I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
6
,
No
.
4
,
Dec
em
b
er
2017
,
p
p
.
1
59
~
16
5
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
6
.
i
4
.
p
p
1
5
9
-
1
6
5
159
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
AI
Cla
ss
ificatio
n of
Ro
a
d Da
m
a
g
e f
ro
m
Dig
ital I
m
a
g
e
U
sing
Ba
ck
pro
pa
g
a
tion
Neura
l Ne
tw
o
rk
Su
t
i
k
no
,
H
el
m
ie
Arif
Wi
ba
w
a
,
P
ri
m
a
Yus
uf
B
ud
ia
rt
o
D
e
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
/
In
f
o
rm
a
ti
c
s
,
Dip
o
n
e
g
o
r
o
Un
iv
e
rsity
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
2
3
,
2
0
1
7
R
ev
i
s
ed
Oct
25
,
2
0
1
7
A
cc
ep
ted
No
v
8
,
2
0
1
7
On
e
o
f
th
e
b
ig
g
e
st
c
a
u
se
s
o
f
d
e
a
th
in
th
e
w
o
rld
is
a
tra
ff
ic
a
c
c
i
d
e
n
t.
Ro
a
d
d
a
m
a
g
e
is
o
n
e
o
f
th
e
c
a
u
se
f
a
c
to
rs
f
ro
m
th
e
traff
ic
a
c
c
id
e
n
t.
T
o
r
e
d
u
c
e
th
is
p
ro
b
lem
is
re
q
u
ired
a
n
e
a
rly
d
e
tec
ti
o
n
a
g
a
in
st
ro
a
d
d
a
m
a
g
e
.
T
h
is
p
a
p
e
r
d
e
sc
rib
e
s
h
o
w
to
c
las
sify
ro
a
d
d
a
m
a
g
e
u
sin
g
ima
g
e
p
ro
c
e
ss
in
g
a
n
d
b
a
c
k
p
ro
p
a
g
a
ti
o
n
n
e
u
ra
l
n
e
tw
o
rk
.
I
m
a
g
e
p
ro
c
e
ss
in
g
is
u
se
d
to
o
b
tain
b
in
a
ry
im
a
g
e
c
o
n
sists
o
f
a
n
o
rm
a
li
z
a
ti
o
n
,
g
ra
y
sc
a
li
n
g
,
e
d
g
e
d
e
te
c
ti
o
n
a
n
d
th
re
sh
o
l
d
i
n
g
,
w
h
il
e
th
e
b
a
c
k
p
ro
p
a
g
a
ti
o
n
n
e
u
ra
l
n
e
tw
o
rk
a
lg
o
rit
h
m
is
u
se
d
f
o
r
c
las
s
i
fy
in
g
.
T
h
e
c
o
n
c
lu
sio
n
o
f
th
is
tes
t
th
a
t
th
e
a
lg
o
rit
h
m
is
a
b
le
to
p
r
o
v
id
e
th
e
a
c
c
u
ra
c
y
r
a
te
o
f
8
3
%
.
T
h
e
re
su
lt
s
o
f
th
is
re
se
a
rc
h
m
a
y
c
o
n
tri
b
u
te
to
t
h
e
d
e
v
e
lo
p
m
e
n
t
o
f
ro
a
d
d
a
m
a
g
e
d
e
t
e
c
ti
o
n
sy
ste
m
b
a
se
d
o
n
th
e
d
ig
it
a
l
im
a
g
e
so
th
a
t
th
e
traff
ic ac
c
id
e
n
ts
c
a
u
s
e
d
b
y
ro
a
d
d
a
m
a
g
e
c
a
n
b
e
re
d
u
c
e
d
.
K
ey
w
o
r
d
:
B
ac
k
p
r
o
p
ag
atio
n
n
eu
r
al
n
et
w
o
r
k
C
las
s
i
f
icatio
n
I
m
ag
e
p
r
o
ce
s
s
i
n
g
R
o
ad
d
am
a
g
e
T
r
af
f
ic
ac
cid
en
ts
Co
p
y
rig
h
t
©
2
0
1
7
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Su
ti
k
n
o
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
/I
n
f
o
r
m
at
ics
,
Dip
o
n
eg
o
r
o
Un
i
v
er
s
it
y
,
P
r
o
f
.
So
ed
ar
t
o
Stre
et,
T
em
b
al
an
g
,
Se
m
ar
a
n
g
,
I
n
d
o
n
e
s
ia
.
E
m
ail:
ti
k
@
u
n
d
ip
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
On
e
o
f
t
h
e
b
i
g
g
e
s
t
ca
u
s
e
s
o
f
d
ea
th
i
n
t
h
e
w
o
r
ld
i
s
t
h
e
tr
af
f
ic
ac
cid
en
t.
Data
o
b
tai
n
ed
f
r
o
m
t
h
e
W
o
r
ld
Hea
lth
Or
g
a
n
izatio
n
(
W
HO)
i
n
2
0
1
2
s
h
o
w
s
t
h
e
b
i
g
g
est
n
u
m
b
er
s
o
f
d
ea
t
h
f
o
r
p
eo
p
le
ag
e
d
1
5
to
1
9
y
ea
r
s
o
ld
is
ca
u
s
ed
b
y
tr
af
f
ic
ac
cid
en
t
s
.
T
h
e
n
u
m
b
er
o
f
p
eo
p
le
d
ied
d
u
e
to
th
e
tr
af
f
ic
ac
cid
e
n
t
in
2
0
1
3
r
ea
ch
ed
1
.
2
5
b
illi
o
n
.
I
t
is
esti
m
a
ted
th
at
t
h
e
n
u
m
b
er
w
i
ll
in
cr
ea
s
e
t
h
r
ee
f
o
ld
in
2
0
3
0
w
h
ic
h
b
ec
a
m
e
3
.
6
b
il
lio
n
ea
ch
y
ea
r
[
1
]
.
On
e
o
f
t
h
e
f
ac
to
r
s
t
h
at
ca
u
s
e
tr
af
f
ic
ac
cid
en
t
is
b
ad
r
o
ad
co
n
d
itio
n
,
e.
g
.
th
e
d
a
m
a
g
ed
r
o
ad
s
,
s
lip
p
er
y
r
o
ad
s
,
w
et
r
o
ad
s
an
d
s
n
o
wy
r
o
ad
s
[
2
]
.
So
m
e
ca
u
s
es
o
f
t
h
e
d
est
r
u
ctio
n
o
f
th
i
s
r
o
ad
ar
e
th
e
in
cr
ea
s
e
o
f
tr
af
f
ic
lo
ad
v
o
lu
m
e,
b
ad
d
r
ain
ag
e
s
y
s
te
m
,
b
ad
p
av
em
e
n
t c
o
n
s
tr
u
ct
io
n
m
a
ter
ial,
cli
m
ate
a
n
d
u
n
s
tab
le
s
o
i
l c
o
n
d
itio
n
[
3
]
.
Fro
m
th
a
t
p
r
o
b
lem
s
,
to
r
ed
u
c
e
tr
af
f
ic
ac
cid
en
t
t
h
at
is
ca
u
s
ed
b
y
r
o
ad
d
a
m
ag
e
is
r
eq
u
ir
e
d
an
ea
r
l
y
d
etec
t
ab
o
u
t
th
e
r
o
ad
d
am
ag
e
.
T
h
er
e
ar
e
s
ev
er
al
w
a
y
s
to
d
etec
t
r
o
ad
d
am
a
g
e,
w
h
ic
h
is
an
al
y
z
in
g
th
e
r
o
ad
co
n
d
itio
n
s
b
ased
o
n
t
h
e
i
m
a
g
e
an
d
an
al
y
zi
n
g
t
h
e
s
i
g
n
al
f
r
o
m
th
e
r
es
u
lt
s
o
f
v
ib
r
atio
n
o
n
th
e
s
tr
ee
t
t
h
at
is
p
ass
ed
[
4
-
9
]
.
