I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
rm
a
t
ics a
nd
Co
m
m
un
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
15
,
No
.
2
,
J
u
n
e
20
26
,
p
p
.
588
~
5
9
6
I
SS
N:
2252
-
8
7
7
6
,
DOI
:
1
0
.
1
1
5
9
1
/iji
ct
.
v15
i
2
.
pp
588
-
5
9
6
588
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ict.
ia
esco
r
e.
co
m
Co
a
stline seg
men
tatio
n on La
ndsa
t
8 O
LI
ima
g
es usi
ng
ma
jority v
o
ting w
ith
dee
p learning
mo
dels
Nur
Na
f
iiy
a
h
1
,
Sa
lwa
Na
bil
a
h
2
,
Nur
Aziz
a
h Af
f
a
nd
y
2
,
Rif
k
y
Ais
y
a
t
ul F
a
r
o
h
3
,
E
s
a
P
ra
k
a
s
a
4
1
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
c
s
,
U
n
i
v
e
r
si
t
a
s I
sl
a
m
L
a
mo
n
g
a
n
,
La
m
o
n
g
a
n
,
I
n
d
o
n
e
s
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
i
v
i
l
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
a
s I
sl
a
m
La
m
o
n
g
a
n
,
L
a
mo
n
g
a
n
,
I
n
d
o
n
e
si
a
3
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
a
s I
sl
a
m
La
m
o
n
g
a
n
,
L
a
m
o
n
g
a
n
,
I
n
d
o
n
e
si
a
4
R
e
s
e
a
r
c
h
C
e
n
t
e
r
f
o
r
D
a
t
a
S
c
i
e
n
c
e
a
n
d
I
n
f
o
r
ma
t
i
o
n
,
N
a
t
i
o
n
a
l
R
e
se
a
r
c
h
a
n
d
I
n
n
o
v
a
t
i
o
n
A
g
e
n
c
y
(
B
R
I
N
)
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
s
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
23
,
2
0
2
5
R
ev
is
ed
Mar
31
,
2
0
2
6
Acc
ep
ted
Ap
r
15
,
2
0
2
6
Co
a
stli
n
e
s
a
re
h
ig
h
ly
d
y
n
a
m
ic
d
u
e
to
b
o
t
h
n
a
tu
ra
l
p
ro
c
e
ss
e
s
a
n
d
a
n
th
ro
p
o
g
e
n
ic
fa
c
to
rs
,
i
n
c
lu
d
in
g
g
lo
b
a
l
wa
rm
in
g
a
n
d
se
a
lev
e
l
rise
.
Ac
c
u
ra
te
c
o
a
stli
n
e
se
g
m
e
n
tatio
n
is
e
ss
e
n
ti
a
l
fo
r
e
ffe
c
ti
v
e
m
o
n
i
to
r
in
g
a
n
d
m
a
n
a
g
e
m
e
n
t.
Alth
o
u
g
h
p
re
v
io
u
s
stu
d
ies
h
a
v
e
a
p
p
li
e
d
d
e
e
p
le
a
rn
in
g
f
o
r
c
o
a
stli
n
e
d
e
tec
ti
o
n
,
m
a
n
y
e
x
isti
n
g
m
o
d
e
ls
stil
l
su
ffe
r
fr
o
m
in
sta
b
i
li
ty
a
c
ro
ss
sc
e
n
e
s,
b
lu
rre
d
b
o
u
n
d
a
ries
,
a
n
d
s
e
g
m
e
n
tatio
n
a
rti
fa
c
ts,
in
d
ica
ti
n
g
t
h
a
t
m
o
d
e
l
g
e
n
e
ra
li
z
a
ti
o
n
re
m
a
in
s a
c
h
a
ll
e
n
g
e
.
Th
is st
u
d
y
a
ims
to
d
e
v
e
lo
p
a
m
o
re
ro
b
u
st
c
o
a
stli
n
e
se
g
m
e
n
tati
o
n
a
p
p
ro
a
c
h
b
y
i
n
tro
d
u
c
i
n
g
a
n
a
u
t
o
m
a
ted
m
a
jo
rit
y
v
o
ti
n
g
stra
teg
y
th
a
t
in
te
g
ra
tes
th
re
e
d
e
e
p
lea
rn
in
g
m
o
d
e
ls:
Re
sN
e
t5
0
,
Re
sN
e
t1
8
,
a
n
d
M
o
b
il
e
Ne
t
-
V2
.
L
a
n
d
sa
t
8
OLI
ima
g
e
ry
is
u
se
d
f
o
r
train
in
g
a
n
d
tes
ti
n
g
.
Th
e
Ja
c
c
a
rd
in
d
e
x
r
e
su
lt
s
sh
o
w
th
a
t
Re
sN
e
t1
8
,
Re
sN
e
t5
0
,
a
n
d
M
o
b
i
leN
e
t
-
V2
a
c
h
iev
e
d
sc
o
re
s
o
f
0
.
9
6
,
0
.
9
8
,
a
n
d
0
.
9
5
re
sp
e
c
ti
v
e
ly
,
wh
il
e
th
e
p
r
o
p
o
se
d
m
a
jo
rit
y
v
o
ti
n
g
m
e
th
o
d
a
lso
a
c
h
iev
e
d
0
.
9
8
.
De
sp
it
e
th
e
p
ro
d
u
c
in
g
a
sim
il
a
r
n
u
m
e
rica
l
sc
o
re
to
th
e
b
e
st
in
d
iv
i
d
u
a
l
m
o
d
e
l
(Re
sN
e
t5
0
),
t
h
e
e
n
se
m
b
le
m
e
th
o
d
imp
ro
v
e
s
se
g
m
e
n
tati
o
n
c
o
n
si
ste
n
c
y
b
y
re
d
u
c
in
g
a
rti
fa
c
ts
su
c
h
a
s
u
n
wa
n
t
e
d
p
e
rip
h
e
ra
l
sh
a
p
e
s
a
n
d
c
ra
c
k
s with
in
lan
d
a
re
a
s.
Th
e
se
fin
d
in
g
s
d
e
m
o
n
str
a
te
th
a
t
c
o
m
b
in
i
n
g
m
u
l
ti
p
le
se
g
m
e
n
tatio
n
o
u
t
p
u
ts
y
ield
s
m
o
re
sta
b
le
a
n
d
re
li
a
b
le
c
o
a
stli
n
e
d
e
tec
ti
o
n
th
a
n
u
si
n
g
sin
g
le
m
o
d
e
ls.
F
u
tu
re
wo
r
k
will
a
p
p
ly
t
h
is
a
p
p
r
o
a
c
h
to
b
ro
a
d
e
r
In
d
o
n
e
si
a
n
c
o
a
sta
l
re
g
io
n
s
t
o
fu
rt
h
e
r
a
ss
e
ss
it
s
g
e
n
e
ra
li
z
a
b
il
it
y
a
c
ro
ss
d
iv
e
rse
sh
o
re
li
n
e
c
o
n
d
i
ti
o
n
s.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
L
an
d
s
at
8
OL
I
Ma
jo
r
ity
v
o
tin
g
Seg
m
en
tatio
n
au
to
m
atio
n
Seg
m
en
tatio
n
co
astl
in
e
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Nu
r
Naf
iiy
ah
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atics,
U
n
iv
er
s
itas
I
s
lam
L
am
o
n
g
an
Vete
r
an
5
3
A
Stre
et
,
L
am
o
n
g
a
n
,
I
n
d
o
n
esia
E
m
ail:
m
y
n
af
f
@
u
n
is
la.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
C
o
astal
ar
ea
s
ar
e
co
n
tin
u
o
u
s
ly
ch
an
g
in
g
d
u
e
to
ab
r
asio
n
,
ac
c
r
etio
n
,
s
ea
lev
el
r
is
e,
an
d
a
n
th
r
o
p
o
g
en
ic
p
r
ess
u
r
es
s
u
ch
as
g
lo
b
al
war
m
in
g
an
d
co
astal
d
ev
elo
p
m
en
t
[
1
]
,
[
2
]
.
Mo
n
ito
r
in
g
th
ese
ch
an
g
es
is
cr
u
cial
f
o
r
co
n
s
e
r
v
atio
n
,
p
la
n
n
in
g
,
an
d
s
u
s
tain
ab
le
m
an
ag
em
en
t.
R
em
o
te
s
en
s
in
g
p
r
o
v
id
es
an
ef
f
icien
t
ap
p
r
o
ac
h
to
d
etec
t c
o
astl
in
e
d
y
n
am
ics with
o
u
t d
ir
ec
t
f
ield
s
u
r
v
e
y
s
.
L
an
d
s
at
8
OL
I
im
ag
e
r
y
h
as
b
ec
o
m
e
wid
ely
u
s
ed
f
o
r
co
astl
in
e
s
tu
d
ies,
s
u
p
p
o
r
tin
g
r
e
s
ea
r
ch
o
n
au
to
m
atic
lan
d
-
s
ea
s
eg
m
en
tati
o
n
[
3
]
,
[
4
]
,
m
ac
h
i
n
e
lear
n
in
g
ap
p
r
o
ac
h
es
[
3
]
,
[
5
]
,
c
o
astl
in
e
ex
tr
a
ctio
n
[
6
]
–
[
9
]
,
an
d
p
r
ac
tical
ap
p
licatio
n
s
s
u
ch
as n
av
ig
atio
n
an
d
lan
d
i
n
g
[
1
0
]
–
[
1
2
]
.
Ho
wev
e
r
,
th
ese
m
eth
o
d
s
co
n
tin
u
e
to
f
ac
e
d
if
f
icu
lties
in
d
is
tin
g
u
is
h
in
g
l
an
d
f
r
o
m
wate
r
in
s
ce
n
es
with
s
tr
o
n
g
s
p
ec
tr
al
s
im
ilar
ity
,
i
m
ag
e
n
o
is
e,
an
d
lo
w
co
n
tr
ast.
T
h
ese
lim
itatio
n
s
o
f
t
en
r
ed
u
ce
th
e
p
r
ec
is
io
n
an
d
s
t
ab
ilit
y
o
f
r
esu
ltin
g
c
o
astl
in
e
b
o
u
n
d
a
r
ies.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
C
o
a
s
tlin
e
s
eg
men
ta
tio
n
o
n
La
n
d
s
a
t 8
OLI
ima
g
es u
s
in
g
ma
j
o
r
ity
vo
tin
g
w
ith
d
ee
p
… (
N
u
r
N
a
fiiya
h
)
589
Dee
p
lear
n
in
g
m
eth
o
d
s
,
p
ar
ticu
lar
ly
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs),
h
av
e
d
em
o
n
s
tr
ated
s
tr
o
n
g
p
er
f
o
r
m
an
ce
in
im
a
g
e
class
if
icat
io
n
an
d
s
eg
m
en
t
atio
n
task
s
[
1
]
.
