I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
,
p
p
.
3267
~
3
2
7
8
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
3
.
pp
3
2
6
7
-
3
2
7
8
3267
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Co
mpa
ra
tive a
na
ly
sis
of YO
L
O
v
8
techniqu
es:
O
pen
CV and
co
o
rdina
te
a
tt
en
ti
o
n weighting
for
dista
nce pe
rcept
i
o
n in blind
na
v
ig
a
tion sy
stems
E
m
a
Ut
a
m
i,
E
rwin
Sy
a
hrud
i
n,
Ang
g
it
Dwi
H
a
rt
a
nto
M
a
s
t
e
r
o
f
I
n
f
o
r
ma
t
i
c
s,
F
a
c
u
l
t
y
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
U
n
i
v
e
r
si
t
a
s
A
mi
k
o
m
Y
o
g
y
a
k
a
r
t
a
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
1
4
,
2
0
2
4
R
ev
is
ed
Dec
1
8
,
2
0
2
4
Acc
ep
ted
J
an
1
6
,
2
0
2
5
Bli
n
d
n
e
ss
is
a
v
e
ry
imp
o
rta
n
t
iss
u
e
to
c
o
n
sid
e
r
in
re
se
a
rc
h
a
ime
d
a
t
a
ss
isti
n
g
v
isio
n
.
Th
is
c
o
n
d
it
io
n
re
q
u
ires
f
u
r
th
e
r
st
u
d
y
t
o
p
r
o
v
i
d
e
so
lu
ti
o
n
s
fo
r
th
e
b
li
n
d
.
Th
is
stu
d
y
e
v
a
lu
a
tes
a
n
d
c
o
m
p
a
re
s
th
e
e
ffe
c
ti
v
e
n
e
ss
o
f
th
e
y
o
u
o
n
l
y
lo
o
k
o
n
c
e
v
8
(YO
LOv
8
)
m
o
d
e
l
in
te
g
ra
ted
with
Op
e
n
CV
a
n
d
t
h
e
c
o
o
r
d
in
a
t
e
a
tt
e
n
ti
o
n
we
ig
h
ti
n
g
(CAW)
tec
h
n
iq
u
e
f
o
r
d
istan
c
e
e
stim
a
ti
o
n
in
a
b
l
in
d
n
a
v
ig
a
ti
o
n
sy
ste
m
.
In
it
ially
,
YO
LOv
8
in
teg
ra
ted
with
Op
e
n
CV
p
ro
d
u
c
e
d
les
s
th
a
n
o
p
ti
m
a
l
re
su
lt
s
,
p
r
o
m
p
ti
n
g
fu
rt
h
e
r
imp
ro
v
e
m
e
n
t
e
ffo
rts
t
o
su
rp
a
ss
th
e
p
e
rfo
rm
a
n
c
e
o
f
CAW.
Th
e
g
o
a
l
i
s
to
e
n
h
a
n
c
e
th
e
a
c
c
u
ra
c
y
a
n
d
e
ff
icie
n
c
y
o
f
d
istan
c
e
p
e
rc
e
p
ti
o
n
wit
h
o
u
t
t
h
e
n
e
e
d
f
o
r
a
d
d
it
i
o
n
a
l
se
n
so
rs.
Th
e
m
a
teria
ls
u
se
d
in
c
l
u
d
e
a
v
a
riet
y
o
f
d
a
tas
e
ts
a
n
n
o
tate
d
with
d
istan
c
e
in
f
o
rm
a
ti
o
n
t
o
train
a
n
d
e
v
a
lu
a
te t
h
e
m
o
d
e
l.
Th
e
m
e
th
o
d
s e
m
p
l
o
y
e
d
i
n
c
lu
d
e
i
n
teg
ra
ti
n
g
YO
LOv
8
with
Op
e
n
CV
fo
r
b
a
se
li
n
e
c
o
m
p
a
riso
n
a
n
d
a
p
p
l
y
in
g
CAW
t
o
imp
ro
v
e
d
istan
c
e
p
e
rc
e
p
ti
o
n
t
h
ro
u
g
h
e
n
h
a
n
c
e
d
fe
a
tu
re
a
tt
e
n
ti
o
n
.
Th
e
re
s
u
lt
s
sh
o
w
th
a
t
YO
LOv
8
+
Op
e
n
CV
Im
p
r
o
v
e
d
a
c
h
iev
e
s
th
e
lo
we
st
m
e
a
n
sq
u
a
re
d
e
rro
r
(M
S
E)
a
c
ro
ss
th
e
e
n
ti
re
d
ista
n
c
e
ra
n
g
e
:
0
-
1
m
(
0
.
4
4
),
1
-
2
m
(0
.
5
0
),
2
-
3
m
(0
.
5
8
),
3
-
4
m
(0
.
6
4
)
,
a
n
d
4
-
5
m
(0
.
7
1
)
.
YO
LOv
8
+
CAW
a
lso
o
u
tp
e
rfo
rm
s
YO
LOv
8
+
Op
e
n
CV
o
r
ig
i
n
a
l
,
d
e
m
o
n
stra
ti
n
g
a
n
o
ta
b
le
e
n
h
a
n
c
e
m
e
n
t
in
a
c
c
u
ra
c
y
.
Th
e
m
o
d
e
l
a
c
h
ie
v
e
s
a
d
e
tec
ti
o
n
a
c
c
u
ra
c
y
o
f
9
5
.
7
%
,
sh
o
w
c
a
sin
g
th
e
e
ffe
c
ti
v
e
n
e
ss
o
f
c
o
m
p
u
ter
v
isio
n
tec
h
n
iq
u
e
s
in
su
p
p
o
rti
n
g
b
li
n
d
n
a
v
ig
a
ti
o
n
sy
ste
m
s,
o
ffe
rin
g
p
re
c
ise
d
istan
c
e
e
stim
a
ti
o
n
c
a
p
a
b
i
li
ti
e
s
a
n
d
re
d
u
c
in
g
th
e
re
li
a
n
c
e
o
n
e
x
tern
a
l
se
n
so
rs.
T
h
e
imp
li
c
a
ti
o
n
s
in
c
l
u
d
e
imp
r
o
v
e
d
re
a
l
-
ti
m
e
p
e
rfo
r
m
a
n
c
e
a
n
d
a
c
c
e
ss
ib
il
it
y
fo
r
th
e
b
li
n
d
,
p
a
v
in
g
t
h
e
wa
y
f
o
r
m
o
r
e
e
fficie
n
t
a
n
d
re
li
a
b
le
n
a
v
i
g
a
ti
o
n
a
ss
istan
c
e
tec
h
n
o
l
o
g
ies
.
K
ey
w
o
r
d
s
:
B
lin
d
p
er
s
o
n
C
o
m
p
u
ter
v
is
io
n
C
o
o
r
d
in
ate
atten
tio
n
weig
h
tin
g
Op
en
C
V
YOL
Ov
8
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
:
E
m
a
Utam
i
Ma
s
ter
o
f
I
n
f
o
r
m
atics,
Facu
lty
o
f
C
o
m
p
u
te
r
Scien
ce
,
Un
iv
e
r
s
itas
Am
ik
o
m
Yo
g
y
ak
ar
ta
Slem
an
,
Yo
g
y
ak
ar
ta,
I
n
d
o
n
esia
E
m
ail:
em
a.
u
@
am
ik
o
m
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
R
ec
en
t a
d
v
an
ce
m
e
n
ts
in
d
ee
p
lear
n
in
g
tech
n
iq
u
es h
av
e
led
t
o
s
ig
n
if
ican
t
p
r
o
g
r
ess
in
co
m
p
u
ter
v
is
io
n
tech
n
o
lo
g
ies,
en
a
b
lin
g
m
o
r
e
e
f
f
icien
t
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
p
atter
n
r
ec
o
g
n
itio
n
[
1
]
,
[
2
]
.
T
h
ese
im
p
r
o
v
em
e
n
ts
h
av
e
s
ig
n
if
ican
tly
en
h
a
n
ce
d
t
h
e
ac
cu
r
ac
y
an
d
r
o
b
u
s
tn
ess
o
f
v
ar
io
u
s
task
s
,
in
clu
d
in
g
o
b
ject
d
etec
tio
n
an
d
d
is
tan
ce
m
ea
s
u
r
em
en
t,
wh
ich
ar
e
ess
en
tial
f
o
r
ap
p
licatio
n
s
r
eq
u
ir
in
g
p
r
ec
is
e
s
p
atial
awa
r
en
ess
[
3
]
.
I
n
p
ar
ticu
lar
,
t
h
e
in
te
g
r
atio
n
o
f
d
ee
p
lear
n
in
g
m
o
d
els
with
o
p
tim
izatio
n
tech
n
iq
u
es
h
as
c
o
n
tr
ib
u
ted
to
b
etter
p
er
f
o
r
m
an
ce
in
r
ea
l
-
wo
r
ld
e
n
v
ir
o
n
m
en
ts
,
m
ak
in
g
th
em
m
o
r
e
r
eliab
le
f
o
r
cr
itical
ap
p
licatio
n
s
s
u
ch
as
au
to
n
o
m
o
u
s
n
av
i
g
atio
n
an
d
as
s
is
tiv
e
tech
n
o
lo
g
ies.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
6
7
-
3
2
7
8
3268
Fo
r
in
d
iv
id
u
als
with
v
is
u
al
im
p
air
m
en
ts
,
th
ese
ca
p
a
b
ilit
ies
ar
e
tr
an
s
f
o
r
m
ativ
e
.
Acc
u
r
ate
o
b
ject
d
etec
tio
n
an
d
d
is
tan
ce
esti
m
atio
n
ca
n
g
r
ea
tly
en
h
an
ce
m
o
b
ilit
y
an
d
in
d
ep
e
n
d
en
ce
,
en
a
b
lin
g
s
af
e
n
av
ig
atio
n
an
d
f
o
s
ter
in
g
a
u
to
n
o
m
y
[
4
]
–
[
6
]
.
B
y
lev
er
ag
in
g
s
tate
-
of
-
th
e
-
ar
t
d
ee
p
lear
n
in
g
m
o
d
els,
co
m
p
u
ter
v
is
io
n
ca
n
p
r
o
v
id
e
in
n
o
v
ativ
e
s
o
lu
tio
n
s
t
o
ad
d
r
ess
th
ese
u
n
iq
u
e
ch
allen
g
es
[
7
]
–
[
9
]
.
On
e
o
f
th
e
lead
i
n
g
m
eth
o
d
s
f
o
r
o
b
ject
d
etec
tio
n
is
th
e
y
o
u
o
n
ly
l
o
o
k
o
n
ce
(
YOL
O)
f
r
am
ewo
r
k
,
r
en
o
wn
ed
f
o
r
its
r
ea
l
-
tim
e
p
er
f
o
r
m
an
ce
a
n
d
h
ig
h
ac
cu
r
ac
y
.
I
ts
latest
iter
atio
n
,
YO
L
Ov
8
[
1
0
]
,
[
1
1
]
,
d
em
o
n
s
tr
ates
s
u
b
s
tan
tial
im
p
r
o
v
em
en
ts
in
d
iv
er
s
e
ap
p
licatio
n
s
.
W
h
en
in
teg
r
ated
with
tr
ad
itio
n
al
co
m
p
u
ter
v
is
io
n
tech
n
iq
u
es
lik
e
th
o
s
e
in
Op
en
C
V,
YOL
Ov
8
ca
n
p
o
ten
tially
en
h
an
ce
o
b
ject
d
etec
tio
n
an
d
d
is
tan
c
e
esti
m
atio
n
f
o
r
ass
is
tiv
e
tech
n
o
lo
g
ies
[
1
2
]
.