Dete
ctin
g
t
h
e
co
n
d
itio
n
o
f
r
o
ad
d
am
a
g
e
b
ased
o
n
i
m
a
g
es
ca
n
b
e
d
o
n
e
b
y
th
e
i
m
a
g
e
s
h
o
o
tin
g
a
t
lo
n
g
a
n
d
clo
s
e
d
is
ta
n
ce
.
I
m
a
g
e
s
h
o
o
ti
n
g
at
lo
n
g
d
is
ta
n
ce
s
c
an
b
e
ta
k
en
f
r
o
m
t
h
e
s
a
tellite.
T
h
is
tec
h
n
iq
u
e
ca
n
u
s
e
s
e
v
er
al
m
eth
o
d
s
:
a
k
n
o
w
led
g
e
-
b
a
s
ed
,
d
e
m
s
ter
s
h
a
f
er
t
h
eo
r
y
,
k
u
llb
ac
k
-
l
eib
ler
a
nd
n
ea
r
e
s
t
n
eig
h
b
o
r
[
5
-
9
]
.
Oth
er
r
esear
ch
to
class
i
f
y
r
o
ad
d
am
a
g
e
ca
n
b
e
d
o
n
e
b
y
an
al
y
zi
n
g
t
h
e
s
ig
n
al
co
n
d
itio
n
o
f
t
h
e
r
o
ad
.
T
h
is
tech
n
iq
u
e
ca
n
u
s
e
s
e
v
e
r
al
m
et
h
o
d
s
:
t
h
e
v
ib
r
atio
n
te
ch
n
iq
u
e,
t
h
e
Si
g
n
als
o
f
S
m
a
r
tp
h
o
n
e
-
E
m
b
ed
d
ed
A
cc
eler
o
m
eter
an
d
w
a
v
elet
tr
an
s
f
o
r
m
tec
h
n
iq
u
e
[
1
0
-
1
2
]
.
On
t
h
e
u
s
e
o
f
v
ib
r
atio
n
m
e
th
o
d
,
th
e
r
o
ad
is
d
i
v
id
ed
in
to
t
h
r
ee
class
e
s
,
n
a
m
el
y
n
o
r
m
al,
b
u
m
p
an
d
p
o
th
o
le
[
1
0
]
.
On
t
h
e
u
s
e
o
f
Si
g
n
al
s
o
f
S
m
a
r
tp
h
o
n
e
-
E
m
b
ed
d
ed
A
cc
eler
o
m
eter
m
et
h
o
d
,
r
o
ad
co
n
d
itio
n
s
ar
e
d
iv
id
ed
in
to
f
o
u
r
r
o
ad
co
n
d
itio
n
s
,
n
a
m
el
y
f
lat,
p
o
s
iti
v
e
s
te
p
d
am
a
g
e
(
P
S),
n
e
g
ati
v
e
s
tep
d
a
m
ag
e
(
NS)
,
a
n
d
co
n
v
ex
s
tep
d
a
m
ag
e
(
C
S)
w
h
ich
g
e
n
er
at
es
ac
cu
r
ac
y
r
ate
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
4
,
Dec
em
b
er
201
7
:
1
5
9
–
1
6
5
160
7
0
%
[
1
1
]
.
W
h
ile
th
e
u
s
e
o
f
w
a
v
elet
tr
an
s
f
o
r
m
tec
h
n
iq
u
e,
r
o
ad
co
n
d
itio
n
s
ar
e
d
iv
id
ed
in
to
t
h
r
ee
co
n
d
itio
n
s
,
n
a
m
e
l
y
p
o
s
iti
v
e
s
tep
(
P
S),
n
eg
ati
v
e
s
tep
(
NS)
an
d
co
n
v
e
x
s
tep
(
C
S)
w
h
ic
h
g
e
n
e
r
ates
th
e
b
est
ac
cu
r
ac
y
r
ate
o
f
8
8
% [
1
2
]
.
I
n
th
o
s
e
p
ap
er
s
,
th
e
r
o
ad
co
n
d
itio
n
s
ar
e
b
ased
o
n
th
e
r
o
ad
s
ig
n
al
t
h
at
is
p
ass
ed
a
n
d
n
o
t
b
ased
o
n
ac
tu
al
r
o
ad
co
n
d
itio
n
s
,
s
o
it
is
p
o
s
s
ib
le
t
h
at
t
h
er
e
i
s
a
ch
a
n
g
ed
er
r
o
r
f
r
o
m
r
o
ad
co
n
d
itio
n
t
o
th
e
f
o
r
m
at
io
n
o
f
a
s
ig
n
al.
T
o
k
n
o
w
th
e
r
ea
l
co
n
d
itio
n
o
f
th
e
r
o
ad
,
th
e
d
ata
u
s
ed
ca
n
b
e
in
th
e
f
o
r
m
o
f
d
ig
ita
l
i
m
ag
e
d
ata
w
h
ic
h
th
en
p
r
o
ce
s
s
i
n
f
u
r
t
h
er
p
r
o
ce
s
s
in
g
p
r
o
ce
s
s
.
I
m
a
g
e
p
r
o
ce
s
s
in
g
h
a
s
b
ee
n
w
id
el
y
u
s
ed
f
o
r
th
e
id
en
t
if
icatio
n
a
n
d
d
etec
tio
n
o
f
o
b
j
ec
ts
,
w
h
ic
h
ar
e
r
ice
v
ar
iet
y
id
en
ti
f
icatio
n
,
b
io
m
etr
ic
p
er
s
o
n
al
id
en
t
if
ica
tio
n
,
w
ater
t
u
r
b
id
it
y
d
etec
tio
n
,
an
d
co
tto
n
d
is
ea
s
es
d
etec
tio
n
[
1
3
-
1
6
]
.
T
h
is
tech
n
iq
u
e
h
as
b
ee
n
d
o
n
e
to
class
i
f
y
r
o
ad
d
am
a
g
e
o
n
d
ig
ital
i
m
a
g
es
b
y
u
s
in
g
r
e
g
io
n
s
p
li
t
m
er
g
er
a
n
d
f
r
ac
tal
d
i
m
en
s
io
n
.
T
h
e
ac
cu
r
ac
y
r
ate
t
h
at
i
s
g
en
er
ated
b
y
u
s
i
n
g
r
eg
io
n
s
p
lit
m
er
g
er
i
s
6
1
,7
% a
n
d
u
s
in
g
f
r
ac
tal
d
i
m
e
n
s
i
o
n
is
8
2
,
9
% [
1
7
]
.
T
h
is
p
ap
er
p
r
o
p
o
s
es c
lass
i
f
y
i
n
g
r
o
ad
d
am
ag
e
b
y
m
ea
n
s
o
f
i
m
a
g
e
p
r
o
ce
s
s
in
g
a
n
d
b
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
et
w
o
r
k
.
I
m
a
g
e
p
r
o
ce
s
s
in
g
is
u
s
ed
to
o
b
tain
a
b
in
ar
y
i
m
ag
e
co
n
s
i
s
ti
n
g
o
f
a
p
r
o
ce
s
s
o
f
n
o
r
m
a
lizati
o
n
,
g
r
a
y
s
ca
li
n
g
,
ed
g
e
d
etec
tio
n
,
an
d
th
r
e
s
h
o
ld
i
n
g
,
w
h
ile
t
h
e
b
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
et
w
o
r
k
al
g
o
r
ith
m
is
u
s
e
d
f
o
r
class
if
y
i
n
g
.
