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.
1
4
,
No
.
5
,
Octo
b
er
20
2
4
,
p
p
.
5
8
2
4
~
5
8
3
3
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
1
4
i
5
.
pp
5
8
2
4
-
5
8
3
3
5824
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Dete
c
tion o
f
e
lem
ents of
perso
na
l s
a
fet
y
f
o
r
the
prev
ention o
f
a
ccidents at
work
wit
h
conv
o
lutiona
l neural ne
twork
s
M
a
ria
Cla
ud
ia
B
o
nfa
nte
1
,
I
v
a
n H
er
na
nd
ez
Ruiz
1
,
J
ua
n C
o
ntr
er
a
s
M
o
nte
s
2
,
E
ug
enia
Arr
iet
a
Ro
dríg
uez
3
,
Alej
a
nd
ro
Ca
m
a
-
P
into
3
1
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
s
i
d
a
d
d
e
l
S
i
n
ú
,
C
a
r
t
a
g
e
n
a
,
C
o
l
o
m
b
i
a
2
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
Es
c
u
e
l
a
N
a
v
a
l
A
l
mi
r
a
n
t
e
P
a
d
i
l
l
a
,
C
a
r
t
a
g
e
n
a
,
C
o
l
o
mb
i
a
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
l
e
c
t
r
o
n
i
c
s,
U
n
i
v
e
r
si
d
a
d
d
e
l
a
C
o
s
t
a
,
B
a
r
r
a
n
q
u
i
l
l
a
,
C
o
l
o
mb
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
3
0
,
2
0
2
3
R
ev
is
ed
Ma
y
1
,
2
0
2
4
Acc
ep
ted
Ma
y
1
8
,
2
0
2
4
Th
e
tas
k
o
f
re
c
o
g
n
izin
g
p
e
rs
o
n
a
l
p
ro
tec
ti
v
e
e
lem
e
n
ts
i
n
wo
rk
p
lac
e
e
n
v
iro
n
m
e
n
ts
i
n
re
a
l
ti
m
e
is
f
u
n
d
a
m
e
n
tal
to
p
r
o
tec
ti
n
g
t
h
e
e
m
p
lo
y
e
e
s
in
c
a
se
o
f
a
n
y
a
c
c
id
e
n
ts.
T
h
is
c
a
n
b
e
a
c
h
iev
e
d
b
y
d
e
p
lo
y
in
g
a
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
(CNN
)
a
lg
o
rit
h
m
th
a
t
c
a
n
e
fficie
n
tl
y
d
e
tec
t
p
ro
tec
ti
v
e
e
lem
e
n
ts
th
ro
u
g
h
s
u
rv
e
il
la
n
c
e
d
e
v
ice
s.
Th
e
re
fo
re
,
t
h
is
w
o
rk
p
r
o
p
o
se
s
th
e
c
o
n
stru
c
ti
o
n
o
f
a
m
o
d
e
l,
imp
lem
e
n
ti
n
g
t
h
e
y
o
u
o
n
ly
lo
o
k
o
n
c
e
(YO
LO)
d
e
tec
t
o
r,
wh
o
se
a
rc
h
it
e
c
tu
re
h
a
s
b
e
e
n
o
n
e
o
f
t
h
e
m
o
st
tes
ted
a
c
c
o
r
d
in
g
to
li
tera
tu
r
e
re
v
iew
.
YO
LOv
5
a
n
d
YO
LOv
7
v
e
rsio
n
s we
re
u
se
d
a
n
d
a
d
a
tas
e
t
o
f
2
,
0
0
0
i
m
a
g
e
s fo
r
fo
u
r
c
las
se
s
c
o
n
si
d
e
re
d
.
T
h
is
d
a
tas
e
t
wa
s
c
o
ll
e
c
ti
o
n
fr
o
m
v
a
ri
o
u
s
so
u
rc
e
s
a
n
d
lab
e
ll
e
d
b
y
th
e
a
u
th
o
rs,
o
f
wh
ich
8
0
%
wa
s
u
se
d
fo
r
trai
n
in
g
,
1
5
%
fo
r
tes
ti
n
g
a
n
d
5
%
f
o
r
m
o
d
e
l
v
a
li
d
a
ti
o
n
.
T
h
e
m
o
st
imp
o
rtan
t
m
e
tri
c
s
a
re
p
re
se
n
ted
,
m
a
k
in
g
a
c
o
m
p
a
riso
n
b
e
twe
e
n
th
e
m
o
d
e
ls,
a
n
d
fin
a
l
ly
it
wa
s
id
e
n
ti
fie
d
t
h
a
t
YO
LOv
7
a
c
h
ie
v
e
d
a
h
ig
h
e
r
s
u
c
c
e
ss
ra
te,
wh
ich
c
o
u
ld
b
e
c
o
n
sid
e
re
d
a
m
o
re
c
o
m
p
lete
so
lu
ti
o
n
fo
r
o
c
c
u
p
a
ti
o
n
a
l
h
e
a
lt
h
a
n
d
sa
fe
ty
m
a
n
a
g
e
m
e
n
t
in
c
o
m
p
a
n
ies
.
K
ey
w
o
r
d
s
:
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
Dete
ctio
n
m
eth
o
d
s
Hea
lth
m
an
ag
em
en
t
Ob
ject
d
etec
tio
n
Saf
ety
m
an
ag
em
e
n
t
YOL
O
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
:
Ma
r
ia
C
lau
d
ia
B
o
n
f
an
te
Facu
lty
o
f
E
n
g
in
ee
r
in
g
,
Un
iv
e
r
s
id
ad
d
el
Sin
ú
B
o
s
q
u
e
Av
en
u
e,
T
r
an
s
v
er
s
e
5
4
No
.
3
0
-
4
5
3
,
C
ar
tag
en
a
1
3
0
0
0
1
,
C
o
lo
m
b
ia
E
m
ail:
m
ar
ia.
b
o
n
f
an
te@
u
n
is
in
u
.
ed
u
.
co
1.
I
NT
RO
D
UCT
I
O
N
T
h
o
s
e
r
esp
o
n
s
ib
le
f
o
r
h
ea
lth
an
d
s
af
ety
in
th
e
wo
r
k
p
lace
f
ac
e
th
e
ch
allen
g
e
o
f
o
b
s
er
v
in
g
,
r
ep
o
r
tin
g
an
d
ass
ess
in
g
th
e
h
az
ar
d
s
ex
i
s
tin
g
in
wo
r
k
p
lace
s
an
d
th
eir
im
p
ac
t
o
n
t
h
e
co
m
p
an
y
,
wh
ile
at
th
e
s
am
e
tim
e
p
r
o
m
p
tly
p
er
ce
iv
in
g
th
e
lack
o
f
u
s
e
o
f
p
er
s
o
n
al
p
r
o
tectiv
e
e
q
u
ip
m
en
t
(
PP
E
)
,
wh
ich
ca
n
h
a
v
e
r
ep
er
cu
s
s
io
n
s
in
th
e
ap
p
ea
r
a
n
ce
o
f
d
is
ea
s
es,
r
ed
u
ce
d
p
r
o
d
u
ctio
n
tim
e
o
r
u
n
f
o
r
tu
n
ate
in
ju
r
ies
in
th
e
wo
r
k
e
r
th
at
d
im
in
is
h
th
eir
q
u
ality
o
f
life
a
n
d
well
-
b
ein
g
.
C
o
n
s
eq
u
en
tly
,
th
e
r
e
is
a
g
r
o
win
g
d
em
a
n
d
f
o
r
co
m
p
u
tati
o
n
al
to
o
ls
b
ased
o
n
co
n
v
o
l
u
tio
n
al
n
e
u
r
al
n
etwo
r
k
s
(
C
NNs)
th
at
ca
n
lear
n
f
r
o
m
ac
cid
en
t
r
ec
o
r
d
s
in
in
d
u
s
tr
ial
s
ec
to
r
s
[
1
]
,
t
o
f
ac
ilit
ate
d
ec
is
io
n
-
m
ak
in
g
th
at
co
n
tr
ib
u
tes
to
im
p
r
o
v
e
th
e
en
f
o
r
ce
m
en
t
o
f
s
tr
ict
r
u
les
a
n
d
c
o
n
s
tan
t
m
o
n
ito
r
in
g
in
wo
r
k
p
lace
s
[
2
]
.
T
h
is
is
b
e
ca
u
s
e
C
NNs
h
av
e
m
a
d
e
s
ig
n
i
f
ican
t
ad
v
an
ce
s
in
th
e
f
ield
o
f
f
ac
ial
r
ec
o
g
n
itio
n
,
f
atig
u
e
d
etec
tio
n
[
3
]
,
im
ag
e
c
lass
if
icatio
n
an
d
r
ea
l
-
tim
e
v
is
u
al
wea
p
o
n
d
etec
tio
n
in
cr
im
e
an
d
s
u
r
v
eillan
c
e
[
4
]
,
a
n
d
ev
en
in
th
e
tr
a
n
s
p
o
r
ta
tio
n
s
ec
to
r
[
5
]
.
I
n
s
ec
to
r
s
s
u
ch
as
co
n
s
tr
u
ctio
n
,
d
ee
p
lea
r
n
in
g
tech
n
i
q
u
es
allo
w
eq
u
ip
m
en
t
tr
ac
k
in
g
an
d
cr
ac
k
d
etec
tio
n
[
6
]
,
b
ased
o
n
an
o
b
ject
d
ataset
th
at
s
er
v
es
as
a
b
asis
f
o
r
tr
ain
in
g
o
b
ject
d
etec
tio
n
m
o
d
els an
d
test
in
g
th
eir
p
er
f
o
r
m
an
ce
[
7
]
.
I
n
r
ec
e
n
t
y
ea
r
s
,
o
b
ject
d
etec
tio
n
h
as
b
ee
n
a
p
o
p
u
lar
t
o
p
ic
am
o
n
g
s
tu
d
ies
an
d
r
esear
ch
aim
in
g
t
o
cr
ea
te
o
b
ject
d
etec
tio
n
m
o
d
els
th
at
ar
e
ca
p
ab
le
o
f
p
er
f
o
r
m
in
g
u
s
ef
u
l
task
s
o
f
ar
ea
s
elec
tio
n
,
f
ea
tu
r
e
ex
tr
ac
tio
n
as
a
k
ey
attr
ib
u
te
an
d
ca
teg
o
r
izatio
n
in
im
ag
es
an
d
v
id
e
o
s
,
b
ased
o
n
tr
ain
in
g
a
s
u
itab
ly
l
ab
elled
d
ataset
[
8
]
.
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
Dete
ctio
n
o
f e
leme
n
ts
o
f p
ers
o
n
a
l sa
fety
fo
r
th
e
p
r
ev
en
tio
n
o
f a
cc
id
en
ts
a
t
…
(
Ma
r
ia
C
la
u
d
ia
B
o
n
f
a
n
te
)
5825
T
h
u
s
,
u
s
in
g
th
e
d
ee
p
lear
n
i
n
g
(
DL
)
ap
p
r
o
ac
h
an
d
h
an
d
i
n
h
an
d
with
co
m
p
u
ter
v
is
io
n
,
s
tu
d
ies
h
av
e
b
ee
n
co
n
d
u
cte
d
to
i
d
en
tify
an
d
lo
c
ate
o
b
jects
o
f
ce
r
tain
class
es
in
im
ag
es
an
d
v
id
eo
s
in
r
ea
l
tim
e,
f
o
r
an
ac
cu
r
ate
ass
es
s
m
en
t o
f
PP
E
wo
r
k
en
v
ir
o
n
m
en
ts
[
9
]
.
C
NN
-
b
ased
alg
o
r
ith
m
s
o
u
tp
er
f
o
r
m
o
t
h
er
tech
n
iq
u
es
in
f
ea
t
u
r
e
ex
tr
ac
tio
n
,
s
o
th
ey
ar
e
th
e
m
o
s
t
u
s
ed
m
eth
o
d
[
1
0
]
.