Sev
er
al
ar
ch
itectu
r
es,
in
clu
d
in
g
en
co
d
er
-
d
ec
o
d
er
m
o
d
els
[
1
3
]
–
[
1
5
]
,
U
-
Net
[
1
6
]
–
[
1
8
]
,
b
i
-
L
ST
M
[
1
9
]
,
e
n
s
em
b
le
lear
n
in
g
[
2
0
]
,
an
d
M
o
b
ileNet
-
V2
[
2
1
]
,
h
av
e
b
ee
n
a
p
p
lied
to
co
astl
in
e
s
eg
m
en
tatio
n
.
Alth
o
u
g
h
t
h
ese
m
o
d
els
im
p
r
o
v
e
f
ea
tu
r
e
ex
tr
ac
tio
n
o
n
lo
w
-
r
eso
lu
tio
n
im
a
g
er
y
[
2
2
]
–
[
2
4
]
,
in
d
iv
id
u
al
m
o
d
els
s
till
s
tr
u
g
g
le
to
p
r
o
d
u
ce
r
o
b
u
s
t
an
d
ar
tifa
ct
-
f
r
ee
r
esu
lts
ac
r
o
s
s
d
iv
er
s
e
co
astal c
o
n
d
itio
n
s
.
E
n
s
em
b
le
ap
p
r
o
ac
h
es
h
av
e
b
e
en
ex
p
lo
r
ed
in
p
r
ev
i
o
u
s
s
tu
d
ie
s
an
d
h
av
e
s
h
o
w
p
o
ten
tial
f
o
r
im
p
r
o
v
in
g
s
eg
m
en
tatio
n
co
n
s
is
ten
cy
[
5
]
,
[
7
]
.
I
n
v
ar
io
u
s
d
o
m
ain
s
s
u
ch
a
s
m
ed
ical
im
ag
in
g
,
r
em
o
te
s
en
s
in
g
class
if
icatio
n
,
an
d
r
o
ad
e
x
tr
ac
tio
n
,
en
s
em
b
le
d
ee
p
lear
n
i
n
g
th
r
o
u
g
h
tech
n
i
q
u
es
lik
e
m
ajo
r
ity
v
o
tin
g
o
r
f
ea
tu
r
e
lev
el
f
u
s
io
n
h
as
b
ee
n
d
em
o
n
s
tr
ated
t
o
e
n
h
an
ce
b
o
u
n
d
ar
y
p
r
ec
is
io
n
an
d
r
ed
u
ce
n
o
is
e
r
elate
d
ar
tifa
cts,
p
ar
ticu
lar
ly
in
ca
s
es
with
am
b
ig
u
o
u
s
o
r
lo
w
co
n
tr
a
s
t
f
ea
tu
r
es.
Ho
wev
er
,
ex
is
tin
g
en
s
em
b
le
s
tu
d
ies
h
av
e
n
o
t
s
p
ec
if
ically
ev
alu
ated
th
e
in
teg
r
atio
n
o
f
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
f
o
r
c
o
astl
in
e
s
eg
m
en
tatio
n
,
d
esp
ite
th
eir
ef
f
ec
tiv
en
ess
in
o
th
er
f
ield
s
[
2
5
]
.
Mo
r
eo
v
e
r
,
n
o
p
r
io
r
r
es
ea
r
ch
h
as
s
y
s
tem
atica
lly
t
e
s
t
ed
m
ajo
r
ity
v
o
tin
g
en
s
em
b
les
f
o
r
co
astl
in
e
s
eg
m
en
tatio
n
u
s
in
g
L
a
n
d
s
at
8
OL
I
,
leav
in
g
a
g
ap
in
u
n
d
er
s
tan
d
in
g
wh
eth
er
co
m
b
in
in
g
m
u
ltip
le
m
o
d
els
c
an
m
itig
ate
b
o
u
n
d
ar
y
n
o
is
e
a
n
d
im
p
r
o
v
e
r
o
b
u
s
tn
ess
ac
r
o
s
s
d
iv
er
s
e
s
h
o
r
elin
e
co
n
d
itio
n
s
.
T
o
ad
d
r
ess
th
is
g
ap
,
th
is
s
tu
d
y
p
r
o
p
o
s
es
an
au
to
m
ated
c
o
a
s
tlin
e
s
eg
m
en
tatio
n
m
eth
o
d
b
ase
d
o
n
a
m
ajo
r
ity
v
o
tin
g
s
tr
ateg
y
th
at
in
teg
r
ates
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
.
T
h
e
aim
is
to
e
n
h
an
ce
s
eg
m
en
tatio
n
r
eliab
ilit
y
b
y
r
e
d
u
cin
g
ar
tifa
cts
an
d
p
r
o
d
u
ci
n
g
m
o
r
e
co
n
s
is
ten
t
co
astl
in
e
b
o
u
n
d
a
r
ies
th
an
t
h
o
s
e
g
en
er
ated
b
y
in
d
iv
id
u
al
d
ee
p
l
ea
r
n
in
g
m
o
d
els.
T
h
is
ap
p
r
o
ac
h
is
ex
p
ec
ted
to
o
f
f
er
a
m
o
r
e
s
tab
le
alte
r
n
ativ
e
f
o
r
o
p
er
atio
n
al
c
o
astl
in
e
m
o
n
ito
r
i
n
g
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
Da
t
a
s
et
T
h
is
s
tu
d
y
u
s
es
a
p
u
b
licly
av
ailab
le
d
ataset
f
r
o
m
r
esear
ch
[
7
]
,
ac
ce
s
s
ib
le
at
h
ttp
://www.
r
em
o
tesen
s
in
g
lab
.
y
ild
iz.
ed
u
.
tr
.
Fro
m
th
is
d
ataset,
8
2
4
im
a
g
es
wer
e
s
elec
ted
f
o
r
tr
ain
in
g
an
d
2
0
im
ag
es
f
o
r
test
in
g
.
E
ac
h
im
ag
e
h
as
a
r
eso
lu
tio
n
o
f
2
2
4
×
2
2
4
p
ix
els,
with
wh
ite
p
ix
els
r
ep
r
esen
tin
g
s
ea
wate
r
an
d
b
lack
p
ix
els
r
ep
r
esen
tin
g
lan
d
.
T
h
e
d
ataset
is
d
er
iv
ed
f
r
o
m
L
an
d
s
at
8
OL
I
im
ag
er
y
,
wh
ich
p
r
o
v
id
es
m
u
ltis
p
ec
tr
al,
atm
o
s
p
h
er
ically
co
r
r
ec
ted
,
an
d
tem
p
o
r
ally
co
n
s
is
ten
t
d
ata
s
u
itab
le
f
o
r
co
astl
in
e
ex
tr
ac
tio
n
.
T
h
e
u
s
e
o
f
L
a
n
d
s
at
im
ag
e
r
y
is
p
ar
ticu
lar
ly
s
ig
n
if
ican
t
b
ec
au
s
e
its
g
lo
b
al
c
o
v
er
ag
e
an
d
lo
n
g
-
ter
m
co
n
tin
u
ity
s
u
p
p
o
r
t
r
ep
r
o
d
u
ci
b
le
an
d
s
ca
l
ab
le
co
astl
in
e
m
o
n
ito
r
i
n
g
.
B
ef
o
r
e
m
o
d
el
tr
ain
in
g
,
s
ev
e
r
al
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
wer
e
ap
p
lied
,
in
clu
d
in
g
r
esizin
g
all
im
ag
es
to
a
u
n
if
o
r
m
r
eso
lu
tio
n
an
d
en
s
u
r
in
g
c
o
n
s
is
ten
t
alig
n
m
en
t
b
etwe
en
in
p
u
t
im
ag
es
an
d
g
r
o
u
n
d
-
tr
u
th
m
ask
s
.
No
ad
d
itio
n
al
en
h
an
ce
m
e
n
t
tech
n
iq
u
es,
s
u
ch
as
h
is
to
g
r
am
eq
u
aliza
tio
n
o
r
n
o
is
e
f
ilter
i
n
g
,
wer
e
u
s
ed
to
p
r
eser
v
e
th
e
o
r
ig
in
al
s
p
ec
tr
al
c
h
ar
ac
t
er
is
tics
o
f
co
astal
b
o
u
n
d
ar
ies.
2
.
2
.
P
r
o
po
s
ed
m
et
ho
d
T
h
e
s
elec
tio
n
o
f
R
esNet5
0
,
R
esNet1
8
,
an
d
M
o
b
ileNet
-
V2
was
b
ased
o
n
th
eir
b
alan
ce
o
f
ac
cu
r
ac
y
,
co
m
p
u
tatio
n
al
e
f
f
icien
cy
,
an
d
p
r
o
v
e
n
ef
f
ec
tiv
en
ess
in
tr
an
s
f
er
lear
n
in
g
ap
p
licatio
n
s
.
R
e
s
Net
ar
ch
itectu
r
es
wer
e
ch
o
s
en
b
ec
au
s
e
th
eir
r
esid
u
al
co
n
n
ec
tio
n
s
h
elp
m
it
ig
ate
th
e
v
an
is
h
in
g
g
r
ad
ien
t
p
r
o
b
lem
an
d
en
ab
l
e
d
ee
p
er
f
ea
tu
r
e
ex
tr
ac
tio
n
.
M
o
b
ileNet
-
V2
was
in
clu
d
e
d
as
a
lig
h
tweig
h
t
alter
n
ativ
e
d
e
s
ig
n
ed
f
o
r
ef
f
icien
t
co
m
p
u
tatio
n
.
Alth
o
u
g
h
o
th
e
r
tr
an
s
f
er
lear
n
i
n
g
C
NNs
s
u
ch
as
VGG
an
d
Den
s
eNe
t
a
r
e
a
v
ailab
le,
th
is
s
tu
d
y
f
o
cu
s
es
o
n
ev
alu
atin
g
r
ep
r
esen
tativ
e
h
ea
v
y
,
m
ed
iu
m
,
an
d
li
g
h
tweig
h
t
ar
ch
itectu
r
es
to
a
n
a
ly
ze
th
eir
b
eh
a
v
io
r
with
in
an
en
s
em
b
le.
Fu
tu
r
e
s
tu
d
ies m
ay
in
co
r
p
o
r
ate
ad
d
itio
n
al
T
L
-
C
NNs to
ex
p
an
d
t
h
e
c
o
m
p
ar
ativ
e
a
n
aly
s
is
.
R
esNet5
0
is
a
5
0
-
lay
er
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
p
r
etr
ain
ed
o
n
I
m
a
g
eNe
t
an
d
c
ap
ab
le
o
f
class
if
y
in
g
im
ag
es
in
to
1
0
0
0
o
b
ject
ca
teg
o
r
ies,
allo
win
g
it
to
lea
r
n
r
ic
h
an
d
t
r
an
s
f
er
ab
le
f
ea
t
u
r
e
r
ep
r
esen
tatio
n
s
[
1
3
]
.
