I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
9
,
No
.
4
,
Dec
em
b
er
2020
,
p
p
.
584
~
5
90
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
9
.i
4
.
p
p
584
-
5
90
584
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
A
real
-
ti
m
e d
ro
w
sines
s a
nd f
a
tigue
rec
o
g
nition us
ing
suppo
rt
v
ector
m
a
chine
Nur
Na
bil
a
h Ab
u M
a
ng
s
ho
r
1
,
I
y
lia
Ash
iqi
n Abd
ul M
a
j
id
2
,
Sh
a
f
a
f
I
bra
hi
m
3
,
Nurba
it
y
Sa
bri
4
1,
3,
4
Ce
n
tre
o
f
V
isi
o
n
a
n
d
A
lg
o
rit
h
m
A
n
a
l
y
ti
c
s Re
se
a
r
c
h
G
ro
u
p
,
F
a
c
u
lt
y
o
f
Co
m
p
u
ter an
d
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
R
A
Ca
w
a
n
g
a
n
M
e
lak
a
(Ka
m
p
u
s Ja
sin
),
M
e
lak
a
,
M
a
la
y
sia
2
F
a
c
u
lt
y
o
f
Co
m
p
u
ter an
d
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
RA
Ca
w
a
n
g
a
n
M
e
lak
a
(
Ka
m
p
u
s Ja
sin
),
M
e
lak
a
,
M
a
la
y
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
A
p
r
20
,
2
0
20
R
ev
i
s
ed
J
u
l
1
2
,
2
0
20
A
cc
ep
ted
A
ug
2
0
,
2
0
20
A
d
ro
w
sin
e
ss
a
n
d
f
a
ti
g
u
e
p
ro
b
le
m
s
a
m
o
n
g
th
e
d
r
iv
e
rs
a
re
th
e
m
a
in
f
a
c
to
r
th
a
t
c
o
n
tri
b
u
tes
to
ro
a
d
a
c
c
id
e
n
ts.
T
h
e
se
p
ro
b
lem
s
a
r
e
v
it
a
l
to
b
e
r
e
so
lv
e
d
a
s
th
e
y
c
o
u
ld
c
o
n
tri
b
u
te
to
d
a
m
a
g
e
o
f
ro
a
d
f
a
c
il
it
ies
,
v
e
h
icle
s
a
n
d
m
o
st
im
p
o
rtan
tl
y
th
e
lo
ss
o
f
li
v
e
s.
In
a
v
o
id
in
g
t
h
e
se
m
a
tt
e
r
s,
a
p
ro
p
e
r
m
e
c
h
a
n
is
m
is
n
e
e
d
e
d
to
a
lert
th
e
d
riv
e
r
to
sta
y
a
wa
k
e
th
ro
u
g
h
o
u
t
t
h
e
d
riv
in
g
jo
u
rn
e
y
.
T
h
u
s,
th
is
st
u
d
y
p
ro
p
o
se
d
a
re
a
l
-
ti
m
e
p
ro
to
ty
p
e
f
o
r
re
c
o
g
n
izin
g
th
e
d
ro
w
sin
e
ss
a
n
d
f
a
ti
g
u
e
f
a
c
e
e
x
p
re
ss
io
n
o
f
th
e
d
riv
e
r.
T
h
e
m
e
th
o
d
o
l
o
g
y
o
f
th
is
st
u
d
y
in
v
o
lv
e
s
f
a
c
ial
f
e
a
tu
re
s
d
e
tec
ti
o
n
u
si
n
g
Vio
la
-
Jo
n
e
s
a
l
g
o
rit
h
m
to
d
e
tec
t
th
e
e
x
a
c
t
p
o
siti
o
n
o
f
b
o
th
lef
t
a
n
d
rig
h
t
e
y
e
s
a
n
d
m
o
u
th
.
Ne
x
t,
b
a
se
d
o
n
th
e
d
e
tec
ted
e
y
e
s
a
n
d
m
o
u
th
b
e
f
o
re
h
a
n
d
,
t
h
e
se
g
m
e
n
tatio
n
p
ro
c
e
ss
e
s
p
e
rf
o
r
m
e
d
o
n
b
o
t
h
e
y
e
s
a
n
d
m
o
u
t
h
u
si
n
g
S
o
b
e
l
e
d
g
e
d
e
tec
ti
o
n
to
o
b
tai
n
f
a
c
ial
re
g
io
n
s.
T
h
e
f
e
a
tu
re
e
x
trac
ti
o
n
p
h
a
se
is
c
o
n
d
u
c
ted
u
sin
g
sh
a
p
e
-
b
a
se
d
f
e
a
tu
re
to
o
b
tai
n
t
h
e
e
x
trac
ti
o
n
v
a
lu
e
s.
S
u
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
(S
V
M
)
c
las
si
f
ier
is
d
e
p
lo
y
e
d
f
o
r
th
e
re
c
o
g
n
it
io
n
tas
k
.
A
to
tal
o
f
1
0
0
im
a
g
e
s
a
re
u
se
d
d
u
ri
n
g
th
e
tes
ti
n
g
sta
g
e
s.
T
h
is
stu
d
y
a
c
h
iev
e
d
a
c
o
m
p
e
tetiv
e
r
e
su
lt
o
f
9
0
.
0
0
%
o
f
a
c
c
u
ra
c
y
.
Y
e
t,
h
y
b
rid
iza
ti
o
n
o
r
i
n
teg
ra
ti
o
n
o
f
m
o
re
i
m
a
g
e
p
ro
c
e
ss
in
g
tec
h
n
iq
u
e
s
w
il
l
b
e
p
e
rf
o
r
m
e
d
in
th
e
f
u
tu
re
t
o
im
p
ro
v
e
th
e
c
u
rre
n
t
a
c
c
u
ra
c
y
o
b
tai
n
e
d
.
K
ey
w
o
r
d
s
:
Dr
o
w
s
i
n
es
s
Face
r
ec
o
g
n
itio
n
Fatig
u
e
Su
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
Vio
la
-
J
o
n
es
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
:
Nu
r
Nab
ilah
A
b
u
Ma
n
g
s
h
o
r
C
en
tr
e
o
f
V
is
io
n
an
d
A
l
g
o
r
ith
m
An
al
y
t
ics R
e
s
ea
r
ch
Gr
o
u
p
Facu
lt
y
o
f
C
o
m
p
u
ter
an
d
Ma
t
h
e
m
a
tical
Scie
n
ce
s
Un
i
v
er
s
iti T
ek
n
o
lo
g
i M
A
R
A
C
a
w
an
g
a
n
Me
lak
a
(
Ka
m
p
u
s
J
asin
)
,
7
7
3
0
0
Me
r
li
m
a
u
,
Me
lak
a,
Ma
la
y
s
ia
E
m
ail:
n
u
r
n
ab
ilah
@
u
it
m
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
T
h
o
u
s
an
d
s
o
f
Ma
la
y
s
ia
n
s
lo
s
e
th
eir
li
v
es
co
n
ti
n
u
o
u
s
l
y
d
u
e
t
o
th
e
r
o
ad
ac
cid
en
t
s
[
1
]
.
R
o
ad
ac
cid
en
ts
ar
e
a
s
itu
atio
n
w
h
er
e
it
in
v
o
lv
es
a
co
llis
io
n
b
et
w
ee
n
th
e
v
eh
ic
les
i
n
an
u
n
d
esira
b
le
o
r
u
n
e
x
p
ec
ted
ev
en
t
w
it
h
o
u
t
a
n
i
n
te
n
tio
n
al
ca
u
s
e
an
d
p
lan
[
2
]
.
Ma
la
y
s
ian
s
a
r
e
n
o
t
a
w
ar
e
t
h
at
e
v
er
y
d
a
y
th
e
y
g
o
t
a
lo
t
o
f
ten
d
en
cie
s
o
f
a
n
ac
cid
en
t
w
h
eth
er
th
e
y
ar
e
th
e
d
r
i
v
er
o
r
p
ass
en
g
er
.
T
h
e
d
r
iv
er
is
th
e
m
ai
n
ac
to
r
in
t
h
is
s
itu
a
tio
n
i
n
k
ee
p
i
n
g
t
h
e
p
ass
e
n
g
er
s
af
e.
T
h
er
e
ar
e
s
ev
er
al
ca
u
s
es
t
h
at
lead
to
ac
cid
en
ts
d
u
e
to
n
eg
li
g
e
n
t
d
r
iv
er
s
u
c
h
as
d
r
o
w
s
i
n
es
s
,
u
n
co
n
s
cio
u
s
d
r
iv
er
,
ex
h
au
s
t
n
e
s
s
,
lack
o
f
s
leep
,
o
r
in
v
o
lv
e
i
n
a
lo
n
g
d
r
i
v
e
w
ith
o
u
t
s
h
o
r
t
r
est
[3
-
4]
.
B
esid
es,
h
av
i
n
g
en
o
u
g
h
s
leep
i
s
i
m
p
o
r
tan
t
to
o
.
T
h
e
m
o
s
t
co
m
m
o
n
av
er
ag
e
ad
u
lt
clo
ck
is
s
e
v
en
to
eig
h
t
h
o
u
r
s
p
er
n
ig
h
t.
