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.
16
,
No
.
5
,
Octo
b
er
20
26
,
p
p
.
2
7
5
0
~
2
7
6
8
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
16
i
5
.
pp
2
7
5
0
-
2
7
6
8
2750
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Rea
l
-
time
multim
o
da
l f
a
tigue de
tec
tion usin
g
f
a
cia
l v
isio
n and
a
lert
int
eg
ra
tion
v
ia
ESP32
f
o
r occ
upa
tiona
l health
a
pplica
tions
Andrés
E
nri
qu
e
Ro
j
a
s
P
rim
o
,
Alf
re
do
L
a
za
ro
G
utie
rr
ez
,
F
elix
P
ucuh
ua
y
la
-
Rev
a
t
t
a
D
e
p
a
r
t
a
m
e
n
t
o
d
e
I
n
g
e
n
i
e
r
í
a
E
l
e
c
t
r
ó
n
i
c
a
,
F
a
c
u
l
t
a
d
d
e
I
n
g
e
n
i
e
r
í
a
,
U
n
i
v
e
r
si
d
a
d
T
e
c
n
o
l
ó
g
i
c
a
d
e
l
P
e
r
ú
,
L
i
ma
,
P
e
r
ú
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
2
1
,
2
0
2
5
R
ev
is
ed
Ma
y
6
,
2
0
2
6
Acc
ep
ted
J
u
l 2
2
,
2
0
2
6
Early
d
e
tec
ti
o
n
o
f
wo
r
k
fa
ti
g
u
e
is a ma
jo
r
c
h
a
ll
e
n
g
e
in
in
d
u
strial
se
tt
in
g
s
d
u
e
to
t
h
e
lac
k
o
f
n
o
n
-
in
v
a
siv
e
,
a
c
c
e
ss
ib
le,
a
n
d
l
o
w
-
c
o
st
s
y
ste
m
s
c
a
p
a
b
le
o
f
o
p
e
ra
ti
n
g
i
n
re
a
l
ti
m
e
.
In
t
h
is
c
o
n
tex
t,
th
is
re
se
a
rc
h
p
ro
p
o
se
s
a
m
u
lt
imo
d
a
l
re
a
l
-
ti
m
e
fa
ti
g
u
e
d
e
tec
ti
o
n
sy
ste
m
u
sin
g
fa
c
ial
v
isi
o
n
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
,
a
ime
d
a
t
risk
p
re
v
e
n
ti
o
n
a
n
d
p
r
o
m
o
ti
n
g
o
c
c
u
p
a
ti
o
n
a
l
h
e
a
lt
h
.
Th
e
sy
ste
m
in
teg
ra
tes
g
e
o
m
e
tri
c
a
n
d
b
e
h
a
v
io
ra
l
p
a
ra
m
e
ters
,
su
c
h
a
s
e
y
e
a
sp
e
c
t
ra
ti
o
(EAR),
h
e
a
d
ti
lt
,
a
n
d
m
o
u
th
o
p
e
n
i
n
g
,
p
ro
c
e
ss
e
d
o
n
a
Ra
sp
b
e
rry
P
i
5
u
sin
g
M
e
d
iaPi
p
e
a
n
d
a
h
y
b
rid
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
(CNN
)
M
o
b
i
leV
iT
m
o
d
e
l.
Visu
a
l
a
n
d
a
u
d
ib
le
a
lerts
a
re
m
a
n
a
g
e
d
b
y
a
n
ES
P
3
2
m
icro
c
o
n
tro
ll
e
r
u
sin
g
t
h
e
m
e
ss
a
g
e
q
u
e
u
i
n
g
tele
m
e
try
tran
s
p
o
rt
(
M
QTT
)
p
ro
t
o
c
o
l,
w
h
il
e
a
g
ra
p
h
ica
l
in
terf
a
c
e
d
e
v
e
lo
p
e
d
i
n
Tk
i
n
ter
a
ll
o
ws
re
a
l
-
ti
m
e
m
o
n
it
o
r
in
g
o
f
o
p
e
ra
to
r
sta
t
u
s.
Ex
p
e
rime
n
tal
re
su
lt
s,
e
v
a
l
u
a
ted
in
a
sim
u
late
d
wo
rk
e
n
v
ir
o
n
m
e
n
t
u
si
n
g
AI
-
g
e
n
e
ra
ted
sy
n
t
h
e
ti
c
v
id
e
o
s
,
sh
o
w
a
n
a
c
c
u
ra
c
y
g
re
a
ter
th
a
n
9
7
%
a
n
d
a
late
n
c
y
o
f
les
s
th
a
n
2
5
0
m
s
,
c
o
n
firmi
n
g
t
h
e
sy
ste
m
'
s
e
ffe
c
ti
v
e
n
e
ss
in
t
h
e
e
a
rly
d
e
tec
ti
o
n
o
f
si
g
n
s
o
f
d
ro
ws
i
n
e
ss
a
n
d
a
tt
e
n
ti
o
n
d
e
ficit.
I
n
c
o
n
c
lu
si
o
n
,
t
h
e
p
ro
p
o
s
a
l
re
p
re
se
n
ts
a
n
o
n
-
i
n
v
a
siv
e
,
sc
a
lab
le,
a
n
d
e
fficie
n
t
so
lu
ti
o
n
th
a
t
c
o
m
b
i
n
e
s
c
o
m
p
u
ter
v
isi
o
n
,
d
e
e
p
lea
rn
i
n
g
,
a
n
d
t
h
e
In
tern
e
t
o
f
Th
in
g
s (Io
T)
to
stre
n
g
t
h
e
n
wo
r
k
p
lac
e
sa
fe
ty
a
n
d
we
ll
-
b
e
in
g
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
in
tellig
en
ce
E
SP
3
2
m
icr
o
co
n
t
r
o
ller
Fatig
u
e
d
etec
tio
n
I
n
ter
n
et
o
f
th
in
g
s
Occ
u
p
atio
n
al
h
ea
lth
R
asp
b
er
r
y
Pi 5
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
:
An
d
r
és E
n
r
iq
u
e
R
o
jas Pr
im
o
Dep
ar
tam
en
to
d
e
I
n
g
en
ier
ía
E
lectr
ó
n
ica,
Facu
ltad
d
e
I
n
g
e
n
i
er
ía,
Un
iv
er
s
id
ad
T
ec
n
o
ló
g
ica
d
el
Per
ú
Av
.
So
l,
B
lo
ck
2
,
San
J
u
an
d
e
L
u
r
ig
an
c
h
o
,
L
im
a,
Per
ú
E
m
ail: U
2
0
2
3
1
6
6
@
u
tp
.
ed
u
.
p
e
1.
I
NT
RO
D
UCT
I
O
N
Fatig
u
e
is
r
ec
o
g
n
ized
as
a
co
m
p
lex
p
h
en
o
m
en
o
n
i
n
v
o
lv
i
n
g
b
o
th
p
h
y
s
io
lo
g
ical
an
d
p
s
y
ch
o
lo
g
ical
ex
h
au
s
tio
n
,
w
h
ich
d
ir
ec
tly
af
f
ec
ts
p
eo
p
le'
s
wo
r
k
p
er
f
o
r
m
an
ce
an
d
well
-
b
ein
g
[
1
]
,
[
2
]
.
I
ts
im
p
ac
t
is
wid
esp
r
ea
d
:
it
is
esti
m
ated
th
a
t
th
e
g
lo
b
al
p
r
e
v
alen
ce
r
ea
ch
e
s
2
4
.
2
%
f
o
r
g
en
er
al
f
atig
u
e
a
n
d
7
.
7
%
f
o
r
ch
r
o
n
ic
f
atig
u
e,
with
a
g
r
ea
ter
im
p
ac
t
o
n
wo
m
en
[
3
]
.
On
a
s
o
cio
ec
o
n
o
m
ic
lev
el,
its
ef
f
ec
ts
a
r
e
e
v
id
en
t
in
h
ig
h
-
r
is
k
s
ec
to
r
s
:
th
e
W
o
r
ld
Hea
lth
Or
g
an
izatio
n
(
W
HO)
esti
m
ates
th
at
f
atig
u
e
co
n
t
r
ib
u
tes
s
ig
n
if
ican
tly
to
th
e
1
.
2
5
m
illi
o
n
an
n
u
al
d
ea
th
s
f
r
o
m
tr
af
f
ic
ac
cid
en
ts
[
4
]
,
in
ad
d
itio
n
to
co
m
p
r
o
m
is
in
g
s
af
ety
in
in
d
u
s
tr
ies
s
u
ch
as
co
n
s
tr
u
ctio
n
,
tr
a
n
s
p
o
r
tatio
n
,
a
n
d
m
in
in
g
[
5
]
.
I
n
th
is
s
en
s
e,
f
atig
u
e
n
o
t
o
n
l
y
af
f
ec
ts
th
e
h
ea
lth
o
f
wo
r
k
er
s
,
b
u
t
also
p
u
b
lic
s
af
ety
an
d
in
d
u
s
tr
ial
p
r
o
ce
s
s
es,
wh
er
e
alter
n
ativ
e
s
o
lu
tio
n
s
ar
e
ev
en
b
ein
g
d
eb
a
ted
[
2
]
.
I
n
th
e
in
d
u
s
tr
ial
s
ec
to
r
,
f
atig
u
e
h
as
b
ee
n
id
en
tifie
d
as
o
n
e
o
f
th
e
m
ain
ca
u
s
es
o
f
h
u
m
an
er
r
o
r
[
6
]
.
Var
io
u
s
s
tu
d
ies
ag
r
ee
th
at
b
et
wee
n
3
0
%
an
d
9
0
%
o
f
s
er
io
u
s
in
cid
en
ts
ar
e
d
ir
ec
tly
r
elate
d
to
wo
r
k
er
f
atig
u
e,
co
n
f
ir
m
in
g
th
at
th
is
f
ac
to
r
co
m
p
r
o
m
is
es
o
p
er
atio
n
al
s
af
ety
in
m
u
ltip
le
s
ec
to
r
s
[
6
]
,
[
7
]
.
Fu
r
th
er
m
o
r
e
,
m
ed
ical
liter
atu
r
e
r
an
k
s
am
o
n
g
th
e
to
p
f
iv
e
ca
u
s
es
o
f
m
ed
ical
co
n
s
u
ltatio
n
,
ass
o
ciate
d
with
s
e
r
io
u
s
illn
ess
es
an
d
f
u
n
ctio
n
al
im
p
air
m
en
t
[
3
]
.
I
t
s
co
n
s
eq
u
en
ce
s
g
o
b
e
y
o
n
d
p
h
y
s
ical
h
ea
lth
:
it
is
ass
o
ciate
d
with
d
ep
r
ess
io
n
,
an
x
iety
,
a
n
d
lo
s
s
o
f
p
r
o
d
u
ctiv
ity
[
8
]
,
ev
e
n
af
f
ec
tin
g
s
p
ec
i
f
ic
p
o
p
u
latio
n
s
s
u
ch
as
f
ar
m
er
s
,
wh
er
e
alm
o
s
t
a
q
u
ar
ter
r
e
p
o
r
t
e
x
h
au
s
tio
n
a
n
d
h
alf
s
u
f
f
er
f
r
o
m
s
leep
p
r
o
b
lem
s
[
9
]
.
I
n
in
d
u
s
tr
ial
wo
r
k
er
s
,
th
e
im
p
ac
t
tr
an
s
lates
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
R
ea
l
-
time
mu
ltimo
d
a
l fa
tig
u
e
d
etec
tio
n
u
s
in
g
fa
ci
a
l v
is
io
n
a
n
d
a
lert
…
(
A
n
d
r
és E
n
r
iq
u
e
R
o
ja
s
P
r
imo
)
2751
in
to
m
u
s
cu
lo
s
k
eleta
l
p
ain
,
a
b
s
en
teeism
,
an
d
lo
wer
jo
b
s
ati
s
f
ac
tio
n
[
1
0
]
.
C
o
n
tr
o
l
r
o
o
m
o
p
er
ato
r
s
,
wh
o
ar
e
s
u
b
jecte
d
to
r
ep
etitiv
e
a
n
d
c
o
g
n
itiv
ely
d
em
an
d
in
g
task
s
,
ar
e
h
ig
h
ly
v
u
ln
er
a
b
le
to
m
e
n
tal
f
atig
u
e,
wh
ich
r
ed
u
ce
s
th
eir
ab
ilit
y
to
co
n
ce
n
tr
ate
an
d
r
esp
o
n
d
[
1
1
]
.
Sim
ilar
ly
,
in
cr
itical
s
ec
to
r
s
s
u
ch
as
n
u
clea
r
p
o
wer
p
lan
ts
,
o
p
er
ato
r
f
atig
u
e
h
as b
e
en
lin
k
ed
t
o
s
er
io
u
s
er
r
o
r
s
with
p
o
ten
tially
ca
tast
r
o
p
h
ic
co
n
s
eq
u
en
ce
s
[
1
2
]
.
At
th
e
n
atio
n
al
lev
el,
s
tu
d
ies
o
n
f
atig
u
e
at
wo
r
k
ar
e
s
till
s
ca
r
ce
,
b
u
t
a
r
ec
e
n
t
s
tu
d
y
r
e
v
ea
led
th
at
u
n
iv
er
s
ity
s
tu
d
en
ts
ex
p
er
ien
ce
d
in
cr
ea
s
ed
lev
els
o
f
f
atig
u
e
d
u
r
in
g
th
e
p
a
n
d
em
ic
d
u
e
to
p
r
o
lo
n
g
ed
ex
p
o
s
u
r
e
to
o
n
lin
e
class
es,
wh
ich
led
to
a
d
ec
lin
e
in
th
eir
ac
ad
em
ic
p
er
f
o
r
m
a
n
ce
[
1
3
]
.
T
h
is
g
ap
h
ig
h
lig
h
ts
th
e
n
ee
d
to
im
p
lem
en
t
n
o
n
-
in
v
asiv
e
tech
n
o
lo
g
ies
f
o
r
h
ea
lth
m
o
n
ito
r
in
g
in
t
h
e
wo
r
k
p
lace
,
wh
ich
li
m
its
th
e
ab
ilit
y
t
o
d
etec
t r
is
k
s
ass
o
ciate
d
with
p
h
y
s
ical
an
d
m
en
tal
ex
h
au
s
tio
n
e
ar
ly
o
n
.