R
esNet1
8
is
a
s
m
aller
1
8
-
lay
er
v
er
s
io
n
th
at
r
etain
s
th
e
r
esid
u
al
lear
n
in
g
m
ec
h
an
is
m
b
u
t
with
f
ewe
r
p
ar
am
eter
s
,
also
b
en
ef
itin
g
f
r
o
m
I
m
ag
eNe
t
p
r
et
r
ain
in
g
to
s
u
p
p
o
r
t
r
o
b
u
s
t
f
ea
tu
r
e
ex
tr
ac
tio
n
[
1
3
]
.
Mo
b
ileNet
-
V2
is
a
5
3
-
lay
er
lig
h
tweig
h
t
C
NN
b
ased
o
n
in
v
er
ted
r
esid
u
al
b
lo
ck
s
an
d
d
ep
th
wis
e
s
ep
ar
ab
le
co
n
v
o
l
u
tio
n
s
,
an
d
its
p
r
etr
ain
e
d
I
m
ag
eNe
t w
eig
h
ts
en
ab
le
ef
f
ec
tiv
e
g
en
er
aliza
tio
n
to
n
ew
i
m
ag
e
s
eg
m
en
tatio
n
task
s
[
1
3
]
.
All
p
r
etr
ain
e
d
m
o
d
els
th
u
s
lev
er
a
g
e
I
m
a
g
eNe
t
-
b
ased
in
itializatio
n
to
im
p
r
o
v
e
co
n
v
er
g
en
ce
an
d
p
er
f
o
r
m
an
ce
.
T
h
is
r
esear
ch
p
r
o
p
o
s
es
a
c
o
astl
in
e
s
eg
m
en
tat
io
n
m
eth
o
d
t
h
at
d
is
tin
g
u
is
h
es
s
ea
wate
r
an
d
la
n
d
u
s
in
g
d
ee
p
lear
n
in
g
co
n
v
o
lu
ti
o
n
al
n
e
u
r
al
n
etwo
r
k
s
,
f
o
llo
wi
n
g
th
e
ap
p
r
o
ac
h
in
tr
o
d
u
ce
d
i
n
s
tu
d
y
[
7
]
.
I
n
t
h
at
s
tu
d
y
,
th
e
W
ater
Net
m
o
d
el
co
m
b
i
n
ed
o
u
tp
u
ts
f
r
o
m
s
e
v
er
al
d
ee
p
lear
n
in
g
s
eg
m
en
t
atio
n
ar
ch
itectu
r
es.
Similar
ly
,
th
e
p
r
esen
t
wo
r
k
ap
p
lies
an
en
s
em
b
le
s
tr
ateg
y
b
y
co
m
b
in
in
g
th
e
s
eg
m
en
tatio
n
r
esu
lts
o
f
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
u
s
in
g
a
m
ajo
r
ity
v
o
tin
g
m
e
ch
an
is
m
,
wh
er
e
th
e
f
in
al
class
f
o
r
ea
ch
p
ix
el
is
d
eter
m
in
ed
b
y
th
e
m
o
s
t
f
r
eq
u
en
t
p
r
ed
ictio
n
am
o
n
g
th
e
t
h
r
ee
m
o
d
els.
Fig
u
r
e
1
illu
s
tr
ates
th
e
p
r
o
p
o
s
ed
s
eg
m
en
tatio
n
wo
r
k
f
lo
w.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
15
,
No
.
2
,
J
u
n
e
20
26
:
5
8
8
-
5
9
6
590
Mo
d
el
p
er
f
o
r
m
a
n
ce
is
e
v
alu
a
ted
u
s
in
g
th
e
J
ac
ca
r
d
Similar
ity
C
o
ef
f
icien
t,
as
s
h
o
wn
in
(
1
)
.
I
n
th
is
eq
u
atio
n
,
A
r
ep
r
esen
ts
th
e
p
r
e
d
icted
s
eg
m
en
tatio
n
m
ask
p
r
o
d
u
ce
d
b
y
th
e
m
o
d
el,
a
n
d
B
r
e
p
r
esen
ts
th
e
g
r
o
u
n
d
-
tr
u
th
lab
el.
T
h
is
m
etr
ic
q
u
a
n
tifie
s
th
e
o
v
er
lap
b
etwe
en
p
r
ed
ictio
n
a
n
d
r
ef
e
r
en
ce
an
d
is
wid
ely
u
s
ed
in
s
em
an
tic
s
eg
m
en
tatio
n
task
s
.
Fig
u
r
e
1
.
Seg
m
e
n
tatio
n
p
r
o
ce
s
s
d
iag
r
am
(
,
)
=
|
∩
|
|
∪
|
(
1
)
B
ased
o
n
Fig
u
r
e
1
,
th
e
m
o
d
el
d
ev
elo
p
m
en
t
p
r
o
ce
s
s
b
e
g
in
s
with
tr
ain
in
g
th
r
ee
d
ee
p
lear
n
in
g
ar
ch
itectu
r
es
R
esNet5
0
,
R
esN
et1
8
,
an
d
Mo
b
ileNet
-
V2
u
s
in
g
8
2
4
s
atellite
im
ag
es.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
ea
ch
m
o
d
el
was
th
e
n
ev
al
u
ated
u
s
in
g
2
0
test
im
ag
es
t
o
an
al
y
ze
t
h
eir
s
eg
m
en
tatio
n
b
e
h
av
io
r
.
F
o
r
ea
ch
test
im
ag
e,
th
e
o
u
tp
u
ts
p
r
o
d
u
ce
d
b
y
th
e
th
r
ee
m
o
d
els
wer
e
r
ec
o
r
d
e
d
,
a
n
d
m
ajo
r
ity
v
o
tin
g
was
ap
p
lied
at
th
e
p
ix
el
lev
el
t
o
g
en
er
ate
th
e
f
in
al
s
eg
m
en
tatio
n
r
esu
lt (
2
)
.
Fig
u
r
e
2
illu
s
tr
ates th
e
wo
r
k
f
lo
w
o
f
th
is
m
ajo
r
it
y
v
o
tin
g
p
r
o
ce
s
s
.
Fig
u
r
e
2
.
Ma
jo
r
ity
v
o
tin
g
s
eg
m
en
tatio
n
f
lo
wch
a
r
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
C
o
a
s
tlin
e
s
eg
men
ta
tio
n
o
n
La
n
d
s
a
t 8
OLI
ima
g
es u
s
in
g
ma
j
o
r
ity
vo
tin
g
w
ith
d
ee
p
… (
N
u
r
N
a
fiiya
h
)
591
T
h
e
s
eg
m
en
tatio
n
task
co
n
s
is
ts
o
f
two
class
e
s
:
0
f
o
r
lan
d
an
d
1
f
o
r
s
ea
wate
r
.
I
n
(
2
)
,
H
r
ep
r
esen
ts
th
e
f
in
al
p
r
e
d
icted
lab
el
f
o
r
ea
ch
p
ix
el
af
ter
th
e
v
o
tin
g
p
r
o
ce
s
s
.
Ma
jo
r
ity
v
o
tin
g
d
eter
m
i
n
es
t
h
e
p
ix
el
class
b
ased
o
n
th
e
m
o
s
t
f
r
eq
u
en
t
p
r
e
d
ictio
n
am
o
n
g
th
e
th
r
ee
m
o
d
els,
th
e
r
eb
y
r
ed
u
ci
n
g
n
o
is
e
-
r
elate
d
er
r
o
r
s
b
y
lev
er
ag
in
g
th
e
co
llectiv
e
ag
r
ee
m
e
n
t o
f
m
u
ltip
le
class
if
ier
s
.
=
{
1
1
0
ℎ
}
(
2
)
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
tu
d
y
tr
ain
ed
a
to
tal
o
f
8
2
4
s
atellite
im
ag
es
u
s
in
g
th
r
ee
d
ee
p
lear
n
in
g
m
o
d
els:
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
.
All
m
o
d
els
wer
e
t
r
ain
ed
u
s
in
g
id
en
tical
h
y
p
er
p
ar
am
eter
s
,
i
n
clu
d
in
g
a
lear
n
in
g
r
ate
(
α
)
o
f
0
.
0
0
1
,
th
e
s
to
ch
ast
ic
g
r
ad
ien
t
d
escen
t
with
m
o
m
en
tu
m
(
SGDM
)
o
p
tim
izer
,
a
n
d
4
0
ep
o
ch
s
.
E
ac
h
m
o
d
el
was
tr
ain
ed
th
r
ee
tim
e
s
to
en
s
u
r
e
s
tab
ilit
y
an
d
co
n
s
is
ten
cy
o
f
r
esu
lts
.
T
h
e
av
e
r
ag
e
lo
s
s
an
d
ac
cu
r
ac
y
o
b
tain
ed
f
r
o
m
th
ese
tr
ain
in
g
s
ess
io
n
s
ar
e
s
u
m
m
ar
ized
in
T
ab
le
1
,
wh
ich
s
h
o
ws
th
at
R
esNet
5
0
ac
h
iev
ed
th
e
lo
west
lo
s
s
v
alu
e
(
0
.
0
1
4
)
an
d
th
e
h
ig
h
est
ac
cu
r
ac
y
(
9
9
.
4
8
%)
,
in
d
icatin
g
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
co
m
p
ar
ed
wit
h
R
esNet1
8
an
d
Mo
b
ileNet
-
V2
.
T
ab
le
1
.
E
v
alu
atio
n
o
f
th
e
tr
ai
n
in
g
p
r
o
ce
s
s
No
M
o
d
e
l
Lo
ss
A
c
c
u
r
a
c
y
(
%)
1
R
e
sN
e
t
5
0
0
.
0
1
4
9
9
.
4
8
2
R
e
sN
e
t
1
8
0
.
0
1
6
9
9
.
3
8
3
M
o
b
i
l
e
N
e
t
-
V2
0
.
0
3
5
9
8
.
7
6
Fig
u
r
e
3
illu
s
tr
ates
th
e
tr
ain
in
g
p
r
o
ce
s
s
f
o
r
ea
c
h
d
ee
p
l
ea
r
n
in
g
m
o
d
el:
Fig
u
r
e
3
(
a)
R
esNet5
0
,
Fig
u
r
e
3
(
b
)
R
esNet1
8
,
an
d
Fig
u
r
e
3
(
c)
Mo
b
ileNet
-
V2
.
T
h
ese
cu
r
v
es
s
h
o
w
ch
a
n
g
es
in
lo
s
s
an
d
ac
cu
r
ac
y
ac
r
o
s
s
4
0
ep
o
ch
s
.