Op
e
n
C
V,
as
an
o
p
en
-
s
o
u
r
ce
lib
r
a
r
y
,
o
f
f
er
s
p
o
we
r
f
u
l
to
o
ls
f
o
r
im
a
g
e
p
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
ex
tr
a
ctio
n
,
an
d
g
eo
m
etr
ic
tr
an
s
f
o
r
m
atio
n
s
,
m
ak
in
g
it
a
v
alu
ab
le
c
o
m
p
lem
en
t
to
d
ee
p
lear
n
in
g
m
o
d
els
[
1
3
]
–
[
1
5
]
.
M
ea
n
wh
ile,
in
n
o
v
atio
n
s
lik
e
co
o
r
d
in
ate
atten
tio
n
weig
h
tin
g
(
C
AW
)
f
u
r
th
er
r
ef
in
e
d
is
tan
ce
esti
m
atio
n
b
y
in
c
o
r
p
o
r
atin
g
s
p
atial
atten
tio
n
m
ec
h
an
is
m
s
.
T
h
is
s
tu
d
y
ex
p
lo
r
es
h
o
w
th
ese
m
eth
o
d
s
ca
n
b
e
tailo
r
ed
to
ass
is
tiv
e
tech
n
o
l
o
g
ies.
Desp
ite
th
ese
ad
v
an
ce
m
e
n
ts
,
ac
h
iev
in
g
p
r
ec
is
e
d
is
tan
ce
p
e
r
ce
p
tio
n
r
em
ain
s
ch
allen
g
in
g
[
1
6
]
–
[
1
8
]
.
T
r
ad
itio
n
al
m
eth
o
d
s
u
s
in
g
Op
en
C
V
o
f
ten
r
ely
o
n
b
asic
g
eo
m
etr
ic
ca
lcu
latio
n
s
,
wh
ich
m
a
y
f
alter
in
co
m
p
lex
en
v
ir
o
n
m
en
ts
.
I
n
co
n
tr
ast,
m
e
th
o
d
s
lik
e
C
AW
lev
er
ag
e
s
p
a
tial
atten
tio
n
m
ec
h
an
is
m
s
to
i
m
p
r
o
v
e
ac
cu
r
ac
y
.
Ho
wev
er
,
its
ap
p
licatio
n
in
ass
is
tiv
e
tech
n
o
lo
g
y
f
o
r
th
e
b
lin
d
h
as n
o
t
b
ee
n
th
o
r
o
u
g
h
ly
r
esear
ch
ed
[
1
1
]
–
[
2
0
]
.
Ad
d
itio
n
ally
,
a
r
elate
d
s
tu
d
y
ex
p
l
o
r
es
ad
v
an
ce
m
en
ts
i
n
v
id
eo
co
m
p
r
ess
io
n
b
y
lev
er
ag
in
g
a
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN)
to
ac
ce
ler
ate
th
e
p
ar
titi
o
n
in
g
o
f
c
o
d
in
g
u
n
it
(
C
U)
b
lo
ck
s
in
h
ig
h
ef
f
icien
cy
v
id
eo
co
d
i
n
g
(
HE
VC
)
v
id
eo
e
n
co
d
in
g
[
2
1
]
.
T
h
is
ap
p
r
o
ac
h
n
o
t
o
n
ly
en
h
an
ce
s
co
m
p
u
tatio
n
al
ef
f
icien
cy
b
u
t
also
s
ig
n
if
ican
tly
r
ed
u
ce
s
h
a
r
d
war
e
r
eso
u
r
ce
co
n
s
u
m
p
tio
n
,
m
ak
in
g
it
well
-
s
u
ited
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
.
T
h
e
s
tu
d
y
d
em
o
n
s
tr
ates
th
at
s
u
ch
o
p
tim
izatio
n
tech
n
iq
u
es
ca
n
im
p
r
o
v
e
p
r
o
ce
s
s
in
g
s
p
ee
d
wh
ile
m
ain
tain
in
g
h
ig
h
v
id
eo
q
u
ality
,
wh
ich
is
p
ar
tic
u
lar
ly
b
en
ef
icial
f
o
r
r
eso
u
r
ce
-
co
n
s
tr
ain
ed
en
v
ir
o
n
m
en
ts
.
T
h
ese
ef
f
icien
cy
g
ain
s
ca
n
b
e
tr
an
s
lated
to
ap
p
licati
o
n
s
b
ey
o
n
d
v
id
e
o
co
m
p
r
ess
io
n
,
s
u
ch
as
ass
is
tiv
e
n
av
ig
atio
n
s
y
s
tem
s
f
o
r
th
e
v
is
u
ally
im
p
air
ed
,
wh
e
r
e
r
ea
l
-
tim
e
s
p
atial
d
ata
an
aly
s
is
is
cr
u
cial.
B
y
in
teg
r
atin
g
s
im
ilar
d
ee
p
lear
n
in
g
-
b
ased
o
p
tim
izatio
n
s
,
ass
is
tiv
e
tech
n
o
lo
g
ies
ca
n
ac
h
iev
e
f
aster
a
n
d
m
o
r
e
ac
c
u
r
ate
o
b
ject
d
et
ec
tio
n
an
d
d
is
tan
ce
esti
m
atio
n
,
en
h
an
cin
g
o
v
e
r
all
s
y
s
tem
p
er
f
o
r
m
a
n
ce
.
A
r
elate
d
s
tu
d
y
p
r
esen
ts
a
d
ata
-
d
r
iv
en
a
p
p
r
o
ac
h
th
at
c
o
m
b
in
es
o
b
ject
d
etec
tio
n
,
tr
ac
k
in
g
,
d
is
tan
ce
esti
m
atio
n
d
etec
tio
n
,
an
d
s
ize
m
ea
s
u
r
em
en
t
u
s
in
g
a
s
ter
eo
v
is
io
n
s
y
s
tem
[
2
2
]
.
B
y
co
m
b
i
n
in
g
YOL
Ov
8
f
o
r
o
b
ject
d
etec
tio
n
with
a
m
u
lti
-
lay
er
p
e
r
ce
p
tr
o
n
(
ML
P)
to
m
o
d
el
th
e
r
elatio
n
s
h
ip
s
b
etwe
en
d
is
tan
ce
,
s
ize,
an
d
d
is
p
ar
ity
,
th
e
alg
o
r
ith
m
ac
h
i
ev
es
u
p
to
9
9
.
9
9
%
ac
cu
r
ac
y
in
d
is
tan
ce
esti
m
atio
n
f
o
r
b
o
th
ca
lib
r
ated
an
d
u
n
ca
lib
r
ated
ca
m
e
r
a
co
n
f
i
g
u
r
a
tio
n
s
.
T
h
ese
r
esu
lts
d
em
o
n
s
tr
ate
th
e
p
o
ten
tial
o
f
ap
p
ly
in
g
s
i
m
ilar
tech
n
iq
u
es
to
im
p
r
o
v
e
ass
is
tiv
e
tech
n
o
lo
g
ies b
lin
d
p
e
o
p
le.
A
s
ep
ar
ate
s
tu
d
y
in
v
esti
g
ates
th
e
u
s
e
o
f
a
h
y
b
r
id
m
o
d
el
c
o
m
b
in
in
g
a
p
r
e
-
tr
ain
e
d
d
u
al
-
p
a
th
r
ec
u
r
r
e
n
t
n
eu
r
al
n
etwo
r
k
(
DPR
NN
)
with
a
tr
an
s
f
o
r
m
er
f
o
r
au
d
io
s
o
u
r
ce
s
ep
ar
atio
n
[
2
3
]
.
T
h
is
m
o
d
el
h
an
d
les
lo
n
g
s
eq
u
en
ce
s
m
o
r
e
ef
f
ec
ti
v
ely
b
y
p
ar
titi
o
n
in
g
in
p
u
t
d
ata
an
d
u
s
i
n
g
th
e
tr
an
s
f
o
r
m
er
'
s
ab
ilit
y
to
u
n
d
er
s
tan
d
c
o
n
tex
t.
T
h
e
ap
p
r
o
ac
h
lead
s
to
im
p
r
o
v
ed
s
ig
n
al
-
to
-
n
o
is
e
r
atio
s
an
d
s
ep
ar
atio
n
q
u
ality
,
wh
ic
h
co
u
ld
in
s
p
ir
e
s
im
ilar
s
tr
ateg
ies
in
p
r
o
ce
s
s
in
g
lo
n
g
-
r
an
g
e
s
p
atial
d
ata
f
o
r
n
av
i
g
atio
n
s
y
s
tem
s
b
lin
d
p
eo
p
le.
T
h
is
s
tu
d
y
aim
s
to
co
n
d
u
ct
a
co
m
p
r
eh
en
s
iv
e
co
m
p
ar
ativ
e
an
aly
s
is
o
f
YOL
Ov
8
in
teg
r
ated
with
Op
en
C
V
an
d
YOL
Ov
8
en
h
an
ce
d
with
C
AW
to
ev
alu
ate
th
eir
ef
f
ec
tiv
en
ess
in
o
b
ject
d
etec
tio
n
an
d
d
is
tan
ce
esti
m
atio
n
.
B
y
ass
ess
in
g
k
ey
p
er
f
o
r
m
an
ce
m
etr
ics
s
u
ch
as
ac
cu
r
ac
y
,
p
r
o
ce
s
s
in
g
s
p
ee
d
,
a
n
d
r
eliab
ilit
y
,
th
is
r
esear
ch
s
ee
k
s
to
d
eter
m
in
e
t
h
e
ad
v
an
tag
es
an
d
lim
itatio
n
s
o
f
ea
ch
ap
p
r
o
ac
h
i
n
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
.
T
h
e
f
in
d
in
g
s
f
r
o
m
t
h
is
s
tu
d
y
ar
e
e
x
p
ec
ted
to
p
r
o
v
id
e
v
alu
ab
le
i
n
s
ig
h
ts
f
o
r
th
e
d
ev
elo
p
m
en
t
o
f
ad
v
a
n
ce
d
ass
is
ti
v
e
tech
n
o
lo
g
ies,
p
ar
ticu
lar
ly
f
o
r
v
is
u
ally
im
p
air
e
d
in
d
iv
id
u
a
ls
wh
o
r
ely
o
n
p
r
ec
is
e
s
p
at
ial
awa
r
en
ess
f
o
r
n
av
ig
atio
n
.
Fu
r
th
er
m
o
r
e,
b
y
an
aly
zin
g
th
e
co
m
p
u
tatio
n
al
ef
f
icien
cy
an
d
ad
ap
tab
ilit
y
o
f
th
ese
m
o
d
els,
th
is
r
esear
ch
aim
s
to
co
n
tr
i
b
u
te
to
th
e
b
r
o
a
d
er
f
ield
o
f
in
tellig
en
t
v
is
io
n
s
y
s
tem
s
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
r
esear
ch
m
eth
o
d
o
lo
g
y
f
o
llo
ws
a
s
y
s
tem
atic
ap
p
r
o
ac
h
to
ev
alu
atin
g
an
d
co
m
p
ar
in
g
th
e
p
er
f
o
r
m
an
ce
o
f
YOL
Ov
8
in
te
g
r
ated
with
Op
en
C
V
a
n
d
YOL
Ov
8
e
n
h
an
ce
d
with
C
AW
f
o
r
d
is
tan
ce
p
er
ce
p
tio
n
in
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
.