R
ec
en
t
s
t
u
d
ies
o
u
t
lin
ed
b
y
[
5
]
s
u
g
g
ested
t
h
at
s
ta
y
i
n
g
u
p
late
at
n
i
g
h
t,
co
n
s
u
m
i
n
g
ex
ce
s
s
iv
e
ca
f
f
ei
n
e
an
d
in
s
o
m
n
ia
m
a
y
co
n
tr
ib
u
te
to
d
r
o
w
s
i
n
ess
.
T
h
e
ter
m
“
d
r
o
w
s
y
”
i
s
s
y
n
o
n
y
m
o
u
s
w
i
th
s
l
ee
p
y
,
w
h
ic
h
s
i
m
p
l
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
A
r
ea
l
-
time
d
r
o
w
s
in
es
s
a
n
d
fa
t
ig
u
e
r
ec
o
g
n
itio
n
u
s
in
g
s
u
p
p
o
r
t
ve
cto
r
ma
ch
in
e
(
N
u
r
N
a
b
ila
h
A
b
u
Ma
n
g
s
h
o
r
)
585
m
ea
n
s
an
i
n
cli
n
atio
n
to
f
all
a
s
leep
.
T
h
e
b
r
ain
m
a
y
s
tar
t
to
g
iv
e
i
n
s
tr
u
ctio
n
to
h
ib
er
n
ate
in
g
etti
n
g
en
o
u
g
h
s
leep
at
an
y
t
i
m
e
[
6
]
.
I
n
an
o
t
h
er
n
o
te,
n
u
m
er
o
u
s
m
et
h
o
d
s
w
er
e
i
m
p
le
m
e
n
ted
to
m
ea
s
u
r
e
th
e
d
r
iv
er
’
s
s
y
m
p
to
m
s
w
h
e
n
ex
p
er
ien
ci
n
g
d
o
w
s
i
n
ess
an
d
f
atig
u
e
w
h
ile
d
r
iv
i
n
g
[
7
-
1
0
]
.
Ma
n
y
r
esear
ch
er
a
ll
o
v
er
th
e
w
o
r
ld
h
a
s
al
s
o
a
g
r
ee
d
th
at
a
co
m
m
o
n
ch
ar
ac
ter
is
tic
s
s
h
ar
ed
b
y
a
d
r
o
w
s
y
a
n
d
f
atig
u
e
d
r
iv
er
is
b
ased
o
n
th
eir
b
o
d
y
attit
u
d
e
an
d
th
e
f
ac
ial
e
x
p
r
ess
io
n
[
1
1
]
.
Kitaj
im
a
’
s
f
ac
ial
ex
p
r
es
s
io
n
e
s
ti
m
a
tio
n
m
et
h
o
d
is
o
n
e
o
f
th
e
est
ab
lis
h
ed
m
et
h
o
d
in
m
ea
s
u
r
in
g
t
h
e
d
r
o
w
s
y
a
n
d
f
at
ig
u
e
d
r
iv
er
f
ac
ial
ex
p
r
ess
io
n
[
1
2
]
.
T
h
e
m
eth
o
d
h
a
s
o
u
tli
n
ed
th
e
ch
ar
ac
ter
is
tic
s
in
r
ec
o
g
n
izin
g
a
d
r
o
w
s
in
e
s
s
le
v
el
u
p
to
f
i
v
e
le
v
el.
A
cc
o
r
d
in
g
l
y
,
t
h
e
u
s
e
au
to
m
ati
o
n
s
y
s
te
m
w
ith
le
s
s
o
r
n
o
h
u
m
an
i
n
ter
f
er
en
ce
is
b
en
e
f
ic
ial
f
o
r
h
u
m
an
s
p
ec
if
icall
y
i
n
d
etec
ti
n
g
an
d
r
ec
o
g
n
izin
g
t
h
e
d
r
iv
er
’
s
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
f
ac
e
e
x
p
r
ess
io
n
[
1
3
]
.
T
h
e
i
m
p
le
m
en
ta
tio
n
o
f
i
m
ag
e
p
r
o
ce
s
s
i
n
g
in
d
escr
ib
i
n
g
th
e
d
r
o
w
s
y
a
n
d
f
a
tig
u
e
f
ac
ial
e
x
p
r
es
s
io
n
ca
n
lead
to
th
e
d
et
ec
tio
n
an
d
r
ec
o
g
n
it
io
n
o
f
t
h
e
d
r
iv
er
’
s
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
ex
p
r
ess
io
n
a
u
to
m
atica
ll
y
an
d
ef
f
ec
tiv
e
l
y
[
1
4
-
1
7
]
.
Hen
ce
,
t
h
is
s
tu
d
y
p
r
o
p
o
s
ed
a
r
ea
l
-
ti
m
e
d
r
o
w
s
in
e
s
s
a
n
d
f
at
i
g
u
e
f
ac
ial
e
x
p
r
ess
io
n
r
ec
o
g
n
i
tio
n
u
s
i
n
g
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
tec
h
n
iq
u
e.
T
h
e
d
etec
tio
n
o
f
f
ac
ia
l
f
ea
t
u
r
es
is
d
o
n
e
u
s
in
g
Vio
la
-
J
o
n
es
al
g
o
r
ith
m
in
d
etec
ti
n
g
th
e
e
x
ac
t
p
o
s
it
io
n
o
f
b
o
th
lef
t
an
d
r
i
g
h
t
e
y
e
s
,
a
n
d
m
o
u
t
h
.
N
ex
t,
a
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
is
p
er
f
o
r
m
ed
to
b
o
th
e
y
es
an
d
m
o
u
t
h
u
s
i
n
g
So
b
el
ed
g
e
d
etec
tio
n
.
T
h
e
s
h
ap
e
-
b
ase
d
f
ea
tu
r
e
ex
tr
ac
tio
n
i
s
th
e
n
c
o
n
d
u
cted
to
an
al
y
s
e
th
e
c
h
ar
ac
ter
is
tic
s
o
f
t
h
e
s
eg
m
en
ted
e
y
e
s
a
n
d
m
o
u
t
h
r
e
g
io
n
s
.
Fin
a
ll
y
,
a
tec
h
n
iq
u
e
o
f
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
is
d
ep
lo
y
ed
f
o
r
t
h
e
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
r
ec
o
g
n
i
tio
n
ta
s
k
.
T
h
is
p
ap
er
is
d
iv
id
ed
in
to
f
i
v
e
s
ec
tio
n
s
.
T
h
e
f
ir
s
t
s
ec
tio
n
c
o
n
s
i
s
ts
o
f
t
h
e
i
n
tr
o
d
u
ctio
n
a
n
d
r
esear
ch
m
o
tiv
a
tio
n
.
Seco
n
d
s
ec
tio
n
co
m
p
r
i
s
es
o
f
t
h
e
r
esear
ch
m
et
h
o
d
s
ad
ap
ted
in
th
i
s
r
esear
ch
.
F
u
r
th
er
m
o
r
e,
s
ec
tio
n
th
r
ee
en
tail
s
th
e
an
al
y
s
is
an
d
f
i
n
d
in
g
s
o
f
th
is
r
esear
ch
.
E
v
e
n
tu
a
ll
y
,
t
h
e
last
s
ec
tio
n
s
u
m
m
ar
izes
th
e
r
esear
ch
f
i
n
d
in
g
s
r
esp
ec
ted
to
r
esear
ch
o
b
j
ec
tio
n
,
as
w
el
l a
s
r
ec
o
m
m
e
n
d
atio
n
s
f
o
r
f
u
t
u
r
e
r
esear
ch
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
ai
m
o
f
th
o
s
s
t
u
d
y
i
s
to
r
e
co
g
n
ize
d
r
o
w
s
y
an
d
f
ati
g
u
e
f
ac
e
ex
p
r
ess
io
n
o
f
t
h
e
d
r
iv
er
u
s
in
g
SVM
tech
n
iq
u
e
a
n
d
to
e
v
al
u
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
r
e
co
g
n
iti
o
n
u
s
i
n
g
co
n
f
u
s
io
n
m
atr
i
x
.
Fi
g
u
r
e
1
d
ep
icts
t
h
e
p
r
o
p
o
s
ed
f
lo
w
c
h
ar
t o
f
t
h
is
s
tu
d
y
.
Fig
u
r
e
1
.
P
r
o
p
o
s
ed
f
lo
w
ch
ar
t
f
o
r
th
is
s
t
u
d
y
T
h
e
p
r
o
p
o
s
ed
f
lo
w
ch
ar
t
o
f
t
h
is
s
t
u
d
y
b
eg
i
n
s
w
it
h
d
ata
co
llectio
n
.
Ne
x
t,
t
h
e
p
r
o
ce
s
s
in
g
p
h
a
s
e
co
n
s
is
ts
o
f
f
o
u
r
s
u
b
-
p
r
o
ce
s
s
es
w
h
ic
h
ar
e
f
ac
e
d
etec
tio
n
,
e
y
es
an
d
m
o
u
t
h
s
e
g
m
e
n
tat
io
n
,
f
ea
t
u
r
e
ex
tr
ac
tio
n
,
a
n
d
class
i
f
icatio
n
.