Alth
o
u
g
h
t
h
e
n
u
m
b
e
r
o
f
iter
atio
n
s
v
ar
ies
d
ep
en
d
in
g
o
n
d
ataset
s
ize,
th
e
ac
cu
r
ac
y
ten
d
s
to
s
tab
ilize
af
ter
ap
p
r
o
x
im
ately
2
0
0
iter
atio
n
s
,
with
m
in
im
al
d
if
f
er
e
n
ce
s
b
e
y
o
n
d
th
at
p
o
in
t.
T
h
is
p
atter
n
d
em
o
n
s
tr
ates th
at
th
e
m
o
d
els co
n
v
er
g
e
ef
f
ec
ti
v
ely
with
in
th
e
g
iv
en
tr
ain
i
n
g
co
n
f
ig
u
r
atio
n
.
T
h
e
co
astl
in
e
s
eg
m
en
tatio
n
r
e
s
u
lts
p
r
o
d
u
ce
d
b
y
th
e
d
ee
p
lea
r
n
in
g
m
o
d
els
(
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
)
ar
e
p
r
esen
ted
s
eq
u
en
tially
in
T
ab
le
2
.
I
n
th
i
s
s
tu
d
y
,
t
h
e
J
ac
ca
r
d
Similar
ity
I
n
d
ex
was
u
s
ed
t
o
q
u
an
tify
th
e
s
im
ilar
ity
b
etwe
e
n
th
e
p
r
ed
icted
s
eg
m
en
tatio
n
an
d
th
e
g
r
o
u
n
d
-
tr
u
th
im
ag
es,
wh
er
e
v
alu
es
r
an
g
e
f
r
o
m
0
to
1
a
n
d
v
alu
es
clo
s
er
to
1
in
d
icate
h
ig
h
er
s
im
ilar
ity
.
T
h
e
s
eg
m
en
tatio
n
r
esu
lts
p
r
o
d
u
ce
d
u
s
in
g
th
e
m
ajo
r
ity
v
o
tin
g
ap
p
r
o
ac
h
ar
e
s
h
o
wn
in
T
ab
le
3
,
wh
ile
th
e
J
ac
ca
r
d
v
alu
es
f
o
r
ea
ch
i
n
d
iv
i
d
u
al
m
o
d
el
a
n
d
th
e
en
s
em
b
le
m
eth
o
d
ar
e
s
u
m
m
a
r
ized
in
T
ab
le
4
.
T
h
e
m
ajo
r
ity
v
o
tin
g
m
eth
o
d
c
o
m
b
in
es
th
e
o
u
tp
u
ts
o
f
all
m
o
d
els
with
th
e
aim
o
f
m
in
im
izi
n
g
e
r
r
o
r
s
;
m
o
d
els
with
h
ig
h
er
ac
cu
r
ac
y
ca
n
co
m
p
en
s
ate
f
o
r
th
o
s
e
wi
th
lo
wer
p
er
f
o
r
m
a
n
ce
,
r
esu
ltin
g
in
m
o
r
e
s
tab
le
p
r
ed
ictio
n
s
.
I
n
th
is
ca
s
e,
th
e
en
s
em
b
le
o
u
t
p
u
t
p
r
o
d
u
ce
d
r
esu
lts
th
at
wer
e
n
o
t
s
ig
n
if
ica
n
tly
d
if
f
e
r
en
t
f
r
o
m
R
esNet5
0
f
o
r
im
ag
es
2
,
3
,
4
,
an
d
5
.
Alth
o
u
g
h
th
e
J
ac
ca
r
d
Similar
ity
I
n
d
ex
is
a
p
o
wer
f
u
l
ev
alu
atio
n
m
et
r
ic,
it
h
as
th
e
lim
itatio
n
o
f
b
ein
g
less
s
en
s
itiv
e
to
v
ar
iatio
n
s
i
n
s
et
s
ize,
wh
ich
s
h
o
u
ld
b
e
co
n
s
id
er
ed
wh
en
in
ter
p
r
etin
g
t
h
e
r
esu
lts
.
B
ased
o
n
th
e
s
eg
m
en
tatio
n
r
esu
lts
o
f
I
m
ag
es
1
-
3
in
T
ab
l
e
2
,
ea
ch
d
ee
p
lea
r
n
in
g
m
o
d
el
ex
h
ib
its
s
p
ec
if
ic
wea
k
n
ess
es.
T
h
e
R
esNet5
0
m
o
d
el
p
r
o
d
u
ce
s
ad
d
itio
n
al
b
o
u
n
d
ar
y
lin
es,
in
d
icatin
g
th
at
its
s
eg
m
en
tatio
n
is
n
o
t
f
u
lly
ac
cu
r
ate.
T
h
e
R
esNet1
8
m
o
d
el
s
h
o
ws
s
m
all
h
o
les
an
d
lin
e
ar
tifa
c
ts
in
I
m
ag
es
1
a
n
d
2
,
s
u
g
g
esti
n
g
i
n
s
tab
ilit
y
in
d
is
tin
g
u
is
h
in
g
lan
d
a
r
ea
s
.
Similar
ly
,
th
e
Mo
b
ileN
et
-
V2
m
o
d
el
g
en
er
ates
lin
e
ar
tifa
cts
an
d
m
is
s
in
g
la
n
d
r
eg
i
o
n
s
in
I
m
ag
e
1
,
wh
ile
I
m
ag
es
2
a
n
d
3
c
o
n
tain
s
m
all
h
o
les,
in
d
icatin
g
th
at
th
is
m
o
d
el
h
as a
ten
d
en
cy
t
o
cr
ea
t
e
g
ap
s
with
in
lan
d
a
r
ea
s
.
B
ec
au
s
e
ea
ch
in
d
iv
id
u
al
m
o
d
el
(
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
)
p
r
o
d
u
ce
s
d
if
f
er
en
t
ty
p
es
o
f
ar
tifa
cts
s
u
ch
as
u
n
wan
ted
lin
es
an
d
h
o
les
s
o
m
e
o
f
t
h
ese
im
p
er
f
ec
tio
n
s
s
till
ap
p
ea
r
in
t
h
e
m
ajo
r
ity
v
o
tin
g
r
esu
lts
.
Ho
wev
er
,
th
e
en
s
em
b
le
ap
p
r
o
ac
h
g
en
er
ally
r
e
d
u
ce
s
th
ese
er
r
o
r
s
b
y
co
m
b
in
in
g
th
e
s
tr
en
g
th
s
o
f
all
th
r
ee
m
o
d
el
s
.
Fo
r
ex
am
p
le,
al
th
o
u
g
h
s
o
m
e
lin
e
ar
tifa
cts
r
e
m
ain
in
I
m
ag
es
1
an
d
2
in
T
a
b
le
3
,
th
eir
n
u
m
b
er
an
d
p
r
o
m
in
en
ce
ar
e
r
e
d
u
ce
d
c
o
m
p
ar
ed
to
th
e
r
esu
lts
p
r
o
d
u
c
ed
b
y
t
h
e
in
d
iv
i
d
u
al
m
o
d
els.
T
ab
le
3
p
r
esen
ts
f
iv
e
ex
am
p
les
o
f
L
an
d
s
at
8
OL
I
in
p
u
t
im
ag
es
alo
n
g
with
th
e
ir
co
r
r
esp
o
n
d
in
g
lab
els
(
b
lack
r
ep
r
esen
tin
g
lan
d
a
n
d
wh
ite
r
ep
r
esen
tin
g
s
ea
)
.
T
h
ese
im
ag
es we
r
e
s
eg
m
en
ted
u
s
in
g
th
e
m
ajo
r
ity
v
o
tin
g
m
eth
o
d
,
w
h
er
e
th
e
o
u
t
p
u
ts
f
r
o
m
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
a
r
e
co
m
b
in
e
d
to
d
ete
r
m
in
e
th
e
f
in
al
p
r
ed
ictio
n
.
Ov
er
all,
th
e
m
ajo
r
ity
v
o
tin
g
m
eth
o
d
s
h
o
ws
g
o
o
d
a
v
er
ag
e
p
er
f
o
r
m
an
ce
;
h
o
wev
e
r
,
s
eg
m
en
tatio
n
q
u
ality
d
ec
r
ea
s
es
wh
en
th
e
in
p
u
t
im
ag
e
c
o
n
tain
s
d
ar
k
er
wate
r
r
eg
i
o
n
s
o
r
co
m
p
lex
s
ea
-
lan
d
in
ter
ac
tio
n
s
.
I
n
s
u
ch
ca
s
es
esp
ec
ially
wh
en
th
e
s
ea
ex
ten
d
s
i
n
to
th
e
lan
d
ar
ea
th
e
en
s
em
b
l
e
m
eth
o
d
m
ay
s
till
p
r
o
d
u
ce
g
ap
s
o
r
less
ac
cu
r
ate
b
o
u
n
d
ar
ies.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
15
,
No
.
2
,
J
u
n
e
20
26
:
5
8
8
-
5
9
6
592
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
T
r
ain
in
g
p
r
o
ce
s
s
(
a)
R
esNet5
0
,
(
b
)
R
esNet1
8
,
an
d
(
c)
Mo
b
ileNet
-
V2
B
ased
o
n
T
ab
le
4
,
th
e
J
ac
ca
r
d
v
alu
es
o
f
R
esNet5
0
an
d
t
h
e
m
ajo
r
ity
v
o
tin
g
m
o
d
el
ar
e
id
en
tical,
in
d
icatin
g
th
at
th
e
en
s
em
b
le
m
eth
o
d
clo
s
ely
m
ir
r
o
r
s
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
tr
o
n
g
est
in
d
iv
id
u
al
m
o
d
el.
Ma
jo
r
ity
v
o
tin
g
o
f
f
er
s
a
n
a
d
v
an
tag
e
b
y
r
ed
u
cin
g
ar
tifa
c
ts
s
u
ch
as
s
m
all
h
o
les
o
r
li
n
e
d
is
to
r
tio
n
s
th
at
co
m
m
o
n
l
y
ap
p
ea
r
in
s
in
g
le
-
m
o
d
el
p
r
ed
ictio
n
s
,
alth
o
u
g
h
th
is
im
p
r
o
v
em
e
n
t
r
eq
u
ir
es
h
ig
h
er
co
m
p
u
tatio
n
al
co
s
t
b
ec
au
s
e
m
u
ltip
le
m
o
d
els m
u
s
t b
e
p
r
o
ce
s
s
ed
b
ef
o
r
e
v
o
tin
g
is
ap
p
lied
.
W
h
en
co
m
p
ar
ed
with
p
r
ev
io
u
s
s
tu
d
ies,
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
is
g
en
er
ally
co
m
p
ar
ab
le.
Fo
r
ex
am
p
le,
th
e
W
ater
Net
en
s
em
b
le
s
y
s
tem
in
[
5
]
ac
h
iev
ed
ex
ce
p
tio
n
all
y
h
ig
h
I
o
U
an
d
F1
-
s
co
r
es
o
f
9
9
.
5
9
%
an
d
9
9
.