T
h
e
p
r
o
b
lem
a
d
d
r
ess
ed
in
t
h
is
s
tu
d
y
is
th
e
lack
o
f
f
atig
u
e
d
etec
tio
n
s
y
s
tem
s
th
at
ar
e
n
o
n
-
in
v
asiv
e,
ac
ce
s
s
ib
le,
lo
w
-
co
s
t,
an
d
ca
p
ab
le
o
f
o
p
er
atin
g
au
t
o
m
atica
lly
in
r
ea
l
tim
e
wi
th
in
wo
r
k
en
v
ir
o
n
m
en
ts
.
R
ec
en
t
s
tu
d
ies
h
ig
h
lig
h
t
th
at
f
atig
u
e
r
em
ain
s
d
if
f
icu
lt
to
id
en
tif
y
u
s
in
g
co
n
v
en
tio
n
al
m
eth
o
d
s
an
d
th
at
th
er
e
is
s
till
a
cr
itical
n
ee
d
f
o
r
m
o
r
e
ef
f
icie
n
t
au
to
m
atic
to
o
ls
f
o
r
its
co
n
tin
u
o
u
s
m
o
n
ito
r
in
g
[
1
1
]
.
T
h
e
s
tu
d
y
p
r
o
p
o
s
es
t
h
e
d
esi
g
n
,
im
p
l
em
e
n
tat
i
o
n
,
a
n
d
v
ali
d
at
io
n
o
f
a
s
y
s
te
m
b
as
ed
o
n
f
a
cial
v
is
io
n
an
d
ar
t
if
ici
al
i
n
t
e
lli
g
e
n
c
e,
i
n
t
e
g
r
ate
d
wit
h
a
n
ale
r
t
d
e
v
ic
e
u
s
i
n
g
an
E
SP
3
2
m
ic
r
o
c
o
n
t
r
o
lle
r
.
T
h
e
s
t
u
d
y
s
e
ek
s
t
o
e
v
al
u
at
e
th
e
t
ec
h
n
o
lo
g
i
ca
l
ef
f
e
cti
v
e
n
ess
o
f
t
h
e
p
r
o
t
o
t
y
p
e,
c
o
n
s
i
d
e
r
i
n
g
p
e
r
f
o
r
m
a
n
c
e
m
e
tr
ics
s
u
ch
as
d
et
ec
ti
o
n
a
cc
u
r
a
cy
,
f
alse
p
o
s
iti
v
e
r
a
te
,
r
es
p
o
n
s
e
l
ate
n
cy
,
an
d
r
e
lia
b
i
li
ty
o
f
t
h
e
al
er
ts
g
e
n
e
r
a
te
d
.
T
h
e
o
b
je
cti
v
e
is
to
v
er
if
y
wh
et
h
e
r
th
e
a
p
p
li
ca
ti
o
n
o
f
th
is
m
u
lti
m
o
d
al
a
p
p
r
o
a
c
h
ca
n
r
ed
u
ce
o
cc
u
p
at
io
n
al
r
is
k
s
ass
o
cia
te
d
wit
h
d
r
o
wsi
n
ess
a
n
d
att
en
ti
o
n
d
e
f
i
cit
d
is
o
r
d
er
,
c
o
n
tr
ib
u
t
in
g
t
o
t
h
e
d
e
v
el
o
p
m
e
n
t
o
f
s
a
f
e
r
an
d
m
o
r
e
p
r
e
v
e
n
t
iv
e
in
d
u
s
t
r
ia
l
en
v
i
r
o
n
m
e
n
ts
.
I
n
ter
m
s
o
f
ju
s
tific
atio
n
,
th
e
li
ter
atu
r
e
h
ig
h
lig
h
ts
th
at
f
atig
u
e
is
d
if
f
icu
lt
to
d
etec
t
au
to
m
atica
lly
d
u
e
to
th
e
s
u
b
tlety
o
f
its
v
is
u
al
i
n
d
icato
r
s
,
r
ei
n
f
o
r
ci
n
g
th
e
n
ee
d
f
o
r
m
u
ltim
o
d
al,
n
o
n
-
in
v
asiv
e
s
y
s
tem
s
f
o
r
r
ea
l
-
tim
e
m
o
n
ito
r
in
g
[
1
4
]
.
Fro
m
a
co
n
ce
p
tu
al
p
o
in
t
o
f
v
iew,
co
g
n
itiv
e
f
atig
u
e
is
d
ef
in
e
d
as
a
s
tate
ch
ar
ac
ter
ized
b
y
tire
d
n
ess
,
lack
o
f
en
er
g
y
,
d
if
f
icu
lty
c
o
n
ce
n
t
r
atin
g
,
an
d
i
m
p
air
ed
ex
ec
u
tiv
e
f
u
n
ctio
n
s
,
wh
ich
in
cr
ea
s
es
th
e
r
is
k
o
f
er
r
o
r
s
in
cr
itical
s
it
u
atio
n
s
[
1
5
]
.
Alth
o
u
g
h
f
atig
u
e
an
d
ex
ce
s
s
iv
e
d
ay
tim
e
s
leep
in
ess
ar
e
o
f
ten
co
n
f
u
s
ed
,
t
h
e
f
o
r
m
er
h
as
a
g
r
e
ater
im
p
ac
t
o
n
p
s
y
c
h
o
lo
g
ical,
s
o
cial,
an
d
o
cc
u
p
atio
n
al
d
im
e
n
s
io
n
s
[
8
]
.
I
ts
m
ai
n
tr
ig
g
er
s
in
clu
d
e
p
ar
tial
s
leep
d
ep
r
iv
atio
n
an
d
cir
ca
d
ia
n
p
r
o
ce
s
s
es,
wh
o
s
e
ef
f
ec
ts
h
av
e
b
ee
n
co
m
p
ar
ed
to
th
o
s
e
o
f
alco
h
o
l
co
n
s
u
m
p
tio
n
i
n
co
g
n
itiv
e
im
p
air
m
e
n
t
[
6
]
.
I
ts
m
an
if
estatio
n
s
in
clu
d
e
m
em
o
r
y
lap
s
es,
er
r
o
r
s
,
an
d
d
ec
r
ea
s
ed
co
n
ce
n
tr
atio
n
,
wh
ich
u
n
d
er
s
co
r
es
its
d
an
g
er
in
h
ig
h
ly
d
em
an
d
i
n
g
en
v
ir
o
n
m
en
ts
[
1
1
]
.
T
h
e
r
ep
er
cu
s
s
io
n
s
o
f
f
atig
u
e
also
ex
ten
d
to
th
e
ec
o
n
o
m
ic
s
p
h
er
e.
I
n
c
o
u
n
tr
ies
s
u
ch
as
th
e
Un
ited
States
,
th
e
an
n
u
al
co
s
t
ass
o
ciate
d
with
ch
r
o
n
ic
f
atig
u
e
s
y
n
d
r
o
m
e
ca
n
ex
ce
ed
$
1
1
,
7
8
0
p
er
p
atien
t
[
3
]
.
Alth
o
u
g
h
s
tr
ateg
ies
s
u
ch
as
er
g
o
n
o
m
ic
in
te
r
v
en
tio
n
s
h
av
e
b
ee
n
im
p
lem
e
n
ted
,
th
eir
ef
f
ec
tiv
e
n
ess
in
r
e
d
u
cin
g
f
atig
u
e
an
d
p
h
y
s
ical
p
ain
h
as b
ee
n
lim
ited
[
1
0
]
.
Ar
tific
ial
in
tellig
en
ce
an
d
c
o
m
p
u
ter
v
is
io
n
h
av
e
estab
li
s
h
ed
th
em
s
elv
es
as
ce
n
tr
al
p
illar
s
o
f
in
n
o
v
atio
n
[
1
3
]
–
[
1
5
]
.
Dee
p
le
ar
n
in
g
a
p
p
lied
to
v
i
d
eo
a
n
aly
s
is
allo
ws
f
atig
u
e
p
atter
n
s
to
b
e
d
etec
ted
in
r
ea
l
tim
e,
ac
h
iev
in
g
ac
c
u
r
ac
y
o
f
o
v
er
9
4
%
[
1
3
]
.
Am
o
n
g
th
e
m
o
s
t
r
eliab
le
m
etr
ics
is
th
e
p
er
ce
n
tag
e
o
f
ey
elid
clo
s
u
r
e
(
PERC
L
OS)
,
wh
o
s
e
ef
f
ec
tiv
en
ess
h
as
b
ee
n
co
r
r
o
b
o
r
ated
in
b
o
th
in
d
u
s
tr
ial
co
n
tex
ts
an
d
n
u
clea
r
co
n
tr
o
l r
o
o
m
s
[
6
]
–
[
1
3
]
.
At
th
e
s
am
e
tim
e,
p
h
y
s
io
lo
g
ical
b
i
o
m
ar
k
er
s
s
u
ch
as
h
ea
r
t
r
ate
v
ar
i
ab
ilit
y
(
HR
V)
h
av
e
p
r
o
v
e
n
to
b
e
r
o
b
u
s
t
p
r
ed
icto
r
s
o
f
co
g
n
itiv
e
f
atig
u
e,
ac
h
iev
in
g
ac
cu
r
ac
y
lev
els
clo
s
e
to
9
7
%
u
s
in
g
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
s
,
alth
o
u
g
h
th
ey
ar
e
in
v
asiv
e
f
o
r
th
e
u
s
er
wh
en
im
p
lem
en
tin
g
th
e
s
y
s
tem
[
1
5
]
.
Alth
o
u
g
h
elec
tr
o
en
ce
p
h
al
o
g
r
am
(
E
E
G)
an
d
elec
tr
o
ca
r
d
i
o
g
r
am
(
E
C
G
)
b
ased
m
eth
o
d
s
ca
n
d
etec
t
d
r
o
wsi
n
ess
wi
th
g
o
o
d
ac
cu
r
ac
y
,
s
ev
er
al
s
tu
d
ies
n
o
te
th
at
th
ey
ar
e
n
o
t
id
ea
l
f
o
r
r
ea
l
wo
r
k
o
r
d
r
iv
in
g
en
v
i
r
o
n
m
e
n
ts
.
T
h
eir
r
elian
ce
o
n
body
-
attac
h
ed
s
en
s
o
r
s
an
d
s
p
ec
ialized
eq
u
ip
m
e
n
t
m
ak
es
th
em
in
tr
u
s
iv
e
an
d
o
f
ten
u
n
co
m
f
o
r
ta
b
le,
wh
ich
r
ed
u
ce
s
th
eir
p
r
ac
ticality
o
u
ts
id
e
co
n
tr
o
lled
c
o
n
d
itio
n
s
[
1
6
]
,
[
1
7
]
.
I
n
th
is
co
n
te
x
t,
s
ev
er
al
s
tu
d
ies
ag
r
ee
th
at
co
m
p
u
ter
v
is
io
n
is
a
m
o
r
e
p
r
ac
tical
an
d
v
iab
le
alter
n
ativ
e
[
1
1
]
–
[
1
4
]
.
R
ec
en
t
r
ev
iews
h
ig
h
lig
h
t
th
at
s
y
s
tem
s
b
ased
o
n
b
eh
a
v
io
r
al
f
ea
tu
r
es
s
u
ch
as
f
ac
ial
tr
ac
k
in
g
ar
e
wid
ely
ad
o
p
ted
d
u
e
to
th
eir
ea
s
e
o
f
u
s
e
an
d
lo
w
co
s
t,
alth
o
u
g
h
th
eir
p
er
f
o
r
m
an
c
e
ca
n
b
e
af
f
ec
ted
b
y
v
ar
iatio
n
s
in
lig
h
tin
g
a
n
d
ca
m
er
a
a
n
g
les
[
1
6
]
.
T
h
is
r
ein
f
o
r
ce
s
th
e
n
ee
d
to
ad
o
p
t
m
u
ltimo
d
al
ap
p
r
o
ac
h
es
th
at
co
m
b
in
e
m
u
ltip
le
v
is
u
al
an
d
p
o
s
tu
r
al
p
ar
a
m
eter
s
to
in
cr
ea
s
e
th
e
r
o
b
u
s
tn
ess
o
f
th
e
s
y
s
tem
in
d
y
n
a
m
ic
wo
r
k
en
v
ir
o
n
m
e
n
ts
.
T
h
e
co
n
tr
ib
u
tio
n
o
f
th
is
r
esear
ch
f
o
cu
s
es
o
n
th
e
d
ev
elo
p
m
en
t
o
f
a
n
o
n
-
i
n
v
asiv
e
m
u
lti
p
ar
am
etr
ic
s
y
s
tem
ca
p
ab
le
o
f
s
im
u
ltan
eo
u
s
ly
in
teg
r
atin
g
an
d
an
aly
zin
g
d
if
f
er
en
t
f
ac
ial
an
d
p
o
s
tu
r
al
i
n
d
icato
r
s
ass
o
ciate
d
with
f
atig
u
e.
Un
lik
e
tr
a
d
itio
n
a
l
ap
p
r
o
ac
h
es
b
ased
s
o
lely
o
n
b
lin
k
in
g
o
r
ey
e
clo
s
u
r
e,
th
e
p
r
o
p
o
s
al
in
co
r
p
o
r
ates
a
b
r
o
ad
er
s
et
o
f
ch
ar
ac
ter
is
tics
,
in
clu
d
in
g
ey
e
asp
ec
t
r
atio
f
o
r
d
etec
tin
g
p
r
o
lo
n
g
ed
clo
s
u
r
es,
th
e
p
r
o
b
a
b
ilit
y
o
f
m
o
u
th
o
p
e
n
in
g
f
o
r
id
e
n
tify
in
g
y
awn
in
g
,
a
n
d
h
ea
d
o
r
ien
tatio
n
b
y
ca
lcu
latin
g
th
e
an
g
le
a
n
d
v
er
tical
i
n
clin
atio
n
r
elativ
e
to
th
e
wo
r
k
s
tatio
n
.
