
ng
a
n
d
s
e
g
m
e
n
t
i
n
g
e
a
c
h
f
r
u
i
t
i
t
e
m
b
a
s
e
d
o
n
t
h
e
i
nput
im
a
ge
.
T
h
e
s
e
t
a
s
ks
po
s
e
c
h
a
ll
e
n
ge
s
due
t
o
v
a
r
i
o
us
f
a
c
t
o
r
s
,
s
uc
h
a
s
t
h
e
pr
e
s
e
n
c
e
o
f
m
u
l
t
i
p
l
e
f
r
u
i
t
i
t
e
m
s
a
n
d
t
h
e
v
a
r
i
a
t
i
o
n
s
i
n
c
o
l
o
r
a
n
d
s
i
z
e
b
e
t
we
e
n
t
h
e
m
.
F
ur
t
h
e
r
m
o
r
e
,
t
hi
s
s
t
udy
i
n
t
r
o
duc
e
s
a
n
o
v
e
l
a
l
go
r
i
t
hm
t
h
a
t
e
nh
a
nc
e
s
m
a
ny
a
ppr
o
a
c
h
e
s
,
s
uc
h
a
s
C
NN
,
r
e
g
i
o
n
pr
o
p
o
s
a
l
n
e
t
wo
r
ks
(
R
P
N)
,
a
n
d
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
s
.
T
h
e
a
im
i
s
to
a
c
hi
e
v
e
o
p
t
i
m
a
l
r
e
s
u
l
t
s
by
a
c
c
ur
a
t
e
l
y
r
e
pr
e
s
e
n
t
i
n
g
f
r
u
i
t
i
m
a
g
e
s
.
4.
1.
T
h
e
e
n
h
a
n
c
e
d
m
u
l
t
ip
l
e
f
r
u
it
s
e
gm
e
n
t
at
ion
al
go
r
it
h
m
A
n
o
v
e
l
a
ppr
o
a
c
h
f
o
r
e
n
h
a
nc
e
d
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
h
a
s
b
e
e
n
de
v
e
l
o
pe
d,
l
e
v
e
r
a
g
i
n
g
a
n
im
pr
o
v
e
d
v
e
r
s
i
o
n
o
f
t
h
e
R
e
s
Ne
t
-
101
a
r
c
hi
t
e
c
t
ur
e
a
s
i
t
s
b
a
c
k
b
o
n
e
.
T
hi
s
m
o
d
i
f
i
e
d
de
e
p
l
e
a
r
ni
ng
m
o
de
l
h
a
s
b
e
e
n
s
pe
c
if
i
c
a
ll
y
o
pt
i
mi
z
e
d
t
o
e
x
t
r
a
c
t
m
o
r
e
n
ua
n
c
e
d
a
n
d
c
o
m
pr
e
h
e
ns
i
ve
f
r
u
i
t
f
e
a
t
ur
e
s
.
T
h
e
i
m
pr
o
v
e
d
R
e
s
Ne
t
-
101
b
a
c
k
b
o
n
e
us
e
s
c
utt
i
n
g
-
e
dge
m
e
t
h
o
ds
l
i
ke
d
i
l
a
t
e
d
c
o
n
v
o
l
ut
i
o
n
s
a
n
d
a
tt
e
n
t
i
o
n
m
e
c
h
a
ni
s
m
s
t
o
ga
th
e
r
b
o
t
h
s
m
a
l
l
d
e
t
a
i
l
s
a
n
d
bi
g
-
p
i
c
t
ur
e
i
nf
o
r
m
a
t
i
o
n
a
t
t
h
e
s
a
m
e
t
i
m
e
.
T
hi
s
r
e
s
u
l
t
s
i
n
m
o
r
e
a
c
c
ur
a
t
e
a
n
d
r
o
b
us
t
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
,
e
v
e
n
i
n
c
o
m
p
l
e
x
s
c
e
n
a
r
i
o
s
w
i
t
h
m
u
l
t
i
p
l
e
o
v
e
r
l
a
pp
i
ng
f
r
u
i
t
s
o
r
v
a
r
y
i
ng
li
g
h
t
i
n
g
c
o
n
d
i
t
i
o
n
s
.
T
h
e
im
pr
o
v
e
d
a
l
go
r
i
t
hm
n
o
t
o
nl
y
e
x
c
e
l
s
i
n
d
i
s
t
i
n
gu
i
s
hi
ng
b
e
t
we
e
n
d
if
f
e
r
e
n
t
t
y
pe
s
o
f
f
r
u
i
t
s
b
ut
a
l
s
o
pr
e
c
i
s
e
l
y
de
l
i
ne
a
t
e
s
t
h
e
i
r
b
o
un
da
r
i
e
s
,
m
a
k
i
ng
it
pa
r
t
i
c
u
l
a
r
l
y
v
a
l
ua
bl
e
f
o
r
a
pp
l
i
c
a
t
i
o
n
s
in
a
ut
o
m
a
t
e
d
h
a
r
v
e
s
t
i
n
g,
qua
l
i
t
y
c
o
n
t
r
o
l
in
t
h
e
f
o
o
d
i
n
dus
t
r
y
,
a
n
d
d
i
e
t
a
r
y
a
n
a
ly
s
i
s
s
ys
t
e
m
s
.
4.
1.
1.
I
m
p
r
ovin
g
Re
s
N
e
t
-
101
b
ac
k
b
on
e
f
o
r
b
e
t
t
e
r
f
e
at
u
r
e
e
x
t
r
ac
t
ion
C
NN
i
s
a
t
y
pe
o
f
de
e
p
l
e
a
r
ni
ng
a
r
c
hi
t
e
c
t
ur
e
t
h
a
t
r
e
l
i
e
s
o
n
a
v
a
r
i
e
t
y
o
f
i
n
put
s
,
s
uc
h
a
s
a
ud
i
o
,
pi
c
t
ur
e
,
t
e
x
t,
a
n
d
vi
de
o
.
T
h
e
hi
g
h
a
c
c
ur
a
c
y
o
f
C
NN
h
a
s
be
e
n
s
h
o
wn
i
n
v
a
r
i
o
us
do
m
a
i
ns
.
S
e
ve
r
a
l
w
i
d
e
l
y
r
e
c
o
gni
z
e
d
C
NN
a
r
c
hi
t
e
c
t
ur
e
s
t
h
a
t
s
e
r
v
e
as
ba
c
k
b
o
n
e
n
e
t
wo
r
k
s
i
nc
l
ude
Al
e
x
Ne
t
,
VGG,
a
n
d
R
e
s
Ne
t
-
101.
T
h
e
R
e
s
Ne
t
-
101
a
r
c
hi
t
e
c
t
ur
e
h
a
s
b
e
e
n
i
de
n
t
i
f
i
e
d
a
s
a
s
ui
t
a
bl
e
a
ppr
o
a
c
h
f
o
r
a
c
hi
e
vi
ng
i
m
pr
o
v
e
d
a
c
c
ur
a
c
y
i
n
o
bj
e
c
t
d
e
t
e
c
t
i
o
n
[
27]
.
T
h
e
R
e
s
Ne
t
-
101
m
o
de
l
a
c
hi
e
v
e
s
hi
g
h
pe
r
f
o
r
m
a
n
c
e
by
e
m
p
l
o
yi
ng
a
de
e
p
n
e
ur
a
l
n
e
t
wor
k
a
r
c
hi
t
e
c
t
ur
e
t
h
a
t
i
nc
o
r
por
a
t
e
s
a
s
e
r
i
e
s
o
f
bl
o
c
ks
de
s
i
g
n
e
d
t
o
s
o
l
v
e
t
h
e
pr
o
bl
e
m
o
f
gr
a
d
i
e
n
t
v
a
ni
s
hi
ng.
T
hi
s
i
s
a
c
hi
e
v
e
d
by
t
h
e
a
dd
i
t
i
o
n
of
s
h
o
r
t
c
u
t
c
o
n
n
e
c
t
i
o
ns
[
27]
.
T
h
e
a
r
c
hi
t
e
c
t
ur
e
of
R
e
s
Ne
t
-
101
i
n
c
o
r
por
a
t
e
s
t
w
o
t
y
pe
s
of
s
h
o
r
t
c
u
t
c
o
n
n
e
c
t
i
o
ns
.
1)
I
de
n
t
i
t
y
bl
o
c
k:
if
t
he
d
i
m
e
ns
i
o
n
s
o
f
t
h
e
i
nput
a
n
d
o
ut
pu
t
a
r
e
t
h
e
s
a
m
e
,
t
h
e
n
t
h
e
i
de
n
t
i
t
y
bl
o
c
k
c
a
n
be
a
pp
l
i
e
d
d
i
r
e
c
t
l
y
.
2)
C
o
n
v
o
l
ut
i
o
na
l
bl
o
c
k:
if
t
h
e
s
i
z
e
of
t
h
e
i
n
put
d
i
f
f
e
r
s
f
r
o
m
t
h
a
t
of
t
h
e
o
u
t
pu
t
,
a
c
o
n
v
o
l
ut
i
o
n
bl
o
c
k
is
us
e
d
to
a
d
j
u
s
t
t
h
e
s
i
z
e
o
f
t
h
e
i
n
put
.
T
h
e
gr
a
d
i
e
n
t
v
a
ni
s
hin
g
pr
o
bl
e
m
h
a
s
b
e
e
n
s
o
l
v
e
d
by
t
h
e
R
e
s
N
e
t
-
101
b
a
c
k
b
o
ne
w
i
t
h
t
h
e
i
m
p
l
e
m
e
n
t
a
t
i
o
n
of
s
h
o
r
t
c
ut
c
o
n
n
e
c
t
i
o
n
s
.
Ho
we
v
e
r
,
R
e
s
Ne
t
-
101
s
h
o
ws
s
e
v
e
r
a
l
i
s
s
ue
s
:
T
h
e
u
t
i
li
z
a
t
i
o
n
of
a
l
a
r
ge
f
i
l
t
e
r
s
i
z
e
i
n
t
h
e
i
ni
t
i
a
l
c
o
nv
o
l
ut
i
o
n
l
a
y
e
r
r
e
s
u
l
t
s
i
n
t
h
e
e
x
t
r
a
c
t
i
o
n
o
f
l
e
s
s
c
o
m
p
l
e
x
f
e
a
t
ur
e
s
.