T
h
is
p
r
o
ce
s
s
b
eg
in
s
with
d
ataset
s
elec
t
io
n
an
d
p
r
ep
r
o
ce
s
s
in
g
,
en
s
u
r
in
g
th
at
th
e
d
ata
u
s
ed
f
o
r
tr
ain
in
g
a
n
d
test
in
g
ac
cu
r
ately
r
ep
r
esen
t
r
ea
l
-
wo
r
ld
s
ce
n
ar
io
s
.
T
h
e
ex
p
er
im
en
tal
s
etu
p
in
clu
d
es
co
n
f
ig
u
r
in
g
b
o
t
h
m
o
d
els,
d
e
f
in
in
g
t
h
eir
h
y
p
er
p
ar
am
eter
s
,
a
n
d
im
p
lem
e
n
tin
g
t
h
em
in
a
c
o
n
t
r
o
lled
en
v
ir
o
n
m
en
t
to
ass
es
s
th
eir
ef
f
ec
tiv
en
ess
.
A
d
d
itio
n
ally
,
ev
alu
atio
n
m
etr
ics
s
u
ch
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
in
f
er
en
c
e
tim
e
ar
e
u
tili
ze
d
to
m
ea
s
u
r
e
th
e
s
tr
en
g
th
s
an
d
lim
itatio
n
s
o
f
ea
ch
ap
p
r
o
ac
h
.
Fin
ally
,
s
tatis
tical
an
aly
s
i
s
is
co
n
d
u
cte
d
to
v
alid
ate
th
e
s
ig
n
if
ican
ce
o
f
t
h
e
r
esu
lts
,
p
r
o
v
id
in
g
d
ee
p
e
r
in
s
ig
h
ts
in
to
th
e
ir
ap
p
licab
ilit
y
f
o
r
ass
is
tiv
e
tech
n
o
lo
g
ies.
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
-
8
7
0
8
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f YOLOv8
tech
n
iq
u
es:
Op
e
n
C
V
a
n
d
co
o
r
d
in
a
te
a
tten
tio
n
…
(
E
m
a
Uta
mi
)
3269
2
.
1
.
Da
t
a
c
o
llect
io
n
A
co
m
p
r
eh
en
s
iv
e
d
ataset
o
f
i
m
ag
es
an
d
d
is
tan
ce
v
id
eo
s
co
v
er
in
g
a
v
ar
iety
o
f
in
d
o
o
r
an
d
o
u
td
o
o
r
en
v
ir
o
n
m
en
ts
will
b
e
u
s
ed
in
T
ab
le
1
.
T
h
e
d
ataset
s
h
o
u
ld
h
av
e
a
v
ar
iety
o
f
o
b
jects,
d
is
tan
ce
s
,
lig
h
tin
g
co
n
d
itio
n
s
,
an
d
o
cc
lu
s
io
n
s
to
e
n
s
u
r
e
th
e
r
eliab
ilit
y
an
d
g
en
e
r
a
lizab
ilit
y
o
f
th
e
r
esu
lts
.
Pu
b
licly
av
ailab
le
d
atasets
s
u
ch
as
co
m
m
o
n
o
b
jects
in
c
o
n
tex
t
(
C
OC
O)
,
an
d
cu
s
to
m
d
ata
s
ets
b
e
u
s
ed
.
I
n
ad
d
itio
n
,
a
s
p
e
cial
d
ataset
tailo
r
ed
to
r
ea
l
-
wo
r
ld
s
ce
n
a
r
io
s
f
ac
ed
b
y
in
d
i
v
id
u
als with
v
is
u
al
im
p
air
m
en
ts
will b
e
cr
ea
ted
[
2
3
]
.
T
ab
le
1
.
Data
s
ets
No
D
a
t
a
s
e
t
A
t
t
r
i
b
u
t
e
s
1.
C
O
C
O
B
o
u
n
d
i
n
g
b
o
x
,
o
b
j
e
c
t
c
a
t
e
g
o
r
y
,
se
g
m
e
n
t
a
t
i
o
n
2.
C
u
s
t
o
m
d
a
t
a
se
t
B
o
u
n
d
i
n
g
b
o
x
,
o
b
j
e
c
t
c
a
t
e
g
o
r
y
,
g
r
o
u
n
d
t
r
u
t
h
d
i
st
a
n
c
e
2
.
2
.
E
x
perim
ent
a
l set
up
T
h
e
ex
p
er
im
en
tal
d
esig
n
f
o
r
t
h
is
s
tu
d
y
f
o
llo
ws
a
th
o
r
o
u
g
h
a
n
d
s
tr
u
ctu
r
ed
m
eth
o
d
o
lo
g
y
to
ass
es
s
an
d
co
m
p
ar
e
th
e
p
er
f
o
r
m
an
ce
o
f
YOL
Ov
8
co
m
b
in
e
d
with
Op
en
C
V
an
d
YOL
Ov
8
au
g
m
e
n
t
ed
with
C
AW
f
o
r
d
is
tan
ce
esti
m
atio
n
in
b
lin
d
n
a
v
ig
atio
n
s
y
s
tem
s
.
T
h
e
f
o
llo
wi
n
g
p
ar
a
g
r
ap
h
s
p
r
o
v
id
e
a
d
etailed
d
escr
ip
tio
n
o
f
t
h
e
s
tep
s
in
v
o
lv
ed
in
t
h
e
ex
p
e
r
im
e
n
tal
s
etu
p
.
T
h
e
ex
p
er
im
e
n
tal
s
etu
p
f
o
r
th
is
s
tu
d
y
in
v
o
lv
es
a
co
m
p
r
eh
e
n
s
iv
e
ev
alu
atio
n
o
f
YOL
Ov
8
in
teg
r
ated
with
Op
en
C
V
an
d
YOL
Ov
8
en
h
an
ce
d
with
C
AW
f
o
r
d
is
tan
ce
p
er
ce
p
tio
n
in
a
b
lin
d
n
a
v
ig
atio
n
s
y
s
tem
as
d
ep
icted
in
Fig
u
r
e
1
.
I
n
itially
,
a
d
iv
er
s
e
d
ataset
o
f
im
ag
es
an
d
v
id
eo
s
f
r
o
m
v
ar
io
u
s
i
n
d
o
o
r
an
d
o
u
td
o
o
r
en
v
ir
o
n
m
en
ts
was
co
llected
an
d
an
n
o
tated
with
g
r
o
u
n
d
tr
u
th
b
o
u
n
d
in
g
b
o
x
es
an
d
d
is
tan
ce
s
.
T
h
e
d
atasets
in
clu
d
ed
p
u
b
licly
av
ailab
le
s
o
u
r
ce
s
s
u
ch
as
C
OC
O
an
d
KI
T
T
I
,
as
well
as
cu
s
to
m
d
ata
tail
o
r
ed
f
o
r
th
is
s
tu
d
y
[
2
4
]
.
Af
ter
p
r
ep
a
r
in
g
t
h
e
d
atas
et,
wh
ich
in
v
o
lv
e
d
o
r
g
an
izin
g
th
e
d
ata
in
t
o
tr
ain
in
g
,
v
alid
atio
n
,
a
n
d
test
in
g
s
ets,
th
e
YOL
Ov
8
m
o
d
el
was im
p
l
em
en
ted
u
s
in
g
O
p
en
C
V
an
d
C
AW
f
o
r
d
is
tan
ce
esti
m
atio
n
.
O
b
ject
d
etec
tio
n
was
p
er
f
o
r
m
ed
o
n
th
e
d
ataset,
an
d
d
is
tan
ce
s
wer
e
esti
m
ated
u
s
in
g
tr
ad
itio
n
al
g
eo
m
etr
ic
ca
lcu
lati
o
n
s
in
th
e
Op
en
C
V
s
etu
p
an
d
a
n
en
h
an
ce
d
s
p
atial
atten
tio
n
m
ec
h
a
n
is
m
in
th
e
C
AW
s
etu
p
.
Per
f
o
r
m
an
ce
was e
v
alu
ated
in
te
r
m
s
o
f
ac
cu
r
ac
y
(
m
ea
n
ab
s
o
lu
te
e
r
r
o
r
(
MA
E
)
,
r
o
o
t
m
ea
n
s
q
u
a
r
ed
er
r
o
r
(
R
MSE
)
)
a
n
d
s
p
ee
d
(
f
r
am
e
s
p
er
s
ec
o
n
d
(
FP
S),
in
f
er
en
ce
tim
e)
[
2
5
]
.
C
o
m
p
ar
ativ
e
s
tu
d
y
was
co
n
d
u
cted
to
ev
alu
ate
th
e
s
tatis
t
ical
s
ig
n
if
ican
ce
o
f
th
e
p
er
f
o
r
m
an
ce
d
if
f
e
r
en
ce
s
b
etw
ee
n
th
e
two
a
p
p
r
o
ac
h
es.
B
ased
o
n
th
e
r
esu
lts
o
f
th
is
s
tu
d
y
,
th
is
r
esear
ch
o
f
f
er
s
s
u
g
g
esti
o
n
s
f
o
r
im
p
r
o
v
i
n
g
ass
is
tiv
e
tech
n
o
lo
g
y
f
o
r
p
eo
p
le
with
v
is
u
al
im
p
air
m
en
ts
.
Fig
u
r
e
1
.
Pro
ce
s
s
ex
p
e
r
im
en
ta
l setu
p
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
6
7
-
3
2
7
8
3270
2
.
3
.
I
m
ple
m
ent
a
t
io
n det
a
ils
T
h
e
i
m
p
l
e
m
e
n
ta
t
i
o
n
o
f
t
h
e
e
x
p
e
r
i
m
e
n
t
a
l
s
et
u
p
i
n
v
o
l
v
e
s
t
w
o
m
a
i
n
c
o
n
f
i
g
u
r
e
r
a
ti
o
n
s
:
Y
O
L
O
v
8
i
n
t
e
g
r
a
t
e
d
w
i
t
h
O
p
e
n
C
V
a
n
d
YO
L
O
v
8
en
h
a
n
c
e
d
w
i
t
h
C
AW
.
E
a
c
h
c
o
n
f
i
g
u
r
e
r
a
t
i
o
n
u
n
d
e
r
g
o
e
s
a
s
e
r
i
es
o
f
s
t
e
p
s
f
o
r
o
b
j
e
ct
d
e
t
e
c
t
i
o
n
a
n
d
d
is
t
a
n
c
e
es
ti
m
a
t
io
n
.
B
e
l
o
w
a
r
e
t
h
e
d
e
t
a
il
e
d
s
t
e
p
s
a
n
d
p
r
o
c
e
s
s
e
s
i
n
v
o
l
v
e
d
i
n
t
h
e
i
m
p
l
e
m
e
n
t
a
ti
o
n
.
Fig
u
r
e
2
s
h
o
ws
th
e
u
s
e
o
f
th
e
YOL
Ov
8
m
o
d
el
to
id
en
tif
y
o
b
j
ec
ts
in
im
ag
es,
with
th
e
m
o
d
el
g
en
er
atin
g
b
o
u
n
d
in
g
b
o
x
es
an
d
class
lab
els
f
o
r
ea
ch
d
etec
ted
o
b
ject.
Pre
-
tr
ain
ed
weig
h
ts
ar
e
u
s
ed
in
itially
,
f
o
llo
wed
b
y
f
in
e
-
tu
n
in
g
o
n
a
c
u
s
to
m
d
atase
t
to
im
p
r
o
v
e
d
et
ec
tio
n
ac
c
u
r
ac
y
f
o
r
s
p
ec
if
ic
s
ce
n
ar
io
s
f
ac
ed
b
y
v
is
u
ally
im
p
ai
r
ed
in
d
iv
id
u
als.