T
h
e
f
ac
e
d
etec
tio
n
is
u
s
ed
to
r
ec
o
g
n
i
ze
th
e
ex
ac
t
p
o
s
itio
n
o
f
b
o
th
lef
t
a
n
d
r
ig
h
t
e
y
es,
an
d
m
o
u
t
h
.
I
n
n
ar
r
o
w
i
n
g
th
e
p
r
o
ce
s
s
i
n
g
ar
ea
,
t
h
e
d
etec
ted
r
eg
io
n
o
f
le
f
t
an
d
r
i
g
h
t
e
y
es,
a
n
d
also
m
o
u
t
h
ar
e
th
e
n
S
tar
t
Da
ta
c
oll
e
c
t
io
n
F
ac
e
detec
t
i
o
n
u
s
i
ng
V
i
o
l
a
J
on
es
a
l
g
or
i
t
hm
E
y
e
s
a
n
d
m
o
ut
h
s
e
g
m
e
n
tat
i
o
n
u
s
ing
S
obe
l
F
e
a
t
ur
e
s
e
x
t
r
a
c
t
i
o
n
u
s
ing
s
h
a
pe
-
ba
s
e
d
f
e
a
t
ur
e
s
C
l
a
s
s
if
i
c
a
t
io
n
u
s
ing
s
u
p
por
t
v
e
c
to
r
m
a
c
hine
E
v
a
lu
a
t
io
n
E
n
d
P
r
oc
e
s
s
ing
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
4
,
Dec
e
m
b
er
20
20
:
5
8
4
–
5
90
586
s
eg
m
e
n
ted
.
A
s
h
ap
e
-
b
ased
f
e
atu
r
e
ex
tr
ac
tio
n
i
s
u
s
ed
to
s
t
u
d
y
th
e
c
h
ar
ac
ter
is
tics
o
f
ea
c
h
s
eg
m
e
n
ted
r
eg
io
n
.
Nex
t,
t
h
e
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
r
ec
o
g
n
itio
n
is
p
er
f
o
r
m
ed
u
s
i
n
g
th
e
SVM
tec
h
n
iq
u
e
w
h
ic
h
p
r
o
d
u
ce
s
th
e
f
i
n
al
o
u
tco
m
e
o
f
t
h
e
class
if
ica
tio
n
s
u
b
s
eq
u
e
n
tl
y
.
T
h
e
d
etail
ex
p
lan
atio
n
o
f
ea
c
h
p
r
o
ce
s
s
in
v
o
lv
ed
is
elab
o
r
ated
f
u
r
t
h
er
in
t
h
e
n
ex
t
s
u
b
s
ec
tio
n
.
2
.
1
.
Da
t
a
c
o
llect
io
n
T
h
e
d
ata
co
llectio
n
w
a
s
co
n
d
u
cted
in
U
n
iv
er
s
iti
T
ec
h
n
o
l
o
g
y
Ma
r
a
(
UiT
M)
C
a
w
an
g
a
n
Me
la
k
a
(
Ka
m
p
u
s
J
asi
n
)
.
T
w
o
f
e
m
ale
s
tu
d
e
n
ts
i
m
itated
v
ar
io
u
s
f
ac
i
al
ex
p
r
ess
io
n
i
n
cl
u
d
es
d
r
o
w
s
y
a
n
d
f
a
tig
u
e
f
ac
ia
l
ex
p
r
ess
io
n
.
T
h
e
len
g
t
h
o
f
t
h
e
ca
p
tu
r
ed
v
id
eo
s
i
s
i
n
th
e
r
an
g
e
o
f
3
0
to
6
0
s
ec
o
n
d
s
.
T
o
tal
o
f
4
0
0
i
m
ag
e
s
o
f
le
f
t
e
y
es,
4
0
0
i
m
ag
e
s
o
f
r
ig
h
t e
y
es
an
d
4
0
0
im
ag
e
s
o
f
m
o
u
t
h
ar
e
g
en
er
ated
f
r
o
m
th
e
s
e
v
id
eo
s
.
Fe
w
n
u
m
b
er
s
o
f
v
id
eo
s
co
m
p
o
s
ed
o
f
p
er
s
o
n
i
m
itates
t
h
e
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
ex
p
r
ess
io
n
s
ar
e
r
ec
o
r
d
e
d
.
T
h
e
ca
p
tu
r
ed
v
id
eo
s
ar
e
s
ep
ar
ated
in
to
co
llectio
n
o
f
s
til
l
i
m
ag
e
s
.
T
h
e
i
m
a
g
es
a
r
e
th
en
u
s
ed
i
n
t
h
e
p
r
o
ce
s
s
in
g
s
ta
g
e.
I
n
g
e
n
er
al,
a
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
d
r
iv
er
w
il
l
b
lin
k
e
y
e
s
s
lo
wy
o
r
r
ath
e
r
clo
s
ed
th
e
e
y
es
a
s
w
ell
t
h
e
m
o
u
t
h
w
ill
y
a
w
n
.
Fi
g
u
r
e
2
s
h
o
w
s
e
x
a
m
p
le
o
f
f
ac
ial
ex
p
r
ess
io
n
s
h
o
w
i
n
g
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
s
tate.
T
h
e
class
i
f
icatio
n
o
f
f
ac
ial
e
x
p
r
ess
io
n
is
ad
ap
ted
f
r
o
m
th
e
Kitaj
a
m
a
’
s
f
ac
ial
ex
p
r
es
s
io
n
m
et
h
o
d
[
1
2
]
.
I
t
class
i
f
ie
s
t
h
e
d
r
iv
er
’
s
f
ac
ial
ex
p
r
ess
io
n
b
ased
o
n
ce
r
tai
n
b
eh
av
io
r
s
.
T
ab
le
1
tab
u
lates
th
e
d
escr
ip
tio
n
o
f
th
e
Kitaj
am
a
’
s
f
ac
ial
e
x
p
r
ess
io
n
m
et
h
o
d
.
Fig
u
r
e
2
.
A
p
er
s
o
n
s
h
o
w
ed
d
r
o
w
s
y
an
d
f
ati
g
u
e
f
ac
ial
e
x
p
r
ess
io
n
w
it
h
clo
s
ed
ey
e
s
an
d
o
p
en
ed
m
o
u
th
T
ab
le
1
.
Descr
ip
tio
n
o
f
Kitaj
im
a
’
s
f
ac
ial
ex
p
r
es
s
io
n
m
et
h
o
d
[
1
2
]
D
r
o
w
si
n
e
ss
L
e
v
e
l
D
e
scri
p
t
i
o
n
B
e
h
a
v
i
o
r
1
N
o
t
S
l
e
e
p
y
Ey
e
s mo
v
e
q
u
i
c
k
l
y
a
n
d
mo
t
i
o
n
i
s
a
c
t
i
v
e
.
2
S
l
i
g
h
t
l
y
S
l
e
e
p
y
Ey
e
s mo
v
e
s
l
o
w
sl
i
g
h
t
l
y
a
n
d
l
i
p
o
p
e
n
s a
l
i
t
t
l
e
.
3
S
l
e
e
p
y
M
o
u
t
h
mo
v
e
s,
t
o
u
c
h
e
s t
h
e
f
a
c
e
a
n
d
r
e
se
a
t
i
n
g
4
R
a
t
h
e
r
S
l
e
e
p
y
H
e
a
d
i
s
sh
a
k
i
n
g
,
f
r
e
q
u
e
n
t
y
a
w
n
i
n
g
a
n
d
b
l
i
n
k
s
a
r
e
sl
o
w
.
5
V
e
r
y
S
l
e
e
p
y
Ey
e
s a
r
e
c
l
o
se
d
a
n
d
h
e
a
d
f
a
l
l
s b
a
c
k
w
a
r
d
.
2
.
2
.
F
a
ce
d
et
ec
t
io
n us
ing
Vio
la
-
J
o
nes
Fro
m
th
e
i
m
a
g
e
g
en
er
ated
,
f
ac
e
d
etec
tio
n
is
t
h
e
n
p
er
f
o
r
m
ed
.
I
t
is
o
n
e
o
f
t
h
e
i
m
p
o
r
tan
t
s
tep
s
f
o
r
f
u
r
t
h
er
d
etec
tin
g
d
r
o
w
s
y
a
n
d
f
atig
u
e
f
ac
e
e
x
p
r
ess
io
n
.
T
h
is
s
tep
is
cr
u
cial
as
it
n
ee
d
s
to
lo
ca
te
th
e
ex
ac
t
p
o
s
itio
n
o
f
t
h
e
f
ac
e.
A
s
s
u
b
s
e
q
u
en
tl
y
,
t
h
e
lo
ca
tio
n
o
f
b
o
th
r
ig
h
t
an
d
le
f
t
e
y
es
an
d
m
o
u
t
h
w
il
l
b
e
d
eter
m
i
n
ed
.
