I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
,
p
p
.
2
5
4
0
~
2
5
4
8
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
8
i
4
.
p
p
2
5
4
0
-
2548
2540
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Rea
l
-
Ti
m
e V
ideo
Process
ing
u
sing
Co
ntour Nu
m
b
er
s a
nd
Ang
les for Non
-
u
rba
n Roa
d Ma
r
ker Clas
sifica
tion
Z
a
m
a
ni M
d Sa
ni
1
,
H
a
dh
ra
m
i A
bd
G
ha
ni
2
,
Ro
s
li B
esa
r
3
,
Aziz
ul Az
iza
n
4
,
H
a
f
iza
Aba
s
5
1
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Un
iv
e
rsiti
T
e
k
n
ik
a
l
M
a
la
y
si
a
M
e
lak
a
,
M
a
la
y
sia
2,
3
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
M
u
lt
im
e
d
ia Un
iv
e
rsit
y
,
M
a
la
y
sia
4,
5
A
d
v
a
n
c
e
d
In
f
o
r
m
a
ti
c
s S
c
h
o
o
l,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
lay
sia
,
M
a
la
y
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
9
,
2
0
1
8
R
ev
i
s
ed
J
u
n
2
0
,
2
0
1
8
A
cc
ep
ted
J
u
l
1
1
,
2
0
1
8
Ro
a
d
u
se
rs
m
a
k
e
v
it
a
l
d
e
c
isio
n
s
to
sa
f
e
l
y
m
a
n
e
u
v
e
r
th
e
ir
v
e
h
icle
s
b
a
se
d
o
n
th
e
ro
a
d
m
a
rk
e
rs,
w
h
ich
n
e
e
d
to
b
e
c
o
rre
c
tl
y
c
la
ss
i
f
ied
.
T
h
e
ro
a
d
m
a
r
k
e
rs
c
las
si
f
ica
ti
o
n
is
sig
n
if
ic
a
n
tl
y
i
m
p
o
rtan
t
e
sp
e
c
iall
y
f
o
r
th
e
a
u
to
n
o
m
o
u
s
c
a
r
tec
h
n
o
l
o
g
y
.
T
h
e
c
u
rre
n
t
p
ro
b
lem
s
o
f
e
x
ten
si
v
e
p
ro
c
e
ss
in
g
ti
m
e
a
n
d
re
lativ
e
l
y
lo
w
e
r
a
v
e
ra
g
e
a
c
c
u
ra
c
y
w
h
e
n
c
las
sify
in
g
u
p
t
o
f
iv
e
t
y
p
e
s
o
f
ro
a
d
m
a
rk
e
r
s
a
re
a
d
d
re
ss
e
d
in
th
is
p
a
p
e
r.
T
w
o
n
o
v
e
l
re
a
l
ti
m
e
v
id
e
o
p
ro
c
e
ss
in
g
m
e
th
o
d
s
a
re
p
ro
p
o
se
d
b
y
e
x
tra
c
ti
n
g
tw
o
f
o
r
m
u
late
d
f
e
a
tu
re
s
n
a
m
e
l
y
th
e
c
o
n
to
u
r
n
u
m
b
e
r,
,
a
n
d
a
n
g
le,
to
c
las
sify
th
e
ro
a
d
m
a
rk
e
rs.
In
it
ially
,
th
e
c
a
m
e
ra
p
o
siti
o
n
is
c
a
li
b
ra
ted
t
o
o
b
tai
n
t
h
e
b
e
st
F
ield
o
f
V
iew
(F
OV
)
f
o
r
id
e
n
ti
fy
in
g
a
c
u
sto
m
ize
d
Re
g
io
n
o
f
In
tere
st
(
ROI).
A
n
a
d
a
p
ti
v
e
sm
o
o
th
in
g
a
l
g
o
rit
h
m
is
p
e
rf
o
r
m
e
d
o
n
t
h
e
ROI
b
e
f
o
re
t
h
e
c
o
n
t
o
u
rs
o
f
th
e
ro
a
d
m
a
rk
e
r
s
a
n
d
t
h
e
c
o
rre
sp
o
n
d
in
g
tw
o
fe
a
tu
re
s
a
r
e
d
e
ter
m
in
e
d
.
It
is
o
b
se
rv
e
d
th
a
t
th
e
a
c
h
iev
a
b
le
a
c
c
u
ra
c
y
o
f
th
e
p
ro
p
o
se
d
m
e
th
o
d
s
a
t
se
v
e
ra
l
n
o
n
-
u
r
b
a
n
ro
a
d
sc
e
n
a
rio
s
is
a
p
p
ro
x
im
a
tel
y
9
6
%
a
n
d
th
e
p
r
o
c
e
ss
in
g
t
im
e
p
e
r
f
ra
m
e
is
si
g
n
if
ica
n
tl
y
re
d
u
c
e
d
w
h
e
n
th
e
v
id
e
o
re
so
lu
t
io
n
in
c
re
a
se
s
a
s
c
o
m
p
a
re
d
to
t
h
a
t
o
f
th
e
e
x
isti
n
g
a
p
p
ro
a
c
h
.
K
ey
w
o
r
d
:
Au
to
ass
is
t d
r
iv
i
n
g
s
y
s
te
m
R
o
ad
m
ar
k
er
clas
s
if
icatio
n
s
Vid
eo
im
a
g
e
p
r
o
ce
s
s
i
n
g
Co
p
y
rig
h
t
©
2
0
1
8
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
:
Z
a
m
an
i M
d
San
i,
Facu
lt
y
o
f
E
lectr
ic
a
n
d
E
lectr
o
n
ic,
Un
i
v
er
s
iti T
ek
n
ik
al
Ma
la
y
s
ia
Me
lak
a,
Han
g
T
u
ah
J
a
y
a,
Du
r
ia
n
T
u
n
g
g
al,
7
5
4
5
0
,
Ma
lay
s
ia.
E
m
ail: z
a
m
an
is
a
n
i
@
u
te
m
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
n
u
m
b
er
o
f
r
o
ad
ac
cid
en
ts
in
Ma
la
y
s
ia
i
n
cr
ea
s
es
at
an
al
ar
m
in
g
r
ate
o
f
9
.
7
%
ev
er
y
y
ea
r
o
v
er
th
e
p
ast th
ir
t
y
y
ea
r
s
[
1
]
co
n
tr
ib
u
te
d
b
y
(
r
o
ad
u
s
er
s
)
er
r
o
r
s
,
f
a
u
lt
y
r
o
ad
en
v
ir
o
n
m
e
n
t,
a
n
d
v
e
h
icl
e
d
ef
ec
ts
.
T
h
e
r
o
ad
m
ar
k
er
,
is
v
er
y
i
m
p
o
r
ta
n
t
i
n
g
u
id
in
g
r
o
ad
u
s
er
s
w
h
ile
d
r
iv
i
n
g
.
Di
f
f
er
e
n
t
t
y
p
es
o
f
r
o
ad
m
ar
k
er
s
i
n
n
o
n
-
u
r
b
a
n
r
o
ad
s
s
u
ch
as
d
o
u
b
le
s
o
lid
(
DD)
,
d
ash
ed
(
D)
,
s
o
lid
-
d
ash
ed
(
SD)
,
d
ash
ed
-
s
o
lid
an
d
s
i
n
g
le
s
o
lid
(
SS
)
,
allo
w
d
r
iv
er
s
to
s
a
f
el
y
d
ec
id
e
eit
h
e
r
to
m
ai
n
tai
n
t
h
e
co
u
r
s
e
i
n
t
h
e
m
id
d
le
o
f
th
e
la
n
e
o
r
to
o
v
er
tak
e
t
h
e
f
r
o
n
t
v
eh
ic
les.
I
n
ea
r
lier
w
o
r
k
s
,
m
an
y
e
f
f
o
r
ts
w
er
e
in
v
es
ted
in
s
o
lv
i
n
g
la
n
e
d
etec
tio
n
an
d
tr
a
ck
in
g
p
r
o
b
le
m
s
f
o
r
Au
to
-
Ass
is
t
Dr
iv
i
n
g
S
y
s
te
m
(
A
D
S)
[
2
]
,
[
3
]
.
T
h
ese
m
et
h
o
d
s
h
ad
b
ee
n
u
s
ed
as
a
w
ar
n
i
n
g
s
y
s
te
m
to
th
e
d
r
i
v
er
[4
]
-
[
8
]
an
d
s
u
r
p
r
is
in
g
l
y
,
t
h
e
lan
e
d
etec
tio
n
also
w
a
s
u
s
ed
as
p
ar
t
o
f
th
e
s
y
s
te
m
to
an
al
y
s
e
th
e
d
r
iv
er
b
eh
av
io
u
r
[
9
]
.
Sy
m
b
o
ls
,
alp
h
ab
et
s
,
an
d
cu
s
to
m
m
ar
k
er
s
w
er
e
also
b
ein
g
e
x
p
lo
r
ed
an
d
u
s
ed
as
th
e
in
f
o
r
m
atio
n
to
aler
t
d
r
iv
er
s
[
10
,
11
]
.
Ho
w
e
v
er
,
r
o
ad
m
ar
k
er
clas
s
i
f
icatio
n
[
12
]
,
[
13
]
s
till
r
em
ai
n
s
an
o
p
en
q
u
es
tio
n
d
u
e
to
t
h
e
v
ar
y
i
n
g
t
y
p
es
o
f
r
o
ad
m
ar
k
er
s
ac
r
o
s
s
th
e
g
lo
b
e
w
it
h
li
m
ited
e
f
f
o
r
t
b
ein
g
p
u
t
to
s
o
lv
e
t
h
e
p
r
o
b
le
m
i
n
r
o
ad
m
ar
k
er
class
i
f
icatio
n
.
C
o
llad
o
et
a
l
.
[
1
4
]
p
r
o
p
o
s
ed
a
f
r
eq
u
en
c
y
an
al
y
s
is
m
et
h
o
d
f
o
r
th
e
u
r
b
an
r
o
ad
m
ar
k
e
r
c
lass
i
f
icatio
n
,
w
h
ic
h
u
s
es
t
h
e
in
v
er
s
e
p
er
s
p
ec
tiv
e
m
ap
p
in
g
(
I
P
M)
to
g
en
er
ate
a
b
ir
d
-
e
y
e
v
i
e
w
o
f
t
h
e
r
o
ad
a
n
d
p
r
o
d
u
ce
a
m
o
d
if
ied
Ho
u
g
h
T
r
an
s
f
o
r
m
(
HT
)
f
o
r
b
etter
lan
e
d
etec
tio
n
.
T
h
e
m
et
h
o
d
t
h
en
ap
p
lies
p
o
w
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
R
ea
l
-
Time
V
id
eo
P
r
o
ce
s
s
in
g
u
s
in
g
C
o
n
to
u
r
N
u
mb
ers
a
n
d
A
n
g
les fo
r
N
o
n
-
u
r
b
a
n
…
(
Za
ma
n
i Md
S
a
n
i
)
2541
s
p
ec
tr
u
m
a
n
d
Fo
u
r
ier
an
al
y
s
is
to
d
etec
t
th
r
ee
ty
p
es
o
f
m
ar
k
er
s
,
w
h
ich
ar
e;
s
in
g
le
s
o
lid
,
d
ash
ed
,
an
d
m
er
g
ed
r
o
ad
m
ar
k
er
s
.