Gen
er
ic
o
b
ject
d
etec
to
r
s
p
r
esen
t
two
ca
te
g
o
r
ies:
two
-
s
h
o
t
an
d
s
in
g
le
-
s
h
o
t
[
1
1
]
.
T
h
e
f
ir
s
t
ac
h
iev
es
th
e
g
o
al
in
two
s
tep
s
:
r
eg
io
n
p
r
o
p
o
s
al
f
o
llo
wed
b
y
t
h
e
class
if
icatio
n
o
f
th
o
s
e
r
eg
i
o
n
s
an
d
r
ef
in
em
en
t
o
f
th
e
lo
ca
tio
n
p
r
ed
ictio
n
.
R
eg
io
n
p
r
o
p
o
s
al
n
etwo
r
k
ca
n
p
r
ed
ict
t
h
e
o
b
ject'
s
b
o
u
n
d
in
g
b
o
x
an
d
class
co
n
f
id
en
ce
s
co
r
es
s
im
u
ltan
eo
u
s
ly
u
s
in
g
f
u
lly
co
n
n
ec
ted
la
y
er
s
at
ea
ch
p
o
s
itio
n
[
1
2
]
.
T
h
e
s
ec
o
n
d
d
etec
to
r
s
a
r
e
ca
p
ab
le
o
f
a
n
aly
zin
g
a
n
im
ag
e
with
a
s
in
g
le
n
etwo
r
k
ass
ess
m
en
t,
th
ese
f
o
cu
s
o
n
all
s
p
at
ial
r
eg
io
n
p
r
o
p
o
s
als
f
o
r
o
b
ject
d
etec
tio
n
th
r
o
u
g
h
a
r
elativ
ely
s
im
p
ler
ar
ch
itectu
r
e,
ex
am
p
les
o
f
th
is
ty
p
e
is
y
o
u
o
n
l
y
lo
o
k
o
n
c
e
(
YOL
O)
[
1
3
]
.
On
e
-
s
tag
e
alg
o
r
ith
m
s
ar
e
f
aster
,
b
u
t
less
p
r
ec
is
e,
wh
ile
two
-
s
tag
e
o
b
ject
d
etec
tio
n
alg
o
r
ith
m
s
ar
e
s
lo
wer
,
b
u
t
m
o
r
e
ac
c
u
r
ate
.
I
n
ad
d
itio
n
t
o
th
e
ab
o
v
e
class
if
icatio
n
,
m
u
lti
-
s
tag
e
m
eth
o
d
d
etec
to
r
s
[
1
4
]
ar
e
in
clu
d
ed
,
wh
ic
h
ar
e
m
ain
l
y
f
o
cu
s
ed
o
n
th
e
s
elec
tiv
e
r
eg
io
n
p
r
o
p
o
s
al
s
tr
ateg
y
th
r
o
u
g
h
a
v
er
y
co
m
p
le
x
ar
ch
itectu
r
e.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
ese
alg
o
r
ith
m
s
ar
e
co
m
p
ar
ed
u
s
in
g
k
n
o
wn
d
atasets
[
1
5
]
,
[
1
6
]
an
d
m
etr
ics s
u
ch
as
p
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e
,
I
n
ter
s
ec
tio
n
o
v
er
U
n
io
n
(
I
o
U)
,
wh
ich
allo
ws
to
co
m
b
i
n
e
th
e
m
ea
s
u
r
es
o
f
ex
h
au
s
tiv
en
ess
an
d
p
r
ec
is
io
n
in
a
s
in
g
le
v
alu
e.
An
o
th
er
ev
alu
atio
n
m
etr
ic
u
s
ed
is
av
er
ag
e
p
r
ec
is
io
n
(
AP)
wh
ich
is
th
e
ar
ea
u
n
d
er
th
e
c
u
r
v
e,
th
at
is
,
u
n
d
er
th
e
g
r
ap
h
g
e
n
er
ated
b
y
th
e
m
etr
ics ax
is
=
p
r
ec
is
io
n
an
d
a
x
is
=
r
ec
all,
g
iv
en
b
y
t
h
e
ex
p
r
ess
io
n
(
1
)
.
=
(
)
/
(
ℎ
)
(
1
)
T
h
e
m
ea
n
av
e
r
ag
e
p
r
ec
is
io
n
(
m
AP)
m
etr
ic
is
also
u
s
ed
,
w
h
ich
ca
lcu
lates
th
e
av
er
ag
e
v
alu
e
o
f
th
e
av
er
a
g
e
p
r
ec
is
io
n
(
AP)
in
all
class
es,
it is
g
iv
en
b
y
ex
p
r
ess
io
n
(
2
)
.
=
1
N
∑
A
Pi
N
i
=
1
(
2
)
I
n
r
e
v
iew
o
f
p
r
ev
i
o
u
s
wo
r
k
,
we
id
en
tifie
d
th
at
YOL
O
p
o
p
u
lar
ized
th
e
o
n
e
-
s
tag
e
a
p
p
r
o
ac
h
b
y
d
em
o
n
s
tr
atin
g
r
ea
l
-
tim
e
p
r
ed
i
ctio
n
s
an
d
ac
h
iev
in
g
r
em
ar
k
a
b
le
d
etec
tio
n
s
p
ee
d
.
T
h
e
n
etwo
r
k
d
iv
id
es a
n
im
ag
e
in
to
a
g
r
id
o
f
s
ize
G
×
G,
a
n
d
ea
ch
g
r
id
g
en
er
ates
p
r
e
d
ictio
n
s
f
o
r
b
o
u
n
d
in
g
b
o
x
es.
E
ac
h
b
o
u
n
d
in
g
b
o
x
is
lim
ited
to
h
a
v
in
g
o
n
l
y
o
n
e
cl
ass
d
u
r
in
g
p
r
ed
ictio
n
.
T
h
is
h
a
s
b
ee
n
o
p
tim
ized
i
n
d
i
f
f
er
en
t
v
er
s
io
n
s
,
YOL
Ov
2
[
1
7
]
is
n
am
e
d
af
ter
th
e
im
p
r
o
v
em
en
t
th
at
in
clu
d
es
b
atch
n
o
r
m
aliza
tio
n
,
h
ig
h
-
r
eso
lu
tio
n
cl
ass
if
ier
an
d
an
ch
o
r
f
r
am
es
[
1
8
]
.
L
ater
,
YOL
Ov
3
was
p
r
o
p
o
s
ed
b
ased
o
n
R
esid
u
al
B
lo
ck
s
[
1
9
]
,
w
h
ich
ar
e
em
p
lo
y
ed
in
f
ea
tu
r
e
lear
n
in
g
an
d
ar
e
co
m
p
o
s
ed
o
f
co
n
v
o
lu
tio
n
al
co
n
n
ec
tio
n
s
,
p
r
o
v
id
i
n
g
th
e
a
b
ilit
y
to
d
etec
t
its
d
is
tin
g
u
is
h
in
g
f
ea
tu
r
e
at
th
r
ee
d
if
f
er
e
n
t le
v
els,
th
u
s
m
ak
in
g
o
b
jects o
f
v
ar
io
u
s
s
izes m
o
r
e
co
r
r
ec
tly
r
ec
o
g
n
ized
.
T
h
e
last
lay
er
o
f
th
e
YOL
O
-
v
3
m
o
d
els
ca
n
b
e
m
o
d
if
ied
t
o
ac
co
m
m
o
d
ate
o
b
ject
class
es
o
f
in
te
r
est
s
u
ch
as
h
elm
et
a
n
d
/o
r
s
af
ety
waistco
at
in
co
n
s
tr
u
ctio
n
en
v
ir
o
n
m
e
n
ts
[
2
0
]
.
A
r
ea
l
ap
p
licatio
n
o
f
t
h
is
d
etec
to
r
is
its
in
teg
r
atio
n
in
to
th
e
Saf
eFac
in
tellig
en
t
s
y
s
te
m
[
2
1
]
,
d
esig
n
ed
f
o
r
s
af
ety
m
an
ag
em
en
t
in
m
an
u
f
ac
tu
r
in
g
en
v
ir
o
n
m
en
ts
.
T
h
i
s
s
y
s
tem
u
s
es
a
s
et
o
f
in
s
talled
ca
m
er
as
to
ca
p
tu
r
e
im
ag
es
o
f
wo
r
k
er
s
ap
p
r
o
ac
h
in
g
m
ac
h
i
n
er
y
in
d
an
g
e
r
o
u
s
s
itu
atio
n
s
.
T
h
e
s
y
s
tem
an
aly
s
e
s
th
ese
im
ag
es
in
r
ea
l
tim
e
an
d
aler
ts
m
an
ag
er
s
wh
en
it
d
ete
cts
u
n
s
af
e
b
eh
a
v
io
r
o
r
lack
o
f
PP
E
u
s
e.
Fu
r
th
er
o
n
,
YOL
Ov
4
em
er
g
es
with
th
e
ab
ilit
y
to
r
ec
o
g
n
ize
m
u
ltip
le
o
b
jects
in
a
s
in
g
le
f
r
am
e
[
2
2
]
.
YOL
Ov
4
-
T
in
y
[
2
3
]
is
lig
h
ter
an
d
d
esig
n
ed
to
r
ed
u
ce
th
e
tim
e
f
o
r
o
b
ject
d
etec
tio
n
.
I
ts
s
u
p
p
o
r
ts
r
ea
l
-
tim
e
im
ag
e
an
aly
s
is
also
wh
en
r
u
n
n
in
g
o
n
em
b
ed
d
ed
s
y
s
tem
s
o
r
d
ev
ices
[
2
4
]
.
A
n
im
p
r
o
v
ed
S
C
M
-
YOL
O
h
elm
et
d
etec
tio
n
is
p
r
o
p
o
s
ed
[
2
5
]
.
T
h
is
v
er
s
io
n
is
o
p
tim
ized
b
y
in
teg
r
atin
g
a
s
p
atial
p
y
r
am
id
p
o
o
lin
g
(
SP
P)
m
o
d
u
le
[
2
6
]
.
Similar
ly
,
SAI
-
YOL
O
[
2
7
]
r
ed
u
ce
s
th
e
n
u
m
b
er
o
f
p
ar
am
eter
s
an
d
c
o
m
p
u
tatio
n
al
d
if
f
icu
lty
an
d
in
cr
ea
s
es th
e
n
etwo
r
k
d
etec
tio
n
s
p
ee
d
wh
ile
m
ain
tain
in
g
ce
r
t
ain
r
ec
o
g
n
itio
n
ac
cu
r
ac
y
[
2
8
]
.
Nex
t,
YOL
Ov
5
[
2
9
]
is
u
s
ed
to
tr
ain
a
d
ataset
o
f
f
ac
e
m
ask
.
Als
o
,
FD
-
YOL
Ov
5
[
3
0
]
in
teg
r
ates
a
f
u
zz
y
-
b
ased
im
ag
e
m
o
d
u
le,
wh
ich
allo
ws
d
if
f
er
e
n
tiatin
g
b
etwe
en
v
ar
io
u
s
ty
p
es
o
f
h
el
m
ets
tr
ain
in
g
a
7
6
4
im
ag
e
d
ataset.
I
t
o
p
tim
izes
YOL
Ov
5
[
3
1
]
to
ad
d
r
ess
lo
w
ac
cu
r
ac
y
an
d
r
o
b
u
s
tn
ess
p
r
o
b
le
m
s
f
o
r
s
m
all
o
b
jects
in
co
m
p
le
x
n
at
u
r
al
e
n
v
ir
o
n
m
e
n
ts
an
d
i
n
teg
r
ates
th
e
Gh
o
s
t
m
o
d
u
le
[
3
2
]
wh
ich
r
eq
u
ir
es
f
ewe
r
p
ar
am
eter
s
a
n
d
less
co
m
p
u
tatio
n
al
co
m
p
le
x
ity
.