Vio
la
-
J
o
n
es
al
g
o
r
ith
m
is
a
l
o
ca
l
f
ea
tu
r
e
tech
n
iq
u
e
w
h
ic
h
ca
teg
o
r
ized
as
a
f
ea
tu
r
e
-
b
as
ed
tech
n
iq
u
e
[
1
8
]
.
T
h
er
ef
o
r
e,
th
is
s
tu
d
y
d
ep
lo
y
e
d
Vio
la
J
o
n
es
alg
o
r
ith
m
in
d
et
ec
tio
n
g
t
h
e
f
ac
e
as
w
ell
t
h
e
lo
ca
ti
o
n
o
f
b
o
th
r
ig
h
t
an
d
lef
t e
y
es a
n
d
th
e
m
o
u
t
h
o
f
th
e
d
r
iv
er
.
T
h
e
Vio
la
-
J
o
n
es
al
g
o
r
ith
m
u
s
es
Haa
r
-
li
k
e
f
ea
t
u
r
es
[
1
9
]
an
d
th
e
f
ir
s
t
s
tep
i
n
t
h
is
alg
o
r
ith
m
i
s
to
co
n
v
er
t
t
h
e
in
p
u
t
i
m
a
g
e
i
n
to
an
i
n
te
g
r
al
i
m
ag
e.
I
n
te
g
r
al
i
m
ag
e
is
a
s
u
m
m
ed
ar
ea
tab
le
f
o
r
t
h
e
p
u
r
p
o
s
e
to
s
p
ee
d
u
p
t
h
e
co
m
p
u
ta
tio
n
o
f
t
h
e
s
u
m
v
al
u
es
i
n
a
r
ec
tan
g
le
s
u
b
s
et
o
f
t
h
e
p
ix
el
g
r
id
.
E
q
u
ati
o
n
(
1
)
d
en
o
tes
th
e
co
n
s
tr
u
ct
io
n
o
f
th
e
in
teg
r
al
i
m
ag
e,
w
h
er
e
t
h
e
i
n
te
g
r
al
i
m
a
g
e
at
lo
ca
tio
n
x,
y
co
n
tai
n
s
th
e
s
u
m
o
f
th
e
p
ix
el
s
ab
o
v
e
an
d
to
th
e
lef
t o
f
x,
y
p
o
s
itio
n
[
2
0
]
.
(
,
)
=
∑
(
′
,
′
)
′
<
,
′
<
(
1
)
Nex
t,
af
ter
th
e
s
u
m
o
f
th
e
r
ec
tan
g
u
lar
ar
ea
is
co
m
p
u
ted
,
th
e
f
ea
t
u
r
e
ass
o
ciate
d
w
i
th
p
atter
n
P
o
f
i
m
a
g
e
I
is
d
ef
i
n
ed
b
y
(
2
)
as
f
o
llo
w
s
.
T
h
is
w
i
ll
allo
w
t
h
e
co
m
p
ar
is
o
n
b
et
w
ee
n
t
h
e
p
atter
n
an
d
i
m
a
g
e.
Eq
u
atio
n
(
2
)
s
h
o
w
s
t
h
e
i
m
p
le
m
en
tatio
n
.
∑
∑
(
,
)
1
(
,
)
ℎ
=
∑
∑
(
,
)
1
≤
≤
1
≤
≤
1
(
,
)
1
≤
≤
1
≤
≤
(
2
)
In
(
2
)
,
I
an
d
P
d
en
o
te
a
n
i
m
a
g
e
a
n
d
a
p
atter
n
r
esp
ec
t
iv
el
y
.
B
o
th
o
f
i
m
ag
e,
I
a
n
d
p
atter
n
,
P
in
t
h
e
s
a
m
e
s
ize
o
f
N
x
N
.
T
h
e
in
te
g
r
al
i
m
ag
e
s
o
b
tain
e
d
f
r
o
m
(
1
)
w
ill
allo
w
in
te
g
r
a
ls
f
o
r
th
e
Haa
r
-
li
k
e
f
ea
t
u
r
es
to
b
e
ca
lcu
la
ted
.
T
o
o
v
er
co
m
e
t
h
e
i
s
s
u
es
i
n
d
i
f
f
er
en
t
lig
h
te
n
i
n
g
co
n
d
itio
n
,
all
i
m
ag
e
s
w
i
ll
b
e
n
o
r
m
alize
d
u
s
in
g
m
ea
n
an
d
v
ar
ian
ce
[
1
9
]
.
Fig
u
r
e
3
s
h
o
w
s
th
e
f
i
v
e
Haa
r
-
l
ik
e
p
atter
n
s
u
s
ed
to
d
escr
ib
e
th
e
p
o
s
itio
n
an
d
p
atter
n
o
f
t
h
e
f
ac
i
al
f
ea
t
u
r
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
A
r
ea
l
-
time
d
r
o
w
s
in
es
s
a
n
d
fa
t
ig
u
e
r
ec
o
g
n
itio
n
u
s
in
g
s
u
p
p
o
r
t
ve
cto
r
ma
ch
in
e
(
N
u
r
N
a
b
ila
h
A
b
u
Ma
n
g
s
h
o
r
)
587
On
th
e
o
th
er
h
a
n
d
,
Fig
u
r
e
4
illu
s
tr
ates
th
e
s
a
m
p
le
d
etec
tio
n
o
f
f
ac
e,
e
y
e
s
an
d
m
o
u
t
h
u
s
in
g
Vio
la
-
J
o
n
es
alg
o
r
ith
m
.
T
h
e
f
i
v
e
d
er
iv
ed
p
atter
n
s
ar
e
co
n
s
id
er
e
d
th
e
f
ac
ial
f
ea
tu
r
e
s
o
n
f
ac
e
[
2
1
]
.
I
n
th
e
i
m
p
le
m
en
ta
tio
n
,
th
e
e
x
tr
ac
te
d
f
ea
tu
r
es
to
r
ep
r
esen
t
th
e
f
ac
ial
f
ea
tu
r
es
ar
e
th
e
h
o
r
izo
n
tal,
v
er
tical
an
d
h
o
r
izo
n
tal
w
it
h
s
p
ac
es
an
d
c
h
ec
k
er
ed
[
2
0
]
.
Hen
ce
th
e
u
s
e
o
f
Haa
r
-
lik
e
f
ea
t
u
r
es
p
atter
n
s
is
j
u
s
ti
f
ied
.
T
h
is
V
io
la
-
J
o
n
es
al
g
o
r
ith
m
w
ill
ef
f
ec
ti
v
el
y
d
etec
t
t
h
e
f
ac
e
f
ir
s
t
an
d
d
etec
t
th
e
p
o
s
itio
n
o
f
th
e
e
y
es
a
n
d
m
o
u
t
h
s
u
b
s
eq
u
en
t
l
y
.
Ne
x
t,
th
e
f
ac
ial
f
ea
t
u
r
es d
etec
ted
n
ee
d
to
b
e
s
eg
m
e
n
ted
.
Fig
u
r
e
3
.
Fiv
e
Haa
r
-
lik
e
f
ea
t
u
r
es p
atter
n
s
to
d
escr
ib
e
th
e
f
ac
ial
f
ea
t
u
r
es [
2
1
]
Fig
u
r
e
4
.
Gr
ap
h
ical
r
ep
r
esen
ta
tio
n
o
f
f
ac
e
d
etec
tio
n
u
s
in
g
Vio
la
J
o
n
es a
l
g
o
r
ith
m
2
.
3
.
E
y
es a
nd
m
o
uth
s
eg
e
m
e
nta
t
io
n us
ing
s
o
bel
I
n
s
eg
m
e
n
tatio
n
p
h
a
s
e,
So
b
el
ed
g
e
d
etec
tio
n
is
u
s
ed
f
o
r
s
e
g
m
en
tin
g
e
y
e
s
an
d
m
o
u
th
o
n
t
h
e
d
r
iv
er
’
s
f
ac
e
d
etec
ted
u
s
in
g
Vio
la
-
J
o
n
e
alg
o
r
ith
m
ea
r
lier
.
So
b
el
ed
g
e
d
etec
tio
n
is
s
u
it
f
o
r
its
h
i
g
h
f
r
eq
u
en
c
y
v
ar
iatio
n
as
e
y
e
b
lin
k
in
g
an
d
m
o
u
th
y
a
w
n
i
n
g
.
I
t
w
o
r
k
s
b
y
co
m
p
u
t
in
g
th
e
g
r
ad
ien
t
o
f
i
m
a
g
e
’
s
i
n
te
n
s
it
y
at
ea
ch
p
ix
e
l
in
th
e
i
m
a
g
e
u
s
in
g
t
w
o
d
i
f
f
er
e
n
t
3
x
3
m
atr
i
x
k
er
n
els.
E
ac
h
k
e
r
n
el
co
n
s
tit
u
te
s
o
f
x
-
d
ir
ec
tio
n
k
er
n
el,
Gx
a
n
d
t
h
e
y
-
d
ir
ec
tio
n
k
er
n
el,
G
y
.
