I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
8
,
No
.
1
,
A
p
r
il
20
2
5
,
p
p
.
393
~
402
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
3
8
.
i
1
.
pp
393
-
4
0
2
393
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Exploring
divers
e
predic
tion mo
del
s in int
ellig
ent
traff
ic
contro
l
Sa
hira
Vila
k
k
um
a
da
t
hil
,
Velum
a
ni
T
hiy
a
g
a
ra
j
a
n
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
R
a
t
h
i
n
a
m
C
o
l
l
e
g
e
o
f
A
r
t
s a
n
d
S
c
i
e
n
c
e
,
C
o
i
m
b
a
t
o
r
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
May
29
,
2
0
2
4
R
ev
is
ed
Oct
18
,
2
0
2
4
Acc
ep
ted
Oc
t
30
,
2
0
2
4
Traffic
c
o
n
g
e
sti
o
n
is
a
m
a
jo
r
c
h
a
ll
e
n
g
e
t
h
a
t
a
ffe
c
ts
e
x
c
e
ll
e
n
c
e
o
f
li
fe
f
o
r
n
u
m
e
ro
u
s p
e
o
p
le
a
c
ro
ss
wo
rl
d
.
T
h
e
fa
st
g
ro
wt
h
i
n
m
a
n
y
v
e
h
icle
s
c
o
n
tri
b
u
tes
to
c
o
n
g
e
stio
n
d
u
r
in
g
p
e
a
k
a
n
d
n
o
n
-
p
e
a
k
h
o
u
rs.
T
h
e
v
e
h
icle
traffic
re
su
lt
e
d
in
m
a
n
y
issu
e
s
li
k
e
a
c
c
id
e
n
ts
a
n
d
i
n
e
fficie
n
c
y
in
traffic
fl
o
w.
M
a
n
y
tr
a
ffic
li
g
h
t
c
o
n
tro
l
sy
ste
m
s
o
p
e
ra
te
o
n
fi
x
e
d
ti
m
e
in
ter
v
a
ls
lea
d
s
to
i
n
e
ffici
e
n
c
y
.
Th
e
fix
e
d
-
ti
m
e
sig
n
a
ls
c
a
u
se
u
n
n
e
c
e
ss
a
ry
d
e
lay
s
o
n
ro
a
d
s
with
m
i
n
imu
m
n
u
m
b
e
r
o
f
q
u
a
n
ti
t
y
v
e
h
icle
s.
In
telli
g
e
n
t
tran
sp
o
rt
s
y
ste
m
s
(IT
S
)
in
tr
o
d
u
c
e
n
e
w
c
o
m
p
re
h
e
n
siv
e
fra
m
e
wo
rk
t
h
a
t
c
o
m
b
in
e
th
e
a
d
v
a
n
c
e
d
tec
h
n
o
lo
g
ies
to
imp
ro
v
e
th
e
tran
s
p
o
rtati
o
n
n
e
two
rk
e
fficie
n
c
y
a
n
d
t
o
o
p
ti
m
ize
t
h
e
traffic
m
a
n
a
g
e
m
e
n
t.
Th
e
h
i
g
h
-
traffic
ro
u
tes
a
re
fo
rc
e
d
t
o
wa
it
e
x
c
e
ss
iv
e
ly
.
M
a
c
h
in
e
lea
rn
in
g
(
ML
)
m
e
th
o
d
s
h
a
v
e
d
e
sig
n
e
d
t
o
e
x
a
m
in
e
th
e
traffi
c
c
o
n
tr
o
l.
Ho
we
v
e
r,
th
e
a
c
c
u
ra
te
d
e
tec
ti
o
n
a
n
d
v
e
h
icle
trac
k
in
g
a
re
e
ss
e
n
ti
a
l
o
n
e
fo
r
e
ffe
c
ti
v
e
ITS
.
I
n
o
r
d
e
r
t
o
m
e
n
ti
o
n
t
h
e
se
p
r
o
b
lem
s,
M
L
a
n
d
d
e
e
p
lea
rn
i
n
g
(
DL
)
m
e
th
o
d
s a
re
i
n
tro
d
u
c
e
d
to
i
m
p
ro
v
e
p
re
d
icti
o
n
p
e
rf
o
rm
a
n
c
e
.
K
ey
w
o
r
d
s
:
Hig
h
-
tr
af
f
ic
r
o
u
tes
I
n
tellig
en
t tr
an
s
p
o
r
t sy
s
tem
s
T
r
af
f
ic
co
n
g
esti
o
n
T
r
an
s
p
o
r
tatio
n
n
etwo
r
k
ef
f
icien
cy
Veh
icle
tr
af
f
ic
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
:
Velu
m
an
i T
h
iy
ag
a
r
ajan
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
R
ath
in
am
C
o
lleg
e
o
f
A
r
ts
an
d
Scien
ce
C
o
im
b
ato
r
e,
T
am
il Na
d
u
,
I
n
d
i
a
E
m
ail:
v
elu
m
an
i4
6
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
tr
af
f
ic
co
n
g
esti
o
n
o
cc
u
r
s
wh
en
r
o
a
d
u
s
ag
e
in
cr
ea
s
es.
Ma
n
y
s
tr
ateg
ies
an
d
tech
n
iq
u
es
b
ased
o
n
m
ac
h
in
e
lear
n
in
g
(
ML
)
as
well
as
co
m
p
u
tatio
n
al
in
tell
ig
en
ce
wer
e
em
p
lo
y
ed
to
p
r
ev
en
t
co
n
g
esti
o
n
.
C
o
m
p
u
tatio
n
al
in
tellig
en
ce
is
u
s
ed
f
o
r
m
a
n
ag
in
g
tr
af
f
ic
a
n
d
f
o
r
m
i
n
im
izin
g
th
e
c
o
n
g
esti
o
n
.
C
o
m
p
u
tatio
n
al
in
tellig
en
ce
is
u
s
ed
f
o
r
h
an
d
lin
g
tr
af
f
ic
an
d
f
o
r
m
in
im
izi
n
g
th
e
co
n
g
esti
o
n
.
An
t,
b
ee
,
as
well
as
g
en
etic
m
eth
o
d
s
em
p
lo
y
ed
f
o
r
tr
a
f
f
ic
m
an
ag
em
e
n
t.
W
ith
d
ev
elo
p
m
en
t
in
co
m
p
u
ter
v
is
io
n
,
M
L
an
d
d
ee
p
lear
n
i
n
g
(
DL
)
m
eth
o
d
s
ar
e
u
s
ed
to
id
e
n
tify
,
r
ec
o
g
n
ize,
ca
teg
o
r
ize
an
d
tr
ac
k
th
e
m
u
ltip
le
o
b
jects
in
im
ag
es
o
r
v
id
e
o
s
.
W
ith
d
ev
elo
p
m
en
t
o
f
tech
n
o
l
o
g
y
,
v
ar
io
u
s
in
tellig
en
t
tr
an
s
p
o
r
t
s
y
s
tem
s
(
I
T
S)
h
a
v
e
in
cr
ea
s
ed
th
eir
d
esire
f
o
r
au
to
m
atio
n
.
W
ith
lar
g
e
d
ev
elo
p
m
en
t
o
f
v
eh
icu
lar
c
o
m
m
u
n
icatio
n
s
er
v
ices,
th
er
e
is
elev
ated
s
t
ip
u
late
f
o
r
in
tellig
en
t
tr
an
s
p
o
r
tatio
n
s
ch
em
e
to
au
to
m
atica
lly
id
e
n
tif
y
th
e
u
n
u
s
u
al
tr
a
f
f
ic
o
f
f
e
n
s
e
an
d
u
n
co
n
tr
o
lled
d
r
iv
in
g
o
n
r
o
a
d
s
.
Veh
icle
lo
ca
lizatio
n
is
an
im
p
o
r
tan
t
p
ar
t
f
o
r
in
tellig
en
t
as
well
as
au
to
n
o
m
o
u
s
s
ch
em
es
lik
e
s
elf
-
d
r
iv
en
d
r
iv
in
g
,
s
u
r
v
eillan
ce
an
d
s
o
o
n
.
Dif
f
er
en
t
v
eh
icle
id
en
tific
atio
n
tech
n
iq
u
es
ar
e
em
p
lo
y
e
d
wid
esp
r
ea
d
wi
th
f
r
am
e
d
if
f
er
e
n
cin
g
,
an
d
g
au
s
s
ian
m
i
x
tu
r
e
m
o
d
el
(
GM
M
)
.
T
r
a
f
f
ic
v
i
d
eo
p
r
o
ce
s
s
in
g
is
ca
r
r
ie
d
o
u
t
to
f
o
llo
w
p
o
i
g
n
an
t
v
eh
icl
es
as
o
f
o
n
e
f
r
am
e
o
f
im
ag
e
s
eq
u
en
ce
to
a
n
o
th
er
f
r
a
m
e.
W
ith
wid
e
-
r
an
g
in
g
r
esear
ch
in
au
to
m
ated
s
u
r
v
eillan
ce
s
y
s
tem
s
,
h
ig
h
-
p
r
ec
is
io
n
v
e
h
icle
r
ec
o
g
n
itio
n
an
d
tr
ac
k
in
g
is
d
em
an
d
in
g
o
n
e
d
u
e
to
co
m
p
lex
r
o
ad
n
etwo
r
k
s
an
d
v
ar
iab
le
illu
m
i
n
atio
n
co
n
d
itio
n
s
.
Fig
u
r
e
1
illu
s
tr
ate
s
th
e
ad
v
an
tag
es
o
f
tr
af
f
ic
co
n
tr
o
l p
r
ed
ictio
n
p
r
o
c
ess
.
ITS
is
u
s
ed
to
b
len
d
n
etw
o
r
k
-
r
elate
d
d
ata
lik
e
v
eh
icu
lar
clo
u
d
co
m
p
u
tin
g
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
3
9
3
-
4
0
2
394
a
s
ea
m
less
m
an
n
er
f
o
r
h
an
d
li
n
g
th
e
tr
af
f
ic
ef
f
icien
tly
.
T
h
e
tr
af
f
ic
c
o
n
g
esti
o
n
h
an
d
lin
g
s
o
lu
tio
n
is
ca
te
g
o
r
ize
d
as
: i)
t
r
af
f
ic
d
ata
co
llectio
n
,
ii)
t
r
af
f
ic
m
an
a
g
em
en
t,
iii)
c
o
n
g
esti
o
n
av
o
id
an
ce
,
an
d
i
v
)
t
r
av
e
l tim
e
p
r
ed
ictio
n
.
Fig
u
r
e
1
.
B
en
ef
its
o
f
t
r
af
f
ic
co
n
tr
o
l p
r
e
d
ictio
n
Fig
u
r
e
2
d
escr
ib
es
th
e
d
if
f
er
en
t
p
r
o
ce
s
s
es
o
f
in
tellig
en
t
tr
af
f
ic
m
an
ag
em
en
t.
T
h
e
tr
af
f
ic
d
ata
co
llectio
n
is
m
ain
an
d
im
p
o
r
tan
t
f
u
n
ctio
n
in
m
an
ag
i
n
g
tr
af
f
ic
m
a
n
ag
em
e
n
t
p
r
o
ce
s
s
f
o
r
ac
c
u
r
ate
tr
a
f
f
ic
p
r
ed
ictio
n
.
T
r
af
f
ic
m
an
ag
em
e
n
t
s
y
s
tem
is
th
e
p
r
o
ce
s
s
o
f
o
r
g
an
izin
g
,
ar
r
a
n
g
in
g
,
g
u
i
d
in
g
an
d
h
a
n
d
lin
g
th
e
tr
af
f
ic
with
m
o
v
i
n
g
v
eh
icles
a
n
d
s
tatio
n
ar
y
v
eh
icles
alo
n
g
with
cy
clis
ts
an
d
p
ed
estrian
s
.
T
r
af
f
ic
m
a
n
ag
em
e
n
t
s
y
s
tem
g
u
ar
an
teed
th
e
s
af
ety
an
d
ef
f
ec
tiv
e
n
ess
d
u
r
in
g
m
o
v
em
en
t
o
f
p
e
o
p
le
an
d
g
o
o
d
s
.
T
h
e
d
esig
n
ed
s
y
s
tem
in
cr
ea
s
ed
th
e
en
v
ir
o
n
m
e
n
t q
u
a
lity
ar
o
u
n
d
tr
af
f
ic
l
o
ca
tio
n
s
.
Ou
r
co
n
tr
ib
u
tio
n
o
f
th
e
w
o
r
k
i
s
g
iv
en
as:
−
W
e
p
r
ed
ict
th
e
in
tellig
en
t
tr
af
f
ic
o
cc
u
r
r
en
ce
u
s
in
g
d
if
f
e
r
en
t
ML
an
d
DL
m
eth
o
d
s
as
well
as
d
em
o
n
s
tr
ate
th
at
m
o
d
el
ca
r
r
y
o
u
t e
n
h
a
n
ce
d
o
n
th
is
d
atab
ase.
−
W
e
in
tr
o
d
u
ce
th
e
in
tellig
en
t
tr
af
f
ic
co
n
tr
o
l
p
r
ed
ictio
n
o
b
jecti
v
es
b
y
s
ix
d
is
s
im
ilar
ML
m
o
d
els
an
d
p
r
esen
t
co
m
p
ar
ativ
e
a
n
aly
s
is
with
co
n
v
en
tio
n
al
m
o
d
els.
−
W
e
co
n
d
u
ct
co
m
p
ar
is
o
n
o
f
r
es
u
lts
o
n
th
e
in
t
ellig
en
t tr
a
f
f
ic
c
o
n
tr
o
l f
o
r
ec
ast b
y
d
is
s
im
ilar
ML
f
r
am
ewo
r
k
s
.
Fig
u
r
e
2
.