R
e
s
Ne
t
-
101
bl
o
c
ks
m
i
g
h
t
h
a
v
e
r
e
c
e
i
ve
d
a
l
im
i
t
e
d
a
m
o
u
nt
o
f
r
e
pe
t
i
t
i
o
n
s
,
whil
e
ot
h
e
r
s
m
a
y
h
a
v
e
b
e
e
n
e
x
c
e
s
s
i
ve
ly
r
e
pe
a
t
e
d.
T
h
e
r
e
r
e
m
a
i
ns
a
ne
c
e
s
s
i
t
y
to
i
m
pr
o
v
e
t
h
e
de
t
e
c
t
i
o
n
r
a
t
e
f
o
r
f
r
u
i
t
r
e
c
o
gni
t
i
o
n
.
T
h
e
e
nh
a
n
c
e
d
R
e
s
Ne
t
-
101
a
r
c
hi
t
e
c
t
ur
e
h
a
s
t
h
r
e
e
s
t
a
ge
s
.
1)
T
h
e
f
i
r
s
t
c
o
n
v
o
l
ut
i
o
n
l
a
y
e
r
s
h
o
u
l
d
b
e
m
o
d
i
f
i
e
d
to
c
o
n
t
a
i
n
a
s
m
a
ll
e
r
f
il
t
e
r
s
i
z
e
.
2)
T
h
e
R
e
s
Ne
t
-
101
bl
o
c
k
s
h
o
u
l
d
b
e
o
p
t
i
mi
z
e
d
by
c
a
r
e
f
u
l
ly
s
e
l
e
c
t
i
n
g
t
h
e
a
ppr
o
pr
i
a
t
e
n
u
m
b
e
r
o
f
r
e
pe
t
i
t
i
o
ns
.
3)
T
o
e
n
h
a
n
c
e
t
h
e
m
o
de
l
’
s
pe
r
f
o
r
m
a
n
c
e
,
a
c
o
n
v
o
l
ut
i
o
na
l
l
a
ye
r
s
h
o
ul
d
b
e
i
n
t
r
o
duc
e
d
a
t
e
a
c
h
l
e
ve
l
.
T
h
e
s
ugge
s
t
e
d
e
n
h
a
nc
e
m
e
n
t
to
t
h
e
R
e
s
Ne
t
101
b
a
c
k
b
o
n
e
i
nv
o
l
v
e
s
r
e
duc
i
n
g
t
h
e
f
i
l
t
e
r
s
i
z
e
in
t
h
e
i
ni
t
i
a
l
c
o
nv
o
l
ut
i
o
n
l
a
y
e
r
a
n
d
a
r
r
a
n
g
i
ng
t
h
e
l
a
y
e
r
s
t
h
a
t
r
e
qui
r
e
m
o
r
e
i
t
e
r
a
t
i
o
ns
.
A
dd
i
t
i
o
n
a
ll
y
,
a
c
o
n
v
o
l
ut
i
o
n
a
l
l
a
y
e
r
i
s
a
dde
d
f
o
r
e
a
c
h
s
t
a
ge
to
e
nh
a
n
c
e
t
h
e
f
e
a
t
ur
e
s
e
l
e
c
t
i
o
n
pr
o
c
e
s
s
by
e
x
t
r
a
c
t
i
n
g
m
u
l
t
i
p
l
e
s
pe
c
i
f
i
c
a
n
d
e
x
a
c
t
f
e
a
t
ur
e
m
a
ps
f
r
o
m
t
h
e
s
h
a
l
l
o
w
l
a
y
e
r
s
.
F
i
gur
e
1
s
h
o
ws
t
h
e
i
m
pr
o
v
e
d
R
e
s
N
e
t
-
101
b
a
c
kb
o
ne
.
T
h
us
,
t
h
e
pr
e
s
e
n
t
s
t
udy
h
a
s
de
v
e
l
o
pe
d
a
n
o
v
e
l
a
ppr
o
a
c
h
f
o
r
de
t
e
c
t
i
n
g
s
e
v
e
r
a
l
f
o
o
ds
,
s
h
o
w
i
n
g
r
e
duc
e
d
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
dur
a
t
i
o
n
a
n
d
e
nh
a
n
c
e
d
a
c
c
ur
a
c
y
,
s
o
l
vi
n
g
t
h
e
a
b
o
v
e
m
e
n
t
i
o
n
e
d
i
s
s
ue
s
.
4.
1.
2.
F
il
t
e
r
s
i
z
e
m
od
if
icat
ion
T
h
e
f
il
t
e
r
s
t
h
a
t
a
r
e
us
e
d
by
t
h
e
C
NN
a
r
e
ut
i
li
z
e
d
to
f
i
nd
t
h
e
f
e
a
t
ur
e
m
a
p
f
r
o
m
t
h
e
i
nput
i
m
a
ge
by
us
i
ng
a
s
u
i
t
a
bl
e
f
i
l
t
e
r
s
i
z
e
.
Va
r
i
o
us
f
il
t
e
r
s
i
z
e
s
a
r
e
us
e
d
to
e
x
t
r
a
c
t
s
pe
c
i
f
i
c
de
t
a
i
l
s
f
r
o
m
t
h
e
i
m
a
ge
s
w
i
t
hi
n
t
h
e
C
NN
l
a
y
e
r
.
T
h
e
go
a
l
o
f
t
hi
s
s
t
ud
y
i
s
to
i
nve
s
t
i
ga
t
e
t
h
e
tec
hni
que
s
ut
i
li
z
e
d
f
o
r
f
r
u
i
t
r
e
c
o
gni
t
i
o
n
.
He
n
c
e
,
s
i
nc
e
m
a
ny
c
o
m
put
a
t
i
o
n
s
a
r
e
r
e
qu
i
r
e
d
in
t
hi
s
l
a
y
e
r
to
p
r
o
duc
e
a
s
i
n
gu
l
a
r
v
a
l
ue
as
a
f
e
a
t
ur
e
m
a
p,
it
is
a
d
vi
s
a
bl
e
to
a
v
o
i
d
us
i
n
g
7
×
7,
as
t
h
e
y
f
r
e
que
n
t
l
y
s
i
m
p
li
f
y
t
h
e
f
e
a
t
ur
e
s
a
n
d
m
a
y
o
v
e
r
l
o
o
k
i
m
po
r
t
a
n
t
de
t
a
i
l
s
c
o
n
t
a
i
ne
d
i
n
t
he
i
m
a
ge
s
.
On
t
h
e
ot
h
e
r
h
a
n
d,
c
h
o
o
s
i
n
g
a
s
m
a
l
l
e
r
f
il
t
e
r
s
i
z
e
,
s
uc
h
a
s
3
×
3,
pr
o
vi
de
s
a
gr
e
a
t
e
r
a
m
o
un
t
o
f
de
t
a
i
l
,
th
e
r
e
f
o
r
e
l
e
a
d
i
ng
t
o
e
n
h
a
n
c
e
d
a
c
c
ur
a
c
y
[
28]
,
[
29]
.
T
h
e
r
e
f
o
r
e
,
to
e
x
t
r
a
c
t
a
gr
e
a
t
e
r
n
u
m
b
e
r
o
f
f
e
a
t
ur
e
s
,
t
h
e
s
ugge
s
t
e
d
b
a
c
k
b
o
ne
e
m
p
l
o
y
e
d
a
s
m
a
ll
e
r
f
il
t
e
r
s
i
z
e
o
f
3
×
3
i
n
s
t
e
a
d
o
f
t
h
e
7
×
7
f
il
t
e
r
s
i
z
e
a
pp
l
i
e
d
i
n
R
e
s
Ne
t
-
101.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
8
,
N
o
.
2
,
M
a
y
20
2
5
:
1
331
-
1
338
1334
F
i
gur
e
1.
E
n
ha
n
c
e
d
R
e
s
Ne
t
-
101
b
a
c
kb
o
ne
4.
1.
3.
Op
t
im
iz
in
g
t
h
e
Re
s
Ne
t
-
101
b
l
oc
k
W
hi
c
h
pr
e
ve
n
t
s
t
h
e
a
c
hi
e
v
e
m
e
n
t
o
f
i
m
pr
o
v
e
d
a
c
c
ur
a
c
y
a
n
d
r
e
duc
e
d
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
t
i
m
e
s
.
T
h
e
f
r
e
que
n
c
y
of
i
t
e
r
a
t
i
o
n
s
is
e
s
t
a
bl
i
s
h
e
d
t
h
r
o
ugh
e
x
pe
r
i
m
e
n
t
a
t
i
o
n
pe
r
f
o
r
m
e
d
on
t
h
e
de
s
i
g
n
e
d
R
e
s
Ne
t
-
101
ba
c
kb
o
ne
a
b
o
ut
f
r
u
i
t
da
t
a
.
T
h
e
e
x
pe
r
i
m
e
n
t
i
nc
l
ude
s
e
v
a
l
ua
t
i
n
g
t
h
e
a
c
c
ur
a
c
y
o
f
t
h
e
R
e
s
Ne
t
bl
o
c
k.
I
f
t
h
e
n
e
w
a
c
c
ur
a
c
y
pa
s
s
e
s
t
h
e
o
l
d
a
c
c
ur
a
c
y
,
t
h
e
R
e
s
Ne
t
bl
o
c
k
i
s
r
e
c
ur
s
i
ve
l
y
r
e
pe
a
t
e
d
u
n
t
i
l
a
hi
g
h
e
r
l
e
v
e
l
o
f
a
c
c
ur
a
c
y
i
s
a
c
hi
e
v
e
d.
T
h
e
r
e
pe
t
i
t
i
o
n
s
s
t
o
p
wh
e
n
t
h
e
n
e
w
a
c
c
ur
a
c
y
c
o
m
e
s
un
de
r
t
h
e
o
l
d
a
c
c
ur
a
c
y
.
As
a
n
o
u
t
c
o
m
e
,
t
h
e
pr
o
c
e
s
s
o
f
o
p
t
i
mi
z
a
t
i
o
n
i
nv
o
l
v
e
s
de
t
e
r
m
i
n
i
ng
t
h
e
a
ppr
o
pr
i
a
t
e
n
u
m
be
r
of
r
e
pe
a
t
s
.
T
h
e
a
f
o
r
e
m
e
n
t
i
o
n
e
d
pr
o
c
e
dur
e
h
a
s
be
e
n
c
o
m
p
l
e
t
e
d
to
i
de
n
t
i
f
y
t
h
e
bl
o
c
k
s
t
h
a
t
r
e
qui
r
e
a
dd
i
t
i
o
n
a
l
i
t
e
r
a
t
i
o
ns
or
f
e
we
r
i
t
e
r
a
t
i
o
ns
.