Ho
w
e
v
er
,
r
o
ad
m
ar
k
er
clas
s
if
icatio
n
f
o
r
n
o
n
-
u
r
b
an
t
y
p
e
r
o
ad
s
was
n
o
t
ad
d
r
ess
ed
.
Sch
u
b
er
t
et
a
l
.
[
1
5
]
s
tu
d
ied
lan
e
ch
a
n
g
in
g
b
ased
o
n
r
o
ad
m
ar
k
er
r
ec
o
g
n
itio
n
s
u
r
r
o
u
n
d
i
n
g
tr
af
f
ic
s
it
u
atio
n
b
u
t
th
e
r
o
ad
m
ar
k
er
s
class
if
ied
wer
e
li
m
ited
to
o
n
l
y
;
d
ash
ed
an
d
s
o
lid
lin
es.
I
n
a
w
o
r
k
b
y
L
i
n
d
n
er
et
a
l
.
[
1
6
]
,
an
ed
g
e
d
etec
to
r
tech
n
iq
u
e
w
as
d
esig
n
ed
to
s
u
b
s
eq
u
e
n
tl
y
s
ea
r
ch
f
o
r
a
g
r
o
u
p
o
f
f
o
u
r
o
b
j
ec
t
s
n
a
m
el
y
t
h
e
li
n
es,
cu
r
v
e
s
,
p
ar
allel
cu
r
v
e
s
an
d
cl
o
s
e
d
o
b
j
ec
ts
in
d
etec
tin
g
r
o
ad
m
ar
k
er
s
.
A
lt
h
o
u
g
h
f
o
u
r
t
y
p
es
o
f
r
o
ad
m
ar
k
er
s
w
er
e
clas
s
i
f
ied
,
m
o
s
t o
f
t
h
e
r
o
ad
m
ar
k
er
s
ar
e
d
ash
ed
li
n
es
with
d
if
f
er
e
n
t size
s
.
Me
an
w
h
ile,
an
o
t
h
er
r
esear
ch
f
r
o
m
Su
c
h
itra
et
a
l
.
clas
s
i
f
ied
t
h
r
ee
r
o
ad
m
ar
k
er
s
n
a
m
el
y
; d
as
h
ed
,
s
o
lid
a
n
d
zig
za
g
u
s
i
n
g
a
m
o
d
u
lar
ap
p
r
o
ac
h
[
1
7
]
.
T
h
e
r
o
ad
m
ar
k
er
s
ar
e
class
i
f
ied
as
eit
h
er
d
as
h
ed
o
r
s
o
lid
u
s
i
n
g
t
h
e
B
asic
L
an
e
Ma
r
k
in
g
(
B
L
M)
,
w
h
ic
h
is
b
ased
o
n
co
n
tin
u
it
y
p
r
o
p
er
ties
.
T
h
is
ap
p
r
o
ac
h
h
o
w
e
v
er
,
ap
p
lies
th
e
te
m
p
o
r
al
in
f
o
r
m
atio
n
i
n
t
h
e
c
lass
i
f
icatio
n
o
p
er
atio
n
,
w
h
ic
h
r
en
d
er
s
s
lo
w
er
d
etec
tio
n
w
h
en
th
e
r
o
ad
m
ar
k
er
t
y
p
e
ch
an
g
e
s
w
h
il
s
t
d
r
iv
i
n
g
o
n
th
e
r
o
ad
.
I
n
ad
d
itio
n
to
th
ese
m
e
th
o
d
s
,
Ned
ev
s
c
h
i
et
a
l
.
[
18
]
u
s
es
a
p
er
io
d
ic
h
is
to
g
r
a
m
to
d
eter
m
i
n
e
t
h
e
t
y
p
e
o
f
r
o
ad
m
ar
k
er
s
.
T
h
is
eg
o
-
lo
ca
lizatio
n
is
o
b
s
er
v
ed
to
en
ab
l
e
th
e
clas
s
if
icatio
n
o
f
r
o
ad
m
ar
k
er
s
in
to
f
o
u
r
t
y
p
e
s
n
a
m
el
y
s
i
n
g
le
s
o
lid
,
d
o
u
b
le
s
o
lid
,
d
ash
ed
,
an
d
m
er
g
ed
.
I
n
a
r
ec
en
t
r
esear
ch
,
P
au
la
e
t
a
l
.
[
1
9
]
p
r
esen
ted
an
au
to
m
ati
c
class
i
f
icatio
n
tech
n
iq
u
e
to
class
i
f
y
f
i
v
e
t
y
p
es
o
f
r
o
ad
m
ar
k
er
s
.
T
h
e
ap
p
r
o
ac
h
u
s
es
b
et
w
ee
n
th
r
ee
to
f
i
v
e
f
ea
tu
r
e
s
ex
tr
ac
ted
f
r
o
m
t
h
e
i
m
a
g
e,
w
h
ich
is
later
f
ed
to
th
e
ar
tific
ial
n
e
u
r
al
n
et
w
o
r
k
.
A
t
w
o
-
s
tag
e
m
eth
o
d
is
m
o
d
elled
f
o
r
t
h
e
f
u
ll
cla
s
s
if
ica
tio
n
,
w
ith
t
h
e
f
ir
s
t
s
ta
g
e
ap
p
lies
t
h
e
B
a
y
es
ia
n
clas
s
i
f
ier
f
o
r
d
as
h
ed
,
s
i
n
g
le
d
ash
ed
an
d
d
o
u
b
le
s
o
lid
w
h
i
le
th
e
s
ec
o
n
d
s
ta
g
e
d
if
f
er
e
n
tiate
s
b
et
w
ee
n
d
ash
ed
-
s
o
lid
an
d
s
o
lid
-
d
ash
ed
l
in
e
s
f
o
r
r
o
ad
m
ar
k
er
d
etec
tio
n
.
Ho
wev
er
,
th
e
r
e
s
u
l
ts
o
f
th
e
clas
s
i
f
icatio
n
w
er
e
f
o
u
n
d
to
h
av
e
c
h
an
g
ed
ab
r
u
p
tl
y
o
n
ea
ch
f
r
a
m
e,
w
h
ic
h
ca
u
s
ed
in
co
n
s
is
te
n
t
r
es
u
lt
s
w
h
ile
d
r
iv
i
n
g
.
Ma
t
h
ib
ela
et
a
l
.
[
20
]
p
r
o
p
o
s
ed
a
n
e
w
ap
p
r
o
ac
h
u
s
i
n
g
a
u
n
iq
u
e
s
e
t
o
f
g
eo
m
e
tr
ic
f
ea
t
u
r
es
w
h
ic
h
f
u
n
ctio
n
s
w
it
h
i
n
a
p
r
o
b
ab
ilis
tic
R
USB
o
o
s
t
an
d
C
o
n
d
itio
n
al
R
an
d
o
m
Field
(
C
R
F)
n
et
w
o
r
k
to
class
if
y
t
h
e
r
o
ad
m
ar
k
er
s
in
to
s
ev
e
n
t
y
p
e
s
,
i
n
clu
d
in
g
;
s
in
g
le
b
o
u
n
d
ar
y
,
d
o
u
b
le
b
o
u
n
d
ar
y
,
s
ep
ar
ato
r
,
zig
-
za
g
,
i
n
ter
s
ec
tio
n
,
b
o
x
ed
j
u
n
ctio
n
a
n
d
s
p
ec
ial
la
n
es.
E
v
e
n
t
h
o
u
g
h
m
o
r
e
t
y
p
es
o
f
r
o
ad
m
ar
k
er
s
w
er
e
clas
s
i
f
ied
u
s
i
n
g
t
h
is
m
e
th
o
d
,
u
n
f
o
r
tu
n
atel
y
,
th
e
cla
s
s
i
f
icat
io
n
o
n
l
y
u
s
ed
s
tatic
i
m
ag
e
s
i
n
u
r
b
an
r
o
ad
s
.
On
th
e
b
asi
s
o
f
t
h
e
co
m
p
r
eh
e
n
s
iv
e
liter
at
u
r
e
r
ev
ie
w
,
t
h
e
ex
te
n
s
i
v
e
clas
s
i
f
icatio
n
o
f
r
o
ad
m
a
r
k
er
s
h
ad
b
ee
n
r
ar
ely
s
tu
d
ied
.
I
t
is
in
t
er
esti
n
g
to
lo
o
k
at
th
e
af
o
r
em
en
tio
n
ed
d
if
f
er
en
t
ap
p
r
o
ac
h
es
f
o
r
r
o
ad
m
ar
k
er
class
i
f
icatio
n
alt
h
o
u
g
h
n
o
s
ta
n
d
ar
d
d
atab
ases
av
ailab
le
to
allo
w
f
air
co
m
p
ar
is
o
n
.
I
n
ad
d
itio
n
,
th
e
d
i
m
en
s
io
n
,
co
lo
u
r
an
d
s
ize
o
f
th
e
m
ar
k
er
s
ar
e
v
ar
ied
ac
r
o
s
s
th
e
w
o
r
ld
.
T
h
is
p
ap
er
p
r
o
p
o
s
es
a
n
o
v
el
ap
p
r
o
ac
h
to
class
i
f
y
th
e
s
e
r
o
a
d
u
s
i
n
g
a
c
u
s
to
m
ized
R
e
g
io
n
o
f
I
n
ter
es
t
(
R
OI
)
in
a
v
id
eo
ac
q
u
ir
ed
f
r
o
m
a
ca
m
er
a
w
it
h
i
ts
c
al
ib
r
ated
p
o
s
itio
n
.
T
w
o
f
ea
t
u
r
es
ar
e
d
er
iv
ed
,
n
a
m
el
y
;
t
h
e
co
n
to
u
r
n
u
m
b
er
,
,
an
d
th
e
co
n
to
u
r
an
g
le,
,
ar
e
later
f
ed
in
to
a
t
w
o
-
la
y
er
class
if
ier
.
T
h
e
f
ir
s
t
la
y
er
class
i
f
ies
th
e
D
a
n
d
SS
m
ar
k
er
ty
p
e
s
b
ased
ca
lcu
lated
,
v
alu
es,
w
h
ile
t
h
e
s
ec
o
n
d
la
y
er
class
i
f
ies
th
e
DD,
DS
o
r
SD
m
ar
k
er
t
y
p
es,
b
y
u
s
i
n
g
v
al
u
es
as
s
h
o
w
n
in
Fi
g
u
r
e
1
.
T
em
p
o
r
al
in
f
o
r
m
atio
n
in
te
g
r
atio
n
i
s
ap
p
lied
to
im
p
r
o
v
e
th
e
cla
s
s
i
f
icatio
n
ac
cu
r
ac
y
b
y
v
alid
ati
n
g
th
e
m
ar
k
er
t
y
p
e’
s
tr
an
s
itio
n
s
o
n
th
e
r
o
ad
.