Pro
ce
s
s
o
f
in
tellig
en
t tr
af
f
ic
m
an
ag
e
m
en
t
2.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
A
s
m
ar
t
tr
af
f
ic
c
o
n
tr
o
l
s
y
s
tem
was
d
ev
elo
p
ed
i
n
[
1
]
wh
ic
h
s
ep
er
ate
im
a
g
e
u
s
in
g
ex
tr
e
m
e
g
r
a
d
ien
t
b
o
o
s
tin
g
(
XGBo
o
s
t
)
class
if
ier
to
ex
to
r
t
f
o
r
ef
r
o
n
t
o
b
jects
f
r
o
m
p
r
ep
r
o
ce
s
s
ed
im
ag
e
f
o
r
ac
c
u
r
ate
v
eh
icle
d
etec
tio
n
an
d
tr
ac
k
in
g
.
H
o
wev
er
,
it
f
ailed
to
ap
p
ly
DL
de
tecto
r
s
to
in
cr
ea
s
e
v
eh
icle
d
etec
tio
n
ac
cu
r
ac
y
.
A
tr
af
f
ic
m
o
n
ito
r
in
g
an
d
co
n
tr
o
llin
g
s
y
s
tem
wer
e
d
ev
el
o
p
ed
in
[
2
]
to
im
p
r
o
v
e
ac
cu
r
ac
y
an
d
m
in
im
ize
co
m
p
u
tatio
n
tim
e.
Ho
we
v
er
,
it
f
ac
es
ch
allen
g
es
in
tr
af
f
ic
m
o
n
ito
r
in
g
an
d
c
o
n
tr
o
llin
g
d
u
e
to
th
e
lar
g
e
av
aila
b
ilit
y
o
f
im
a
g
es.
An
a
u
to
m
ated
tr
af
f
ic
m
o
n
ito
r
in
g
s
y
s
tem
,
ca
lled
T
R
AM
ON,
was
d
ev
elo
p
ed
in
[
3
]
f
o
r
ac
cu
r
ately
p
r
ed
ictin
g
tr
af
f
ic
a
n
d
tr
ac
k
in
g
v
eh
icle
p
o
s
itio
n
s
.
Ho
wev
er
,
th
e
s
y
s
tem
d
id
n
o
t
ad
d
r
ess
th
e
is
s
u
e
o
f
tim
e
co
m
p
lex
ity
in
tr
af
f
ic
p
r
ed
ictio
n
.
A
m
o
d
el
-
b
ased
r
ein
f
o
r
ce
m
en
t
lear
n
i
n
g
(
R
L
)
m
et
h
o
d
ca
lled
d
ee
p
Q
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
xp
lo
r
in
g
d
ivers
e
p
r
ed
ictio
n
mo
d
els in
in
tellig
en
t tra
ffic
co
n
tr
o
l
…
(
S
a
h
ir
a
V
ila
kk
u
ma
d
a
th
il)
395
n
etwo
r
k
(
DQN)
was
d
e
v
elo
p
ed
in
[
4
]
to
en
h
a
n
ce
ac
c
u
r
ac
y
o
f
tr
af
f
ic
s
ig
n
al
co
n
tr
o
l.
B
u
t
th
e
er
r
o
r
r
ate
was
ac
cu
r
ately
r
ed
u
ce
d
.
E
n
d
-
to
-
e
n
d
DL
n
etwo
r
k
n
am
ed
Pair
in
g
Net
was
d
ev
elo
p
ed
[
5
]
to
i
m
p
r
o
v
e
ac
c
u
r
ac
y
o
f
tr
af
f
ic
an
aly
s
is
th
r
o
u
g
h
m
u
ltip
le
o
b
jects
tr
ac
k
in
g
(
MO
T
)
an
d
o
b
ject
d
etec
tio
n
.
Ho
wev
er
,
ac
cu
r
ate
tr
af
f
ic
f
lo
w
p
r
ed
ictio
n
r
em
ain
ed
a
ch
allen
g
in
g
is
s
u
e.
A
c
o
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN
)
-
b
ased
class
if
ier
w
as
d
ev
elo
p
ed
in
[
6
]
to
im
p
r
o
v
e
v
eh
icle
tr
af
f
ic
f
lo
w
m
o
n
ito
r
i
n
g
th
r
o
u
g
h
tr
ajec
to
r
y
tr
ac
k
i
n
g
an
d
o
b
j
ec
t
d
etec
t
io
n
.
Ho
wev
er
,
it
f
ailed
to
attain
h
ig
h
est
ac
cu
r
ac
y
in
tr
a
f
f
ic
f
lo
w
esti
m
atio
n
.
A
n
ew
g
r
o
u
n
d
tr
af
f
ic
m
an
ag
em
en
t
ap
p
r
o
ac
h
was
in
tr
o
d
u
ce
d
in
[
7
]
with
m
o
b
ilit
y
,
f
le
x
ib
ilit
y
an
d
m
u
ltip
le
u
n
m
a
n
n
ed
ae
r
ial
v
eh
icles
(
UAVs)
.
T
h
e
m
ain
aim
o
f
d
esig
n
ed
a
p
p
r
o
ac
h
was
to
en
h
an
ce
th
e
n
av
ig
atio
n
an
d
d
r
iv
in
g
ex
p
er
ien
ce
th
r
o
u
g
h
u
s
in
g
UAVs
th
at
av
o
id
co
n
g
ested
r
o
u
tes.
C
NN
-
b
ased
ap
p
r
o
ac
h
te
r
m
ed
L
ig
h
tSp
aN
was
in
tr
o
d
u
ce
d
in
[
8
]
f
o
r
v
e
h
icle
id
en
tific
atio
n
with
s
p
ar
s
e
d
ata
to
attain
co
m
p
lex
s
o
lu
tio
n
.
T
h
e
d
esig
n
ed
ap
p
r
o
ac
h
m
in
im
iz
ed
waitin
g
tim
e
a
n
d
tr
av
elin
g
tim
e
with
h
ig
h
ac
cu
r
ac
y
.
A
n
ew
f
r
am
ewo
r
k
was
in
tr
o
d
u
ce
d
i
n
[
9
]
f
o
r
m
o
n
ito
r
in
g
h
ig
h
way
tr
af
f
ic
-
s
tr
ea
m
m
ea
s
u
r
es
with
q
u
ality
tr
ajec
to
r
y
d
ata.
T
h
e
d
esig
n
ed
f
r
am
ewo
r
k
co
m
p
r
is
ed
th
e
m
ea
s
u
r
e
th
at
r
ef
lects
f
o
ll
o
wer
d
r
iv
er
te
r
m
ed
r
ec
ep
tiv
en
ess
an
g
le.
A
n
ew
r
o
ad
tr
a
f
f
ic
n
o
is
e
m
o
d
el
(
R
T
NM
)
was
in
tr
o
d
u
ce
d
i
n
[
1
0
]
f
o
r
d
y
n
am
ically
ass
es
s
in
g
r
o
ad
tr
af
f
ic
s
o
u
n
d
s
tag
es
as
o
f
r
eliab
le
d
ata.
R
T
NM
s
u
p
p
o
r
ted
o
r
r
ep
lace
d
th
e
n
o
is
e
s
en
s
o
r
n
etwo
r
k
s
th
r
o
u
g
h
s
o
lv
i
n
g
n
o
is
e
p
o
llu
tio
n
c
o
n
ce
r
n
s
.
C
NN
an
d
R
L
m
eth
o
d
was
d
esig
n
ed
i
n
[
1
1
]
.
T
h
e
d
esig
n
e
d
tech
n
iq
u
e
r
e
d
u
ce
d
th
e
o
v
er
h
ea
d
o
n
o
b
s
er
v
ed
en
titi
es
with
elev
ated
b
it
r
ate.
T
h
e
r
ec
u
r
s
iv
e
n
etwo
r
k
d
esig
n
was
co
n
s
tr
u
cted
in
[
1
2
]
to
m
o
n
ito
r
tr
af
f
ic
f
lo
w
f
o
r
a
n
o
m
aly
id
en
tific
atio
n
.
T
h
e
d
e
s
ig
n
in
cr
ea
s
ed
th
e
cy
b
er
-
attac
k
d
etec
tio
n
in
S
DN.
T
h
e
d
is
tr
ib
u
ted
d
en
ial
-
of
-
s
er
v
ice
(
DDo
S)
attac
k
w
as
av
o
id
ed
th
r
o
u
g
h
elim
in
atin
g
th
e
n
etwo
r
k
f
o
r
w
ar
d
in
g
p
er
f
o
r
m
an
ce
d
eg
r
ad
ati
o
n
.
E
f
f
icien
tDet
ar
ch
itectu
r
e
an
d
T
en
s
o
r
Flo
w
lite
was
em
p
lo
y
ed
in
[
1
3
]
to
em
p
lo
y
r
ea
l
-
tim
e
liv
e
v
id
eo
g
iv
en
as
o
f
ca
m
er
as
at
in
ter
s
ec
t
io
n
s
.
T
h
e
d
esig
n
ed
ar
ch
itectu
r
e
ca
r
r
ie
d
o
u
t
in
s
ta
n
tan
eo
u
s
tr
af
f
ic
b
u
lk
in
ess
co
m
p
u
tatio
n
th
r
o
u
g
h
im
ag
e
p
r
o
ce
s
s
in
g
an
d
v
eh
icle
d
etec
tio
n
.
T
h
e
ac
o
u
s
tic
n
o
is
e
m
o
n
it
o
r
in
g
s
y
s
tem
was
in
tr
o
d
u
ce
d
in
[
1
4
]
f
o
r
r
o
ad
tr
a
f
f
ic
m
o
n
ito
r
in
g
with
d
r
iv
er
s
af
ety
.
T
h
e
d
esig
n
e
d
s
y
s
tem
em
p
lo
y
ed
v
e
h
icle
ty
p
e
a
n
d
w
ea
th
er
-
r
elate
d
p
av
em
en
t
c
o
n
d
itio
n
d
e
p
en
d
in
g
o
n
au
d
io
lev
el
m
ea
s
u
r
em
e
n
t.
A
n
in
teg
r
ate
d
f
o
g
a
n
d
clo
u
d
co
m
p
u
tin
g
f
r
a
m
ewo
r
k
was
d
esig
n
ed
in
[
1
5
]
to
m
in
im
ize
th
e
laten
cy
a
n
d
n
et
wo
r
k
co
n
g
esti
o
n
f
o
r
tr
af
f
ic
m
o
n
ito
r
in
g
.
A
n
ew
b
o
u
n
d
in
g
b
o
x
(
B
b
o
x
)
-
b
ased
v
eh
icle
tr
ac
k
in
g
alg
o
r
ith
m
wa
s
d
esig
n
ed
in
[
1
6
]
with
v
eh
icl
e
o
b
ject
p
atter
n
s
co
llected
f
r
o
m
h
ig
h
way
v
i
d
eo
s
.
T
h
e
ML
class
if
ier
an
d
C
NN
c
la
s
s
if
ier
wer
e
s
elec
ted
r
eal
-
ti
m
e
h
ig
h
way
tr
af
f
ic
m
o
n
ito
r
in
g
s
y
s
tem
.
Fo
r
s
m
ar
t
tr
af
f
ic
m
o
n
ito
r
in
g
a
n
d
m
a
n
a
g
em
en
t,
th
e
s
elf
-
p
o
wer
ed
tr
i
b
o
elec
tr
ic
s
en
s
o
r
(
C
N
-
STS)
with
elec
tr
o
s
p
u
n
co
m
p
o
s
ite
n
an
o
f
ib
er
s
was
in
tr
o
d
u
ce
d
in
[
1
7
]
.
T
r
an
s
f
er
r
e
d
ch
ar
g
e
d
en
s
ity
was
u
s
e
d
to
ad
d
r
ess
th
e
f
ast
-
r
esp
o
n
s
e
an
d
h
ig
h
s
en
s
itiv
ity
n
ee
d
s
f
o
r
s
m
ar
t
tr
af
f
ic
m
an
a
g
em
en
t.
A
lo
w
-
c
o
s
t
in
ter
n
et
o
f
th
in
g
s
(
I
o
T
)
s
y
s
tem
was
d
esig
n
ed
in
[
1
8
]
f
o
r
tr
af
f
ic
f
lo
w
m
o
n
ito
r
in
g
an
d
air
q
u
ality
in
d
ex
(
AQI
)
.
T
r
a
f
f
ic
f
l
o
w
was
p
er
f
o
r
m
ed
th
r
o
u
g
h
v
id
eo
p
r
o
ce
s
s
in
g
i
n
c
o
m
p
r
ess
ed
d
o
m
ain
.
I
t
was
d
et
er
m
in
ed
i
n
r
ea
l
-
tim
e
o
v
e
r
em
b
ed
d
ed
ar
ch
itectu
r
e.
Air
p
o
llu
tio
n
g
au
g
e
s
tatio
n
w
as
co
m
p
u
te
d
in
[
1
9
]
to
f
u
lf
ill
with
v
al
u
es.
T
h
e
co
n
tr
ib
u
tio
n
was
em
p
l
o
y
ed
to
ex
am
in
e
th
e
c
u
r
r
e
n
t
d
is
tr
ib
u
ti
o
n
o
f
air
e
x
ce
llen
ce
o
b
s
er
v
in
g
s
ta
tio
n
s
f
o
r
r
o
ad
tr
a
n
s
p
o
r
t.
T
h
e
d
y
n
am
ic
u
p
d
ates
o
f
tr
af
f
ic
n
o
is
e
m
a
p
was
ca
r
r
i
ed
o
u
t
in
[
2
0
]
f
o
r
n
o
is
e
m
o
n
it
o
r
in
g
with
tr
af
f
ic
s
p
ee
d
d
ata
to
f
o
r
ec
ast
th
e
n
o
is
e
em
is
s
io
n
o
f
r
o
a
d
n
etwo
r
k
.