Geo
m
etr
ic
ca
lcu
l
atio
n
s
ar
e
a
p
p
lied
t
o
esti
m
ate
th
e
d
is
tan
ce
o
f
d
etec
ted
o
b
je
cts
u
s
in
g
Op
en
C
V
f
u
n
ctio
n
s
.
T
ec
h
n
iq
u
es
s
u
ch
a
s
m
o
n
o
cu
lar
v
is
io
n
-
b
ased
d
e
p
th
esti
m
atio
n
o
r
s
ter
eo
v
is
i
o
n
ar
e
u
s
ed
f
o
r
m
o
r
e
ac
cu
r
ate
r
esu
lts
.
Fig
u
r
e
2
.
YOL
Ov
8
with
Op
en
C
V
Fig
u
r
e
3
s
h
o
ws
th
at
th
e
in
teg
r
atio
n
o
f
th
e
YOL
Ov
8
m
o
d
el
in
v
o
lv
es
in
clu
d
in
g
a
C
AW
lay
er
th
at
en
h
an
ce
s
s
p
atial
atten
tio
n
,
i
m
p
r
o
v
i
n
g
th
e
m
o
d
el'
s
ab
ilit
y
to
f
o
c
u
s
o
n
r
ele
v
an
t
f
ea
t
u
r
es.
T
h
e
r
ef
i
n
ed
YOL
Ov
8
+CAW
m
o
d
el
is
tr
ain
ed
o
n
a
c
u
s
to
m
d
ataset
to
o
p
tim
ize
d
et
ec
tio
n
p
er
f
o
r
m
a
n
ce
.
T
h
e
o
u
tp
u
t
o
f
th
e
YOL
Ov
8
+CAW
m
o
d
el
i
s
th
e
n
u
tili
ze
d
f
o
r
m
o
r
e
ac
cu
r
ate
d
is
tan
ce
esti
m
atio
n
.
B
y
in
co
r
p
o
r
atin
g
th
e
C
AW
m
ec
h
an
is
m
,
th
e
m
o
d
el
ac
h
iev
es
b
etter
s
p
atial
u
n
d
er
s
tan
d
in
g
,
lead
in
g
to
m
o
r
e
p
r
ec
is
e
d
is
tan
ce
ca
lcu
latio
n
s
.
W
h
ile
s
im
ilar
g
eo
m
etr
y
an
d
v
i
s
io
n
-
b
ased
tech
n
iq
u
es
as
in
th
e
Op
en
C
V
ap
p
r
o
ac
h
ar
e
e
m
p
l
o
y
ed
,
th
e
e
n
h
an
ce
d
atten
tio
n
to
o
b
ject
f
ea
tu
r
es r
es
u
lts
in
s
u
p
er
io
r
ac
c
u
r
ac
y
.
Fig
u
r
e
3
.
YOL
Ov
8
with
C
AW
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
r
esu
lts
o
f
th
is
s
tu
d
y
ar
e
p
r
esen
ted
in
two
m
ain
s
ec
tio
n
s
:
e
v
alu
atio
n
m
etr
ics
an
d
s
tatis
tica
l
an
aly
s
is
,
ea
ch
o
f
f
er
in
g
a
co
m
p
r
e
h
en
s
iv
e
co
m
p
ar
is
o
n
o
f
YOL
Ov
8
in
teg
r
ated
with
Op
en
C
V
an
d
YOL
Ov
8
en
h
an
ce
d
with
C
AW
f
o
r
d
is
tan
ce
p
er
ce
p
tio
n
in
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
.
T
h
e
ev
alu
atio
n
m
etr
ics
s
ec
tio
n
ex
am
in
es
k
ey
p
er
f
o
r
m
an
ce
in
d
icato
r
s
s
u
ch
a
s
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
in
f
e
r
en
ce
tim
e
t
o
d
eter
m
i
n
e
th
e
ef
f
icien
c
y
o
f
ea
ch
ap
p
r
o
ac
h
.
Ad
d
itio
n
ally
,
q
u
alitativ
e
o
b
s
er
v
atio
n
s
ar
e
in
clu
d
ed
to
h
ig
h
lig
h
t
th
e
r
ea
l
-
w
o
r
ld
ap
p
licab
ilit
y
o
f
b
o
th
m
et
h
o
d
s
.
T
h
e
s
tatis
tical
an
aly
s
is
s
ec
tio
n
f
u
r
th
er
v
alid
at
es
th
e
f
in
d
in
g
s
b
y
ap
p
ly
i
n
g
a
p
p
r
o
p
r
iate
s
tatis
tica
l
test
s
to
ass
e
s
s
th
e
s
ig
n
if
ican
ce
o
f
p
er
f
o
r
m
an
ce
d
if
f
e
r
en
ce
s
.
T
h
ese
r
esu
lts
p
r
o
v
id
e
v
alu
ab
le
in
s
ig
h
ts
in
to
th
e
s
tr
en
g
th
s
an
d
l
im
itatio
n
s
o
f
e
ac
h
ap
p
r
o
ac
h
,
co
n
tr
i
b
u
tin
g
t
o
th
e
ad
v
a
n
ce
m
en
t
o
f
ass
is
tiv
e
tech
n
o
lo
g
ies
f
o
r
v
is
u
ally
im
p
air
ed
in
d
iv
id
u
als.
3
.
1
.
E
v
a
lua
t
i
o
n m
et
rics
I
n
ev
alu
atin
g
th
e
p
er
f
o
r
m
a
n
c
e
o
f
YOL
Ov
8
in
teg
r
ated
wit
h
Op
en
C
V
an
d
YOL
Ov
8
e
n
h
an
ce
d
with
C
AW
,
s
ev
er
al
k
ey
m
etr
ics
wer
e
co
n
s
id
er
e
d
:
MA
E
,
R
MSE
,
FP
S,
an
d
in
f
er
en
ce
tim
e.
T
h
e
ev
alu
atio
n
in
cl
u
d
es
MA
E
,
R
MSE
,
FP
S,
an
d
in
f
er
e
n
ce
tim
e.
Par
am
eter
s
u
s
ed
f
o
r
th
e
ex
p
er
i
m
e
n
ts
in
clu
d
e:
a.
Data
s
et:
C
OC
O
(
cu
s
to
m
d
atas
ets)
b.
I
n
p
u
t r
eso
lu
tio
n
: 6
4
0
×6
4
0
p
i
x
els.
c.
Har
d
war
e:
NVI
DI
A
GeFo
r
ce
R
T
X
3
0
9
0
.
d.
B
atch
s
ize:
1
6
.
e.
L
ea
r
n
in
g
r
ate:
0
.
0
0
1
.
f.
T
r
ain
in
g
e
p
o
ch
s
: 1
0
.
T
h
ese
p
ar
am
eter
s
wer
e
o
p
tim
ized
to
b
alan
ce
m
o
d
el
p
er
f
o
r
m
an
ce
an
d
c
o
m
p
u
tatio
n
al
ef
f
ici
en
cy
.
3
.
2
.
Scena
rio
-
ba
s
ed
t
esting
Ad
d
itio
n
al
s
ce
n
ar
io
s
wer
e
in
tr
o
d
u
ce
d
t
o
ev
alu
ate
th
e
r
o
b
u
s
tn
ess
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
els.
T
h
ese
in
clu
d
ed
lo
w
lig
h
t,
p
ar
tial
o
cc
l
u
s
io
n
,
h
i
g
h
o
b
ject
d
en
s
ity
,
an
d
co
m
p
lex
b
ac
k
g
r
o
u
n
d
en
v
i
r
o
n
m
en
ts
.
R
esu
lts
s
h
o
w
th
at
YOL
Ov
8
+
C
AW
d
em
o
n
s
tr
ated
co
n
s
is
ten
t a
cc
u
r
ac
y
im
p
r
o
v
em
en
ts
ac
r
o
s
s
all
s
ce
n
ar
io
s
.
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
-
8
7
0
8
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f YOLOv8
tech
n
iq
u
es:
Op
e
n
C
V
a
n
d
co
o
r
d
in
a
te
a
tten
tio
n
…
(
E
m
a
Uta
mi
)
3271
T
ab
le
2
in
teg
r
ate
d
with
Op
en
C
V
an
d
YOL
Ov
8
en
h
an
ce
d
with
C
AW
u
n
d
er
d
if
f
er
en
t
s
c
en
ar
io
s
.
T
h
e
r
esu
lts
in
d
icate
th
at
YOL
Ov
8
+
C
AW
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
s
YOL
Ov
8
+
Op
en
C
V
in
ter
m
s
o
f
ac
cu
r
ac
y
,
ac
h
iev
in
g
lo
wer
MA
E
a
n
d
R
MSE
v
alu
es
ac
r
o
s
s
all
test
ed
co
n
d
itio
n
s
.
Sp
ec
if
ically
,
th
e
C
AW
m
ec
h
an
is
m
im
p
r
o
v
es
f
ea
tu
r
e
atten
tio
n
,
en
ab
lin
g
th
e
m
o
d
el
to
m
o
r
e
p
r
e
cisely
esti
m
ate
d
is
tan
ce
s
ev
en
u
n
d
er
ch
allen
g
i
n
g
co
n
d
itio
n
s
s
u
ch
as
lo
w
lig
h
t,
o
cc
lu
s
io
n
,
an
d
h
ig
h
o
b
ject
d
en
s
ity
.
Fo
r
in
s
tan
ce
,
in
lo
w
-
lig
h
t
s
ce
n
ar
io
s
,
th
e
MA
E
f
o
r
YOL
Ov
8
+
C
AW
is
0
.
4
0
m
eter
s
co
m
p
ar
e
d
to
0
.
6
0
m
eter
s
f
o
r
YOL
Ov
8
+
Op
e
n
C
V,
h
ig
h
lig
h
tin
g
th
e
r
o
b
u
s
tn
ess
o
f
C
AW
in
s
ce
n
ar
io
s
wh
er
e
v
is
ib
ilit
y
is
r
ed
u
ce
d
.
T
h
is
im
p
r
o
v
e
d
ac
cu
r
ac
y
d
e
m
o
n
s
tr
ates
th
e
p
o
ten
tial
o
f
C
AW
to
en
h
an
ce
d
is
tan
ce
p
er
ce
p
tio
n
r
eliab
ilit
y
in
r
ea
l
-
wo
r
ld
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
.
a.
MA
E
=
1
∑
|
−
̂
|
=
1
(
1
)
w
h
er
e
is
th
e
ac
tu
al
d
is
tan
ce
,
̂
is
th
e
esti
m
ated
d
is
tan
ce
,
an
d
is
th
e
n
u
m
b
e
r
o
f
o
b
s
er
v
atio
n
s
[
2
6
]
.
b.
R
MSE
R
MSE
i
s
co
m
p
u
ted
b
y
tak
in
g
t
h
e
s
q
u
ar
e
r
o
o
t
o
f
th
e
av
er
ag
e
o
f
th
e
s
q
u
ar
ed
d
if
f
er
e
n
ce
s
b
etwe
en
th
e
o
b
s
er
v
ed
v
alu
es
,
an
d
p
r
ed
icted
v
alu
es
̂
[
2
7
]
.
=
√
1
∑
=
1
(
−
̂
)
2
(
2
)
c.
FPS
=
(
)
(
3
)
d.
I
n
f
er
en
ce
tim
e
(
)
=
(
)
(
4
)
FP
S
an
d
in
f
er
en
ce
tim
e
ar
e
cr
itical
f
o
r
r
ea
l
-
tim
e
a
p
p
licatio
n
s
[
2
8
]
as
d
escr
ib
e
d
in
T
ab
le
3
.