T
hi
s
pr
o
c
e
dur
e
is
e
x
e
c
ut
e
d
to
e
n
h
a
n
c
e
t
h
e
c
o
m
put
a
t
i
o
n
a
l
t
i
m
e
e
f
f
i
c
i
e
n
c
y
.
Af
t
e
r
t
h
a
t,
t
h
e
r
e
s
u
l
t
s
o
f
e
a
c
h
s
t
e
p
a
r
e
c
o
m
bi
ne
d
w
i
t
h
a
n
e
x
t
r
a
c
o
n
v
o
l
ut
i
o
n
a
l
l
a
y
e
r
to
m
a
ke
th
e
s
ugge
s
t
e
d
R
e
s
Ne
t
-
101
b
a
c
k
b
o
ne
m
o
r
e
a
c
c
ur
a
t
e
.
T
o
l
o
c
a
li
z
e
a
n
d
s
e
g
m
e
n
t
f
o
o
d,
t
h
e
e
nh
a
n
c
e
d
m
u
l
t
ip
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
i
nc
o
r
por
a
t
e
s
t
h
e
R
P
N
a
n
d
t
h
e
m
a
s
k
b
r
a
n
c
h
s
ugge
s
t
e
d
by
c
i
t
e
l
i
2017
f
u
l
ly
.
T
hi
s
a
l
go
r
i
t
hm
e
x
t
r
a
c
t
s
f
e
a
t
ur
e
s
s
pe
c
i
f
i
c
t
o
e
a
c
h
R
o
I
wh
e
r
e
f
r
u
i
t
i
s
i
de
n
t
i
f
i
e
d.
T
h
e
m
a
s
k
b
r
a
n
c
h
pr
o
duc
e
s
a
m
a
s
k
w
i
t
h
d
i
m
e
ns
i
o
n
s
’
m
x
m
f
o
r
e
v
e
r
y
R
o
I
a
c
c
o
r
di
n
g
to
[
4
]
.
T
h
e
tot
a
l
n
u
m
be
r
of
c
l
a
s
s
if
i
c
a
t
i
o
n
s
in
t
hi
s
c
on
t
e
x
t
can
be
r
e
pr
e
s
e
n
t
e
d
as
K
+
1,
wh
e
r
e
K
de
n
o
t
e
s
th
e
c
o
un
t
of
de
t
e
c
t
a
bl
e
i
t
e
m
s
pr
e
s
e
n
t
in
t
h
e
im
a
ge
.
T
he
a
dd
it
i
o
n
a
l
c
l
a
s
s
is
s
pe
c
if
i
c
a
ll
y
d
e
s
i
g
n
a
t
e
d
f
o
r
t
h
e
ba
c
kgr
o
un
d.
A
C
N
N
-
b
a
s
e
d
m
a
s
k
c
l
a
s
s
if
i
e
r
i
s
e
m
p
l
o
y
e
d
t
o
c
l
a
s
s
i
f
y
t
h
e
f
i
xe
d
f
e
a
t
ur
e
s
a
f
t
e
r
t
h
e
i
r
c
r
e
a
t
i
o
n
.
M
u
l
t
i
p
le
m
a
s
ks
a
r
e
pr
o
duc
e
d
f
o
r
di
f
f
e
r
e
n
t
o
bj
e
c
t
s
,
wi
t
h
n
o
c
om
pe
t
i
t
i
o
n
o
b
s
e
r
v
e
d
b
e
t
we
e
n
c
l
a
s
s
e
s
.
C
o
nv
e
n
t
i
o
na
l
n
e
ur
a
l
n
e
t
wo
r
ks
t
y
p
i
c
a
l
ly
e
m
p
l
o
y
t
h
e
r
e
g
i
o
ns
c
h
o
s
e
n
by
R
o
I
c
l
a
s
s
if
i
e
r
s
a
s
i
n
put
to
ge
n
e
r
a
t
e
a
m
a
s
k.
T
hi
s
e
n
s
ur
e
s
t
h
a
t
s
pa
t
i
a
l
l
o
c
a
t
i
o
n
s
a
n
d
f
e
a
t
ur
e
s
a
r
e
s
u
i
t
a
bly
s
i
z
e
d.
T
h
e
r
e
s
u
l
t
i
n
g
m
a
t
r
i
x
c
o
ns
i
s
t
s
of
e
l
e
m
e
n
t
s
wi
t
h
a
v
a
l
ue
of
1
i
n
d
i
c
a
t
i
n
g
t
h
e
l
i
ke
li
h
o
o
d
of
an
o
bj
e
c
t
’
s
pr
e
s
e
n
c
e
,
w
hi
l
e
e
l
e
m
e
n
t
s
w
i
t
h
a
v
a
l
ue
of
0
i
n
d
i
c
a
t
e
t
h
e
a
b
s
e
nc
e
o
f
a
ny
i
t
e
m
s
.
T
hi
s
m
a
t
r
i
x
i
s
c
o
m
m
o
nly
r
e
f
e
r
r
e
d
to
a
s
a
bi
n
a
r
y
m
a
s
k.
T
h
e
m
a
s
k
h
e
a
d
ut
i
l
i
z
e
s
t
h
e
f
u
l
ly
c
o
n
v
o
l
ut
i
o
n
a
l
n
e
t
wo
r
k
(
F
C
N)
to
pr
e
di
c
t
t
h
e
m
a
s
k
a
t
a
pi
x
e
l
l
e
ve
l
f
o
r
e
a
c
h
R
o
I
.
T
h
e
pr
e
s
e
n
t
s
t
udy
e
mp
l
o
y
s
t
h
e
s
pe
c
i
a
li
z
e
d
c
l
a
s
s
if
i
c
a
t
i
o
n
b
r
a
n
c
h
pr
o
p
o
s
e
d
by
[
4]
to
m
a
ke
pr
e
d
i
c
t
i
o
n
s
r
e
ga
r
d
i
ng
t
h
e
i
de
n
t
i
f
i
c
a
t
i
o
n
o
f
m
a
s
k
I
Ds
.
T
h
e
o
u
t
pu
t
o
f
t
h
e
e
nha
n
c
e
d
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
e
n
c
o
m
pa
s
s
e
s
c
l
a
s
s
i
f
i
c
a
t
i
o
n
l
a
b
e
l
s
,
b
o
un
d
i
ng
b
o
x
e
s
,
a
n
d
m
a
s
k
s
f
o
r
e
a
c
h
i
de
n
t
i
f
i
e
d
f
r
u
i
t
i
t
e
m
,
a
s
i
ll
us
t
r
a
t
e
d
i
n
F
i
gur
e
1.
T
h
e
l
o
s
s
f
u
n
c
t
i
o
n
f
o
r
t
hi
s
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:
2
5
0
2
-
4
7
52
A
c
c
ur
ate
s
e
gme
ntat
ion
of
f
r
uit
bas
e
d
on
de
e
p
lear
ning
(
E
s
r
aa
A
bu
E
ls
oud
)
1335
m
o
de
l
c
o
m
pr
i
s
e
s
t
h
r
e
e
c
o
m
po
n
e
n
t
s
:
c
l
a
s
s
if
i
c
a
t
i
o
n
l
o
s
s
(
L
c
l
s
)
,
b
o
un
d
i
ng
b
o
x
l
o
s
s
(
L
b
o
x
)
,
a
n
d
m
a
s
k
l
o
s
s
(
L
m
a
s
k)
.
T
h
e
m
a
s
k
l
o
s
s
is
c
a
l
c
u
l
a
t
e
d
as
t
h
e
a
v
e
r
a
ge
bi
na
r
y
c
r
o
s
s
-
e
n
t
r
o
py
l
o
s
s
,
as
de
f
i
ne
d
by
[
4]
.
T
h
e
o
v
e
r
a
ll
l
o
s
s
f
u
n
c
t
i
o
n
f
o
r
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
c
a
n
be
m
a
t
h
e
m
a
t
i
c
a
ll
y
r
e
pr
e
s
e
n
t
e
d
as
(
1
)
.
=
+
+
(1
)
5.
E
XP
E
RI
M
E
NT
A
L
RE
S
UL
T
S
AND
RE
S
UL
T
T
h
e
e
x
pe
r
im
e
n
t
a
l
r
e
s
u
l
t
s
a
r
e
de
t
a
i
l
e
d
i
n
t
hi
s
s
e
c
t
i
o
n
,
a
n
d
t
h
e
e
v
a
l
ua
t
i
o
n
o
f
t
h
e
a
l
go
r
i
t
hm
i
s
d
i
s
c
u
s
s
e
d
in
t
h
e
pr
e
c
e
d
i
ng
s
e
c
t
i
o
ns
.
T
h
e
e
v
a
l
u
a
t
i
o
n
of
t
h
e
m
u
lt
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
w
a
s
c
o
n
du
c
t
e
d
u
s
i
n
g
t
wo
m
e
a
s
ur
e
m
e
n
t
s
:
T
h
e
s
t
ud
y
e
m
p
l
o
y
e
d
s
e
ve
r
a
l
t
h
r
e
s
h
o
l
ds
o
f
a
v
e
r
a
ge
pr
e
c
i
s
i
o
n
(
A
P
)
to
a
s
s
e
s
s
a
n
d
c
o
m
pa
r
e
t
h
e
pe
r
f
o
r
m
a
n
c
e
of
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
s
.
A
dd
i
t
i
o
na
l
ly
,
t
h
e
s
t
ud
y
m
e
a
s
ur
e
d
t
h
e
t
i
m
e
r
e
qu
i
r
e
d
f
o
r
b
o
t
h
t
h
e
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
ph
a
s
e
s
.
5.
1.
E
x
p
e
r
im
e
n
t
a
l
p
ar
a
m
e
t
e
r
s
An
a
l
go
r
i
t
hm
f
o
r
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
on
wa
s
i
m
p
l
e
m
e
n
t
e
d
us
i
ng
T
e
n
s
o
r
F
l
o
w.
T
h
e
m
e
t
h
o
do
l
o
g
i
e
s
we
r
e
a
s
s
e
s
s
e
d
w
i
t
h
Am
a
z
o
n
W
e
b
S
e
r
vi
c
e
s
(
AW
S
)
a
n
d
Am
a
z
o
n
M
a
c
hi
ne
I
m
a
ge
(
A
M
I
)
,
e
m
p
l
o
yi
ng
t
h
e
GPU
-
Us
-
T
e
s
l
a
V100
w
i
t
h
16
GB
o
f
m
e
m
o
r
y
a
n
d
VC
P
U
s
-
8
c
o
r
e
s
wi
t
h
61
GB
o
f
m
e
m
o
r
y
.