T
h
e
tr
af
f
ic
s
p
ee
d
was
em
p
lo
y
e
d
f
o
r
af
f
ec
tin
g
tr
a
f
f
ic
n
o
is
e.
T
h
e
tr
af
f
ic
s
p
ee
d
was
co
m
p
u
ted
to
u
p
d
ate
n
o
is
e
s
o
u
r
ce
in
ten
s
ity
with
r
ea
l
-
tim
e
n
o
is
e
m
o
n
ito
r
in
g
d
ata.
Su
p
er
v
is
ed
lear
n
in
g
f
r
am
ewo
r
k
was
d
esig
n
ed
[
2
1
]
f
o
r
s
tr
u
ctu
r
al
h
ea
lth
m
o
n
ito
r
in
g
(
SHM
)
-
s
en
s
o
r
-
b
ased
tr
af
f
ic
lo
ad
esti
m
atio
n
.
T
h
e
s
h
o
r
t
r
ec
o
r
d
in
g
s
ess
io
n
was
co
m
p
u
ted
f
r
o
m
s
m
ar
t
ca
m
er
a
to
lab
el
ac
ce
ler
atio
n
d
a
ta
with
eq
u
i
v
alen
t
n
u
m
b
er
o
f
p
ass
in
g
v
eh
icles.
Fu
zz
y
in
cid
en
ce
g
r
a
p
h
was
co
n
s
tr
u
cted
in
[
2
2
]
with
f
u
z
zy
in
cid
en
ce
ch
r
o
m
atic
n
u
m
b
er
s
.
Fu
zz
y
in
cid
en
ce
co
lo
r
i
n
g
m
o
n
ito
r
ed
th
e
h
u
m
an
lo
s
s
d
u
r
in
g
ac
cid
en
ts
th
r
o
u
g
h
a
d
h
er
i
n
g
to
tr
af
f
ic
f
lo
w
laws
with
m
in
im
u
m
tr
a
f
f
ic
f
lo
w
waitin
g
tim
e
.
T
h
e
ac
cid
en
t
m
o
n
i
to
r
in
g
o
f
ch
o
s
en
ar
ea
was
p
er
f
o
r
m
ed
[
2
3
]
t
o
co
m
m
u
n
icate
with
th
e
d
ata
an
d
s
h
o
wed
p
a
r
ticu
lar
tr
af
f
ic
ac
cid
en
ts
.
A
s
in
g
le
-
n
o
d
e
tr
af
f
ic
m
ea
s
u
r
em
e
n
t
s
ch
em
e
ter
m
ed
Flex
M
o
n
was
in
tr
o
d
u
ce
d
i
n
[
2
4
]
t
o
d
eter
m
in
e
f
in
e
-
g
r
ain
ed
f
lo
ws
at
s
o
lit
ar
y
n
etwo
r
k
n
o
d
e.
Flex
Mo
n
d
iv
id
ed
th
e
lar
g
e
f
lo
ws
f
r
o
m
s
m
all
o
n
es
w
ith
f
lo
w
r
u
les,
s
k
etch
es.
ML
f
r
am
ewo
r
k
was
d
esig
n
ed
in
[
2
5
]
to
f
in
d
th
e
tr
af
f
ic
co
n
g
esti
o
n
d
ep
en
d
in
g
o
n
m
u
ltip
le
p
ar
am
eter
s
lik
e
d
elay
co
n
s
tr
ain
ts
an
d
s
p
ee
d
th
r
o
u
g
h
GSP
v
eh
icle
tr
ajec
to
r
y
.
A
tr
af
f
ic
ev
en
t
r
ep
o
r
tin
g
s
ch
em
e
wa
s
d
esig
n
ed
f
o
r
ef
f
icien
t
e
v
en
t
d
etec
tio
n
an
d
d
ata
s
o
u
r
ce
r
ep
u
tatio
n
m
ec
h
an
is
m
s
.
I
n
co
r
p
o
r
ated
m
o
n
ito
r
in
g
p
lat
f
o
r
m
was
in
tr
o
d
u
ce
d
.
Air
q
u
a
lity
o
b
s
er
v
in
g
u
n
it
co
m
b
in
ed
o
p
en
-
s
o
u
r
ce
tec
h
n
o
lo
g
y
with
lo
w
-
co
s
t
an
d
h
ig
h
-
r
eso
lu
tio
n
s
en
s
o
r
s
.
A
n
e
x
p
o
n
en
tially
-
weig
h
ted
m
o
v
in
g
av
er
a
g
e
(
E
W
MA
)
m
o
n
ito
r
in
g
s
ch
em
e
was
in
tr
o
d
u
c
ed
in
t
o
in
c
r
ea
s
e
th
e
r
o
b
u
s
tn
ess
an
d
m
i
n
im
ize
th
e
f
alse
alar
m
s
b
ec
au
s
e
o
f
m
o
d
elin
g
er
r
o
r
.
I
o
T
b
ased
wir
ele
s
s
s
en
s
o
r
s
y
s
tem
was
d
esig
n
ed
in
with
wir
eles
s
ac
ce
ler
o
m
eter
f
o
r
tr
af
f
ic
an
d
v
eh
icle
class
if
icatio
n
m
o
n
ito
r
in
g
.
L
ab
o
r
ato
r
y
test
s
,
f
ield
test
s
an
d
n
u
m
er
ical
s
im
u
latio
n
wer
e
p
er
f
o
r
m
ed
to
v
alid
ate
ac
cu
r
ac
y
o
f
m
o
n
ito
r
i
n
g
s
y
s
tem
.
An
ad
ap
tiv
e
len
g
th
(
AL
)
b
itm
ap
was
in
t
r
o
d
u
ce
d
in
to
co
n
s
tr
u
ct
th
e
tr
af
f
ic
s
u
m
m
ar
y
.
AL
-
b
itm
a
p
cr
ea
ted
th
e
b
itm
ap
with
AL
f
o
r
ev
er
y
h
o
s
t.
T
h
e
b
itm
ap
len
g
th
au
t
o
m
atica
lly
in
cr
ea
s
ed
with
n
u
m
b
er
o
f
h
o
s
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
3
9
3
-
4
0
2
396
3.
M
E
T
H
O
D
Au
to
m
atio
n
as
well
as
in
tellig
en
t
co
n
tr
o
l
tech
n
o
lo
g
ies
ar
e
m
an
n
er
to
r
ev
o
lu
tio
n
ize
f
lo
w
o
f
tr
af
f
ic
as
well
as
s
ec
u
r
ity
in
m
o
d
er
n
t
r
an
s
p
o
r
tatio
n
s
ch
em
es
th
at
e
s
co
r
t
to
s
u
g
g
ested
s
ch
em
e,
m
icr
o
co
n
tr
o
ller
an
d
ca
m
er
as
em
p
lo
y
e
d
to
f
o
llo
w
n
u
m
b
er
o
f
v
eh
icles,
p
er
m
itti
n
g
f
o
r
tim
e
-
b
asis
o
f
o
b
s
er
v
in
g
o
f
s
ch
em
e.
T
r
af
f
ic
m
ig
h
t
h
ap
p
e
n
b
e
ca
u
s
e
o
f
w
eig
h
ty
tr
af
f
ic
jam
s
in
in
ter
s
ec
tio
n
s
.
T
h
er
e
ar
e
d
iv
er
s
e
tr
a
f
f
ic
ad
m
in
is
tr
atio
n
ap
p
r
o
ac
h
es wh
ich
in
tr
in
s
ically
s
elf
-
ch
an
g
in
g
to
ev
a
d
e
co
n
g
e
s
tio
n
.
3
.1
.
A
s
ma
rt
t
r
a
f
f
ic
c
o
ntr
o
l
s
chem
e
depend
o
n p
ix
el
-
la
belin
g
a
nd
SO
RT
t
ra
ck
e
r
Au
to
n
o
m
o
u
s
v
eh
icle
r
e
co
g
n
it
io
n
as
well
as
tr
ac
k
in
g
wer
e
ess
en
tial
o
n
e
f
o
r
in
tellig
en
t
tr
an
s
p
o
r
t
m
an
ag
em
en
t
as
well
as
co
n
t
r
o
l
s
ch
em
es.
Nu
m
er
o
u
s
m
eth
o
d
s
wer
e
em
p
lo
y
ed
to
d
esig
n
th
e
s
m
ar
t
tr
af
f
ic
s
ch
em
es.
T
h
e
v
e
h
icle
r
ec
o
g
n
i
tio
n
as
well
as
tr
ac
k
in
g
was
ca
r
r
ied
o
u
t
th
r
o
u
g
h
p
i
x
el
-
lab
elin
g
an
d
r
ea
l
-
tim
e
tr
ac
k
in
g
.
A
n
ew
s
m
ar
t
tr
af
f
i
c
co
n
tr
o
l
s
y
s
tem
was
in
tr
o
d
u
ce
d
to
p
ar
titi
o
n
th
e
im
ag
e
th
r
o
u
g
h
XGBo
o
s
)
class
if
ier
f
o
r
ex
tr
ac
tin
g
th
e
f
o
r
eg
r
o
u
n
d
o
b
jects.
T
h
e
d
esig
n
ed
m
o
d
el
was
p
ar
titi
o
n
ed
i
n
to
s
ev
en
s
tep
s
.
I
n
p
r
im
ar
y
s
tep
,
ev
er
y
im
ag
es
p
r
ep
r
o
ce
s
s
ed
to
elim
in
ate
th
e
n
o
is
e.
I
n
s
ec
o
n
d
s
tep
,
th
e
p
ix
el
-
l
ab
elin
g
was
ca
r
r
ied
o
u
t
p
er
f
o
r
m
ed
th
r
o
u
g
h
XGBo
o
s
t
c
lass
if
ier
to
d
iv
id
e
b
ac
k
g
r
o
u
n
d
f
r
o
m
f
o
r
eg
r
o
u
n
d
.
I
n
t
h
ir
d
s
tep
,
all
p
ix
els
class
if
ied
was
ex
tr
ac
ted
an
d
c
o
n
v
er
ted
in
to
b
in
ar
y
im
ag
e.
T
h
e
b
lo
b
ex
tr
ac
tio
n
m
eth
o
d
wa
s
em
p
lo
y
ed
to
lim
it
ev
er
y
v
e
h
icle.
I
n
f
o
u
r
th
s
tep
,
in
ter
s
ec
tio
n
o
v
er
u
n
i
o
n
(
I
o
U
)
s
co
r
e
was
co
m
p
u
ted
th
r
o
u
g
h
d
etec
ted
v
e
h
icles
with
g
r
o
u
n
d
tr
u
t
h
.
I
n
f
if
th
s
tep
,
all
v
er
if
ied
v
eh
icles
p
er
f
o
r
m
ed
v
is
u
al
g
eo
m
etr
y
g
r
o
u
p
(
VGG)
f
ea
tu
r
e
ex
tr
ac
tio
n
.
A
u
n
iq
u
e
id
en
tifie
r
was
allo
ca
ted
to
allo
w
m
u
lti
-
o
b
ject
tr
ac
k
in
g
ac
r
o
s
s
im
ag
e
f
r
am
es.
I
n
s
ix
th
s
tep
,
th
e
v
eh
icl
es we
r
e
co
u
n
ted
an
d
class
if
ied
in
to
s
tatio
n
ar
y
as we
l
l a
s
m
o
v
in
g
ca
r
s
th
r
o
u
g
h
d
etec
tin
g
m
o
tio
n
by
f
ar
n
eb
ac
k
o
p
tical
f
lo
w
alg
o
r
ith
m
.
Simp
le
o
n
lin
e
a
n
d
r
ea
l
-
tim
e
tr
ac
k
er
(
SOR
T
)
was
e
m
p
lo
y
ed
f
o
r
e
f
f
icien
t
tr
ac
k
in
g
.
T
h
e
d
esig
n
e
d
m
o
d
el
in
cr
ea
s
ed
p
r
ec
is
io
n
f
o
r
d
etec
ti
o
n
with
v
is
io
n
m
ee
ts
d
r
o
n
e
s
in
g
le
o
b
ject
-
tr
ac
k
i
n
g
(
Vis
Dr
o
n
e)
d
ataset.
3
.2
.
Rea
l
-
t
im
e
t
ra
f
f
ic
co
ntr
o
l a
nd
m
o
nito
ring
C
o
n
g
esti
o
n
h
as
b
ec
o
m
e
k
e
y
p
r
o
b
lem
.
An
a
u
to
m
o
b
ile
r
o
a
d
tr
af
f
ic
d
en
s
ity
is
an
ess
en
tial
o
n
e
f
o
r
en
h
an
ce
d
tr
af
f
ic
s
ig
n
al
co
n
tr
o
l
an
d
ef
f
icien
t
tr
af
f
ic
m
a
n
ag
em
en
t.
T
r
a
f
f
ic
co
n
g
esti
o
n
o
cc
u
r
r
e
d
d
u
e
to
in
s
u
f
f
icien
t
ca
p
ab
ilit
y
.
C
ap
ac
i
ty
lim
itatio
n
s
as
well
as
d
em
an
d
co
n
s
tr
ain
ts
ar
e
co
n
n
ec
ted
.
E
v
er
y
s
ig
n
al
h
o
ld
u
p
is
f
ir
m
c
o
d
ed
as
well
as
t
r
af
f
i
c
in
d
e
p
en
d
e
n
t.
An
im
ag
e
p
r
o
c
ess
in
g
as
well
as
s
u
r
v
e
illan
ce
s
ch
em
es
p
er
f
o
r
m
ed
b
y
p
ass
en
g
er
d
ata,
an
d
s
o
o
n
.
Mo
v
in
g
au
t
o
m
o
b
ile
tr
ac
in
g
im
ag
e
p
r
o
v
i
d
es
th
e
q
u
an
titat
iv
e
ex
p
lan
atio
n
o
f
tr
af
f
ic
f
lo
w.