T
h
is
alg
o
r
ith
m
h
as
b
ee
n
d
e
m
o
n
s
tr
ated
an
o
v
er
all
ac
cu
r
ac
y
o
f
ap
p
r
o
x
im
a
tel
y
9
5
%
an
d
r
ed
u
c
ed
r
ed
u
ctio
n
in
th
e
p
r
o
ce
s
s
in
g
t
i
m
e
b
y
~5
0
%
s
h
o
r
ter
th
an
t
h
e
ex
is
ti
n
g
m
et
h
o
d
.
Fig
u
r
e
1
.
R
o
ad
m
ar
k
er
s
f
o
u
n
d
at
t
w
o
-
w
a
y
n
o
n
-
u
r
b
an
r
o
ad
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
Ca
m
er
a
s
et
u
p f
o
r
re
g
io
n o
f
i
nte
re
s
t
s
elec
t
io
n
Ma
n
y
e
x
is
tin
g
m
eth
o
d
s
f
o
r
s
e
lectin
g
th
e
R
OI
s
u
c
h
as
t
h
e
v
an
is
h
i
n
g
p
o
in
t,
ar
ea
-
b
ased
d
etec
tio
n
an
d
ar
ea
-
b
ased
tr
ac
k
i
n
g
m
et
h
o
d
s
h
ad
b
ee
n
u
s
ed
in
th
e
p
a
s
t.
A
t
ec
h
n
iq
u
e
o
f
ca
lcu
lat
in
g
t
h
e
v
a
n
is
h
i
n
g
p
o
i
n
t
u
s
i
n
g
Ho
u
g
h
T
r
an
s
f
o
r
m
(
HT
)
is
ca
r
r
ied
o
u
t
b
y
o
b
tain
i
n
g
th
e
in
ter
s
ec
tio
n
li
n
e
an
d
t
h
e
R
OI
at
t
h
e
lo
w
er
h
al
f
o
f
t
h
e
i
m
a
g
e
[
2
1
]
.
I
n
[
22
]
,
[
23
]
,
th
e
R
OI
is
s
elec
ted
o
v
er
th
e
w
h
o
le
h
o
r
izo
n
tal
ax
i
s
an
d
a
li
m
i
ted
r
an
g
e
alo
n
g
t
h
e
v
er
tical
ax
is
,
b
ef
o
r
e
th
e
f
ea
tu
r
e
ex
tr
ac
tio
n
a
n
d
r
o
ad
m
ar
k
er
o
r
lan
e
class
i
f
icat
io
n
ta
k
e
p
lace
.
I
n
th
i
s
p
r
o
j
ec
t,
a
n
e
w
m
et
h
o
d
in
s
elec
t
in
g
t
h
e
R
OI
b
ased
o
n
its
ca
lib
r
ated
ca
m
er
a’
s
h
e
ig
h
t
an
d
Field
o
f
Vie
w
(
F
OV)
is
p
r
o
p
o
s
ed
.
Fo
r
th
e
i
n
itial
s
et
u
p
,
a
ca
m
er
a
w
ith
a
r
eso
lu
tio
n
o
f
1
2
8
0
x
7
2
0
lo
ca
ted
at
th
e
ce
n
ter
o
f
t
h
e
ca
r
is
ca
lib
r
ated
o
n
its
p
o
s
itio
n
w
it
h
it
s
FOV
a
d
j
u
s
ted
to
w
ar
d
s
th
e
p
la
n
ar
r
o
ad
s
u
r
f
ac
e
ca
p
t
u
r
ed
b
y
t
h
e
ca
m
er
a
as
s
h
o
w
n
i
n
Fig
u
r
e
2
(
a
)
an
d
Fi
g
u
r
e
2
(
b
)
.
T
o
ca
lib
r
ate
th
e
p
o
s
itio
n
o
f
th
e
ca
m
er
a,
t
w
o
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
2
5
4
0
–
2
5
4
8
2542
h
o
r
izo
n
tal
li
n
es,
h1
an
d
h2
,
ar
e
s
et
o
n
th
e
FOV
to
eq
u
all
y
d
iv
id
e
in
to
th
r
ee
s
ec
tio
n
s
as
i
n
Fig
u
r
e
2
(
c
)
,
w
h
er
e
th
e
h
1
h
o
r
izo
n
tal
li
n
e
in
t
h
e
i
m
ag
e
m
u
s
t b
e
at
t
h
e
s
a
m
e
lev
e
l
w
it
h
t
h
e
m
o
s
t r
ig
h
t la
n
e
in
t
h
e
i
m
ag
e.
(
a)
(
b
)
(
c)
Fig
u
r
e
2
.
T
h
e
ca
m
er
a
s
et
u
p
(
a)
C
a
m
er
a
p
o
s
itio
n
(
b
)
3
d
ca
m
er
a
co
o
r
d
in
ate
s
y
s
te
m
(
c)
C
a
m
er
a
h
eig
h
t
ad
j
u
s
t
m
e
n
t
w
it
h
h
1
an
d
h
2
h
o
r
izo
n
tal
li
n
es
T
h
e
i
m
a
g
e
ca
p
t
u
r
ed
is
th
e
n
d
iv
id
ed
in
to
4
0
eq
u
all
y
s
ized
s
e
ctio
n
s
o
r
s
q
u
ar
es
,
ea
ch
o
f
w
h
i
ch
h
av
in
g
a
s
ize
o
f
1
6
0
x
1
4
4
p
ix
el
s
a
s
s
h
o
w
n
i
n
Fi
g
u
r
e
3
(
a
)
an
d
Fig
u
r
e
3
(
b
)
.
T
h
e
R
OI
s
elec
tio
n
m
et
h
o
d
is
ap
p
lied
to
ev
er
y
v
id
eo
f
r
a
m
e
to
ch
o
o
s
e
(
x,
y
)
as
th
e
R
OI
,
w
h
ic
h
co
n
tai
n
s
th
e
r
o
ad
m
ar
k
er
,
as
s
h
o
w
n
i
n
Fig
u
r
e
3
(
c
)
.
I
t
ca
n
b
e
o
b
s
er
v
ed
th
at
(
x,
y
)
is
th
e
n
ea
r
e
s
t
to
th
e
ca
r
w
h
er
e
th
e
ef
f
ec
t
s
o
f
th
e
r
o
ad
cu
r
v
e
o
n
th
e
r
o
ad
m
ar
k
er
is
lo
w
.
In
th
e
ca
s
e
o
f
lan
e
d
ep
ar
tu
r
e
,
th
e
R
OI
w
i
ll
c
o
n
tain
n
o
m
ar
k
e
r
f
o
r
a
lo
n
g
er
d
u
r
at
io
n
w
h
ic
h
in
d
icate
s
t
h
at
t
h
e
v
eh
ic
le
is
d
ep
ar
ted
f
r
o
m
t
h
e
r
i
g
h
t
tr
ac
k
.
I
n
o
u
r
ap
p
r
o
ac
h
,
r
1
,
co
n
tain
i
n
g
t
h
e
R
GB
in
f
o
r
m
a
ti
o
n
,
is
co
n
v
er
ted
to
g
r
e
y
s
ca
le
a
n
d
f
ilter
ed
t
h
r
o
u
g
h
G
a
u
s
s
ia
n
f
ilter
[
24
]
b
ef
o
r
e
u
n
d
er
g
o
i
n
g
t
h
e
t
h
r
es
h
o
ld
in
g
p
r
o
ce
s
s
u
s
in
g
Ots
u
T
h
r
esh
o
ld
in
g
m
et
h
o
d
[
25
]
f
o
r
b
in
ar
y
co
n
v
er
s
io
n
.
T
h
en
th
e
t
w
o
f
ea
t
u
r
es,
w
h
ich
ar
e
th
e
co
n
to
u
r
n
u
m
b
er
an
d
th
e
an
g
le
s
,
w
ill b
e
ex
tr
ac
ted
to
co
m
p
lete
t
h
e
clas
s
i
f
icatio
n
a
s
d
is
cu
s
s
ed
n
e
x
t.
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
(
a)
R
o
w
an
d
c
o
lu
m
n
p
o
s
t c
am
er
a
p
o
s
itio
n
ca
lib
r
atio
n
(
b
)
R
OI
r
1
(
c
)
R
OI
s
co
n
s
is
t
o
f
r
o
ad
m
ar
k
er
s
2
.
2
.
T
he
nu
m
ber
o
f
co
nto
urs
,
,
a
nd
t
he
a
ng
le
in
M
et
ho
d A
a
nd
B
A
c
o
n
to
u
r
is
a
lin
e
th
at
co
n
n
e
cts
all
p
o
in
ts
alo
n
g
t
h
e
b
o
u
n
d
ar
y
o
f
a
n
o
b
j
ec
t
o
f
th
e
s
a
m
e
c
o
lo
u
r
.
T
h
e
co
n
to
u
r
n
u
m
b
er
,
in
th
e
R
OI
is
ca
lcu
lated
b
y
co
u
n
ti
n
g
t
h
e
n
u
m
b
er
o
f
all
co
n
to
u
r
s
alo
n
g
its
h
o
r
izo
n
tal
ax
is
.
T
h
e
ca
lcu
latio
n
o
f
is
ca
r
r
ied
o
u
t
in
th
e
R
OI
s
,
(
x,
y)
,
f
o
r
ev
er
y
f
r
a
m
es
i
n
a
co
m
p
lete
c
y
cle,
w
h
ich
is
d
ef
i
n
e
d
as
th
e
ti
m
e
tak
e
n
b
y
all
co
n
s
ec
u
ti
v
e
f
r
a
m
e
s
co
n
tai
n
in
g
a
co
m
p
lete
r
o
ad
m
ar
k
er
p
atter
n
.
T
h
is
p
atter
n
w
il
l r
ep
ea
t its
el
f
af
ter
ev
er
y
co
m
p
lete
c
y
cle
u
n
til t
h
e
r
o
ad
m
a
r
k
er
t
y
p
e
ch
a
n
g
e
s
o
r
en
d
s
as i
n
Fi
g
u
r
e
4
.
T
h
e
p
r
o
p
o
s
ed
r
o
ad
m
ar
k
er
cl
ass
i
f
icatio
n
is
b
a
s
ed
o
n
t
h
e
c
o
n
to
u
r
n
u
m
b
er
,
,
an
d
th
e
co
n
to
u
r
an
g
le,
,
i
m
p
le
m
e
n
ted
u
s
in
g
t
w
o
m
eth
o
d
s
o
f
t
w
o
-
la
y
er
class
if
icatio
n
as
d
ep
icted
in
Fi
g
u
r
e
5
.
T
h
e
f
o
r
m
u
latio
n
o
f
an
d
f
o
r
b
o
th
o
f
th
e
s
e
m
eth
o
d
s
,
w
h
ic
h
ar
e
n
a
m
ed
as
Me
th
o
d
A
a
n
d
Me
th
o
d
B
,
w
il
l
b
e
p
r
esen
ted
n
ex
t.