Similar
ly
,
tr
ain
in
g
o
f
d
if
f
er
en
t
v
er
s
io
n
s
YOL
Ov
5
is
d
o
n
e
u
s
in
g
a
d
ataset
o
f
1
.
4
8
5
im
a
g
es,
co
m
p
r
is
in
g
f
o
u
r
PP
E
in
ed
u
ca
tio
n
al
lab
o
r
ato
r
ies,
T
h
e
Y
OL
Ov
5
n
ap
p
r
o
ac
h
ac
h
iev
e
d
th
e
h
ig
h
est
m
AP
o
f
7
7
.
4
0
%
f
o
r
s
m
all
an
d
lar
g
e
in
s
tan
ce
s
[
3
3
]
.
R
eg
ar
d
in
g
r
ela
ted
wo
r
k
i
n
v
o
lv
i
n
g
an
en
h
a
n
ce
m
en
t
o
f
e
x
is
tin
g
d
atasets
,
in
s
tu
d
y
[
3
4
]
p
r
esen
t
s
a
ca
s
e
d
etec
tio
n
m
o
d
el
with
5
,
0
0
0
im
a
g
es,
T
h
e
d
ataset
was
tr
ain
ed
with
th
e
YOL
Ov
3
,
YOL
Ov
4
,
YOL
Ov
5
,
o
b
tain
in
g
b
etter
r
esu
lts
f
o
r
YOL
OR
.
Al
s
o
,
a
co
m
p
ar
i
s
o
n
o
f
th
e
p
er
f
o
r
m
an
ce
o
f
Y
OL
Ov
7
w
ith
o
th
er
s
m
o
d
el
s
i
s
m
ad
e
[
3
5
]
,
i
t
s
c
an
d
et
e
ct
th
e
ab
s
en
ce
o
f
s
af
e
ty
h
el
m
et
s
o
n
d
ar
k
-
s
k
in
n
ed
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.
1
4
,
No
.
5
,
Octo
b
e
r
2
0
2
4
:
5
8
2
4
-
5
8
3
3
5826
p
eo
p
l
e
in
lo
w
l
ig
h
t
co
n
d
i
tio
n
s
,
t
h
i
s
i
s
b
ec
au
s
e
i
t
s
ar
c
h
i
tec
tu
r
e
c
o
n
s
i
s
t
s
o
f
th
r
ee
im
p
o
r
tan
t
p
ar
t
s
:
th
e
s
p
in
e,
th
e
n
ec
k
an
d
th
e
h
ea
d
.
T
h
e
b
ac
k
b
o
n
e
i
s
r
esp
o
n
s
ib
le
f
o
r
ex
t
r
a
ct
in
g
f
e
atu
r
e
s
f
r
o
m
th
e
g
iv
en
in
p
u
t
im
a
g
e
s
,
th
e
n
ec
k
m
a
in
ly
g
en
er
at
e
s
f
ea
tu
r
e
p
y
r
am
id
s
,
an
d
th
e
h
ea
d
p
er
f
o
r
m
s
th
e
f
in
a
l
d
et
ec
tio
n
a
s
o
u
tp
u
t
.
I
t
s
d
e
s
i
g
n
i
s
b
as
ed
o
n
ef
f
ic
ien
t
la
y
er
ag
g
r
eg
a
tio
n
n
et
wo
r
k
(
E
L
A
N
)
,
wh
i
ch
m
ak
es
u
s
e
o
f
ex
p
an
s
io
n
,
s
h
u
f
f
lin
g
a
n
d
ca
r
d
in
a
li
ty
f
u
s
io
n
to
in
cr
e
as
e
th
e
l
ea
r
n
in
g
c
ap
a
ci
ty
o
f
t
h
e
n
et
wo
r
k
w
i
th
o
u
t
co
m
p
r
o
m
i
s
in
g
th
e
o
r
ig
in
al
g
r
ad
i
en
t
p
at
h
.
O
th
e
r
m
o
d
e
l
s
wi
th
th
e
o
b
jec
t
s
o
f
in
t
er
e
s
t
o
f
th
i
s
s
t
u
d
y
em
p
lo
y
in
g
Y
OL
O
v
5
h
av
e
b
ee
n
d
o
n
e
in
[
3
6
]
–
[
3
8
]
.
Ho
wev
er
,
th
ese
p
r
ev
io
u
s
r
esear
ch
es
o
n
ly
b
u
ild
th
e
m
o
d
els
with
a
s
in
g
le
ty
p
e
o
f
o
b
ject;
th
is
wo
r
k
p
r
o
p
o
s
es
th
e
d
etec
tio
n
o
f
ele
m
en
ts
s
u
ch
as
in
d
u
s
tr
ial
h
elm
ets
an
d
f
ac
e
m
ask
s
in
im
a
g
e,
m
ak
i
n
g
u
s
e
o
f
a
d
ataset
o
f
2
,
0
0
0
im
a
g
es
o
f
p
eo
p
le
with
an
d
with
o
u
t
th
ese
p
r
o
tectiv
e
elem
en
ts
co
llected
f
r
o
m
d
if
f
e
r
en
t
s
o
u
r
ce
s
an
d
lab
elin
g
s
ev
e
r
al
class
es
in
o
n
e
im
ag
e.
T
h
e
s
elec
tio
n
o
f
i
m
ag
es
co
v
e
r
in
g
a
wid
e
v
a
r
iety
o
f
s
ce
n
ar
io
s
,
f
r
o
m
cr
o
wd
s
to
m
o
r
e
f
o
cu
s
ed
s
itu
a
tio
n
s
an
d
v
ar
y
in
g
th
e
d
is
tan
ce
b
etwe
en
p
eo
p
le
to
ac
h
iev
e
th
e
g
en
er
ality
o
f
th
e
m
o
d
el.
Alth
o
u
g
h
YOL
O
h
as
b
ee
n
o
p
tim
ized
in
d
if
f
er
e
n
t
v
er
s
io
n
s
,
th
e
ex
p
e
r
im
en
t
is
p
e
r
f
o
r
m
e
d
o
n
l
y
with
YOL
O
v
5
an
d
YOL
O
v
7
,
t
h
e
l
atter
b
ein
g
th
e
m
o
s
t
r
ec
en
t
at
th
e
tim
e
o
f
th
e
d
ev
elo
p
m
en
t
o
f
th
is
wo
r
k
,
we
p
r
esen
t th
e
p
ar
a
m
eter
izatio
n
a
n
d
an
aly
s
is
o
f
th
e
r
esu
lts
o
f
th
e
two
tr
ain
ed
m
o
d
els.
2.
M
E
T
H
O
D
Mo
d
el
b
u
ild
in
g
f
o
r
e
n
d
p
o
in
t
p
r
o
tectio
n
p
latf
o
r
m
(
E
PP
)
d
et
ec
tio
n
r
elies
o
n
s
ev
e
r
al
cr
itic
al
p
h
ases
:
d
ata
co
llectio
n
,
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
,
d
ata
p
ar
titi
o
n
in
g
an
d
m
o
d
el
tr
ain
in
g
[
3
9
]
.
E
ac
h
o
f
th
ese
p
h
ases
i
s
ess
en
tial
to
en
s
u
r
e
th
e
p
er
f
o
r
m
an
ce
an
d
ac
cu
r
a
cy
o
f
th
e
f
in
al
m
o
d
el.
I
n
th
e
f
o
llo
win
g
,
we
ex
p
lain
i
n
d
etail
th
e
task
s
p
er
f
o
r
m
ed
,
th
e
s
o
u
r
ce
an
d
s
ize
o
f
th
e
d
ataset,
to
o
ls
u
s
e
d
,
tr
ain
in
g
p
ar
am
ete
r
s
,
an
d
th
e
p
r
o
d
u
cts
o
b
tain
ed
in
ea
ch
p
h
ase.
2
.1
.
Da
t
a
c
o
llect
io
n
Fo
r
th
is
s
tu
d
y
,
th
e
lab
eled
d
ataset
was
as
s
em
b
led
b
y
c
o
llectin
g
2
,
0
0
0
im
ag
es
o
b
tain
e
d
f
r
e
ely
o
n
lin
e,
b
o
th
f
r
o
m
Kag
g
le
a
n
d
o
th
er
in
ter
n
et
s
o
u
r
ce
s
,
in
clu
d
in
g
th
o
s
e
ca
p
tu
r
ed
b
y
th
e
a
u
th
o
r
s
th
em
s
elv
es.
I
n
th
e
lab
elin
g
p
r
o
ce
s
s
,
th
e
to
o
l
h
ttp
s
:
//w
w
w
.
ma
ke
s
en
s
e.
a
i/
wa
s
u
s
ed
,
s
p
ec
if
y
in
g
th
e
class
as
s
h
o
wn
in
T
a
b
le
1
an
d
co
o
r
d
in
ates.
T
h
e
d
ataset
is
a
v
ailab
le
in
h
ttp
s
:
//w
w
w
.
ka
g
g
le.
co
m/d
a
ta
s
ets/
iva
n
h
ern
a
n
d
ezr
u
iz
/s
a
fety
-
h
elme
t
-
and
-
ma
s
k
,
wh
er
e
it
h
as
b
ee
n
d
u
ly
p
u
b
lis
h
ed
.
An
o
th
e
r
asp
ec
t
to
co
n
s
id
er
is
th
at
all
im
ag
es
m
u
s
t
h
av
e
th
e
s
am
e
p
ix
el
s
ize
in
h
eig
h
t a
n
d
wid
th
(
6
4
0
p
p
)
.
T
ab
le
1
.
Def
in
itio
n
o
f
class
es
C
l
a
s
s
D
e
scri
p
t
i
o
n
0
w
i
t
h
h
e
l
m
e
t
-
w
i
t
h
mas
k
1
w
i
t
h
h
e
l
m
e
t
-
w
i
t
h
o
u
t
mas
k
2
w
i
t
h
o
u
t
h
e
l
m
e
t
-
w
i
t
h
mas
k
3
w
i
t
h
o
u
t
h
e
l
m
e
t
-
w
i
t
h
o
u
t
m
a
s
k
2
.2
.
Div
is
io
n o
f
d
a
t
a
T
h
e
lab
elled
im
ag
es we
r
e
r
an
d
o
m
ly
d
iv
id
e
d
in
to
th
e
f
o
llo
win
g
p
er
ce
n
tag
e
d
is
tr
ib
u
tio
n
,
8
0
%
tr
ain
in
g
,
1
5
%
test
in
g
an
d
th
e
r
em
ain
in
g
5
%
f
o
r
v
alid
atio
n
.
T
h
e
tr
ain
in
g
s
am
p
le
is
co
m
p
o
s
ed
o
f
2
5
%
f
o
r
ea
ch
class
,
s
o
it
ca
n
b
e
s
tated
th
at
i
t
is
a
b
a
lan
ce
d
s
am
p
le,
th
e
s
am
e
p
r
o
p
o
r
tio
n
was
u
s
ed
in
test
in
g
an
d
v
alid
atio
n
.
E
ac
h
class
is
r
ep
r
esen
ted
u
n
if
o
r
m
l
y
in
t
h
e
s
u
b
s
ets,
wh
ich
is
cr
u
cial
to
a
v
o
id
b
iases
in
th
e
m
o
d
el.
R
an
d
o
m
p
ar
titi
o
n
in
g
an
d
class
b
alan
cin
g
h
elp
en
s
u
r
e
th
at
th
e
m
o
d
el
g
en
er
alize
s
well
an
d
is
n
o
t
o
v
er
ly
d
ep
en
d
e
n
t
o
n
a
s
p
ec
if
ic
class
.
2
.3
.