T
h
e
r
ea
l
-
tim
e
liv
e
v
id
eo
f
ee
d
s
wer
e
u
s
ed
f
r
o
m
ca
m
er
as
at
in
ter
s
ec
tio
n
s
to
ex
ec
u
te
in
s
tan
tan
eo
u
s
tr
af
f
ic
b
u
lk
in
ess
co
m
p
u
tatio
n
s
th
r
o
u
g
h
im
ag
e
p
r
o
ce
s
s
in
g
an
d
v
eh
icle
d
etec
tio
n
with
h
elp
o
f
E
f
f
icien
tDet
ar
ch
itectu
r
e
a
n
d
T
en
s
o
r
Flo
w
l
ite.
T
h
e
m
ain
o
b
jectiv
e
o
f
t
h
e
ar
ch
itectu
r
e
was
to
m
in
im
ize
th
e
tr
af
f
ic
jam
s
an
d
ac
cid
en
ts
th
at
s
witch
s
ig
n
al
lig
h
ts
b
ased
o
n
v
e
h
icle
d
en
s
it
y
o
n
r
o
ad
a
n
d
p
r
io
r
ity
s
et
f
o
r
p
a
r
ticu
lar
em
er
g
en
c
y
v
eh
icles.
T
h
e
d
esig
n
ed
ar
ch
it
ec
tu
r
e
p
r
esen
ted
th
e
p
eo
p
le
w
ith
s
af
e
tr
an
s
p
o
r
tatio
n
,
r
ed
u
ce
d
f
u
el
co
n
s
u
m
p
tio
n
an
d
waitin
g
tim
e.
Veh
icle
r
ec
o
g
n
itio
n
was
p
er
f
o
r
m
ed
d
u
r
in
g
s
y
s
tem
f
r
o
m
im
ag
es.
T
h
e
r
ec
o
g
n
itio
n
w
as
n
o
t
p
er
f
o
r
m
ed
d
u
r
in
g
elec
tr
o
n
ic
s
en
s
o
r
s
m
o
u
n
ted
o
n
r
o
ad
way
.
T
h
e
ca
m
er
a
in
s
tallatio
n
was
p
er
f
o
r
m
ed
n
e
x
t
to
tr
af
f
ic
lig
h
t
th
at
g
ath
er
e
d
th
e
v
id
eo
f
ee
d
s
en
t
to
R
asp
b
er
r
y
Pi.
T
h
e
d
esig
n
ed
ar
ch
itectu
r
e
h
an
d
led
t
h
e
tr
af
f
ic
lig
h
t tim
in
g
to
p
er
f
o
r
m
au
t
o
m
atic
co
n
tr
o
l o
f
tr
af
f
ic
s
itu
atio
n
s
.
3
.3
.
T
RAMON
:
a
n a
uto
ma
t
ed
t
ra
f
f
ic
m
o
nito
ring
s
y
s
t
em
f
o
r
hig
h dens
it
y
,
m
ix
ed
a
nd
la
ne
-
f
re
e
t
ra
f
f
ic
T
r
af
f
ic
co
n
g
esti
o
n
is
co
n
s
id
er
ed
as
th
e
k
ey
p
r
o
b
lem
in
citi
es.
T
r
af
f
ic
d
ata
lik
e
tr
a
f
f
ic
m
ea
n
s
p
ee
d
,
f
lo
w,
d
e
n
s
ity
an
d
tr
av
el
t
im
e
id
en
tifie
s
th
e
tr
af
f
ic
co
n
g
esti
o
n
h
o
ts
p
o
ts
an
d
attain
s
th
e
p
o
ten
tial
s
o
lu
tio
n
s
.
T
r
af
f
ic
d
ata
a
r
e
g
ath
e
r
ed
th
r
o
u
g
h
h
u
m
a
n
s
u
r
v
e
y
o
r
s
,
r
o
ad
tu
b
es,
in
d
u
ctio
n
lo
o
p
an
d
p
i
ez
o
elec
tr
ic
s
en
s
o
r
s
.
T
h
e
d
esig
n
ed
m
et
h
o
d
c
o
llected
th
e
d
ata
with
m
in
im
al
m
ain
ten
an
ce
co
s
ts
.
W
ith
d
ev
elo
p
m
en
t
in
tr
af
f
ic
d
ata
co
llectio
n
,
th
e
co
m
p
u
ter
v
is
io
n
o
r
ig
in
ated
with
a
v
ailab
ilit
y
o
f
DL
.
Vid
eo
an
d
im
ag
e
p
r
o
ce
s
s
in
g
u
s
in
g
DL
ex
tr
ac
ted
m
ac
r
o
s
co
p
ic
d
ata
an
d
m
icr
o
s
co
p
ic
d
ata.
No
v
e
l
v
is
u
al
d
atab
ase
ap
p
r
o
ac
h
was
in
tr
o
d
u
ce
d
f
o
r
ass
is
tin
g
T
R
AM
ON
in
en
h
an
ce
d
d
en
s
ity
,
ass
o
r
ted
as we
ll a
s
lan
e
-
f
r
ee
tr
af
f
ic.
An
ad
v
a
n
ce
d
DL
alg
o
r
ith
m
was
in
tr
o
d
u
ce
d
to
i
d
en
tify
a
n
d
t
r
ac
k
th
e
v
e
h
icles
f
r
o
m
tr
af
f
ic
v
id
eo
s
.
T
h
e
d
esig
n
ed
m
o
n
ito
r
in
g
m
eth
o
d
s
p
r
esen
te
d
ac
cu
r
ate
tr
af
f
ic
m
o
n
ito
r
in
g
i
n
m
ix
e
d
tr
a
f
f
ic.
T
h
e
m
ix
ed
tr
af
f
ic
f
l
o
ws
in
d
ev
elo
p
in
g
c
o
u
n
tr
ies
co
m
p
r
is
ed
v
eh
icles
ty
p
es.
T
h
e
co
m
p
u
ter
v
is
io
n
alg
o
r
ith
m
s
ex
p
er
ien
ce
d
d
if
f
icu
lties
in
id
en
tify
in
g
an
d
tr
a
ck
in
g
th
e
h
ig
h
d
en
s
ity
o
f
v
e
h
icles.
A
co
m
p
r
e
h
en
s
iv
e
f
r
a
m
ewo
r
k
was
em
p
lo
y
ed
to
tr
ain
d
ee
p
-
lear
n
in
g
-
b
as
ed
co
m
p
u
ter
v
is
io
n
alg
o
r
ith
m
s
d
etec
tin
g
t
h
e
v
eh
ic
les in
h
ig
h
d
e
n
s
ity
,
h
eter
o
g
en
e
o
u
s
an
d
lan
e
-
f
r
ee
tr
a
f
f
ic.
3
.4
.
I
ma
g
e
-
ba
s
ed
t
ra
f
f
ic
s
ig
na
l c
o
ntr
o
l v
ia
wo
rld
m
o
del
W
ith
g
r
o
wth
o
f
in
tellig
en
ce
tech
n
o
lo
g
ies,
tr
a
f
f
ic
s
ig
n
al
c
o
n
tr
o
l
p
er
f
o
r
m
ed
r
en
o
v
atio
n
as
o
f
th
e
p
r
ed
eter
m
in
e
d
-
tim
e
co
n
tr
o
l
to
p
r
o
ac
tiv
e
co
n
tr
o
l.
A
f
ix
ed
-
ti
m
e
co
n
tr
o
ller
p
h
ase
s
eq
u
en
ce
an
d
d
u
r
atio
n
wer
e
p
r
e
-
d
eter
m
i
n
ed
th
r
o
u
g
h
p
r
o
f
ess
io
n
al
en
g
in
ee
r
s
th
r
o
u
g
h
e
x
p
e
r
ien
ce
o
r
r
u
les.
T
r
af
f
ic
s
ig
n
al
co
n
tr
o
l
was
s
h
if
ted
as
o
f
p
ass
iv
e
to
p
r
o
ac
tiv
e
c
o
n
t
r
o
ls
f
o
r
allo
win
g
co
n
tr
o
ller
to
s
tr
aig
h
t
p
r
esen
t
tr
af
f
ic
f
lo
w
to
attain
d
esti
n
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
xp
lo
r
in
g
d
ivers
e
p
r
ed
ictio
n
mo
d
els in
in
tellig
en
t tra
ffic
co
n
tr
o
l
…
(
S
a
h
ir
a
V
ila
kk
u
ma
d
a
th
il)
397
An
ef
f
ec
tiv
e
p
r
ed
ictio
n
m
o
d
el
was
r
eq
u
ir
ed
f
o
r
p
e
r
f
o
r
m
in
g
s
ig
n
al
co
n
tr
o
lle
r
s
.
An
im
ag
e
was
em
p
lo
y
ed
with
v
eh
icle
p
o
s
itio
n
s
to
e
x
p
lain
th
e
in
ter
s
ec
tio
n
t
r
af
f
ic
s
tates.
A
m
o
d
el
-
b
ased
RL
m
eth
o
d
ter
m
ed
Dr
ea
m
er
V
2
was
in
tr
o
d
u
ce
d
f
o
e
lea
r
n
in
g
-
b
ased
tr
af
f
ic
wo
r
ld
a
n
aly
s
is
.
T
h
e
tr
af
f
ic
wo
r
ld
m
o
d
el
ex
p
lain
ed
t
h
e
tr
af
f
ic
d
y
n
am
ics
in
im
ag
e
f
o
r
m
.
T
h
e
d
esig
n
e
d
m
o
d
el
was
em
p
lo
y
ed
as
a
b
s
tr
ac
t
alter
n
ativ
e
to
cr
ea
te
m
u
lti
-
s
tep
p
lan
n
in
g
in
f
o
r
m
atio
n
.
W
o
r
ld
m
eth
o
d
w
as e
m
p
lo
y
ed
to
f
o
r
ec
ast th
e
im
p
ac
t o
f
d
iv
er
s
e
co
n
tr
o
l b
eh
a
v
io
r
s
o
n
f
u
tu
r
e
tr
a
f
f
ic
co
n
d
itio
n
s
.
3
.5
.
P
a
iring
Net
:
a
m
ulti
-
f
ra
m
e
ba
s
ed
v
ehi
cle
t
ra
j
ec
t
o
r
y
predict
io
n dee
p lea
rni
ng
net
wo
rk
T
r
af
f
ic
d
ata
co
llectio
n
as
wel
l
as
in
v
esti
g
atio
n
ar
e
co
n
s
id
e
r
ed
as
ess
en
tial
elem
en
ts
f
o
r
r
ea
l
-
tim
e
ad
m
in
is
tr
atio
n
o
f
tr
an
s
p
o
r
tat
io
n
n
etwo
r
k
s
.
I
T
S
was
em
p
lo
y
ed
to
p
er
f
o
r
m
ac
cu
r
ate
an
d
co
s
t
-
ef
f
ec
tiv
e
tech
n
iq
u
e
s
f
o
r
tr
a
f
f
ic
d
ata
co
l
lectio
n
.
UAV
with
v
id
eo
s
tr
ea
m
in
g
ca
p
ab
ilit
y
co
v
er
ed
wid
e
ar
ea
,
m
in
im
ized
in
s
tallatio
n
co
s
t
as
well
as
er
r
o
r
s
.
UAV
s
u
p
p
o
r
ted
b
y
i
m
ag
e
p
r
o
ce
s
s
in
g
g
ath
er
ed
m
icr
o
s
co
p
ic
tr
af
f
ic
in
f
o
r
m
atio
n
lik
e
v
eh
icle
ty
p
e,
d
ir
ec
tio
n
,
tr
ajec
to
r
y
,
s
p
ee
d
,
an
d
d
r
i
v
in
g
b
e
h
av
io
r
.
An
in
tellig
en
t
im
ag
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
we
r
e
n
ec
ess
ar
y
f
o
r
r
ec
o
g
n
izin
g
an
d
t
r
ac
k
in
g
th
e
o
b
jects
g
ath
er
ed
i
n
th
e
UAV
v
id
e
o
s
.
C
ity
tr
af
f
ic
in
f
r
astru
ctu
r
e
r
eq
u
ir
ed
ev
alu
atio
n
an
d
en
h
an
ce
m
en
t
th
r
o
u
g
h
lar
g
e
q
u
a
n
tity
o
f
d
ata
an
aly
s
is
.
T
h
e
co
n
s
tr
u
ctio
n
a
n
d
la
b
o
r
io
u
s
wo
r
k
p
e
r
f
o
r
m
ed
v
is
io
n
en
h
an
ce
m
en
t
in
tr
a
f
f
ic
an
al
y
s
is
.
Am
o
n
g
d
iv
er
s
e
i
n
tellig
en
t
tr
an
s
p
o
r
tatio
n
s
y
s
tem
,
MO
T
was
ca
r
r
ied
o
u
t.
MO
T
was
e
m
p
lo
y
ed
in
tr
af
f
ic
an
aly
s
is
with
v
eh
icle
s
p
ee
d
,
m
o
v
em
en
t
d
ir
ec
tio
n
as
well
as
co
n
s
id
er
atio
n
o
f
s
in
g
le
f
a
cto
r
in
t
r
ajec
to
r
y
tr
ac
k
in
g
.
An
en
d
-
to
-
en
d
DL
n
etwo
r
k
ter
m
e
d
Pair
in
g
Net
w
as
in
tr
o
d
u
ce
d
.
Pair
in
g
Net
co
m
b
in
ed
th
e
v
eh
icle
tr
ajec
to
r
y
to
n
etwo
r
k
d
u
r
i
n
g
f
ea
tu
r
e
s
y
n
th
esis
o
f
s
u
cc
ess
iv
e
im
ag
es.
Pair
in
g
Net
was
d
e
s
ig
n
ed
to
f
o
r
ec
ast
p
r
o
g
r
ess
d
ir
ec
tio
n
as
well
as
v
eh
icle
s
p
ee
d
th
r
o
u
g
h
r
etain
in
g
f
u
n
ctio
n
an
d
ac
cu
r
ac
y
.