I
n
Me
th
o
d
A
,
r
o
ad
m
ar
k
er
D
is
d
etec
ted
w
h
e
n
t
h
e
m
in
i
m
u
m
o
v
er
th
e
c
y
cle
is
ze
r
o
a
n
d
th
e
m
ax
i
m
u
m
is
o
n
e.
I
f
o
v
er
a
co
m
p
lete
c
y
cle,
t
h
en
t
h
e
r
o
ad
m
ar
k
er
is
S
S.
Oth
er
w
i
s
e,
if
t
h
e
m
ax
i
m
u
m
o
v
er
th
e
cy
cle
i
s
t
w
o
,
th
e
n
th
e
s
ec
o
n
d
la
y
er
clas
s
if
ica
tio
n
u
s
i
n
g
t
h
e
an
g
le
o
f
th
e
ce
n
tr
o
id
s
w
il
l
b
e
r
u
n
to
clas
s
i
f
y
th
e
m
ar
k
er
a
s
SD,
D
S
o
r
DD.
T
h
e
an
g
le
θ,
w
h
ic
h
i
s
m
ea
s
u
r
ed
b
et
w
ee
n
a
li
n
e
co
n
n
ec
ti
n
g
t
h
e
t
w
o
ce
n
tr
o
i
d
s
an
d
t
h
e
h
o
r
izo
n
tal
a
x
i
s
,
i
s
ca
l
cu
lated
o
n
l
y
o
n
th
e
f
ir
s
t
f
r
a
m
e
in
a
s
et
o
f
f
r
a
m
es
s
h
o
w
i
n
g
a
p
ar
tic
u
lar
r
o
ad
m
a
r
k
er
p
atter
n
t
h
at
r
ep
ea
ts
it
s
el
f
af
ter
th
e
last
f
r
a
m
e
i
n
th
e
s
et
b
ef
o
r
e
th
e
p
atter
n
ch
an
g
es.
T
h
is
ca
n
b
e
id
en
tifie
d
f
r
o
m
t
h
e
c
h
a
n
g
e
in
v
alu
e,
w
h
ic
h
i
s
f
r
o
m
1
0
or
1
2
.
T
h
er
e
ar
e
th
r
ee
th
r
es
h
o
ld
an
g
le
s
an
d
,
w
h
ich
ar
e
d
eter
m
i
n
ed
f
r
o
m
tr
ain
i
n
g
d
ataset,
ap
p
lied
f
o
r
th
e
class
i
f
icatio
n
.
If
,
th
en
th
e
r
o
ad
m
ar
k
er
i
s
D
S.
I
f
,
th
e
n
t
h
e
r
o
ad
m
ar
k
er
i
s
SD
an
d
if
o
r
,
th
e
n
th
e
r
o
ad
m
ar
k
er
is
DD.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
R
ea
l
-
Time
V
id
eo
P
r
o
ce
s
s
in
g
u
s
in
g
C
o
n
to
u
r
N
u
mb
ers
a
n
d
A
n
g
les fo
r
N
o
n
-
u
r
b
a
n
…
(
Za
ma
n
i Md
S
a
n
i
)
2543
(
a)
(
b
)
Fig
u
r
e
2
.
T
h
e
co
n
to
u
r
n
u
m
b
er
ca
lcu
lated
f
r
o
m
t
h
e
f
r
a
m
e
s
(
a)
Dash
ed
(
b
)
So
lid
-
d
ash
ed
Fig
u
r
e
5
.
T
w
o
-
la
y
er
clas
s
i
f
icat
io
n
d
iag
r
a
m
(
a)
Me
th
o
d
A
(
b
)
Me
th
o
d
B
I
n
Me
t
h
o
d
B
,
th
e
s
ig
n
i
f
ica
n
t
d
if
f
er
e
n
t
i
s
w
h
e
n
d
etec
ti
n
g
D
D.
I
n
s
tead
o
f
d
etec
ti
n
g
DD
at
th
e
s
ec
o
n
d
la
y
er
u
s
i
n
g
th
e
an
g
le
,
DD
is
d
etec
ted
at
th
e
f
ir
s
t
la
y
er
b
ased
o
n
,
w
h
en
it
is
eq
u
al
to
t
w
o
o
v
er
a
co
m
p
lete
c
y
cle
.
Hen
ce
,
o
n
l
y
o
n
e
t
h
r
esh
o
ld
a
n
g
le,
,
u
s
ed
to
d
etec
t D
S a
n
d
SD.
I
n
o
r
d
er
to
f
i
n
d
th
e
co
n
to
u
r
an
g
le,
a
s
s
ee
n
i
n
Fi
g
u
r
e
6
(
d
)
a
n
d
f
o
r
m
u
lated
as
f
o
llo
w
s
:
[
]
(1
)
T
w
o
ce
n
tr
o
id
s
,
an
d
,
ar
e
c
alcu
lated
f
r
o
m
t
h
e
m
o
m
en
t
s
o
f
th
e
m
ar
k
er
s
,
,
,
s
u
r
r
o
u
n
d
ed
b
y
th
e
co
n
to
u
r
s
as
s
h
o
w
n
in
Fig
u
r
e
6
(
a)
an
d
Fig
u
r
e
6
(
b
)
.
T
h
e
r
ed
lin
e
r
ep
r
esen
t
s
th
e
co
n
to
u
r
s
a
n
d
t
h
e
r
ed
d
o
ts
d
en
o
te
th
e
ce
n
tr
o
id
s
.
T
h
e
m
o
m
en
t
s
a
r
e
ca
lcu
lated
a
s
∑
,
w
h
er
e
r
ep
r
esen
ts
th
e
r
eg
io
n
in
th
e
co
n
to
u
r
(
th
e
r
ed
lin
es)
[
2
6
]
.
I
n
th
e
n
ex
t
s
ec
tio
n
,
t
h
e
th
r
es
h
o
ld
an
g
le
s
w
ill
b
e
d
eter
m
i
n
ed
to
class
i
f
y
t
h
e
r
o
ad
m
ar
k
er
as
DD
,
DS
or
SD
(
f
o
r
Me
th
o
d
A
)
,
w
h
er
e
=1
,
2
,
3
f
o
r
Me
th
o
d
A
a
n
d
o
n
l
y
is
u
s
ed
f
o
r
Me
th
o
d
B
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
2
5
4
0
–
2
5
4
8
2544
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
6
.
(
a)
-
(
c)
R
OI
w
it
h
ce
n
tr
o
id
s
f
o
u
n
d
(
d
)
v
alu
e
ca
lc
u
l
atio
n
f
r
o
m
t
w
o
ce
n
tr
o
id
s
2
.
3
.
T
he
t
hres
ho
ld a
ng
les
f
o
r
m
ul
a
t
io
n f
o
r
t
he
s
ec
o
nd
-
la
y
er
cla
s
s
if
ica
t
io
n in
m
et
ho
d A
a
nd
B
I
n
Me
th
o
d
A
,
t
h
e
t
h
r
es
h
o
ld
an
g
les
ar
e
f
o
r
m
u
lated
b
ased
o
n
th
e
an
g
les
b
et
w
ee
n
t
w
o
ce
n
t
r
o
id
s
o
f
ea
ch
o
f
th
e
3
0
0
f
r
a
m
e
s
ex
tr
ac
ted
f
r
o
m
f
o
u
r
v
id
eo
clip
s
(
Vid
2
2
_
USB
,
Vid
2
4
_
USB
,
Vid
2
6
_
USB
,
Vid
2
7
_
USB
)
as
in
T
ab
le
2
.
T
h
e
an
g
le
s
ar
e
th
en
p
lo
tted
a
cc
o
r
d
in
g
l
y
i
n
F
ig
u
r
e
7
.
As
f
o
r
DD
,
th
e
v
al
u
es
ar
e
f
o
u
n
d
to
be
d
is
tr
ib
u
ted
b
et
w
ee
n
th
e
n
e
g
ati
v
e
an
g
les
(
d
en
o
ted
as
DDlo
w
er
)
an
d
th
e
po
s
itiv
e
an
g
le
s
(
d
en
o
ted
as
DDu
p
p
er
)
d
u
e
to
th
e
ch
a
n
g
e
in
t
h
e
lo
ca
tio
n
o
f
th
e
t
w
o
ce
n
tr
o
id
s
s
h
i
f
t
in
g
alo
n
g
a
cu
r
v
y
r
o
ad
.
T
h
e
an
g
le
s
f
o
r
th
e
1
st
f
r
a
m
e
s
i
n
ev
er
y
s
et
o
f
f
r
a
m
es
f
o
r
DS,
SD
an
d
DD
ar
e
f
o
u
n
d
to
b
e
s
ep
ar
ated
w
ith
ea
ch
o
th
er
,
as
s
ee
n
in
Fig
u
r
e
7
.
I
t
is
clea
r
f
r
o
m
th
is
f
ig
u
r
e
th
at
th
e
th
r
esh
o
ld
an
g
le
s
,
ar
e
th
e
s
ep
ar
atin
g
an
g
le
s
o
r
lin
es b
et
w
ee
n
f
o
u
r
g
r
o
u
p
s
o
f
an
g
les
w
h
er
e
f
o
r
Me
th
o
d
A
.
Fig
u
r
e
3
.
Dis
tr
ib
u
tio
n
o
f
t
h
e
f
o
r
T
h
e
th
r
ee
t
h
r
esh
o
ld
an
g
le
s
,
an
d
w
h
ic
h
ar
e
r
ep
r
esen
ted
b
y
t
h
e
b
lu
e
h
o
r
izo
n
tal
li
n
e
s
in
Fi
g
u
r
e
7
,
ar
e
ca
lcu
lated
s
u
ch
th
at
th
e
an
g
les
o
r
v
al
u
es
ar
e
s
ep
ar
ated
o
r
class
if
ie
d
in
tr
o
th
r
ee
t
y
p
es
;
n
a
m
e
l
y
DD,
SD
a
n
d
DS
.
Hen
ce
,
t
he
se
th
r
es
h
o
l
d
v
al
u
es
th
e
m
id
p
o
in
t
b
et
w
ee
n
th
e
m
a
x
i
m
u
m
a
n
g
le
o
f
a
m
ar
k
er
t
y
p
e
a
n
d
th
e
m
in
i
m
u
m
a
n
g
le
o
f
th
e
n
e
x
t
n
ei
g
h
b
o
u
r
in
g
m
ar
k
er
t
y
p
e
.
T
h
e
m
i
n
i
m
u
m
an
d
m
a
x
i
m
u
m
an
g
le
s
o
f
DDu
p
p
er
,
SD,
DS
an
d
DDlo
w
er
ar
e
s
to
r
ed
as
th
e
f
ir
s
t
r
o
w
(
m
i
n
i
m
u
m
a
n
g
le
s
)
an
d
th
e
th
e
s
ec
o
n
d
r
o
w
(
m
a
x
i
m
u
m
a
n
g
les)
o
f
m
atr
i
x
r
esp
ec
tiv
el
y
.