T
ra
ini
ng
t
he
det
ec
t
io
n
m
o
del
T
h
e
h
y
p
er
p
ar
a
m
eter
s
u
s
ed
in
th
e
tr
ain
in
g
p
r
o
ce
s
s
wer
e
s
e
lecte
d
co
n
s
id
er
in
g
th
e
r
ec
o
m
m
en
d
atio
n
f
o
u
n
d
in
th
e
liter
atu
r
e
r
e
v
iew
[
3
0
]
,
in
ter
m
s
o
f
ep
o
ch
s
an
d
b
atch
wer
e
ad
j
u
s
ted
in
o
u
r
e
x
p
er
im
en
t
o
n
a
tr
ial
-
an
d
-
er
r
o
r
b
asis
.
A
b
atch
s
ize
o
f
1
6
,
th
e
s
elec
ted
o
p
tim
izer
i
s
s
to
ch
asti
c
g
r
ad
ien
t
d
escen
t
(
SGD)
,
with
lo
g
is
tic
r
eg
r
ess
io
n
(
L
r
=
0
.
0
1
)
t
h
at
p
r
o
v
id
ed
a
n
ap
p
r
o
p
r
iate
b
alan
ce
b
etwe
en
co
n
v
er
g
e
n
ce
s
p
ee
d
a
n
d
tr
ain
i
n
g
s
tab
ilit
y
.
I
n
ad
d
itio
n
,
o
th
er
im
p
o
r
tan
t
p
ar
am
eter
s
wer
e
co
n
s
id
er
ed
to
o
p
tim
ize
th
e
tr
ain
in
g
p
r
o
ce
s
s
,
wh
ich
a
r
e
d
etailed
in
T
ab
le
2
.
T
h
ese
ad
ju
s
tm
en
ts
allo
wed
th
e
m
o
d
el
n
o
t
o
n
ly
t
o
lear
n
ef
f
ec
tiv
ely
,
b
u
t
also
to
g
en
er
alize
well
o
n
u
n
s
ee
n
d
ata,
a
n
d
th
e
ass
u
r
an
ce
o
f
o
b
tain
in
g
a
r
o
b
u
s
t
an
d
ef
f
icien
t
m
o
d
el,
s
u
itab
l
e
f
o
r
th
e
r
eq
u
i
r
ed
class
if
icatio
n
task
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
Dete
ctio
n
o
f e
leme
n
ts
o
f p
ers
o
n
a
l sa
fety
fo
r
th
e
p
r
ev
en
tio
n
o
f a
cc
id
en
ts
a
t
…
(
Ma
r
ia
C
la
u
d
ia
B
o
n
f
a
n
te
)
5827
T
ab
le
2
.
C
o
n
f
i
g
u
r
atio
n
p
ar
am
eter
s
o
f
th
e
alg
o
r
ith
m
s
P
a
r
a
me
t
e
r
v
a
l
u
e
Y
O
LO
v
5
Y
O
LO
v
7
Ép
o
c
h
1
7
0
70
B
a
t
c
h
16
16
B
o
x
l
o
ss
g
a
i
n
0
.
0
1
9
5
1
0
.
0
2
4
0
3
C
l
a
s
s
l
o
ss
g
a
i
n
0
.
0
0
0
3
8
8
0
.
0
0
1
0
9
2
O
b
j
e
c
t
l
o
ss
g
a
i
n
0
.
0
1
8
5
7
0
.
0
0
9
5
0
1
Le
a
r
n
i
n
g
r
a
t
e
0
.
0
1
0
.
0
1
W
a
r
mu
p
m
o
me
n
t
u
m
0
.
8
0
.
8
W
a
r
mu
p
e
p
o
c
h
s
3
.
0
3
.
0
O
p
t
i
mi
z
e
r
w
e
i
g
h
t
d
e
c
a
y
0
.
0
0
0
5
0
.
0
0
0
5
W
a
r
mu
p
b
i
a
s
0
.
1
0
.
1
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
Fig
u
r
e
1
s
h
o
ws
th
e
co
n
f
u
s
io
n
m
atr
ix
o
f
th
e
YOL
Ov
5
m
o
d
e
l.
I
t
s
u
m
m
ar
izes
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
tr
ain
ed
m
o
d
el
w
h
en
p
er
f
o
r
m
i
n
g
th
e
o
b
ject
d
etec
tio
n
task
f
o
r
th
e
f
o
u
r
d
ef
in
e
d
class
es.
W
h
en
an
aly
zin
g
th
e
co
n
f
u
s
io
n
m
atr
ix
,
th
e
tr
ain
ed
m
o
d
el
ca
n
d
etec
t a
ll c
lass
es a
b
o
v
e
8
0
%.
I
n
Fig
u
r
e
2
,
th
e
p
lo
t
o
f
F1
ag
ain
s
t
co
n
f
id
en
ce
s
h
o
ws
th
at
th
e
m
o
d
el
tr
ain
ed
with
YOL
Ov
5
ac
h
iev
es
an
F1
-
s
co
r
e
o
f
0
.
8
8
ac
r
o
s
s
all
class
es
(
b
lu
e
cu
r
v
e
)
.
T
h
e
F1
-
s
co
r
e
is
a
v
alu
e
th
at
r
ep
r
esen
ts
a
tr
ad
e
-
o
f
f
b
etwe
e
n
th
e
ac
cu
r
ac
y
r
ate
an
d
th
e
r
ec
al
l
r
ate
o
f
th
e
m
o
d
el.
R
ef
er
r
in
g
to
th
e
g
r
ap
h
s
p
lo
tted
p
r
ec
is
io
n
v
er
s
u
s
co
n
f
id
en
ce
is
1
.
0
0
t
o
0
.
9
4
,
an
d
r
ec
all
v
e
r
s
u
s
co
n
f
id
e
n
ce
o
f
all
class
es
is
0
.
9
5
,
an
d
th
e
m
etr
ic
m
AP@
0
.
5
is
0
.
8
7
.
T
h
e
m
etr
ics
ar
e
s
h
o
wn
in
T
ab
le
3
.
Fig
u
r
e
3
s
h
o
w
th
e
c
o
n
f
u
s
io
n
m
atr
ix
o
f
YOL
Ov
7
,
wh
er
e
th
e
tr
ain
ed
m
o
d
el
ca
n
d
etec
t c
lass
es e
q
u
al
to
o
r
g
r
ea
t
er
th
an
8
5
%.
Fig
u
r
e
4
s
h
o
ws
th
e
YOL
O
v
7
p
er
f
o
r
m
an
ce
p
l
o
ts
:
F1
v
s
co
n
f
id
en
ce
f
r
o
m
0
.
8
9
,
Pre
cisi
o
n
v
s
.
co
n
f
id
en
ce
f
r
o
m
1
.
0
0
to
0
.
9
3
,
r
ec
all
v
s
co
n
f
id
en
ce
o
f
0
.
9
8
an
d
m
AP@
0
.
5
is
0
.
8
9
.
I
n
ter
m
s
o
f
th
e
F1
-
s
co
r
e
m
etr
ic,
YOL
Ov
5
o
b
tain
ed
a
s
co
r
e
o
f
0
.
8
8
wh
ile
YOL
Ov
7
o
b
tain
ed
0
.
8
9
.
T
h
is
in
d
icate
s
th
at
YOL
Ov
7
h
as
a
b
etter
b
alan
ce
b
etwe
en
ac
cu
r
ac
y
r
ate
an
d
r
ec
all
r
ate
th
an
YOL
Ov
5
.
On
th
e
o
th
er
h
an
d
,
th
e
ac
cu
r
ac
y
v
s
co
n
f
id
en
ce
g
r
ap
h
s
s
h
o
w
th
at
b
o
th
m
o
d
els
h
av
e
a
v
er
y
h
ig
h
a
cc
u
r
ac
y
r
ate,
YOL
Ov
5
s
co
r
ed
0
.
9
5
in
all
class
es
wh
ile
YOL
Ov
7
s
co
r
ed
0
.
9
8
.
R
eg
ar
d
in
g
th
e
m
etr
ic
m
AP@
0
.
5
,
YOL
Ov
5
s
co
r
ed
0
.
8
7
wh
ile
YOL
Ov
7
s
co
r
ed
0
.
8
9
.
T
h
e
p
e
r
f
o
r
m
an
ce
m
etr
ic
s
f
o
r
ea
ch
class
ar
e
s
h
o
wn
i
n
T
ab
le
4
.
Fig
u
r
e
1
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
th
e
YOL
Ov
5
m
o
d
el
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.
1
4
,
No
.
5
,
Octo
b
e
r
2
0
2
4
:
5
8
2
4
-
5
8
3
3
5828
Fig
u
r
e
5
s
h
o
ws
th
e
v
alid
atio
n
o
f
b
o
t
h
tr
ain
ed
m
o
d
els
.
I
n
Fi
g
u
r
e
5
(
a)
,
YOL
Ov
5
s
h
o
ws
th
e
r
esu
lts
o
f
im
ag
e
p
r
ep
r
o
ce
s
s
in
g
with
ed
g
e
d
etec
tio
n
f
o
r
p
eo
p
le
wea
r
in
g
PP
E
.
T
h
is
ap
p
r
o
ac
h
allo
ws
to
clea
r
ly
id
en
tify
th
e
co
n
to
u
r
s
o
f
wo
r
k
e
r
s
an
d
th
eir
p
r
o
tectiv
e
eq
u
ip
m
en
t,
h
ig
h
lig
h
tin
g
its
ef
f
ec
tiv
e
n
ess
in
s
itu
atio
n
s
wh
er
e
th
e
elem
en
ts
ar
e
clea
r
ly
v
is
ib
le.
On
th
e
o
th
er
h
an
d
,
Fig
u
r
e
5
(
b
)
p
r
esen
ts
th
e
r
esu
lts
o
b
tain
ed
with
YOL
Ov
7
,
d
em
o
n
s
tr
atin
g
a
g
r
ea
ter
ab
ilit
y
to
d
etec
t
s
m
all
o
b
jects,
ev
en
wh
en
th
ey
ar
e
p
ar
tially
h
id
d
e
n
b
y
o
th
er
elem
e
n
ts
an
d
ar
e
in
v
ar
io
u
s
p
o
s
itio
n
s
.
YOL
Ov
7
m
an
ag
es
to
co
r
r
ec
tl
y
id
en
tify
PP
E
d
esp
ite
th
e
ad
d
ed
d
if
f
icu
lties
d
u
e
to
o
b
s
tr
u
ctio
n
s
an
d
v
ar
iab
ilit
y
i
n
o
b
ject
p
o
s
itio
n
s
.
C
o
m
p
a
r
is
o
n
o
f
f
ea
tu
r
es
b
etwe
en
YOL
Ov
5
an
d
YOL
Ov
7
ca
n
b
e
s
ee
n
in
T
ab
le
5
.
C
o
m
p
ar
ed
to
o
th
er
wo
r
k
s
,
o
u
r
ex
p
er
im
e
n
t
u
s
ed
s
u
p
er
i
o
r
v
er
s
io
n
s
o
f
YOL
O
u
s
ed
in
[
3
4
]
to
d
etec
t
PP
E
.
C
o
m
p
ar
ed
to
o
th
er
r
elate
d
[
2
9
]
–
[
3
1
]
an
d
[
3
5
]
,
o
u
r
e
x
p
er
im
en
t
ac
h
ie
v
ed
a
m
AP@
0
.
5
o
f
0
.
8
7
an
d
a
m
AP@
0
.
5
:0
.
9
5
o
f
0
.
6
4
in
t
h
e
o
v
er
all
class
r
an
k
in
g
.
I
f
we
co
m
p
a
r
e
th
e
r
esu
lts
with
th
e
p
ap
er
[
3
5
]
w
h
ich
em
p
lo
y
YOL
Ov
7
,
o
u
r
s
ac
h
ie
v
ed
a
m
AP@
0
.
5
o
f
0
.
8
9
a
n
d
a
m
AP@
0
.
5
:0
.
9
5
o
f
0
.