7
9
%,
r
esp
ec
tiv
ely
,
u
s
in
g
a
co
m
b
in
atio
n
o
f
U
-
Net
v
ar
ian
ts
,
FC
-
Den
s
eNe
t,
an
d
Pix
2
P
ix
.
Similar
ly
,
th
e
co
m
p
a
r
ativ
e
f
r
am
ewo
r
k
in
[
6
]
r
ep
o
r
t
ed
m
ea
n
I
o
U
v
alu
es
ab
o
v
e
9
2
%
wh
en
ev
alu
atin
g
m
u
ltip
le
s
tate
-
of
-
th
e
-
ar
t D
C
NN
ar
ch
itectu
r
es f
o
r
s
ea
lan
d
s
eg
m
en
tatio
n
.
Alth
o
u
g
h
th
e
J
ac
ca
r
d
v
alu
es
in
th
is
s
tu
d
y
ar
e
s
lig
h
tly
l
o
wer
,
th
e
f
in
d
i
n
g
s
d
em
o
n
s
tr
a
te
th
at
a
lig
h
tweig
h
t
en
s
em
b
le
c
o
m
p
o
s
ed
o
f
R
esNet5
0
,
R
esNet1
8
,
an
d
Mo
b
ileNet
-
V2
ca
n
s
till
p
r
o
d
u
ce
c
o
m
p
etitiv
e
an
d
co
n
s
is
ten
t
s
eg
m
en
tatio
n
,
p
ar
ticu
lar
ly
in
r
ed
u
cin
g
b
o
u
n
d
ar
y
n
o
is
e
an
d
im
p
r
o
v
in
g
co
astl
in
e
s
tab
ilit
y
.
T
h
u
s
,
m
ajo
r
ity
v
o
tin
g
r
e
m
ain
s
a
v
iab
le
alt
er
n
ativ
e
f
o
r
s
ce
n
ar
io
s
th
at
r
eq
u
ir
e
r
o
b
u
s
t seg
m
en
tatio
n
with
lim
ited
m
o
d
el
co
m
p
lex
ity
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
C
o
a
s
tlin
e
s
eg
men
ta
tio
n
o
n
La
n
d
s
a
t 8
OLI
ima
g
es u
s
in
g
ma
j
o
r
ity
vo
tin
g
w
ith
d
ee
p
… (
N
u
r
N
a
fiiya
h
)
593
T
ab
le
2
.
Seg
m
e
n
tatio
n
re
s
u
lts
No
I
n
p
u
t
G
r
o
u
n
d
t
r
u
t
h
(
l
a
b
e
l
)
R
e
s
u
l
t
m
o
d
e
l
R
e
sN
e
t
5
0
R
e
s
u
l
t
m
o
d
e
l
R
e
sN
e
t
1
8
R
e
s
u
l
t
m
o
d
e
l
M
o
b
i
l
e
N
e
t
-
V2
1
2
3
4
5
T
ab
le
3
.
Ma
jo
r
ity
v
o
tin
g
s
eg
m
en
tatio
n
r
esu
lts
No
I
n
p
u
t
G
r
o
u
n
d
t
r
u
t
h
(
l
a
b
e
l
)
R
e
s
u
l
t
ma
j
o
r
i
t
y
v
o
t
i
n
g
1
2
3
4
5
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
15
,
No
.
2
,
J
u
n
e
20
26
:
5
8
8
-
5
9
6
594
T
h
is
s
tu
d
y
h
as
s
ev
er
al
lim
it
atio
n
s
.
First,
th
e
d
atase
t
s
ize
is
r
elativ
ely
s
m
all
an
d
r
estr
icted
to
a
s
p
ec
if
ic
g
eo
g
r
a
p
h
ic
r
e
g
io
n
,
wh
ich
m
ay
a
f
f
ec
t
th
e
m
o
d
el’
s
g
en
er
aliza
b
ilit
y
to
d
if
f
er
e
n
t
c
o
astl
in
e
co
n
d
itio
n
s
.
Seco
n
d
,
th
e
m
o
d
el
was
tr
ain
e
d
ex
clu
s
iv
ely
o
n
L
a
n
d
s
at
8
O
L
I
im
ag
er
y
,
leav
i
n
g
its
p
e
r
f
o
r
m
an
ce
ac
r
o
s
s
o
th
er
s
en
s
o
r
s
(
e.
g
.
,
Sen
tin
el
-
2
,
P
lan
etSco
p
e)
u
n
v
er
if
ied
.
T
h
i
r
d
,
th
e
m
ajo
r
ity
v
o
tin
g
e
n
s
em
b
le
in
cr
ea
s
es
co
m
p
u
tatio
n
al
co
s
t c
o
m
p
a
r
ed
with
s
in
g
le
-
m
o
d
el
ap
p
r
o
ac
h
es.
Ad
d
r
ess
in
g
th
ese
lim
itatio
n
s
will b
e
th
e
f
o
cu
s
o
f
f
u
tu
r
e
r
esear
c
h
.
T
ab
le
4
.
Mo
d
el
ev
alu
atio
n
p
er
f
o
r
m
an
ce
J
ac
ca
r
d
No
M
o
d
e
l
A
v
e
r
a
g
e
1
R
e
sN
e
t
5
0
0
.
9
8
2
R
e
sN
e
t
1
8
0
.
9
6
3
M
o
b
i
l
e
N
e
t
-
V2
0
.
9
5
4
M
a
j
o
r
i
t
y
V
o
t
i
n
g
(
R
e
sN
e
t
5
0
,
R
e
sN
e
t
1
8
,
M
o
b
i
l
e
N
e
t
-
V
2
)
0
.
9
8
4.
CO
NCLU
SI
O
N
T
h
is
s
t
u
d
y
in
tr
o
d
u
ce
d
a
n
e
n
s
e
m
b
le
-
b
ase
d
co
astl
in
e
s
e
g
m
e
n
t
ati
o
n
a
p
p
r
o
a
c
h
t
h
a
t
i
n
te
g
r
a
tes
R
esNet
5
0
,
R
esNet
1
8
,
a
n
d
M
o
b
il
eNe
t
-
V2
u
s
i
n
g
a
m
aj
o
r
it
y
v
o
ti
n
g
s
t
r
a
te
g
y
.
T
h
e
f
in
d
i
n
g
s
s
h
o
w
t
h
at
t
h
is
en
s
em
b
l
e
p
r
o
v
i
d
es
m
o
r
e
s
t
ab
le
a
n
d
c
o
n
s
is
t
en
t
c
o
astl
i
n
e
b
o
u
n
d
a
r
ies
t
h
a
n
in
d
i
v
i
d
u
al
m
o
d
els
,
p
ar
t
ic
u
l
ar
ly
b
y
r
e
d
u
ci
n
g
b
o
u
n
d
a
r
y
n
o
is
e
a
n
d
m
in
im
i
zi
n
g
a
r
t
if
ac
ts
s
u
c
h
as
s
m
a
ll
h
o
les
o
r
li
n
e
d
i
s
to
r
ti
o
n
s
.
T
h
es
e
r
es
u
lts
h
ig
h
l
i
g
h
t
t
h
e
p
o
te
n
ti
al
o
f
lig
h
t
wei
g
h
t
e
n
s
e
m
b
les
f
o
r
im
p
r
o
v
i
n
g
t
h
e
r
o
b
u
s
tn
ess
o
f
c
o
astl
in
e
e
x
t
r
ac
ti
o
n
i
n
r
e
m
o
te
s
e
n
s
i
n
g
a
p
p
lic
ati
o
n
s
.
Ho
w
ev
er
,
t
h
is
s
t
u
d
y
h
as
s
e
v
e
r
al
l
im
i
tat
io
n
s
.
T
h
e
d
at
aset
s
i
ze
is
r
e
lat
iv
el
y
s
m
all
a
n
d
li
m
it
ed
t
o
a
s
p
e
ci
f
ic
g
eo
g
r
a
p
h
ic
r
e
g
i
o
n
,
wh
i
ch
m
ay
r
est
r
ic
t
m
o
d
e
l
g
e
n
e
r
a
liz
ab
ilit
y
.
I
n
a
d
d
iti
o
n
,
t
h
e
e
x
p
e
r
i
m
e
n
ts
w
e
r
e
co
n
d
u
ct
ed
e
x
cl
u
s
i
v
el
y
o
n
L
a
n
d
s
at
8
OL
I
im
ag
er
y
,
l
ea
v
i
n
g
p
er
f
o
r
m
a
n
c
e
a
c
r
o
s
s
o
t
h
er
s
at
elli
te
s
e
n
s
o
r
s
u
n
t
est
ed
.
T
h
e
en
s
em
b
l
e
m
e
th
o
d
als
o
i
n
c
r
e
ases
c
o
m
p
u
ta
ti
o
n
al
c
o
s
t
c
o
m
p
a
r
e
d
wit
h
s
i
n
g
le
-
m
o
d
e
l i
m
p
le
m
e
n
ta
ti
o
n
s
.
Fu
tu
r
e
r
esear
ch
s
h
o
u
ld
th
er
e
f
o
r
e
ex
p
an
d
th
e
d
ataset
to
in
clu
d
e
m
u
ltip
le
r
eg
io
n
s
,
ev
alu
ate
th
e
m
o
d
el
ac
r
o
s
s
d
if
f
er
en
t
s
atellite
s
en
s
o
r
s
(
e.
g
.
,
Sen
tin
el
-
2
,
Plan
etSco
p
e)
,
an
d
in
v
esti
g
ate
m
o
r
e
ef
f
icien
t
en
s
em
b
l
e
s
tr
ateg
ies.
Ap
p
ly
in
g
th
e
m
eth
o
d
to
a
d
d
itio
n
al
b
en
c
h
m
ar
k
d
atasets
o
r
n
ewly
lab
eled
c
o
astal
im
ag
er
y
will
also
h
elp
ass
ess
it
s
ad
ap
tab
ilit
y
to
d
iv
er
s
e
s
h
o
r
elin
e
c
o
n
d
itio
n
s
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
an
k
y
o
u
to
th
e
r
esear
ch
an
d
s
er
v
ice
in
s
titu
te
o
f
U
n
iv
er
s
itas
I
s
lam
L
am
o
n
g
an
,
wh
ich
p
r
o
v
id
ed
th
e
r
esear
ch
g
r
a
n
t.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
a
u
th
o
r
s
s
tate
th
at
n
o
f
u
n
d
i
n
g
was
in
v
o
lv
e
d
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Nu
r
Naf
iiy
ah
✓
✓
✓
✓
✓
✓
✓
✓
Salwa
Nab
ilah
✓
✓
✓
✓
✓
Nu
r
Aziz
ah
Af
f
an
d
y
✓
✓
✓
✓
R
if
k
y
Ais
y
atu
l Far
o
h
✓
✓
E
s
a
Pra
k
asa
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
ter
est.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
C
o
a
s
tlin
e
s
eg
men
ta
tio
n
o
n
La
n
d
s
a
t 8
OLI
ima
g
es u
s
in
g
ma
j
o
r
ity
vo
tin
g
w
ith
d
ee
p
… (
N
u
r
N
a
fiiya
h
)
595
DATA AV
AI
L
AB
I
L
I
T
Y
L
an
d
s
at
8
OL
I
(
Op
e
r
atio
n
a
l
L
an
d
Ma
n
ag
er
)
im
ag
es
a
r
e
p
u
b
licly
av
ailab
le
o
n
th
e
web
s
ite
h
ttp
://www.
r
em
o
tesen
s
in
g
lab
.
y
ild
iz.
ed
u
.
tr
.