F
ur
t
h
e
r
m
o
r
e
,
t
h
e
o
p
t
i
mi
z
a
t
i
o
n
a
l
go
r
i
t
hm
ut
i
l
i
z
e
d
i
n
t
hi
s
wo
r
k
wa
s
s
to
c
h
a
s
t
i
c
gr
a
d
i
e
n
t
de
s
c
e
n
t
(
S
G
D)
.
T
h
e
we
i
g
h
t
de
c
a
y
wa
s
e
s
t
a
bli
s
he
d
a
t
0
.
0002,
wi
t
h
a
l
e
a
r
ni
ng
m
o
m
e
n
t
u
m
o
f
0.
8
a
n
d
a
l
e
a
r
ni
n
g
r
a
t
e
o
f
0.
002,
o
v
e
r
t
h
e
c
o
u
r
s
e
o
f
40
e
po
c
h
s
.
T
h
e
e
po
c
h
c
o
n
s
i
s
t
e
d
of
1000
i
t
e
r
a
t
i
o
n
s
.
5.
2.
Op
t
im
iz
at
ion
Re
s
Ne
t
-
101
b
ac
k
b
on
e
T
h
e
pr
e
c
i
s
i
o
n
of
f
r
u
i
t
s
e
gm
e
n
t
a
t
i
o
n
is
a
ug
m
e
n
t
e
d
by
i
n
c
o
r
por
a
t
i
n
g
s
upp
l
e
m
e
n
t
a
r
y
c
o
nv
o
l
ut
i
o
n
a
l
l
a
y
e
r
s
a
t
e
a
c
h
t
i
e
r
a
n
d
s
e
l
e
c
t
i
n
g
a
n
o
pt
i
mi
s
e
d
b
a
c
k
b
o
ne
t
h
a
t
a
c
c
o
un
t
s
f
o
r
t
h
e
e
x
a
c
t
n
u
m
be
r
o
f
r
e
pe
t
i
t
i
o
ns
w
i
t
hi
n
e
a
c
h
n
e
ur
a
l
bl
o
c
k.
T
h
e
o
p
t
i
mi
s
a
t
i
o
n
pr
o
c
e
dur
e
wa
s
e
x
e
c
ut
e
d
us
i
n
g
f
r
u
i
t
ph
o
to
g
r
a
ph
s
a
s
t
h
e
b
a
s
i
s
.
T
h
e
i
n
t
e
gr
a
t
i
o
n
o
f
c
o
n
v
o
l
ut
i
o
n
a
l
l
a
y
e
r
s
a
t
e
v
e
r
y
l
e
v
e
l
e
nh
a
n
c
e
s
th
e
a
c
c
ur
a
c
y
o
f
f
r
u
i
t
r
e
c
o
gni
t
i
o
n
.
T
hi
s
t
e
c
hni
qu
e
i
nv
o
l
ve
s
ga
t
h
e
r
i
n
g
l
o
c
a
l
f
e
a
t
ur
e
m
a
ps
f
r
o
m
e
a
c
h
s
t
a
ge
,
s
ub
s
e
que
n
t
l
y
l
e
a
d
i
ng
t
o
t
h
e
e
x
t
r
a
c
t
i
o
n
o
f
t
h
e
f
i
na
l
f
e
a
t
ur
e
m
a
ps
.
T
h
e
u
l
t
i
m
a
t
e
f
e
a
t
ur
e
m
a
p
i
s
o
b
t
a
i
n
e
d
by
s
u
mm
i
n
g
t
h
e
l
o
c
a
l
f
e
a
t
ur
e
m
a
p
s
f
r
o
m
e
a
c
h
s
t
a
ge
.
T
h
e
r
e
a
f
t
e
r
,
t
h
e
f
e
a
t
ur
e
m
a
p
f
u
nc
t
i
o
ns
a
s
t
he
i
n
put
f
o
r
t
h
e
s
u
b
s
e
q
ue
n
t
e
l
e
m
e
n
t
i
n
t
h
e
m
u
l
t
i
-
f
r
u
i
t
r
e
c
o
gni
t
i
o
n
pr
o
c
e
dur
e
.
T
h
e
R
P
N
f
u
n
c
t
i
o
n
s
w
i
t
hi
n
t
h
e
f
a
s
t
e
r
R
-
C
NN
a
r
c
hi
t
e
c
t
ur
e
.
T
h
e
e
f
f
i
c
a
c
y
o
f
t
h
e
a
ugm
e
n
t
e
d
R
e
s
Ne
t
-
101
m
o
de
l
i
s
i
ll
us
t
r
a
t
e
d
i
n
F
i
gur
e
1.
E
x
pe
r
i
m
e
n
t
s
we
r
e
pe
r
f
o
r
m
e
d
f
o
r
e
a
c
h
l
e
v
e
l
o
f
R
e
s
Ne
t
-
101,
a
n
d
t
h
e
i
de
a
l
n
u
m
be
r
o
f
r
e
pe
t
i
t
i
o
ns
o
f
t
h
e
R
e
s
N
e
t
-
101
bl
o
c
k
wa
s
e
s
t
a
bli
s
h
e
d
po
s
t
-
t
r
a
i
ni
ng.
S
ubs
e
que
n
t
l
y
,
s
upp
l
e
m
e
n
t
a
r
y
c
o
nv
o
l
ut
i
o
n
a
l
l
a
y
e
r
s
we
r
e
i
n
t
e
gr
a
t
e
d
i
n
t
o
e
a
c
h
l
e
v
e
l
.
T
h
e
a
d
vi
s
e
d
n
u
m
be
r
o
f
r
e
pe
t
i
t
i
o
n
s
f
o
r
t
h
e
a
ug
m
e
n
t
e
d
R
e
s
Ne
t
-
101
bl
o
c
k,
f
o
l
l
o
w
i
ng
t
h
e
i
n
c
o
r
por
a
t
i
o
n
o
f
t
h
e
c
o
n
v
o
l
ut
i
o
n
l
a
y
e
r
,
i
s
de
t
a
i
l
e
d
be
l
o
w:
T
h
e
pr
o
p
o
s
e
d
a
r
c
hi
t
e
c
t
ur
e
h
a
s
o
n
e
c
o
n
v
o
l
ut
i
o
n
bl
o
c
k
s
uc
c
e
e
de
d
by
t
h
r
e
e
r
e
pe
a
t
i
n
g
i
d
e
n
t
i
t
y
bl
o
c
ks
.
T
h
e
pr
o
p
o
s
e
d
a
r
c
hi
t
e
c
t
u
r
e
h
a
s
o
n
e
c
o
n
v
o
l
ut
i
o
n
bl
o
c
k
s
uc
c
e
e
d
e
d
by
t
h
r
e
e
r
e
pe
a
t
i
n
g
i
de
n
t
i
t
y
bl
o
c
ks
.
T
h
e
pr
o
po
s
e
d
a
r
c
hi
t
e
c
t
ur
e
h
a
s
o
n
e
c
o
nv
o
l
ut
i
o
n
bl
o
c
k
s
uc
c
e
e
de
d
by
13
r
e
pe
a
t
i
n
g
i
de
n
t
i
t
y
bl
o
c
k
s
.
T
h
e
pr
o
p
o
s
e
d
a
r
c
hi
t
e
c
t
ur
e
h
a
s
o
n
e
c
o
nv
o
l
u
t
i
o
n
bl
o
c
k
s
uc
c
e
e
de
d
by
t
wo
r
e
pe
a
t
e
d
i
d
e
n
t
i
t
y
bl
o
c
ks
.
5.
3.
E
val
u
at
ion
r
e
s
u
l
t
s
of
t
h
e
e
n
h
a
n
c
e
d
m
u
l
t
ip
l
e
f
r
u
it
s
e
gm
e
n
t
at
ion
al
gor
it
h
m
wit
h
d
if
f
e
r
e
n
t
s
t
at
e
-
of
-
t
h
e
-
ar
t
al
go
r
it
h
m
s
A
c
o
m
p
a
r
a
t
i
v
e
a
n
a
ly
s
i
s
wa
s
c
a
r
r
i
e
d
o
u
t
to
e
v
a
l
ua
t
e
t
h
e
pe
r
f
o
r
m
a
n
c
e
o
f
t
h
e
s
ugge
s
t
e
d
a
l
go
r
i
t
hm
i
n
r
e
l
a
t
i
o
n
to
t
h
e
m
o
s
t
a
dv
a
n
c
e
d
a
l
go
r
i
t
hm
s
.
S
ub
s
e
qu
e
n
t
l
y
,
t
h
e
s
e
a
l
go
r
i
t
hm
s
u
n
de
r
we
n
t
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
us
i
ng
d
i
f
f
e
r
e
n
t
c
r
i
t
e
r
i
a
.
T
h
e
c
o
n
t
r
a
s
t
i
s
s
h
o
wn
i
n
T
a
bl
e
1.
T
h
e
r
e
s
u
l
t
de
m
o
n
s
t
r
a
t
e
s
t
h
a
t
t
h
e
s
ugge
s
t
e
d
a
l
go
r
i
t
hm
d
i
s
p
l
a
y
s
s
upe
r
i
o
r
r
e
s
u
l
t
s
.
T
h
e
s
ugge
s
t
e
d
e
n
h
a
nc
e
d
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
yi
e
l
ds
o
p
t
i
m
a
l
r
e
s
u
l
t
s
o
wi
n
g
t
o
t
h
e
r
e
f
i
ne
d
R
e
s
N
e
t
-
101
b
a
c
kb
o
ne
,
a
c
hi
e
v
e
d
by
s
e
l
e
c
t
i
n
g
bl
o
c
k
dup
l
i
c
a
t
e
s
a
n
d
r
e
duc
in
g
f
il
t
e
r
si
ze.
A
dd
i
t
i
o
na
l
ly
,
by
i
nc
o
r
po
r
a
t
i
n
g
t
h
e
c
o
n
v
o
l
ut
i
o
n
a
l
l
a
y
e
r
.
T
h
e
e
f
f
i
c
a
c
y
o
f
t
h
e
i
m
pr
o
v
e
d
m
u
l
t
ip
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
in
c
o
m
pa
r
i
s
o
n
to
ot
h
e
r
l
e
a
d
i
n
g
a
l
go
r
i
t
hm
s
.