B
ased
o
n
th
e
v
al
u
es
r
ec
o
r
d
ed
in
Fi
g
u
r
e
7
,
th
e
co
r
r
esp
o
n
d
in
g
m
a
tr
ix
is
g
i
v
en
b
elo
w
.
w
h
er
e
th
e
t
h
r
es
h
o
ld
an
g
les ar
e
ca
lcu
lated
u
s
in
g
t
h
e
f
o
llo
w
i
n
g
f
o
r
m
u
la
(
2
)
w
it
h
°,
an
d
,
as
s
h
o
w
n
in
T
ab
le
1
(
a)
f
o
r
m
ak
i
n
g
t
h
e
class
i
f
icatio
n
d
ec
is
io
n
.
A
s
f
o
r
Me
th
o
d
B
,
o
n
l
y
ap
p
lied
f
o
r
th
e
clas
s
if
ica
tio
n
d
ec
is
io
n
,
as
s
h
o
wn
i
n
T
ab
le
1
(
b
)
b
elo
w
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
R
ea
l
-
Time
V
id
eo
P
r
o
ce
s
s
in
g
u
s
in
g
C
o
n
to
u
r
N
u
mb
ers
a
n
d
A
n
g
les fo
r
N
o
n
-
u
r
b
a
n
…
(
Za
ma
n
i Md
S
a
n
i
)
2545
T
ab
le
1
.
v
alu
e
f
o
r
(
a)
Me
th
o
d
A
an
d
(
b
)
Me
th
o
d
B
(
a)
(
b
)
2
.
4
.
Reso
lv
ing
er
ro
rs in c
la
s
s
if
ica
t
io
n
E
r
r
o
r
s
in
r
o
ad
m
ar
k
er
class
i
f
i
ca
tio
n
ten
d
to
h
ap
p
en
d
u
r
in
g
th
e
m
ar
k
er
tr
an
s
i
tio
n
p
er
io
d
.
T
h
e
f
r
a
m
e
d
elay
t
i
m
e
m
i
g
h
t
also
ca
u
s
e
e
r
r
o
r
s
i
n
class
i
f
icatio
n
.
B
o
th
o
f
th
ese
ca
s
e
s
ca
n
b
e
s
o
l
v
ed
b
y
u
s
i
n
g
th
e
te
m
p
o
r
al
in
f
o
r
m
atio
n
b
et
w
ee
n
t
h
e
clas
s
if
icatio
n
r
es
u
lt
s
d
u
r
i
n
g
t
h
e
tr
a
n
s
it
io
n
p
er
io
d
.
T
h
e
m
ar
k
er
s
t
en
d
to
b
e
s
i
m
i
lar
l
y
lab
elled
f
o
r
a
s
et
o
f
ad
j
ac
en
t
f
r
a
m
es.
K
n
o
w
in
g
t
h
is
te
n
d
en
c
y
,
th
e
p
r
ev
io
u
s
s
et
o
f
clas
s
i
f
ic
atio
n
r
esu
lt
w
ill
b
e
s
to
r
ed
a
n
d
if
th
er
e
is
a
tr
a
n
s
it
i
o
n
o
f
m
ar
k
er
t
y
p
es
,
it
w
i
ll
w
a
it
o
n
th
e
n
ex
t
s
et
o
f
f
r
a
m
e
s
cl
ass
i
f
icatio
n
r
esu
l
ts
f
o
r
a
f
in
a
l v
alid
atio
n
to
av
o
id
u
n
n
ec
es
s
ar
y
cla
s
s
i
f
icatio
n
er
r
o
r
s
.
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
T
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
w
as
t
ested
w
i
th
m
u
ltip
le
v
id
eo
s
eq
u
en
ce
s
ca
p
tu
r
ed
b
y
t
w
o
d
if
f
er
en
t
d
ev
ice
s
w
it
h
d
i
f
f
er
en
t
ca
r
v
e
lo
cities
u
s
in
g
a
lo
w
er
r
an
g
e
U
SB
ca
m
er
a
(
L
o
g
itec
h
c3
1
0
)
an
d
a
h
i
g
h
er
r
a
n
g
e
ca
m
er
a
(
Go
P
r
o
Her
o
Sil
v
er
)
.
T
o
ev
al
u
ate
th
e
r
o
b
u
s
t
n
es
s
o
f
th
e
n
e
w
al
g
o
r
ith
m
,
o
th
er
s
ets
o
f
v
i
d
eo
s
ac
q
u
ir
ed
w
it
h
d
if
f
er
e
n
t
d
ev
ices
at
d
i
f
f
er
en
t
r
eso
lu
tio
n
s
a
n
d
t
h
e
v
elo
cit
y
o
f
t
h
e
v
eh
ic
le
is
b
et
w
ee
n
6
0
-
9
0
k
m
/h
f
o
r
t
h
e
test
v
id
eo
i
m
ag
e
s
,
w
h
ic
h
m
ee
ts
t
h
e
s
p
ee
d
l
i
m
i
t
i
n
t
h
e
s
u
b
u
r
b
an
r
o
ad
ar
ea
.
T
he
v
id
eo
s
ar
e
a
ls
o
tak
e
n
w
h
e
n
th
e
v
eh
ic
le
w
as
m
o
v
in
g
o
n
d
if
f
er
en
t
r
o
ad
co
n
d
itio
n
s
in
cl
u
d
i
n
g
f
lat,
h
ill
y
a
n
d
n
ar
r
o
w
r
o
ad
s
.
T
h
e
v
id
eo
s
u
s
ed
to
d
eter
m
in
e
t
h
e
t
h
r
es
h
o
ld
an
g
les
w
er
e
tak
e
n
w
it
h
a
m
o
d
er
ate
v
e
h
icle
s
p
ee
d
b
et
w
ee
n
5
0
-
6
0
k
m
.
T
h
e
ex
p
e
r
i
m
e
n
ts
w
er
e
ca
r
r
ied
o
u
t
u
s
in
g
a
d
e
s
k
to
p
co
m
p
u
ter
eq
u
ip
p
ed
w
it
h
I
n
tel®
co
r
e
™
i7
-
4
7
9
0
C
P
U
@
3
6
0
GHz
an
d
1
6
GB
R
A
M.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
w
a
s
i
m
p
le
m
en
ted
b
y
u
s
in
g
th
e
Vi
s
u
a
l
Stu
d
io
C
++
2
0
1
0
co
m
p
iler
w
i
th
t
h
e
lib
r
ar
y
f
r
o
m
Op
en
C
V
v
er
s
io
n
2
.
4
.
1
1
.
I
n
th
e
class
i
f
icat
io
n
p
r
o
ce
s
s
,
if
,
th
e
s
o
f
t
w
ar
e
w
ill
class
i
f
y
it
as
eith
er
D
o
r
SS
d
ep
en
d
in
g
o
n
f
o
u
n
d
in
t
h
e
n
e
x
t
s
et
o
f
f
r
a
m
es
.
Fo
r
e
x
a
m
p
le,
i
f
t
h
e
n
ex
t
s
et
o
f
f
r
a
m
e
s
d
etec
ted
w
it
h
,
th
e
s
y
s
te
m
w
i
ll
class
i
f
y
i
t
as
a
D
t
y
p
es
w
h
er
ea
s
i
f
co
n
s
is
te
n
tl
y
d
etec
ted
o
v
er
th
e
co
m
p
lete
c
y
cle
,
th
e
s
y
s
te
m
w
i
ll
clas
s
i
f
y
it
a
s
SS
t
y
p
es.
I
f
t
h
e
m
a
x
i
m
u
m
v
alu
e
is
t
w
o
o
v
er
t
h
e
co
m
p
lete
cy
cle
,
th
e
r
o
ad
m
ar
k
er
w
ill
b
e
class
if
ied
as
eith
er
DS,
SD
o
r
DD
,
b
y
u
s
i
n
g
t
h
e
v
alu
e
ca
lc
u
lated
at
th
e
1
st
f
r
a
m
e
f
r
o
m
th
e
s
e
t
o
f
f
r
a
m
es
.
T
h
e
class
i
f
icatio
n
r
esu
lt
w
i
ll
r
e
m
ai
n
a
n
d
w
i
ll
o
n
l
y
b
e
r
ec
las
s
i
f
ied
if
,
ch
an
g
es
f
r
o
m
2
1
.
Fo
r
th
e
ca
s
e
o
f
DD
t
y
p
e
w
h
ic
h
ex
p
ec
ts
co
n
s
is
ten
tl
y
,
a
s
a
m
p
lin
g
u
p
to
its
1
0
th
f
r
a
m
e
s
w
il
l
b
e
p
er
f
o
r
m
ed
w
h
en
o
n
l
y
t
w
o
co
n
to
u
r
s
ar
e
d
etec
ted
an
d
class
if
i
ca
tio
n
r
esu
lt
w
i
ll
b
e
u
p
d
ated
b
y
u
s
in
g
th
e
v
al
u
e
o
n
its
1
1
th
f
r
a
m
e
.
T
h
e
ac
cu
r
ac
y
o
f
th
e
m
ar
k
er
class
i
f
icatio
n
is
ca
lcu
lated
u
s
i
n
g
E
q
u
atio
n
(
3
)
.
T
h
e
to
tal
s
ets
o
f
f
r
a
m
e
s
class
i
f
ied
co
r
r
ec
tl
y
,
Ʃ
⍴
,
w
ill
b
e
d
iv
id
ed
b
y
th
e
to
tal
s
ets
o
f
f
r
a
m
e
s
in
th
e
v
id
eo
i
m
a
g
e
Ʃ
ξ,
ex
clu
d
i
n
g
t
h
e
to
ta
l
n
u
m
b
er
o
f
1
st
s
et
s
o
f
f
r
a
m
e
s
wh
en
t
h
e
r
o
ad
m
ar
k
er
tr
an
s
it
io
n
h
ap
p
en
s
,
Ʃ
τ
1
.
T
h
e
ex
clu
s
io
n
o
f
Ʃ
τ
1
is
d
u
e
to
th
e
r
o
ad
m
ar
k
er
tr
a
n
s
i
tio
n
i
n
w
h
i
ch
t
h
e
an
g
le
ca
lc
u
lat
io
n
i
s
te
m
p
o
r
ar
il
y
p
au
s
ed
.
T
h
e
cli
p
s
an
d
th
e
ac
cu
r
ac
y
r
esu
lt
s
ar
e
s
h
o
w
n
in
T
ab
le
2
.
∑
⍴
∑
∑
∑
(
3
)
As
s
h
o
w
n
i
n
T
ab
le
1
,
clip
s
4
to
7
w
er
e
tr
ain
ed
f
o
r
id
en
tify
in
g
.
C
lip
s
9
an
d
1
0
ar
e
re
c
o
r
d
ed
w
h
e
n
th
e
v
e
h
icle
v
elo
cit
y
is
b
et
w
ee
n
6
0
an
d
8
0
k
m
/h
.