T
h
is
co
m
b
in
ed
an
aly
s
is
allo
ws
f
o
r
a
m
o
r
e
r
o
b
u
s
t
ass
ess
m
en
t
o
f
t
h
e
u
s
er
'
s
co
n
d
itio
n
,
r
e
d
u
cin
g
d
e
p
en
d
e
n
ce
o
n
a
s
in
g
le
p
a
r
am
eter
t
h
at
,
in
r
ea
l
c
o
n
d
itio
n
s
,
ca
n
b
e
a
f
f
ec
ted
b
y
ex
ter
n
al
f
ac
to
r
s
s
u
ch
as
lig
h
tin
g
,
th
e
o
p
er
ato
r
'
s
n
atu
r
al
m
o
v
em
en
t,
o
r
in
d
iv
i
d
u
al
v
a
r
iab
ilit
y
.
T
h
an
k
s
to
th
is
m
u
ltimo
d
a
l
in
teg
r
atio
n
,
th
e
s
y
s
tem
in
cr
ea
s
es its
r
eliab
ilit
y
in
th
e
ea
r
ly
d
e
tectio
n
o
f
f
atig
u
e
in
o
cc
u
p
atio
n
al
en
v
ir
o
n
m
en
ts
.
T
o
th
is
en
d
,
th
e
r
esear
ch
ai
m
s
to
co
n
tr
ib
u
te
t
o
th
e
f
ield
o
f
o
cc
u
p
atio
n
al
s
af
ety
b
y
p
r
o
v
id
in
g
an
ac
ce
s
s
ib
le,
s
ca
lab
le
to
o
l
th
at
ca
n
b
e
ad
ap
ted
to
d
if
f
e
r
en
t
wo
r
k
en
v
ir
o
n
m
en
ts
wh
er
e
p
r
o
d
u
ctiv
ity
an
d
r
is
k
p
r
ev
en
tio
n
ar
e
p
r
io
r
ities
.
I
n
a
d
d
itio
n
,
th
e
m
u
lti
-
p
ar
a
m
eter
a
p
p
r
o
ac
h
allo
ws
f
o
r
th
e
o
p
tim
i
za
tio
n
o
f
d
etec
tio
n
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
7
5
0
-
2
7
6
8
2752
alg
o
r
ith
m
s
an
d
th
e
in
co
r
p
o
r
a
tio
n
o
f
n
ew
s
o
u
r
ce
s
o
f
in
f
o
r
m
atio
n
th
at
s
tr
en
g
th
en
th
e
s
y
s
tem
's
p
r
ed
ictiv
e
ca
p
ab
ilit
ies.
T
h
e
m
ain
co
n
tr
ib
u
tio
n
s
o
f
th
is
wo
r
k
ar
e
as f
o
llo
ws:
a.
T
h
e
d
ev
elo
p
m
en
t
o
f
a
r
ea
l
-
t
im
e
m
u
ltimo
d
al
f
atig
u
e
d
ete
ctio
n
s
y
s
tem
in
teg
r
atin
g
o
cu
lar
,
f
ac
ial,
an
d
p
o
s
tu
r
al
f
ea
tu
r
es with
in
a
u
n
if
i
ed
co
m
p
u
ter
v
is
io
n
f
r
am
ewo
r
k
.
b.
T
h
e
im
p
lem
en
tatio
n
o
f
an
ef
f
i
cien
t
ed
g
e
-
b
ased
s
o
lu
tio
n
o
n
a
R
asp
b
er
r
y
Pi
5
,
en
ab
lin
g
r
ea
l
-
tim
e
o
p
er
atio
n
with
o
u
t r
eq
u
i
r
in
g
h
ig
h
-
p
er
f
o
r
m
an
ce
co
m
p
u
tin
g
r
eso
u
r
ce
s
.
c.
T
h
e
i
n
t
e
g
r
a
ti
o
n
o
f
a
n
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
(
I
o
T
)
b
a
s
e
d
a
l
e
r
t
s
y
s
t
e
m
u
s
i
n
g
m
e
s
s
a
g
e
q
u
e
u
i
n
g
te
le
m
e
t
r
y
t
r
a
n
s
p
o
r
t
(
M
Q
T
T
)
a
n
d
a
n
E
SP
3
2
m
o
d
u
l
e
,
p
r
o
v
i
d
i
n
g
i
m
m
e
d
i
a
t
e
v
is
u
a
l
a
n
d
a
u
d
i
t
o
r
y
f
e
e
d
b
a
c
k
u
p
o
n
f
a
t
i
g
u
e
d
e
t
e
c
t
i
o
n
.
d.
T
h
e
p
r
o
p
o
s
al
o
f
a
n
o
n
-
in
tr
u
s
iv
e
an
d
co
s
t
-
ef
f
ec
tiv
e
s
o
lu
tio
n
f
o
r
o
cc
u
p
atio
n
al
h
ea
lth
m
o
n
ito
r
in
g
in
r
ea
l
-
wo
r
ld
en
v
ir
o
n
m
e
n
ts
.
e.
T
h
e
v
alid
atio
n
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
th
r
o
u
g
h
co
n
tr
o
lled
ex
p
er
im
en
ts
u
s
in
g
s
y
n
th
e
tic
v
id
eo
d
ata,
ac
h
iev
in
g
h
ig
h
d
etec
tio
n
ac
cu
r
ac
y
an
d
l
o
w
laten
cy
.
I
n
th
is
co
n
tex
t,
th
e
m
ain
o
b
je
ctiv
e
o
f
th
is
s
tu
d
y
is
to
d
ev
el
o
p
an
ar
tific
ial
in
tellig
en
ce
-
b
a
s
ed
s
y
s
tem
ca
p
ab
le
o
f
d
etec
tin
g
f
atig
u
e
in
r
ea
l
tim
e
an
d
is
s
u
in
g
i
n
teg
r
at
ed
aler
ts
v
ia
a
m
icr
o
c
o
n
tr
o
ller
,
with
a
clea
r
f
o
cu
s
o
n
o
cc
u
p
atio
n
al
h
ea
lt
h
ap
p
l
icatio
n
s
.
T
o
ac
h
iev
e
th
is
,
t
h
e
wo
r
k
in
co
r
p
o
r
ates
s
ev
er
al
co
m
p
lem
e
n
tar
y
co
m
p
o
n
en
ts
:
a
co
m
p
u
ter
v
is
io
n
m
o
d
el
th
at
an
aly
ze
s
f
ac
ial
s
ig
n
als
to
id
en
tif
y
ea
r
l
y
s
ig
n
s
o
f
f
atig
u
e,
a
p
r
o
to
ty
p
e
aler
t
m
ec
h
an
is
m
t
h
at
ac
tiv
ates
au
to
m
atica
lly
u
p
o
n
d
etec
tin
g
cr
itical
s
tates,
an
d
a
n
in
tu
itiv
e
in
ter
f
ac
e
th
at
allo
ws
f
o
r
co
n
tin
u
o
u
s
m
o
n
ito
r
in
g
o
f
th
e
o
p
er
a
to
r
'
s
co
n
d
itio
n
.
Fin
ally
,
th
e
s
y
s
tem
is
ev
alu
ated
in
co
n
tr
o
lled
s
ce
n
a
r
io
s
to
ass
ess
its
p
er
f
o
r
m
an
ce
in
ter
m
s
o
f
d
etec
tio
n
ac
cu
r
ac
y
,
f
a
ls
e
p
o
s
itiv
e
r
ate,
co
m
m
u
n
icatio
n
laten
cy
,
a
n
d
t
h
e
r
eliab
ilit
y
o
f
t
h
e
g
en
e
r
ated
aler
ts
.
2.
M
E
T
H
O
D
T
h
e
m
e
t
h
o
d
o
l
o
g
y
e
m
p
l
o
y
e
d
i
n
t
h
is
r
es
e
a
r
c
h
i
s
b
a
s
e
d
o
n
a
c
o
n
t
r
o
l
l
e
d
e
x
p
e
r
i
m
e
n
t
a
l
a
p
p
r
o
a
c
h
d
e
s
i
g
n
e
d
t
o
v
a
l
i
d
a
t
e
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
t
h
e
f
a
t
i
g
u
e
d
et
e
c
ti
o
n
s
y
s
t
e
m
w
i
t
h
o
u
t
r
e
q
u
i
r
i
n
g
t
h
e
d
i
r
e
ct
p
a
r
t
i
c
i
p
a
ti
o
n
o
f
r
e
a
l
i
n
d
i
v
i
d
u
a
l
s
.
T
h
e
p
r
o
p
o
s
e
d
s
y
s
te
m
i
m
p
l
e
m
e
n
ts
a
m
u
l
ti
p
a
r
a
m
et
r
i
c
c
o
m
p
u
t
e
r
v
i
s
i
o
n
m
o
d
e
l
t
h
at
i
n
t
e
g
r
at
e
s
v
a
r
i
o
u
s
v
i
s
u
a
l
i
n
d
i
c
at
o
r
s
f
o
r
r
e
a
l
-
t
im
e
f
a
t
i
g
u
e
d
e
t
e
c
ti
o
n
,
w
i
t
h
d
i
r
e
c
t
a
p
p
l
i
c
at
i
o
n
s
i
n
o
c
c
u
p
a
t
i
o
n
a
l
h
e
a
lt
h
.
T
h
e
m
e
t
h
o
d
o
l
o
g
i
c
a
l
p
r
o
c
es
s
w
as
s
t
r
u
c
t
u
r
e
d
i
n
t
o
s
e
v
e
r
a
l s
e
q
u
e
n
ti
a
l
s
t
a
g
es
,
as
s
h
o
w
n
i
n
F
i
g
u
r
e
1
,
f
o
r
m
i
n
g
a
w
o
r
k
f
l
o
w
t
h
a
t
s
p
a
n
s
f
r
o
m
t
h
e
i
d
e
n
t
i
f
i
c
a
tio
n
o
f
r
e
l
e
v
a
n
t
f
a
c
i
a
l
p
a
r
a
m
et
e
r
s
t
o
t
h
e
te
c
h
n
i
c
a
l
v
al
i
d
a
ti
o
n
o
f
t
h
e
m
o
d
e
l
.
Fig
u
r
e
1
.
Flo
wch
ar
t
o
f
th
e
m
et
h
o
d
o
lo
g
ical
p
r
o
ce
s
s
o
f
th
e
f
ati
g
u
e
d
etec
tio
n
s
y
s
tem
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
R
ea
l
-
time
mu
ltimo
d
a
l fa
tig
u
e
d
etec
tio
n
u
s
in
g
fa
ci
a
l v
is
io
n
a
n
d
a
lert
…
(
A
n
d
r
és E
n
r
iq
u
e
R
o
ja
s
P
r
imo
)
2753
2
.
1
.
L
it
er
a
t
ure
re
v
iew
C
o
m
p
u
ter
v
is
io
n
-
b
ased
f
atig
u
e
d
etec
tio
n
m
eth
o
d
s
ar
e
co
m
m
o
n
ly
class
if
ied
in
to
o
cu
lar
,
f
ac
ial,
an
d
p
o
s
tu
r
al
ap
p
r
o
ac
h
es.
Ocu
lar
m
eth
o
d
s
r
ely
o
n
in
d
icato
r
s
s
u
ch
as
th
e
ey
e
asp
ec
t
r
atio
(
E
A
R
)
an
d
PER
C
L
OS
to
ev
alu
ate
ey
e
clo
s
u
r
e
an
d
b
li
n
k
in
g
b
e
h
av
io
r
,
wh
ic
h
ar
e
s
tr
o
n
g
ly
ass
o
ciate
d
with
f
atig
u
e
[
1
2
]
,
[
1
8
]
.
Facial
ap
p
r
o
ac
h
es
ex
ten
d
t
h
is
an
aly
s
is
b
y
in
co
r
p
o
r
atin
g
m
o
u
th
m
o
v
em
en
ts
to
d
etec
t
y
awn
i
n
g
u
s
in
g
f
ac
ial
lan
d
m
a
r
k
s
[
1
9
]
,
wh
ile
p
o
s
tu
r
al
m
eth
o
d
s
an
aly
ze
h
ea
d
an
d
n
ec
k
p
o
s
itio
n
,
as
h
ea
d
tilt
in
g
is
clo
s
ely
li
n
k
ed
t
o
d
r
o
wsi
n
ess
d
u
e
to
th
e
r
ed
u
ce
d
ab
ilit
y
to
m
ain
tain
an
u
p
r
ig
h
t
p
o
s
tu
r
e
[
2
0
]
.
T
h
ese
ap
p
r
o
ac
h
es
b
el
o
n
g
to
th
e
b
r
o
ad
er
ca
teg
o
r
y
o
f
b
eh
a
v
io
r
-
b
ased
m
eth
o
d
s
,
as
th
ey
r
ely
o
n
e
x
ter
n
ally
o
b
s
er
v
ab
le
f
ea
tu
r
es
ca
p
tu
r
ed
th
r
o
u
g
h
co
m
p
u
ter
v
is
io
n
s
y
s
tem
s
.
T
h
i
s
clas
s
if
icatio
n
alig
n
s
with
p
r
ev
io
u
s
s
tu
d
ies
th
at
g
r
o
u
p
d
r
o
wsi
n
ess
d
etec
tio
n
tech
n
iq
u
es in
to
b
eh
av
io
r
al,
p
h
y
s
io
lo
g
ical,
an
d
h
y
b
r
id
ap
p
r
o
a
ch
es
[
1
6
]
.
B
u
ild
in
g
u
p
o
n
th
ese
ap
p
r
o
ac
h
es,
s
ev
er
al
s
tu
d
ies
h
av
e
em
p
l
o
y
ed
g
e
o
m
etr
ic
d
escr
ip
to
r
s
d
er
i
v
ed
f
r
o
m
f
ac
ial
lan
d
m
ar
k
s
to
c
h
ar
ac
ter
i
ze
f
atig
u
e
-
r
elate
d
b
eh
av
i
o
r
s
.