6
6
,
in
b
o
th
ex
p
er
im
en
ts
a
d
ataset
s
ig
n
if
ican
t
co
m
p
ar
ed
t
o
th
e
im
ag
e
s
et
u
s
ed
in
th
e
s
tu
d
ies
[
2
9
]
,
[
3
0
]
,
[
3
3
]
wh
ich
a
r
e
s
m
aller
an
d
th
is
d
o
es
n
o
t
g
u
a
r
an
tee
a
n
ac
c
u
r
ac
y
in
th
e
v
alid
atio
n
.
I
n
co
n
tr
ast
,
o
u
r
tr
ain
e
d
m
o
d
el
ca
n
p
er
f
ec
tly
id
en
tify
f
ea
tu
r
es
in
v
id
eo
s
tr
ea
m
s
to
g
eth
er
with
s
till
im
ag
es.
Dif
f
er
en
t
f
r
o
m
o
th
er
wo
r
k
s
th
at
p
r
esen
t
a
c
o
m
p
ar
is
o
n
o
f
d
if
f
e
r
en
t
v
er
s
io
n
s
o
f
YOL
O
in
class
if
icatio
n
s
y
s
tem
s
o
f
a
s
in
g
le
class
,
in
o
u
r
w
o
r
k
a
c
o
m
p
ar
is
o
n
b
e
twee
n
two
v
er
s
io
n
s
o
f
YOL
O
f
o
r
a
m
u
lticlas
s
s
y
s
t
em
is
ca
r
r
ied
o
u
t
an
d
h
y
p
er
p
a
r
am
eter
s
h
av
e
b
ee
n
u
s
ed
th
at
o
p
tim
ize
th
e
u
s
e
o
f
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
.
Fig
u
r
e
2
.
YOL
Ov
5
p
er
f
o
r
m
a
n
ce
m
etr
ics
T
ab
le
3
.
YOL
Ov
5
p
er
f
o
r
m
an
c
e
m
etr
ics
C
l
a
s
s
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
M
a
p
@
.
5
M
a
p
@
0
.
5
:
0
.
9
5
0
0
.
64
0
.
82
0
.
68
0
.
36
1
0
.
94
0
.
91
0
.
94
0
.
78
2
0
.
95
0
.
84
0
.
89
0
.
58
3
0
.
95
0
.
97
0
.
98
0
.
85
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
Dete
ctio
n
o
f e
leme
n
ts
o
f p
ers
o
n
a
l sa
fety
fo
r
th
e
p
r
ev
en
tio
n
o
f a
cc
id
en
ts
a
t
…
(
Ma
r
ia
C
la
u
d
ia
B
o
n
f
a
n
te
)
5829
Fig
u
r
e
3
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
th
e
YOL
Ov
7
m
o
d
el
Fig
u
r
e
4
.
YOL
Ov
7
p
er
f
o
r
m
a
n
ce
m
etr
ics
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.
1
4
,
No
.
5
,
Octo
b
e
r
2
0
2
4
:
5
8
2
4
-
5
8
3
3
5830
T
ab
le
4
.
YOL
Ov
7
p
er
f
o
r
m
an
c
e
m
etr
ics
C
l
a
s
s
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
M
a
p
@
.
5
M
a
p
@
0
.
5
:
0
.
9
5
0
0
.
62
0
.
85
0
.
71
0
.
38
1
0
.
95
0
.
94
0
.
97
0
.
83
2
0
.
93
0
.
87
0
.
92
0
.
58
3
0
.
94
0
.
98
0
.
98
0
.
85
(
a)
(
b
)
Fig
u
r
e
5
.
Valid
atio
n
o
f
t
h
e
m
o
d
els:
(
a)
Y
OL
O
v
5
an
d
(
b
)
Y
OL
O
v7
T
ab
le
5
.
C
o
m
p
a
r
is
o
n
o
f
f
ea
tu
r
es b
etwe
en
YOL
Ov
5
an
d
YO
L
Ov
7
F
e
a
t
u
r
e
s
Y
O
LO
v
5
Y
O
LO
v
7
mA
P
@
0
.
5
l
o
w
e
r
H
i
g
h
e
r
Ep
o
c
h
s
h
i
g
h
e
r
Lo
w
e
r
Tr
a
i
n
i
n
g
p
r
o
c
e
ss
l
o
w
e
r
H
i
g
h
e
r
Tr
a
i
n
e
d
m
o
d
e
l
si
z
e
h
i
g
h
e
r
Lo
w
e
r
I
n
f
e
r
e
n
c
e
i
n
C
P
U
s
y
st
e
ms
F
a
st
e
r
S
l
o
w
e
r
U
se
o
f
mem
o
r
y
i
n
t
r
a
i
n
i
n
g
S
t
a
b
l
e
U
n
st
a
b
l
e
B
a
c
k
b
o
n
e
(
c
o
mp
u
t
a
t
i
o
n
a
l
B
l
o
c
k
)
a
r
c
h
i
t
e
c
t
u
r
e
D
a
r
k
n
e
t
w
i
t
h
c
r
o
ss
st
a
g
e
p
a
r
t
i
a
l
n
e
t
w
o
r
k
(
C
S
P
N
e
t
)
Ex
t
e
n
d
e
d
e
f
f
i
c
i
e
n
t
l
a
y
e
r
a
g
g
r
e
g
a
t
i
o
n
n
e
t
w
o
r
k
(
E
-
ELA
N
)
F
l
o
a
t
i
n
g
p
o
i
n
t
o
p
e
r
a
t
i
o
n
s
l
o
w
e
r
H
i
g
h
e
r
r
e
sp
o
n
s
e
t
i
m
e
(
t
e
st
o
r
i
n
f
e
r
e
n
c
e
)
h
i
g
h
e
r
Lo
w
e
r
4.
CO
NCLU
SI
O
N
A
s
u
b
s
tan
tial
co
n
tr
ib
u
tio
n
o
f
o
u
r
wo
r
k
lies
in
th
e
co
ll
ec
tio
n
o
f
im
ag
es
r
ep
r
esen
tin
g
v
ar
i
o
u
s
s
itu
atio
n
s
,
all
ac
cu
r
ately
lab
eled
an
d
ass
ig
n
ed
to
s
p
ec
if
i
c
class
es.
Sev
er
al
f
ac
to
r
s
wer
e
co
n
s
id
er
e
d
in
d
eter
m
in
in
g
th
e
q
u
ality
an
d
d
iv
er
s
ity
o
f
th
e
d
ata
to
ac
h
iev
e
a
h
ig
h
l
y
g
en
e
r
alize
d
m
o
d
el.
Key
co
n
s
id
er
atio
n
s
in
clu
d
ed
:
i
)
T
h
e
in
clu
s
io
n
o
f
im
ag
es
ca
p
tu
r
ed
u
n
d
e
r
v
ar
i
o
u
s
lig
h
tin
g
c
o
n
d
itio
n
s
,
allo
win
g
th
e
m
o
d
el
to
ef
f
ec
tiv
ely
ad
a
p
t
to
r
ea
l
-
w
o
r
ld
s
itu
atio
n
s
.
T
h
is
is
ess
en
tial
to
en
s
u
r
e
th
at
th
e
m
o
d
el
ca
n
p
e
r
f
o
r
m
o
p
tim
ally
in
en
v
ir
o
n
m
en
ts
with
ch
a
n
g
in
g
lig
h
t
lev
els.
ii
)
T
h
e
in
c
o
r
p
o
r
atio
n
o
f
im
ag
es
th
at
r
ep
r
ese
n
t
b
o
th
in
d
o
o
r
an
d
o
u
td
o
o
r
en
v
ir
o
n
m
en
ts
,
ex
ten
d
in
g
th
e
a
p
p
licab
ilit
y
o
f
th
e
al
g
o
r
ith
m
t
o
a
wid
e
r
an
g
e
o
f
r
e
al
-
wo
r
ld
s
itu
atio
n
s
an
d
o
f
f
er
in
g
v
er
s
atility
in
its
p
er
f
o
r
m
an
ce
.
An
d
iii
)
T
h
e
c
o
n
s
id
er
atio
n
th
e
p
r
esen
ce
o
f
co
m
p
lex
o
b
jects
o
r
b
ac
k
g
r
o
u
n
d
s
,
v
ar
iatio
n
s
i
n
p
e
o
p
le'
s
clo
th
in
g
,
an
d
d
if
f
e
r
en
t
t
im
es
o
f
d
a
y
.
T
h
is
a
p
p
r
o
ac
h
n
o
t
o
n
ly
en
s
u
r
es
th
e
q
u
ality
an
d
d
i
v
er
s
ity
o
f
th
e
d
a
taset
u
s
ed
in
tr
ain
i
n
g
,
b
u
t
also
s
tr
en
g
th
en
s
t
h
e
m
o
d
el'
s
ab
ilit
y
to
g
en
er
alize
an
d
p
er
f
o
r
m
ef
f
icien
tly
i
n
a
wid
e
v
ar
iety
o
f
s
itu
atio
n
s
.
I
n
f
u
t
u
r
e
wo
r
k
,
m
o
d
if
icatio
n
s
will
b
e
m
ad
e
to
t
h
e
d
ataset
b
y
ad
d
in
g
class
es
o
f
o
th
e
r
PP
E
s
u
ch
as:
waistco
ats,
g
lo
v
es,
an
d
b
o
o
ts
.
I
n
ad
d
itio
n
,
t
h
e
p
o
s
s
ib
ilit
y
o
f
u
s
in
g
n
ew
v
er
s
io
n
s
o
f
YOL
O
co
u
ld
b
e
ex
p
lo
r
ed
,
s
in
ce
it
i
s
ev
id
en
t
th
at
th
e
f
ield
o
f
o
b
ject
d
etec
to
r
s
is
im
p
r
o
v
in
g
,
an
d
n
ew
p
r
o
p
o
s
als
ar
e
co
n
s
tan
tly
ap
p
ea
r
in
g
,
s
o
it
co
u
ld
b
e
co
n
s
id
er
e
d
a
m
o
r
e
c
o
m
p
lete
s
o
lu
tio
n
f
o
r
th
e
m
an
a
g
em
en
t
o
f
h
ea
lth
an
d
s
af
ety
at
wo
r
k
in
co
m
p
an
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
Dete
ctio
n
o
f e
leme
n
ts
o
f p
ers
o
n
a
l sa
fety
fo
r
th
e
p
r
ev
en
tio
n
o
f a
cc
id
en
ts
a
t
…
(
Ma
r
ia
C
la
u
d
ia
B
o
n
f
a
n
te
)
5831
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
is
p
r
o
ject
h
as
b
ee
n
f
in
an
c
ed
with
th
e
r
esear
ch
f
u
n
d
s
s
tip
u
lated
f
o
r
th
e
2
0
2
2
an
n
u
ity
b
y
th
e
Un
iv
er
s
id
ad
d
el
Sin
ú
Secc
io
n
al
C
ar
tag
en
a
C
o
lo
m
b
ia.
RE
F
E
R
E
NC
E
S
[
1
]
C
.
M
o
r
a
i
s,
K
.
L.
Y
u
n
g
,
K
.
Jo
h
n
so
n
,
R
.
M
o
u
r
a
,
M
.
B
e
e
r
,
a
n
d
E
.
P
a
t
e
l
l
i
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
h
u
m
a
n
e
r
r
o
r
s
a
n
d
i
n
f
l
u
e
n
c
i
n
g
f
a
c
t
o
r
s
:
a
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
S
a
f
e
t
y
S
c
i
e
n
c
e
,
v
o
l
.
1
4
6
,
n
o
.
F
e
b
r
u
a
r
y
2
0
2
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
ssc
i
.
2
0
2
1
.
1
0
5
5
2
8
.
[
2
]
A
.
M
.
V
u
k
i
c
e
v
i
c
,
M
.
D
j
a
p
a
n
,
V
.
I
sai
l
o
v
i
c
,
D
.
M
i
l
a
s
i
n
o
v
i
c
,
M
.
S
a
v
k
o
v
i
c
,
a
n
d
P
.