F
ig
u
r
e
5
s
h
o
w
s
t
h
e
k
er
n
el
u
s
ed
in
t
h
e
So
b
el
ed
g
e
d
etec
tio
n
w
h
ile
Fig
u
r
e
6
d
ep
icts
th
e
se
g
m
e
n
tatio
n
o
f
e
y
e
s
an
d
m
o
u
th
u
s
i
n
g
So
b
el
ed
g
e
d
etec
tio
n
.
Fig
u
r
e
5
.
Ker
n
el
u
s
ed
i
n
th
e
So
b
el
ed
g
e
d
etec
tio
n
Fig
u
r
e
6
.
Seg
e
m
en
ta
tio
n
o
f
e
y
es a
n
d
m
o
u
th
u
s
in
g
So
b
el
ed
g
e
d
etec
tio
n
2
.
4
.
F
e
a
t
ures e
x
t
ra
ct
io
n us
ing
s
ha
pe
-
ba
s
ed
f
ea
t
ures
Featu
r
es
e
x
tr
ac
tio
n
i
s
co
n
d
u
c
ted
n
ex
t
to
ex
tr
ac
t
t
h
e
u
s
e
f
u
l
in
f
o
r
m
atio
n
f
r
o
m
t
h
e
i
m
ag
es
f
o
r
th
e
class
i
f
icatio
n
p
u
r
p
o
s
e.
T
h
is
s
t
u
d
y
u
s
es
e
y
es
a
n
d
m
o
u
t
h
f
ea
t
u
r
es
t
h
at
h
a
v
e
b
ee
n
s
eg
m
e
n
te
d
u
s
i
n
g
So
b
el
ed
g
e
d
etec
tio
n
.
E
y
es
a
n
d
m
o
u
t
h
f
ea
tu
r
es
e
x
tr
ac
tio
n
i
s
p
er
f
o
r
m
ed
u
s
i
n
g
s
h
ap
e
f
ea
t
u
r
es
to
o
b
tain
th
e
f
ea
tu
r
e
v
ec
to
r
s
th
at
w
i
ll
b
e
u
s
ed
in
t
h
e
cla
s
s
i
f
icatio
n
later
.
T
h
e
lis
t
o
f
f
ea
t
u
r
e
v
ec
to
r
s
ca
lc
u
lated
a
r
e
ar
ea
,
p
er
im
eter
,
eq
u
iv
ala
n
ce
d
ia
m
eter
,
m
aj
o
r
ax
is
a
n
d
m
i
n
o
r
ax
is
.
T
h
es
e
f
ea
t
u
r
es
ar
e
s
elec
ted
d
u
e
to
is
s
u
itab
ilit
y
i
n
r
ep
r
esen
tin
g
t
h
e
s
tate
o
f
e
y
es
an
d
m
o
u
t
h
.
R
eg
io
n
p
r
o
p
s
(
)
i
n
Ma
tlab
is
u
s
ed
f
o
r
ex
tr
ac
tin
g
t
h
ese
f
ea
tu
r
e
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
4
,
Dec
e
m
b
er
20
20
:
5
8
4
–
5
90
588
2
.
5
.
Cla
s
s
if
ica
t
io
n us
ing
s
up
po
rt
v
ec
t
o
r
m
a
chine
T
h
is
s
ec
tio
n
d
i
s
cu
s
s
e
s
o
n
t
h
e
class
i
f
icatio
n
u
s
i
n
g
th
e
s
tate
-
of
-
th
e
-
ar
t
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
class
i
f
ier
.
SVM
cla
s
s
i
f
ier
is
a
v
er
y
u
s
e
f
u
l
m
ac
h
i
n
e
lear
n
i
n
g
to
o
l
[
2
2
-
2
3
]
.
I
n
th
is
s
tag
e
class
i
f
icatio
n
u
s
in
g
SVM
is
a
p
r
o
ce
s
s
to
d
eter
m
in
e
w
h
et
h
er
th
e
d
r
iv
er
is
i
n
d
r
o
w
s
y
o
r
f
ati
g
u
e
co
n
d
itio
n
b
ased
o
n
th
e
s
et
o
f
i
m
ag
e
s
o
b
tain
ed
f
r
o
m
t
h
e
v
id
eo
.
T
h
e
i
m
p
le
m
en
ta
tio
n
o
f
SVM
cla
s
s
i
f
ier
in
v
o
les
th
e
u
s
e
o
f
tr
a
i
n
I
ma
g
eCl
a
s
s
ifier
(
)
f
u
n
ctio
n
[
2
4
]
w
h
er
e
t
w
o
d
i
f
f
e
r
en
t
ca
teg
o
r
y
o
f
cla
s
s
es
ar
e
b
u
ild
.
T
h
e
class
e
s
ar
e
eith
er
t
h
e
d
r
iv
er
is
d
r
o
w
s
y
an
d
f
ati
g
u
e,
o
r
th
e
d
r
iv
er
is
a
w
a
k
e.
T
h
e
m
o
d
el
is
tr
ain
ed
a
n
d
th
e
s
u
p
p
o
r
t
v
ec
to
r
s
(
S
Vs)
f
o
r
th
e
s
e
clas
s
es
ar
e
g
en
er
ated
.
He
n
ce
,
th
e
test
i
n
g
ca
n
b
e
co
n
d
u
cted
b
ased
o
n
t
h
e
tr
ain
ed
m
o
d
el
b
u
i
lt.
Fi
g
u
r
e
7
illu
s
tr
ate
s
ex
a
m
p
le
o
f
k
er
n
el
u
s
ed
i
n
SVM.
Fig
u
r
e
7
.
E
x
a
m
p
le
o
f
k
er
n
el
S
VM
T
h
e
p
r
e
d
ictio
n
o
f
d
r
o
w
s
y
a
n
d
f
atig
u
e
f
ac
e
e
x
p
r
ess
io
n
o
n
th
e
d
r
iv
er
’
s
f
ac
e
ar
e
b
ased
o
n
th
e
b
o
th
clo
s
e
lef
t
an
d
r
ig
h
t
e
y
e
an
d
o
p
en
/clo
s
ed
m
o
u
t
h
.
T
h
ese
s
y
m
p
h
to
m
s
i
n
d
icate
s
t
h
e
d
r
iv
er
is
y
a
w
n
i
n
g
an
d
th
e
e
y
es
ar
e
s
h
u
t.
S
u
b
s
eq
u
e
n
tl
y
,
i
f
t
h
e
f
r
a
m
e
s
f
r
o
m
t
h
e
v
id
o
e
s
h
o
w
t
h
e
tr
u
e
p
r
ed
icted
s
ig
n
s
o
f
d
r
o
w
s
y
an
d
f
ati
g
u
e
s
tate
co
n
ti
n
u
o
u
s
l
y
in
3
s
ec
o
n
d
s
an
d
ab
o
v
e,
th
e
alar
m
w
i
ll
b
e
tu
r
n
ed
o
n
[
3
]
.
T
h
is
alar
m
is
ai
m
to
w
ak
e
t
h
e
d
r
iv
er
u
p
f
r
o
m
f
ee
li
n
g
s
leep
y
,
to
b
e
m
o
r
e
ca
u
tio
u
s
an
d
k
ee
p
f
o
cu
s
w
h
ile
d
r
iv
i
n
g
.
2
.
6
.
E
v
a
lua
t
i
o
n
T
h
e
p
r
o
p
o
s
ed
s
tu
d
y
is
e
v
alu
a
t
ed
u
s
i
n
g
co
n
f
u
s
io
n
m
atr
ix
.
T
h
e
test
o
u
tco
m
e
ca
n
b
e
p
o
s
iti
v
e
w
h
ic
h
is
th
e
r
ec
o
g
n
it
io
n
r
e
s
u
lt
co
r
r
ec
tl
y
r
ec
o
g
n
ized
a
s
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
co
n
d
itio
n
.
W
h
ile
t
h
e
n
eg
a
tiv
e
r
es
u
lt
s
in
d
icate
as
a
w
ak
e.
Ho
w
ev
er
,
th
e
p
o
s
itiv
e
ca
n
b
e
f
u
r
th
er
d
escr
ib
ed
as
tr
u
e
p
o
s
itiv
e
(
T
P
)
an
d
f
alse
p
o
s
iti
v
e
(
FP
)
.
T
P
is
w
h
e
n
th
e
r
ec
o
g
n
it
io
n
r
esu
lt
(
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
)
ar
e
c
o
r
r
ec
tly
r
ec
o
g
n
ized
as
th
e
ex
p
ec
ted
r
esu
lt
(
d
r
o
w
s
y
a
n
d
f
ati
g
u
e)
.
Me
an
wh
ile,
FP
is
w
h
e
n
th
e
r
ec
o
g
n
it
i
o
n
r
esu
lt
(
a
w
a
k
e)
ar
e
in
co
r
r
ec
t
l
y
r
ec
o
g
n
ized
as
t
h
e
ex
p
ec
ted
r
esu
lt (
d
r
o
w
s
y
a
n
d
f
a
tig
u
e)
.
Nex
t,
t
h
e
n
eg
at
iv
e
ca
n
b
e
tr
u
e
n
eg
a
tiv
e
(
T
N)
an
d
f
alse
n
e
g
ati
v
e
(
FN)
.