I
n
o
cu
lar
a
n
aly
s
is
,
th
e
E
AR
h
as
b
ee
n
wid
ely
u
s
ed
to
d
etec
t
b
lin
k
in
g
p
atter
n
s
an
d
p
r
o
lo
n
g
ed
e
y
e
clo
s
u
r
e,
as
it
is
co
m
p
u
ted
f
r
o
m
E
u
clid
ea
n
d
is
tan
ce
s
b
etwe
en
ey
e
lan
d
m
ar
k
s
an
d
e
x
h
ib
its
s
tab
le
v
alu
es
wh
en
th
e
ey
es
ar
e
o
p
en
,
d
ec
r
ea
s
in
g
d
u
r
in
g
clo
s
u
r
e
[
2
1
]
.
Similar
ly
,
f
ac
ial
-
b
ased
ap
p
r
o
ac
h
es
u
tili
z
e
th
e
m
o
u
t
h
asp
ec
t
r
atio
(
MA
R
)
to
d
etec
t
y
awn
in
g
th
r
o
u
g
h
th
r
esh
o
ld
-
b
ased
class
if
icatio
n
o
f
s
u
s
tain
ed
m
o
u
th
o
p
en
in
g
[
1
9
]
.
Desp
ite
th
eir
co
m
p
u
tatio
n
al
e
f
f
icien
cy
an
d
s
u
itab
ilit
y
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
,
th
ese
g
e
o
m
etr
ic
m
eth
o
d
s
m
a
y
b
e
af
f
ec
te
d
b
y
illu
m
in
atio
n
c
h
an
g
es,
h
e
ad
p
o
s
e
v
a
r
iatio
n
s
,
an
d
in
d
i
v
id
u
al
f
ac
ial
d
if
f
er
en
c
es,
wh
ich
ca
n
r
e
d
u
ce
d
etec
tio
n
r
eliab
ilit
y
in
u
n
c
o
n
s
tr
ain
ed
c
o
n
d
itio
n
s
[
2
1
]
.
I
n
ad
d
itio
n
,
s
tu
d
ies
b
ased
o
n
p
h
y
s
ical
m
ea
s
u
r
em
e
n
ts
h
av
e
s
h
o
wn
th
at
r
ely
in
g
o
n
a
s
in
g
le
p
ar
am
eter
,
s
u
ch
as
h
ea
d
p
o
s
tu
r
e,
m
ay
b
e
i
n
s
u
f
f
icien
t
to
ac
cu
r
ately
d
etec
t
d
r
o
wsi
n
ess
,
s
in
ce
n
o
r
m
al
d
r
i
v
in
g
b
e
h
av
io
r
s
ca
n
p
r
o
d
u
ce
s
im
ilar
p
atter
n
s
a
n
d
lead
to
f
alse
alar
m
s
[
2
0
]
.
T
o
ad
d
r
ess
th
ese
lim
itatio
n
s
,
r
ec
en
t
wo
r
k
s
h
a
v
e
in
co
r
p
o
r
ated
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs)
to
au
to
m
atica
lly
lear
n
d
is
cr
im
in
a
tiv
e
f
ea
tu
r
es
f
r
o
m
im
ag
e
d
ata,
im
p
r
o
v
in
g
d
etec
tio
n
p
e
r
f
o
r
m
an
ce
u
n
d
er
co
m
p
lex
co
n
d
itio
n
s
.
I
n
p
ar
ticu
lar
,
h
y
b
r
id
ap
p
r
o
ac
h
es
co
m
b
in
in
g
C
NN
-
b
ased
ey
e
s
tate
clas
s
if
icatio
n
with
MA
R
-
b
ased
y
awn
in
g
d
etec
tio
n
h
av
e
d
em
o
n
s
tr
ated
im
p
r
o
v
e
d
r
o
b
u
s
tn
ess
b
y
in
teg
r
atin
g
lear
n
ed
r
ep
r
esen
tatio
n
s
with
g
eo
m
etr
ic
f
ea
tu
r
es
[
1
9
]
.
E
x
ten
d
in
g
b
e
y
o
n
d
v
is
io
n
-
b
as
ed
m
eth
o
d
s
,
r
ec
en
t
r
esear
ch
h
as
ex
p
lo
r
ed
f
atig
u
e
d
etec
tio
n
s
tr
ateg
ies
b
ased
o
n
p
h
y
s
io
lo
g
ical,
b
e
h
a
v
io
r
al,
v
eh
ic
u
lar
,
an
d
h
y
b
r
id
f
ea
tu
r
es,
em
p
h
asizin
g
th
at
n
o
s
in
g
le
ap
p
r
o
ac
h
is
s
u
f
f
icien
t
to
a
d
d
r
ess
all
r
ea
l
-
wo
r
ld
co
n
d
itio
n
s
[
1
6
]
.
Ph
y
s
io
lo
g
ical
-
b
ased
m
et
h
o
d
s
,
p
ar
tic
u
lar
ly
th
o
s
e
r
ely
in
g
o
n
E
E
G,
E
C
G,
an
d
elec
tr
o
o
c
u
lo
g
r
am
(
E
OG)
s
ig
n
als,
o
f
f
er
h
ig
h
ac
cu
r
ac
y
b
y
d
ir
ec
tly
ca
p
tu
r
in
g
th
e
d
r
iv
er
’
s
in
ter
n
al
s
tate;
h
o
wev
er
,
th
ei
r
in
tr
u
s
iv
e
n
atu
r
e
an
d
d
ep
en
d
en
ce
o
n
wea
r
ab
le
s
en
s
o
r
s
l
im
it
th
eir
p
r
ac
tical
ap
p
licab
ilit
y
.
I
n
c
o
n
tr
ast,
v
is
io
n
-
b
ased
ap
p
r
o
ac
h
es
p
r
o
v
id
e
a
n
o
n
-
in
tr
u
s
iv
e
alter
n
ativ
e
b
u
t
r
em
ain
s
en
s
itiv
e
to
en
v
ir
o
n
m
en
tal
f
ac
to
r
s
s
u
ch
as illu
m
in
atio
n
,
o
cc
lu
s
io
n
s
,
an
d
s
ce
n
e
v
ar
iab
ilit
y
[
2
2
]
.
Fu
r
th
er
m
o
r
e
,
co
m
p
a
r
ativ
e
s
tu
d
ies
h
av
e
s
h
o
wn
th
at
p
h
y
s
io
lo
g
ical
s
ig
n
als
ten
d
to
b
e
m
o
r
e
r
eliab
le
b
u
t
in
tr
o
d
u
ce
ch
allen
g
es
s
u
ch
as
s
ig
n
al
v
ar
iab
ilit
y
,
co
m
p
u
tatio
n
al
co
m
p
lex
ity
,
an
d
lim
ited
s
ca
lab
ilit
y
,
wh
er
ea
s
b
eh
av
io
r
al
m
eth
o
d
s
ar
e
ea
s
ier
to
d
e
p
lo
y
b
u
t
less
r
o
b
u
s
t
u
n
d
er
u
n
co
n
s
tr
ain
ed
co
n
d
itio
n
s
[
2
3
]
.
T
o
o
v
er
c
o
m
e
th
ese
lim
itatio
n
s
,
h
y
b
r
id
a
n
d
m
u
ltimo
d
al
ap
p
r
o
ac
h
es
h
av
e
em
er
g
ed
as
a
p
r
o
m
is
in
g
d
i
r
ec
tio
n
,
co
m
b
in
in
g
m
u
ltip
le
s
o
u
r
ce
s
o
f
in
f
o
r
m
atio
n
to
im
p
r
o
v
e
r
o
b
u
s
tn
ess
an
d
d
etec
tio
n
ac
cu
r
ac
y
.
I
n
p
ar
ticu
la
r
,
th
e
in
teg
r
atio
n
o
f
b
eh
av
io
r
al
a
n
d
p
h
y
s
io
lo
g
ical
f
ea
tu
r
es
h
as
b
ee
n
s
h
o
wn
to
s
ig
n
if
ican
tly
en
h
an
ce
p
e
r
f
o
r
m
an
ce
,
in
cr
ea
s
in
g
d
etec
tio
n
ac
c
u
r
ac
y
f
r
o
m
m
o
d
er
ate
lev
els
to
n
ea
r
h
ig
h
-
p
er
f
o
r
m
a
n
ce
r
a
n
g
es,
with
s
o
m
e
h
y
b
r
id
s
y
s
tem
s
ac
h
iev
in
g
r
esu
lts
ap
p
r
o
ac
h
i
n
g
n
ea
r
-
p
er
f
ec
t
ac
c
u
r
ac
y
u
n
d
er
co
n
tr
o
lled
co
n
d
itio
n
s
[
1
6
]
.
Ho
wev
er
,
th
ese
s
y
s
tem
s
also
in
tr
o
d
u
ce
ch
alle
n
g
es
r
elate
d
to
d
ata
f
u
s
io
n
,
c
o
m
p
u
tatio
n
al
co
s
t,
a
n
d
r
ea
l
-
ti
m
e
im
p
lem
en
tatio
n
,
wh
ich
m
u
s
t b
e
ca
r
e
f
u
lly
ad
d
r
e
s
s
ed
to
en
s
u
r
e
p
r
ac
tical
d
e
p
lo
y
m
en
t
[
2
2
]
.
Desp
ite
th
e
s
ig
n
if
ican
t
p
r
o
g
r
ess
ac
h
iev
ed
in
f
atig
u
e
d
etec
tio
n
s
y
s
tem
s
,
s
ev
er
al
lim
ita
ti
o
n
s
r
em
ain
u
n
ad
d
r
ess
ed
in
cu
r
r
en
t
ap
p
r
o
a
ch
es.
I
n
p
ar
ticu
lar
,
v
is
io
n
-
b
as
ed
m
eth
o
d
s
,
wh
ile
n
o
n
-
in
tr
u
s
i
v
e
an
d
s
u
itab
le
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
,
ar
e
s
ti
ll
h
ig
h
ly
s
en
s
itiv
e
to
en
v
ir
o
n
m
en
tal
co
n
d
itio
n
s
s
u
ch
as
illu
m
in
atio
n
c
h
an
g
es,
o
cc
lu
s
io
n
s
,
an
d
h
e
ad
p
o
s
e
v
a
r
iatio
n
s
,
wh
ich
ca
n
r
e
d
u
ce
th
eir
r
eliab
ilit
y
in
p
r
ac
tical
s
ce
n
ar
io
s
.
On
th
e
o
t
h
er
h
an
d
,
p
h
y
s
io
lo
g
ical
-
b
ased
m
e
th
o
d
s
p
r
o
v
id
e
h
i
g
h
er
ac
c
u
r
ac
y
b
y
d
ir
ec
tly
ca
p
tu
r
in
g
th
e
d
r
i
v
er
’
s
in
ter
n
al
s
tate;
h
o
wev
er
,
th
eir
in
tr
u
s
iv
e
n
at
u
r
e
an
d
d
e
p
en
d
e
n
ce
o
n
s
p
ec
ialized
s
en
s
o
r
s
lim
it
th
eir
u
s
ab
ilit
y
in
r
ea
l
-
wo
r
ld
d
ep
lo
y
m
e
n
ts
.
Alth
o
u
g
h
h
y
b
r
id
an
d
m
u
ltim
o
d
al
ap
p
r
o
ac
h
es h
av
e
b
ee
n
p
r
o
p
o
s
ed
to
o
v
er
c
o
m
e
th
ese
lim
itatio
n
s
b
y
co
m
b
i
n
in
g
m
u
ltip
le
s
o
u
r
ce
s
o
f
in
f
o
r
m
atio
n
,
m
an
y
ex
is
tin
g
s
o
lu
tio
n
s
r
ely
o
n
co
m
p
lex
ar
ch
itectu
r
es,
h
ig
h
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
,
o
r
r
eq
u
ir
e
h
eter
o
g
en
eo
u
s
d
ata
s
o
u
r
ce
s
th
at
ar
e
n
o
t
alwa
y
s
av
ailab
le
in
r
ea
l
-
tim
e
en
v
ir
o
n
m
en
ts
.
T
h
er
ef
o
r
e,
th
er
e
is
a
clea
r
n
ee
d
f
o
r
ef
f
icien
t
an
d
n
o
n
-
i
n
t
r
u
s
iv
e
s
y
s
tem
s
ca
p
ab
le
o
f
in
teg
r
atin
g
m
u
ltip
le
b
eh
a
v
io
r
al
in
d
icato
r
s
wh
ile
m
ain
tain
in
g
lo
w
co
m
p
u
tatio
n
al
co
s
t
an
d
r
ea
l
-
tim
e
p
er
f
o
r
m
a
n
ce
.
I
n
th
is
co
n
tex
t,
th
e
p
r
esen
t
wo
r
k
p
r
o
p
o
s
es
a
m
u
ltimo
d
al
f
atig
u
e
d
etec
tio
n
s
y
s
tem
b
ased
ex
clu
s
iv
ely
o
n
c
o
m
p
u
ter
v
is
io
n
,
in
teg
r
atin
g
o
c
u
lar
,
f
ac
ial,
an
d
p
o
s
tu
r
al
f
ea
tu
r
es
with
in
a
u
n
if
ied
f
r
am
ewo
r
k
.
B
y
co
m
b
in
in
g
co
m
p
lem
en
tar
y
in
d
icato
r
s
s
u
ch
as
ey
e
clo
s
u
r
e,
y
awn
in
g
b
e
h
av
io
r
,
a
n
d
h
ea
d
p
o
s
itio
n
,
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
aim
s
to
im
p
r
o
v
e
d
etec
tio
n
r
o
b
u
s
tn
ess
wh
ile
av
o
id
in
g
th
e
lim
itatio
n
s
ass
o
ciate
d
with
s
in
g
le
-
m
o
d
ality
s
y
s
tem
s
an
d
in
tr
u
s
iv
e
s
en
s
in
g
tech
n
iq
u
es.
Ad
d
itio
n
ally
,
th
e
s
y
s
tem
in
co
r
p
o
r
ates
a
r
ea
l
-
tim
e
aler
t
m
ec
h
an
is
m
th
at
p
r
o
v
id
es
im
m
ed
iate
f
ee
d
b
ac
k
to
th
e
u
s
er
wh
en
f
atig
u
e
-
r
el
ated
ev
en
ts
ar
e
d
etec
ted
,
en
h
an
cin
g
its
p
r
ac
tical
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
7
5
0
-
2
7
6
8
2754
ap
p
licab
ilit
y
in
s
af
ety
-
cr
itical
en
v
ir
o
n
m
e
n
ts
.