Alth
o
u
g
h
YOL
Ov
8
with
Op
e
n
C
V
h
as
h
ig
h
er
FP
S
an
d
lo
wer
in
f
er
e
n
c
e
tim
e,
th
e
d
if
f
e
r
en
ce
is
n
o
t
s
i
g
n
if
ican
t,
a
n
d
b
o
th
co
n
f
ig
u
r
atio
n
s
ar
e
ca
p
a
b
le
o
f
d
eliv
er
in
g
r
ea
l
-
tim
e
p
e
r
f
o
r
m
an
ce
.
T
h
e
s
lig
h
t
s
p
ee
d
u
p
f
o
r
YOL
Ov
8
+
C
AW
i
s
o
f
f
s
et
b
y
its
in
cr
ea
s
ed
ac
c
u
r
ac
y
.
T
ab
le
2
.
MA
E
an
d
R
MSE
u
n
d
er
d
if
f
er
e
n
t scen
ar
io
s
C
o
n
d
i
t
i
o
n
Y
O
LO
v
8
+
O
p
e
n
C
V
M
A
E
Y
O
LO
v
8
+
C
A
W
M
A
E
Y
O
LO
v
8
+
O
p
e
n
C
V
R
M
S
E
Y
O
LO
v
8
+
C
A
W
R
M
S
E
Lo
w
l
i
g
h
t
0
.
6
0
0
.
4
0
0
.
7
2
0
.
5
0
O
c
c
l
u
d
e
d
o
b
j
e
c
t
0
.
5
5
0
.
3
8
0
.
6
7
0
.
4
6
H
i
g
h
o
b
j
e
c
t
d
e
n
s
i
t
y
0
.
6
5
0
.
4
5
0
.
7
6
0
.
5
2
C
o
m
p
l
e
x
b
a
c
k
g
r
o
u
n
d
0
.
6
8
0
.
4
8
0
.
8
0
0
.
5
5
T
ab
le
3
.
Sp
ee
d
d
etec
tio
n
M
e
t
r
i
c
Y
O
LO
v
8
+
O
p
e
n
C
V
Y
O
LO
v
8
+
C
A
W
FPS
25
20
I
n
f
e
r
e
n
c
e
t
i
me
(
ms)
40
50
3
.
3
.
E
nh
a
nced
YO
L
O
v
8
wit
h O
penCV
inte
g
ra
t
io
n f
o
r
im
pro
v
ed
dis
t
a
nce
det
ec
t
io
n
T
o
im
p
r
o
v
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
YOL
Ov
8
f
o
r
d
is
tan
ce
d
etec
tio
n
u
s
in
g
Op
e
n
C
V
an
d
b
r
in
g
it
clo
s
er
to
th
e
p
er
f
o
r
m
a
n
ce
o
f
C
AW
,
s
e
v
er
al
s
tr
ateg
ies
ca
n
b
e
im
p
le
m
en
ted
to
en
h
an
ce
b
o
th
ac
cu
r
ac
y
an
d
d
etec
tio
n
ca
p
ab
ilit
ies.
T
h
ese
ap
p
r
o
ac
h
e
s
m
ay
in
clu
d
e
o
p
tim
izin
g
th
e
m
o
d
el'
s
ar
ch
itectu
r
e
b
y
in
co
r
p
o
r
atin
g
ad
d
itio
n
al
lay
er
s
o
r
m
o
d
if
y
in
g
e
x
is
tin
g
o
n
es
to
im
p
r
o
v
e
f
ea
tu
r
e
ex
tr
ac
ti
o
n
.
Fu
r
th
er
m
o
r
e,
in
teg
r
atin
g
a
d
v
an
ce
d
tech
n
iq
u
es
s
u
ch
as
m
u
lti
-
s
ca
le
tr
ain
in
g
,
d
ata
au
g
m
en
tatio
n
,
o
r
p
o
s
t
-
p
r
o
c
ess
in
g
m
eth
o
d
s
lik
e
Kalm
an
f
i
lter
s
ca
n
h
elp
r
ef
in
e
th
e
m
o
d
el'
s
ab
ilit
y
to
p
r
e
d
ic
t
d
is
tan
ce
s
m
o
r
e
ac
cu
r
ately
.
Ad
d
itio
n
ally
,
f
in
e
-
tu
n
in
g
h
y
p
er
p
ar
am
eter
s
an
d
lev
er
ag
in
g
tr
a
n
s
f
er
lear
n
in
g
f
r
o
m
p
r
etr
ain
e
d
m
o
d
els
ca
n
h
elp
b
o
o
s
t
p
er
f
o
r
m
an
ce
,
en
a
b
l
in
g
YOL
Ov
8
with
Op
en
C
V
to
ap
p
r
o
ac
h
th
e
r
o
b
u
s
tn
ess
o
f
C
A
W
in
r
ea
l
-
tim
e
ap
p
licatio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
6
7
-
3
2
7
8
3272
T
h
e
im
p
r
o
v
em
e
n
ts
to
YOL
O
v
8
with
Op
en
C
V
f
o
c
u
s
o
n
en
h
an
cin
g
d
is
tan
ce
d
etec
tio
n
ac
cu
r
ac
y
,
p
r
o
v
id
i
n
g
a
s
tr
o
n
g
alter
n
ativ
e
t
o
C
AW
with
o
u
t
th
e
n
ee
d
f
o
r
e
x
tr
a
s
en
s
o
r
s
,
as
s
h
o
wn
in
Fig
u
r
e
4
.
T
h
is
in
teg
r
atio
n
in
clu
d
es
b
etter
ca
m
er
a
ca
lib
r
a
tio
n
,
a
s
ter
eo
ca
m
er
a
s
etu
p
f
o
r
d
ep
th
p
er
ce
p
tio
n
,
ad
v
a
n
ce
d
d
ep
th
esti
m
atio
n
f
o
r
g
en
er
atin
g
d
e
p
th
m
ap
s
,
a
n
d
co
m
b
in
in
g
th
e
m
with
p
o
in
t
clo
u
d
d
ata
f
o
r
p
r
ec
is
e
d
is
tan
ce
m
ea
s
u
r
em
e
n
ts
.
Ad
d
itio
n
al
tech
n
iq
u
es
s
u
ch
as
p
o
s
t
-
p
r
o
ce
s
s
in
g
,
f
ilter
in
g
,
an
d
f
in
e
-
tu
n
in
g
YOL
Ov
8
with
an
n
o
tated
d
is
tan
c
e
d
ata
f
u
r
th
er
im
p
r
o
v
e
ac
cu
r
ac
y
.
Usi
n
g
Op
en
C
V,
a
p
o
p
u
la
r
o
p
e
n
-
s
o
u
r
ce
lib
r
ar
y
k
n
o
wn
f
o
r
r
ea
l
-
tim
e
im
ag
e
p
r
o
ce
s
s
in
g
an
d
d
ep
th
esti
m
atio
n
,
en
s
u
r
es
h
ig
h
p
er
f
o
r
m
an
ce
wh
ile
b
ein
g
co
s
t
-
ef
f
ec
tiv
e
f
o
r
d
ev
elo
p
in
g
ass
is
tiv
e
tech
n
o
lo
g
ies f
o
r
th
e
v
is
u
ally
im
p
air
ed
.
Fig
u
r
e
4
.
YOL
Ov
8
with
Op
en
C
V
f
o
r
d
is
tan
ce
d
etec
tio
n
3
.
4
.
Co
m
pa
ra
t
iv
e
re
s
ults
T
h
is
co
m
p
ar
ativ
e
an
aly
s
is
aim
s
to
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
YOL
Ov
8
in
teg
r
ated
wit
h
Op
en
C
V
v
er
s
u
s
YOL
Ov
8
en
h
an
ce
d
with
C
AW
,
f
o
cu
s
in
g
s
p
ec
if
ically
o
n
th
eir
a
b
ilit
y
to
p
er
ce
i
v
e
d
is
tan
ce
s
in
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
.
T
h
e
s
tu
d
y
co
m
p
ar
es
k
ey
p
e
r
f
o
r
m
an
ce
m
e
tr
ics
s
u
ch
as
ac
cu
r
ac
y
,
r
ea
l
-
tim
e
in
f
er
en
ce
s
p
ee
d
,
an
d
r
o
b
u
s
tn
ess
to
en
v
ir
o
n
m
en
t
al
v
ar
iab
les,
wh
ich
ar
e
cr
u
cial
f
o
r
th
e
ef
f
ec
tiv
en
ess
o
f
ass
is
ti
v
e
tech
n
o
lo
g
ies.
B
y
th
o
r
o
u
g
h
ly
an
aly
zi
n
g
b
o
th
m
o
d
els
in
ter
m
s
o
f
th
eir
ab
ilit
y
to
d
etec
t
o
b
jects
an
d
esti
m
ate
d
is
tan
ce
s
with
p
r
ec
is
io
n
,
th
is
r
esear
ch
s
ee
k
s
to
id
en
tify
th
e
s
tr
en
g
th
s
an
d
w
ea
k
n
ess
es
o
f
ea
ch
a
p
p
r
o
ac
h
in
p
r
o
v
id
in
g
r
eliab
le
s
p
atial
awa
r
en
ess
f
o
r
v
is
u
ally
im
p
air
ed
in
d
i
v
id
u
als.
I
n
T
ab
le
4
,
MA
E
p
r
o
v
id
es a
co
m
p
r
eh
e
n
s
iv
e
v
iew
o
f
th
e
av
e
r
ag
e
ab
s
o
lu
te
er
r
o
r
in
d
is
tan
ce
esti
m
atio
n
ac
r
o
s
s
v
ar
io
u
s
d
is
tan
ce
r
an
g
e
s
f
o
r
th
r
ee
d
is
tin
ct
m
o
d
els:
YOL
Ov
8
+
Op
en
C
V
o
r
ig
in
al
,
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
,
an
d
YOL
Ov
8
+
C
A
W
.
T
h
e
f
in
d
i
n
g
s
u
n
eq
u
iv
o
ca
lly
d
em
o
n
s
tr
ate
th
at
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
s
b
o
t
h
YOL
Ov
8
+
Op
en
C
V
o
r
ig
in
a
l
an
d
YOL
Ov
8
+
C
AW
,
s
h
o
w
ca
s
in
g
s
ig
n
if
ican
tly
lo
wer
MA
E
v
alu
es
ac
r
o
s
s
a
ll
s
p
ec
if
ied
d
is
tan
ce
ca
teg
o
r
ies.
T
h
is
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
u
n
d
er
s
co
r
es
its
ca
p
ab
ilit
y
in
ac
h
iev
in
g
m
o
r
e
ac
cu
r
ate
d
is
tan
ce
esti
m
atio
n
s
,
wh
ich
is
cr
u
cial
f
o
r
ap
p
licatio
n
s
lik
e
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
wh
er
e
p
r
e
cisi
o
n
is
p
ar
am
o
u
n
t.
T
ab
le
4
.
C
o
m
p
a
r
ativ
e
r
esu
lt
MA
E
M
o
d
e
l
M
A
E
(
0
-
1
m)
M
A
E
(1
-
2
m)
M
A
E
(2
-
3
m)
M
A
E
(3
-
4
m)
M
A
E
(4
-
5
m)
Y
O
LO
v
8
+
O
p
e
n
C
V
o
r
i
g
i
n
a
l
0
.
4
2
0
.
4
8
0
.
5
6
0
.
6
1
0
.
7
3
Y
O
LO
v
8
+
O
p
e
n
C
V
i
m
p
r
o
v
e
d
0
.