RE
F
E
R
E
NC
E
S
[
1
]
Y
.
Z
h
a
n
g
a
n
d
X
.
H
o
u
,
“
C
h
a
r
a
c
t
e
r
i
st
i
c
s
o
f
c
o
a
s
t
l
i
n
e
c
h
a
n
g
e
s
o
n
S
o
u
t
h
e
a
s
t
A
s
i
a
i
s
l
a
n
d
s
f
r
o
m
2
0
0
0
t
o
2
0
1
5
,
”
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
2
,
n
o
.
3
,
p
.
5
1
9
,
F
e
b
.
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s1
2
0
3
0
5
1
9
.
[
2
]
C
.
C
h
e
n
e
t
a
l
.
,
“
T
h
e
a
p
p
l
i
c
a
t
i
o
n
o
f
t
h
e
t
a
ss
e
l
e
d
c
a
p
t
r
a
n
sf
o
r
ma
t
i
o
n
a
n
d
f
e
a
t
u
r
e
k
n
o
w
l
e
d
g
e
f
o
r
t
h
e
e
x
t
r
a
c
t
i
o
n
o
f
c
o
a
st
l
i
n
e
i
n
f
o
r
mat
i
o
n
f
r
o
m
r
e
m
o
t
e
s
e
n
si
n
g
i
mag
e
s,”
Ad
v
a
n
c
e
s
i
n
S
p
a
c
e
Re
se
a
rch
,
v
o
l
.
6
4
,
n
o
.
9
,
p
p
.
1
7
8
0
–
1
7
9
1
,
N
o
v
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
sr
.
2
0
1
9
.
0
7
.
0
3
2
.
[
3
]
M
.
S
c
a
r
p
e
t
t
a
,
M
.
S
p
a
d
a
v
e
c
c
h
i
a
,
V
.
I
.
D
’
A
l
e
ssan
d
r
o
,
L.
D
e
P
a
l
m
a
,
a
n
d
N
.
G
i
a
q
u
i
n
t
o
,
“
A
n
e
w
d
a
t
a
set
o
f
sa
t
e
l
l
i
t
e
i
ma
g
e
s
f
o
r
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
c
o
a
st
l
i
n
e
mea
s
u
r
e
m
e
n
t
,
”
i
n
2
0
2
2
I
EEE
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
e
t
r
o
l
o
g
y
f
o
r
E
x
t
e
n
d
e
d
Re
a
l
i
t
y
,
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
N
e
u
ra
l
En
g
i
n
e
e
r
i
n
g
(
Me
t
ro
XR
AI
N
E)
,
I
EEE,
O
c
t
.
2
0
2
2
,
p
p
.
6
3
5
–
6
4
0
.
d
o
i
:
1
0
.
1
1
0
9
/
M
e
t
r
o
X
R
A
I
N
E5
4
8
2
8
.
2
0
2
2
.
9
9
6
7
5
7
4
.
[
4
]
E.
M
c
A
l
l
i
s
t
e
r
,
A
.
P
a
y
o
,
A
.
N
o
v
e
l
l
i
n
o
,
T.
D
o
l
p
h
i
n
,
a
n
d
E.
M
e
d
i
n
a
-
Lo
p
e
z
,
“
M
u
l
t
i
s
p
e
c
t
r
a
l
sa
t
e
l
l
i
t
e
i
m
a
g
e
r
y
a
n
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
t
h
e
e
x
t
r
a
c
t
i
o
n
o
f
sh
o
r
e
l
i
n
e
i
n
d
i
c
a
t
o
r
s,”
C
o
a
st
a
l
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
7
4
,
p
.
1
0
4
1
0
2
,
Ju
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
a
st
a
l
e
n
g
.
2
0
2
2
.
1
0
4
1
0
2
.
[
5
]
F
.
Er
d
e
m,
B
.
B
a
y
r
a
m,
T.
B
a
k
i
r
ma
n
,
O
.
C
.
B
a
y
r
a
k
,
a
n
d
B
.
A
k
p
i
n
a
r
,
“
A
n
e
n
s
e
mb
l
e
d
e
e
p
l
e
a
r
n
i
n
g
b
a
s
e
d
s
h
o
r
e
l
i
n
e
se
g
m
e
n
t
a
t
i
o
n
a
p
p
r
o
a
c
h
(
W
a
t
e
r
N
e
t
)
f
r
o
m
L
a
n
d
s
a
t
8
O
LI
i
m
a
g
e
s,
”
A
d
v
a
n
c
e
s
i
n
S
p
a
c
e
Re
s
e
a
r
c
h
,
v
o
l
.
6
7
,
n
o
.
3
,
p
p
.
9
6
4
–
9
7
4
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
sr
.
2
0
2
0
.
1
0
.
0
4
3
.
[
6
]
T.
Y
a
n
g
e
t
a
l
.
,
“
S
e
a
-
l
a
n
d
se
g
me
n
t
a
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
La
n
d
sa
t
-
8
O
LI
i
ma
g
e
r
y
,
”
Ma
r
i
n
e
G
e
o
d
e
sy
,
v
o
l
.
4
3
,
n
o
.
2
,
p
p
.
1
0
5
–
1
3
3
,
M
a
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
8
0
/
0
1
4
9
0
4
1
9
.
2
0
2
0
.
1
7
1
3
2
6
6
.
[
7
]
C
.
C
h
e
n
,
J
.
F
u
,
S
.
Z
h
a
n
g
,
a
n
d
X
.
Z
h
a
o
,
“
C
o
a
s
t
l
i
n
e
i
n
f
o
r
ma
t
i
o
n
e
x
t
r
a
c
t
i
o
n
b
a
s
e
d
o
n
t
h
e
t
a
sse
l
e
d
c
a
p
t
r
a
n
sf
o
r
ma
t
i
o
n
o
f
L
a
n
d
s
a
t
-
8
O
LI
i
mag
e
s
,
”
Est
u
a
ri
n
e
,
C
o
a
st
a
l
a
n
d
S
h
e
l
f
S
c
i
e
n
c
e
,
v
o
l
.
2
1
7
,
p
p
.
2
8
1
–
2
9
1
,
F
e
b
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
c
ss.
2
0
1
8
.
1
0
.
0
2
1
.
[
8
]
Y
.
Li
u
,
X
.
W
a
n
g
,
F
.
Li
n
g
,
S
.
X
u
,
a
n
d
C
.
W
a
n
g
,
“
A
n
a
l
y
s
i
s
o
f
c
o
a
st
l
i
n
e
e
x
t
r
a
c
t
i
o
n
f
r
o
m
La
n
d
sa
t
-
8
O
LI
i
ma
g
e
r
y
,
”
W
a
t
e
r
,
v
o
l
.
9
,
n
o
.
1
1
,
p
.
8
1
6
,
O
c
t
.
2
0
1
7
,
d
o
i
:
1
0
.
3
3
9
0
/
w
9
1
1
0
8
1
6
.
[
9
]
C
.
C
h
e
n
e
t
a
l
.
,
“
T
e
mp
o
r
a
l
a
n
d
sp
a
t
i
a
l
v
a
r
i
a
t
i
o
n
o
f
c
o
a
s
t
l
i
n
e
u
si
n
g
r
e
mo
t
e
se
n
s
i
n
g
i
ma
g
e
s
f
o
r
Z
h
o
u
s
h
a
n
a
r
c
h
i
p
e
l
a
g
o
,
C
h
i
n
a
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
p
p
l
i
e
d
Ea
r
t
h
O
b
s
e
rv
a
t
i
o
n
a
n
d
G
e
o
i
n
f
o
rm
a
t
i
o
n
,
v
o
l
.
1
0
7
,
p
.
1
0
2
7
1
1
,
M
a
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
a
g
.
2
0
2
2
.
1
0
2
7
1
1
.
[
1
0
]
S
.
H
o
ż
y
ń
a
n
d
J
.
Za
l
e
w
s
k
i
,
“
S
h
o
r
e
l
i
n
e
d
e
t
e
c
t
i
o
n
a
n
d
l
a
n
d
se
g
me
n
t
a
t
i
o
n
f
o
r
a
u
t
o
n
o
mo
u
s
s
u
r
f
a
c
e
v
e
h
i
c
l
e
n
a
v
i
g
a
t
i
o
n
w
i
t
h
t
h
e
u
s
e
o
f
a
n
o
p
t
i
c
a
l
s
y
s
t
e
m,
”
S
e
n
s
o
rs
,
v
o
l
.
2
0
,
n
o
.
1
0
,
p
.
2
7
9
9
,
M
a
y
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
0
1
0
2
7
9
9
.
[
1
1
]
D
.
C
h
e
n
g
,
G
.
M
e
n
g
,
S
.
X
i
a
n
g
,
a
n
d
C
.
P
a
n
,
“
F
u
si
o
n
N
e
t
:
e
d
g
e
a
w
a
r
e
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
s
f
o
r
sem
a
n
t
i
c
se
g
me
n
t
a
t
i
o
n
o
f
r
e
mo
t
e
se
n
si
n
g
h
a
r
b
o
r
i
ma
g
e
s,”
I
EE
E
J
o
u
r
n
a
l
o
f
S
e
l
e
c
t
e
d
T
o
p
i
c
s
i
n
Ap
p
l
i
e
d
Ea
r
t
h
O
b
ser
v
a
t
i
o
n
s
a
n
d
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
0
,
n
o
.
1
2
,
p
p
.
5
7
6
9
–
5
7
8
3
,
D
e
c
.
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
TA
R
S
.
2
0
1
7
.
2
7
4
7
5
9
9
.
[
1
2
]
A
.
G
a
o
e
t
a
l
.
,
“
A
v
e
c
t
o
r
-
b
a
se
d
c
o
a
st
l
i
n
e
sh
a
p
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
p
p
r
o
a
c
h
u
si
n
g
se
q
u
e
n
t
i
a
l
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ap
p
l
i
e
d
Ea
r
t
h
O
b
s
e
rv
a
t
i
o
n
a
n
d
G
e
o
i
n
f
o
rm
a
t
i
o
n
,
v
o
l
.
1
2
9
,
p
.
1
0
3
8
1
0
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
a
g
.
2
0
2
4
.