T
h
is
d
esig
n
s
ee
k
s
to
ac
h
iev
e
a
b
alan
ce
b
etwe
en
ac
cu
r
ac
y
,
ef
f
icien
cy
,
a
n
d
p
r
ac
tical
ap
p
li
ca
b
ilit
y
,
m
ak
in
g
it su
itab
le
f
o
r
r
ea
l
-
wo
r
ld
f
atig
u
e
m
o
n
ito
r
in
g
s
ce
n
ar
io
s
.
Un
lik
e
p
r
ev
io
u
s
ap
p
r
o
ac
h
es
th
at
r
ely
o
n
eith
er
s
in
g
le
-
m
o
d
ality
d
etec
tio
n
o
r
in
t
r
u
s
iv
e
p
h
y
s
io
lo
g
ical
s
en
s
o
r
s
,
th
e
p
r
o
p
o
s
ed
s
y
s
tem
in
teg
r
ates
m
u
ltip
le
b
e
h
av
io
r
al
in
d
icato
r
s
with
in
a
lig
h
tweig
h
t
ed
g
e
-
co
m
p
u
tin
g
ar
ch
itectu
r
e.
Fu
r
th
er
m
o
r
e,
th
e
in
teg
r
atio
n
o
f
r
ea
l
-
tim
e
I
o
T
-
b
ased
aler
ts
u
s
in
g
MQ
T
T
an
d
E
SP
3
2
en
ab
les
im
m
ed
iate
r
esp
o
n
s
e,
m
a
k
in
g
t
h
e
s
y
s
tem
s
u
itab
le
f
o
r
p
r
ac
tical
d
ep
lo
y
m
e
n
t in
o
cc
u
p
atio
n
al
en
v
ir
o
n
m
en
ts
.
2
.
2
.
Select
io
n o
f
v
a
ri
a
bles
a
nd
det
ec
t
io
n c
rit
er
ia
B
ased
o
n
th
e
f
i
n
d
i
n
g
s
o
f
th
e
li
ter
at
u
r
e
r
ev
iew
,
t
h
e
m
o
s
t
w
id
e
ly
u
s
e
d
q
u
a
n
ti
tat
iv
e
m
et
r
ics
i
n
s
ci
e
n
ti
f
ic
s
tu
d
ies
o
n
f
a
ti
g
u
e
a
n
d
d
r
o
ws
in
ess
we
r
e
a
d
o
p
t
e
d
[
1
2
]
–
[
1
8
]
.
T
h
es
e
m
et
r
ics
we
r
e
s
ele
ct
ed
to
en
s
u
r
e
t
h
e
a
cc
u
r
ac
y
an
d
r
eli
ab
ilit
y
o
f
t
h
e
v
is
u
al
s
i
g
n
s
co
m
p
u
te
d
b
y
t
h
e
f
ati
g
u
e
d
et
ec
ti
o
n
s
y
s
t
em
.
T
h
e
p
a
r
a
m
et
e
r
s
in
c
lu
d
ed
:
a.
E
AR
f
o
r
d
etec
tin
g
b
lin
k
f
r
eq
u
en
cy
an
d
p
r
o
lo
n
g
e
d
ey
e
cl
o
s
u
r
e,
b.
Mo
u
th
o
p
en
in
g
as a
n
in
d
icato
r
o
f
d
r
o
wsi
n
ess
an
d
lo
s
s
o
f
atten
tio
n
,
c.
Hea
d
tilt
an
g
le
f
o
r
id
en
tify
i
n
g
p
o
s
tu
r
al
f
atig
u
e
i
n
d
icato
r
s
.
T
h
ese
m
etr
ics
wer
e
in
teg
r
ate
d
in
to
th
e
s
y
s
tem
to
estab
lis
h
d
etec
tio
n
cr
iter
ia
th
at
g
u
id
e
th
e
r
ea
l
-
tim
e
ac
tiv
atio
n
o
f
aler
ts
.
2
.
3
.
M
a
t
er
ia
ls
a
nd
t
o
o
ls
Fig
u
r
e
2
s
u
m
m
ar
izes
th
e
p
h
y
s
ical
an
d
lo
g
ical
co
m
p
o
n
en
ts
u
s
ed
in
th
e
s
y
s
tem
im
p
lem
en
t
atio
n
.
T
h
e
alg
o
r
ith
m
was
d
ev
elo
p
ed
u
s
in
g
th
e
Py
th
o
n
p
r
o
g
r
am
m
i
n
g
lan
g
u
ag
e
alo
n
g
with
s
p
ec
ialized
lib
r
ar
ies
f
o
r
im
ag
e
p
r
o
ce
s
s
in
g
an
d
r
ea
l
-
tim
e
v
i
d
e
o
an
aly
s
is
.
Op
en
C
V
an
d
Me
d
iaPip
e
wer
e
u
s
ed
f
o
r
f
ac
ial
la
n
d
m
ar
k
ex
tr
ac
tio
n
an
d
p
r
e
p
r
o
ce
s
s
in
g
,
w
h
ile
T
en
s
o
r
Flo
w
was
em
p
lo
y
ed
to
im
p
lem
en
t
th
e
n
eu
r
al
n
etwo
r
k
–
b
ased
m
o
u
th
-
s
tate
class
if
ier
.
Fo
r
th
is
p
u
r
p
o
s
e,
a
p
u
b
licly
av
ailab
le
y
awn
in
g
d
a
taset
o
b
tain
ed
f
r
o
m
Kag
g
le
was
u
s
ed
to
tr
ain
a
n
d
v
alid
ate
th
e
m
o
u
th
-
o
p
en
i
n
g
m
o
d
el,
en
ab
lin
g
th
e
d
is
cr
im
in
atio
n
b
etwe
en
o
p
e
n
an
d
clo
s
ed
m
o
u
th
s
tates
u
n
d
er
d
if
f
er
en
t
c
o
n
d
itio
n
s
[
2
4
]
.
M
QT
T
s
er
v
ed
as
th
e
co
m
m
u
n
icatio
n
p
r
o
to
c
o
l
f
o
r
ex
ch
a
n
g
in
g
aler
t
m
ess
ag
es
b
etwe
en
th
e
R
asp
b
er
r
y
Pi
5
,
th
e
E
SP
3
2
m
icr
o
co
n
tr
o
ller
,
a
n
d
th
e
PC
in
ter
f
ac
e.
Fu
r
th
er
m
o
r
e,
t
h
e
g
r
ap
h
ical
in
ter
f
ac
e
was
im
p
lem
en
ted
u
s
in
g
th
e
T
k
i
n
ter
lib
r
ar
y
,
wh
ich
allo
ws
f
o
r
r
ea
l
-
tim
e
v
is
u
aliza
tio
n
o
f
th
e
d
etec
tio
n
r
esu
lts
.
R
eg
ar
d
in
g
h
ar
d
war
e,
th
e
s
y
s
t
em
in
clu
d
es
a
2
2
0
V
AC
to
5
V
DC
p
o
wer
s
u
p
p
ly
to
e
n
s
u
r
e
s
tab
le
o
p
er
atio
n
o
f
all
co
m
p
o
n
en
ts
.
A
L
o
g
itech
C
5
0
5
e
HD
web
ca
m
was
s
elec
ted
as
th
e
d
ata
-
ac
q
u
is
itio
n
d
ev
ice
d
u
e
to
its
s
u
itab
ili
ty
f
o
r
n
o
n
-
i
n
v
a
s
iv
e
f
atig
u
e
-
m
o
n
ito
r
in
g
ap
p
licatio
n
s
,
f
o
llo
win
g
th
e
cu
r
r
en
t
tr
en
d
o
f
av
o
i
d
in
g
wea
r
ab
le
o
r
in
tr
u
s
iv
e
s
en
s
in
g
ap
p
r
o
ac
h
es
[
1
4
]
.
All
in
co
m
in
g
v
id
eo
f
r
am
es
ar
e
p
r
o
ce
s
s
ed
o
n
a
R
asp
b
er
r
y
Pi
5
,
wh
ich
f
u
n
ctio
n
s
as
th
e
m
ain
co
m
p
u
tatio
n
al
u
n
it
an
d
p
r
o
v
i
d
es
s
u
f
f
icien
t
p
er
f
o
r
m
an
ce
to
ex
ec
u
te
lig
h
tweig
h
t
co
m
p
u
ter
-
v
is
io
n
an
d
m
ac
h
in
e
-
lear
n
in
g
alg
o
r
ith
m
s
with
o
u
t
r
eq
u
ir
in
g
ex
te
r
n
al
p
r
o
ce
s
s
in
g
h
ar
d
war
e
[
2
5
]
.
T
h
e
E
SP
3
2
m
icr
o
co
n
tr
o
ller
is
u
s
ed
as
an
ex
ter
n
al
aler
t
m
o
d
u
le,
ac
tiv
atin
g
L
E
Ds
an
d
a
b
u
zz
er
in
r
esp
o
n
s
e
to
f
atig
u
e
ev
en
ts
.
Fig
u
r
e
2
.
Flo
wch
ar
t
o
f
th
e
m
et
h
o
d
o
lo
g
ical
p
r
o
ce
s
s
o
f
th
e
f
ati
g
u
e
d
etec
tio
n
s
y
s
tem
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
R
ea
l
-
time
mu
ltimo
d
a
l fa
tig
u
e
d
etec
tio
n
u
s
in
g
fa
ci
a
l v
is
io
n
a
n
d
a
lert
…
(
A
n
d
r
és E
n
r
iq
u
e
R
o
ja
s
P
r
imo
)
2755
2
.
4
.
Sy
s
t
e
m
des
ig
n
Fig
u
r
e
3
p
r
esen
ts
th
e
o
v
er
all
e
x
ec
u
tio
n
f
lo
w
o
f
th
e
p
r
o
p
o
s
ed
f
atig
u
e
-
m
o
n
ito
r
in
g
s
y
s
tem
.
T
h
e
p
r
o
ce
s
s
b
eg
in
s
with
th
e
i
n
itializatio
n
o
f
th
e
ca
m
er
a
a
n
d
s
o
f
twar
e
m
o
d
u
les,
f
o
llo
we
d
b
y
co
n
tin
u
o
u
s
f
r
am
e
ac
q
u
is
itio
n
an
d
p
r
e
p
r
o
ce
s
s
in
g
.
On
ce
a
f
ac
e
is
d
etec
ted
,
th
e
s
y
s
tem
ev
alu
ates
th
e
ex
tr
ac
ted
f
ea
tu
r
e
s
th
r
o
u
g
h
th
e
e
y
e,
m
o
u
th
,
a
n
d
p
o
s
tu
r
e
d
etec
tio
n
s
u
b
m
o
d
u
les.
W
h
en
an
y
s
ig
n
o
f
f
atig
u
e
is
id
en
tifie
d
,
t
h
e
co
r
r
esp
o
n
d
i
n
g
aler
t
is
tr
an
s
m
itted
v
ia
MQ
T
T
to
b
o
t
h
th
e
E
SP
3
2
an
d
th
e
g
r
a
p
h
ica
l
in
ter
f
ac
e,
e
n
ab
lin
g
th
e
ac
tiv
atio
n
o
f
v
is
u
al
a
n
d
au
d
ib
le
n
o
tific
atio
n
s
.
T
h
e
lo
o
p
co
n
tin
u
es u
n
til th
e
u
s
er
s
to
p
s
th
e
m
o
n
ito
r
i
n
g
s
ess
io
n
.
Fig
u
r
e
3
.
Sy
s
tem
ex
ec
u
tio
n
f
l
o
w
T
o
co
m
p
lem
en
t
th
e
f
lo
wch
ar
t,
Fig
u
r
e
4
illu
s
tr
ates
th
e
in
ter
n
al
s
tr
u
ctu
r
e
o
f
t
h
e
d
etec
tio
n
m
o
d
u
les
r
esp
o
n
s
ib
le
f
o
r
an
al
y
zin
g
f
ac
i
al
ch
ar
ac
ter
is
tics
as
s
o
ciate
d
w
ith
f
atig
u
e.
T
h
is
d
iag
r
am
d
etai
ls
th
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
—
im
p
lem
en
ted
with
Op
en
C
V
an
d
Me
d
iaPip
e
an
d
th
e
ex
tr
ac
tio
n
o
f
k
ey
f
ea
t
u
r
es
s
u
ch
as
th
e
ey
e
asp
ec
t
r
atio
,
p
r
o
lo
n
g
ed
ey
e
clo
s
u
r
e,
h
ea
d
-
tilt
an
g
le,
a
n
d
m
o
u
t
h
-
o
p
en
in
g
s
eg
m
en
tatio
n
.
Me
d
iaPip
e
is
in
co
r
p
o
r
ate
d
as
th
e
p
r
im
ar
y
lan
d
m
a
r
k
-
e
x
tr
ac
tio
n
f
r
am
ew
o
r
k
d
u
e
to
its
ab
ilit
y
to
g
en
e
r
ate
h
ig
h
-
r
eso
lu
tio
n
f
ac
ial
k
ey
p
o
in
ts
in
r
ea
l
tim
e
an
d
m
ain
tain
s
tab
le
t
r
ac
k
in
g
ev
e
n
u
n
d
er
p
a
r
tial
o
cc
lu
s
io
n
s
,
m
ak
in
g
it
s
u
itab
le
f
o
r
id
en
tify
in
g
s
u
b
tle
m
icr
o
-
ex
p
r
ess
io
n
s
an
d
p
o
s
tu
r
al
v
ar
iatio
n
s
r
elev
an
t
to
f
atig
u
e
ass
ess
m
en
t
[
2
2
]
.
T
h
ese
p
ar
am
eter
s
ar
e
th
en
p
r
o
ce
s
s
ed
in
th
e
f
ea
tu
r
e
-
an
aly
s
is
s
tag
e,
wh
er
e
th
e
m
o
u
th
-
s
t
ate
class
if
ier
is
im
p
lem
en
ted
u
s
in
g
a
lig
h
tweig
h
t
C
NN
co
m
b
in
ed
with
tr
an
s
f
o
r
m
er
-
b
ased
c
o
m
p
o
n
en
ts
.