T
h
e
r
e
m
ain
in
g
test
clip
s
co
n
tai
n
at
least
t
w
o
t
y
p
e
s
o
f
r
o
ad
m
ar
k
er
s
w
it
h
d
i
f
f
er
e
n
t
r
eso
lu
tio
n
s
a
n
d
f
r
a
m
e
r
ates.
I
t
ca
n
b
e
o
b
s
er
v
ed
t
h
at
th
e
ac
c
u
r
ac
y
ac
h
ie
v
ed
b
y
Me
th
o
d
B
is
b
etter
co
m
p
ar
ed
t
o
m
et
h
o
d
A
as
o
n
l
y
o
n
e
t
h
r
es
h
o
ld
an
g
le
is
u
s
ed
.
T
h
e
p
r
o
ce
s
s
in
g
ti
m
e
p
er
f
r
a
m
e
f
o
r
Me
t
h
o
d
B
is
s
li
g
h
tl
y
lo
w
e
r
o
n
a
v
er
ag
e
as
co
m
p
ar
ed
w
it
h
Me
t
h
o
d
A
.
I
t
ca
n
also
b
e
s
e
en
i
n
T
ab
le
1
t
h
at
d
if
f
er
e
n
t
t
y
p
es
o
f
v
id
eo
co
m
p
r
ess
io
n
f
o
r
m
at
s
lik
e
.
wm
v
r
en
d
er
s
a
lo
n
g
er
p
r
o
ce
s
s
in
g
ti
m
e,
a
s
s
h
o
w
n
f
o
r
clip
6
an
d
7
.
Ou
r
m
et
h
o
d
is
a
ls
o
tes
t
ed
ag
ain
s
t
t
h
e
e
x
i
s
ti
n
g
ap
p
r
o
a
ch
p
r
esen
ted
b
y
P
au
la
at
[
1
9
]
u
s
i
n
g
a
f
e
w
v
id
eo
s
w
it
h
d
i
f
f
er
en
t
r
eso
lu
t
io
n
s
.
T
h
e
r
es
u
lted
a
v
er
ag
e
p
r
o
ce
s
s
i
n
g
ti
m
e
p
er
f
r
a
m
e
i
s
r
ec
o
r
d
ed
in
T
ab
le
3
.
Ou
r
ap
p
r
o
ac
h
h
as
b
ee
n
o
b
s
er
v
ed
t
o
p
er
f
o
r
m
b
etter
at
lar
g
er
r
eso
lu
tio
n
s
.
T
h
is
is
co
n
tr
ib
u
ted
m
ain
l
y
b
y
t
h
e
s
m
alle
r
R
OI
ap
p
lied
i
n
o
u
r
ap
p
r
o
ac
h
,
b
esid
es
t
h
e
t
w
o
f
o
r
m
u
la
ted
f
e
atu
r
es
n
a
m
el
y
t
h
e
co
n
to
u
r
n
u
m
b
er
an
d
an
g
le
u
s
ed
to
class
i
f
y
t
h
e
r
o
ad
m
ar
k
er
t
y
p
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
2
5
4
0
–
2
5
4
8
2546
T
ab
le
2
.
C
lip
s
I
n
f
o
r
m
atio
n
an
d
th
e
A
cc
u
r
ac
y
R
es
u
lt
s
f
o
r
b
o
th
m
et
h
o
d
s
T
ab
le
3
: A
v
er
ag
e
T
i
m
e
E
x
ec
u
tio
n
f
o
r
E
ac
h
Fra
m
e
R
e
so
l
u
t
i
o
n
S
i
z
e
A
v
e
r
a
g
e
T
i
me
p
e
r
f
r
a
me
(
ms)
M
.
B
.
d
e
P
a
u
l
a
a
n
d
C
.
R
.
J
u
n
g
(
1
5
)
N
e
w
p
r
o
p
o
se
d
M
e
t
h
o
d
640X
480
3
.
7
1
7
.
1
1
2
8
0
X
7
2
0
6
.
5
3
8
.
7
8
1
9
2
0
X
1
0
8
0
1
6
.
5
9
1
1
.
1
9
4.
CO
NCLU
SI
O
N
A
n
o
v
el
ap
p
r
o
ac
h
o
n
th
e
r
o
a
d
m
ar
k
er
s
clas
s
i
f
icatio
n
p
r
o
ce
s
s
is
p
r
ese
n
ted
in
t
h
i
s
p
ap
er
.
A
s
m
al
ler
R
OI
i
s
u
s
ed
to
p
r
o
ce
s
s
t
h
e
v
i
d
eo
f
r
a
m
es
a
n
d
ex
tr
ac
t
t
h
e
f
o
r
m
u
la
ted
f
ea
tu
r
es,
w
h
ich
ar
e
t
h
e
co
n
to
u
r
n
u
m
b
er
an
d
an
g
le,
as
t
h
e
i
n
p
u
t
s
f
o
r
th
e
p
r
o
p
o
s
ed
t
w
o
-
la
y
er
cla
s
s
i
f
ie
r
alg
o
r
ith
m
.
T
h
e
av
er
ag
e
ac
c
u
r
ac
y
o
f
~9
6
%
h
a
s
b
ee
n
ac
h
iev
ed
u
s
in
g
o
u
r
p
r
o
p
o
s
ed
alg
o
r
ith
m
,
w
it
h
a
r
elati
v
e
l
y
lo
w
er
p
r
o
ce
s
s
in
g
ti
m
e
p
er
f
r
a
m
e
at
lar
g
e
v
id
eo
r
eso
lu
tio
n
a
s
co
m
p
ar
ed
w
ith
th
e
ex
is
ti
n
g
ap
p
r
o
ac
h
.
T
h
e
er
r
o
r
s
f
o
u
n
d
i
n
t
h
e
r
o
ad
m
ar
k
e
r
class
i
f
icatio
n
ar
e
ca
u
s
ed
b
y
t
h
e
ab
r
u
p
t
ch
a
n
g
es
in
i
llu
m
i
n
atio
n
,
v
an
i
s
h
in
g
r
o
ad
m
ar
k
er
,
w
h
i
te
ele
m
e
n
ts
o
n
t
h
e
r
o
ad
an
d
m
ar
k
er
s
b
lo
ck
in
g
b
y
t
h
e
o
th
er
v
e
h
icle
s
.
Fu
t
u
r
e
w
o
r
k
w
i
ll
f
o
cu
s
o
n
th
e
illu
m
i
n
atio
n
a
n
d
f
ai
n
t
m
ar
k
e
r
is
s
u
e
s
th
a
t
af
f
ec
t
th
e
f
ea
tu
r
es e
x
tr
ac
tio
n
f
o
r
clas
s
if
y
i
n
g
th
e
r
o
ad
m
ar
k
er
t
y
p
e
s
.
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
au
th
o
r
s
w
o
u
ld
lik
e
to
a
ck
n
o
w
led
g
e
th
e
f
u
n
d
in
g
s
u
p
p
o
r
t
r
ec
eiv
ed
f
r
o
m
Un
iv
er
s
it
i
T
ek
n
ik
al
Ma
la
y
s
ia
Me
lak
a
(
UT
eM
)
f
r
o
m
P
J
P
/2
0
1
4
/FKE(
1
5
D)
/S0
1
3
5
7
–
R
o
ad
L
an
e
R
ec
o
g
n
it
io
n
Usi
n
g
Vis
io
n
an
d
A
r
ti
f
icial
I
n
te
lli
g
en
ce
f
o
r
Au
to
A
s
s
is
t
Dr
i
v
er
A
ler
t
S
y
s
t
e
m
.
I
n
ad
d
iti
o
n
,
a
s
p
ec
ial
t
h
an
k
s
to
U
n
i
v
er
s
it
i
T
ek
n
o
lo
g
i
Ma
la
y
s
ia
(
UT
M)
f
o
r
f
u
n
d
in
g
th
i
s
p
ap
er
u
n
d
er
P
AS
Gr
a
n
t
N
u
m
b
er
P
Y/2
0
1
6
/0
7
7
3
2
.
T
h
e
au
th
o
r
s
also
w
o
u
ld
li
k
e
to
ac
k
n
o
w
led
g
e
Mo
h
a
m
m
ed
Fa
u
zi
Mo
h
a
m
m
ed
Hu
s
s
ei
n
a
s
an
in
d
u
s
tr
y
e
x
p
er
t
w
h
o
h
as
g
i
v
en
ad
v
ice
an
d
s
u
p
p
o
r
t th
r
o
u
g
h
o
u
t
th
e
f
u
ll p
r
o
j
ec
t c
o
m
p
letio
n
.
RE
F
E
R
E
NC
E
S
[1
]
M
u
sta
f
a
,
M
.
N.
,
“
Ov
e
rv
ie
w
o
f
C
u
rre
n
t
R
o
a
d
S
a
f
e
t
y
S
it
u
a
ti
o
n
In
M
a
la
y
sia
”
,
Hig
h
w
a
y
P
lan
n
i
n
g
U
n
it
,
Ro
a
d
S
a
f
e
t
y
S
e
c
ti
o
n
,
M
in
istry
o
f
W
o
rk
s
,
2
0
0
5
.
[2
]
M
it
a
,
S
.
,
&
M
c
A
ll
e
ste
r,
D
.
,
“
Lan
e
d
e
tec
ti
o
n
a
n
d
trac
k
in
g
in
c
h
a
ll
e
n
g
in
g
e
n
v
iro
n
m
e
n
ts
b
a
se
d
o
n
a
w
e
i
g
h
ted
g
ra
p
h
a
n
d
in
teg
ra
ted
c
u
e
s
”
,
2
0
1
0
IEE
E/
RS
J
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
In
telli
g
e
n
t
Ro
b
o
ts
a
n
d
S
y
ste
ms
,
p
p
.
5
5
4
3
-
5
5
5
0
,
2
0
1
0
.
[3
]
Zh
e
,
X.,
&
Zh
if
e
n
g
,
L
.
,
“
A
ro
b
u
st
lan
e
d
e
tec
ti
o
n
m
e
th
o
d
in
t
h
e
d
if
fe
re
n
t
sc
e
n
a
rio
s
”
,
2
0
1
2
IE
E
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
M
e
c
h
a
tro
n
ics
a
n
d
Au
to
m
a
ti
o
n
,
p
p
.
1
3
5
8
-
1
3
6
3
,
2
0
1
2
.
[4
]
F
a
n
Ch
a
o
,
Xu
Jin
g
-
b
o
&
DI
S
h
u
a
i
,
“
L
a
n
e
De
tec
ti
o
n
Ba
se
d
o
n
M
a
c
h
in
e
L
e
a
rn
in
g
A
lg
o
rit
h
m
”
,
T
EL
KOM
NIKA
(
T
e
lec
o
mu
u
n
ica
ti
o
n
,
Co
mp
u
ti
n
g,
El
e
c
tro
n
ics
a
n
d
C
o
n
t
ro
l)
,
v
ol
.
12
-
2
,
p
p
.
1
4
0
3
-
1
4
0
9
,
2
0
1
4
.