T
h
is
alg
o
r
it
h
m
h
as
b
ee
n
w
id
e
l
y
u
s
ed
to
cla
s
s
i
f
y
w
h
ic
h
g
en
er
a
tes
t
h
e
f
air
l
y
g
o
o
d
d
eg
r
ee
o
f
ac
cu
r
ac
y
r
ate,
a
m
o
n
g
o
f
it
s
a
r
e
to
class
i
f
y
b
ati
k
m
o
ti
f
,
b
r
ain
ca
n
ce
r
,
h
ar
u
m
m
an
is
m
an
g
o
,
r
ea
l
-
ti
m
e
i
s
ch
e
m
ic
b
ea
t,
m
o
v
in
g
v
e
h
icle
n
o
is
e,
an
d
g
e
n
d
er
[
1
8
-
2
3
]
.
T
h
e
r
esu
lt
s
o
f
th
is
r
esear
ch
m
a
y
co
n
t
r
ib
u
te
to
t
h
e
d
ev
elo
p
m
e
n
t
o
f
r
o
ad
d
a
m
a
g
e
d
etec
tio
n
s
y
s
te
m
b
ased
o
n
th
e
d
ig
ital
i
m
a
g
e
ca
p
tu
r
ed
b
y
t
h
e
ca
m
er
a
t
h
at
is
p
air
ed
to
th
e
v
eh
icle
s
o
th
a
t
t
h
e
tr
af
f
ic
ac
cid
en
t
s
ca
u
s
ed
b
y
r
o
ad
d
am
a
g
e
ca
n
b
e
r
ed
u
ce
d
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
Sy
s
t
e
m
P
ro
ce
s
s
S
y
s
te
m
p
r
o
ce
s
s
t
h
at
w
as
d
o
n
e
in
t
h
i
s
r
esear
ch
,
g
e
n
er
all
y
d
iv
id
ed
in
to
t
w
o
s
u
b
-
p
r
o
ce
s
s
,
n
a
m
el
y
s
u
b
p
r
o
ce
s
s
tr
ain
i
n
g
a
n
d
s
u
b
p
r
o
ce
s
s
test
in
g
.
E
ac
h
s
u
b
p
r
o
ce
s
s
th
at
s
h
o
w
n
i
n
Fi
g
u
r
e
1
.
Su
b
p
r
o
ce
s
s
tr
ain
in
g
co
n
s
is
tin
g
o
f
r
esizi
n
g
,
g
r
a
y
s
ca
lin
g
,
ed
g
e
d
etec
tio
n
,
t
h
r
es
h
o
ld
in
g
an
d
tr
ai
n
i
n
g
.
S
u
b
p
r
o
ce
s
s
t
esti
n
g
co
n
s
i
s
ti
n
g
o
f
r
esizin
g
,
g
r
a
y
s
ca
li
n
g
,
ed
g
e
d
e
tectio
n
,
th
r
es
h
o
ld
in
g
,
an
d
test
in
g
.
T
h
e
p
r
o
ce
s
s
o
f
tr
ain
in
g
a
n
d
test
i
n
g
is
u
s
i
n
g
b
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
et
w
o
r
k
.
Fig
u
r
e
1
.
Flo
w
c
h
ar
t o
f
s
y
s
te
m
p
r
o
ce
s
s
E
x
p
lan
atio
n
o
f
ea
ch
p
r
o
ce
s
s
i
s
as f
o
llo
w
s
:
1.
R
esize
T
h
e
in
p
u
t
o
f
t
h
i
s
r
esize
s
p
r
o
ce
s
s
i
s
t
h
e
i
m
a
g
e
co
n
d
itio
n
o
f
d
am
a
g
ed
r
o
ad
an
d
n
o
t
d
a
m
a
g
ed
w
it
h
a
v
ar
ie
t
y
o
f
s
izes.
T
h
is
p
r
o
ce
s
s
ch
an
g
e
s
th
e
i
m
a
g
e
o
f
v
ar
io
u
s
s
izes
i
n
to
5
0
x
5
0
p
ix
els
s
ized
im
a
g
e
.
T
h
is
i
m
a
g
e
is
u
s
ed
f
o
r
tr
ain
i
n
g
an
d
tes
tin
g
.
2.
Gr
a
y
s
ca
li
n
g
I
m
ag
e
f
r
o
m
t
h
e
r
es
u
lt
o
f
r
esiz
e
p
r
o
ce
s
s
is
an
i
m
ag
e
w
i
th
a
c
o
lo
r
r
ep
r
esen
tatio
n
o
f
R
GB
(
R
ed
Gr
ee
n
B
lu
e)
.
Gr
a
y
s
ca
li
n
g
p
r
o
ce
s
s
is
u
s
ed
t
o
s
i
m
p
li
f
y
R
GB
co
lo
r
i
m
a
g
e
in
to
th
e
i
m
a
g
e
o
f
eig
h
t
b
it
s
o
r
2
5
6
p
r
im
ar
y
co
lo
r
s
.
3.
E
d
g
e
Dete
ctio
n
E
d
g
e
Dete
ctio
n
is
t
h
e
i
n
te
n
s
it
y
o
f
g
r
a
y
d
eg
r
ee
t
h
at
is
s
u
d
d
en
l
y
ch
a
n
g
ed
in
a
s
h
o
r
t d
is
tan
c
e.
T
h
e
p
u
r
p
o
s
e
o
f
th
is
p
r
o
ce
s
s
is
to
en
h
a
n
ce
t
h
e
ap
p
ea
r
an
ce
o
f
th
e
b
o
u
n
d
ar
y
li
n
e
b
et
w
ee
n
t
h
e
i
m
a
g
e
p
ar
t th
at
is
d
a
m
a
g
ed
an
d
n
o
t d
a
m
ag
ed
.
T
h
e
o
p
er
ato
r
th
at
u
s
ed
in
t
h
i
s
ed
g
e
d
etec
ti
o
n
i
s
th
e
s
o
b
el
o
p
er
ato
r
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
la
s
s
i
fica
tio
n
o
f R
o
a
d
Da
ma
g
e
fr
o
m
Dig
ita
l I
ma
g
e
Usi
n
g
…
(
S
u
tikn
o
)
161
4.
T
h
r
esh
o
ld
in
g
T
h
r
esh
o
ld
in
g
is
u
s
ed
to
s
et
t
h
e
a
m
o
u
n
t
o
f
g
r
a
y
d
eg
r
ee
at
an
i
m
a
g
e.
T
h
e
r
esu
l
t
f
r
o
m
t
h
i
s
p
r
o
ce
s
s
is
th
e
b
in
ar
y
i
m
a
g
e.
T
h
e
b
in
ar
y
i
m
ag
e
i
s
a
n
i
m
a
g
e
t
h
at
h
as
t
w
o
v
al
u
es,
n
a
m
el
y
b
lac
k
a
n
d
wh
ite.
T
h
e
b
i
n
ar
y
i
m
a
g
e
t
h
at
h
as
b
ee
n
g
e
n
er
ated
is
s
to
r
ed
in
th
e
d
atab
ase
to
d
o
th
e
tr
ai
n
i
n
g
p
r
o
ce
s
s
.
E
x
a
m
p
les
t
h
e
c
h
an
g
e
s
o
f
r
o
ad
d
a
m
ag
e
i
m
ag
e
at
a
p
r
o
ce
s
s
o
f
g
r
a
y
s
ca
li
n
g
,
ed
g
e
d
etec
tio
n
an
d
th
r
es
h
o
ld
in
g
ca
n
b
e
s
h
o
w
n
i
n
Fig
u
r
e
2
.
Fig
u
r
e
2
.
T
h
e
ch
an
g
e
s
o
f
r
o
ad
d
am
a
g
e
i
m
a
g
e
at
p
r
o
ce
s
s
o
f
g
r
a
y
s
ca
ll
in
g
,
ed
g
e
d
etec
tio
n
a
n
d
th
r
es
h
o
ld
in
g
5.