T
o
g
e
th
er
,
th
ese
s
u
b
m
o
d
u
les
p
r
o
v
i
d
e
th
e
m
u
ltimo
d
al
f
o
u
n
d
atio
n
th
at
e
n
ab
les th
e
s
y
s
tem
to
d
etec
t f
atig
u
e
in
d
icato
r
s
in
r
ea
l tim
e.
Fig
u
r
e
4
.
Featu
r
e
an
aly
s
is
an
d
tr
ain
in
g
s
tag
e
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
7
5
0
-
2
7
6
8
2756
Fig
u
r
e
s
5
a
n
d
6
s
h
o
w
t
h
e
c
o
m
m
u
n
i
ca
t
io
n
an
d
ale
r
t
s
u
b
s
y
s
te
m
t
h
at
s
u
p
p
o
r
ts
th
e
s
o
f
t
wa
r
e
a
r
c
h
it
ec
t
u
r
e
.
T
h
e
R
asp
b
er
r
y
Pi
5
e
x
ec
u
te
s
th
e
d
etec
tio
n
alg
o
r
ith
m
s
a
n
d
p
u
b
lis
h
es
f
atig
u
e
e
v
en
ts
t
h
r
o
u
g
h
th
e
MQ
T
T
p
r
o
to
co
l,
wh
ile
th
e
E
SP
3
2
r
ec
eiv
es
th
ese
m
es
s
ag
es
an
d
m
an
ag
es
th
e
co
r
r
esp
o
n
d
in
g
aler
t
ac
tio
n
s
.
T
h
e
v
is
u
al
aler
t
is
g
en
er
ated
th
r
o
u
g
h
h
ig
h
-
in
ten
s
ity
L
E
Ds,
an
d
th
e
au
d
ib
le
aler
t
is
ac
tiv
ated
th
r
o
u
g
h
a
b
u
zz
er
co
n
n
ec
te
d
to
th
e
E
SP
3
2
.
T
h
is
m
o
d
u
lar
c
o
m
m
u
n
icatio
n
d
esig
n
en
s
u
r
es
th
at
aler
ts
ca
n
b
e
tr
ig
g
er
ed
with
m
in
im
al
laten
cy
,
in
d
ep
en
d
en
tly
o
f
th
e
co
m
p
u
tat
io
n
al
lo
ad
o
f
th
e
d
etec
tio
n
m
o
d
u
les.
F
i
g
u
r
e
5
.
M
Q
T
T
-
b
as
e
d
c
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
a
l
e
r
t
s
u
b
s
y
s
t
e
m
b
e
t
w
e
e
n
t
h
e
R
as
p
b
e
r
r
y
P
i
5
a
n
d
t
h
e
E
S
P
3
2
m
o
d
u
l
e
Fig
u
r
e
6
.
MQ
T
T
-
p
h
y
s
ical
p
r
o
t
o
ty
p
e
o
f
th
e
E
SP
3
2
-
b
ased
aler
t m
o
d
u
le
u
s
ed
d
u
r
in
g
r
ea
l
-
tim
e
v
alid
atio
n
2
.
5
.
Va
lid
a
t
i
o
n
pro
ce
du
re
A
s
et
o
f
s
y
n
th
etic
v
id
eo
s
g
e
n
e
r
ated
u
s
in
g
d
if
f
u
s
io
n
-
b
ased
m
o
d
els
(
AI
-
d
r
i
v
en
f
ac
ial
g
e
n
er
a
tio
n
to
o
ls
)
was
u
s
ed
f
o
r
s
y
s
tem
v
alid
atio
n
.
T
h
is
s
tu
d
y
r
elies
ex
clu
s
iv
ely
o
n
s
y
n
th
etic
v
id
e
o
s
to
en
s
u
r
e
co
n
tr
o
lled
a
n
d
r
ep
r
o
d
u
cib
le
v
alid
atio
n
co
n
d
itio
n
s
.
T
h
ese
v
id
eo
s
wer
e
d
esig
n
ed
to
r
ep
r
esen
t
co
n
tr
o
lled
v
is
u
al
s
tates
ass
o
ciate
d
with
b
o
th
f
atig
u
e
an
d
n
o
n
-
f
atig
u
e
co
n
d
itio
n
s
,
in
clu
d
in
g
h
alf
-
clo
s
ed
ey
es,
y
awn
in
g
,
a
n
d
h
ea
d
tilt
in
g
.
Af
ter
g
en
er
atio
n
,
all
im
ag
es
wer
e
m
an
u
ally
r
e
v
iewe
d
an
d
class
if
ied
to
estab
lis
h
s
y
n
th
etic
g
r
o
u
n
d
tr
u
th
,
wh
ich
s
er
v
ed
as
an
o
b
je
ctiv
e
r
ef
er
en
ce
f
o
r
co
m
p
ar
in
g
th
e
s
y
s
tem
’
s
o
u
tp
u
ts
an
d
ev
al
u
atin
g
its
d
etec
tio
n
ca
p
ab
ilit
y
u
n
d
er
s
tab
le
an
d
r
ep
r
o
d
u
cib
le
v
is
u
al
co
n
d
itio
n
s
.
T
h
is
ap
p
r
o
ac
h
alig
n
s
with
m
eth
o
d
o
lo
g
ical
p
r
ac
tices
r
ep
o
r
ted
in
t
h
e
lit
er
atu
r
e,
wh
ich
h
ig
h
lig
h
t
th
e
lim
ited
a
v
ailab
ilit
y
o
f
s
ta
n
d
ar
d
ized
d
atasets
s
p
ec
if
ically
d
esig
n
ed
f
o
r
o
cc
u
p
atio
n
al
f
atig
u
e
d
etec
tio
n
u
n
d
er
r
ea
l
-
wo
r
ld
co
n
d
itio
n
s
,
p
ar
ticu
lar
ly
f
o
r
b
li
n
k
an
d
m
u
ltimo
d
al
b
e
h
av
io
r
al
an
aly
s
is
.
C
o
n
s
eq
u
en
tly
,
m
an
u
al
an
n
o
tatio
n
r
em
ai
n
s
a
co
m
m
o
n
ly
ad
o
p
ted
s
tr
ateg
y
f
o
r
co
n
s
tr
u
ctin
g
g
r
o
u
n
d
-
tr
u
th
l
ab
els in
ex
p
er
im
e
n
tal
ev
alu
ati
o
n
s
[
2
6
]
,
[
2
7
]
On
ce
th
is
r
ef
er
en
ce
d
ataset
was
d
ef
in
ed
,
th
e
s
y
s
tem
o
u
tp
u
ts
wer
e
co
m
p
ar
ed
with
th
e
m
a
n
u
al
lab
els
to
co
m
p
u
te
class
ical
p
er
f
o
r
m
an
ce
m
etr
ics,
in
clu
d
i
n
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
th
e
F1
-
s
co
r
e
f
o
r
ea
ch
ev
en
t
an
d
ca
teg
o
r
y
.
I
n
p
ar
a
llel,
th
e
MQ
T
T
p
r
o
to
c
o
l
w
as
em
p
lo
y
ed
as
a
lig
h
tweig
h
t
co
m
m
u
n
icatio
n
m
ec
h
an
is
m
b
etwe
en
th
e
m
ain
p
r
o
ce
s
s
in
g
m
o
d
u
le
(
R
asp
b
er
r
y
Pi)
an
d
th
e
E
SP
3
2
m
icr
o
co
n
tr
o
ller
r
esp
o
n
s
ib
le
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
R
ea
l
-
time
mu
ltimo
d
a
l fa
tig
u
e
d
etec
tio
n
u
s
in
g
fa
ci
a
l v
is
io
n
a
n
d
a
lert
…
(
A
n
d
r
és E
n
r
iq
u
e
R
o
ja
s
P
r
imo
)
2757
f
o
r
ex
ec
u
tin
g
p
h
y
s
ical
aler
ts
.
T
h
e
s
elec
tio
n
o
f
MQ
T
T
is
s
u
p
p
o
r
ted
b
y
its
ex
ten
s
iv
e
u
s
e
in
r
ea
l
-
tim
e
m
o
n
ito
r
in
g
s
y
s
tem
s
an
d
its
p
r
o
v
e
n
r
eliab
ilit
y
f
o
r
tr
an
s
m
itti
n
g
cr
itical
d
ata
in
I
o
T
en
v
ir
o
n
m
en
ts
,
as
d
o
cu
m
e
n
ted
in
p
r
ev
i
o
u
s
s
tu
d
ies
o
n
v
is
io
n
-
b
ased
f
atig
u
e
d
e
tectio
n
s
y
s
tem
s
[
2
8
]
.
Du
e
to
its
lo
w
laten
cy
an
d
m
in
im
al
r
eso
u
r
ce
co
n
s
u
m
p
tio
n
,
MQ
T
T
en
s
u
r
es
th
at
d
etec
t
ed
f
atig
u
e
ev
en
ts
ar
e
tr
an
s
m
i
tted
im
m
ed
iately
,
en
ab
lin
g
tim
ely
ac
tiv
atio
n
o
f
v
is
u
al
an
d
au
d
ito
r
y
aler
ts
o
n
th
e
E
SP
3
2
[
6
]
.
2
.
6
.
E
v
a
lua
t
i
o
n
cr
it
er
ia
T
o
ass
ess
th
e
s
y
s
tem
’
s
p
er
f
o
r
m
an
ce
,
th
e
f
o
llo
win
g
k
ey
m
etr
ics we
r
e
co
n
s
id
er
ed
:
a.
Dete
ctio
n
ac
cu
r
ac
y
: th
e
p
er
ce
n
tag
e
o
f
c
o
r
r
ec
tly
i
d
en
tifie
d
f
a
tig
u
e
-
r
elate
d
ev
e
n
ts
,
b.
Fals
e
p
o
s
itiv
e
r
ate:
th
e
p
er
ce
n
t
ag
e
o
f
in
c
o
r
r
ec
t a
ler
ts
g
e
n
er
at
ed
,
c.
R
esp
o
n
s
e
laten
cy
: th
e
elap
s
ed
tim
e
b
etwe
en
d
etec
tio
n
a
n
d
al
er
t a
ctiv
atio
n
,
d.
Aler
t r
eliab
ilit
y
: th
e
ef
f
ec
tiv
en
ess
o
f
v
is
u
al
an
d
au
d
ito
r
y
ale
r
t d
eliv
er
y
.
I
n
a
d
d
itio
n
,
th
e
aler
t
ac
tiv
ati
o
n
th
r
esh
o
ld
was
f
in
e
-
t
u
n
ed
t
o
en
s
u
r
e
tim
ely
an
d
ac
cu
r
ate
r
esp
o
n
s
es
u
n
d
er
v
ar
y
i
n
g
lig
h
tin
g
co
n
d
it
io
n
s
an
d
f
ac
ial
o
r
ien
tatio
n
s
,
t
h
er
eb
y
e
n
h
an
cin
g
th
e
r
o
b
u
s
tn
ess
an
d
s
tab
ilit
y
o
f
th
e
s
y
s
tem
ac
r
o
s
s
d
iv
er
s
e
wo
r
k
p
lace
en
v
ir
o
n
m
e
n
ts
2
.
7
.
M
a
t
hema
t
ica
l
f
o
rm
ula
t
i
o
n o
f
t
he
det
ec
t
io
n a
l
g
o
rit
h
m
s
T
h
e
p
r
o
p
o
s
ed
d
etec
tio
n
s
y
s
te
m
r
elies
o
n
a
s
et
o
f
g
e
o
m
etr
ic
an
d
s
tatis
tical
m
o
d
els
d
e
r
iv
ed
f
r
o
m
f
ac
ial
lan
d
m
ar
k
s
.
T
h
ese
f
o
r
m
u
latio
n
s
d
escr
ib
e
h
o
w
ey
e
o
p
en
n
ess
,
m
o
u
th
b
eh
av
io
r
,
an
d
h
ea
d
o
r
ien
tatio
n
a
r
e
q
u
an
tifie
d
f
r
o
m
ea
c
h
v
id
eo
f
r
a
m
e.
All
f
ac
ial
co
o
r
d
in
ates
ar
e
o
b
tain
ed
u
s
in
g
t
h
e
Me
d
iaPip
e
Face
Me
s
h
m
o
d
el,
wh
ich
p
r
o
v
i
d
es
n
o
r
m
alize
d
l
an
d
m
ar
k
p
o
s
itio
n
s
th
at
ar
e
later
co
n
v
er
ted
in
t
o
p
ix
el
-
b
as
ed
co
o
r
d
i
n
ates
f
o
r
p
r
o
ce
s
s
in
g
.
2
.
7
.
1
.
E
uclid
ea
n di
s
t
a
nce
T
h
e
E
u
clid
ea
n
d
is
tan
ce
is
u
s
ed
to
co
m
p
u
te
th
e
g
eo
m
etr
ic
s
ep
ar
atio
n
b
etwe
en
f
ac
ial
lan
d
m
ar
k
s
i
n
ea
ch
f
r
am
e,
f
o
r
m
in
g
th
e
b
asis
f
o
r
th
e
f
ea
tu
r
es e
x
tr
ac
ted
i
n
th
e
ey
e
an
d
h
ea
d
-
p
o
s
e
m
o
d
u
les.
Fo
r
an
y
two
p
o
in
ts
=
(
,
)
an
d
=
(
,
)
,
th
e
E
u
clid
ea
n
d
is
tan
ce
i
s
d
ef
in
ed
as:
(
,
)
=
√
(
−
)
2
+
(
−
)
2
(
1
)
W
h
er
e
an
d
r
ep
r
esen
t
two
f
ac
ial
lan
d
m
ar
k
s
s
elec
ted
ac
co
r
d
in
g
to
th
e
in
d
ex
p
o
s
itio
n
s
d
ef
in
ed
b
y
th
e
Face
Me
s
h
m
o
d
el
(
e.
g
.