3
6
0
.
4
1
0
.
4
9
0
.
5
5
0
.
6
2
Y
O
LO
v
8
+
C
A
W
0
.
3
9
0
.
4
5
0
.
5
2
0
.
5
8
0
.
6
7
Mo
r
eo
v
er
,
wh
ile
YOL
Ov
8
+
C
AW
ex
h
ib
its
n
o
tab
le
im
p
r
o
v
e
m
en
ts
o
v
er
YOL
Ov
8
+
Op
e
n
C
V
o
r
ig
in
al,
it f
alls
s
h
o
r
t o
f
m
atch
in
g
th
e
p
r
ec
is
io
n
ac
h
iev
ed
b
y
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
.
T
h
is
co
m
p
ar
is
o
n
h
i
g
h
lig
h
ts
th
e
p
iv
o
tal
r
o
le
o
f
a
d
v
an
ce
d
f
ea
tu
r
e
atten
tio
n
m
ec
h
a
n
is
m
s
s
u
ch
as
C
AW
in
e
n
h
an
cin
g
th
e
p
e
r
ce
p
tu
al
ca
p
ab
ilit
ies
o
f
co
m
p
u
ter
v
is
io
n
m
o
d
els,
p
ar
ticu
lar
ly
wh
e
n
in
teg
r
ated
with
r
o
b
u
s
t
f
r
am
ewo
r
k
s
lik
e
Op
en
C
V.
T
h
e
s
y
n
er
g
y
b
etwe
en
th
ese
tech
n
iq
u
es
n
o
t
o
n
ly
en
h
a
n
ce
s
ac
cu
r
ac
y
b
u
t
also
v
alid
ate
s
th
eir
ef
f
icac
y
i
n
r
ea
l
-
wo
r
ld
s
ce
n
a
r
io
s
,
wh
e
r
e
r
eliab
le
d
is
tan
c
e
esti
m
atio
n
is
v
ital
f
o
r
p
r
o
v
id
in
g
e
f
f
ec
tiv
e
n
av
ig
atio
n
al
aid
s
to
v
is
u
ally
im
p
air
ed
in
d
iv
id
u
als.
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
-
8
7
0
8
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f YOLOv8
tech
n
iq
u
es:
Op
e
n
C
V
a
n
d
co
o
r
d
in
a
te
a
tten
tio
n
…
(
E
m
a
Uta
mi
)
3273
T
ab
le
5
R
MSE
p
r
o
v
id
es
in
s
ig
h
t
in
to
th
e
m
ea
n
s
q
u
ar
e
d
er
r
o
r
i
n
d
is
tan
ce
esti
m
atio
n
b
y
th
e
th
r
ee
m
o
d
els
at
f
iv
e
d
if
f
er
e
n
t
d
is
tan
ce
r
an
g
es.
R
MSE
p
lace
s
g
r
ea
ter
em
p
h
asis
o
n
lar
g
er
er
r
o
r
s
,
s
o
h
ig
h
er
v
alu
es
in
d
icate
lar
g
er
er
r
o
r
v
ar
iatio
n
s
.
Fro
m
t
h
is
d
iag
r
am
,
we
ca
n
s
ee
th
at
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
ed
co
n
s
is
ten
tly
h
as
t
h
e
lo
west R
M
SE
ac
r
o
s
s
all
d
is
tan
ce
r
an
g
es,
in
d
icatin
g
t
h
at
th
is
m
o
d
el
h
as th
e
b
est p
er
f
o
r
m
an
ce
in
r
ed
u
cin
g
lar
g
e
er
r
o
r
v
ar
iatio
n
s
.
YOL
Ov
8
+
C
AW
p
er
f
o
r
m
s
b
etter
th
a
n
Y
OL
Ov
8
+
Op
en
C
V
o
r
ig
i
n
al,
b
u
t
s
till
h
ig
h
e
r
th
a
n
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
ed
.
T
h
is
s
u
g
g
ests
th
at
im
p
r
o
v
e
m
en
ts
in
O
p
en
C
V
ca
n
p
r
o
d
u
ce
m
o
r
e
s
tab
le
a
n
d
r
eliab
le
d
is
tan
ce
esti
m
atio
n
,
r
ed
u
cin
g
s
ig
n
if
ica
n
t
er
r
o
r
s
th
at
ca
n
af
f
ec
t
th
e
o
v
er
all
p
er
f
o
r
m
an
ce
o
f
th
e
b
lin
d
n
av
ig
atio
n
s
y
s
tem
.
T
ab
le
5
.
C
o
m
p
a
r
ativ
e
R
MSE
M
o
d
e
l
R
M
S
E
(
0
-
1
m)
R
M
S
E
(1
-
2
m)
R
M
S
E
(2
-
3
m)
R
M
S
E
(3
-
4
m)
R
M
S
E
(4
-
5
m)
Y
O
LO
v
8
+
O
p
e
n
C
V
o
r
i
g
i
n
a
l
0
.
5
2
0
.
5
7
0
.
6
4
0
.
6
9
0
.
7
8
Y
O
LO
v
8
+
O
p
e
n
C
V
i
m
p
r
o
v
e
d
0
.
4
4
0
.
5
0
0
.
5
8
0
.
6
4
0
.
7
1
Y
O
LO
v
8
+
C
A
W
0
.
4
8
0
.
5
3
0
.
6
0
0
.
6
6
0
.
7
4
3
.
5
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
wit
h pre
v
io
us
re
s
ea
rc
hers
A
d
etailed
co
m
p
ar
is
o
n
was
co
n
d
u
cte
d
b
y
an
aly
zin
g
th
e
r
e
s
u
lts
f
r
o
m
s
im
ilar
s
tu
d
ies
to
p
r
o
v
id
e
a
co
m
p
r
eh
e
n
s
iv
e
u
n
d
er
s
tan
d
in
g
o
f
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
s
.
I
n
th
is
c
o
m
p
ar
is
o
n
,
v
ar
io
u
s
s
tate
-
of
-
th
e
-
ar
t
ap
p
r
o
ac
h
es
we
r
e
co
n
s
id
er
ed
to
h
i
g
h
lig
h
t
th
e
s
tr
en
g
th
s
an
d
lim
itatio
n
s
o
f
th
e
cu
r
r
en
t
m
eth
o
d
s
.
T
ab
le
6
p
r
esen
ts
a
co
m
p
ar
ativ
e
an
aly
s
is
o
f
ac
cu
r
ac
y
d
etec
tio
n
,
s
h
o
wca
s
in
g
h
o
w
th
e
p
e
r
f
o
r
m
an
ce
o
f
YOL
Ov
8
in
teg
r
ated
with
Op
e
n
C
V
an
d
YOL
Ov
8
en
h
an
ce
d
with
C
AW
co
m
p
ar
es
to
o
th
e
r
lead
in
g
m
o
d
els
in
ter
m
s
o
f
de
tectio
n
ac
cu
r
ac
y
,
p
r
o
v
id
i
n
g
a
clea
r
er
p
ictu
r
e
o
f
th
ei
r
r
elativ
e
ef
f
ec
tiv
en
ess
in
r
ea
l
-
w
o
r
ld
ap
p
licatio
n
s
.
T
ab
le
6
.
C
o
m
p
a
r
ativ
e
ac
cu
r
ac
y
M
o
d
e
l
/
S
t
u
d
y
D
a
t
a
s
e
t
A
c
c
u
r
a
c
y
Y
O
LO
v
8
+
O
p
e
n
C
V
o
r
i
g
i
n
a
l
C
O
C
O
w
i
t
h
(
c
u
st
o
m
d
a
t
a
s
e
t
)
9
0
.
4
%
Y
O
LO
v
8
+
C
A
W
C
O
C
O
w
i
t
h
(
c
u
st
o
m
d
a
t
a
s
e
t
)
9
2
.
2
%
Y
O
LO
v
8
+
O
p
e
n
C
V
i
m
p
r
o
v
e
d
C
O
C
O
w
i
t
h
(
c
u
st
o
m
d
a
t
a
s
e
t
)
9
5
.
7
%
Ta
n
g
e
t
a
l
.
[
2
9
]
(
Y
O
LO
v
5
+
C
A
W
)
C
O
C
O
&
P
a
sc
a
l
8
9
.
8
%
S
u
r
e
s
h
e
t
a
l
.
[
3
0
]
(OFR
C
N
N
+
O
p
e
n
C
V
)
C
O
C
O
9
7
.
8
%
I
n
T
a
b
le
6
,
th
e
co
m
p
a
r
is
o
n
r
ev
ea
ls
t
h
at
OFR
C
NN
+
O
p
e
n
C
V
[
3
0
]
a
ch
ie
v
es
t
h
e
h
ig
h
est
a
cc
u
r
ac
y
o
f
9
7
.
8
%,
d
e
m
o
n
s
tr
ati
n
g
its
e
x
c
e
p
ti
o
n
al
p
e
r
f
o
r
m
a
n
c
e
in
t
h
e
c
o
n
tex
t
o
f
s
ta
tic
o
b
je
ct
d
ete
cti
o
n
.
T
h
is
h
i
g
h
ac
c
u
r
ac
y
is
i
n
d
i
ca
ti
v
e
o
f
t
h
e
m
o
d
el'
s
e
f
f
e
cti
v
e
n
ess
w
h
en
w
o
r
k
i
n
g
w
it
h
well
-
d
e
f
i
n
e
d
,
u
n
c
h
a
n
g
i
n
g
e
n
v
i
r
o
n
m
e
n
ts
.
H
o
we
v
e
r
,
it
is
i
m
p
o
r
ta
n
t
to
n
o
t
e
t
h
at
th
is
s
t
u
d
y
is
p
r
i
m
a
r
i
ly
f
o
c
u
s
e
d
o
n
o
f
f
l
in
e
d
et
ec
t
io
n
,
m
ea
n
i
n
g
t
h
at
c
o
m
p
u
tati
o
n
ti
m
e
an
d
r
ea
l
-
tim
e
p
e
r
f
o
r
m
an
ce
co
n
s
t
r
ai
n
ts
w
e
r
e
n
o
t
as
c
r
it
ic
al
i
n
t
h
e
e
v
al
u
at
io
n
.
T
h
e
r
e
f
o
r
e,
w
h
il
e
t
h
e
a
cc
u
r
ac
y
is
im
p
r
ess
i
v
e
,
t
h
e
m
o
d
el'
s
a
p
p
li
c
ab
i
l
it
y
t
o
r
ea
l
-
ti
m
e
s
y
s
te
m
s
,
s
u
c
h
as
t
h
o
s
e
u
s
e
d
in
ass
is
ti
v
e
n
a
v
i
g
ati
o
n
f
o
r
t
h
e
v
is
u
al
ly
i
m
p
ai
r
e
d
,
m
a
y
b
e
li
m
i
ted
b
y
i
ts
p
r
o
ce
s
s
i
n
g
s
p
ee
d
a
n
d
r
eso
u
r
c
e
d
e
m
a
n
d
s
i
n
d
y
n
am
ic
en
v
i
r
o
n
m
e
n
ts
.
I
n
co
n
tr
ast,
th
e
m
o
d
els
ev
alu
at
ed
in
th
is
s
tu
d
y
,
s
u
ch
as
Y
OL
O
v
8
in
teg
r
ated
with
C
AW
,
ar
e
s
p
ec
if
ically
o
p
tim
ized
f
o
r
r
ea
l
-
tim
e
o
b
ject
d
etec
tio
n
task
s
,
wh
er
e
m
ain
t
ain
in
g
a
b
alan
ce
b
etwe
en
ac
c
u
r
ac
y
an
d
in
f
er
e
n
ce
s
p
ee
d
is
cr
itical.