T
r
ain
in
g
C
las
s
i
f
icatio
n
al
g
o
r
ith
m
t
h
at
i
s
u
s
ed
in
t
h
i
s
r
esear
ch
is
u
s
i
n
g
b
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
et
w
o
r
k
al
g
o
r
ith
m
.
T
h
is
alg
o
r
ith
m
n
ee
d
s
t
h
e
tr
ai
n
in
g
b
e
f
o
r
e
d
o
in
g
th
e
te
s
ti
n
g
.
T
h
e
in
p
u
t
f
r
o
m
tr
ai
n
i
n
g
p
r
o
ce
s
s
is
t
h
e
p
atter
n
o
f
th
e
b
in
ar
y
i
m
a
g
e
th
at
h
as
b
ee
n
s
to
r
ed
in
th
e
d
atab
ase
an
d
th
e
r
esu
lt
o
f
th
i
s
p
r
o
ce
s
s
is
w
ei
g
h
ted
in
p
u
t
n
et
w
o
r
k
la
y
er
to
w
ar
d
s
h
id
d
en
la
y
er
an
d
w
ei
g
h
ts
n
et
w
o
r
k
f
r
o
m
h
id
d
en
la
y
er
to
w
ar
d
s
o
u
tp
u
t
la
y
er
.
T
h
e
r
esu
lt
f
r
o
m
t
h
is
w
ei
g
h
ts
i
s
s
to
r
ed
in
th
e
d
atab
ase.
6.
T
esti
n
g
T
h
e
test
in
g
p
r
o
ce
s
s
i
s
t
h
e
las
t
p
r
o
ce
s
s
f
r
o
m
s
y
s
te
m
p
r
o
ce
s
s
o
f
th
is
r
esear
c
h
.
T
h
e
in
p
u
t
f
r
o
m
t
h
is
p
r
o
ce
s
s
is
w
ei
g
h
ts
t
h
at
r
es
u
lt
in
g
f
r
o
m
tr
ain
i
n
g
p
r
o
ce
s
s
a
n
d
test
i
n
g
i
m
a
g
e
th
a
t
h
a
s
b
ee
n
d
o
n
e
r
esize
p
r
o
ce
s
s
,
g
r
a
y
s
ca
lin
g
,
ed
g
e
d
etec
tio
n
a
n
d
th
r
es
h
o
ld
in
g
.
T
h
e
r
esu
lts
f
r
o
m
test
in
g
p
r
o
ce
s
s
ar
e
to
t
est
th
e
i
m
ag
e
,
w
h
et
h
er
en
ter
ed
in
to
d
a
m
a
g
ed
r
o
ad
class
o
r
n
o
t.
2
.
2
B
a
ck
pro
pa
g
a
t
io
n Ne
ura
l N
et
w
o
rk
Arc
hite
ct
ure
C
las
s
i
f
icatio
n
p
r
o
ce
s
s
o
f
r
o
a
d
d
am
a
g
e
i
n
t
h
i
s
r
esear
ch
is
u
s
i
n
g
b
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
et
w
o
r
k
alg
o
r
ith
m
.
T
h
is
n
eu
r
al
n
et
w
o
r
k
ar
ch
itec
tu
r
e
co
n
s
is
ts
o
f
2
5
0
0
in
p
u
ts
,
o
n
e
h
id
d
en
la
y
er
wh
ich
co
n
s
i
s
ts
o
f
1
0
n
eu
r
o
n
s
an
d
1
o
u
tp
u
t a
s
d
is
p
la
y
ed
in
Fi
g
u
r
e
3
.
T
h
e
ac
tiv
atio
n
f
u
n
ctio
n
u
s
ed
is
a
b
in
ar
y
s
i
g
m
o
id
.
Fig
u
r
e
3
.
B
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
et
w
o
r
k
ar
c
h
itect
u
r
e
in
t
h
e
r
esear
ch
3.
RE
SU
L
T
S
A
ND
D
I
SCU
SS
I
O
N
T
h
is
r
esear
ch
is
co
n
d
u
cted
test
i
n
g
b
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
et
w
o
r
k
alg
o
r
it
h
m
to
class
i
f
y
t
h
e
co
n
d
itio
n
o
f
d
a
m
a
g
ed
r
o
ad
an
d
g
o
o
d
r
o
ad
.
I
n
th
e
tr
ai
n
in
g
p
r
o
ce
s
s
is
g
i
v
en
th
e
f
i
x
ed
i
n
p
u
t
v
ar
i
ab
le
v
al
u
e
t
h
at
is
ep
o
ch
b
o
u
n
d
ar
y
o
f
1
0
.
0
0
0
an
d
er
r
o
r
b
o
u
n
d
ar
y
o
f
0
.
0
0
0
1
.
W
h
ile
t
h
e
v
alu
e
o
f
i
n
p
u
t
v
ar
iab
les
f
o
r
lear
n
in
g
r
ate
an
d
th
e
a
m
o
u
n
t
o
f
tr
ain
in
g
d
ata
is
g
i
v
e
n
v
ar
ied
.
L
ea
r
n
i
n
g
r
ate
v
ar
iatio
n
v
alu
e
s
t
h
at
a
r
e
g
iv
e
n
ar
e
0
.
1
,
0
.
5
an
d
0
.
9
.
W
h
ile
th
e
a
m
o
u
n
t
o
f
tr
ain
in
g
d
ata
th
at
ar
e
g
iv
e
n
ar
e
7
5
,
1
5
0
,
2
2
5
an
d
3
0
0
d
ata.
T
h
e
r
esu
lts
o
f
w
ei
g
h
ts
f
r
o
m
n
et
w
o
r
k
tr
ain
in
g
o
n
ea
c
h
v
ar
iatio
n
o
f
th
e
s
e
v
ar
iab
les
i
s
s
to
r
ed
in
t
h
e
d
at
ab
ase
an
d
u
s
ed
f
o
r
test
i
n
g
r
o
ad
d
am
a
g
e
cla
s
s
i
f
ica
tio
n
.
Gr
a
y
s
c
a
li
ng
E
d
g
e
De
tec
t
i
o
n
T
h
r
e
s
h
o
ld
ing
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
4
,
Dec
em
b
er
201
7
:
1
5
9
–
1
6
5
162
T
h
e
d
ata
u
s
ed
in
th
e
test
i
n
g
co
n
s
is
ts
o
f
t
w
o
r
o
ad
co
n
d
itio
n
s
,
n
a
m
e
l
y
g
o
o
d
r
o
a
d
an
d
d
am
ag
ed
r
o
ad
.
T
h
e
g
o
o
d
r
o
ad
is
d
iv
id
ed
in
to
t
w
o
,
th
ese
ar
e
u
n
m
ar
k
ed
g
o
o
d
r
o
ad
an
d
g
o
o
d
r
o
ad
m
ar
k
in
g
.
T
h
e
a
m
o
u
n
t
o
f
d
ata
u
s
ed
in
th
i
s
r
esear
ch
is
3
0
d
ata,
as
s
h
o
w
n
i
n
T
ab
le
1
.
T
h
e
r
esu
lts
o
f
test
i
n
g
o
n
ea
c
h
lear
n
in
g
r
ate
v
ar
iatio
n
i
s
s
h
o
w
n
i
n
T
ab
le
2
,
T
ab
le
3
,
an
d
T
ab
le
4
.
T
ab
le
1
.
T
esti
n
g
Data
U
n
mar
k
e
d
G
o
o
d
R
o
a
d
G
o
o
d
R
o
a
d
M
a
r
k
i
n
g
s
D
a
mag
e
d
R
o
a
d
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
la
s
s
i
fica
tio
n
o
f R
o
a
d
Da
ma
g
e
fr
o
m
Dig
ita
l I
ma
g
e
Usi
n
g
…
(
S
u
tikn
o
)
163
T
ab
le
2
.