M
i
l
o
se
v
i
c
,
“
G
e
n
e
r
i
c
c
o
mp
l
i
a
n
c
e
o
f
i
n
d
u
st
r
i
a
l
P
P
E
b
y
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
S
a
f
e
t
y
S
c
i
e
n
c
e
,
v
o
l
.
1
4
8
,
n
o
.
J
u
n
e
2
0
2
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
ss
c
i
.
2
0
2
1
.
1
0
5
6
4
6
.
[
3
]
M
.
G
h
a
z
a
l
,
Y
.
A
b
u
H
a
e
y
e
h
,
A
.
A
b
e
d
,
a
n
d
S
.
G
h
a
z
a
l
,
“
E
mb
e
d
d
e
d
f
a
t
i
g
u
e
d
e
t
e
c
t
i
o
n
u
s
i
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
w
i
t
h
mo
b
i
l
e
i
n
t
e
g
r
a
t
i
o
n
,
”
Pr
o
c
e
e
d
i
n
g
s
-
2
0
1
8
I
EEE
6
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
F
u
t
u
r
e
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
a
n
d
C
l
o
u
d
W
o
rks
h
o
p
s
,
W
-
Fi
C
l
o
u
d
2
0
1
8
,
p
p
.
1
2
9
–
1
3
3
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
W
-
F
i
C
l
o
u
d
.
2
0
1
8
.
0
0
0
2
6
.
[
4
]
P
.
Y
a
d
a
v
,
N
.
G
u
p
t
a
,
a
n
d
P
.
K
.
S
h
a
r
ma,
“
A
c
o
m
p
r
e
h
e
n
si
v
e
st
u
d
y
t
o
w
a
r
d
s
h
i
g
h
-
l
e
v
e
l
a
p
p
r
o
a
c
h
e
s
f
o
r
w
e
a
p
o
n
d
e
t
e
c
t
i
o
n
u
s
i
n
g
c
l
a
ss
i
c
a
l
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
,
”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
1
2
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
2
.
1
1
8
6
9
8
.
[
5
]
D
.
P
.
Le
st
a
r
i
a
n
d
R
.
K
o
sas
i
h
,
“
C
o
m
p
a
r
i
s
o
n
o
f
t
w
o
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
f
o
r
d
e
t
e
c
t
i
n
g
f
i
r
e
h
o
t
sp
o
t
s,”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
2
,
n
o
.
3
,
p
p
.
3
1
1
8
–
3
1
2
8
,
J
u
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
2
i
3
.
p
p
3
1
1
8
-
3
1
2
8
.
[
6
]
R
.
K
h
a
l
l
a
f
a
n
d
M
.
K
h
a
l
l
a
f
,
“
A
u
t
o
ma
t
i
o
n
i
n
c
o
n
s
t
r
u
c
t
i
o
n
c
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
a
n
a
l
y
s
i
s o
f
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
l
i
c
a
t
i
o
n
s
i
n
c
o
n
s
t
r
u
c
t
i
o
n
:
a
sy
st
e
ma
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
Au
t
o
m
a
t
i
o
n
i
n
C
o
n
st
ru
c
t
i
o
n
,
v
o
l
.
1
2
9
,
n
o
.
M
a
y
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
u
t
c
o
n
.
2
0
2
1
.
1
0
3
7
6
0
.
[
7
]
P
.
K
u
mar
a
n
d
G
.
G
o
sw
a
m
i
,
“
A
c
o
mp
r
e
h
e
n
si
v
e
r
e
v
i
e
w
o
n
r
e
a
l
t
i
me
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
,
”
2
0
2
2
1
1
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
y
s
t
e
m
Mo
d
e
l
i
n
g
&
A
d
v
a
n
c
e
m
e
n
t
i
n
Re
se
a
rc
h
T
re
n
d
s
(
S
MA
RT)
,
p
p
.
1
4
9
9
–
1
5
0
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
S
M
A
R
T
5
5
8
2
9
.
2
0
2
2
.
1
0
0
4
6
9
7
2
.
[
8
]
W.
-
C
.
C
h
e
r
n
,
J.
H
y
e
o
n
,
T.
V
N
g
u
y
e
n
,
V
.
K
.
A
sar
i
,
a
n
d
H
.
K
i
m,
“
C
o
n
t
e
x
t
-
a
w
a
r
e
s
a
f
e
t
y
a
ssessm
e
n
t
s
y
s
t
e
m
f
o
r
f
a
r
-
f
i
e
l
d
mo
n
i
t
o
r
i
n
g
,
”
A
u
t
o
m
a
t
i
o
n
i
n
C
o
n
s
t
r
u
c
t
i
o
n
,
v
o
l
.
1
4
9
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
u
t
c
o
n
.
2
0
2
3
.
1
0
4
7
7
9
.
[
9
]
M
.
B
o
u
k
a
b
o
u
s
a
n
d
M
.
A
z
i
z
i
,
“
I
mag
e
a
n
d
v
i
d
e
o
-
b
a
se
d
c
r
i
me
p
r
e
d
i
c
t
i
o
n
u
si
n
g
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
,
”
Bu
l
l
e
t
i
n
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
3
,
p
p
.
1
6
3
0
–
1
6
3
8
,
Ju
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
2
i
3
.
5
1
5
7
.
[
1
0
]
R
.
A
l
t
u
r
k
i
,
M
.
A
l
h
a
r
b
i
,
F
.
A
l
a
n
z
i
,
a
n
d
S
.
A
l
b
a
h
l
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
d
e
t
e
c
t
i
n
g
a
n
d
r
e
c
o
g
n
i
z
i
n
g
f
a
c
e
mask
s
:
a
su
r
v
e
y
,
”
Fro
n
t
i
e
rs
i
n
P
u
b
l
i
c
H
e
a
l
t
h
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
u
b
h
.
2
0
2
2
.
9
5
5
3
3
2
.
[
1
1
]
V
i
b
h
u
t
i
,
N
.
J
i
n
d
a
l
,
H
.
S
i
n
g
h
,
a
n
d
P
.
S
.
R
a
n
a
,
“
F
a
c
e
m
a
s
k
d
e
t
e
c
t
i
o
n
i
n
C
O
V
I
D
-
1
9
:
a
st
r
a
t
e
g
i
c
r
e
v
i
e
w
,
”
M
u
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
1
,
n
o
.
2
8
,
p
p
.
4
0
0
1
3
–
4
0
0
4
2
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
2
-
1
2
9
9
9
-
6.
[
1
2
]
L.
A
z
i
z
,
M
.
S
.
B
i
n
H
a
j
i
S
a
l
a
m,
U
.
U
.
S
h
e
i
k
h
,
a
n
d
S
.
A
y
u
b
,
“
E
x
p
l
o
r
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
se
d
a
r
c
h
i
t
e
c
t
u
r
e
,
s
t
r
a
t
e
g
i
e
s,
a
p
p
l
i
c
a
t
i
o
n
s
a
n
d
c
u
r
r
e
n
t
t
r
e
n
d
s
i
n
g
e
n
e
r
i
c
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
:
a
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
8
,
p
p
.
1
7
0
4
6
1
–
1
7
0
4
9
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
3
0
2
1
5
0
8
.
[
1
3
]
J.
R
e
d
m
o
n
,
S
.
D
i
v
v
a
l
a
,
R
.
G
i
r
sh
i
c
k
,
a
n
d
A
.
F
a
r
h
a
d
i
,
“
Y
o
u
o
n
l
y
l
o
o
k
o
n
c
e
:
u
n
i
f
i
e
d
,
r
e
a
l
-
t
i
me
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
,
”
i
n
2
0
1
6
I
E
E
E
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
a
n
d
Pa
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
(
C
VP
R)
,
J
u
n
.
2
0
1
6
,
p
p
.
7
7
9
–
7
8
8
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
6
.
9
1
.
[
1
4
]
M
.
G
u
e
r
r
i
e
r
i
a
n
d
G
.
P
a
r
l
a
,
“
R
e
a
l
-
t
i
me
s
o
c
i
a
l
d
i
st
a
n
c
e
m
e
a
s
u
r
e
me
n
t
a
n
d
f
a
c
e
mas
k
d
e
t
e
c
t
i
o
n
i
n
p
u
b
l
i
c
t
r
a
n
s
p
o
r
t
a
t
i
o
n
s
y
st
e
ms
d
u
r
i
n
g
t
h
e
C
O
V
I
D
-
1
9
p
a
n
d
e
m
i
c
a
n
d
p
o
s
t
-
p
a
n
d
e
mi
c
e
r
a
:
T
h
e
o
r
e
t
i
c
a
l
a
p
p
r
o
a
c
h
a
n
d
c
a
s
e
s
t
u
d
y
i
n
I
t
a
l
y
,
”
T
ra
n
s
p
o
rt
a
t
i
o
n
R
e
se
a
rc
h
I
n
t
e
r
d
i
sci
p
l
i
n
a
r
y
P
e
rs
p
e
c
t
i
v
e
s
,
v
o
l
.
1
6
,
n
o
.
Ju
n
e
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
r
i
p
.
2
0
2
2
.
1
0
0
6
9
3
.
[
1
5
]
M
.
V
.
R
a
m
a
d
h
a
n
e
t
a
l
.
,
“
C
o
m
p
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
o
f
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
f
o
r
d
e
t
e
c
t
i
n
g
f
a
c
e
m
a
sk
mas
k
,
”
i
n
7
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
2
0
2
3
,
v
o
l
.
0
0
,
n
o
.
2
0
2
2
.
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
2
.
1
2
.
1
1
0
.
[
1
6
]
I
.
Jav
e
d
e
t
a
l
.
,
“
F
a
c
e
m
a
s
k
d
e
t
e
c
t
i
o
n
a
n
d
s
o
c
i
a
l
d
i
st
a
n
c
e
m
o
n
i
t
o
r
i
n
g
s
y
st
e
m
f
o
r
C
O
V
I
D
-
1
9
p
a
n
d
e
mi
c
,
”
Mu
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
2
,
n
o
.
9
,
p
p
.
1
4
1
3
5
–
1
4
1
5
2
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
2
-
1
3
9
1
3
-
w.
[
1
7
]
S
.
S
e
t
h
i
,
M
.
K
a
t
h
u
r
i
a
,
a
n
d
T.
K
a
u
sh
i
k
,
“
F
a
c
e
mas
k
d
e
t
e
c
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
:
a
n
a
p
p
r
o
a
c
h
t
o
r
e
d
u
c
e
r
i
s
k
o
f
C
o
r
o
n
a
v
i
r
u
s
sp
r
e
a
d
,
”
J
o
u
r
n
a
l
o
f
Bi
o
m
e
d
i
c
a
l
I
n
f
o
r
m
a
t
i
c
s
,
v
o
l
.
1
2
0
,
A
u
g
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
i
.
2
0
2
1
.
1
0
3
8
4
8
.
[
1
8
]
M
.
L
o
e
y
,
G
.
M
a
n
o
g
a
r
a
n
,
M
.
H
a
me
d
,
a
n
d
N
.
Ta
h
a
,
“
F
i
g
h
t
i
n
g
a
g
a
i
n
st
C
O
V
I
D
-
1
9
:
a
n
o
v
e
l
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
b
a
sed
o
n
Y
O
LO
-
v2
w
i
t
h
R
e
sN
e
t
-
5
0
f
o
r
me
d
i
c
a
l
f
a
c
e
mask
d
e
t
e
c
t
i
o
n
,
”
S
u
s
t
a
i
n
a
b
l
e
C
i
t
i
e
s
a
n
d
S
o
c
i
e
t
y
,
v
o
l
.
6
5
,
n
o
.
Ju
n
e
2
0
2
0
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
scs.
2
0
2
0
.
1
0
2
6
0
0
.
[
1
9
]
N
.
K
.
A
n
u
s
h
k
a
n
n
a
n
,
V
.
R
.
K
u
mb
h
a
r
,
S
.
K
.
M
a
d
d
i
l
a
,
C
.
S
.
K
o
l
l
i
,
B
.