T
h
ese
m
o
d
els
ar
e
d
esig
n
ed
to
p
r
o
ce
s
s
d
ata
q
u
ick
ly
,
wh
ich
is
ess
en
tial
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
lik
e
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
,
wh
er
e
d
elay
s
ca
n
af
f
ec
t
th
e
s
y
s
tem
's
ef
f
ec
tiv
en
ess
.
Ho
wev
er
,
th
e
an
aly
s
is
also
h
ig
h
lig
h
ts
th
at
th
e
tim
e
ef
f
icien
cy
o
f
th
e
n
ewly
p
r
o
p
o
s
ed
alg
o
r
ith
m
s
,
s
u
ch
as
YOL
Ov
8
+
C
AW
an
d
YOL
Ov
8
+
Op
en
C
V,
h
as
n
o
t
b
ee
n
th
o
r
o
u
g
h
ly
s
tu
d
ied
in
ter
m
s
o
f
in
f
er
e
n
ce
tim
e
a
n
d
th
ei
r
r
e
al
-
tim
e
ap
p
licab
ilit
y
in
d
y
n
am
ic
en
v
ir
o
n
m
en
ts
.
Un
d
er
s
tan
d
in
g
th
ese
f
ac
to
r
s
is
ess
en
tial,
as
d
y
n
am
ic
en
v
ir
o
n
m
en
ts
o
f
ten
p
r
esen
t
ad
d
itio
n
al
ch
allen
g
es
th
at
r
eq
u
ir
e
b
o
th
f
ast
p
r
o
ce
s
s
in
g
an
d
h
ig
h
ac
cu
r
ac
y
f
o
r
r
eliab
le
d
is
t
an
ce
esti
m
atio
n
an
d
o
b
ject
d
etec
tio
n
.
a.
YOL
Ov
8
+
Op
en
C
V
o
r
ig
in
al
with
9
0
.
4
%
ac
cu
r
ac
y
p
r
o
v
i
d
e
s
a
s
o
lid
f
o
u
n
d
atio
n
f
o
r
r
ea
l
-
t
im
e
ap
p
licatio
n
s
,
ac
h
iev
in
g
a
r
ea
s
o
n
ab
le
tr
a
d
e
-
o
f
f
b
etwe
en
s
p
ee
d
an
d
ac
c
u
r
ac
y
.
b.
YOL
Ov
8
+
C
AW
,
with
an
ac
cu
r
ac
y
o
f
9
2
.
2
%,
s
h
o
ws a
n
o
tab
le
im
p
r
o
v
em
en
t in
s
p
atial
atten
tio
n
,
lead
in
g
to
m
o
r
e
p
r
ec
is
e
d
is
tan
ce
esti
m
atio
n
,
m
ak
in
g
it
a
b
etter
f
it
f
o
r
d
y
n
am
ic
e
n
v
ir
o
n
m
en
ts
,
s
u
ch
as
th
o
s
e
en
co
u
n
ter
e
d
in
b
lin
d
n
a
v
ig
atio
n
s
y
s
tem
s
.
c.
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
,
ac
h
iev
in
g
an
ac
c
u
r
ac
y
o
f
9
5
.
7
%,
r
ep
r
esen
ts
th
e
h
ig
h
est
-
p
er
f
o
r
m
in
g
r
ea
l
-
tim
e
m
o
d
el
in
th
is
co
m
p
ar
is
o
n
.
T
h
is
im
p
r
o
v
em
e
n
t
h
ig
h
lig
h
ts
th
e
ef
f
ec
tiv
e
n
ess
o
f
th
e
en
h
an
ce
m
e
n
ts
m
ad
e
to
th
e
o
r
ig
in
al
Op
en
C
V
in
teg
r
atio
n
,
p
ar
ticu
lar
l
y
in
r
ef
i
n
in
g
th
e
d
etec
tio
n
p
i
p
elin
e
an
d
im
p
r
o
v
in
g
th
e
co
n
s
is
ten
cy
o
f
p
r
ed
ictio
n
s
u
n
d
er
v
ar
y
in
g
c
o
n
d
itio
n
s
.
T
h
e
h
ig
h
er
ac
cu
r
ac
y
m
a
k
es
it
well
-
s
u
i
ted
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
r
eq
u
i
r
in
g
h
ig
h
p
r
e
cisi
o
n
,
s
u
ch
as
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
,
with
o
u
t
co
m
p
r
o
m
is
in
g
th
e
m
o
d
el'
s
in
f
er
en
ce
s
p
ee
d
o
r
s
u
itab
ilit
y
f
o
r
r
eso
u
r
ce
-
co
n
s
tr
ain
e
d
en
v
i
r
o
n
m
en
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
6
7
-
3
2
7
8
3274
I
n
r
ea
l
-
tim
e
a
p
p
licatio
n
s
,
th
e
g
o
al
is
to
b
alan
ce
s
p
ee
d
an
d
ac
cu
r
ac
y
.
W
h
ile
OFR
C
NN
+
Op
en
C
V
ac
h
iev
es
th
e
h
ig
h
est
ac
cu
r
ac
y
,
its
r
elian
ce
o
n
o
f
f
lin
e
p
r
o
ce
s
s
in
g
m
ay
lim
it
its
ap
p
licab
ilit
y
in
d
y
n
am
ic,
r
ea
l
-
wo
r
ld
en
v
ir
o
n
m
e
n
ts
.
Mo
d
els
lik
e
YOL
Ov
8
+
Op
en
C
V
i
m
p
r
o
v
e
d
an
d
YOL
Ov
8
+
C
AW
p
r
o
v
id
e
t
h
e
o
p
tim
al
b
alan
ce
f
o
r
r
ea
l
-
tim
e
u
s
e,
o
f
f
er
in
g
b
o
t
h
h
ig
h
ac
cu
r
ac
y
a
n
d
r
esp
o
n
s
iv
en
ess
ess
en
tial
f
o
r
task
s
lik
e
b
lin
d
n
av
ig
atio
n
.
I
n
Fig
u
r
e
5
d
escr
ib
es th
e
v
a
r
ia
tio
n
o
f
th
e
m
ea
n
s
q
u
ar
e
er
r
o
r
i
n
d
is
tan
ce
esti
m
ates b
y
th
e
th
r
ee
m
o
d
els
o
v
er
d
if
f
er
en
t
d
is
tan
ce
r
an
g
es.
T
h
is
d
iag
r
am
s
h
o
ws
th
at
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
c
o
n
s
is
ten
tly
h
as
th
e
lo
west
R
MSE
v
alu
es
ac
r
o
s
s
all
d
is
tan
ce
r
an
g
es,
in
d
icatin
g
th
at
it
p
er
f
o
r
m
s
th
e
b
est
in
ter
m
s
o
f
r
ed
u
cin
g
s
ig
n
if
ican
t
er
r
o
r
s
in
d
is
tan
ce
e
s
tim
atio
n
.
T
h
is
im
p
lies
th
at
th
e
im
p
r
o
v
ed
Op
e
n
C
V
in
teg
r
ati
o
n
p
r
o
v
id
es
a
m
o
r
e
s
tab
le
an
d
r
eliab
le
d
is
tan
ce
est
im
atio
n
,
m
in
im
izin
g
lar
g
e
d
is
c
r
ep
an
cies
m
o
r
e
ef
f
ec
tiv
ely
th
a
n
t
h
e
o
th
er
m
o
d
els.
YOL
Ov
8
+
C
AW
s
h
o
ws
b
etter
p
er
f
o
r
m
a
n
ce
th
an
YOL
Ov
8
+
Op
en
C
V
o
r
ig
in
al
b
u
t
f
alls
s
h
o
r
t
co
m
p
ar
e
d
to
th
e
im
p
r
o
v
e
d
Op
en
C
V
v
er
s
io
n
,
h
i
g
h
lig
h
tin
g
th
e
ef
f
icac
y
o
f
t
h
e
Op
en
C
V
en
h
an
ce
m
en
ts
.
Fig
u
r
e
5
.
R
o
o
t m
ea
n
s
q
u
ar
e
er
r
o
r
co
m
p
ar
is
o
n
I
n
Fig
u
r
e
6
s
h
o
ws
th
e
p
er
f
o
r
m
an
ce
o
f
th
r
ee
m
o
d
els:
YOL
Ov
8
+O
p
en
C
V
o
r
ig
in
al,
YOL
Ov
8
+O
p
en
C
V
im
p
r
o
v
e
d
,
a
n
d
YOL
Ov
8
+CAW
in
d
is
tan
ce
esti
m
atio
n
at
v
a
r
io
u
s
d
is
tan
ce
r
a
n
g
es (
0
-
1
m
,
1
-
2
m
,
2
-
3
m
,
3
-
4
m
,
4
-
5
m
)
.
Fro
m
th
is
d
iag
r
am
,
it
is
clea
r
th
at
YOL
Ov
8
+O
p
en
C
V
im
p
r
o
v
ed
c
o
n
s
is
ten
tly
ac
h
iev
es
th
e
lo
west
MA
E
v
alu
es
ac
r
o
s
s
all
d
is
tan
ce
r
an
g
es,
in
d
icatin
g
th
at
it
is
th
e
m
o
s
t
ac
cu
r
ate
m
o
d
el
in
ter
m
s
o
f
m
in
im
izin
g
th
e
ab
s
o
lu
te
er
r
o
r
in
d
is
tan
ce
esti
m
atio
n
.
YOL
Ov
8
+CA
W
p
er
f
o
r
m
s
b
etter
t
h
an
YOL
Ov
8
+O
p
en
C
V
o
r
ig
in
al,
b
u
t
n
o
t
as
w
ell
as
YOL
Ov
8
+O
p
en
C
V
im
p
r
o
v
e
d
,
s
u
g
g
esti
n
g
th
at
th
e
im
p
r
o
v
em
en
ts
in
Op
en
C
V
s
ig
n
if
ican
tly
en
h
an
ce
th
e
ac
cu
r
ac
y
o
f
d
is
tan
ce
esti
m
atio
n
.
Fig
u
r
e
6
.
Me
an
ab
s
o
lu
te
er
r
o
r
co
m
p
ar
is
o
n
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
-
8
7
0
8
C
o
mp
a
r
a
tive
a
n
a
lysi
s
o
f YOLOv8
tech
n
iq
u
es:
Op
e
n
C
V
a
n
d
co
o
r
d
in
a
te
a
tten
tio
n
…
(
E
m
a
Uta
mi
)
3275
I
n
Fig
u
r
e
7
s
h
o
ws
th
e
r
elatio
n
s
h
ip
b
etwe
en
p
r
o
ce
s
s
in
g
s
p
ee
d
(
FP
S)
an
d
in
f
e
r
en
ce
tim
e
f
o
r
th
r
ee
m
o
d
els:
YOL
Ov
8
+
Op
en
C
V
o
r
ig
in
al,
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
,
an
d
YOL
Ov
8
+
C
AW
.
Fro
m
th
is
d
iag
r
am
,
it
ca
n
b
e
s
ee
n
th
at
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
co
n
s
is
ten
tly
ac
h
iev
es
h
ig
h
er
FP
S
an
d
lo
wer
in
f
er
en
ce
tim
es
ac
r
o
s
s
all
d
is
tan
ce
r
an
g
es,
m
a
k
in
g
it
th
e
m
o
s
t
ef
f
icien
t
m
o
d
el.