T
N
is
w
h
en
th
e
r
ec
o
g
n
itio
n
r
esu
lt
(
a
w
ak
e)
ar
e
co
r
r
ec
tl
y
r
ec
o
g
n
ized
a
s
t
h
e
e
x
p
ec
ted
r
es
u
lt
(
a
w
a
k
e)
.
O
n
th
e
o
th
er
h
a
n
d
,
FN
i
s
w
h
e
n
t
h
e
r
ec
o
g
n
itio
n
r
esu
lt
(
d
r
o
w
s
y
a
n
d
f
atig
u
e)
ar
e
in
co
r
r
ec
tly
r
ec
o
g
n
ized
as
th
e
e
x
p
ec
ted
r
esu
l
t
(
a
w
ak
e)
.
T
ab
le
2
s
h
o
w
s
t
h
e
co
n
f
u
s
io
n
m
atr
ix
o
u
tco
m
e
f
r
o
m
t
h
e
e
x
p
er
i
m
e
n
t
co
n
d
u
cted
.
T
h
e
to
tal
te
s
ted
i
m
ag
e
s
ar
e
co
m
p
o
s
ed
o
f
5
0
im
a
g
es
f
o
r
d
r
o
w
s
y
an
d
f
atig
u
e
co
n
d
itio
n
a
n
d
5
0
ea
ch
f
o
r
aw
a
k
e
co
n
d
itio
n
.
T
ab
le
2
.
C
o
n
f
u
s
io
n
m
atr
i
x
r
es
u
lt
P
r
e
d
i
c
t
e
d
C
l
a
ss
i
f
i
c
a
t
i
o
n
D
r
o
w
s
y
a
n
d
F
a
t
i
g
u
e
A
w
a
k
e
A
c
t
u
a
l
C
o
n
d
i
t
i
o
n
D
r
o
w
s
y
a
n
d
F
a
t
i
g
u
e
4
6
(
T
P
)
4
(
F
P
)
A
w
a
k
e
6
(
F
N
)
4
4
(
T
P
)
Valid
atio
n
i
n
v
o
l
v
es
ca
lcu
la
ti
n
g
f
o
u
r
o
b
j
e
c
t
i
v
e
m
e
a
s
u
r
e
s
o
f
t
e
s
t
p
e
r
f
o
r
m
a
n
c
e
,
n
a
m
e
l
y
,
s
en
s
it
iv
i
t
y
,
s
p
ec
if
icit
y
,
p
o
s
iti
v
e
p
r
ed
ictiv
e
v
alu
e
(
P
P
V
)
a
n
d
n
e
g
a
t
i
v
e
p
r
e
d
i
c
t
i
v
e
v
a
l
u
e
(
N
P
V
)
.
Nex
t,
t
h
e
v
alid
atio
n
i
n
v
o
lv
es
ca
lcu
la
tin
g
f
i
v
e
o
b
j
ec
tiv
e
m
ea
s
u
r
es
o
f
test
p
er
f
o
r
m
a
n
ce
,
n
a
m
e
l
y
,
ac
cu
r
ac
y
,
s
e
n
s
iti
v
it
y
,
s
p
ec
if
ic
it
y
,
p
o
s
iti
v
e
p
r
ed
ict
iv
e
v
al
u
e
(
P
P
V)
an
d
n
eg
at
iv
e
p
r
ed
ictiv
e
v
al
u
e
(
NP
V)
[
2
5
-
2
6
]
ar
e
co
m
p
u
ted
.
A
cc
u
r
ac
y
m
ea
s
u
r
es
th
e
o
v
er
al
l p
er
f
o
r
m
an
ce
o
f
t
h
e
tes
tin
g
.
S
e
n
s
it
iv
i
t
y
m
ea
s
u
r
es
t
h
e
p
r
o
p
o
r
tio
n
o
f
th
e
co
r
r
ec
t
p
r
ed
icted
p
o
s
itiv
e
class
e
s
wh
ile
s
p
ec
i
f
icit
y
m
ea
s
u
r
es
th
e
p
r
o
p
o
r
tio
n
o
f
th
e
n
eg
at
iv
e
clas
s
es
t
h
at
h
a
v
e
c
o
r
r
ec
tly
c
lass
if
ied
[
1
9
-
20]
.
E
q
u
atio
n
s
(
3
-
5
)
s
h
o
w
s
t
h
e
f
o
r
m
u
la
to
ca
lc
u
late
ac
cu
r
ac
y
,
s
e
n
s
i
tiv
it
y
a
n
d
s
p
ec
i
f
icit
y
r
esp
ec
ti
v
el
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
A
r
ea
l
-
time
d
r
o
w
s
in
es
s
a
n
d
fa
t
ig
u
e
r
ec
o
g
n
itio
n
u
s
in
g
s
u
p
p
o
r
t
ve
cto
r
ma
ch
in
e
(
N
u
r
N
a
b
ila
h
A
b
u
Ma
n
g
s
h
o
r
)
589
=
×
100
(
3
)
=
+
(
4
)
=
+
(
5
)
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
T
ab
le
3
s
h
o
w
s
t
h
e
s
u
m
m
ar
iza
tio
n
o
f
t
h
e
r
es
u
lt
s
r
es
u
lts
ac
h
i
ev
ed
b
y
t
h
e
s
t
u
d
y
in
cl
u
d
i
n
g
t
h
e
p
o
s
iti
v
e
p
r
ed
ictio
n
v
alu
e
(
P
P
V)
an
d
n
eg
at
iv
e
p
r
ed
ictio
n
v
al
u
e
(
N
P
V)
.
B
ased
o
n
th
e
ex
p
er
i
m
e
n
t
co
n
d
u
cted
,
th
e
ac
cu
r
ac
y
ac
h
ie
v
ed
b
y
th
is
s
tu
d
y
is
9
0
.
0
0
%.
On
t
h
e
to
p
o
f
t
h
at,
s
e
n
s
i
tiv
it
y
a
n
d
s
p
ec
i
f
icit
y
r
ate
ac
h
iev
ed
ar
e
0
.
8
8
5
an
d
0
.
9
1
7
r
esp
e
ctiv
ely
.
T
ab
le
3
.
Su
m
m
ar
y
r
es
u
lt
M
e
a
su
r
e
me
n
t
R
e
su
l
t
A
c
c
u
r
a
c
y
9
0
.
0
0
%
S
e
n
si
t
i
v
i
t
y
0
.
8
8
5
S
p
e
c
i
f
i
c
i
t
y
0
.
9
1
7
P
o
si
t
i
v
e
P
r
e
d
i
c
t
i
o
n
V
a
l
u
e
(
P
P
V
)
0
.
9
2
0
N
e
g
a
t
i
v
e
P
r
e
d
i
c
t
i
o
n
V
a
l
u
e
(
N
P
V
)
0
.
8
8
0
4.
CO
NCLU
SI
O
N
T
h
is
s
t
u
d
y
p
r
o
p
o
s
ed
a
r
ea
l
-
ti
m
e
r
ec
o
g
n
itio
n
o
f
d
r
o
w
s
y
an
d
f
ati
g
u
e
f
ac
ial
ex
p
r
ess
io
n
.
I
t
d
e
m
o
n
s
tr
ates
a
p
r
o
m
i
s
i
n
g
r
es
u
lt
i
n
r
ec
o
g
n
i
zin
g
d
r
o
w
s
y
an
d
f
ati
g
u
e
f
ac
e
ex
p
r
ess
io
n
o
f
t
h
e
d
r
iv
er
u
s
i
n
g
t
h
e
Vio
la
-
J
o
n
es
alg
o
r
ith
m
,
s
h
ap
e
-
b
ased
f
ea
t
u
r
es
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
w
it
h
a
n
ac
cu
r
ac
y
o
f
9
0
.
0
0
%.
I
t
is
also
s
tated
th
at
th
i
s
s
t
u
d
y
ac
h
ie
v
e
d
a
p
o
s
itiv
e
p
r
ed
ictio
n
v
al
u
e
(
PP
V)
as
0
.
9
2
0
an
d
n
eg
ati
v
e
p
r
ed
ictio
n
v
alu
e
(
NP
V)
as
0
.
8
8
0
.
B
esid
es,
b
o
t
h
th
e
s
e
n
s
i
tiv
it
y
a
n
d
s
p
ec
if
icit
y
ac
h
ie
v
e
a
p
r
o
m
i
s
in
g
v
al
u
e
o
f
0
.
8
8
5
an
d
0
.
9
1
7
.
B
o
th
o
f
th
e
s
en
s
iti
v
it
y
a
n
d
s
p
ec
if
icit
y
v
al
u
e
s
o
b
atian
ed
is
h
ig
h
a
n
d
it
ca
n
b
e
co
n
clu
d
ed
th
e
p
r
o
p
o
s
ed
s
tu
d
y
ab
le
to
d
is
tin
g
u
i
s
h
b
et
w
ee
n
d
r
o
w
s
y
a
n
d
f
ati
g
u
e
ex
p
r
ess
io
n
an
d
a
w
a
k
e
f
ac
e
e
x
p
r
ess
io
n
ac
c
o
r
d
in
g
l
y
.