P
er
f
o
r
m
a
n
ce
o
f
b
ac
k
p
r
o
p
ag
atio
n
n
eu
r
al
n
et
w
o
r
k
with
lear
n
in
g
r
ate
0
.
9
T
h
e
A
m
o
u
n
t
o
f
T
r
a
i
n
i
n
g
D
a
t
a
T
h
e
A
c
c
u
r
a
c
y
o
f
U
n
mark
e
d
G
o
o
d
R
o
a
d
(
%)
T
h
e
A
c
c
u
r
a
c
y
o
f
G
o
o
d
R
o
a
d
M
a
r
k
i
n
g
s (
%)
T
h
e
A
c
c
u
r
a
c
y
o
f
D
a
mag
e
d
R
o
a
d
(
%)
A
v
e
r
a
g
e
(
%)
75
1
0
0
30
60
63
1
5
0
1
0
0
60
60
73
2
2
5
1
0
0
60
60
73
3
0
0
1
0
0
60
60
73
T
ab
le
3
.
P
er
f
o
r
m
a
n
ce
o
f
b
ac
k
p
r
o
p
ag
atio
n
n
eu
r
al
n
et
w
o
r
k
with
lear
n
in
g
r
ate
0
.
5
T
h
e
A
m
o
u
n
t
o
f
T
r
a
i
n
i
n
g
D
a
t
a
T
h
e
A
c
c
u
r
a
c
y
o
f
U
n
mark
e
d
G
o
o
d
R
o
a
d
(
%)
T
h
e
A
c
c
u
r
a
c
y
o
f
G
o
o
d
R
o
a
d
M
a
r
k
i
n
g
s (
%)
T
h
e
A
c
c
u
r
a
c
y
o
f
D
a
mag
e
d
R
o
a
d
(
%)
A
v
e
r
a
g
e
(
%)
75
1
0
0
50
50
67
1
5
0
1
0
0
60
70
77
2
2
5
1
0
0
60
70
77
3
0
0
1
0
0
60
70
77
T
ab
le
4
.
P
er
f
o
r
m
a
n
ce
o
f
b
ac
k
p
r
o
p
ag
atio
n
n
eu
r
al
n
et
w
o
r
k
with
lear
n
in
g
r
ate
0
.
1
T
h
e
A
m
o
u
n
t
o
f
T
r
a
i
n
i
n
g
D
a
t
a
T
h
e
A
c
c
u
r
a
c
y
o
f
U
n
mark
e
d
G
o
o
d
R
o
a
d
(
%)
T
h
e
A
c
c
u
r
a
c
y
o
f
G
o
o
d
R
o
a
d
M
a
r
k
i
n
g
s (
%)
T
h
e
A
c
c
u
r
a
c
y
o
f
D
a
mag
e
d
R
o
a
d
(
%)
A
v
e
r
a
g
e
(
%)
75
1
0
0
20
50
57
1
5
0
1
0
0
40
70
70
2
2
5
1
0
0
70
70
80
3
0
0
1
0
0
80
70
83
Fro
m
T
ab
le
2
,
T
ab
le
3
,
an
d
T
ab
le
4
s
h
o
w
t
h
at
th
e
b
est
ac
c
u
r
ac
y
is
o
b
tain
ed
b
y
8
3
%.
T
h
is
v
al
u
e
is
o
b
tain
ed
at
t
h
e
ti
m
e
g
i
v
en
lear
n
in
g
r
ate
v
al
u
e
o
f
0
.
1
an
d
th
e
a
m
o
u
n
t
o
f
tr
ain
i
n
g
d
ata
b
y
3
0
0
.
T
h
e
alg
o
r
ith
m
is
ab
le
to
r
ec
o
g
n
ize
all
u
n
m
ar
k
e
d
g
o
o
d
r
o
ad
w
it
h
a
n
ac
c
u
r
ac
y
r
ate
o
f
1
0
0
%.
T
h
is
is
d
u
e
to
g
o
o
d
r
o
ad
d
ata
th
at
u
s
ed
i
n
th
e
p
r
o
ce
s
s
o
f
tr
ain
in
g
a
n
d
tes
tin
g
h
a
v
e
al
m
o
s
t
t
h
e
s
a
m
e
co
lo
r
,
s
o
t
h
e
al
g
o
r
ith
m
is
ab
le
to
id
en
t
if
y
ea
s
il
y
.
B
e
s
id
es
t
h
at,
t
h
i
s
al
g
o
r
ith
m
is
ab
le
to
r
ec
o
g
n
ize
g
o
o
d
r
o
ad
m
ar
k
i
n
g
s
w
it
h
t
h
e
b
e
s
t
ac
cu
r
ac
y
r
ate
o
f
8
0
%
an
d
th
is
al
g
o
r
ith
m
i
s
ab
le
to
r
ec
o
g
n
ize
d
a
m
ag
ed
r
o
ad
w
it
h
t
h
e
b
est
ac
cu
r
ac
y
r
ate
o
f
7
0
%.
T
h
ese
r
esu
lt
s
ar
e
d
u
e
to
th
e
alg
o
r
ith
m
th
a
t
u
s
ed
h
a
s
n
o
t
b
ee
n
ab
le
to
p
r
o
p
er
ly
d
is
t
in
g
u
is
h
b
et
w
ee
n
t
h
e
m
ar
k
i
n
g
lin
e
s
w
it
h
th
e
d
a
m
a
g
ed
r
o
ad
s
o
th
at
th
e
m
ar
k
i
n
g
li
n
es i
n
th
e
g
o
o
d
r
o
ad
is
co
n
s
id
er
ed
d
a
m
ag
ed
r
o
ad
.
Fro
m
th
e
te
s
ti
n
g
r
es
u
lt
s
s
h
o
w
n
th
at
t
h
e
m
o
r
e
t
h
e
ad
d
itio
n
s
o
f
v
ar
y
i
n
g
a
m
o
u
n
ts
o
f
tr
a
i
n
i
n
g
d
ata
w
i
ll
en
h
a
n
ce
th
e
ac
c
u
r
ac
y
r
ate
th
at
is
g
en
er
ated
.
T
h
is
is
d
u
e
to
th
e
v
ar
iatio
n
s
d
ata
th
at
ar
e
ad
d
e
d
im
p
r
o
v
es
w
ei
g
h
ts
o
f
tr
ain
in
g
r
es
u
lt
s
i
n
d
is
t
in
g
u
i
s
h
i
n
g
b
et
w
ee
n
d
a
m
a
g
ed
r
o
ad
an
d
g
o
o
d
r
o
ad
.
Oth
er
th
an
t
h
at,
t
h
e
s
m
al
ler
th
e
ad
d
ed
v
alu
e
o
f
lear
n
i
n
g
r
ate
h
as
a
ten
d
en
c
y
ca
n
i
n
cr
ea
s
e
th
e
ac
cu
r
ac
y
r
ate
b
ec
au
s
e
ad
d
in
g
g
r
ea
t
lear
n
in
g
r
ate
v
alu
e
h
as a
te
n
d
en
c
y
w
ill r
u
in
th
e
p
atter
n
o
f
tr
ai
n
i
n
g
d
ata.
T
h
e
r
esu
lt
s
o
f
tes
tin
g
in
t
h
is
r
esear
ch
co
m
p
ar
ed
to
o
t
h
er
r
es
ea
r
ch
i
n
cla
s
s
i
f
y
in
g
r
o
ad
d
a
m
ag
e
ca
n
b
e
s
ee
n
as
i
n
T
ab
le
5
.