1
0
3
8
1
0
.
[
1
3
]
L.
-
C
.
C
h
e
n
,
Y
.
Zh
u
,
G
.
P
a
p
a
n
d
r
e
o
u
,
F
.
S
c
h
r
o
f
f
,
a
n
d
H
.
A
d
a
m,
“
E
n
c
o
d
e
r
-
d
e
c
o
d
e
r
w
i
t
h
a
t
r
o
u
s
s
e
p
a
r
a
b
l
e
c
o
n
v
o
l
u
t
i
o
n
f
o
r
sem
a
n
t
i
c
i
ma
g
e
s
e
g
me
n
t
a
t
i
o
n
,
”
i
n
C
o
m
p
u
t
e
r
V
i
si
o
n
–
EC
C
V
2
0
1
8
,
V
.
F
e
r
r
a
r
i
,
M
.
H
e
b
e
r
t
,
C
.
S
m
i
n
c
h
i
ses
c
u
,
a
n
d
Y
.
W
e
i
ss
,
Ed
s
.
,
S
p
r
i
n
g
e
r
,
C
h
a
m,
2
0
1
8
,
p
p
.
8
3
3
–
8
5
1
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
0
1
2
3
4
-
2
_
4
9
.
[
1
4
]
M
.
A
l
a
m
,
J.
-
F
.
W
a
n
g
,
C
.
G
u
a
n
g
p
e
i
,
L.
Y
u
n
r
o
n
g
,
a
n
d
Y
.
C
h
e
n
,
“
C
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
t
h
e
se
m
a
n
t
i
c
s
e
g
m
e
n
t
a
t
i
o
n
o
f
r
e
m
o
t
e
se
n
s
i
n
g
i
m
a
g
e
s
,
”
M
o
b
i
l
e
N
e
t
w
o
rk
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
6
,
n
o
.
1
,
p
p
.
200
-
2
1
5
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
3
6
-
020
-
01703
-
3.
[
1
5
]
K
.
B
.
D
a
n
g
e
t
a
l
.
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s
t
o
d
e
t
e
c
t
c
o
a
st
l
i
n
e
s
a
n
d
s
h
o
r
e
l
i
n
e
s,”
J
o
u
rn
a
l
o
f
E
n
v
i
r
o
n
m
e
n
t
a
l
Ma
n
a
g
e
m
e
n
t
,
v
o
l
.
3
2
0
,
p
.
1
1
5
7
3
2
,
O
c
t
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
e
n
v
ma
n
.
2
0
2
2
.
1
1
5
7
3
2
.
[
1
6
]
C
.
S
e
a
l
e
,
T.
R
e
d
f
e
r
n
,
P
.
C
h
a
t
f
i
e
l
d
,
C
.
Lu
o
,
a
n
d
K
.
D
e
m
p
se
y
,
“
C
o
a
s
t
l
i
n
e
d
e
t
e
c
t
i
o
n
i
n
sat
e
l
l
i
t
e
i
ma
g
e
r
y
:
A
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
o
n
n
e
w
b
e
n
c
h
mar
k
d
a
t
a
,
”
R
e
m
o
t
e
S
e
n
si
n
g
o
f
E
n
v
i
ro
n
m
e
n
t
,
v
o
l
.
2
7
8
,
p
.
1
1
3
0
4
4
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
se.
2
0
2
2
.
1
1
3
0
4
4
.
[
1
7
]
R
.
Li
e
t
a
l
.
,
“
D
e
e
p
U
N
e
t
:
a
d
e
e
p
f
u
l
l
y
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
f
o
r
p
i
x
e
l
-
l
e
v
e
l
se
a
-
l
a
n
d
s
e
g
m
e
n
t
a
t
i
o
n
,
”
I
EEE
J
o
u
rn
a
l
o
f
S
e
l
e
c
t
e
d
T
o
p
i
c
s
i
n
Ap
p
l
i
e
d
Ea
r
t
h
O
b
ser
v
a
t
i
o
n
s
a
n
d
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
1
,
n
o
.
1
1
,
p
p
.
3
9
5
4
–
3
9
6
2
,
N
o
v
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
TA
R
S
.
2
0
1
8
.
2
8
3
3
3
8
2
.
[
1
8
]
H
.
S
u
,
S
.
W
e
i
,
J
.
Q
i
u
,
a
n
d
W
.
W
u
,
“
R
a
f
t
N
e
t
:
a
n
e
w
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
c
o
a
st
a
l
r
a
f
t
a
q
u
a
c
u
l
t
u
r
e
e
x
t
r
a
c
t
i
o
n
f
r
o
m
La
n
d
s
a
t
8
O
LI
d
a
t
a
,
”
Re
m
o
t
e
S
e
n
s
i
n
g
,
v
o
l
.
1
4
,
n
o
.
1
8
,
p
.
4
5
8
7
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s
1
4
1
8
4
5
8
7
.
[
19]
P
.
M
a
r
t
i
-
P
u
i
g
,
M
.
S
e
r
r
a
-
S
e
r
r
a
,
F
.
R
i
b
a
s,
G
.
S
i
marr
o
,
a
n
d
M
.
C
a
b
a
l
l
e
r
i
a
,
“
A
u
t
o
ma
t
i
c
s
h
o
r
e
l
i
n
e
d
e
t
e
c
t
i
o
n
b
y
p
r
o
c
e
ssi
n
g
p
l
a
n
v
i
e
w
t
i
m
e
x
i
ma
g
e
s
u
s
i
n
g
b
i
-
LST
M
n
e
t
w
o
r
k
s,
”
Ex
p
e
r
t
S
y
st
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
4
0
,
p
.
1
2
2
5
6
6
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
2
2
5
6
6
.
[
2
0
]
M
.
B
a
l
d
e
o
n
C
a
l
i
st
o
a
n
d
S
.
K
.
L
a
i
-
Y
u
e
n
,
“
A
d
a
E
n
-
N
e
t
:
a
n
e
n
sem
b
l
e
o
f
a
d
a
p
t
i
v
e
2
D
–
3
D
f
u
l
l
y
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
s
f
o
r
m
e
d
i
c
a
l
i
ma
g
e
s
e
g
me
n
t
a
t
i
o
n
,
”
N
e
u
ra
l
N
e
t
w
o
r
k
s
,
v
o
l
.
1
2
6
,
p
p
.
7
6
–
9
4
,
J
u
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
u
n
e
t
.
2
0
2
0
.
0
3
.
0
0
7
.
[
2
1
]
N
.
N
a
f
i
’
i
y
a
h
,
C
.
F
a
t
i
c
h
a
h
,
D
.
H
e
r
u
m
u
r
t
i
,
E.
R
.
A
s
t
u
t
i
,
a
n
d
R
.
H
.
P
u
t
r
a
,
“
M
o
b
i
l
e
N
e
t
V
2
e
n
s
e
m
b
l
e
s
e
g
me
n
t
a
t
i
o
n
f
o
r
ma
n
d
i
b
u
l
a
r
o
n
p
a
n
o
r
a
mi
c
r
a
d
i
o
g
r
a
p
h
y
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
En
g
i
n
e
e
r
i
n
g
a
n
d
S
y
s
t
e
m
s
,
v
o
l
.
1
6
,
n
o
.
2
,
p
p
.
5
4
6
–
5
6
0
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
2
2
2
6
6
/
i
j
i
e
s
2
0
2
3
.
0
4
3
0
.
4
5
.
[
2
2
]
Z.
M
i
a
o
,
K
.
F
u
,
H
.
S
u
n
,
X
.
S
u
n
,
a
n
d
M
.
Y
a
n
,
“
A
u
t
o
ma
t
i
c
w
a
t
e
r
-
b
o
d
y
se
g
men
t
a
t
i
o
n
f
r
o
m h
i
g
h
-
r
e
s
o
l
u
t
i
o
n
s
a
t
e
l
l
i
t
e
i
ma
g
e
s
v
i
a
d
e
e
p
n
e
t
w
o
r
k
s,”
I
EE
E
G
e
o
s
c
i
e
n
c
e
a
n
d
R
e
m
o
t
e
S
e
n
si
n
g
L
e
t
t
e
rs
,
v
o
l
.
1
5
,
n
o
.
4
,
p
p
.
6
0
2
–
6
0
6
,
A
p
r
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
LG
R
S
.
2
0
1
8
.
2
7
9
4
5
4
5
.
[
2
3
]
L.
Y
a
o
,
D
.
K
a
n
o
u
l
a
s
,
Z
.
J
i
,
a
n
d
Y
.
L
i
u
,
“
S
h
o
r
e
l
i
n
e
N
e
t
:
a
n
e
f
f
i
c
i
e
n
t
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
s
h
o
r
e
l
i
n
e
s
e
ma
n
t
i
c
se
g
me
n
t
a
t
i
o
n
f
o
r
u
n
ma
n
n
e
d
s
u
r
f
a
c
e
v
e
h
i
c
l
e
s,
”
i
n
2
0
2
1
I
E
EE/
RS
J
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
t
e
l
l
i
g
e
n
t
Ro
b
o
t
s
a
n
d
S
y
st
e
m
s
(
I
RO
S
)
,
I
EEE
,
S
e
p
.
2
0
2
1
,
p
p
.
5
4
0
3
–
5
4
0
9
.
d
o
i
:
1
0
.
1
1
0
9
/
I
R
O
S
5
1
1
6
8
.
2
0
2
1
.
9
6
3
6
6
1
4
.
[
2
4
]
Z.
Li
,
R
.
W
a
n
g
,
W
.
Z
h
a
n
g
,
F
.
H
u
,
a
n
d
L.
M
e
n
g
,
“
M
u
l
t
i
s
c
a
l
e
f
e
a
t
u
r
e
s
s
u
p
p
o
r
t
e
d
D
e
e
p
L
a
b
V
3
+
o
p
t
i
m
i
z
a
t
i
o
n
sc
h
e
me
f
o
r
a
c
c
u
r
a
t
e
w
a
t
e
r
s
e
ma
n
t
i
c
se
g
me
n
t
a
t
i
o
n
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
7
,
p
p
.
1
5
5
7
8
7
–
1
5
5
8
0
4
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
4
9
6
3
5
.
[
2
5
]
N
.
N
a
f
i
i
y
a
h
,
C
.
F
a
t
i
c
h
a
h
,
D
.
H
e
r
u
m
u
r
t
i
,
E
.
R
.
A
st
u
t
i
,
R
.
H
.
P
u
t
r
a
,
a
n
d
E.
P
r
a
k
a
s
a
,
“
M
a
n
d
i
b
u
l
a
r
se
g
me
n
t
a
t
i
o
n
o
n
p
a
n
o
r
a
mi
c
r
a
d
i
o
g
r
a
p
h
s
w
i
t
h
C
N
N
Tr
a
n
sf
e
r
L
e
a
r
n
i
n
g
,
”
i
n
2
0
2
2
I
EEE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
,
N
e
t
w
o
rks
a
n
d
S
a
t
e
l
l
i
t
e
(
C
O
MN
ET
S
AT)
,
I
EEE,
N
o
v
.