Ho
w
ev
er
,
it
is
b
elie
v
ed
t
h
at
t
h
e
h
y
b
r
id
i
za
tio
n
o
r
in
teg
r
atio
n
o
f
a
n
y
e
x
is
t
in
g
tec
h
n
iq
u
e
s
f
o
r
b
o
th
f
e
atu
r
e
ex
tr
ac
tio
n
a
n
d
class
i
f
icatio
n
ca
n
i
m
p
r
o
v
e
th
e
ac
cu
r
ac
y
r
es
u
lt i
n
f
u
t
u
r
e.
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
r
esear
ch
w
as
s
u
p
p
o
r
ted
b
y
Min
is
tr
y
o
f
E
d
u
ca
tio
n
Ma
la
y
s
ia
(
Mo
E
)
,
an
d
Un
i
v
er
s
it
i
T
ek
n
o
lo
g
i
MA
R
A
t
h
r
o
u
g
h
t
h
e
F
u
n
d
a
m
e
n
tal
R
e
s
ea
r
ch
Gr
a
n
t Sc
h
e
m
e
(
FR
GS)
(
6
0
0
-
I
R
MI
/F
R
GS 5
/3
(
2
1
5
/2
0
1
9
)
)
.
RE
F
E
R
E
NC
E
S
[1
]
Ha
sh
im
H
H,
Ra
h
i
m
S
A
.
T
h
e
Co
n
stru
c
ti
o
n
o
f
Ro
a
d
A
c
c
id
e
n
t
A
n
a
l
y
si
s
a
n
d
Da
tab
a
se
S
y
ste
m
i
n
M
a
lay
sia
.
4
th
IRT
AD
Co
n
fer
e
n
c
e
,
p
p
.
2
3
5
-
2
4
1
,
2
0
0
9
.
[2
]
S
a
sik
a
la
R,
S
u
re
sh
S
,
Ch
a
n
d
ra
m
o
h
a
n
J,
V
a
lan
ra
jk
u
m
a
r
M
.
Dr
iv
e
r
Dro
w
sin
e
ss
D
e
tec
ti
o
n
S
y
st
e
m
u
sin
g
Im
a
g
e
P
r
o
c
e
ss
in
g
T
e
c
h
n
iq
u
e
b
y
th
e
Hu
m
a
n
V
isu
a
l
S
y
ste
m
.
In
t.
J
.
Eme
rg
.
T
e
c
h
n
o
l.
En
g
.
Res
.
,
v
o
l.
6
,
n
o
.
6
,
p
p
.
1
-
1
1
,
2
0
1
8
.
[3
]
S
e
f
ia
A
M
,
S
e
lv
i
J
A
G
.
Driv
e
r
S
tate
A
n
a
l
y
sis
a
n
d
D
ro
w
sin
e
ss
De
t
e
c
ti
o
n
Us
in
g
I
m
a
g
e
P
ro
c
e
ss
in
g
.
In
t.
J
.
S
c
i.
En
g
.
Ap
p
l
.
S
c
i.
,
v
o
l.
2
,
n
o
.
6
,
p
p
.
2
3
9
5
-
3
4
7
0
,
2
0
1
6
.
[4
]
Bo
u
m
e
h
e
d
M
,
A
lsh
a
q
a
q
i
B,
Ba
q
u
h
a
ize
l
A
S
,
Ou
is
M
E
A
.
Driv
e
r
d
ro
w
sin
e
ss
d
e
tec
ti
o
n
sy
ste
m
.
Ad
v
.
S
y
st.
S
c
i
.
Ap
p
l
.
,
v
o
l.
1
6
,
n
o
.
2
,
p
p
.
1
0
1
-
1
0
4
,
2
0
1
6
.
[5
]
Ka
u
r
H.
Driv
e
r
Dro
ws
in
e
ss
De
te
c
ti
o
n
S
y
ste
m
Us
in
g
I
m
a
g
e
P
ro
c
e
ss
in
g
.
Dr
iv.
Dr
o
wsin
e
ss
De
tec
t.
S
y
st.
Us
in
g
Im
a
g
e
Pro
c
e
ss
.
,
v
o
l.
4
,
n
o
.
5
,
p
.
4
0
,
2
0
1
5
.
[6
]
S
a
in
i
V.
Driv
e
r
Dro
w
sin
e
ss
De
t
e
c
ti
o
n
S
y
ste
m
a
n
d
T
e
c
h
n
iq
u
e
s :
A
R
e
v
ie
w
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
mp
u
ter
S
c
ien
c
e
a
n
d
In
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
ies
,
5
(
3
),
4
2
4
5
-
4
2
4
9
,
2
0
1
4
.
[7
]
A
ru
n
S
.
Ke
n
n
e
th
S
,
M
u
r
u
g
a
p
p
a
n
M
.
De
tec
ti
n
g
Driv
e
r
Dro
w
sin
e
ss
Ba
se
d
o
n
S
e
n
s
o
rs:
A
Re
v
ie
w
.
S
e
n
so
rs
,
12
(1
2
),
p
p
.
1
6
9
3
7
-
1
6
9
5
3
,
2
0
1
2
.
[8
]
X
u
x
in
Z,
Xu
e
so
n
g
W
,
X
iao
h
a
n
Y,
Ch
u
a
n
X
,
X
iao
h
u
i
Z
,
Jia
o
h
u
a
W
.
Driv
e
r
Dro
w
sin
e
ss
De
te
c
ti
o
n
Us
in
g
M
ix
e
d
-
e
ffe
c
t
Ord
e
re
d
L
o
g
it
M
o
d
e
l
C
o
n
s
id
e
rin
g
T
im
e
Cu
m
u
lativ
e
E
ff
e
c
t.
An
a
lytic M
e
th
o
d
s i
n
Acc
id
e
n
t
Res
e
a
rc
h
,
2
0
2
0
.
[9
]
Oliv
e
ira
L
,
Ca
rd
o
so
J
S
,
L
o
u
re
n
ç
o
A
,
A
h
lströ
m
C.
Driv
e
r
d
ro
w
s
in
e
ss
d
e
tec
ti
o
n
:
a
c
o
m
p
a
riso
n
b
e
tw
e
e
n
in
tru
siv
e
a
n
d
n
o
n
-
in
tr
u
siv
e
sig
n
a
l
a
c
q
u
isi
t
io
n
m
e
th
o
d
s.
2
0
1
8
7
t
h
E
u
ro
p
e
a
n
W
o
rk
sh
o
p
o
n
Vi
su
a
l
I
n
f
o
rm
a
ti
o
n
Pro
c
e
ss
in
g
(
EUVI
P)
,
T
a
m
p
e
r
e
,
p
p
.
1
-
6
.
,
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
4
,
Dec
e
m
b
er
20
20
:
5
8
4
–
5
90
590
[1
0
]
P
ia
M
F
,
Bry
a
n
J
V
,
Ro
b
e
rt
A
S
,
Ch
risto
p
h
e
r
G
M
,
Ha
n
s
P
A
,
V
a
n
D.
Ef
f
icie
n
t
d
riv
e
r
d
ro
ws
in
e
ss
d
e
tec
ti
o
n
a
t
m
o
d
e
ra
te l
e
v
e
ls o
f
d
ro
w
sin
e
ss
.
Ac
c
id
e
n
t
A
n
a
lys
is &
Pre
v
e
n
ti
o
n
,
Vo
lu
m
e
5
0
,
p
p
.
3
4
1
-
3
5
0
,
2
0
1
3
.
[
1
1
]
A
b
u
S
,
S
a
a
d
A
S
.
D
r
i
v
e
r
D
r
o
w
s
i
n
e
s
s
D
e
t
e
c
t
i
o
n
u
s
i
n
g
F
a
c
e
M
o
n
i
t
o
r
i
n
g
a
n
d
P
r
e
s
s
u
r
e
M
e
a
s
u
r
e
m
e
n
t
.
R
e
s
e
a
r
c
h
&
R
e
v
i
e
w
s
:
A
J
o
u
r
n
a
l
o
f
E
m
b
e
d
d
e
d
S
y
s
t
e
m
&
A
p
p
l
i
c
a
t
i
o
n
s
.
5
(
3
)
:
p
p
.
12
-
1
8
,
2
0
1
7
.
[1
2
]
Hiro
k
i
K,
Na
k
a
h
o
N,
Ke
ii
c
h
i
Y,
Yo
sh
ih
ir
o
G.
P
re
d
ictio
n
of
A
u
to
m
o
b
il
e
Driv
e
r
S
lee
p
in
e
ss
.
1
st
Re
p
o
rt,
Ra
ti
n
g
of
S
lee
p
in
e
ss
Ba
se
d
on
F
a
c
ial
Ex
p
re
ss
io
n
a
n
d
Ex
a
m
in
a
ti
o
n
of
E
ff
e
c
ti
v
e
P
re
d
icto
r
In
d
e
x
e
s
of
S
lee
p
in
e
ss
,
Ja
p
a
n
S
o
c
iety
of
M
e
c
h
a
n
ica
l
En
g
i
n
e
e
rs
M
e
m
o
irs
(C),
V
o
l
.