T
h
e
r
e
s
u
l
t
s
i
n
d
i
c
a
t
e
t
h
a
t
t
h
e
s
ugge
s
t
e
d
e
nh
a
nc
e
d
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
o
f
f
e
r
s
a
s
upe
r
i
o
r
s
o
l
ut
i
o
n
.
T
h
e
pr
o
p
o
s
e
d
a
ppr
o
a
c
h
e
nh
a
n
c
e
s
v
a
r
i
o
us
f
r
u
i
t
e
x
a
m
p
l
e
s
by
ut
i
li
z
i
ng
t
h
e
vi
s
ua
l
t
e
s
t
i
n
g
r
e
s
u
l
t
s
f
r
o
m
ot
h
e
r
a
l
go
r
i
t
hm
s
,
as
i
l
l
u
s
t
r
a
t
e
d
in
t
h
e
F
i
gur
e
2.
T
a
bl
e
1.
E
v
a
l
ua
t
i
o
n
r
e
s
u
l
t
s
o
f
t
h
e
e
nha
n
c
e
d
m
u
l
t
i
p
le
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
a
l
go
r
i
t
hm
w
i
t
h
d
i
f
f
e
r
e
n
t
a
l
go
r
i
t
hm
s
A
lg
o
r
i
th
ms
A
P
50
A
P
75
A
P
90
M
a
s
k
R
-
C
N
N
91.02
77.82
68.97
Y
O
L
A
C
T
89.27
77.48
68.83
C
A
S
C
A
D
E
R
-
C
N
N
94.41
82.71
71.02
P
r
o
p
o
s
e
d
a
lg
o
r
it
hm
97.13
86.22
72.99
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
8
,
N
o
.
2
,
M
a
y
20
2
5
:
1
331
-
1
338
1336
F
i
gur
e
2.
T
h
e
vi
s
ua
l
t
e
s
t
i
n
g
f
i
nd
i
n
g
s
o
f
t
h
e
pr
o
p
o
s
e
d
a
ppr
o
a
c
h
de
m
o
ns
t
r
a
t
e
t
h
e
i
m
pr
o
v
e
d
de
t
e
c
t
i
o
n
o
f
n
u
m
e
r
o
us
f
r
u
i
t
i
ns
t
a
n
c
e
s
wh
e
n
c
o
m
pa
r
e
d
to
e
xi
s
t
i
ng
a
l
go
r
i
t
hm
s
6.
CONC
L
USI
ON
T
h
e
pr
e
s
e
n
t
s
t
udy
a
tt
e
m
pt
s
to
de
v
e
l
o
p
de
e
p
l
e
a
r
ni
n
g
m
e
t
h
o
ds
f
o
r
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
f
o
r
hi
g
h
pe
r
f
o
r
m
a
n
c
e
.
T
hi
s
s
t
udy
us
e
s
de
e
p
l
e
a
r
ni
ng
t
o
i
m
pr
o
v
e
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
I
m
pr
o
v
e
m
e
nt
s
to
t
h
e
R
e
s
Ne
t
-
101
b
a
c
k
b
o
n
e
im
pr
o
v
e
d
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
S
e
l
e
c
t
i
o
n
of
t
h
e
r
i
g
h
t
n
u
m
be
r
of
r
e
pe
t
i
t
i
o
n
bl
o
c
ks
a
n
d
f
il
t
e
r
s
i
z
e
r
e
duc
t
i
o
n
i
n
t
h
e
e
a
r
ly
s
t
e
p
a
c
hi
e
v
e
d
t
hi
s
.
T
hi
s
m
e
t
h
o
d
e
x
t
r
a
c
t
e
d
m
o
r
e
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
f
r
o
m
i
nput
ph
o
to
s
.
T
h
us
,
t
hi
s
s
t
ud
y
ge
n
e
r
a
t
e
d
a
n
i
m
pr
o
v
e
d
R
e
s
Ne
t
-
101
b
a
c
k
b
o
n
e
f
o
r
m
u
l
t
i
p
l
e
f
r
u
i
t
s
e
g
m
e
n
t
a
t
i
o
n
.
AP
a
c
c
ur
a
c
y
a
c
r
o
s
s
t
h
r
e
s
h
o
l
d
s
i
s
b
e
t
t
e
r
wi
t
h
t
h
e
pr
o
p
os
e
d
a
l
go
r
i
t
hm
.
T
hi
s
s
t
ud
y
’
s
t
h
r
e
s
h
o
l
ds
we
r
e
hi
g
h
e
r
t
h
a
n
t
h
e
t
h
r
e
e
s
e
g
m
e
n
t
a
t
i
o
n
m
e
t
h
o
ds
.
W
i
t
h
a
c
c
ur
a
c
y
a
n
d
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
t
i
m
e
,
t
h
e
a
l
go
r
i
t
hm
u
n
de
r
c
o
n
s
i
de
r
a
t
i
o
n
c
a
n
qu
i
c
k
ly
a
n
d
a
c
c
ur
a
t
e
l
y
i
de
n
t
i
f
y
,
de
t
e
c
t,
a
n
d
s
e
g
m
e
n
t
v
a
r
i
o
us
f
r
u
i
t
i
t
e
m
s
.
Ho
we
v
e
r
,
t
hi
s
s
t
ud
y
us
e
d
27,
000
ph
o
to
s
f
r
o
m
t
r
a
i
ni
ng
a
n
d
t
e
s
t
i
n
g
s
e
t
s
.
T
h
us
,
to
i
m
pr
o
v
e
a
l
go
r
i
t
hm
pr
e
c
i
s
i
o
n
a
n
d
r
e
s
i
li
e
n
c
e
,
m
o
r
e
da
t
a
b
a
s
e
im
a
ge
s
s
h
o
u
l
d
be
e
nl
a
r
ge
d.
AC
K
NOWL
E
DGE
M
E
NT
T
hi
s
r
e
s
e
a
r
c
h
is
f
u
n
d
e
d
by
t
h
e
De
a
ns
hi
p
of
R
e
s
e
a
r
c
h
a
n
d
Gr
a
dua
t
e
S
t
udi
e
s
in
Z
a
r
qa
Uni
ve
r
s
i
t
y
/
J
o
r
da
n
.
RE
F
E
R
E
NC
E
S
[
1]
N
.
A
ls
hda
if
a
t,
M
.
A
.
O
s
ma
n,
a
nd
A
.
Z
.
T
a
li
b,
“
A
n
im
p
r
ov
e
d
m
ul
ti
-
o
bj
e
c
t
in
s
ta
nc
e
s
e
gm
e
nt
a
ti
o
n
ba
s
e
d
o
n
de
e
p
l
e
a
r
ni
ng,”
K
u
w
ai
t
J
our
nal
of
S
c
ie
nc
e
, v
o
l.
49, n
o
. 2, 2022
,
d
o
i:
10.48129/kj
s
.1087
9
.
[
2]
A
.
A
l
-
M
o
ma
ni
,
M
.
N
.
A
l
-
R
e
f
a
i,
S
.
A
buo
w
a
id
a
,
M
.
A
r
a
bi
a
t,
N
.
A
ls
hda
if
a
t,
a
nd
M
.
N
.
A
.
R
a
hma
n,
“
T
he
e
f
f
e
c
t
of
t
e
c
hn
o
l
ogi
c
a
l
c
o
nt
e
x
t
o
n s
ma
r
t
h
o
m
e
a
d
o
pt
i
o
n i
n j
or
da
n,”
I
ndone
s
ia
n J
our
nal
of
E
le
c
tr
ic
al
E
ngi
ne
e
r
in
g and C
om
put
e
r
Sc
ie
nc
e
, v
ol
. 33, n
o
. 2
, p.
1186
-
1
195, 2024
, do
i
:
10.11591/i
j
e
e
c
s
.
v
33.i
2.pp1186
-
1195
.
[
3]
B
.
A
l
-
N
a
a
mi
,
H
.
A
bu
O
w
id
a
,
M
.
A
bu
M
a
ll
o
uh,
F
.
A
l
-
N
a
im
a
t,
M
.
A
gha
,
a
nd
A
.
-
R
.
A
l
-
H
in
na
w
i,
“
A
ne
w
pr
o
t
o
t
y
p
e
of
s
ma
r
t
w
e
a
r
a
bl
e
m
o
ni
t
o
r
in
g
s
y
s
t
e
m s
o
lu
ti
o
n
f
or
a
l
z
h
e
im
e
r
’
s
pa
ti
e
nt
s
,”
M
e
di
c
al
D
e
v
ic
e
s
:
E
v
id
e
nc
e
and R
e
s
e
ar
c
h
, pp. 423
–
433, 2021
,
do
i:
10.2147%
2F
M
D
E
R
.S
339855
.
[
4]
Y
.
L
i,
H
.
Q
i,
J
.
D
a
i,
X
.
J
i,
a
nd
Y
.
W
e
i,
“
F
ul
ly
c
o
n
vo
lu
ti
o
na
l
in
s
ta
nc
e
-
a
w
a
r
e
s
e
ma
nt
i
c
s
e
gme
nt
a
ti
o
n,”
in
P
r
oc
e
e
di
ngs
o
f
th
e
I
E
E
E
c
onf
e
r
e
nc
e
on c
om
put
e
r
v
is
io
n and patt
e
r
n
r
e
c
ogni
ti
on
,
2017, p
p. 2359
–
2367
,
do
i:
10.48550/ar
X
i
v
.1611.07709
.
[
5]
K
.
H
e
,
G
.
G
ki
ox
a
r
i,
P
.
D
o
ll
a
´
r
,
a
nd
R
.
G
ir
s
hi
c
k,
“
M
a
s
k
r
-
c
nn,”
in
P
r
oc
.
I
E
E
E
I
nt
.
C
onf
.
C
om
put
.
V
is
,
2017,
pp.
2980
–
2988
,
do
i:
10.1109/I
C
C
V
.2017.322
.
[
6]
D
.
B
o
l
y
a
,
C
.
Z
ho
u,
F
.
X
ia
o
,
a
nd
Y
.
J
.
L
e
e
,
“
Y
o
la
c
t:
R
e
a
l
-
ti
me
in
s
ta
nc
e
s
e
gm
e
nt
a
ti
o
n,”
in
P
r
oc
e
e
di
ngs
o
f
th
e
I
E
E
E
/
C
V
F
in
te
r
nat
io
nal
c
onf
e
r
e
nc
e
on c
om
put
e
r
v
is
io
n
, 2019, pp. 9157
–
9
166
, do
i:
10.48550/ar
X
i
v
.1904.02689
.
[
7]
H
.
A
.
O
w
id
a
,
N
.
A
ls
hda
i
f
a
t,
A
.