I
n
d
e
s
ig
n
ed
n
etwo
r
k
,
ad
d
itio
n
al
f
ea
tu
r
es
wer
e
u
s
ed
to
tr
ac
k
th
e
v
e
h
icle
tr
ajec
to
r
y
.
A
p
ip
elin
e
was
u
s
ed
to
m
in
im
ize
th
e
lo
ad
in
g
laten
c
y
ac
q
u
ir
e
d
th
r
o
u
g
h
co
n
s
ec
u
tiv
e
f
r
am
es f
o
r
Pair
in
g
Net.
Pair
in
g
Net
in
cr
ea
s
ed
th
e
ac
cu
r
ac
y
r
ate
d
u
r
in
g
v
e
h
icle
tr
ajec
to
r
y
p
lan
n
in
g
.
3
.6
.
A
re
a
l
-
t
im
e
v
ehicle
re
co
g
nitio
n
a
nd
new
v
ehicle
t
ra
c
k
ing
s
chem
es
f
o
r
det
er
m
ini
ng
a
nd
m
o
nito
rin
g
t
ra
f
f
ic
o
n hig
hwa
y
s
R
ea
l
-
tim
e
h
ig
h
way
tr
af
f
ic
m
o
n
ito
r
in
g
s
y
s
tem
s
was
a
n
ess
en
tial
p
r
o
b
lem
in
r
o
ad
tr
af
f
ic
ad
m
in
is
tr
atio
n
an
d
s
o
o
n
.
T
h
e
tr
af
f
ic
m
o
n
ito
r
i
n
g
s
y
s
tem
d
e
p
en
d
s
o
n
o
n
lin
e
tr
af
f
ic
f
lo
w
f
r
o
m
tim
e
-
d
e
p
en
d
e
n
t
v
eh
icle
tr
ajec
to
r
ies.
Veh
icle
r
o
u
tes
wer
e
ex
to
r
te
d
as
o
f
v
eh
i
cle
r
ec
o
g
n
itio
n
an
d
d
ata
tr
ac
k
i
n
g
attain
ed
t
h
r
o
u
g
h
r
o
ad
-
s
id
e
ca
m
e
r
a
im
ag
e
p
r
o
ce
s
s
in
g
.
Yo
u
o
n
l
y
lo
o
k
o
n
ce
(
Y
OL
O
)
was
f
av
o
r
ed
as
it
p
r
es
en
t
elev
ated
f
r
am
es
p
er
s
ec
o
n
d
(
FPS
)
co
n
ce
r
t
as
well
as
o
b
ject
lo
ca
lizatio
n
f
u
n
ctio
n
ality
.
T
h
e
d
esig
n
ed
s
y
s
tem
in
cr
ea
s
ed
th
e
v
eh
icle
class
if
icatio
n
ac
cu
r
ac
y
f
o
r
tr
af
f
ic
f
lo
w
m
o
n
ito
r
in
g
.
T
h
e
B
b
o
x
-
b
ased
v
e
h
icle
tr
ac
k
in
g
alg
o
r
it
h
m
in
cr
ea
s
ed
th
e
v
eh
icle
class
i
f
icatio
n
ac
cu
r
ac
y
o
f
YOL
O
.
A
n
ew
v
eh
icle
d
ataset
was
g
ath
er
ed
with
th
e
o
b
ject
p
atter
n
s
g
ath
er
ed
f
r
o
m
h
ig
h
way
v
id
eo
s
.
3
.7
.
D
a
t
a
s
et
us
ed
Fo
r
co
n
d
u
ctin
g
th
e
e
x
p
er
im
e
n
t
an
aly
s
is
,
tr
af
f
ic
im
ag
e
d
ataset
is
u
s
ed
.
T
h
e
n
am
e
o
f
th
e
d
ataset
is
tr
af
f
ic
im
ag
es
o
f
v
e
h
icles
.
T
h
e
UR
L
o
f
th
e
d
ataset
is
h
ttp
s
://www.
k
ag
g
le.
co
m
/d
atasets
/th
er
ea
ls
h
ih
ab
/tra
f
f
ic
-
d
etec
tio
n
-
f
o
r
-
y
o
lo
v
5
.
T
r
af
f
ic
i
m
ag
es
o
f
v
eh
icles
d
ataset
co
m
p
r
is
es
th
e
s
et
o
f
tr
ain
a
n
d
v
alid
atio
n
im
ag
es
with
lab
els
o
f
tr
af
f
ic
co
n
d
itio
n
in
Dh
ak
a
city
f
o
r
tr
af
f
ic
im
ag
e
d
etec
tio
n
.
T
h
e
ca
p
ital
city
o
f
Dh
ak
a
h
as
o
n
ly
7
%
tr
af
f
ic
r
o
a
d
s
in
e
x
is
ten
ce
o
f
a
r
o
u
n
d
8
m
illi
o
n
c
o
m
p
u
ter
s
d
ay
.
Sit
u
atio
n
o
f
D
h
ak
a
tr
a
f
f
ic
is
d
is
tin
ctiv
e
with
in
s
u
r
m
o
u
n
ta
b
le
ch
allen
g
e
f
o
r
tr
af
f
ic
m
an
ag
em
e
n
t
s
y
s
tem
s
t
o
co
n
tr
o
l
an
d
m
ain
tain
th
e
s
m
o
o
th
f
lo
w
f
o
r
m
an
y
v
eh
icles.
An
au
to
m
atio
n
o
f
tr
af
f
ic
p
r
o
ce
s
s
is
m
o
s
t
o
p
tim
al
r
o
u
te
to
av
o
id
th
e
is
s
u
e
u
s
in
g
ad
v
an
ce
s
i
n
ar
tific
ial
in
tellig
en
ce
(
AI
)
-
b
as
ed
tech
n
o
lo
g
y
.
A
s
elf
ad
a
p
tiv
e
city
tr
af
f
ic
co
n
t
r
o
l sy
s
tem
is
ca
r
r
ied
o
u
t
b
ased
o
n
o
b
ject
d
etec
tio
n
a
n
d
DL
.
3
.8
.
E
v
a
lua
t
i
o
n m
et
rics
I
t
is
im
p
o
r
tan
t
to
co
m
p
u
te
th
e
tr
af
f
ic
co
n
tr
o
l
p
r
ed
ictio
n
p
er
f
o
r
m
an
ce
f
o
r
id
e
n
tify
in
g
h
o
w
ac
cu
r
ately
th
e
p
r
ed
icted
r
esu
lts
m
atch
t
h
e
ac
tu
al
o
n
es.
E
v
alu
atio
n
m
etr
ics
ar
e
u
s
ed
to
co
m
p
u
te
th
e
tr
af
f
ic
co
n
tr
o
l
p
r
ed
ictio
n
m
o
d
el.
T
h
e
m
etr
ics
ch
o
ice
d
ep
en
d
s
o
n
m
o
d
el
ty
p
e.
Pre
d
ictio
n
ac
cu
r
ac
y
,
p
r
e
d
ictio
n
tim
e
an
d
f
alse
p
o
s
itiv
e
r
ate
(
FP
R
)
u
s
ed
to
ca
lcu
late
tr
af
f
ic
co
n
tr
o
l
p
r
ed
icti
o
n
.
T
r
a
f
f
ic
co
n
tr
o
l
p
r
ed
ictio
n
ac
cu
r
ac
y
(
T
C
PA
)
is
d
escr
ib
ed
as
r
atio
o
f
n
u
m
b
e
r
o
f
tr
af
f
ic
im
ag
es
th
at
ar
e
a
cc
u
r
ately
p
r
e
d
icted
to
to
tal
t
r
af
f
ic
im
ag
es.
I
t
is
co
m
p
u
ted
in
p
e
r
ce
n
tag
e
(
%).
W
h
en
p
r
ed
ictio
n
ac
cu
r
ac
y
i
s
h
ig
h
er
,
th
e
m
eth
o
d
is
m
o
r
e
ef
f
icien
t.
FP
R
is
d
escr
ib
ed
as
th
e
r
atio
o
f
th
e
n
u
m
b
er
o
f
tr
af
f
ic
im
ag
es
t
h
at
ar
e
in
co
r
r
ec
tly
p
r
ed
icted
.
I
t
is
co
m
p
u
ted
in
p
er
ce
n
tag
e
(
%).
W
h
en
th
e
er
r
o
r
r
ate
is
less
er
,
th
e
m
eth
o
d
is
m
o
r
e
ef
f
icien
t.
T
h
e
tr
af
f
ic
c
o
n
tr
o
l
p
r
ed
ictio
n
ti
m
e
(
T
C
PT
)
is
d
escr
ib
ed
as
p
r
o
d
u
ct
o
f
n
u
m
b
e
r
o
f
tr
af
f
ic
im
ag
e
s
an
d
am
o
u
n
t
o
f
tim
e
co
n
s
u
m
ed
to
p
r
ed
ict
o
n
e
tr
af
f
ic
im
ag
e.
I
t is m
ea
s
u
r
ed
in
m
illi
s
ec
o
n
d
s
(
m
s
)
.
W
h
en
th
e
T
C
PT
is
le
s
s
er
,
th
e
m
eth
o
d
is
m
o
r
e
ef
f
icien
t
.
TCPA
=
n
umb
e
r
of
t
r
af
f
i
c
i
m
ag
es
t
hat
ar
e
accu
r
at
el
y
p
r
ed
i
ct
ed
n
umb
e
r
of
t
r
af
f
i
c
i
m
g
es
∗
100
(
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
3
9
3
-
4
0
2
398
F
PR
=
n
umb
er
of
t
r
af
f
i
c
i
m
ag
es
t
hat
ar
e
i
n
co
r
r
e
ct
el
y
p
r
ed
i
c
t
ed
n
umb
e
r
of
t
r
af
f
i
c
i
m
g
es
∗
100
(
2
)
TCPT
=
n
umb
e
r
of
tr
a
ff
ic
imge
s
∗
Ti
me
(
pr
e
dic
tin
g
on
e
ima
ge
)
(
3
)
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
r
esu
lt
is
ca
lcu
lated
in
T
C
PA
,
T
C
PT
,
an
d
FPR
b
y
u
s
in
g
d
if
f
er
e
n
t
class
if
icatio
n
m
eth
o
d
s
.
T
h
e
r
esu
lt
d
e
p
en
d
s
o
n
h
o
w
ac
cu
r
ate
p
r
o
p
o
s
ed
m
o
d
el
g
ets
tr
ain
ed
.
T
h
e
o
v
e
r
all
p
er
f
o
r
m
a
n
c
e
r
esu
lts
m
ea
s
u
r
es
ar
e
s
h
o
wn
i
n
tab
le
an
d
g
r
ap
h
s
.
I
n
T
a
b
le
1
,
e
v
alu
atio
n
o
f
d
if
f
er
en
t
tr
af
f
ic
co
n
tr
o
l
p
r
e
d
ictio
n
tech
n
i
q
u
es
is
ca
r
r
ied
o
n
tr
a
f
f
ic
im
ag
e
d
ataset.
T
h
e
ta
b
le
attain
ed
th
e
ac
cu
r
ac
y
v
al
u
es
o
f
8
6
.
2
5
%
to
9
6
.
8
6
%
o
n
o
r
i
g
in
al
d
ataset.
New
s
m
ar
t
tr
a
f
f
ic
co
n
tr
o
l
s
y
s
tem
h
as
attain
ed
le
ast
ac
cu
r
ac
y
o
f
8
6
.
2
5
%.
T
r
a
f
f
i
c
m
o
n
ito
r
in
g
s
y
s
tem
h
as
p
r
o
d
u
ce
d
h
i
g
h
est
T
C
PA
9
6
.
8
6
%
o
n
tr
af
f
ic
im
ag
e
d
ata
s
et.
T
h
e
T
C
PT
v
alu
es
r
an
g
es
f
r
o
m
2
9
m
s
to
5
5
m
s
.
T
r
a
f
f
ic
m
o
n
ito
r
in
g
s
y
s
tem
co
n
s
u
m
ed
less
er
T
C
PT
th
an
co
n
v
en
tio
n
al
m
eth
o
d
s
.
T
h
e
lea
r
n
in
g
cu
r
v
e
s
o
f
all
ex
is
tin
g
m
eth
o
d
s
with
tr
af
f
ic
im
ag
e
d
ataset
d
escr
ib
e
th
e
tr
a
f
f
ic
co
n
tr
o
l
p
r
ed
ictio
n
r
esu
lts
.
T
ab
le
1
.
Ov
e
r
all
r
esu
lts
f
o
r
tr
a
f
f
ic
co
n
tr
o
p
r
e
d
ictio
n
m
eth
o
d
Ex
i
s
t
i
n
g
t
e
c
h
n
i
q
u
e
s
TC
P
A
(
%)
TC
P
T
(
ms)
FPR
(
%)
N
e
w
smar
t
t
r
a
f
f
i
c
c
o
n
t
r
o
l
sy
s
t
e
m
8
6
.
2
5
55
1
3
.
7
5
Ef
f
i
c
i
e
n
t
D
e
t
a
r
c
h
i
t
e
c
t
u
r
e
8
8
.
9
2
51
1
1
.
0
8
V
i
su
a
l
d
a
t
a
s
e
t
f
r
a
mew
o
r
k
9
1
.
5
6
45
8
.
4
4
D
r
e
a
merV
2
9
2
.
7
8
40
7
.
2
2
P
a
i
r
i
n
g
N
e
t
9
5
.
2
1
36
4
.
7
9
Tr
a
f
f
i
c
m
o
n
i
t
o
r
i
n
g
s
y
s
t
e
m
9
6
.
8
9
29
3
.
1
1
Fig
u
r
e
3
illu
s
tr
ates
th
e
lear
n
i
n
g
cu
r
v
e
o
f
n
ew
s
m
ar
t
tr
af
f
i
c
co
n
tr
o
l
s
ch
em
e
with
tr
a
f
f
ic
im
ag
es
o
f
v
eh
icles
.