V
i
d
h
y
a
,
a
n
d
R
.
G
.
V
i
d
h
y
a
,
“
Y
O
LO
a
l
g
o
r
i
t
h
m
f
o
r
h
e
l
me
t
d
e
t
e
c
t
i
o
n
i
n
i
n
d
u
st
r
i
e
s
f
o
r
saf
e
t
y
p
u
r
p
o
se,
”
i
n
3
r
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
m
a
rt
El
e
c
t
r
o
n
i
c
s
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
,
I
C
O
S
EC
2
0
2
2
-
Pr
o
c
e
e
d
i
n
g
s
,
2
0
2
2
,
n
o
.
I
c
o
se
c
,
p
p
.
2
2
5
–
2
3
0
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
O
S
EC
5
4
9
2
1
.
2
0
2
2
.
9
9
5
2
1
5
4
.
[
2
0
]
N
.
D
.
N
a
t
h
,
A
.
H
.
B
e
h
z
a
d
a
n
,
a
n
d
S
.
G
.
P
a
a
l
,
“
A
u
t
o
m
a
t
i
o
n
i
n
c
o
n
st
r
u
c
t
i
o
n
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
si
t
e
sa
f
e
t
y
:
r
e
a
l
-
t
i
me
d
e
t
e
c
t
i
o
n
o
f
p
e
r
so
n
a
l
p
r
o
t
e
c
t
i
v
e
e
q
u
i
p
me
n
t
,
”
Au
t
o
m
a
t
i
o
n
i
n
C
o
n
st
ru
c
t
i
o
n
,
v
o
l
.
1
1
2
,
n
o
.
Ju
l
y
2
0
1
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
u
t
c
o
n
.
2
0
2
0
.
1
0
3
0
8
5
.
[
2
1
]
J.
A
h
n
,
J
.
P
a
r
k
,
S
.
S
i
k
,
K
.
L
e
e
,
H
.
D
o
,
a
n
d
J.
K
o
,
“
S
a
f
e
F
a
c
:
v
i
d
e
o
-
b
a
s
e
d
smar
t
safe
t
y
m
o
n
i
t
o
r
i
n
g
f
o
r
p
r
e
v
e
n
t
i
n
g
i
n
d
u
st
r
i
a
l
w
o
r
k
a
c
c
i
d
e
n
t
s,
”
E
x
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
.
2
1
5
,
n
o
.
N
o
v
e
m
b
e
r
2
0
2
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
2
.
1
1
9
3
9
7
.
[
2
2
]
S
.
C
h
e
n
,
K
.
D
e
ma
c
h
i
,
a
n
d
F
.
D
o
n
g
,
“
A
u
t
o
m
a
t
i
o
n
i
n
C
o
n
s
t
r
u
c
t
i
o
n
G
r
a
p
h
-
b
a
s
e
d
l
i
n
g
u
i
st
i
c
a
n
d
v
i
s
u
a
l
i
n
f
o
r
mat
i
o
n
i
n
t
e
g
r
a
t
i
o
n
f
o
r
on
-
s
i
t
e
o
c
c
u
p
a
t
i
o
n
a
l
h
a
z
a
r
d
s
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
Au
t
o
m
a
t
i
o
n
i
n
C
o
n
st
r
u
c
t
i
o
n
,
v
o
l
.
1
3
7
,
n
o
.
M
a
r
c
h
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
u
t
c
o
n
.
2
0
2
2
.
1
0
4
1
9
1
.
[
2
3
]
K
.
Li
,
L.
Q
i
n
,
Q
.
Li
,
F
.
Z
h
a
o
,
Z
.
X
u
,
a
n
d
K
.
L
i
u
,
“
I
mp
r
o
v
e
d
e
d
g
e
l
i
g
h
t
w
e
i
g
h
t
Y
O
LO
v
4
a
n
d
i
t
s
a
p
p
l
i
c
a
t
i
o
n
i
n
o
n
-
si
t
e
p
o
w
e
r
s
y
s
t
e
m
w
o
r
k
,
”
G
l
o
b
a
l
E
n
e
rg
y
I
n
t
e
r
c
o
n
n
e
c
t
i
o
n
,
v
o
l
.
5
,
n
o
.
2
,
p
p
.
1
6
8
–
1
8
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
g
l
o
e
i
.
2
0
2
2
.
0
4
.
0
1
4
.
[
2
4
]
G
.
G
a
l
l
o
,
F
.
D
I
R
i
e
n
z
o
,
P
.
D
u
c
a
n
g
e
,
V
.
F
e
r
r
a
r
i
,
A
.
T
o
g
n
e
t
t
i
,
a
n
d
C
.
V
a
l
l
a
t
i
,
“
sm
a
r
t
sy
s
t
e
m
f
o
r
p
e
r
so
n
a
l
p
r
o
t
e
c
t
i
v
e
e
q
u
i
p
m
e
n
t
d
e
t
e
c
t
i
o
n
i
n
i
n
d
u
st
r
i
a
l
e
n
v
i
r
o
n
m
e
n
t
s
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
,
”
Pro
c
e
e
d
i
n
g
s
-
2
0
2
1
I
EEE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
m
a
r
t
C
o
m
p
u
t
i
n
g
,
S
M
ARTCO
MP
2
0
2
1
,
p
p
.
2
2
2
–
2
2
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
S
M
A
R
T
C
O
M
P
5
2
4
1
3
.
2
0
2
1
.
0
0
0
5
1
.
[
2
5
]
B
.
Z
h
a
n
g
,
C
.
-
F
.
S
u
n
,
S
.
-
Q
.
F
a
n
g
,
Y
.
-
H
.
Zh
a
o
,
a
n
d
S
.
S
u
,
“
W
o
r
k
s
h
o
p
s
a
f
e
t
y
h
e
l
m
e
t
w
e
a
r
i
n
g
d
e
t
e
c
t
i
o
n
m
o
d
e
l
b
a
s
e
d
o
n
S
C
M
-
Y
O
LO
,
”
S
e
n
so
rs
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
2
1
7
6
7
0
2
.
[
2
6
]
A
.
K
u
mar
,
A
.
K
a
l
i
a
,
A
.
S
h
a
r
ma
,
a
n
d
M
.
K
a
u
s
h
a
l
,
“
A
h
y
b
r
i
d
t
i
n
y
Y
O
LO
v
4
‑
S
P
P
m
o
d
u
l
e
b
a
s
e
d
i
m
p
r
o
v
e
d
f
a
c
e
mas
k
d
e
t
e
c
t
i
o
n
v
i
s
i
o
n
s
y
s
t
e
m,
”
J
o
u
r
n
a
l
o
f
Am
b
i
e
n
t
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
H
u
m
a
n
i
ze
d
C
o
m
p
u
t
i
n
g
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
2
6
5
2
-
021
-
0
3
5
4
1
-
x.
[
2
7
]
Z.
Z
h
a
o
e
t
a
l
.
,
“
S
A
I
-
Y
O
LO
:
a
l
i
g
h
t
w
e
i
g
h
t
n
e
t
w
o
r
k
f
o
r
r
e
a
l
-
t
i
m
e
d
e
t
e
c
t
i
o
n
o
f
d
r
i
v
e
r
mas
k
-
w
e
a
r
i
n
g
s
p
e
c
i
f
i
c
a
t
i
o
n
o
n
r
e
s
o
u
r
c
e
-
c
o
n
st
r
a
i
n
e
d
d
e
v
i
c
e
s,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
N
e
u
ro
s
c
i
e
n
c
e
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
4
5
2
9
1
0
7
.
[
2
8
]
Z.
H
a
n
,
H
.
H
u
a
n
g
,
Q
.
F
a
n
,
Y
.
L
i
,
Y
.
L
i
,
a
n
d
X
.
C
h
e
n
,
“
S
M
D
-
Y
O
LO
:
a
n
e
f
f
i
c
i
e
n
t
a
n
d
l
i
g
h
t
w
e
i
g
h
t
d
e
t
e
c
t
i
o
n
met
h
o
d
f
o
r
mas
k
w
e
a
r
i
n
g
st
a
t
u
s
d
u
r
i
n
g
t
h
e
C
O
V
I
D
-
1
9
p
a
n
d
e
mi
c
,
”
C
o
m
p
u
t
e
r
Me
t
h
o
d
s
a
n
d
Pro
g
r
a
m
s
i
n
Bi
o
m
e
d
i
c
i
n
e
,
v
o
l
.
2
2
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
m
p
b
.
2
0
2
2
.
1
0
6
8
8
8
.
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.
1
4
,
No
.
5
,
Octo
b
e
r
2
0
2
4
:
5
8
2
4
-
5
8
3
3
5832
[
2
9
]
M
.
M
.
A
.
J
o
y
e
t
a
l
.
,
“
C
o
n
t
a
c
t
l
e
ss
s
u
r
v
e
i
l
l
a
n
c
e
f
o
r
p
r
e
v
e
n
t
i
n
g
w
i
n
d
-
b
o
r
n
e
d
i
se
a
se
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
3
,
n
o
.
1
1
,
p
p
.
7
7
5
–
7
8
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
2
.
0
1
3
1
1
9
0
.
[
3
0
]
M
.
S
a
d
i
q
,
S
.
M
a
so
o
d
,
a
n
d
O
.
P
a
l
,
“
F
D
-
Y
O
L
O
v
5
:
a
f
u
z
z
y
i
mag
e
e
n
h
a
n
c
e
men
t
b
a
sed
r
o
b
u
st
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
mo
d
e
l
f
o
r
sa
f
e
t
y
h
e
l
met
d
e
t
e
c
t
i
o
n
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Fu
zzy
S
y
st
e
m
s
,
v
o
l
.
2
4
,
n
o
.
5
,
p
p
.
2
6
0
0
–
2
6
1
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s4
0
8
1
5
-
022
-
01267
-
2.
[
3
1
]
L.
Z
h
a
o
,
D
.
Zh
a
n
g
,
Y
.
Li
u
,
J.
G
u
o
,
a
n
d
Z.
S
h
i
,
“
I
mp
r
o
v
e
d
Y
O
LO
v
5
s
n
e
t
w
o
r
k
f
o
r
m
u
l
t
i
-
sc
a
l
e
s
a
f
e
t
y
h
e
l
me
t
d
e
t
e
c
t
i
o
n
,
”
1
1
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s,
C
i
r
c
u
i
t
s
a
n
d
S
y
s
t
e
m
s,
I
C
C
C
A
S
2
0
2
2
,
p
p
.
2
6
2
–
2
6
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
A
S
5
5
2
6
6
.
2
0
2
2
.
9
8
2
5
0
3
7
.
[
3
2
]
W
.
C
h
e
n
,
C
.
L
i
,
a
n
d
H
.
G
u
o
,
“
A
l
i
g
h
t
w
e
i
g
h
t
f
a
c
e
-
a
ss
i
st
e
d
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
mo
d
e
l
f
o
r
w
e
l
d
i
n
g
h
e
l
me
t
u
s
e
,
”
Ex
p
e
rt
S
y
s
t
e
m
s
W
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
2
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
1
9
7
6
4
.
[
3
3
]
L.
A
l
i
,
F
.
A
l
n
a
j
j
a
r
,
M
.
M
.
A
.
P
a
r
a
mb
i
l
,
M
.
I
.
Y
o
u
n
e
s,
Z.
I
.
A
b
d
e
l
h
a
l
i
m
,
a
n
d
H
.
A
l
j
a
ssm
i
,
“
D
e
v
e
l
o
p
me
n
t
o
f
Y
O
LO
v
5
-
b
a
s
e
d
r
e
a
l
-
t
i
m
e
smar
t
m
o
n
i
t
o
r
i
n
g
sy
s
t
e
m
f
o
r
i
n
c
r
e
a
si
n
g
l
a
b
s
a
f
e
t
y
a
w
a
r
e
n
e
ss
i
n
e
d
u
c
a
t
i
o
n
a
l
i
n
st
i
t
u
t
i
o
n
s,
”
S
e
n
so
rs
,
v
o
l
.