[5
]
C
h
u
n
y
a
n
g
M
u
,
X
i
n
g
M
a
.
,
“
L
a
n
e
De
tec
ti
o
n
Ba
se
d
o
n
Ob
jec
t
S
e
g
m
e
n
tatio
n
a
n
d
P
iec
e
w
ise
F
it
ti
n
g
”
,
T
EL
KOM
NIKA
(
T
e
lec
o
mu
u
n
ica
ti
o
n
,
Co
mp
u
ti
n
g,
El
e
c
tro
n
ics
a
n
d
C
o
n
t
ro
l)
,
v
ol
.
12
-
5
,
p
p.
3
4
9
1
-
3
5
0
0
,
2
0
1
4
.
[6
]
X
iao
d
a
n
Hu
a
n
g
,
W
e
i
W
a
n
g
.
,
“
T
ra
ff
ic
P
re
d
ictio
n
Ba
se
d
o
n
C
o
rre
latio
n
o
f
Ro
a
d
S
e
c
ti
o
n
s
”
,
T
EL
KOM
NIKA
(
T
e
lec
o
mu
u
n
ica
ti
o
n
,
Co
mp
u
ti
n
g,
El
e
c
tro
n
ics
a
n
d
C
o
n
t
ro
l)
,
v
o
l.
1
1
,
n
o
.
1
0
,
p
p
.
5
5
2
3
-
5
5
2
9
,
Oc
to
b
e
r
2
0
1
3
.
[7
]
S
o
n
,
J.,
Yo
o
,
H.,
Kim
,
S
.
,
&
S
o
h
n
,
K.
,
“
Ex
p
e
rt
S
y
ste
m
s
w
it
h
Ap
p
li
c
a
ti
o
n
s
Re
a
l
-
ti
m
e
il
lu
m
in
a
ti
o
n
in
v
a
rian
t
lan
e
d
e
tec
ti
o
n
f
o
r
la
n
e
d
e
p
a
rt
u
re
w
a
rn
in
g
sy
ste
m
”
,
Exp
e
rt S
y
ste
ms
w
it
h
Ap
p
li
c
a
ti
ons
,
v
ol
.
42
-
4
,
pp
.
1
8
1
6
-
1
8
2
4
.
2
0
1
5
.
[8
]
A
n
,
X
.
,
S
h
a
n
g
,
E.
,
S
o
n
g
,
J
.
,
L
i,
J.
,
&
He
,
H.
,
“
Re
a
l
-
ti
m
e
lan
e
d
e
p
a
rtu
re
w
a
rn
in
g
sy
st
e
m
b
a
se
d
o
n
a
sin
g
le
F
P
G
A
”
,
EURA
S
IP
J
o
u
rn
a
l
o
n
Ima
g
e
a
n
d
Vi
d
e
o
Pr
o
c
e
ss
in
g
2
0
1
3
,
p
p
.
1
-
1
8
,
2
0
1
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
R
ea
l
-
Time
V
id
eo
P
r
o
ce
s
s
in
g
u
s
in
g
C
o
n
to
u
r
N
u
mb
ers
a
n
d
A
n
g
les fo
r
N
o
n
-
u
r
b
a
n
…
(
Za
ma
n
i Md
S
a
n
i
)
2547
[9
]
Ku
m
tep
e
,
O.,
A
k
a
r,
G
.
B.
,
&
Yu
n
c
u
,
E
.
,
“
Driv
e
r
a
g
g
re
s
siv
e
n
e
ss
d
e
tec
ti
o
n
v
ia
m
u
lt
ise
n
so
ry
d
a
ta
f
u
sio
n
”
,
EURA
S
IP
J
o
u
rn
a
l
o
n
Im
a
g
e
a
n
d
Vi
d
e
o
Pro
c
e
ss
in
g
,
v
ol
.1
,
p
p
.
5
,
2
0
1
6
.
[1
0
]
Ch
ira,
I.
M
.
,
Ch
ib
u
lcu
tea
n
,
&
Da
n
e
sc
u
,
R.
G
.
,
“
Re
a
l
-
ti
m
e
d
e
t
e
c
ti
o
n
o
f
ro
a
d
m
a
rk
in
g
s
f
o
r
d
riv
in
g
a
ss
i
sta
n
c
e
a
p
p
li
c
a
ti
o
n
s
”
,
Co
m
p
u
ter
E
n
g
i
n
e
e
rin
g
a
n
d
S
y
ste
ms
ICCES
2
0
1
0
In
t
e
rn
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
,
p
p
.
1
5
8
-
1
6
3
,
2
0
1
0
[1
1
]
W
u
,
T
.
,
&
Ra
n
g
a
n
a
th
a
n
,
A
.
,
“
A
p
ra
c
ti
c
a
l
s
y
ste
m
f
o
r
ro
a
d
m
a
r
k
in
g
d
e
tec
ti
o
n
a
n
d
re
c
o
g
n
it
io
n
”
,
IEE
E
In
telli
g
e
n
t
Veh
icle
s S
y
mp
o
si
u
m,
Pr
o
c
e
e
d
in
g
s
,
p
p
.
25
-
30
,
2
0
1
2
.
[1
2
]
M
c
Ca
ll
,
J
.
C.
,
&
T
riv
e
d
i,
M
.
M
.
,
“
V
i
d
e
o
-
Ba
se
d
L
a
n
e
Esti
m
a
ti
o
n
a
n
d
T
ra
c
k
in
g
f
o
r
Driv
e
r
A
ss
is
tan
c
e
:
S
u
rv
e
y
,
S
y
st
e
m
,
a
n
d
Ev
a
lu
a
ti
o
n
”
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
I
n
telli
g
e
n
t
T
ra
n
sp
o
rta
ti
o
n
S
y
ste
ms
,
v
ol
.
7
-
1
,
p
p
.
20
-
37
,
2
0
0
6
.
[1
3
]
Ye
n
ik
a
y
a
,
S
.
,
Ye
n
ik
a
y
a
,
G
.
,
&
D
ü
v
e
n
,
E.
,
“
Ke
e
p
in
g
th
e
V
e
h
icle
o
n
th
e
Ro
a
d
:
A
S
u
rv
e
y
o
n
On
-
ro
a
d
Lan
e
De
te
c
ti
o
n
S
y
st
e
m
s
”
,
ACM
Co
mp
u
t.
S
u
rv
.,
v
o
ls.
4
6
-
1,
p
p
.
2
:
1
--
2
:
4
3
,
2
0
1
3
.
[1
4
]
Co
ll
a
d
o
,
J.,
Hi
lario
,
C
.
,
De
L
a
Esc
a
lera
,
&
A
r
m
in
g
o
l,
J.
,
“
A
d
a
p
tativ
e
ro
a
d
lan
e
s
d
e
tec
ti
o
n
a
n
d
c
las
si
f
ica
ti
o
n
”
,
Ad
v
a
n
c
e
d
Co
n
c
e
p
ts
fo
r I
n
telli
g
e
n
t
Vi
sio
n
S
y
ste
ms
,
p
p
.
1
1
5
1
-
1
1
6
2
,
2
0
0
6
.
[1
5
]
S
c
h
u
b
e
rt
,
R.
,
S
c
h
u
lze
,
K.,
&
W
a
n
ielik
,
G
.
,
“
S
it
u
a
ti
o
n
a
ss
e
ss
m
e
n
t
f
o
r
a
u
to
m
a
ti
c
lan
e
-
c
h
a
n
g
e
m
a
n
e
u
v
e
rs
”
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
In
tell
ig
e
n
t
T
ra
n
s
p
o
rta
ti
o
n
S
y
ste
ms
,
v
o
l
.
1
1
-
3
,
p
p
.
6
0
7
-
6
1
6
,
2
0
1
0
.
[1
6
]
L
in
d
n
e
r,
P
.
,
Blo
k
z
y
l,
S
.
,
W
a
n
ielik
,
G
.
,
&
S
c
h
e
u
n
e
rt,
U.
,
“
A
p
p
ly
in
g
m
u
lt
i
lev
e
l
p
ro
c
e
ss
in
g
f
o
r
ro
b
u
s
t
g
e
o
m
e
tri
c
lan
e
f
e
a
tu
re
e
x
trac
ti
o
n
”
,
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
M
u
lt
ise
n
so
r
Fu
sio
n
a
n
d
I
n
teg
ra
t
io
n
fo
r
In
tell
ig
e
n
t
S
y
ste
ms
,
p
p
.
2
4
8
-
2
5
4
,
2
0
1
0
.
[1
7
]
S
u
c
h
it
ra
,
S
.
,
S
a
tzo
d
a
,
R.
K
.
,
&
S
rik
a
n
th
a
n
,
T
.
,
“
Id
e
n
ti
fy
in
g
lan
e
ty
p
e
s:
A
m
o
d
u
lar
a
p
p
ro
a
c
h
”
,
1
6
th
In
ter
n
a
ti
o
n
a
l
IEE
E
Co
n
fer
e
n
c
e
o
n
In
telli
g
e
n
t
T
ra
n
sp
o
rta
ti
o
n
S
y
ste
ms
(
IT
S
C
2
0
1
3
)
,
p
p
.
1
9
2
9
-
1
9
3
4
,
2
0
1
3
.
[1
8
]
Ne
d
e
v
sc
h
i,
S
.
,
P
o
p
e
sc
u
,
V
.
,
Da
n
e
sc
u
,
R.
,
M
a
rit
a
,
T
.
,
&
O
n
ig
a
,
F
.
,
“
A
c
c
u
ra
te
e
g
o
-
v
e
h
icle
g
lo
b
a
l
lo
c
a
li
z
a
ti
o
n
a
t
in
ters
e
c
ti
o
n
s
t
h
ro
u
g
h
a
li
g
n
m
e
n
t
o
f
v
isu
a
l
d
a
ta
w
it
h
d
ig
it
a
l
m
a
p
”
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
In
tel
li
g
e
n
t
T
ra
n
sp
o
rta
ti
o
n
S
y
ste
ms
,
v
o
ls.
1
4
-
2,
p
p
.
6
7
3
-
6
8
7
,
2
0
1
3
.
[1
9
]
P
a
u
la,
M
.
B.
,
&
Ju
n
g
,
C.
R.
,
“
A
u
to
m
a
ti
c
De
tec
ti
o
n
a
n
d
Clas
sif
ica
ti
o
n
o
f
Ro
a
d
L
a
n
e
M
a
rk
in
g
s
Us
in
g
On
b
o
a
rd
V
e
h
icu
lar Cam
e
r
a
s
”
,
IEE
E
T
ra
n
s
a
c
ti
o
n
s o
n
In
telli
g
e
n
t
T
ra
n
sp
o
rta
t
io
n
S
y
ste
ms
,
v
ol
.
1
6
-
6
,
p
p
.
3
1
6
0
–
3
1
6
9
,
2
0
1
5
.
[2
0
]
M
a
th
ib
e
la,
B.
,
Ne
wm
a
n
,
P
.
,
&
P
o
sn
e
r,
I.