,
3
3
,
1
6
0
,
1
5
8
f
o
r
t
h
e
ey
e
r
eg
i
o
n
)
,
,
d
e
n
o
te
th
e
h
o
r
izo
n
tal
p
ix
el
c
o
o
r
d
in
ates
o
f
th
e
lan
d
m
ar
k
s
,
r
ep
r
esen
t
t
h
eir
v
e
r
tical
p
ix
el
co
o
r
d
in
ates,
an
d
,
(
,
)
is
th
e
g
e
o
m
etr
ic
m
ea
s
u
r
em
e
n
t
u
s
ed
to
co
m
p
u
te
th
e
ey
elid
d
is
tan
ce
s
in
th
e
E
AR
f
o
r
m
u
latio
n
an
d
th
e
v
er
tical
an
d
h
o
r
izo
n
tal
co
m
p
o
n
en
ts
in
v
o
lv
e
d
in
h
ea
d
-
tilt
esti
m
atio
n
.
T
h
is
m
etr
ic
is
n
o
t
ap
p
lied
to
th
e
m
o
u
th
s
u
b
m
o
d
u
le,
as
m
o
u
th
o
p
en
i
n
g
is
ev
alu
ated
u
s
in
g
a
C
NN
-
b
ased
class
if
ier
r
ath
er
th
an
g
e
o
m
etr
ic
r
atio
s
.
2
.
7
.
2
.
E
AR
Af
ter
co
m
p
u
tin
g
th
e
lan
d
m
a
r
k
d
is
tan
ce
s
,
th
e
E
AR
is
u
s
ed
to
q
u
an
tify
ey
elid
o
p
en
n
ess
.
T
h
is
g
eo
m
etr
ic
in
d
icato
r
h
as
b
ee
n
wid
ely
ad
o
p
ted
i
n
f
atig
u
e
an
d
d
r
o
wsi
n
ess
d
etec
tio
n
b
ec
au
s
e
it
p
r
o
v
id
es
a
s
tab
le
r
ep
r
esen
tatio
n
o
f
ey
e
clo
s
u
r
e
ev
en
u
n
d
er
v
ar
y
i
n
g
illu
m
in
a
tio
n
o
r
h
ea
d
p
o
s
e
co
n
d
itio
n
s
[
2
9
]
.
T
h
e
E
AR
is
d
ef
in
ed
as
(
2
)
:
(
)
=
+
(
2
)
W
h
er
e
A
an
d
B
d
en
o
te
th
e
v
er
tical
d
is
tan
ce
s
b
etwe
en
th
e
ey
elid
s
,
C
is
th
e
h
o
r
izo
n
tal
d
is
ta
n
ce
b
etwe
en
th
e
ey
e
co
r
n
er
s
,
an
d
(
)
r
ep
r
esen
ts
th
e
e
y
elid
-
o
p
e
n
in
g
r
atio
at
tim
e
.
Hig
h
er
E
AR
v
alu
es
co
r
r
esp
o
n
d
to
o
p
en
ey
es,
wh
ile
lo
wer
v
alu
es
in
d
ic
ate
ey
elid
clo
s
u
r
e,
a
b
eh
a
v
io
r
s
tr
o
n
g
ly
ass
o
ciate
d
with
ea
r
ly
f
atig
u
e
s
y
m
p
to
m
s
[
2
9
]
.
2
.
7
.
3
.
Ada
ptiv
e
E
AR
ba
s
elin
e
m
o
del
Sin
ce
n
atu
r
al
ey
e
o
p
e
n
in
g
v
ar
ies
f
r
o
m
p
er
s
o
n
t
o
p
e
r
s
o
n
an
d
ca
n
ch
an
g
e
s
lig
h
tly
o
v
er
tim
e,
th
e
s
y
s
tem
m
ain
tain
s
a
d
y
n
am
ic
r
ef
er
en
ce
v
alu
e
th
at
ad
ap
ts
to
t
h
e
u
s
er
'
s
ty
p
ical
ey
elid
b
eh
av
i
o
r
.
T
h
is
r
ef
er
e
n
ce
is
u
p
d
ated
u
s
in
g
th
e
in
s
tan
tan
eo
u
s
E
AR
o
b
tain
ed
f
r
o
m
(
2
)
,
wh
ich
p
r
o
v
id
es
th
e
ey
e
-
o
p
en
i
n
g
m
ea
s
u
r
em
e
n
t
in
ea
ch
f
r
am
e.
T
h
e
ad
a
p
tatio
n
p
r
o
ce
s
s
f
o
llo
ws an
ex
p
o
n
en
tial sm
o
o
th
in
g
f
o
r
m
u
latio
n
:
r
ef
(
)
=
(
1
−
)
r
ef
(
−
1
)
+
(
)
(
3
)
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
7
5
0
-
2
7
6
8
2758
W
h
er
e
(
)
r
ep
r
esen
ts
th
e
u
p
d
ated
r
ef
er
en
ce
lev
el
at
tim
e
,
wh
ile
(
)
co
r
r
esp
o
n
d
s
to
th
e
cu
r
r
en
t
E
AR
co
m
p
u
ted
f
r
o
m
(
2
)
.
T
h
e
p
ar
am
eter
ac
ts
as
a
s
m
o
o
th
in
g
f
ac
to
r
th
at
co
n
tr
o
ls
h
o
w
q
u
i
ck
ly
th
e
r
ef
e
r
en
ce
ad
ju
s
ts
to
g
r
ad
u
al
v
ar
iatio
n
s
.
T
h
is
ap
p
r
o
ac
h
s
tab
ilizes
th
e
ey
e
-
o
p
e
n
in
g
b
aselin
e
an
d
p
r
ev
en
ts
s
h
o
r
t
-
ter
m
f
lu
ctu
atio
n
s
s
u
ch
as
m
icr
o
b
lin
k
s
o
r
n
o
is
e
in
la
n
d
m
ar
k
d
etec
tio
n
f
r
o
m
b
ei
n
g
m
is
in
ter
p
r
eted
as
s
ig
n
s
o
f
f
atig
u
e.
2
.
7
.
4
.
P
ro
lo
ng
ed
ey
e
clo
s
ure
cr
it
er
io
n
On
ce
th
e
in
s
tan
tan
eo
u
s
E
AR
h
as
b
ee
n
co
m
p
u
ted
u
s
in
g
(
2
)
an
d
th
e
ad
a
p
tiv
e
r
ef
er
e
n
ce
lev
el
h
as
b
ee
n
u
p
d
ated
t
h
r
o
u
g
h
(
3
)
,
th
e
s
y
s
tem
ev
alu
ates
wh
eth
er
t
h
e
ey
e
is
ab
n
o
r
m
ally
clo
s
ed
.
T
h
e
d
ec
is
io
n
r
elies
o
n
co
m
p
ar
in
g
th
e
u
s
er
’
s
n
atu
r
al
ey
e
–
o
p
e
n
in
g
b
aselin
e
with
th
e
cu
r
r
en
t
e
y
elid
co
n
f
ig
u
r
atio
n
.
T
h
is
co
m
p
ar
is
o
n
is
ex
p
r
ess
ed
as
(
4
)
:
r
ef
(
)
−
(
)
≥
∆
ℎ
(
4
)
I
n
th
is
e
x
p
r
ess
io
n
,
(
)
r
ep
r
esen
ts
th
e
i
n
s
tan
tan
eo
u
s
e
y
e
-
o
p
en
in
g
r
atio
d
ef
i
n
ed
i
n
(
2
)
,
wh
ile
r
ef
(
)
is
th
e
s
m
o
o
th
e
d
b
aselin
e
o
b
tai
n
ed
f
r
o
m
(
3
)
.
T
h
e
s
u
b
tr
ac
tio
n
r
ef
(
)
−
(
)
q
u
an
tifie
s
h
o
w
m
u
ch
t
h
e
cu
r
r
en
t
e
y
elid
p
o
s
itio
n
d
ev
iat
es
f
r
o
m
th
e
u
s
er
’
s
ty
p
ical
ey
e
-
o
p
en
in
g
lev
el.
W
h
en
th
is
d
e
v
iatio
n
ex
ce
e
d
s
th
e
th
r
esh
o
ld
∆
ℎ
,
th
e
s
y
s
tem
in
ter
p
r
e
ts
th
e
ev
en
t a
s
a
clo
s
u
r
e
d
ee
p
e
r
th
an
ex
p
ec
ted
f
r
o
m
n
atu
r
al
b
lin
k
in
g
.
2
.
7
.
5
.
H
ea
d
-
t
ilt
a
ng
le
f
o
r
po
s
t
ure
a
s
s
ess
m
ent
T
h
e
p
o
s
tu
r
e
d
etec
tio
n
m
o
d
u
le
ev
alu
ates
wh
eth
er
th
e
u
s
er
’
s
h
ea
d
d
e
v
iates
f
r
o
m
an
u
p
r
i
g
h
t
p
o
s
itio
n
b
y
c
o
m
b
in
in
g
a
n
g
u
lar
an
d
v
er
tical
d
is
p
lace
m
en
t
m
ea
s
u
r
es.
T
wo
f
ac
ial
la
n
d
m
ar
k
s
,
th
e
n
o
s
e
tip
an
d
th
e
ch
in
,
ar
e
u
s
ed
to
esti
m
ate
h
ea
d
o
r
ien
tatio
n
in
ea
ch
f
r
am
e
.
T
h
e
h
ea
d
-
tilt
an
g
le
is
co
m
p
u
ted
f
r
o
m
th
eir
r
elativ
e
p
o
s
itio
n
s
as
(
5
)
:
(
)
=
|
−
1
(
ℎ
−
ℎ
−
)
|
(
5
)
W
h
er
e,
(
ℎ
,
ℎ
)
an
d
(
,
)
r
ep
r
esen
t
th
e
c
o
o
r
d
in
ates
o
f
th
e
ch
in
a
n
d
n
o
s
e,
r
esp
ec
tiv
ely
.
T
h
e
r
esu
ltin
g
an
g
le
r
ef
lects
th
e
d
e
v
iatio
n
f
r
o
m
a
n
u
p
r
i
g
h
t
p
o
s
tu
r
e
an
d
is
co
m
p
ar
ed
ag
ai
n
s
t
a
p
r
ed
ef
in
ed
th
r
esh
o
ld
to
id
en
tify
a
b
n
o
r
m
al
h
ea
d
tilt
.
I
n
ad
d
itio
n
to
a
n
g
u
lar
a
n
aly
s
is
,
v
er
tical
h
ea
d
m
o
v
em
e
n
t
is
ass
es
s
ed
b
y
tr
ac
k
in
g
th
e
d
is
p
l
ac
em
en
t
o
f
th
e
n
o
s
e
r
elativ
e
to
a
r
ef
er
en
ce
p
o
s
itio
n
r
ec
o
r
d
ed
u
n
d
er
n
eu
tr
al
co
n
d
itio
n
s
.
T
h
is
d
is
p
lace
m
en
t
is
co
m
p
u
ted
u
s
in
g
th
e
E
u
clid
ea
n
d
is
tan
ce
d
ef
in
ed
in
(
1
)
,
ass
u
m
in
g
m
in
im
al
h
o
r
izo
n
tal
v
ar
iatio
n
.
W
h
en
th
e
v
er
tical
d
is
p
lace
m
en
t
ex
ce
e
d
s
a
p
r
ed
e
f
in
ed
t
h
r
esh
o
ld
o
v
er
a
s
u
s
tain
ed
p
e
r
io
d
,
it
is
class
if
ied
as
a
n
ab
n
o
r
m
al
p
o
s
tu
r
e.
B
y
co
m
b
in
in
g
b
o
th
a
n
g
u
la
r
an
d
v
e
r
tical
cr
iter
ia,
t
h
e
s
y
s
tem
d
is
tin
g
u
is
h
es
n
o
r
m
al
h
ea
d
m
o
v
em
en
ts
f
r
o
m
f
atig
u
e
-
r
elate
d
d
ev
iatio
n
s
.
2
.
7
.
6
.
M
o
uth
-
o
penin
g
pro
ba
bil
it
y
m
o
del
T
h
e
m
o
u
th
-
s
tate
esti
m
atio
n
is
p
er
f
o
r
m
ed
u
s
in
g
a
h
y
b
r
id
m
o
d
el
th
at
co
m
b
in
es
co
n
v
o
l
u
tio
n
al
an
d
tr
an
s
f
o
r
m
er
-
b
ased
co
m
p
o
n
en
t
s
to
an
aly
ze
p
atter
n
s
f
r
o
m
th
e
cr
o
p
p
e
d
m
o
u
th
r
e
g
io
n
.
As
s
h
o
wn
in
Fig
u
r
e
4
,
th
e
in
p
u
t
im
ag
e
is
f
ir
s
t
p
r
o
ce
s
s
ed
b
y
a
Mo
b
ileNetV2
b
ac
k
b
o
n
e
to
ex
tr
ac
t
s
p
atial
f
ea
tu
r
es,
wh
ich
ar
e
th
en
r
ef
in
ed
b
y
a
Mo
b
ileViT
b
lo
ck
to
ca
p
t
u
r
e
co
n
tex
tu
al
in
f
o
r
m
atio
n
[
3
0
]
,
[
3
1
]
.
A
g
lo
b
al
av
er
a
g
e
p
o
o
lin
g
lay
er
r
ed
u
ce
s
th
e
f
ea
tu
r
e
m
ap
s
,
an
d
th
e
f
in
al
p
r
ed
ictio
n
is
o
b
tain
ed
th
r
o
u
g
h
a
lig
h
tweig
h
t
d
e
n
s
e
class
if
ier
co
m
p
o
s
ed
o
f
a
R
eL
U
-
ac
tiv
ated
lay
er
,
a
d
r
o
p
o
u
t
s
tag
e,
an
d
a
s
ig
m
o
id
o
u
tp
u
t.
T
h
is
p
r
o
ce
s
s
y
ield
s
a
p
r
o
b
a
b
ilit
y
v
alu
e
r
ep
r
esen
tin
g
th
e
lik
elih
o
o
d
th
a
t th
e
m
o
u
th
is
o
p
en
.
(
|
)
=
−
(
)
(
6
)
I
n
th
is
ex
p
r
ess
io
n
,
d
en
o
tes
th
e
m
o
u
th
im
a
g
e
at
tim
e
,
an
d
−
(
)
r
ep
r
esen
ts
th
e
m
ap
p
i
n
g
f
r
o
m
t
h
e
in
p
u
t
r
e
g
io
n
o
f
in
ter
est
to
a
p
r
o
b
ab
ilit
y
v
al
u
e
in
th
e
r
a
n
g
e
[
0
,
1
]
.