T
h
is
m
o
d
el
s
h
o
ws
s
ig
n
if
ica
n
t
im
p
r
o
v
em
e
n
ts
in
s
p
ee
d
with
o
u
t
s
ac
r
if
icin
g
ac
cu
r
ac
y
,
wh
ich
is
cr
itical
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
s
u
ch
as
n
av
ig
atio
n
s
y
s
tem
s
f
o
r
th
e
b
lin
d
.
YOL
Ov
8
+
C
AW
o
u
tp
er
f
o
r
m
s
YOL
Ov
8
+
Op
en
C
V
o
r
ig
in
al
b
u
t is st
ill in
f
er
io
r
to
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
.
T
h
is
s
h
o
ws
th
at
th
e
im
p
r
o
v
em
en
ts
in
O
p
en
C
V
p
r
o
v
id
e
s
u
b
s
tan
tial
b
en
ef
its
in
ter
m
s
o
f
ef
f
icien
cy
an
d
o
v
er
all
p
er
f
o
r
m
a
n
ce
.
Fig
u
r
e
8
co
m
p
ar
es
th
e
p
e
r
f
o
r
m
an
ce
o
f
th
r
ee
m
o
d
els:
YOL
Ov
8
+O
p
en
C
V
o
r
ig
in
al,
YOL
Ov
8
+O
p
en
C
V
im
p
r
o
v
e
d
,
an
d
YOL
Ov
8
+CAW
at
f
iv
e
d
is
tan
ce
r
an
g
es.
Fro
m
th
is
h
ea
tm
ap
,
it
ca
n
b
e
s
ee
n
th
at
YOL
Ov
8
+O
p
en
C
V
im
p
r
o
v
ed
co
n
s
is
ten
tly
h
as
th
e
lo
wes
t
M
SE
v
alu
e
at
all
d
i
s
tan
ce
r
an
g
es,
in
d
icatin
g
th
at
th
is
m
o
d
el
is
th
e
m
o
s
t
ac
cu
r
ate
in
m
in
im
izin
g
th
e
esti
m
atio
n
er
r
o
r
.
L
ig
h
ter
c
o
lo
r
s
in
th
e
h
ea
t
m
ap
in
d
icate
s
m
aller
er
r
o
r
s
,
w
h
ile
d
ar
k
er
co
lo
r
s
in
d
icate
lar
g
er
er
r
o
r
s
.
YOL
Ov
8
+CAW
p
er
f
o
r
m
s
b
etter
th
an
YOL
Ov
8
+O
p
en
C
V
o
r
ig
in
al
b
u
t
s
till
l
ag
s
b
eh
i
n
d
YOL
Ov
8
+O
p
en
C
V
im
p
r
o
v
ed
,
h
ig
h
lig
h
tin
g
th
e
b
en
ef
its
o
f
in
cr
ea
s
in
g
Op
en
C
V
in
teg
r
atio
n
in
im
p
r
o
v
in
g
d
is
tan
ce
esti
m
atio
n
ac
cu
r
ac
y
.
Fig
u
r
e
7
.
Scatter
p
l
o
t f
o
r
FP
S v
s
.
in
f
er
en
ce
tim
e
I
n
Fig
u
r
e
8
co
m
p
a
r
es
th
e
p
er
f
o
r
m
an
ce
o
f
th
r
ee
m
o
d
els:
YOL
Ov
8
+
Op
en
C
V
o
r
ig
in
al,
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
ed
,
an
d
YOL
Ov
8
+
C
AW
a
t
f
iv
e
d
is
tan
ce
r
an
g
es.
Fro
m
th
is
h
ea
tm
ap
,
i
t
ca
n
b
e
s
ee
n
th
at
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
co
n
s
is
ten
tly
h
as
th
e
lo
west
MSE
v
alu
e
at
all
d
is
tan
ce
r
an
g
es,
in
d
icatin
g
th
at
th
is
m
o
d
el
is
th
e
m
o
s
t
ac
cu
r
a
te
in
m
in
im
izin
g
th
e
esti
m
atio
n
er
r
o
r
.
L
ig
h
ter
co
l
o
r
s
in
th
e
h
ea
tm
ap
in
d
icate
s
m
aller
er
r
o
r
s
,
wh
ile
d
ar
k
e
r
c
o
lo
r
s
in
d
icate
lar
g
er
e
r
r
o
r
s
.
Y
OL
Ov
8
+
C
AW
p
er
f
o
r
m
s
b
etter
th
an
YOL
Ov
8
+
Op
en
C
V
o
r
ig
i
n
al
b
u
t
s
till
lag
s
b
eh
in
d
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
,
h
i
g
h
lig
h
tin
g
th
e
b
e
n
ef
its
o
f
in
c
r
ea
s
in
g
Op
en
C
V
in
teg
r
atio
n
in
im
p
r
o
v
in
g
d
is
tan
ce
esti
m
atio
n
ac
cu
r
ac
y
.
Fig
u
r
e
8
.
Hea
tm
ap
f
o
r
er
r
o
r
an
aly
s
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
6
7
-
3
2
7
8
3276
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
h
ig
h
lig
h
ts
t
h
e
s
ig
n
i
f
ican
t
p
o
te
n
tial
o
f
in
teg
r
atin
g
ad
v
an
ce
d
f
ea
tu
r
e
atten
tio
n
m
e
ch
an
is
m
s
,
s
u
ch
as
C
AW
,
with
d
ee
p
lear
n
in
g
m
o
d
els
s
u
ch
as
YOL
Ov
8
f
o
r
d
is
tan
ce
esti
m
atio
n
in
b
lin
d
n
av
ig
atio
n
s
y
s
tem
s
.
T
h
e
s
tu
d
y
s
h
o
ws
th
at
wh
ile
YOL
Ov
8
co
m
b
in
e
d
with
Op
en
C
V
in
itially
y
ield
s
s
u
b
o
p
ti
m
al
r
esu
lts
,
f
u
r
th
er
r
ef
in
em
en
t
o
f
th
e
in
teg
r
atio
n
t
h
r
o
u
g
h
C
AW
an
d
Op
e
n
C
V
en
h
an
ce
m
en
ts
r
esu
lts
in
r
e
m
ar
k
a
b
le
im
p
r
o
v
em
e
n
ts
in
ac
cu
r
ac
y
.
YOL
Ov
8
+
Op
en
C
V
im
p
r
o
v
e
d
o
u
tp
er
f
o
r
m
s
th
e
o
r
ig
in
al
in
teg
r
atio
n
,
ap
p
r
o
ac
h
in
g
th
e
p
e
r
f
o
r
m
an
ce
o
f
YOL
Ov
8
+
C
AW
,
with
a
s
ig
n
if
ican
t
r
ed
u
ctio
n
in
MSE
.
T
h
i
s
s
tu
d
y
n
o
t
o
n
ly
im
p
r
o
v
es
th
e
a
cc
u
r
ac
y
o
f
d
is
tan
ce
p
er
ce
p
tio
n
b
u
t
also
r
ed
u
ce
s
th
e
r
elian
ce
o
n
ad
d
itio
n
al
s
en
s
o
r
s
,
an
d
o
f
f
er
s
co
s
t
s
av
in
g
s
.
T
h
e
s
e
ad
v
an
ce
s
p
r
esen
t
a
p
r
o
m
is
in
g
f
u
tu
r
e
f
o
r
im
p
r
o
v
in
g
ac
ce
s
s
ib
ilit
y
a
n
d
i
n
d
ep
e
n
d
en
ce
f
o
r
in
d
iv
id
u
als
with
v
is
u
al
im
p
air
m
en
ts
,
en
s
u
r
in
g
th
at
n
av
ig
atio
n
s
y
s
tem
s
ar
e
r
eliab
le
an
d
e
f
f
icien
t.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
s
wo
u
l
d
lik
e
to
a
ck
n
o
wled
g
e
v
alu
a
b
le
f
u
n
d
in
g
p
r
o
v
id
ed
b
y
Kem
en
ter
ian
P
en
d
id
ik
an
,
Keb
u
d
ay
aa
n
,
R
is
et,
d
an
T
ek
n
o
lo
g
i
in
Hib
ah
Pen
elitian
Pas
ca
s
ar
jan
a
-
Pen
elitian
T
esis
Ma
g
is
ter
o
f
t
h
e
f
is
ca
l
y
ea
r
2
0
2
4
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
i
s
r
es
e
a
r
c
h
wa
s
f
u
n
d
e
d
b
y
K
em
e
n
t
e
r
i
a
n
Pe
n
d
i
d
i
k
a
n
,
K
e
b
u
d
ay
a
a
n
,
R
is
e
t
,
d
a
n
T
e
k
n
o
l
o
g
i
t
h
r
o
u
g
h
H
i
b
a
h
P
e
n
e
l
i
ti
a
n
P
a
s
c
a
s
a
r
j
a
n
a
-
P
e
n
e
l
i
t
i
a
n
T
e
s
i
s
M
a
g
is
t
er
b
a
s
e
d
o
n
S
u
r
a
t
K
e
p
u
t
u
s
a
n
N
u
m
b
e
r
0
4
5
9
/
E
5
/
P
G
.
0
2
.
0
0
/
2
0
2
4
a
n
d
P
e
r
j
a
n
j
i
a
n
/K
o
n
tr
a
k
N
u
m
b
e
r
1
0
7
/
E
5
/
P
G
.
0
2
.
0
0
.
P
L
/
2
0
2
4
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
E
m
a
Utam
i
✓
✓
✓
✓
✓
✓
✓
✓
E
r
win
Sy
ah
r
u
d
in
✓
✓
✓
✓
✓
An
g
g
it Dw
i H
ar
tan
to
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
t
h
o
r
s
d
ec
lar
e
th
at
th
e
y
h
av
e
n
o
k
n
o
wn
c
o
m
p
etin
g
f
in
an
cial
in
ter
ests
o
r
p
er
s
o
n
al
r
el
atio
n
s
h
ip
s
th
at
co
u
ld
h
av
e
ap
p
ea
r
ed
t
o
in
f
lu
en
ce
th
e
wo
r
k
r
e
p
o
r
te
d
in
t
h
is
p
ap
er
.
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
ter
est.
I
NF
O
RM
E
D
CO
NS
E
N
T
T
h
is
s
tu
d
y
d
id
n
o
t
in
v
o
lv
e
h
u
m
an
p
ar
ticip
a
n
ts
o
r
th
e
c
o
llectio
n
o
f
p
er
s
o
n
al
d
ata;
th
er
ef
o
r
e
,
in
f
o
r
m
ed
co
n
s
en
t w
as n
o
t r
eq
u
ir
ed
.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
is
r
esear
ch
d
id
n
o
t
in
v
o
l
v
e
h
u
m
an
o
r
an
im
al
s
u
b
jects,
an
d
t
h
u
s
,
eth
ical
ap
p
r
o
v
al
was
n
o
t
ap
p
licab
le.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
i
n
g
s
o
f
th
is
s
tu
d
y
ar
e
o
p
en
ly
a
v
ailab
le
in
th
e
f
o
llo
win
g
r
ep
o
s
ito
r
ies
an
d
ca
n
b
e
ac
ce
s
s
ed
th
r
o
u
g
h
th
eir
r
esp
ec
tiv
e
DOI
lin
k
s
:
-
Do
i: 1
0
.
1
0
8
0
/0
2
5
6
4
6
0
2
.
2
0
2
0
.
1
8
1
9
8
9
3
-
Do
i: 1
0
.
2
9
2
0
7
/
r
esti
.
v
8
i2
.
5
5
2
9
Evaluation Warning : The document was created with Spire.PDF for Python.