T
h
e
g
r
a
p
h
s
h
o
ws
th
e
tr
ain
in
g
s
co
r
e
a
n
d
v
alid
atio
n
s
co
r
e
f
o
r
d
iv
e
r
s
e
n
u
m
b
er
o
f
tr
ain
in
g
ex
a
m
p
les.
W
h
en
th
e
tr
ain
i
n
g
ex
am
p
les
a
r
e
less
er
,
th
e
c
r
o
s
s
-
v
alid
atio
n
s
co
r
e
is
‘
0
.
7
5
’
an
d
tr
ain
in
g
s
c
o
r
e
is
‘
0
.
9
9
’
.
W
h
en
th
e
tr
ain
in
g
ex
am
p
les
g
et
in
c
r
ea
s
ed
,
th
e
cr
o
s
s
-
v
alid
atio
n
s
co
r
e
g
ets
s
lo
wly
in
c
r
ea
s
ed
an
d
r
ea
ch
es
th
e
v
alu
e
‘
0
.
8
4
’
.
T
h
e
tr
ain
in
g
s
co
r
e
g
ets
s
lo
wly
r
ed
u
ce
d
an
d
r
ea
ch
es t
h
e
v
alu
e
‘
0
.
8
6
’
.
Fig
u
r
e
3
.
L
ea
r
n
in
g
cu
r
v
e
o
f
n
e
w
s
m
ar
t tr
af
f
ic
co
n
tr
o
l sch
em
e
with
tr
af
f
ic
im
ag
es o
f
v
eh
icle
s
Fig
u
r
e
4
illu
s
tr
ates
th
e
lear
n
i
n
g
cu
r
v
e
o
f
E
f
f
icien
tDet
a
r
ch
itectu
r
e
with
tr
af
f
ic
im
ag
es
o
f
v
eh
icles
.
T
h
e
g
r
a
p
h
s
h
o
ws
th
e
tr
ain
i
n
g
s
co
r
e
an
d
v
alid
atio
n
s
co
r
e
f
o
r
d
iv
e
r
s
e
n
u
m
b
e
r
o
f
tr
ain
in
g
ex
am
p
les
r
an
g
in
g
f
r
o
m
2
0
0
to
2
,
0
0
0
.
W
h
en
th
e
tr
ain
in
g
e
x
am
p
le
is
2
0
0
,
th
e
c
r
o
s
s
-
v
alid
atio
n
s
co
r
e
is
‘
0
.
8
’
a
n
d
tr
ai
n
in
g
s
co
r
e
is
‘
0
.
9
5
’
.
W
h
en
th
e
tr
ain
i
n
g
ex
a
m
p
les
g
et
in
cr
ea
s
ed
to
2
,
0
0
0
,
t
h
e
cr
o
s
s
-
v
alid
atio
n
s
co
r
e
g
ets
s
lo
wly
in
cr
ea
s
ed
to
th
e
v
alu
e
‘
0
.
8
6
’
.
T
h
e
tr
ain
in
g
s
co
r
e
g
ets s
lo
wly
r
ed
u
ce
d
an
d
r
ea
ch
es th
e
v
alu
e
‘
0
.
8
8
’.
Fig
u
r
e
5
illu
s
tr
ates
th
e
lear
n
i
n
g
cu
r
v
e
o
f
v
is
u
al
d
ataset
f
r
a
m
ewo
r
k
with
tr
a
f
f
ic
im
ag
es
o
f
v
eh
icles
.
T
h
e
g
r
ap
h
s
h
o
ws
th
e
tr
ain
in
g
s
co
r
e
an
d
cr
o
s
s
v
alid
atio
n
s
co
r
e
f
o
r
d
iv
e
r
s
e
n
u
m
b
er
o
f
tr
ain
in
g
e
x
am
p
les
0
.
7
0
.
7
5
0
.
8
0
.
8
5
0
.
9
0
.
9
5
1
1
.
0
5
2
5
0
5
0
0
7
5
0
1
0
0
0
1
2
5
0
1
5
0
0
1
7
5
0
2
0
0
0
S
c
o
r
e
T
r
a
i
n
i
n
g
E
x
a
m
p
l
e
s
Tr
a
i
n
i
n
g
S
c
o
r
e
C
r
o
ss-
V
a
l
i
d
a
t
i
o
n
S
c
o
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
xp
lo
r
in
g
d
ivers
e
p
r
ed
ictio
n
mo
d
els in
in
tellig
en
t tra
ffic
co
n
tr
o
l
…
(
S
a
h
ir
a
V
ila
kk
u
ma
d
a
th
il)
399
r
an
g
in
g
f
r
o
m
2
0
0
t
o
2
,
0
0
0
.
W
h
en
th
e
tr
ain
in
g
ex
am
p
le
is
2
0
0
,
th
e
cr
o
s
s
-
v
alid
atio
n
s
co
r
e
is
‘
0
.
8
2
’
an
d
tr
ai
n
in
g
s
co
r
e
is
‘
0
.
9
9
’
.
W
h
en
th
e
t
r
a
in
in
g
ex
am
p
les
g
et
in
cr
ea
s
ed
to
2
,
0
0
0
,
t
h
e
cr
o
s
s
-
v
alid
atio
n
s
co
r
e
g
ets
s
lo
wly
in
cr
ea
s
ed
to
‘
0
.
9
0
’
an
d
th
e
tr
ai
n
in
g
s
co
r
e
g
ets s
lo
wly
r
ed
u
ce
d
an
d
r
ea
ch
es
th
e
v
al
u
e
‘
0
.
9
1
’
.
Fig
u
r
e
4
.
L
ea
r
n
in
g
cu
r
v
e
o
f
E
f
f
icien
tDet
a
r
ch
itectu
r
e
with
tr
af
f
ic
im
ag
es o
f
v
eh
icles
Fig
u
r
e
5
.
L
ea
r
n
in
g
cu
r
v
e
o
f
v
i
s
u
al
d
ataset
f
r
am
ewo
r
k
with
tr
af
f
ic
im
ag
es o
f
v
eh
icles
Fig
u
r
e
6
illu
s
tr
ates
th
e
lear
n
in
g
cu
r
v
e
o
f
Dr
ea
m
e
r
V2
with
tr
af
f
ic
im
ag
es
o
f
v
eh
icles
.
T
h
e
g
r
a
p
h
s
h
o
ws th
e
tr
ain
in
g
s
co
r
e
an
d
c
r
o
s
s
v
alid
atio
n
s
co
r
e
f
o
r
d
iv
e
r
s
e
n
u
m
b
er
o
f
tr
ain
i
n
g
ex
am
p
les r
an
g
in
g
f
r
o
m
2
0
0
to
2
,
0
0
0
.
W
h
en
th
e
tr
ain
in
g
ex
am
p
le
is
2
0
0
,
th
e
cr
o
s
s
-
v
alid
atio
n
s
co
r
e
is
‘
0
.
8
5
’
an
d
tr
ain
in
g
s
co
r
e
is
‘
1
’
.
W
h
en
th
e
t
r
ain
in
g
ex
am
p
les
g
et
in
cr
ea
s
ed
to
2
,
0
0
0
,
th
e
c
r
o
s
s
-
v
alid
atio
n
s
co
r
e
g
ets
s
lo
wly
in
cr
ea
s
ed
to
‘
0
.
9
0
’
an
d
th
e
tr
ain
i
n
g
s
co
r
e
g
ets s
lo
wly
r
ed
u
ce
d
an
d
r
ea
ch
es th
e
v
alu
e
‘
0
.
9
2
’
.
Fig
u
r
e
7
illu
s
tr
ates
th
e
lear
n
in
g
cu
r
v
e
o
f
Pair
in
g
Net
with
tr
af
f
ic
im
ag
es
o
f
v
e
h
icles
.
T
h
e
g
r
ap
h
s
h
o
ws th
e
tr
ain
in
g
s
co
r
e
an
d
c
r
o
s
s
v
alid
atio
n
s
co
r
e
f
o
r
d
iv
e
r
s
e
n
u
m
b
er
o
f
tr
ain
i
n
g
ex
am
p
les r
an
g
in
g
f
r
o
m
2
0
0
to
2
,
0
0
0
.
W
h
en
th
e
tr
ain
in
g
e
x
am
p
le
is
2
0
0
,
th
e
cr
o
s
s
-
v
alid
atio
n
s
co
r
e
is
‘
0
.
8
8
’
an
d
tr
ain
in
g
s
co
r
e
is
‘
0
.
9
9
’
.
W
h
en
th
e
t
r
ain
in
g
ex
am
p
les
g
et
in
cr
ea
s
ed
to
2
,
0
0
0
,
th
e
c
r
o
s
s
-
v
alid
atio
n
s
co
r
e
g
ets
s
lo
wly
in
cr
ea
s
ed
to
‘
0
.
9
4
’
an
d
th
e
tr
ain
i
n
g
s
co
r
e
g
ets s
lo
wly
r
ed
u
ce
d
an
d
r
ea
ch
es th
e
v
alu
e
‘
0
.
9
6
’
.
F
ig
u
r
e
8
s
y
m
b
o
lizes
th
e
l
ea
r
n
in
g
cu
r
es
o
f
tr
af
f
ic
m
o
n
ito
r
in
g
s
y
s
tem
with
tr
af
f
ic
im
ag
es
o
f
v
eh
icles.
T
h
e
g
r
ap
h
s
h
o
ws
th
e
tr
ain
in
g
s
co
r
e
an
d
cr
o
s
s
v
alid
atio
n
s
co
r
e
f
o
r
d
iv
e
r
s
e
n
u
m
b
er
o
f
tr
ain
in
g
e
x
am
p
les
r
an
g
in
g
f
r
o
m
2
0
0
t
o
2
,
0
0
0
.
W
h
en
th
e
tr
ain
in
g
ex
am
p
le
is
2
0
0
,
th
e
cr
o
s
s
-
v
alid
at
io
n
s
co
r
e
is
‘
0
.
8
9
’
an
d
tr
ai
n
in
g
s
co
r
e
is
‘
0
.
9
9
’
.
W
h
en
th
e
t
r
a
in
in
g
ex
am
p
les
g
et
in
cr
ea
s
ed
to
2
,
0
0
0
,
t
h
e
cr
o
s
s
-
v
alid
atio
n
s
co
r
e
g
ets
s
lo
wly
in
cr
ea
s
ed
to
‘
0
.
9
4
’
a
n
d
th
e
tr
a
in
in
g
s
co
r
e
g
ets
s
lo
wly
r
ed
u
ce
d
an
d
r
ea
c
h
es
th
e
v
alu
e
‘
0
.
9
5
’
.
Fro
m
th
e
lear
n
in
g
cu
r
v
e
,
th
e
tr
af
f
ic
c
o
n
tr
o
l
p
r
ed
i
ctio
n
ac
cu
r
ac
y
v
alu
es a
r
e
o
b
ta
in
ed
.
0
.
7
0
.
7
5
0
.
8
0
.
8
5
0
.
9
0
.
9
5
1
2
5
0
5
0
0
7
5
0
1
0
0
0
1
2
5
0
1
5
0
0
1
7
5
0
2
0
0
0
S
c
o
r
e
T
r
a
i
n
i
n
g
E
x
a
m
p
l
e
s
Tr
a
i
n
i
n
g
S
c
o
r
e
C
r
o
ss-
V
a
l
i
d
a
t
i
o
n
S
c
o
r
e
0
.
8
0
.
8
2
0
.
8
4
0
.
8
6
0
.
8
8
0
.
9
0
.
9
2
0
.
9
4
0
.
9
6
0
.
9
8
1
2
5
0
5
0
0
7
5
0
1
0
0
0
1
2
5
0
1
5
0
0
1
7
5
0
2
0
0
0
S
c
o
r
e
T
r
a
i
n
i
n
g
E
x
a
m
p
l
e
s
Tr
a
i
n
i
n
g
S
c
o
r
e
C
r
o
ss-
V
a
l
i
d
a
t
i
o
n
S
c
o
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
3
9
3
-
4
0
2
400
Fig
u
r
e
6
.
L
ea
r
n
in
g
cu
r
v
e
o
f
D
r
ea
m
er
V2
with
tr
af
f
ic
im
ag
es
o
f
v
eh
icles
Fig
u
r
e
7
.
L
ea
r
n
in
g
cu
r
v
e
o
f
Pa
ir
in
g
Net
with
tr
af
f
ic
im
ag
es o
f
v
eh
icles
Fig
u
r
e
8
.
L
ea
r
n
in
g
cu
r
v
e
o
f
t
r
af
f
ic
m
o
n
ito
r
in
g
s
y
s
tem
s
with
tr
af
f
ic
im
ag
es o
f
v
eh
icles
0
.
7
5
0
.
8
0
.
8
5
0
.
9
0
.9
5
1
1
.
0
5
2
5
0
5
0
0
7
5
0
1
0
0
0
1
2
5
0
1
5
0
0
1
7
5
0
2
0
0
0
S
c
o
r
e
T
r
a
i
n
i
n
g
E
x
a
m
p
l
e
s
Tr
a
i
n
i
n
g
S
c
o
r
e
C
r
o
ss-
V
a
l
i
d
a
t
i
o
n
S
c
o
r
e
0
.
8
2
0
.
8
4
0
.
8
6
0
.
8
8
0
.
9
0
.
9
2
0
.
9
4
0
.
9
6
0
.
9
8
1
2
5
0
5
0
0
7
5
0
1
0
0
0
1
2
5
0
1
5
0
0
1
7
5
0
2
0
0
0
S
c
o
r
e
T
r
a
i
n
i
n
g
E
x
a
m
p
l
e
s
Tr
a
i
n
i
n
g
S
c
o
r
e
C
r
o
ss-
V
a
l
i
d
a
t
i
o
n
S
c
o
r
e
0
.