6
3
,
No
.
6
1
3
,
p
p
.
9
3
-
1
0
0
,
1
9
9
7
.
[
1
3
]
P
a
r
k
S
,
P
a
n
F
,
K
a
n
g
S
,
Y
o
o
C
D
.
D
r
i
v
e
r
D
r
o
w
s
i
n
e
s
s
D
e
t
e
c
t
i
o
n
S
y
s
t
e
m
B
a
s
e
d
o
n
F
e
a
t
u
r
e
R
e
p
r
e
s
e
n
t
a
t
i
o
n
L
e
a
r
n
i
n
g
U
s
i
n
g
V
a
r
i
o
u
s
D
e
e
p
N
e
t
w
o
r
k
s
.
I
n
:
C
h
e
n
C
S
.
,
L
u
J
.
,
M
a
K
K
.
(
e
d
s
)
C
o
m
p
u
t
e
r
V
i
s
i
o
n
-
A
C
C
V
2
0
1
6
W
o
r
k
s
h
o
p
s
.
A
C
C
V
2
0
1
6
.
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
1
0
1
1
8
.
2
0
1
7
.
[1
4
]
S
h
u
y
a
n
H,
G
a
n
g
ti
e
Z,
Driv
e
r
d
ro
w
sin
e
ss
d
e
tec
ti
o
n
w
it
h
e
y
e
li
d
re
late
d
p
a
ra
m
e
ter
s
b
y
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
.
Exp
e
rt S
y
ste
ms
wit
h
A
p
p
l
ica
ti
o
n
s
,
Vo
lu
m
e
3
6
,
Iss
u
e
4
,
p
p
.
7
6
5
1
-
7
6
5
8
,
2
0
0
9
.
[1
5
]
Ra
tn
a
K
M
,
Ra
m
y
a
V
,
F
ra
n
k
li
n
R
G
.
A
lert
S
y
ste
m
f
o
r
Driv
e
r’s
Dro
w
sin
e
ss
Us
in
g
Im
a
g
e
P
ro
c
e
ss
in
g
.
2
0
1
9
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
V
isio
n
T
o
w
a
rd
s
Eme
rg
i
n
g
T
re
n
d
s
i
n
Co
mm
u
n
ica
t
io
n
a
n
d
Ne
two
rk
i
n
g
(
Vi
T
EC
o
N)
,
V
e
ll
o
re
,
In
d
ia,
p
p
.
1
-
5
,
2
0
1
9
.
[1
6
]
Ku
m
a
r
R
P
,
S
a
n
g
e
e
th
M
,
V
a
i
d
h
y
a
n
a
th
a
n
K
S
,
P
a
n
d
ian
A
.
T
ra
ff
ic
S
ig
n
a
n
d
Dro
w
sin
e
ss
De
tec
ti
o
n
Us
in
g
Op
e
n
-
CV.
In
ter
n
a
t
io
n
a
l
Res
e
a
rc
h
J
o
u
r
n
a
l
o
f
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
(
IRJTE
)
.
V
o
l.
0
6
,
Iss
u
e
0
3
,
p
p
.
1
3
9
8
,
2
0
1
9
.
[1
7
]
Bh
o
y
a
r
A
M
,
S
a
w
a
lk
a
r
S
N.
Im
p
lem
e
n
tatio
n
o
n
V
isu
a
l
A
n
a
ly
sis
o
f
E
y
e
S
tate
Us
in
g
Im
a
g
e
P
r
o
c
e
ss
in
g
f
o
r
Driv
e
r
F
a
ti
g
u
e
De
tec
ti
o
n
.
In
ter
n
a
ti
o
n
a
l
Res
e
a
rc
h
J
o
u
rn
a
l
o
f
En
g
i
n
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
(
IRJET
).
V
o
lu
m
e
0
6
,
Iss
u
e
0
4
,
p
p
.
4
3
4
0
,
2
0
1
9
.
[1
8
]
Ib
ra
h
im
S
,
Ja
m
a
lu
d
d
in
K
R,
S
a
m
a
h
K
A
F
A
.
S
e
c
u
rit
y
A
u
th
e
n
ti
c
a
ti
o
n
f
o
r
S
t
u
d
e
n
t
Ca
rd
s’
Bi
o
m
e
tri
c
R
e
c
o
g
n
it
io
n
Us
in
g
V
i
o
la
-
Jo
n
e
s
A
lg
o
rit
h
m
.
In
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
E
n
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
.
Vo
l.
1
1
,
No
.
1
,
p
p
.
2
4
1
-
2
4
7
,
2
0
1
8
.
[1
9
]
Yi
-
Qin
g
W
.
A
n
A
n
a
l
y
sis o
f
th
e
Vio
la
-
Jo
n
e
s F
a
c
e
De
tec
ti
o
n
A
lg
o
rit
h
m
.
Ima
g
e
Pro
c
e
ss
in
g
On
L
i
n
e
(
I
POL
)
,
p
p
.
1
2
8
-
1
4
8
,
2
0
1
4
.
[2
0
]
Eh
sa
n
S
,
Clark
A
F
,
Re
h
m
a
n
N
U,
M
c
Do
n
a
ld
-
M
a
ier
K
D.
I
n
teg
ra
l
Im
a
g
e
s:
Eff
icie
n
t
A
lg
o
rit
h
m
s
f
o
r
T
h
e
ir
Co
m
p
u
tatio
n
a
n
d
S
to
ra
g
e
in
Re
so
u
rc
e
-
Co
n
stra
i
n
e
d
Em
b
e
d
d
e
d
V
isio
n
S
y
ste
m
s.
S
e
n
so
rs
.
V
o
l
.
1
5
,
Iss
u
e
7
,
p
p
.
1
6
8
0
4
-
1
6
8
3
0
,
2
0
1
5
.
[2
1
]
Da
tcu
D,
Ro
t
h
k
ra
n
tz
L
.
M
u
lt
im
o
d
a
l
W
e
b
b
a
se
d
sy
ste
m
f
o
r
h
u
m
a
n
e
m
o
ti
o
n
re
c
o
g
n
it
io
n
.
5
t
h
I
n
ter
n
a
ti
o
n
a
l
I
n
d
u
stria
l
S
imu
l
a
ti
o
n
C
o
n
fer
e
n
c
e
2
0
0
7
,
I
S
C
2
0
0
7
.
[2
2
]
S
h
a
f
a
f
I,
Nu
rn
a
z
ih
a
h
W
,
A
h
m
a
d
F
A
F
,
Nu
r
N
A
M
,
Zaa
b
a
A
.
A
u
to
m
a
ti
c
Clas
si
f
ica
ti
o
n
o
f
P
a
d
d
y
Lea
f
Dise
a
se
.
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
tri
c
a
l
En
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
1
6
,
p
p
.
7
6
4
7
7
2
,
M
a
y
2
0
1
9
.
[2
3
]
S
h
a
f
a
f
I,
Nu
ru
l
A
Z,
Nu
rb
a
it
y
S
,
A
n
is
A
S
,
M
o
h
d
R
M
N.
Rice
g
ra
in
c
las
sif
ic
a
ti
o
n
u
sin
g
m
u
lt
i
-
c
las
s
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
(S
VM).
IA
ES
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Art
if
icia
l
In
tell
ig
e
n
c
e
(
I
J
-
AI)
.
V
o
l.
8
,
No
.
3
,
p
p
2
1
5
-
2
2
0
,
2
0
1
9
.
[2
4
]
A
h
m
e
d
R,
E
m
o
n
K
E
K,
Ho
ss
a
in
M
F
.
Ro
b
u
st
Driv
e
r
F
a
ti
g
u
e
Re
c
o
g
n
it
io
n
Us
in
g
Im
a
g
e
P
ro
c
e
ss
in
g
.
In
2
0
1
4
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
I
n
fo
r
ma
ti
c
s,
El
e
c
tro
n
ics
&
Vi
sio
n
(
ICIEV
)
p
p
.
1
-
6
,
2
0
1
4
.
[2
5
]
A
b
d
u
l
G
L
,
A
n
th
o
n
y
M
.
Cli
n
ica
l
T
e
st:
S
e
n
siti
v
it
y
a
n
d
S
p
e
c
if
icit
y
.
Co
n
ti
n
u
in
g
Ed
u
c
a
ti
o
n
in
A
n
a
e
sth
e
sia
Criti
c
a
l
Ca
re
&
Pa
in
,
V
o
l
.
8
,
Iss
u
e
.
6
,
p
p
.
2
2
1
-
2
2
3
,
De
c
e
m
b
e
r
2
0
0
8
.
[2
6
]
W
o
n
g
H
B,
L
im
G
H.
M
e
a
su
re
s
o
f
Dia
g
n
o
stic
A
c
c
u
ra
c
y
:
S
e
n
siti
v
it
y
,
S
p
e
c
if
icit
y
,
P
P
V
a
n
d
N
P
V
.
Pro
c
e
e
d
in
g
o
f
S
in
g
a
p
o
re
He
a
l
t
h
c
a
re
,
V
o
l
.
2
0
,
N
o
.
4
,
2
0
1
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.