T
h
e
alg
o
r
it
h
m
u
s
ed
i
n
th
i
s
r
esear
ch
w
er
e
co
m
p
ar
ed
u
s
in
g
t
h
e
m
et
h
o
d
r
eg
io
n
s
p
li
t
m
er
g
er
s
an
d
f
r
ac
tal
d
i
m
e
n
s
io
n
[
1
7
]
.
Fr
o
m
t
h
i
s
tab
le
s
h
o
w
s
th
at
b
ac
k
p
r
o
p
ag
atio
n
n
eu
r
al
n
et
w
o
r
k
al
g
o
r
ith
m
u
s
ed
in
t
h
i
s
r
esear
ch
ca
n
in
cr
ea
s
e
th
e
cla
s
s
i
f
icatio
n
r
es
u
lts
th
a
n
u
s
i
n
g
th
e
al
g
o
r
ith
m
o
f
r
eg
io
n
s
p
lit
-
m
er
g
er
a
n
d
f
r
ac
tal
d
i
m
en
s
io
n
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
i
s
alg
o
r
ith
m
ca
n
b
e
in
cr
ea
s
ed
b
y
ad
d
in
g
a
v
ar
iet
y
o
f
tr
ain
i
n
g
d
ata
b
ec
au
s
e
th
e
i
n
cr
ea
s
ed
a
v
ar
iet
y
o
f
tr
ai
n
in
g
d
ata
w
ill i
m
p
r
o
v
e
t
h
e
w
ei
g
h
ts
o
f
tr
ain
in
g
t
h
at
a
r
e
r
esu
lti
n
g
in
d
is
ti
n
g
u
i
s
h
i
n
g
d
am
a
g
ed
r
o
ad
an
d
g
o
o
d
r
o
a
d
.
T
ab
le
5
.
C
o
m
p
ar
is
o
n
o
f
alg
o
r
i
th
m
s
f
o
r
class
i
f
icatio
n
o
f
r
o
ad
d
am
a
g
e
A
l
g
o
r
i
t
h
m
T
h
e
R
e
su
l
t
s o
f
A
c
c
u
r
a
c
y
R
a
t
e
(
%)
R
e
g
i
o
n
sp
l
i
t
-
me
r
g
e
r
6
1
.
7
F
r
a
c
t
a
l
d
i
me
n
si
o
n
8
2
.
9
I
mag
e
p
r
o
c
e
ssi
n
g
a
n
d
b
a
c
k
p
r
o
p
a
g
a
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
8
3
.
0
4.
CO
NCLU
SI
O
N
C
o
n
cl
u
s
io
n
o
b
tain
ed
f
r
o
m
th
i
s
r
esear
ch
i
s
r
o
ad
d
am
a
g
e
cl
ass
i
f
icatio
n
u
s
in
g
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
a
n
d
b
ac
k
p
r
o
p
ag
atio
n
n
eu
r
al
n
et
w
o
r
k
al
g
o
r
ith
m
p
r
o
v
id
es
ac
cu
r
ac
y
r
ate
o
f
8
3
%.
T
h
is
alg
o
r
ith
m
p
r
o
v
id
es
en
h
a
n
ce
m
en
t
ac
c
u
r
ac
y
r
esu
lts
co
m
p
ar
ed
w
it
h
t
h
e
u
s
e
o
f
t
h
e
alg
o
r
it
h
m
o
f
r
eg
io
n
s
p
lit
–
m
er
g
er
a
n
d
f
r
ac
tal
d
i
m
en
s
io
n
.
T
h
e
al
g
o
r
ith
m
is
a
b
le
to
r
ec
o
g
n
ize
all
t
h
e
u
n
m
ar
k
ed
g
o
o
d
r
o
ad
co
n
d
itio
n
s
,
b
u
t
o
n
l
y
p
ar
tiall
y
ab
le
to
r
ec
o
g
n
ize
th
e
g
o
o
d
r
o
ad
m
a
r
k
in
g
s
an
d
d
a
m
a
g
ed
r
o
ad
.
T
h
e
ac
cu
r
ac
y
r
ate
ca
n
b
e
i
m
p
r
o
v
e
d
b
y
i
n
cr
ea
s
i
n
g
t
h
e
a
m
o
u
n
t o
f
d
ata
v
ar
ie
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
4
,
Dec
em
b
er
201
7
:
1
5
9
–
1
6
5
164
RE
F
E
R
E
NC
E
S
[1
]
W
o
rld
He
a
lt
h
Org
a
n
iza
ti
o
n
,
“
G
lo
b
a
l
S
tatu
s
Re
p
o
rt
o
n
Ro
a
d
S
a
f
e
t
y
2
0
1
5
,
”
W
o
rl
d
He
a
lt
h
Or
g
a
n
iza
ti
o
n
(
W
HO
)
,
201
6.
[2
]
F
.
S
a
g
b
e
rg
,
“
Ro
a
d
A
c
c
id
e
n
ts
Ca
u
se
d
b
y
Driv
e
rs
F
a
ll
in
g
A
sle
e
p
,
”
Acc
id
e
n
t
A
n
a
lys
is
&
Pre
v
e
n
ti
o
n
,
v
o
l.
3
1
,
p
p
.
6
3
9
-
6
4
9
,
1
9
9
9
.
[3
]
M
a
rd
ian
u
s,
“
S
t
u
d
i
P
e
n
a
n
g
a
n
a
n
Ja
lan
b
e
rd
a
sa
rk
a
n
T
in
g
k
a
t
Ke
ru
sa
k
a
n
P
e
rk
e
ra
sa
n
J
a
lan
(S
tu
d
i
Ka
su
s:
Ja
la
n
Ku
a
la
Du
a
Ka
b
u
p
a
ten
Ku
b
u
Ra
y
a
),
”
J
u
r
n
a
l
T
e
k
n
ik S
i
p
il
UN
T
AN
,
v
o
l
.
1
3
,
p
p
.
1
4
9
-
1
6
0
,
2
0
1
3
.
[4
]
Q.
Qin
,
e
t
a
l.
,
“
Da
m
a
g
e
De
te
c
ti
o
n
a
n
d
A
ss
e
ss
m
e
n
t
S
y
ste
m
o
f
Ro
a
d
s
f
o
r
De
c
isio
n
S
u
p
p
o
rt
f
o
r
Disa
ste
r,
”
K
e
y
En
g
i
n
e
e
rin
g
M
a
ter
ia
ls
,
v
o
l
s
4
6
7
-
4
6
9
,
p
p
.
1
1
4
4
-
1
1
4
9
,
2
0
1
1
.
[5
]
J.
W
a
n
g
,
e
t
a
l.
,
“
A
Kn
o
w
led
g
e
-
B
a
se
d
M
e
th
o
d
f
o
r
Ro
a
d
Da
m
a
g
e
D
e
tec
ti
o
n
u
sin
g
Hig
h
-
Re
so
lu
t
io
n
Re
m
o
te
S
e
n
sin
g
Im
a
g
e
,
”
2
0
1
5
IEE
E
In
ter
n
a
ti
o
n
a
l
Ge
o
sc
ien
c
e
a
n
d
Rem
o
te
S
e
n
sin
g
S
y
mp
o
siu
m
(
IGARS
S
)
,
IEE
E,
p
p
.
3
5
6
4
-
3
5
6
7
,
2
0
1
5
.
[6
]
M
.
O.
S
g
h
a
ier,
e
t
a
l.
,
“
Ro
a
d
Da
m
a
g
e
D
e
tec
ti
o
n
f
ro
m
V
HR
Re
m
o
te
S
e
n
si
n
g
Im
a
g
e
s
b
a
se
d
o
n
M
u
lt
isc
a
le
T
e
x
tu
re
A
n
a
l
y
si
s
a
n
d
De
m
p
ste
r
S
h
a
f
e
r
T
h
e
o
r
y
,
”
2
0
1
5
IE
EE
I
n
ter
n
a
t
io
n
a
l
Ge
o
sc
ien
c
e
a
n
d
Rem
o
te
S
e
n
s
in
g
S
y
mp
o
siu
m
(
IGARS
S
)
,
IEE
E
,
p
p
.