8
2
0
.8
4
0
.
8
6
0
.
8
8
0
.
9
0
.
9
2
0
.
9
4
0
.
9
6
0
.
9
8
1
2
5
0
5
0
0
7
5
0
1
0
0
0
1
2
5
0
1
5
0
0
1
7
5
0
2
0
0
0
S
c
o
r
e
T
r
a
i
n
i
n
g
E
x
a
m
p
l
e
s
Tr
a
i
n
i
n
g
S
c
o
r
e
C
r
o
ss-
V
a
l
i
d
a
t
i
o
n
S
c
o
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
xp
lo
r
in
g
d
ivers
e
p
r
ed
ictio
n
mo
d
els in
in
tellig
en
t tra
ffic
co
n
tr
o
l
…
(
S
a
h
ir
a
V
ila
kk
u
ma
d
a
th
il)
401
5.
CO
NCLU
SI
O
N
Sev
er
al
r
esear
ch
was
p
er
f
o
r
m
ed
o
n
tr
a
f
f
ic
m
an
a
g
em
en
t
s
y
s
tem
.
I
n
tellig
en
t
tr
af
f
ic
m
o
n
ito
r
in
g
is
an
ac
tiv
e
r
esear
ch
to
p
ic
b
ec
au
s
e
o
f
em
er
g
in
g
tech
n
o
lo
g
ies
lik
e
I
o
T
as
wel
l
a
s
AI
.
T
h
e
in
teg
r
atio
n
o
f
tech
n
o
lo
g
ies
was
em
p
lo
y
ed
f
o
r
d
ec
is
io
n
m
ak
in
g
an
d
attai
n
ed
u
r
b
a
n
g
r
o
wth
.
I
n
th
is
w
o
r
k
,
tr
af
f
ic
co
n
tr
o
l
p
r
ed
ictio
n
was
attem
p
ted
u
s
in
g
d
iv
er
s
e
ML
tech
n
iq
u
es.
T
h
r
ee
p
er
f
o
r
m
a
n
ce
p
ar
am
eter
s
w
er
e
u
s
ed
to
an
aly
ze
r
esu
lts
o
f
d
esig
n
ed
m
o
d
els
im
p
lem
en
ted
f
o
r
tr
af
f
ic
c
o
n
tr
o
l
p
r
ed
ictio
n
with
tr
a
f
f
ic
im
ag
e
d
atasets
.
W
h
en
co
m
p
ar
in
g
th
e
r
esu
lt
with
an
o
t
h
er
r
ec
en
t
s
tu
d
y
,
t
r
af
f
ic
m
o
n
ito
r
in
g
s
y
s
tem
h
as
p
r
o
d
u
ce
d
h
ig
h
est
tr
af
f
ic
co
n
tr
o
l
p
r
ed
ictio
n
ac
cu
r
ac
y
o
f
9
6
.
8
9
%
th
an
o
th
er
f
iv
e
ex
is
tin
g
m
e
th
o
d
s
.
T
h
ese
o
u
tco
m
es
p
o
wer
f
u
lly
p
r
o
p
o
s
e
wh
ich
g
r
ad
ien
t
tr
a
n
s
f
er
lear
n
i
n
g
m
o
d
el
ca
n
b
e
im
p
lem
e
n
ted
f
o
r
tr
af
f
ic
c
o
n
tr
o
l
p
r
ed
ictio
n
i
n
s
tead
o
f
th
e
o
th
er
ex
is
tin
g
ML
class
if
ier
s
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
A
l
o
n
a
z
i
e
t
a
l
.
,
“
A
sm
a
r
t
t
r
a
f
f
i
c
c
o
n
t
r
o
l
s
y
st
e
m
b
a
se
d
o
n
p
i
x
e
l
-
l
a
b
e
l
i
n
g
a
n
d
S
O
R
T
t
r
a
c
k
e
r
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
8
0
9
7
3
–
8
0
9
8
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
2
9
9
4
8
8
.
[
2
]
M
.
G
u
p
t
a
,
H
.
M
i
g
l
a
n
i
,
P
.
D
e
o
,
a
n
d
A
.
B
a
r
h
a
t
t
e
,
“
R
e
a
l
-
t
i
me
t
r
a
f
f
i
c
c
o
n
t
r
o
l
a
n
d
mo
n
i
t
o
r
i
n
g
,
”
e
-
Pri
m
e
-
Ad
v
a
n
c
e
s
i
n
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
,
E
l
e
c
t
r
o
n
i
c
s
a
n
d
E
n
e
rg
y
,
v
o
l
.
5
,
p
.
1
0
0
2
1
1
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
i
me
.
2
0
2
3
.
1
0
0
2
1
1
.
[
3
]
D
.
M
.
T
a
n
a
n
d
L
.
-
M
.
K
i
e
u
,
“
T
R
A
M
O
N
:
a
n
a
u
t
o
mat
e
d
t
r
a
f
f
i
c
mo
n
i
t
o
r
i
n
g
sy
s
t
e
m
f
o
r
h
i
g
h
d
e
n
s
i
t
y
,
mi
x
e
d
a
n
d
l
a
n
e
-
f
r
e
e
t
r
a
f
f
i
c
,
”
I
ATS
S
Re
s
e
a
rc
h
,
v
o
l
.
4
7
,
n
o
.
4
,
p
p
.
4
6
8
–
4
8
1
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
a
t
ss
r
.
2
0
2
3
.
1
0
.
0
0
1
.
[
4
]
X
.
D
a
i
,
C
.
Z
h
a
o
,
X
.
W
a
n
g
,
Y
.
L
v
,
Y
.
L
i
n
,
a
n
d
F
.
-
Y
.
W
a
n
g
,
“
I
mag
e
-
b
a
s
e
d
t
r
a
f
f
i
c
si
g
n
a
l
c
o
n
t
r
o
l
v
i
a
w
o
r
l
d
m
o
d
e
l
s,
”
Fro
n
t
i
e
rs
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
&
El
e
c
t
ro
n
i
c
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
3
,
n
o
.
1
2
,
p
p
.
1
7
9
5
–
1
8
1
3
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
6
3
1
/
F
I
TEE.
2
2
0
0
3
2
3
.
[
5
]
G.
-
W
.
C
h
e
n
,
M
.
-
T.
S
u
n
,
a
n
d
T.
-
U
.
İ
k
,
“
P
a
i
r
i
n
g
N
e
t
:
a
m
u
l
t
i
-
f
r
a
m
e
b
a
s
e
d
v
e
h
i
c
l
e
t
r
a
j
e
c
t
o
r
y
p
r
e
d
i
c
t
i
o
n
d
e
e
p
l
e
a
r
n
i
n
g
n
e
t
w
o
r
k
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
2
9
5
6
6
–
2
9
5
7
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
2
6
0
8
3
2
.
[
6
]
J
.
A
z
i
m
j
o
n
o
v
a
n
d
A
.
Ö
z
men
,
“
A
r
e
a
l
-
t
i
me
v
e
h
i
c
l
e
d
e
t
e
c
t
i
o
n
a
n
d
a
n
o
v
e
l
v
e
h
i
c
l
e
t
r
a
c
k
i
n
g
s
y
s
t
e
m
s
f
o
r
e
st
i
m
a
t
i
n
g
a
n
d
m
o
n
i
t
o
r
i
n
g
t
r
a
f
f
i
c
f
l
o
w
o
n
h
i
g
h
w
a
y
s
,
”
A
d
v
a
n
c
e
d
En
g
i
n
e
e
ri
n
g
I
n
f
o
r
m
a
t
i
c
s
,
v
o
l
.
5
0
,
p
.
1
0
1
3
9
3
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
e
i
.
2
0
2
1
.
1
0
1
3
9
3
.
[
7
]
S
.
Li
u
a
n
d
Y
.
B
a
i
,
“
M
u
l
t
i
p
l
e
U
A
V
s
c
o
l
l
a
b
o
r
a
t
i
v
e
t
r
a
f
f
i
c
mo
n
i
t
o
r
i
n
g
w
i
t
h
i
n
t
e
n
t
i
o
n
-
b
a
se
d
c
o
mm
u
n
i
c
a
t
i
o
n
,
”
C
o
m
p
u
t
e
r
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
2
1
0
,
p
p
.
1
1
6
–
1
2
9
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mc
o
m
.
2
0
2
3
.
0
8
.
0
0
5
.
[
8
]
R
.
B
a
r
b
o
sa
e
t
a
l
.
,
“
I
o
T
b
a
s
e
d
r
e
a
l
-
t
i
me
t
r
a
f
f
i
c
m
o
n
i
t
o
r
i
n
g
s
y
s
t
e
m
u
si
n
g
i
m
a
g
e
s
se
n
s
o
r
s
b
y
s
p
a
r
se
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m,
”
C
o
m
p
u
t
e
r
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
2
1
0
,
p
p
.
3
2
1
–
3
3
0
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mc
o
m
.
2
0
2
3
.
0
8
.
0
0
7
.
[
9
]
N
.
R
a
j
u
,
S
.
A
r
k
a
t
k
a
r
,
a
n
d
S
.
Ea
s
a
,
“
R
e
c
e
p
t
i
v
e
n
e
ss
a
n
g
l
e
:
a
n
e
w
su
r
r
o
g
a
t
e
saf
e
t
y
me
a
s
u
r
e
f
o
r
mo
n
i
t
o
r
i
n
g
t
r
a
f
f
i
c
s
a
f
e
t
y
,
”
I
ATS
S
Re
s
e
a
rc
h
,
v
o
l
.
4
7
,
n
o
.
4
,
p
p
.
5
2
6
–
5
3
4
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
a
t
ss
r
.
2
0
2
3
.
1
1
.
0
0
4
.
[
1
0
]
A
.
P
a
sc
a
l
e
e
t
a
l
.
,
“
R
o
a
d
t
r
a
f
f
i
c
n
o
i
se
mo
n
i
t
o
r
i
n
g
i
n
a
smar
t
c
i
t
y
:
se
n
s
o
r
a
n
d
m
o
d
e
l
-
b
a
se
d
a
p
p
r
o
a
c
h
,
”
T
r
a
n
sp
o
r
t
a
t
i
o
n
Re
se
a
rc
h
Pa
rt
D
:
T
ra
n
s
p
o
rt
a
n
d
E
n
v
i
ro
n
m
e
n
t
,
v
o
l
.
1
2
5
,
p
.
1
0
3
9
7
9
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
r
d
.
2
0
2
3
.
1
0
3
9
7
9
.
[
1
1
]
R
.
M
o
r
e
i
r
a
,
L
.
F
.
R
.
M
o
r
e
i
r
a
,
a
n
d
F
.
d
e
O
l
i
v
e
i
r
a
S
i
l
v
a
,
“
A
n
i
n
t
e
l
l
i
g
e
n
t
n
e
t
w
o
r
k
m
o
n
i
t
o
r
i
n
g
a
p
p
r
o
a
c
h
f
o
r
o
n
l
i
n
e
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
D
a
r
k
n
e
t
t
r
a
f
f
i
c
,
”
C
o
m
p
u
t
e
rs
a
n
d
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
1
0
,
p
.
1
0
8
8
5
2
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
e
l
e
c
e
n
g
.
2
0
2
3
.
1
0
8
8
5
2
.
[
1
2
]
P
.
K
.
S
h
u
k
l
a
,
P
.
M
a
h
e
s
h
w
a
r
y
,
E.
K
.
S
u
b
r
a
ma
n
i
a
n
,
V
.
J.
S
h
i
l
p
a
,
a
n
d
P
.
R
.
K
.
V
a
r
ma
,
“
Tr
a
f
f
i
c
f
l
o
w
m
o
n
i
t
o
r
i
n
g
i
n
so
f
t
w
a
r
e
-
d
e
f
i
n
e
d
n
e
t
w
o
r
k
u
si
n
g
m
o
d
i
f
i
e
d
r
e
c
u
r
s
i
v
e
l
e
a
r
n
i
n
g
,
”
P
h
y
s
i
c
a
l
C
o
m
m
u
n
i
c
a
t
i
o
n
,
v
o
l
.
5
7
,
p
.
1
0
1
9
9
7
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
h
y
c
o
m
.
2
0
2
2
.
1
0
1
9
9
7
.
[
1
3
]
K
.
M
a
r
c
i
n
i
u
k
a
n
d
B
.
K
o
s
t
e
k
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
l
i
e
d
t
o
a
c
o
u
s
t
i
c
-
b
a
s
e
d
r
o
a
d
t
r
a
f
f
i
c
mo
n
i
t
o
r
i
n
g
,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
0
7
,
p
p
.
1
0
8
7
–
1
0
9
5
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
2
.
0
9
.
1
6
4
.
[
1
4
]
S
.
D
h
i
n
g
r
a
,
R
.
B
.
M
a
d
d
a
,
R
.
P
a
t
a
n
,
P
.
Ji
a
o
,
K
.
B
a
r
r
i
,
a
n
d
A
.
H
.
A
l
a
v
i
,
“
I
n
t
e
r
n
e
t
o
f
t
h
i
ngs
-
b
a
s
e
d
f
o
g
a
n
d
c
l
o
u
d
c
o
mp
u
t
i
n
g
t
e
c
h
n
o
l
o
g
y
f
o
r
smar
t
t
r
a
f
f
i
c
mo
n
i
t
o
r
i
n
g
,
”
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
,
v
o
l
.
1
4
,
p
.
1
0
0
1
7
5
,
Ju
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
o
t
.
2
0
2
0
.
1
0
0
1
7
5
.
[1
5
]
X
.
Y
a
n
g
e
t
a
l
.
,
“
Tr
i
b
o
e
l
e
c
t
r
i
c
se
n
s
o
r
a
r
r
a
y
f
o
r
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
b
a
se
d
smar
t
t
r
a
f
f
i
c
m
o
n
i
t
o
r
i
n
g
a
n
d
ma
n
a
g
e
m
e
n
t
s
y
s
t
e
m
,
”
N
a
n
o
E
n
e
r
g
y
,
v
o
l
.