,
“
Re
a
d
in
g
th
e
Ro
a
d
:
Ro
a
d
M
a
rk
in
g
Cl
a
ss
i
f
ica
ti
o
n
a
n
d
In
terp
re
tati
o
n
”
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
In
telli
g
e
n
t
T
ra
n
sp
o
rta
ti
o
n
S
y
ste
ms
,
v
ol
.
1
6
-
4
,
p
p
.
1
-
1
0
.
2
0
1
5
.
[2
1
]
Zh
a
n
g
,
G
.
,
Zh
e
n
g
,
N.,
Cu
i,
C.
,
Ya
n
,
Y.,
&
Yu
a
n
,
Z.
,
“
A
n
e
ff
icie
n
t
ro
a
d
d
e
tec
ti
o
n
m
e
th
o
d
in
n
o
isy
u
rb
a
n
e
n
v
iro
n
m
e
n
t
”
,
IEE
E
In
t
e
ll
ig
e
n
t
V
e
h
icle
s S
y
mp
o
si
u
m,
Pro
c
e
e
d
i
n
g
s,
pp.
5
5
6
-
561
,
2
0
0
9
.
[2
2
]
Br
o
g
g
i,
A
.
,
&
Ca
tt
a
n
i,
S
.
,
“
A
n
a
g
e
n
t
b
a
se
d
e
v
o
l
u
ti
o
n
a
ry
a
p
p
ro
a
c
h
to
p
a
th
d
e
tec
ti
o
n
f
o
r
o
f
f
-
ro
a
d
v
e
h
icle
g
u
id
a
n
c
e
”
,
Pa
tt
e
rn
Rec
o
g
n
it
io
n
L
e
tt
e
rs
,
v
o
l
.
27
,
n
o
.
11
,
p
p
.
1
1
6
4
-
1
1
7
3
.
[2
3
]
S
a
tzo
d
a
,
R.
K.,
&
T
ri
v
e
d
i,
M
.
M
.
,
“
V
isio
n
-
b
a
se
d
lan
e
a
n
a
ly
sis:
Ex
p
lo
ra
ti
o
n
o
f
issu
e
s
a
n
d
a
p
p
r
o
a
c
h
e
s
f
o
r
e
m
b
e
d
d
e
d
re
a
li
z
a
ti
o
n
”
,
IEE
E
C
o
mp
u
ter
S
o
c
iety
Co
n
fer
e
n
c
e
o
n
C
o
mp
u
te
r
Vi
sio
n
a
n
d
Pa
tt
e
rn
Rec
o
g
n
it
i
o
n
W
o
rk
sh
o
p
s
,
pp.
6
0
4
-
6
0
9
.
[2
4
]
J.
C.
R
u
ss
a
n
d
J.
C
.
R
u
ss
,
In
t
ro
d
u
c
ti
o
n
t
o
Im
a
g
e
P
ro
c
e
ss
in
g
a
n
d
A
n
a
ly
sis
,
(CRC
P
re
ss
,
B
o
c
a
Ra
to
n
2
0
0
8
)
,
p
p
.
7
2
-
7
9
,
2
0
0
8
.
[2
5
]
N.
Otsu
,
“
A
th
re
sh
o
ld
se
lec
ti
o
n
m
e
th
o
d
f
ro
m
g
ra
y
-
lev
e
l
h
isto
g
ra
m
s
”
,
IEE
E
T
ra
n
s.
S
y
st.,
M
a
n
Cy
b
e
rn
.
,
v
o
l.
S
M
C*
9
,
Fu
ll
-
T
e
x
t,
v
ol
.
2
0
-
1
,
pp
.
6
2
-
6
6
,
1
9
7
9
.
[
26]
G
.
Bra
d
sk
i
a
n
d
A
.
Ka
e
h
le,
L
e
a
rn
in
g
Op
e
n
C
V
.
(O'
Re
il
ly
M
e
d
ia,
In
c
,
Ca
li
f
o
rn
ia,
2
0
0
8
)
,
p
p
.
2
4
1
-
2
5
5
,
2
0
0
8
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Za
m
a
n
i
M
d
S
a
n
i
re
c
e
iv
e
d
h
is
Ba
c
h
e
lo
r
a
n
d
M
a
ste
r
i
n
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
in
e
e
rin
g
f
ro
m
Un
iv
e
rsit
y
S
a
in
s
M
a
la
y
sia
in
2
0
0
0
a
n
d
2
0
0
9
.
C
u
rre
n
tl
y
,
h
e
is
p
u
rsu
in
g
h
is
P
h
.
D
.
d
e
g
re
e
f
ro
m
M
M
U
a
n
d
a
lso
a
sta
f
f
m
e
m
b
e
r
a
t
Un
iv
e
r
siti
T
e
k
n
ik
a
l
M
a
la
y
sia
M
e
lak
a
.
H
a
d
h
r
a
m
i
A
b
G
h
a
n
i
re
c
e
iv
e
d
h
is
b
a
c
h
e
l
o
r
d
e
g
re
e
in
e
lec
tro
n
ics
e
n
g
in
e
e
rin
g
f
ro
m
M
u
lt
im
e
d
ia
Un
iv
e
rsit
y
M
a
la
y
sia
(M
M
U)
in
2
0
0
2
.
I
n
2
0
0
4
,
h
e
c
o
m
p
lete
d
h
is
m
a
ste
rs
d
e
g
re
e
in
T
e
lec
o
m
m
u
n
ica
ti
o
n
En
g
in
e
e
rin
g
a
t
T
h
e
Un
iv
e
rsit
y
o
f
M
e
lb
o
u
rn
e
.
He
th
e
n
p
u
rsu
e
d
h
is
P
h
.
D.
a
t
Im
p
e
rial
Co
ll
e
g
e
L
o
n
d
o
n
in
th
e
sa
m
e
st
u
d
y
a
re
a
a
n
d
c
o
m
p
le
ted
h
is
P
h
.
D.
re
se
a
rc
h
in
2
0
1
1
.
Cu
rre
n
tl
y
,
h
e
se
rv
e
s
a
s
o
n
e
o
f
th
e
a
c
a
d
e
m
ic an
d
re
se
a
rc
h
sta
ff
m
e
m
b
e
rs at M
M
U.
Ro
sli
B
e
sa
r
is
c
u
rre
n
tl
y
a
ss
o
c
iat
e
p
ro
f
e
ss
o
r
a
t
Un
iv
e
rsit
y
o
f
M
u
lt
im
e
d
ia,
re
c
e
iv
e
d
th
e
B.
En
g
(Ho
n
s)
a
n
d
M
.
S
c
d
e
g
re
e
s
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
S
c
ien
c
e
M
a
la
y
sia
(U
S
M
),
M
a
lay
sia
,
in
1
9
9
0
a
n
d
1
9
9
3
,
re
sp
e
c
ti
v
e
l
y
a
n
d
th
e
P
h
.
D.
d
e
g
re
e
f
ro
m
th
e
M
u
lt
im
e
d
ia
Un
iv
e
r
s
it
y
,
M
a
la
y
sia
,
in
2
0
0
4
.
His
c
u
rr
e
n
t
in
tere
sts in
c
lu
d
e
S
ig
n
a
l
a
n
d
Im
a
g
e
P
r
o
c
e
ss
in
g
,
M
e
d
ica
l
Im
a
g
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
2
5
4
0
–
2
5
4
8
2548
Az
izu
l
Az
iz
a
n
o
b
tain
e
d
h
is
P
h
D
q
u
a
li
f
ica
ti
o
n
(2
0
0
9
),
f
ro
m
Un
iv
e
r
sity
o
f
S
u
rre
y
re
se
a
rc
h
in
g
o
n
3
.
5
G
p
h
y
sic
a
l
la
y
e
r
a
d
a
p
tatio
n
f
o
r
sa
telli
te
s
y
ste
m
s.
He
lat
e
r
jo
in
e
d
t
h
e
M
a
lay
si
a
n
Co
m
m
u
n
ica
ti
o
n
a
n
d
M
u
lt
im
e
d
ia
Co
m
m
issio
n
f
o
r
m
o
re
th
a
n
6
y
e
a
r
s
o
v
e
rse
e
in
g
sp
e
c
tru
m
a
n
d
n
u
m
b
e
rin
g
p
o
li
c
ies
a
n
d
re
so
u
rc
e
m
a
n
a
g
e
m
e
n
t.
He
is
c
u
rr
e
n
tl
y
w
it
h
A
d
v
a
n
c
e
d
In
f
o
r
m
a
ti
c
s
S
c
h
o
o
l
(A
IS
),
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
la
y
sia
Ku
a
la
L
u
m
p
u
r
w
h
e
re
h
is
re
se
a
rc
h
a
r
e
a
s
in
c
lu
d
e
s
Tele
c
o
m
m
u
n
ica
ti
o
n
s,
Cy
b
e
rp
h
y
si
c
a
l
S
y
st
e
m
s (i
n
c
Io
T
),
Bu
sin
e
ss
In
tell
ig
e
n
c
e
a
n
d
En
g
in
e
e
rin
g
E
d
u
c
a
ti
o
n
.
H
a
fiza
A
b
a
s
is
a
re
se
a
rc
h
e
r,
se
n
io
r
lec
tu
re
r,
train
e
r
a
n
d
w
rit
e
r
a
t
A
d
v
a
n
c
e
d
In
f
o
rm
a
ti
c
s
S
c
h
o
o
l
(A
IS
)
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
lay
sia
,
Ku
a
la
L
u
m
p
u
r.
Ha
f
iza
h
a
s
p
ro
d
u
c
e
d
a
n
u
m
b
e
r
o
f
b
o
o
k
s,
tea
c
h
i
n
g
m
o
d
u
les
,
a
tran
sla
ti
o
n
b
o
o
k
,
a
n
d
jo
u
r
n
a
ls.
S
h
e
h
a
s
p
a
rti
c
i
p
a
ted
in
m
a
n
y
c
o
n
f
e
re
n
c
e
s
a
n
d
re
se
a
rc
h
e
x
h
ib
it
io
n
s
a
n
d
in
n
a
ti
o
n
a
l
a
n
d
in
tern
a
ti
o
n
a
l
lev
e
ls.
H
e
r
re
se
a
r
c
h
in
tere
sts
is
re
l
a
ted
to
e
d
u
c
a
ti
o
n
tec
h
n
o
l
o
g
y
(a
u
g
m
e
n
ted
re
a
li
t
y
,
a
p
p
s,
m
u
lt
im
e
d
ia)
a
n
d
lea
rn
in
g
d
is
a
b
il
it
ies
(d
y
sle
x
ia,
re
m
e
d
ial
stu
d
e
n
ts
a
n
d
a
u
ti
sm
).
S
h
e
is
a
lso
a
c
e
rti
f
ie
d
so
f
tw
a
r
e
tes
ter
in
CP
-
M
A
T
(Ce
rti
f
ied
P
r
o
f
e
ss
io
n
a
l
–
M
a
ste
r
A
g
i
le
T
e
stin
g
).
Evaluation Warning : The document was created with Spire.PDF for Python.