2
6
-
3
1
,
2
0
1
5
.
[7
]
P
.
L
i,
e
t
a
l.
,
“
A
No
v
e
l
M
e
th
o
d
fo
r
Urb
a
n
Ro
a
d
Da
m
a
g
e
De
tec
ti
o
n
u
sin
g
V
e
ry
Hi
g
h
Re
so
lu
ti
o
n
S
a
telli
te
I
m
a
g
e
r
y
a
n
d
Ro
a
d
M
a
p
,
”
Ph
o
to
g
ra
mm
e
tri
c
E
n
g
i
n
e
e
rin
g
&
Rem
o
te S
e
n
sin
g
,
v
o
l.
7
7
,
p
p
.
1
0
5
7
-
1
0
6
6
,
2
0
1
1
.
[8
]
L
.
G
o
n
g
,
“
Ro
a
d
Da
m
a
g
e
De
te
c
ti
o
n
f
ro
m
Hig
h
-
Re
so
lu
ti
o
n
RS
Im
a
g
e
,
”
2
0
1
2
IEE
E
In
ter
n
a
ti
o
n
a
l
Ge
o
sc
ien
c
e
a
n
d
Rem
o
te S
e
n
si
n
g
S
y
mp
o
si
u
m
,
IEE
E,
p
p
.
9
9
0
-
9
9
3
,
2
0
1
2
.
[9
]
X
.
Z
h
a
n
g
,
e
t
a
l.
,
“
T
h
e
S
tu
d
y
o
f
Ro
a
d
Da
m
a
g
e
D
e
tec
ti
o
n
Ba
se
d
o
n
Hig
h
-
Re
so
l
u
ti
o
n
S
A
R
Im
a
g
e
,
”
2
0
1
3
IEE
E
In
ter
n
a
t
io
n
a
l
Ge
o
sc
ien
c
e
a
n
d
R
e
mo
te S
e
n
si
n
g
S
y
mp
o
si
u
m
-
IGAR
S
S
,
IEE
E,
p
p
.
2
6
3
3
-
2
6
3
6
,
2
0
1
3
.
[1
0
]
F
.
E.
G
u
n
a
w
a
n
,
e
t
a
l.
,
“
A
V
i
b
ra
to
ry
-
b
a
s
e
d
M
e
th
o
d
f
o
r
R
o
a
d
Da
m
a
g
e
Clas
si
f
ica
ti
o
n
,
”
In
telli
g
e
n
t
T
e
c
h
n
o
lo
g
y
a
n
d
It
s
Ap
p
li
c
a
ti
o
n
s (
IS
IT
IA)
,
2
0
1
5
In
ter
n
a
ti
o
n
a
l
S
e
min
a
r
o
n
,
IEE
E,
p
p
.
1
-
4
,
2
0
1
5
.
[1
1
]
Y.
Ko
b
a
n
a
,
e
t
a
l.
,
“
De
tec
ti
o
n
o
f
Ro
a
d
Da
m
a
g
e
u
sin
g
S
ig
n
a
ls
o
f
S
m
a
rtp
h
o
n
e
-
Em
b
e
d
d
e
d
A
c
c
e
l
e
ro
m
e
t
e
r
w
h
il
e
C
y
c
li
n
g
,
”
2
0
1
4
In
ter
n
a
ti
o
n
a
l
W
o
r
k
sh
o
p
o
n
W
e
b
In
telli
g
e
n
c
e
a
n
d
S
ma
rt S
e
n
si
n
g
,
A
CM
,
p
p
.
1
-
2
,
2
0
1
4
.
[1
2
]
Y.
Ko
b
a
n
a
,
e
t
a
l.
,
“
A
c
c
u
ra
te
Ro
a
d
Da
m
a
g
e
Clas
sif
i
c
a
ti
o
n
Ba
se
d
o
n
Re
a
l
S
ig
n
a
l
M
o
th
e
r
W
a
v
e
let
o
f
Ac
c
e
ler
a
ti
o
n
S
ig
n
a
l,
”
2
0
1
5
IEE
E/
S
ICE
I
n
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m o
n
S
y
ste
m In
teg
ra
ti
o
n
(
S
II)
,
I
EE
E,
p
p
.
9
0
0
-
9
0
5
,
2
0
1
5
.
[1
3
]
L
.
S
u
m
a
r
y
a
n
ti
,
e
t
a
l.
,
“
Dig
it
a
l
Im
a
g
e
b
a
s
e
d
Id
e
n
ti
f
ica
ti
o
n
o
f
Rice
V
a
riety
u
sin
g
I
m
a
g
e
P
ro
c
e
ss
in
g
a
n
d
Ne
u
ra
l
Ne
tw
o
rk
,
”
T
e
le
c
o
mm
u
n
ica
ti
o
n
C
o
mp
u
t
in
g
El
e
c
tro
n
ics
a
n
d
C
o
n
tr
o
l
(
T
EL
KOM
NIKA)
,
Vo
l.
1
6
,
p
p
.
1
8
2
-
1
9
0
,
2
0
1
5
.
[1
4
]
Y.
P
a
n
d
it
a
n
d
C.
S
.
D.
Ra
w
a
t,
“
B
io
m
e
tri
c
P
e
rso
n
a
l
I
d
e
n
ti
f
ica
ti
o
n
b
a
se
d
o
n
Ir
is
P
a
tt
e
rn
s,”
J
o
u
rn
a
l
o
f
T
e
lem
a
ti
c
s
a
n
d
In
fo
rm
a
t
ics
(
J
T
I),
V
o
l
.
2
,
p
p
.
7
-
1
4
,
2
0
1
4
.
[1
5
]
C.
En
,
e
t
a
l.
,
“
A
u
to
m
a
ti
c
De
tec
ti
o
n
a
n
d
A
ss
e
ss
m
e
n
t
S
y
ste
m
o
f
W
a
ter
T
u
rb
id
it
y
b
a
se
d
o
n
Im
a
g
e
P
ro
c
e
ss
in
g
,
”
T
EL
KOM
NIKA,
V
o
l.
1
1
,
p
p
.
1
5
0
6
-
1
5
1
3
,
2
0
1
3
.
[1
6
]
Q.
He
,
e
t
a
l.
,
“
Co
tt
o
n
P
e
sts
a
n
d
Dise
a
se
s
De
te
c
ti
o
n
b
a
se
d
o
n
Im
a
g
e
P
ro
c
e
ss
in
g
,
”
T
e
lec
o
mm
u
n
ica
ti
o
n
Co
mp
u
ti
n
g
El
e
c
tro
n
ics
a
n
d
C
o
n
tro
l
(
T
EL
KO
M
NIKA)
,
Vo
l.
1
1
,
p
p
.
3
4
4
5
-
3
4
5
0
,
2
0
1
3
.
[1
7
]
Z.
Q.
S
h
e
n
,
e
t
a
l.
,
“
Ro
a
d
Da
m
a
g
e
F
e
a
tu
re
Ex
trac
ti
o
n
in
Im
a
g
e
B
a
se
d
o
n
F
ra
c
tal
Dim
e
n
sio
n
,
”
Ap
p
li
e
d
M
e
c
h
a
n
ics
a
n
d
M
a
ter
ia
ls
,
v
o
ls
2
5
6
-
2
5
9
,
p
p
.
2
9
7
1
-
2
9
7
5
,
2
0
1
2
.
[1
8
]
N.
S
u
c
iati,
e
t
a
l.