A
lm
a
ght
ha
w
i,
S
.
A
bu
o
w
a
id
a
,
A
.
A
bur
o
m
ma
n,
A
.
A
l
-
M
o
ma
ni
,M
.
A
r
a
bi
a
t,
a
nd
H
.
Y
.
C
ha
n,
“
I
mpr
ov
e
d
d
e
e
p
l
e
a
r
ni
ng
a
r
c
hi
t
e
c
tu
r
e
f
or
s
ki
n
c
a
nc
e
r
c
la
s
s
i
f
i
c
a
ti
o
n,”
I
ndone
s
ia
n
J
our
nal
of
E
le
c
tr
ic
al
E
ngi
ne
e
r
in
g
and
C
om
put
e
r
Sc
ie
nc
e
, v
o
l.
3
6, n
o
. 1, p
p
. 501
–
508, 2024
,
d
o
i:
10.11591
/i
j
e
e
c
s
.v
36.i
1.pp501
-
508
.
[
8]
H
.
O
w
id
a
,
O
.
S
.
M
.
H
e
mi
e
d,
R
.
S
.
A
lk
ha
w
a
ld
e
h,
N
.
F
.
F
.
A
ls
hda
if
a
t,
A
nd
S
.
F
.
A
.
A
bu
o
w
a
id
a
,
“
I
mpr
ov
e
d
de
e
p
l
e
a
r
n
in
g
a
ppr
o
a
c
he
s
f
or
c
ov
id
-
19
r
e
c
o
gni
ti
o
n
in
c
t
im
a
g
e
s
,”
J
our
nal
of
T
he
or
e
t
ic
al
and
A
ppl
ie
d
I
nf
or
m
at
io
n
T
e
c
hnol
ogy
,
vo
l.
100,
n
o
.
13,
pp. 4925
–
4931, 2022.
[
9]
H
.
A
.
O
w
id
a
,
M
.
R
.
H
a
s
s
a
n,
A
.
M
.
A
li
,
F
.
A
ln
a
im
a
t,
A
.
A
l
S
ha
r
a
h,
S
.
A
buo
w
a
id
a
,
a
nd
N
.
A
ls
hda
if
a
t,
“
T
h
e
p
e
r
f
or
ma
nc
e
of
a
r
ti
f
ic
ia
l
in
t
e
ll
ig
e
nc
e
in
p
r
o
s
ta
te
ma
gn
e
ti
c
r
e
s
o
na
n
c
e
im
a
gi
n
g
s
c
r
e
e
ni
ng,”
I
n
te
r
nat
io
nal
J
our
nal
of
E
le
c
tr
ic
al
and
C
om
p
ut
e
r
E
ngi
ne
e
r
in
g
, vo
l.
14, n
o
. 2, p. 2234
–
2241, 2024
, d
o
i:
10.1159
1/
ij
e
c
e
.
v
14i
2.pp2234
-
2241
.
[
10]
H
.
A
.
O
w
id
a
,
G
.
A
lM
a
ha
di
n,
J
.
I
.
A
l
-
N
a
bul
s
i,
N
.
T
ur
a
b,
S
.
A
buo
w
a
id
a
,
a
nd
N
.
A
l
s
hda
if
a
t,
“
A
u
to
ma
te
d
c
la
s
s
if
i
c
a
ti
o
n
of
br
a
in
tu
mor
-
ba
s
e
d
ma
gne
ti
c
r
e
s
o
na
n
c
e
im
a
gi
ng
us
in
g
de
e
p
le
a
r
ni
n
g
a
ppr
o
a
c
h.”
I
nt
e
r
nat
io
nal
J
our
nal
o
f
E
le
c
t
r
ic
al
and
C
om
p
ut
e
r
E
ngi
ne
e
r
in
g
, vo
l.
14, n
o
. 3, 2024
, d
o
i:
10.11591/i
j
e
c
e
.
v
14i
3.pp3
150
-
3158
.
[
11]
A
.
Y
.
A
lh
us
e
na
t,
H
.
A
.
O
w
id
a
,
H
.
A
.
R
a
ba
ba
h,
J
.
I
.
A
l
-
N
a
bul
s
i,
a
nd
S
.
A
buo
w
a
id
a
,
“
A
s
e
c
ur
e
d
mul
t
i
-
s
ta
ge
s
a
ut
he
nt
ic
a
ti
o
n
pr
o
t
oc
o
l
f
o
r
i
o
t
d
e
v
i
c
e
s
,”
M
at
he
m
at
ic
al
M
ode
ll
in
g
of
E
ngi
ne
e
r
in
g
P
r
obl
e
m
s
,
vo
l.
10,
n
o
.
4,
p
p
.
1352
–
1358,
2023
,
do
i:
10.18280/m
me
p.100429
.
[
12]
L
.
D
e
ng,
D
.
Y
u
e
t
a
l.
,
“
D
e
e
p
le
a
r
n
in
g:
me
th
o
ds
a
nd
a
ppl
ic
a
ti
o
n
s
,”
F
oundati
ons
and
tr
e
nds
®
in
s
ig
nal
pr
oc
e
s
s
in
g
,
v
o
l.
7,
n
o
.
3
–
4,
pp. 197
–
387, 2014
, do
i:
10.1561/200000003
9
.
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:
2
5
0
2
-
4
7
52
A
c
c
ur
ate
s
e
gme
ntat
ion
of
f
r
uit
bas
e
d
on
de
e
p
lear
ning
(
E
s
r
aa
A
bu
E
ls
oud
)
1337
[
13]
W
.
S
hi
m
o
da
a
nd
K
.
Y
a
na
i,
“
C
nn
-
ba
s
e
d
f
oo
d
im
a
ge
s
e
gm
e
n
ta
ti
o
n
w
it
h
o
ut
pi
xe
l
-
w
is
e
a
nn
o
ta
ti
o
n,”
i
n
N
e
w
T
r
e
nds
in
I
m
age
A
nal
y
s
is
and
P
r
oc
e
s
s
in
g
–
I
C
I
A
P
2015
W
or
k
s
hops
:
I
C
I
A
P
2
015
I
nt
e
r
nat
io
nal
W
or
k
s
hops
,
B
io
F
or
,
C
T
M
R
,
R
H
E
U
M
A
,
I
SC
A
,
M
A
D
iM
a,
SB
M
I
, and
Q
oE
M
,
G
e
noa, I
ta
ly
,
S
e
pt
e
mbe
r
7
-
8,
201
5, P
r
oc
e
e
di
ngs
18.
S
pr
in
ge
r
, 2015,
pp. 449
–
457
,
d
o
i:
10.1007/9
78
-
3
-
319
-
2
3222
-
5 55
.
[
14]
K
.
J
.
P
f
is
t
e
r
e
r
e
t
al
.
,
“
A
ut
o
ma
te
d
f
oo
d
in
ta
k
e
tr
a
c
ki
ng
r
e
qui
r
e
s
de
pt
h
-
r
e
f
in
e
d
s
e
ma
nt
i
c
s
e
gm
e
nt
a
ti
o
n
t
o
r
e
c
ti
f
y
v
is
ua
l
-
vo
lu
me
di
s
c
or
da
nc
e
i
n l
o
ng
-
te
r
m
c
a
r
e
h
ome
s
,”
Sc
ie
nt
if
ic
R
e
por
ts
,
v
o
l.
1
2, no
. 83,
20
22,
d
o
i:
10.1038/s
41598
-
021
-
03972
-
8
.
[
15]
D
.
Y
a
ng,
Y
.
D
u,
H
.
Y
a
o
,
a
nd
L
.
B
a
o
,
“
I
ma
g
e
s
e
ma
nt
i
c
s
e
g
me
nt
a
ti
o
n
w
it
h
hi
e
r
a
r
c
hi
c
a
l
f
e
a
tu
r
e
f
us
i
o
n
ba
s
e
d
o
n
d
e
e
p
n
e
ur
a
l
ne
tw
o
r
k,”
C
onne
c
ti
on Sc
ie
nc
e
, v
o
l.
34, n
o
. 1, pp. 1772
–
1784, 2
022
,
do
i:
10.1080/09540091.2
022.2082384
.
[
16]
[
16]
V
.
H
e
ma
ma
li
ni
,
S
.
R
a
ja
r
a
je
s
w
a
r
i,
S
.
N
a
c
hi
y
a
ppa
n,
M
.
S
a
mba
th
,
T
.
D
e
v
i,
B
.
K
.
S
in
gh,
a
nd
A
.
R
a
ghuv
a
ns
hi
,
“
F
o
o
d
qua
li
t
y
in
s
pe
c
ti
o
n
a
nd
gr
a
di
ng
us
in
g
e
f
f
i
c
i
e
nt
im
a
ge
s
e
gm
e
nt
a
ti
o
n
a
n
d
ma
c
hi
n
e
l
e
a
r
ni
ng
-
ba
s
e
d
s
y
s
t
e
m,”
J
our
nal
of
F
ood
Q
ual
it
y
,
vo
l.
2022, no
. 1, p. 5262294, 2022
, d
o
i
:
10.1155/2022/
5262294
.
[
17]
E
.
A
gui
la
r
,
B
.
R
e
m
e
s
e
ir
o
,
M
.
B
o
la
n˜
o
s
,
a
nd
P
.
R
a
de
v
a
,
“
G
r
a
b,
pa
y
,
a
nd
e
a
t:
S
e
ma
nt
i
c
f
oo
d
d
e
t
e
c
t
i
o
n
f
o
r
s
ma
r
t
r
e
s
ta
ur
a
nt
s
,”
I
E
E
E
T
r
ans
ac
ti
ons
on M
ul
ti
m
e
di
a
, v
o
l.
20, n
o
. 12, pp. 3266
–
3275, 2
018
, do
i:
10.1109/
T
M
M
.2018.2831627
.
[
18]
T
.
L
e
,
“
M
a
s
k
r
-
c
nn
w
it
h
da
ta
a
ugme
nt
a
ti
o
n
f
or
f
oo
d
de
t
e
c
ti
o
n
a
nd
r
e
c
o
gni
ti
o
n,”
C
om
put
e
r
Sc
ie
nc
e
,
A
gr
ic
ul
tu
r
al
and
F
ood
Sc
ie
nc
e
s
, 202
0, do
i:
10.36227/t
e
c
hr
x
i
v
.11974362.
[
19]
Y
.
L
i,
X
.
X
u,
a
nd
C
.