9
2
,
p
.
1
0
6
7
5
7
,
F
e
b
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
a
n
o
e
n
.
2
0
2
1
.
1
0
6
7
5
7
.
[
1
6
]
J.
Á
.
M
a
r
t
í
n
-
B
a
o
s,
L
.
R
o
d
r
i
g
u
e
z
-
B
e
n
i
t
e
z
,
R
.
G
a
r
c
í
a
-
R
ó
d
e
n
a
s,
a
n
d
J
.
Li
u
,
“
I
o
T
b
a
se
d
m
o
n
i
t
o
r
i
n
g
o
f
a
i
r
q
u
a
l
i
t
y
a
n
d
t
r
a
f
f
i
c
u
si
n
g
r
e
g
r
e
ss
i
o
n
a
n
a
l
y
si
s,”
A
p
p
l
i
e
d
S
o
f
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
1
5
,
p
.
1
0
8
2
8
2
,
J
a
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
so
c
.
2
0
2
1
.
1
0
8
2
8
2
.
[
1
7
]
V
.
H
a
r
a
n
t
o
v
á
,
J
.
G
n
a
p
,
M
.
D
o
č
k
a
l
i
k
,
a
n
d
M
.
Lo
m
a
n
,
“
M
o
n
i
t
o
r
i
n
g
o
f
t
e
r
r
i
t
o
r
y
p
o
l
l
u
t
i
o
n
b
y
p
a
r
t
i
c
u
l
a
t
e
ma
t
t
e
r
i
n
r
e
l
a
t
i
o
n
t
o
r
o
a
d
t
r
a
f
f
i
c
,
”
T
r
a
n
sp
o
rt
a
t
i
o
n
Re
s
e
a
rc
h
Pr
o
c
e
d
i
a
,
v
o
l
.
7
4
,
p
p
.
2
2
4
–
2
3
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
r
p
r
o
.
2
0
2
3
.
1
1
.
1
3
8
.
[
1
8
]
Z.
La
n
a
n
d
M
.
C
a
i
,
“
D
y
n
a
mi
c
t
r
a
f
f
i
c
n
o
i
se
ma
p
s
b
a
se
d
o
n
n
o
i
se
mo
n
i
t
o
r
i
n
g
a
n
d
t
r
a
f
f
i
c
sp
e
e
d
d
a
t
a
,
”
T
ra
n
s
p
o
r
t
a
t
i
o
n
Re
s
e
a
r
c
h
P
a
r
t
D
:
T
ra
n
s
p
o
rt
a
n
d
E
n
v
i
ro
n
m
e
n
t
,
v
o
l
.
9
4
,
p
.
1
0
2
7
9
6
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
r
d
.
2
0
2
1
.
1
0
2
7
9
6
.
[
1
9
]
A
.
B
u
r
r
e
l
l
o
e
t
a
l
.
,
“
Tr
a
f
f
i
c
L
o
a
d
Es
t
i
mat
i
o
n
f
r
o
m
S
t
r
u
c
t
u
r
a
l
H
e
a
l
t
h
M
o
n
i
t
o
r
i
n
g
s
e
n
s
o
r
s
u
s
i
n
g
s
u
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
,
”
S
u
s
t
a
i
n
a
b
l
e
C
o
m
p
u
t
i
n
g
:
I
n
f
o
rm
a
t
i
c
s
a
n
d
S
y
s
t
e
m
s
,
v
o
l
.
3
5
,
p
.
1
0
0
7
0
4
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
s
u
sc
o
m.
2
0
2
2
.
1
0
0
7
0
4
.
[
2
0
]
V
.
Y
a
mu
n
a
a
n
d
K
.
A
.
P
r
a
k
a
s
h
,
“
A
f
u
z
z
y
i
n
c
i
d
e
n
c
e
c
o
l
o
r
i
n
g
t
o
mo
n
i
t
o
r
t
r
a
f
f
i
c
f
l
o
w
w
i
t
h
m
i
n
i
ma
l
w
a
i
t
i
n
g
t
i
me
,
”
Ex
p
e
rt
S
y
st
e
m
s w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
8
6
,
p
.
1
1
5
6
6
4
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
1
.
1
1
5
6
6
4
.
[
2
1
]
R
.
K
me
ť
a
n
d
M
.
K
v
e
t
,
“
Tr
a
f
f
i
c
a
c
c
i
d
e
n
t
m
o
n
i
t
o
r
i
n
g
i
n
f
o
r
mat
i
o
n
s
y
s
t
e
m
o
f
t
h
e
s
e
l
e
c
t
e
d
r
e
g
i
o
n
,
”
T
r
a
n
sp
o
rt
a
t
i
o
n
Re
s
e
a
r
c
h
Pro
c
e
d
i
a
,
v
o
l
.
5
5
,
p
p
.
1
4
5
2
–
1
4
5
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
r
p
r
o
.
2
0
2
1
.
0
7
.
1
3
2
.
[
2
2
]
Y
.
W
a
n
g
e
t
a
l
.
,
“
F
l
e
x
M
o
n
:
A
f
l
e
x
i
b
l
e
a
n
d
f
i
n
e
-
g
r
a
i
n
e
d
t
r
a
f
f
i
c
mo
n
i
t
o
r
f
o
r
p
r
o
g
r
a
m
ma
b
l
e
n
e
t
w
o
r
k
s,
”
J
o
u
r
n
a
l
o
f
N
e
t
w
o
rk
a
n
d
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
0
1
,
p
.
1
0
3
3
4
4
,
M
a
y
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
n
c
a
.
2
0
2
2
.
1
0
3
3
4
4
.
[2
3
]
S
.
J.
K
a
m
b
l
e
a
n
d
M
.
R
.
K
o
u
n
t
e
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
o
n
t
r
a
f
f
i
c
c
o
n
g
e
s
t
i
o
n
m
o
n
i
t
o
r
i
n
g
sy
s
t
e
m
i
n
i
n
t
e
r
n
e
t
o
f
v
e
h
i
c
l
e
s,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
7
1
,
p
p
.
2
2
3
5
–
2
2
4
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
0
.
0
4
.
2
4
1
.
[2
4
]
A
.
H
.
A
b
d
u
l
l
a
h
,
S
.
B
i
n
S
u
d
i
n
,
F
.
S
.
A
.
S
a
a
d
,
M
.
K
.
A
.
H
a
s
s
a
n
,
M
.
I
.
A
h
mad
,
a
n
d
K
.
A
.
b
i
n
A
.
K
h
a
l
i
d
,
“
A
q
u
a
c
u
l
t
u
r
e
m
o
n
i
t
o
r
i
n
g
sy
st
e
m
u
si
n
g
m
u
l
t
i
-
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
a
n
d
a
d
a
p
t
i
v
e
n
e
u
r
o
f
u
z
z
y
i
n
f
e
r
e
n
c
e
sy
s
t
e
m
,
”
I
n
d
o
n
e
s
i
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
E
EC
S)
,
v
o
l
.
3
3
,
n
o
.
1
,
p
.
7
1
,
Ja
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.v
3
3
.
i
1
.
p
p
7
1
-
8
1
.
[2
5
]
H
.
Y
a
mas
h
i
k
a
a
n
d
S
y
a
f
i
i
,
“
C
h
a
r
a
c
t
e
r
i
st
i
c
s
o
f
sy
m
met
r
i
c
a
l
c
o
m
p
o
n
e
n
t
s
f
o
r
h
i
g
h
i
m
p
e
d
a
n
c
e
f
a
u
l
t
s
i
n
d
i
s
t
r
i
b
u
t
i
o
n
n
e
t
w
o
r
k
s
w
i
t
h
p
h
o
t
o
v
o
l
t
a
i
c
i
n
v
e
r
t
e
r
s
,
”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
E
l
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
E
EC
S
)
,
v
o
l
.
3
3
,
n
o
.
2
,
p
p
.
7
2
7
–
7
3
5
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
3
3
.
i
2
.
p
p
7
2
7
-
7
3
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
8
,
No
.
1
,
Ap
r
il
20
2
5
:
3
9
3
-
4
0
2
402
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Sa
hira
Vila
k
k
um
a
da
t
hil
is
re
se
a
rc
h
s
c
h
o
lar
Ra
th
in
a
m
C
o
ll
e
g
e
o
f
Arts
a
n
d
S
c
ien
c
e
a
n
d
h
a
s
a
tea
c
h
in
g
e
x
p
e
rien
c
e
o
f
1
0
y
e
a
rs
a
n
d
h
o
l
d
th
e
M
.
Sc
.
d
e
g
re
e
in
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
a
lso
v
e
ry
p
a
ss
io
n
a
t
e
to
wa
rd
s
re
se
a
rc
h
w
o
rk
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sa
h
iran
a
jee
b
@g
m
a
il
.
c
o
m
.
Dr
.
Ve
lu
m
a
n
i
T
h
iy
a
g
a
r
a
j
a
n
h
a
s
c
o
m
p
lete
d
h
is
M
.
S
c
.
(C
S
)
,
M
.
P
h
i
l
.
,
M
BA,
B.
Ed
.
,
M.Sc
.
(
P
sy
c
h
o
lo
g
y
)
,
P
h
.
D
.
,
h
e
is
wo
r
k
in
g
a
s
a
n
a
ss
istan
t
p
ro
fe
ss
o
r
i
n
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ie
n
c
e
a
t
Ra
th
in
a
m
Co
ll
e
g
e
o
f
Arts
a
n
d
S
c
ien
c
e
(Au
to
n
o
m
o
u
s),
Co
imb
a
t
o
re
6
4
1
0
2
1
fro
m
1
4
-
09
-
2
0
2
0
to
ti
ll
d
a
te.
He
wo
rk
e
d
a
s
a
ss
istan
t
p
ro
fe
ss
o
r
in
De
p
a
rtme
n
t
o
f
C
o
m
p
u
te
r
S
c
ien
c
e
Ko
n
g
u
Arts
a
n
d
S
c
ien
c
e
Co
ll
e
g
e
(Au
to
n
o
m
o
u
s),
Ero
d
e
fro
m
1
1
-
06
-
2
0
0
9
to
3
0
-
03
-
2
0
2
0
Tam
il
Na
d
u
,
I
n
d
ia.
He
h
a
s
g
o
t
m
o
re
th
a
n
1
2
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
He
h
a
s
o
b
tai
n
e
d
h
is
B.
S
c
.
(CT),
M
.
S
c
.
(CS
),
M
.
P
h
il
.
a
n
d
M
BA
De
g
re
e
s
fro
m
P
e
riy
a
r
Un
iv
e
rsity
S
a
lem
,
Tam
il
Na
d
u
,
In
d
ia.
He
h
a
s
o
b
tain
e
d
h
is
B.
Ed
.
d
e
g
re
e
fro
m
In
d
ira
G
a
n
d
h
i
Na
ti
o
n
a
l
Op
e
n
Un
iv
e
rsit
y
(IG
NO
U)
a
t
De
l
h
i,
h
is
M
.
S
c
.
(P
sy
c
h
o
l
o
g
y
)
fro
m
M
a
d
ra
s
Un
iv
e
rsity
M
a
d
ra
s,
Tam
il
Na
d
u
,
I
n
d
ia.
He
h
a
s
o
b
t
a
in
e
d
h
is
P
h
.
D
.
i
n
c
o
m
p
u
ter
sc
ien
c
e
fro
m
M
a
n
o
n
m
a
n
iam
S
u
n
d
a
ra
n
a
r
Un
i
v
e
rsity
a
t
Ti
ru
n
e
lv
e
ll
i,
Tam
il
Na
d
u
,
I
n
d
ia.
His
a
re
a
o
f
in
tere
st
is
ima
g
e
p
ro
c
e
ss
in
g
re
se
a
rc
h
d
irec
ti
o
n
s
in
v
o
lv
e
d
k
n
o
wle
d
g
e
d
isc
o
v
e
ry
a
n
d
d
a
ta
m
in
in
g
,
p
a
tt
e
r
n
re
c
o
g
n
it
i
o
n
,
k
n
o
wle
d
g
e
-
b
a
se
d
n
e
u
ra
l
n
e
two
rk
s,
s
o
ftwa
re
e
n
g
in
e
e
rin
g
,
a
n
d
I
o
T.
He
h
a
s
p
u
b
li
sh
e
d
m
o
re
th
a
n
1
5
p
a
p
e
rs
in
to
p
m
o
st
i
n
tern
a
ti
o
n
a
l
p
e
e
r
-
re
v
iew
e
d
jo
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
.
He
is
se
rv
in
g
a
s
a
n
e
d
it
o
r
in
4
5
in
ter
n
a
ti
o
n
a
l
p
e
e
r
-
re
v
i
e
we
d
jo
u
r
n
a
ls
a
n
d
a
c
ti
n
g
a
s
a
re
v
iew
e
r
in
1
5
i
n
tern
a
ti
o
n
a
l
p
e
e
r
-
re
v
iew
e
d
jo
u
r
n
a
ls.
He
se
rv
e
s
a
s
a
re
se
a
rc
h
su
p
e
rv
i
so
r
i
n
Co
m
p
u
ter
S
c
ien
c
e
Un
d
e
r
B
h
a
ra
th
iar
Un
iv
e
rsity
.
He
g
o
t
a
p
a
ten
t
rig
h
t
in
-
f
li
g
h
t
ti
c
k
e
t
p
rice
a
n
a
ly
sis
a
n
d
p
re
d
ictio
n
u
sin
g
m
a
c
h
in
e
lea
r
n
i
n
g
.
He
wro
te
a
b
o
o
k
a
n
d
p
u
b
li
s
h
e
d
wit
h
I
S
BN
n
u
m
b
e
r.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
v
e
lu
m
a
n
i4
6
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.