2
2
,
n
o
.
2
2
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
2
2
2
8
8
2
0
.
[
3
4
]
M
.
O
t
g
o
n
b
o
l
d
e
t
a
l
.
,
“
S
H
E
L5
K
:
a
n
e
x
t
e
n
d
e
d
d
a
t
a
se
t
a
n
d
b
e
n
c
h
mar
k
i
n
g
f
o
r
safet
y
h
e
l
met
d
e
t
e
c
t
i
o
n
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
2
,
p
p
.
1
–
2
3
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
2
0
6
2
3
1
5
.
[
3
5
]
N
.
D
.
T
.
Y
u
n
g
,
W
.
K
.
W
o
n
g
,
F
.
H
.
Ju
w
o
n
o
,
a
n
d
Z.
A
.
S
i
m
,
“
S
a
f
e
t
y
h
e
l
met
d
e
t
e
c
t
i
o
n
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
:
i
mp
l
e
me
n
t
a
t
i
o
n
a
n
d
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
u
si
n
g
Y
O
LO
v
5
,
Y
O
LO
v
6
a
n
d
Y
O
LO
v
7
,
”
i
n
2
0
2
2
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
G
re
e
n
E
n
e
r
g
y
,
C
o
m
p
u
t
i
n
g
a
n
d
S
u
s
t
a
i
n
a
b
l
e
T
e
c
h
n
o
l
o
g
y
(
G
E
C
O
S
T
)
,
O
c
t
.
2
0
2
2
,
p
p
.
1
6
4
–
1
7
0
.
d
o
i
:
1
0
.
1
1
0
9
/
G
EC
O
S
T5
5
6
9
4
.
2
0
2
2
.
1
0
0
1
0
4
9
0
.
[
3
6
]
N
.
K
w
a
k
a
n
d
D
.
K
i
m,
“
D
e
t
e
c
t
i
o
n
o
f
w
o
r
k
e
r
’
s
safe
t
y
h
e
l
m
e
t
a
n
d
m
a
s
k
a
n
d
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
w
o
r
k
e
r
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
C
o
m
p
u
t
e
rs
,
M
a
t
e
r
i
a
l
s &
C
o
n
t
i
n
u
a
,
2
0
2
3
,
d
o
i
:
1
0
.
3
2
6
0
4
/
c
mc.
2
0
2
3
.
0
3
5
7
6
2
.
[
3
7
]
H
.
Z
h
a
n
g
,
N
.
Y
a
n
g
,
H
.
C
h
e
n
,
X
.
L
u
a
n
,
M
.
I
l
y
a
s,
a
n
d
B
.
A
d
ma
d
,
“
S
a
f
e
t
y
h
e
l
me
t
a
n
d
m
a
s
k
d
e
t
e
c
t
i
o
n
i
n
su
b
s
t
a
t
i
o
n
b
a
se
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
,
”
i
n
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
E
l
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
n
t
r
o
l
T
e
c
h
n
o
l
o
g
i
e
s
,
2
0
2
2
,
p
p
.
1
4
5
–
1
5
0
.
d
o
i
:
1
0
.
1
1
0
9
/
C
EEC
T5
5
9
6
0
.
2
0
2
2
.
1
0
0
3
0
6
1
8
.
[
3
8
]
P
.
Zo
u
a
n
d
J.
Z
h
a
n
g
,
“
I
n
t
e
l
l
i
g
e
n
t
h
e
l
me
t
d
e
t
e
c
t
i
o
n
sy
st
e
m
b
a
se
d
o
n
t
h
e
i
mp
r
o
v
e
d
Y
O
LO
v
5
,
”
i
n
2
0
2
2
I
EE
E
5
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
B
i
g
D
a
t
a
,
I
C
AI
BD
2
0
2
2
,
2
0
2
2
,
p
p
.
3
1
0
–
3
1
4
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
I
B
D
5
5
1
2
7
.
2
0
2
2
.
9
8
2
0
4
7
9
.
[
3
9
]
S
.
B
e
r
a
a
n
d
P
.
K
u
mar,
“
F
a
c
e
m
a
sk
d
e
t
e
c
t
i
o
n
:
a
d
e
e
p
l
e
a
r
n
i
n
g
c
o
n
c
e
p
t
,
”
i
n
3
rd
G
l
o
b
a
l
C
o
n
f
e
re
n
c
e
f
o
r
A
d
v
a
n
c
e
m
e
n
t
i
n
T
e
c
h
n
o
l
o
g
y
(
G
C
AT)
,
2
0
2
2
,
p
p
.
4
3
7
–
4
4
1
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
EM
5
4
2
2
1
.
2
0
2
2
.
9
8
5
3
0
1
4
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ma
r
ia
Cl
a
u
d
i
a
Bo
n
f
a
n
te
re
c
e
iv
e
d
a
d
e
g
re
e
in
sy
ste
m
s
e
n
g
i
n
e
e
rin
g
,
sp
e
c
ialist
i
n
sy
ste
m
s
a
u
d
it
in
g
,
m
a
ste
r’s
in
b
u
sin
e
ss
p
ro
c
e
ss
m
a
n
a
g
e
m
e
n
t
a
n
d
a
P
h
D
d
e
g
re
e
in
so
f
twa
re
e
n
g
in
e
e
rin
g
fr
o
m
th
e
Un
i
v
e
rsid
a
d
P
o
n
ti
ficia
d
e
S
a
lam
a
n
c
a
.
He
is
c
u
rre
n
tl
y
a
re
se
a
rc
h
p
ro
fe
ss
o
r
o
f
th
e
Un
i
v
e
rsid
a
d
d
e
l
S
in
ú
in
Ca
rtag
e
n
a
-
C
o
lo
m
b
ia.
Hi
s fi
e
ld
o
f
in
tere
st i
s so
ftwa
re
e
n
g
in
e
e
rin
g
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t:
m
a
ria.b
o
n
fa
n
te@
u
n
isi
n
u
.
e
d
u
.
c
o
.
Iv
a
n
H
e
r
n
a
n
d
e
z
Rui
z
re
c
e
iv
e
d
a
d
e
g
re
e
in
tec
h
n
o
l
o
g
ist
i
n
i
n
f
o
rm
a
ti
o
n
s
y
ste
m
s
a
n
d
c
o
m
p
u
ter
n
e
two
r
k
s
a
n
d
a
d
e
g
re
e
i
n
sy
ste
m
s
e
n
g
i
n
e
e
ri
n
g
p
r
o
fe
ss
io
n
a
l
fro
m
th
e
Un
iv
e
rsid
a
d
d
e
l
S
in
ú
Ca
rtag
e
n
a
.
He
is
c
u
rre
n
tl
y
a
re
q
u
irem
e
n
t
a
n
a
ly
st
o
f
t
h
e
so
ftwa
re
c
o
m
p
a
n
y
S
YSNET
S
A
S
,
a
n
d
h
is
field
o
f
in
tere
st
is
d
a
ta
sc
ien
c
e
a
n
d
m
a
c
h
in
e
lea
rn
i
n
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
i
v
a
n
h
e
rn
a
n
d
e
z
1
2
2
6
@g
m
a
il
.
c
o
m
.
J
u
a
n
C
o
n
tr
e
r
a
s
M
o
n
te
s
r
e
c
e
iv
e
d
a
d
e
g
re
e
i
n
e
lec
tri
c
a
l
e
n
g
i
n
e
e
r
a
n
d
a
s
a
sp
e
c
ialist
in
in
d
u
strial
a
u
t
o
m
a
ti
o
n
.
He
is
P
h
D
fr
o
m
t
h
e
In
st
it
u
t
o
S
u
p
e
rio
r
P
o
li
téc
n
ico
Jo
sé
An
to
n
io
Ech
e
v
e
rría,
in
Ha
v
a
n
a
,
Cu
b
a
.
He
c
u
rre
n
tl
y
tea
c
h
e
s
a
t
th
e
Alm
iran
te
P
a
d
il
la
Na
v
a
l
S
c
h
o
o
l
a
n
d
is
a
lso
li
n
k
e
d
to
In
sti
tu
c
ió
n
Un
i
v
e
rsitaria
d
e
Ba
rra
n
q
u
i
ll
a
a
s
P
ro
jec
t
Dire
c
to
r.
His
field
s
o
f
in
tere
st
in
c
lu
d
e
i
n
telli
g
e
n
t
c
o
n
tr
o
l
sy
ste
m
s
a
n
d
th
e
d
e
v
e
lo
p
m
e
n
t
o
f
fu
z
z
y
m
o
d
e
li
n
g
a
n
d
c
o
n
tro
l
m
e
th
o
d
s
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
j
u
a
n
c
o
n
trera
sm
@itsa
.
e
d
u
.
c
o
.
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
Dete
ctio
n
o
f e
leme
n
ts
o
f p
ers
o
n
a
l sa
fety
fo
r
th
e
p
r
ev
en
tio
n
o
f a
cc
id
en
ts
a
t
…
(
Ma
r
ia
C
la
u
d
ia
B
o
n
f
a
n
te
)
5833
Eu
g
e
n
ia
Ar
r
ieta
Ro
d
r
íg
u
e
z
re
c
e
iv
e
d
a
d
e
g
re
e
in
sy
ste
m
s
e
n
g
in
e
e
r
fr
o
m
Un
iv
e
rsid
a
d
Ra
fa
e
l
Nú
ñ
e
z
a
n
d
M
a
ste
r'
s
d
e
g
re
e
fro
m
Un
iv
e
rsid
a
d
T
e
c
n
o
ló
g
ica
d
e
Bo
lí
v
a
r.
S
h
e
is
c
u
rre
n
tl
y
a
d
o
c
to
ra
l
stu
d
e
n
t
in
i
n
fo
rm
a
ti
o
n
a
n
d
c
o
m
m
u
n
i
c
a
ti
o
n
tec
h
n
o
lo
g
ies
a
t
th
e
Un
iv
e
rsid
a
d
d
e
la
Co
sta
d
e
Co
lo
m
b
ia
a
n
d
is
c
u
rre
n
tl
y
wo
r
k
i
n
g
o
n
h
is
d
o
c
t
o
ra
te.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
e
u
g
e
n
iaa
rrieta
ro
d
ri
g
u
e
z
@g
m
a
il
.
c
o
m
.
Ale
ja
n
d
r
o
C
a
m
a
-
Pi
n
t
o
re
c
e
iv
e
d
a
d
e
g
re
e
i
n
in
d
u
str
ial
a
n
d
e
n
v
ir
o
n
m
e
n
tal
e
n
g
in
e
e
r
fro
m
th
e
Un
iv
e
rsity
o
f
Alm
e
ria
S
p
a
in
;
h
e
tea
c
h
e
s
a
t
th
e
F
a
c
u
lt
y
o
f
E
n
g
in
e
e
rin
g
a
n
d
th
e
P
h
D
in
i
n
fo
rm
a
ti
o
n
a
n
d
c
o
m
m
u
n
ica
ti
o
n
tec
h
n
o
lo
g
ies
in
th
e
Un
iv
e
rsid
a
d
d
e
la
C
o
sta
in
Ba
rra
n
q
u
il
la
Co
l
o
m
b
ia.
His
to
p
i
c
s
o
f
in
tere
st
a
re
:
wire
les
s
c
o
m
m
u
n
ica
ti
o
n
s,
wire
les
s
se
n
s
o
r
n
e
two
rk
,
n
e
tw
o
rk
a
rc
h
it
e
c
tu
re
,
c
o
m
m
u
n
ica
ti
o
n
e
n
g
i
n
e
e
rin
g
,
in
tern
e
t
o
f
th
i
n
g
s,
a
n
d
e
n
v
iro
n
m
e
n
tal
sc
ien
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
c
a
m
a
1
@c
u
c
.
e
d
u
.
c
o
.
Evaluation Warning : The document was created with Spire.PDF for Python.