Y
u
a
n,
“
E
nha
nc
e
d
ma
s
k
r
-
c
nn
f
o
r
c
hi
n
e
s
e
f
oo
d
im
a
ge
d
e
t
e
c
t
i
o
n,”
M
at
he
m
at
ic
al
P
r
obl
e
m
s
in
E
ngi
ne
e
r
i
ng
,
vo
l.
2020, n
o
. 1, p. 6253827, 2020
, d
o
i:
10.1155/2020
/6
253827
.
[
20]
R
.
D
.
Y
o
ga
s
w
a
r
a
,
E
.
M
.
Y
uni
a
r
n
o
,
a
nd
A
.
D
.
W
ib
a
w
a
,
“
I
ns
ta
nc
e
-
a
w
a
r
e
s
e
ma
nt
i
c
s
e
gm
e
nt
a
ti
o
n
f
or
f
ood
c
a
lo
r
i
e
e
s
ti
ma
ti
o
n
us
in
g
ma
s
k
r
-
c
nn,”
in
2019
I
nt
e
r
nat
io
nal
Se
m
in
ar
on
I
nt
e
ll
ig
e
nt
T
e
c
hnol
ogy
and
I
ts
A
ppl
ic
at
io
ns
(
I
SI
T
I
A
)
.
I
E
E
E
,
2019,
pp.
416
–
4
21
,
do
i:
10.1109/
I
S
I
T
I
A
.2019.8937129
.
[
21]
S
.
N
a
r
it
o
mi
a
nd
K
.
Y
a
na
i,
“
C
a
lo
r
ie
c
a
pt
o
r
gl
a
s
s
:
F
oo
d
c
a
lo
r
i
e
e
s
ti
ma
ti
o
n
ba
s
e
d
o
n
a
c
tu
a
l
s
iz
e
us
in
g
ho
l
o
l
e
ns
a
nd
de
e
p
l
e
a
r
ni
ng,
”
in
2020
I
E
E
E
C
onf
e
r
e
nc
e
on
V
ir
tu
al
R
e
al
it
y
and
3D
U
s
e
r
I
nt
e
r
fa
c
e
s
A
bs
t
r
ac
ts
and
W
or
k
s
hops
(
V
R
W
)
.
I
E
E
E
,
2020,
pp.
818
–
819
,
do
i:
10.1109/VR
W
50115.2020.00260
.
[
22]
C.
-
H
.
H
s
ia
,
T
.
-
H
.
W
. C
ha
ng, C
.
-
Y
. C
hi
a
ng, a
nd H
.
-
T
. C
ha
n, “
M
a
s
k r
-
c
nn w
it
h n
e
w
da
t
a
a
ugme
n
ta
ti
o
n
f
e
a
tu
r
e
s
f
o
r
s
ma
r
t
de
t
e
c
ti
o
n
of
r
e
ta
il
pr
o
du
c
ts
,”
A
ppl
ie
d Sc
ie
nc
e
s
, v
o
l.
12, n
o
. 6, p. 2902, 20
22
, do
i:
10.3390/a
pp12062902
.
[
23]
W
.
J
ia
,
J
.
W
e
i,
Q
.
Z
ha
ng,
N
.
P
a
n,
Y
.
N
iu
,
X
.
Y
in
,
Y
.
D
in
g,
a
n
d
X
.
G
e
,
“
A
c
c
u
r
a
te
s
e
gm
e
nt
a
ti
o
n
of
g
r
e
e
n
f
r
ui
t
ba
s
e
d
o
n
o
pt
im
iz
e
d
ma
s
k r
c
nn a
ppl
ic
a
ti
o
n i
n
c
o
mpl
e
x
o
r
c
ha
r
d,”
F
r
ont
ie
r
s
i
n
P
la
nt
Sc
ie
nc
e
, v
o
l.
13, p. 955256, 2022
, d
o
i:
10.3389
/
f
pl
s
.2022.9552
56
.
[
24]
S
.
P
.
M
o
ha
nt
y
,
G
.
S
in
gha
l,
E
.
A
.
S
c
u
c
c
im
a
r
r
a
,
D
.
K
e
ba
il
i,
H
.
H
e
´
r
it
i
e
r
,
V
.
B
o
ul
a
nge
r
,
a
nd
M
.
S
a
la
th
e
´
,
“
T
h
e
f
oo
d
r
e
c
o
gni
t
io
n
be
n
c
hma
r
k:
U
s
in
g
de
e
p
l
e
a
r
ni
ng
t
o
r
e
c
o
gni
z
e
f
oo
d
in
i
ma
ge
s
,”
F
r
ont
ie
r
s
in
N
ut
r
it
io
n
,
vo
l.
9,
p.
875143,
2022
,
do
i:
10.3389/f
nu
t.
2022.875143
.
[
25]
T.
-
Y
.
L
in
,
M
.
M
a
ir
e
,
S
.
B
e
l
o
ngi
e
,
J
.
H
a
y
s
,
P
.
P
e
r
o
na
,
D
.
R
a
ma
na
n,
P
.
D
ol
la
´
r
,
a
nd
C
.
L
.
Z
it
n
i
c
k,
“
M
ic
r
o
s
of
t
c
o
c
o
:
C
om
mo
n
o
bj
e
c
ts
in
c
o
nt
e
x
t,
”
in
C
om
put
e
r
V
is
io
n
–
E
C
C
V
2014:
13t
h
E
ur
ope
an
C
on
f
e
r
e
nc
e
,
Z
ur
i
c
h,
Sw
it
z
e
r
la
nd,
Se
pt
e
m
be
r
6
-
12,
2014,
P
r
oc
e
e
di
ngs
, P
ar
t
V
13. Spr
in
ge
r
, 2014, pp. 740
–
755
, doi
:
10.1
007/
978
-
3
-
319
-
10602
-
1 48
.
[
26]
S
.
F
.
A
.
A
buo
w
a
id
a
a
nd
H
.
Y
.
C
ha
n,
“
I
mpr
ove
d
de
e
p
le
a
r
ni
n
g
f
r
a
m
e
w
o
r
k
f
o
r
mul
ti
f
oo
d
in
s
ta
nc
e
s
e
gm
e
nt
a
ti
o
ns
,”
I
nt
e
r
nat
io
nal
J
our
nal
, v
ol
. 9, n
o
. 4, 2020
, d
o
i:
10.30534/i
ja
t
c
s
e
/2
020/
308942
020
.
[
27]
K
.
H
e
,
X
.
Z
ha
ng,
S
.
R
e
n,
a
nd
J
.
S
un,
“
D
e
e
p
r
e
s
id
ua
l
l
e
a
r
ni
ng
f
or
im
a
ge
r
e
c
o
gn
it
i
o
n,”
in
P
r
oc
e
e
di
ngs
of
th
e
I
E
E
E
c
onf
e
r
e
nc
e
on
c
om
put
e
r
v
is
io
n and patt
e
r
n r
e
c
ogni
ti
on
, 2016, pp. 770
–
778
, do
i:
10.48550/ar
X
i
v
.1512.03385
.
[
28]
Y
.
C
a
mgo¨
z
lu
¨
a
nd
Y
.
K
ut
lu
,
“
A
na
l
y
s
is
of
f
il
t
e
r
s
iz
e
e
f
f
e
c
t
in
de
e
p
l
e
a
r
ni
ng,”
J
our
nal
of
A
r
ti
f
ic
ia
l
I
nt
e
ll
ig
e
nc
e
w
it
h
A
ppl
ic
at
i
ons
,
vo
l.
1, n
o
. 1, pp. 20
-
29,
2020
, d
oi
:
10.48550/ar
X
i
v
.2101.01115
.
[
29]
W
.
S
.
A
hme
d
a
nd
A
.
a
.
A
.
K
a
r
im
,
“
T
h
e
im
pa
c
t
of
f
il
t
e
r
s
i
z
e
a
nd
numbe
r
of
f
i
lt
e
r
s
o
n
c
la
s
s
if
i
c
a
ti
o
n
a
c
c
u
r
a
c
y
in
c
nn,”
in
2
020
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on
C
om
pu
te
r
Sc
ie
n
c
e
and
Sof
tw
ar
e
E
ngi
ne
e
r
in
g
(
C
SA
SE
)
,
2020,
pp.
88
–
93
,
do
i
:
10.1109/C
S
A
S
E
48920.2020.9142089
.
B
I
OG
RA
P
HI
E
S
OF
AU
T
HO
RS
Es
ra
a
A
bu
El
s
o
u
d
recei
v
ed
t
h
e
B.
S
c.
d
eg
r
ee
i
n
E
l
ec
t
ri
c
al
E
n
g
i
n
ee
r
i
n
g
fro
m
T
h
e
H
as
h
emi
t
e
U
n
i
v
e
rs
i
t
y
,
J
o
r
d
an
,
i
n
2
0
1
3
an
d
M.
S
c.
i
n
C
y
b
e
r
S
ec
u
r
i
t
y
fr
o
m
T
h
e
H
as
h
em
i
t
e
U
n
i
v
e
rs
i
t
y
i
n
2
0
2
3
.
E
n
g
.
E
s
raa’s
c
u
rr
en
t
re
s
e
ar
ch
i
n
t
ere
s
t
s
i
n
cl
u
d
e
cy
b
e
r
s
ec
u
r
i
t
y
,
m
a
c
h
i
n
e
l
e
ar
n
i
n
g
,
b
i
g
d
at
a
an
d
mo
b
i
l
e
n
e
t
w
o
rk
.
Sh
e
i
s
c
u
rr
en
t
l
y
a
L
ec
t
u
r
e
r
i
n
Z
arq
a
U
n
i
v
e
rs
i
t
y
,
Z
arq
a
,
J
o
rd
an
.
S
h
e
c
a
n
b
e
c
o
n
t
ac
t
e
d
at
em
ai
l
:
e
ab
u
e
l
s
o
u
d
@
zu
.
ed
u
.
j
o
.
Om
a
r
A
l
i
d
m
a
t
rece
i
v
e
d
h
i
s
B.
Sc.
d
eg
r
ee
i
n
Co
m
p
u
t
e
r
I
n
f
o
r
m
at
i
o
n
S
y
s
t
ems
,
M.
Sc.
d
eg
r
ee
i
n
C
o
m
p
u
t
e
r
S
c
i
en
ce
fro
m
A
l
al
-
Ba
y
t
U
n
i
v
e
rs
i
t
y
,
J
o
r
d
an
,
an
d
t
h
e
P
h
.
D
.
d
eg
r
ee
i
n
Co
m
p
u
t
e
r
S
c
i
en
c
e
fr
o
m
U
n
i
v
e
rs
i
t
i
Sai
n
s
Mal
a
y
s
i
a,
Mal
ay
s
i
a
.