,
“
Ba
ti
k
M
o
ti
f
Clas
sif
ic
a
ti
o
n
u
sin
g
Co
l
o
r
-
T
e
x
tu
re
-
Ba
se
d
F
e
a
tu
re
Ex
trac
ti
o
n
a
n
d
B
a
c
k
p
ro
p
a
g
a
ti
o
n
Ne
u
ra
l
Ne
t
w
o
rk
,
Ad
v
a
n
c
e
d
Ap
p
l
ied
In
fo
rm
a
ti
c
s
(
IIA
IAA
I)
-
2
0
1
4
I
IAI
3
rd
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
,
IEE
E
,
p
p
.
517
-
5
2
1
,
2
0
1
4
.
[1
9
]
V
.
G
u
p
ta
a
n
d
K.S
.
S
a
g
a
le,
“
Im
p
le
m
e
n
tatio
n
o
f
Clas
sif
ica
ti
o
n
S
y
ste
m
f
o
r
Bra
in
Ca
n
c
e
r
u
si
n
g
Ba
c
k
p
ro
p
a
g
a
ti
o
n
Ne
tw
o
rk
a
n
d
M
RI,
”
2
0
1
2
Nirma
Un
ive
rs
it
y
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
En
g
i
n
e
e
rin
g
(
NUiCONE)
,
IEE
E,
p
p
.
1
-
4
,
2
0
1
2
.
[2
0
]
Y.M
.
Ya
c
o
b
,
e
t
a
l.
,
“
Ha
ru
m
M
a
n
is
M
a
n
g
o
W
e
e
v
il
In
f
e
st
a
ti
o
n
Clas
sif
ica
ti
o
n
u
sin
g
Ba
c
k
p
ro
p
a
g
a
ti
o
n
Ne
u
ra
l
Ne
tw
o
rk
”
,
El
e
c
tro
n
ic De
sig
n
2
0
0
8
-
ICED
2
0
0
8
.
In
ter
n
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
,
IEE
E
,
p
p
.
1
-
6
,
2
0
0
8
.
[2
1
]
M
.
M
o
h
e
b
b
i
a
n
d
H.A
.
M
o
g
h
a
d
a
m
,
“
Re
a
l
-
T
i
m
e
Isc
h
e
m
ic
Be
a
t
Clas
sif
ica
ti
o
n
u
sin
g
Ba
c
k
p
ro
p
a
g
a
ti
o
n
Ne
u
ra
l
Ne
tw
o
rk
,
”
2
0
0
7
IEE
E
1
5
th
S
i
g
n
a
l
Pro
c
e
ss
in
g
a
n
d
C
o
mm
u
n
ic
a
ti
o
n
s A
p
p
l
ica
ti
o
n
s
,
I
EE
E,
p
p
.
1
-
4
,
2
0
0
7
.
[2
2
]
N.A
.
Ra
h
i
m
,
e
t
a
l.
,
“
M
o
v
in
g
V
e
h
icle
No
ise
Clas
sif
ica
ti
o
n
u
sin
g
Ba
c
k
p
ro
p
a
g
a
ti
o
n
A
lg
o
rit
h
m
,
”
S
i
g
n
a
l
Pro
c
e
ss
in
g
a
n
d
Its
A
p
p
li
c
a
ti
o
n
s (
CS
PA
)
-
2
0
1
0
6
th
In
ter
n
a
ti
o
n
a
l
C
o
ll
o
q
u
iu
m
o
n
,
IEE
E,
p
p
.
1
-
6
,
2
0
1
0
.
[2
3
]
S
.
M
.
R.
A
z
g
h
a
d
i,
e
t
a
l.
,
“
G
e
n
d
e
r
Clas
sif
ic
a
ti
o
n
Ba
se
d
o
n
F
e
e
d
F
o
rw
a
rd
Ba
c
k
p
ro
p
a
g
a
ti
o
n
Ne
u
ra
l
Ne
tw
o
rk
,
”
4
th
IFI
P
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
on
Arti
fi
c
ia
l
In
telli
g
e
n
c
e
Ap
p
li
c
a
ti
o
n
s
a
n
d
I
n
n
o
v
a
ti
o
n
s
,
S
p
ri
n
g
e
r
U
S
,
p
p
.
2
9
9
-
3
0
5
,
2
0
0
7
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
S
u
ti
k
n
o
.
He
h
a
d
h
is
u
n
d
e
rg
ra
d
u
a
te
(S
.
T
.
)
in
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
f
ro
m
Dip
o
n
e
g
o
ro
Un
iv
e
rsit
y
,
S
e
m
a
ra
n
g
,
In
d
o
n
e
sia
in
2
0
0
6
,
a
n
d
M
a
ste
r
(M
.
Cs.)
in
Co
m
p
u
ter
S
c
ien
c
e
f
ro
m
th
e
G
a
d
jah
M
a
d
a
Un
iv
e
rsity
,
Yo
g
y
a
k
a
rta,
In
d
o
n
e
sia
in
2
0
0
9
.
His
re
se
a
rc
h
a
re
a
s
o
f
in
tere
st
a
re
a
rti
f
icia
l
in
telli
g
e
n
c
e
,
n
e
u
ra
l
n
e
tw
o
rk
,
im
a
g
e
p
ro
c
e
ss
in
g
,
a
n
d
r
o
b
o
ti
c
s.
Em
a
il
:
ti
k
@u
n
d
ip
.
a
c
.
i
d
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
la
s
s
i
fica
tio
n
o
f R
o
a
d
Da
ma
g
e
fr
o
m
Dig
ita
l I
ma
g
e
Usi
n
g
…
(
S
u
tikn
o
)
165
He
l
m
ie
A
ri
f
W
ib
a
wa
.
He
h
a
d
h
is
u
n
d
e
rg
ra
d
u
a
te
i
n
M
a
t
h
e
m
a
ti
c
De
p
a
rtme
n
t
in
Di
p
o
n
e
g
o
ro
Un
iv
e
rsit
y
,
S
e
m
a
ra
n
g
,
In
d
o
n
e
sia
in
2
0
0
1
,
a
n
d
M
a
ste
r
o
f
Co
m
p
u
t
e
r
S
c
ien
c
e
in
G
a
d
jah
M
a
d
a
Un
iv
e
rist
y
,
Yo
g
y
a
k
a
rta,
In
d
o
n
e
sia
in
2
0
0
8
.
His
re
se
a
rc
h
a
re
a
s
o
f
in
tere
st
a
re
A
rti
f
icia
l
In
telli
g
e
n
c
e
,
Im
a
g
e
P
ro
c
e
ss
in
g
,
a
n
d
C
o
m
p
u
ter V
isi
o
n
.
Em
a
il
:
h
e
l
m
ie.ari
f
@g
m
a
il
.
c
o
m
P
rim
a
Yu
su
f
Bu
d
iarto
.
He
w
a
s
b
o
rn
i
n
Ja
k
a
rta
o
n
De
c
e
m
b
e
r
6
,
1
9
9
1
.
He
g
ra
d
u
a
ted
e
d
u
c
a
ti
o
n
a
t
S
e
n
io
r
Hig
h
S
c
h
o
o
l
lev
e
l
in
2
0
0
9
a
t
S
M
A
N
3
S
u
k
o
h
a
rjo
,
In
d
o
n
e
sia
.
He
is
n
o
w
stu
d
y
in
g
in
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
/In
f
o
rm
a
ti
c
s,
F
a
c
u
lt
y
o
f
S
c
ien
c
e
a
n
d
m
a
th
e
m
a
ti
c
s,
Dip
o
n
e
g
o
r
o
Un
iv
e
rsit
y
,
S
e
m
a
ra
n
g
,
In
d
o
n
e
sia
.
Em
a
il
:
p
ri
m
a
y
u
su
f
b
@g
m
a
il
.
c
o
m
Evaluation Warning : The document was created with Spire.PDF for Python.