2
0
2
2
,
p
p
.
1
9
0
–
1
9
4
.
d
o
i
:
1
0
.
1
1
0
9
/
C
O
M
N
ETSA
T5
6
0
3
3
.
2
0
2
2
.
9
9
9
4
4
0
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
15
,
No
.
2
,
J
u
n
e
20
26
:
5
8
8
-
5
9
6
596
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Nur
Na
fii
y
a
h
re
c
e
iv
e
d
h
e
r
Ba
c
h
e
lo
r
o
f
In
f
o
rm
a
ti
c
s
En
g
in
e
e
rin
g
fro
m
Un
i
v
e
rsitas
Isla
m
Lam
o
n
g
a
n
,
In
d
o
n
e
sia
(2
0
0
5
-
2
0
0
9
)
a
n
d
M
a
ste
r
o
f
I
n
fo
rm
a
ti
o
n
Tec
h
n
o
lo
g
y
fr
o
m
t
h
e
S
e
k
o
lah
Ti
n
g
g
i
Tek
n
i
k
S
u
ra
b
a
y
a
,
In
d
o
n
e
sia
in
(2
0
1
1
-
2
0
1
3
).
S
h
e
is
t
h
e
P
h
.
D.
d
e
g
re
e
in
c
o
m
p
u
ter
sc
ien
c
e
with
t
h
e
De
p
a
rtme
n
t
o
f
In
f
o
rm
a
ti
c
s,
In
stit
u
t
Te
k
n
o
lo
g
i
S
e
p
u
l
u
h
No
p
e
m
b
e
r
2
0
2
3
.
S
h
e
is
c
u
rre
n
tl
y
i
n
tere
ste
d
in
a
rti
ficia
l
in
telli
g
e
n
c
e
,
d
e
e
p
lea
rn
in
g
,
a
n
d
c
o
m
p
u
ter
v
isio
n
.
S
h
e
is
c
u
rre
n
tl
y
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
a
t
th
e
De
p
a
rtme
n
t
o
f
In
fo
rm
a
ti
c
s
a
t
Un
iv
e
rsitas
Is
lam
Lam
o
n
g
a
n
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
y
n
a
ff@u
n
isla.a
c
.
i
d
.
S
a
lwa
Na
b
il
a
h
is
c
u
rre
n
tl
y
w
o
rk
i
n
g
a
s
a
n
As
sista
n
t
P
ro
fe
ss
o
r
i
n
t
h
e
De
p
a
rtme
n
t
o
f
Ci
v
il
En
g
in
e
e
rin
g
,
F
a
c
u
lt
y
o
f
En
g
i
n
e
e
rin
g
,
Un
iv
e
rsitas
Isla
m
L
a
m
o
n
g
a
n
,
In
d
o
n
e
sia
.
S
h
e
d
i
d
h
e
r
Ba
c
h
e
lo
r
De
g
re
e
a
t
G
e
o
m
a
ti
c
s
En
g
i
n
e
e
rin
g
o
f
I
n
stit
u
t
Tek
n
o
lo
g
i
S
e
p
u
l
u
h
N
o
p
e
m
b
e
r.
S
h
e
c
o
n
ti
n
u
e
d
h
e
r
M
a
ste
r'
s
De
g
re
e
a
t
th
e
De
p
a
rtme
n
t
o
f
G
e
o
m
a
ti
c
s,
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
Na
ti
o
n
a
l
Ch
e
n
g
K
u
n
g
Un
i
v
e
rsity
,
Taiwa
n
.
S
h
e
to
o
k
th
e
G
e
o
sp
a
ti
a
l
field
,
wh
ic
h
sh
e
lea
rn
e
d
a
b
o
u
t
re
m
o
te
se
n
sin
g
a
n
d
g
e
o
s
p
a
ti
a
l
in
f
o
rm
a
ti
o
n
sy
ste
m
s
(G
I
S
).
S
h
e
a
lso
lea
rn
e
d
a
b
o
u
t
meas
u
rin
g
th
e
field
.
S
h
e
h
a
s
a
lmo
st
3
y
e
a
rs
o
f
Tea
c
h
in
g
e
x
p
e
rien
c
e
;
sh
e
h
a
s
p
u
b
li
sh
e
d
a
ro
u
n
d
6
j
o
u
r
n
a
ls
wh
il
e
w
o
rk
i
n
g
a
t
Un
iv
e
rsitas
Isla
m
Lam
o
n
g
a
n
.
He
r
re
s
e
a
rc
h
a
re
a
is
G
e
o
m
a
ti
c
s,
in
p
a
rti
c
u
lar ab
o
u
t
G
e
o
sp
a
ti
a
l.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sa
lwa
n
a
b
il
a
h
@u
n
i
sla
.
a
c
.
id
.
Nur
Az
iza
h
Affa
n
d
y
is
c
u
rre
n
tl
y
wo
r
k
i
n
g
a
s
a
n
As
sista
n
t
P
ro
fe
ss
o
r
in
t
h
e
De
p
a
rtme
n
t
o
f
Civ
il
En
g
i
n
e
e
rin
g
,
F
a
c
u
lt
y
o
f
En
g
in
e
e
ri
n
g
,
Un
i
v
e
rsitas
Isla
m
La
m
o
n
g
a
n
,
In
d
o
n
e
sia
.
S
h
e
c
o
m
p
lete
d
h
e
r
u
n
d
e
rg
ra
d
u
a
te
stu
d
ies
a
n
d
e
a
rn
e
d
a
Ba
c
h
e
lo
r
o
f
En
g
i
n
e
e
rin
g
d
e
g
re
e
in
2
0
0
2
in
t
h
e
field
o
f
C
iv
il
E
n
g
in
e
e
rin
g
a
t
th
e
F
a
c
u
lt
y
o
f
En
g
i
n
e
e
rin
g
,
Un
i
v
e
rsitas
Isla
m
M
a
lan
g
.
In
2
0
0
8
,
s
h
e
p
u
rs
u
e
d
a
M
a
ste
r'
s
p
ro
g
ra
m
in
En
g
i
n
e
e
rin
g
a
n
d
g
ra
d
u
a
ted
i
n
2
0
1
1
fro
m
In
stit
u
t
Tek
n
o
l
o
g
i
S
e
p
u
lu
h
No
p
e
m
b
e
r
S
u
ra
b
a
y
a
with
a
c
o
n
c
e
n
tratio
n
in
Wate
r
Re
so
u
rc
e
s
M
a
n
a
g
e
m
e
n
t.
In
2
0
1
9
,
sh
e
c
o
n
ti
n
u
e
d
h
e
r
d
o
c
to
ra
l
p
r
o
g
ra
m
in
th
e
sa
m
e
d
e
p
a
rtme
n
t
a
t
In
stit
u
t
Tek
n
o
lo
g
i
S
e
p
u
lu
h
No
p
e
m
b
e
r
.
He
r
re
se
a
r
c
h
in
tere
sts
in
c
lu
d
e
wa
ter
re
so
u
rc
e
s
m
a
n
a
g
e
m
e
n
t,
g
re
e
n
b
u
il
d
in
g
stru
c
t
u
re
,
a
n
d
m
o
d
e
li
n
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
u
ra
z
iza
h
@u
n
isla.a
c
.
id
.
Rifk
y
Ais
y
a
tu
l
F
a
r
o
h
o
b
tain
e
d
h
e
r
Ba
c
h
e
lo
r
o
f
S
c
ien
c
e
d
e
g
re
e
fro
m
U
n
iv
e
rsitas
Isla
m
Ne
g
e
ri
M
a
u
lan
a
M
a
li
k
I
b
ra
h
im
M
a
lan
g
,
In
d
o
n
e
sia
(2
0
1
2
–
2
0
1
7
),
a
n
d
h
e
r
M
a
ste
r’s
d
e
g
re
e
in
M
a
th
e
m
a
ti
c
s
fro
m
Un
iv
e
rsitas
Bra
wijay
a
M
a
lan
g
,
I
n
d
o
n
e
sia
(2
0
1
7
–
2
0
1
9
).
He
r
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
ti
m
e
se
ries
fo
re
c
a
stin
g
,
f
u
z
z
y
ti
m
e
se
ries
,
a
p
p
li
e
d
sta
ti
stics
,
a
n
d
d
a
ta
a
n
a
ly
sis.
S
h
e
c
a
n
b
e
re
a
c
h
e
d
a
t
rif
k
y
a
isy
a
t
u
lfaro
h
@
u
n
isla.a
c
.
id
.
Sh
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
rifk
y
a
isy
a
t
u
lfaro
h
@u
n
isla.a
c
.
id
.
Esa
Pra
k
a
sa
h
a
s
c
o
m
p
lete
d
P
h
D
stu
d
y
i
n
El
e
c
tri
c
a
l
a
n
d
El
e
c
tr
o
n
ics
E
n
g
i
n
e
e
rin
g
fro
m
Un
iv
e
rsiti
Tek
n
o
lo
g
i
P
ET
R
ON
AS,
M
a
lay
sia
,
in
2
0
1
4
.
Bo
t
h
m
a
ste
r
a
n
d
b
a
c
h
e
lo
r
d
e
g
re
e
s
h
a
v
e
b
e
e
n
p
re
v
i
o
u
sl
y
fin
ish
e
d
fr
o
m
th
e
Un
iv
e
rsit
y
o
f
G
a
d
jah
M
a
d
a
,
Yo
g
y
a
k
a
rta.
He
h
a
s j
o
in
e
d
Re
se
a
rc
h
C
e
n
ter
fo
r
In
fo
rm
a
ti
c
s,
I
n
d
o
n
e
sia
n
In
stit
u
te
o
f
S
c
ien
c
e
s
(LI
P
I)
sin
c
e
2
0
0
5
.
T
h
e
in
stit
u
te
is
n
o
w
tran
sf
o
rm
e
d
in
to
th
e
Na
ti
o
n
a
l
Re
se
a
rc
h
a
n
d
I
n
n
o
v
a
ti
o
n
Ag
e
n
c
y
(BRIN).
His
re
se
a
rc
h
in
tere
st
is
m
a
in
ly
in
d
ig
i
tal
ima
g
e
p
ro
c
e
ss
in
g
,
c
o
m
p
u
ter
v
isi
o
n
,
a
n
d
p
a
tt
e
r
n
r
e
c
o
g
n
it
i
o
n
.
He
h
a
s
p
u
b
l
ish
e
d
m
a
n
y
re
se
a
rc
h
p
a
p
e
rs
i
n
i
n
tern
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
s
a
n
d
j
o
u
r
n
a
ls.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
e
sa
.
p
ra
k
a
sa
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.