C
u
rr
e
n
t
l
y
,
D
r.
A
l
i
d
m
at
s
e
r
v
e
s
as
an
A
s
s
i
s
t
an
t
Pro
f
e
s
s
o
r
i
n
t
h
e
Co
m
p
u
t
e
r
S
ci
e
n
ce
D
e
p
art
me
n
t
at
Z
arq
a
U
n
i
v
e
rs
i
t
y
,
J
o
rd
a
n
.
H
i
s
Ph
.
D
.
r
e
s
e
ar
c
h
fo
cu
s
ed
o
n
A
rt
i
fi
ci
al
I
n
t
e
l
l
i
g
e
n
ce,
s
p
eci
a
l
i
zi
n
g
i
n
m
o
d
e
l
l
i
n
g
c
r
o
w
d
mo
v
emen
t
s
i
n
p
an
i
c
s
i
t
u
at
i
o
n
s
,
s
u
c
h
as
fi
r
e
s
.
H
e
c
an
b
e
co
n
t
ac
t
e
d
at
em
ai
l
:
o
m
ar.
a
l
i
d
m
at
@
g
m
a
i
l
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2
5
0
2
-
4
7
52
In
do
n
e
s
i
a
n
J
E
l
e
c
E
n
g
&
C
o
m
p
S
c
i
,
Vo
l
.
3
8
,
N
o
.
2
,
M
a
y
20
2
5
:
1
331
-
1
338
1338
Suh
a
i
l
a
A
b
u
o
w
a
i
da
rece
i
v
e
d
t
h
e
B.
S
c.
d
eg
r
ee
s
i
n
co
m
p
u
t
e
r
i
n
f
o
r
m
at
i
o
n
s
y
s
t
em,
M.
Sc.
d
e
g
r
ee
s
i
n
co
m
p
u
t
e
r
s
c
i
en
ce
fro
m
A
L
al
Ba
y
t
U
n
i
v
e
rs
i
t
y
,
J
o
r
d
an
,
i
n
2
0
1
2
an
d
2
0
1
5
,
re
s
p
ec
t
i
v
el
y
,
a
n
d
t
h
e
P
h
.
D
.
d
e
g
r
ee
i
n
Co
m
p
u
t
e
r
S
ci
en
ce
fro
m
U
n
i
v
e
rs
i
t
i
Sai
n
s
Mal
a
y
s
i
a,
Mal
a
y
s
i
a,
i
n
2
0
2
3
.
H
e
r
re
s
e
arch
i
n
t
e
r
e
s
t
s
i
n
-
cl
u
d
e
d
ee
p
l
e
ar
n
i
n
g
,
d
e
p
t
h
e
s
t
i
m
at
i
o
n
,
p
o
i
n
t
cl
o
u
d
an
d
co
m
p
u
t
e
r
v
i
s
i
o
n
.
Sh
e
c
an
b
e
co
n
t
a
c
t
e
d
at
em
a
i
l
:
s
u
h
i
l
ao
w
i
d
a
@
g
m
a
i
l
.
c
o
m
.
Es
ra
a
A
l
h
e
n
a
w
i
i
s
a
d
e
d
i
c
at
e
d
a
c
ad
emi
c
w
i
t
h
a
d
i
s
t
i
n
g
u
i
s
h
e
d
b
ack
g
r
o
u
n
d
i
n
c
o
m
p
u
t
e
r
s
ci
e
n
ce
.
Sh
e
e
arn
ed
h
e
r
Ph
.
D
.
i
n
co
m
p
u
t
e
r
s
c
i
en
ce
w
i
t
h
h
o
n
o
rs
i
n
2
0
2
2
,
fo
l
l
o
w
i
n
g
h
e
r
m
as
t
e
r's
d
eg
r
ee
i
n
t
h
e
s
a
me
fi
e
l
d
i
n
2
0
1
5
.
Sh
e
al
s
o
h
o
l
d
s
a
b
ac
h
el
o
r’s
d
e
g
r
ee
i
n
co
m
p
u
t
er
e
n
g
i
n
ee
ri
n
g
fr
o
m
2
0
1
0
,
cu
rren
t
l
y
s
e
rv
i
n
g
as
a
n
as
s
i
s
t
a
n
t
p
ro
f
e
s
s
o
r
at
Z
arq
a
U
n
i
v
e
rs
i
t
y
s
i
n
ce
Se
p
t
em
b
e
r
2
0
2
3
.
Sh
e
c
an
b
e
co
n
t
a
c
t
ed
at
em
ai
l
:
e
a
l
h
en
a
w
i
@
z
u
.
ed
u
.
j
o
.
N
a
w
a
f
A
l
s
h
da
i
f
a
t
recei
v
ed
t
h
e
B
.
S
c
d
e
g
r
ee
s
i
n
co
m
p
u
t
e
r
s
ci
e
n
ce
fro
m
A
L
al
-
Bay
t
U
n
i
v
e
rs
i
t
y
an
d
M.
S
c.
d
eg
r
ee
s
i
n
co
m
p
u
t
e
r
s
c
i
en
ce
fro
m
t
h
e
U
n
i
v
e
rs
i
t
y
o
f
J
o
r
d
an
,
J
o
r
d
an
,
i
n
2
0
0
2
an
d
2
0
1
1
,
re
s
p
e
c
t
i
v
e
l
y
,
an
d
t
h
e
P
h
.
D
.
d
e
g
r
ee
i
n
Co
m
p
u
t
e
r
S
c
i
e
n
ce
fr
o
m
U
n
i
v
e
rs
i
t
i
Sai
n
s
Mal
a
y
s
i
a,
Mal
a
y
s
i
a
,
i
n
2
0
2
3
.
H
i
s
re
s
e
arch
i
n
t
e
r
es
t
s
i
n
c
l
u
d
e
d
ee
p
l
e
arn
i
n
g
an
d
m
a
ch
i
n
e
l
e
ar
n
i
n
g
.
H
e
c
an
b
e
co
n
t
ac
t
e
d
at
em
a
i
l
:
n
_
al
s
h
d
a
i
fat
@
as
u
.
e
d
u
.
j
o
.
A
hm
a
d
A
b
u
ro
m
m
a
n
i
s
cu
rr
e
n
t
l
y
a
F
u
l
l
Pr
o
fe
s
s
o
r
o
f
A
rt
i
fi
ci
al
I
n
t
el
l
i
g
en
ce
i
n
t
h
e
D
e
p
art
men
t
o
f
D
at
a
S
ci
e
n
ce
an
d
A
rt
i
fi
c
i
a
l
In
t
el
l
i
g
en
ce
i
n
t
h
e
Z
arq
a
U
n
i
v
e
rs
i
t
y
.
H
e
r
ecei
v
ed
t
h
e
B
.
S
c
.
d
e
g
r
ee
i
n
In
fo
r
m
at
i
o
n
T
ech
n
o
l
o
g
y
fr
o
m
A
l
-
Bal
q
a
A
p
p
l
i
e
d
U
n
i
v
e
rs
i
t
y
(B
A
U
)
,
J
o
rd
an
,
i
n
2
0
0
5
,
an
d
t
h
e
M
.
S
c
.
d
e
g
r
ee
i
n
i
n
f
o
r
m
at
i
o
n
t
e
ch
n
o
l
o
g
y
(e
-
Se
r
v
i
ce
s
)
fr
o
m
G
ri
ffi
t
h
U
n
i
v
e
rs
i
t
y
,
A
u
s
t
ral
i
a,
i
n
2
0
0
9
.
H
e
g
o
t
t
h
e
P
h
.
D
.
d
e
g
r
ee
i
n
t
i
me
p
r
e
d
i
c
t
i
o
n
i
n
ru
n
n
i
n
g
b
u
s
i
n
e
s
s
p
ro
ce
s
s
e
s
w
i
t
h
t
h
e
R
e
s
e
ar
ch
C
en
t
e
r
i
n
In
t
el
l
i
g
en
t
T
ech
n
o
l
o
g
i
e
s
,
U
n
i
v
e
rs
i
t
y
o
f
Sa
n
t
i
ag
o
d
e
Co
m
p
o
s
t
e
l
a
(C
i
T
IU
S).
H
e
c
an
b
e
co
n
t
a
c
t
ed
at
em
ai
l
:
aa
b
u
ro
mm
a
n
@
z
u
.
ed
u
.
j
o
.
Hu
a
h
Y
o
n
g
C
h
a
n
i
s
a
s
e
n
i
o
r
l
ec
t
u
r
e
r
i
n
t
h
e
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
S
c
i
en
c
e
s
,
U
n
i
v
e
rs
i
t
i
Sa
i
n
s
Ma
l
a
y
s
i
a
(
U
SM).
He
is
a
c
t
i
v
el
y
i
n
v
o
l
v
e
d
in
g
ri
d
c
o
m
p
u
t
i
n
g
r
e
s
e
ar
ch
ac
t
i
v
i
t
i
e
s
,
b
o
t
h
at
t
h
e
n
at
i
o
n
a
l
an
d
i
n
t
e
rn
at
i
o
n
a
l
l
ev
el
.
H
e
i
s
al
s
o
o
n
e
o
f
t
h
e
p
ro
j
ec
t
l
e
ad
e
rs
of
t
h
e
e
-
S
ci
e
n
ce
G
ri
d
p
i
o
n
ee
r
p
ro
j
ec
t
in
Ma
l
a
y
s
i
a
,
f
u
n
d
e
d
by
t
h
e
M
i
n
i
s
t
ry
of
S
c
i
en
ce,
T
e
ch
n
o
l
o
g
y
,
a
n
d
In
n
o
v
at
i
o
n
.
H
i
s
r
e
s
e
ar
ch
s
p
an
s
fro
m
g
r
i
d
c
o
m
p
u
t
i
n
g
t
o
m
o
r
e
s
p
ec
i
f
i
c
i
s
s
u
e
s
s
u
c
h
as
r
e
s
o
u
r
ce
al
l
o
c
at
i
o
n
,
l
o
ad
b
al
an
ci
n
g
,
s
o
ft
w
ar
e
a
g
en
t
s
,
an
d
mi
d
d
l
ew
are
e
n
g
i
n
ee
ri
n
g
.
He
c
an
be
c
o
n
t
ac
t
e
d
at
em
ai
l
:
h
y
ch
a
n
@
u
s
m
.
my
.
Evaluation Warning : The document was created with Spire.PDF for Python.