T
h
is
p
r
o
b
a
b
ilis
tic
o
u
tp
u
t
is
u
s
ed
t
o
d
is
tin
g
u
is
h
b
etwe
en
n
o
r
m
al
m
o
u
th
m
o
v
em
en
ts
an
d
y
awn
in
g
-
r
elate
d
p
atter
n
s
.
2
.
7
.
7
.
Ya
wn
-
det
ec
t
io
n t
hresh
o
ld
On
ce
th
e
p
r
o
b
a
b
ilit
y
o
f
m
o
u
t
h
o
p
e
n
in
g
h
as
b
ee
n
co
m
p
u
ted
b
y
(
6
)
,
th
e
s
y
s
tem
d
eter
m
i
n
e
s
wh
eth
er
th
e
u
s
er
is
ex
h
ib
itin
g
y
awn
-
lik
e
b
eh
av
io
r
b
y
co
m
p
a
r
in
g
th
i
s
p
r
o
b
ab
ilit
y
to
a
p
r
ed
ef
in
e
d
d
ec
is
io
n
th
r
esh
o
ld
.
T
h
is
r
u
le
is
ex
p
r
ess
ed
as
(
7
)
:
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
R
ea
l
-
time
mu
ltimo
d
a
l fa
tig
u
e
d
etec
tio
n
u
s
in
g
fa
ci
a
l v
is
io
n
a
n
d
a
lert
…
(
A
n
d
r
és E
n
r
iq
u
e
R
o
ja
s
P
r
imo
)
2759
P
(
ope
n
|
I
t
)
≥
(
7
)
Her
e,
d
en
o
tes
th
e
th
r
esh
o
l
d
t
h
at
s
ep
ar
ates
n
o
r
m
al
o
r
al
ac
ti
v
ity
f
r
o
m
s
ig
n
if
ican
t
o
p
en
-
m
o
u
th
s
tates.
Fo
r
th
e
r
esear
ch
,
a
v
alu
e
o
f
0
.
5
was
estab
lis
h
ed
.
W
h
en
th
e
p
r
o
b
ab
ilit
y
co
n
s
is
ten
tly
ex
ce
ed
s
th
is
v
alu
e
in
co
n
s
ec
u
tiv
e
f
r
am
es,
th
e
s
y
s
tem
in
ter
p
r
ets
th
e
b
eh
av
i
o
r
as
a
y
awn
r
ath
e
r
th
an
a
tr
an
s
ien
t
ar
ticu
latio
n
o
r
r
a
n
d
o
m
m
o
u
t
h
m
o
v
em
en
t.
T
h
is
th
r
esh
o
l
d
-
b
ased
cr
iter
io
n
all
o
ws
f
o
r
r
eliab
le
y
awn
d
etec
tio
n
wh
ile
m
ain
tain
in
g
co
m
p
u
tatio
n
al
ef
f
icien
c
y
,
as
it
av
o
id
s
th
e
n
ee
d
f
o
r
a
d
d
itio
n
al
tem
p
o
r
al
m
o
d
elin
g
o
r
s
eq
u
e
n
ce
-
b
ased
class
if
icatio
n
.
2
.
8
.
Dec
is
io
n
cr
it
er
ia
a
nd
t
h
re
s
ho
ld s
elec
t
io
n
T
h
e
f
atig
u
e
d
etec
tio
n
s
y
s
tem
co
m
b
in
es
th
e
m
ath
em
atica
l
f
o
r
m
u
latio
n
s
d
escr
ib
e
d
in
Sectio
n
2
.
7
with
a
s
et
o
f
em
p
ir
ically
d
ef
in
ed
th
r
esh
o
ld
s
to
d
eter
m
in
e
f
atig
u
e
-
r
elate
d
ev
en
ts
in
r
ea
l
tim
e.
Fo
r
o
cu
lar
an
aly
s
is
,
a
d
y
n
am
ic
E
AR
b
aselin
e
is
co
m
p
u
ted
u
s
in
g
ex
p
o
n
e
n
tial
s
m
o
o
th
in
g
with
a
s
m
o
o
th
in
g
f
a
cto
r
o
f
0
.
0
2
.
A
clo
s
u
r
e
ev
en
t
is
d
etec
ted
wh
en
th
e
d
if
f
er
en
ce
b
etwe
en
th
e
b
aselin
e
an
d
th
e
in
s
tan
tan
eo
u
s
E
AR
ex
ce
ed
s
a
th
r
esh
o
ld
o
f
0
.
0
4
f
o
r
at
least
two
co
n
s
ec
u
tiv
e
f
r
a
m
es.
Sh
o
r
t
clo
s
u
r
es
last
i
n
g
less
th
an
0
.
4
0
s
ec
o
n
d
s
a
r
e
class
if
ied
as
n
o
r
m
al
b
lin
k
s
,
wh
ile
p
r
o
l
o
n
g
e
d
ey
e
cl
o
s
u
r
e
is
d
ef
in
ed
as a
d
u
r
atio
n
ex
ce
ed
in
g
5
s
ec
o
n
d
s
.
Fo
r
m
o
u
t
h
-
s
tate
an
aly
s
is
,
a
p
r
o
b
ab
ilit
y
th
r
esh
o
ld
o
f
0
.
5
is
u
s
ed
to
class
if
y
th
e
o
u
tp
u
t
o
f
t
h
e
C
NN
-
b
ased
m
o
d
el
as
o
p
en
o
r
clo
s
ed
.
A
y
awn
ev
en
t
is
d
etec
te
d
wh
en
th
e
m
o
u
th
r
em
ai
n
s
o
p
en
f
o
r
at
least
2
0
co
n
s
ec
u
tiv
e
f
r
am
es.
I
n
th
e
p
o
s
tu
r
e
m
o
d
u
le,
f
atig
u
e
-
r
elate
d
d
ev
iatio
n
s
ar
e
id
e
n
tifie
d
u
s
in
g
b
o
th
an
g
u
lar
a
n
d
v
e
r
tical
cr
iter
ia.
A
h
ea
d
-
tilt
an
g
le
g
r
e
ater
th
an
1
3
°
is
co
n
s
id
er
ed
ab
n
o
r
m
al,
wh
ile
v
er
tical
d
is
p
lac
em
en
t
is
ev
alu
ated
u
s
in
g
a
th
r
esh
o
ld
o
f
4
0
p
ix
els
r
elativ
e
t
o
a
r
ef
er
en
ce
p
o
s
itio
n
.
I
n
b
o
th
ca
s
es,
th
e
c
o
n
d
itio
n
m
u
s
t
p
e
r
s
is
t
f
o
r
at
least 6
s
ec
o
n
d
s
to
b
e
class
if
ied
as a
f
atig
u
e
-
r
elate
d
p
o
s
tu
r
e
e
v
en
t.
3.
RE
SU
L
T
S
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
r
esu
l
ts
o
b
tain
ed
d
u
r
in
g
th
e
d
e
v
elo
p
m
en
t
an
d
im
p
lem
en
tatio
n
o
f
th
e
f
atig
u
e
d
etec
tio
n
s
y
s
tem
b
ased
o
n
c
o
m
p
u
ter
v
is
io
n
.
T
h
e
f
i
n
d
in
g
s
ar
e
s
tr
u
ctu
r
ed
ac
co
r
d
in
g
to
th
e
s
p
ec
if
ic
o
b
jectiv
es o
f
th
e
s
tu
d
y
,
co
v
e
r
in
g
th
e
e
v
alu
atio
n
o
f
th
e
m
u
ltimo
d
al
d
etec
tio
n
m
o
d
el,
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
MQ
T
T
aler
t
s
y
s
tem
,
th
e
v
alid
atio
n
o
f
th
e
s
y
s
tem
in
a
s
im
u
lated
en
v
ir
o
n
m
en
t
with
s
y
n
th
etic
v
id
eo
s
,
an
d
th
e
v
er
if
icatio
n
o
f
th
e
f
u
n
ctio
n
ality
o
f
t
h
e
g
r
a
p
h
i
ca
l in
ter
f
ac
e
d
u
r
in
g
r
ea
l
-
tim
e
m
o
n
ito
r
in
g
.
3
.
1
.
P
er
f
o
r
m
a
nce
o
f
t
he
im
p
lem
ent
ed
a
lg
o
rit
hm
s
T
o
ev
al
u
ate
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
p
r
o
p
o
s
ed
f
atig
u
e
-
d
etec
t
io
n
s
y
s
tem
,
ea
ch
d
etec
tio
n
m
o
d
u
le
was
ass
es
s
ed
in
d
ep
en
d
e
n
tly
u
s
in
g
v
alid
atio
n
d
atasets
r
ep
r
esen
tat
iv
e
o
f
r
ea
l
o
p
er
atin
g
co
n
d
itio
n
s
.
T
h
e
m
o
u
th
-
s
tate
class
if
icatio
n
m
o
d
u
le,
b
ased
o
n
th
e
h
y
b
r
id
Mo
b
ileNetV2
–
Mo
b
ileViT
ar
ch
itectu
r
e,
d
em
o
n
s
tr
ated
h
i
g
h
r
eliab
ilit
y
in
d
is
tin
g
u
is
h
in
g
b
e
twee
n
o
p
en
an
d
clo
s
ed
m
o
u
th
s
tates.
T
ab
le
1
s
u
m
m
ar
izes
t
h
e
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
o
b
tain
ed
f
o
r
ea
c
h
class
wh
en
ap
p
ly
in
g
a
d
ec
is
i
o
n
th
r
esh
o
l
d
o
f
0
.
5
.
T
ab
le
1
.
C
lass
if
icatio
n
m
etr
ics f
o
r
th
e
m
o
u
t
h
-
s
tate
class
if
ier
C
l
a
s
s
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
S
u
p
p
o
r
t
C
l
o
se
d
0
.
9
9
0
.
9
8
0
.
9
8
5
0
1
O
p
e
n
0
.
9
8
0
.
9
9
0
.
9
8
5
0
2
T
h
e
class
if
ier
ac
h
iev
ed
an
o
v
er
all
ac
cu
r
ac
y
o
f
0
.
9
8
o
n
th
e
v
alid
atio
n
s
et.
T
h
is
r
esu
lt
in
d
icate
s
th
at
b
o
th
class
es
ar
e
r
ec
o
g
n
ized
with
a
s
im
ilar
lev
el
o
f
r
el
iab
ilit
y
,
wh
ich
is
ex
p
ec
ted
g
iv
en
th
e
b
alan
ce
d
d
is
tr
ib
u
tio
n
o
f
s
am
p
les in
th
e
d
ataset.
T
h
e
lear
n
in
g
b
eh
a
v
io
r
o
f
th
e
m
o
d
el
is
illu
s
tr
ated
in
Fig
u
r
e
7
an
d
Fig
u
r
e
8
,
w
h
ich
s
h
o
w
t
h
e
ac
cu
r
ac
y
an
d
lo
s
s
cu
r
v
es
r
ec
o
r
d
ed
d
u
r
in
g
tr
ain
in
g
.
As
o
b
s
er
v
ed
,
th
e
tr
ain
in
g
an
d
v
alid
atio
n
ac
c
u
r
ac
y
p
r
o
g
r
ess
iv
ely
in
cr
ea
s
e
an
d
co
n
v
er
g
e
to
war
d
s
tab
le
v
alu
es,
in
d
icatin
g
th
at
th
e
m
o
d
el
ca
p
tu
r
es
th
e
r
elev
a
n
t
p
atter
n
s
with
o
u
t
s
ig
n
s
o
f
u
n
d
e
r
f
itti
n
g
.
Similar
ly
,
th
e
lo
s
s
cu
r
v
es
ex
h
i
b
it
a
co
n
s
is
ten
t
d
o
wn
war
d
tr
e
n
d
with
n
o
a
b
r
u
p
t
f
lu
ctu
atio
n
s
,
s
u
g
g
esti
n
g
th
at
o
v
er
f
itti
n
g
is
well
-
co
n
tr
o
lle
d
d
e
s
p
ite
th
e
u
s
e
o
f
r
elativ
ely
c
o
m
p
ac
t a
r
ch
itectu
r
e.
Fig
u
r
e
9
p
r
esen
ts
th
e
co
n
f
u
s
io
n
m
atr
i
x
co
r
r
esp
o
n
d
in
g
to
th
e
v
alid
atio
n
r
esu
lts
o
f
th
e
m
o
u
th
-
s
tate
class
if
ier
.
T
h
e
m
atr
ix
s
h
o
ws
a
clea
r
d
iag
o
n
al
d
o
m
in
an
ce
,
i
n
d
icatin
g
th
at
t
h
e
m
o
d
el
co
r
r
ec
tly
d
is
tin
g
u
is
h
es
b
etwe
en
o
p
e
n
an
d
clo
s
ed
m
o
u
th
s
tates
in
m
o
s
t
ca
s
es.
Ou
t
o
f
5
0
1
i
n
s
tan
ce
s
lab
eled
as
“c
lo
s
ed
,
”
th
e
class
if
ier
co
r
r
ec
tly
p
r
e
d
icted
4
9
2
a
n
d
m
is
clas
s
if
ied
o
n
ly
9
.
Similar
ly
,
f
o
r
th
e
5
0
2
“o
p
e
n
”
in
s
tan
ce
s
,
it
co
r
r
ec
tly
id
en
tifie
d
4
9
5
an
d
p
r
o
d
u
ce
d
7
m
is
class
if
icatio
n
s
.
T
h
ese
r
e
s
u
lts
alig
n
with
th
e
p
r
ec
is
io
n
an
d
r
ec
all
v
alu
e
s
r
ep
o
r
ted
in
T
ab
le
1
,
co
n
f
ir
m
in
g
th
at
b
o
t
h
class
es a
r
e
r
ec
o
g
n
i
ze
d
with
a
h
ig
h
lev
el
o
f
co
n
s
is
ten
cy
.
Evaluation Warning : The document was created with Spire.PDF for Python.