I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
,
p
p
.
2
4
3
3
~
2
4
4
1
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
8
i
4
.
p
p
2
4
3
3
-
2441
2433
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Info
r
m
a
t
io
n Re
tr
iev
a
l f
ro
m
E
m
o
ti
o
ns a
nd Ey
e Blin
k
s w
ith
help
o
f
Senso
r N
o
des
P
un
ee
t
Sin
g
h L
a
m
ba
1
,
Dee
p
a
li Vir
m
a
ni
2
1
Un
iv
e
rsit
y
S
c
h
o
o
l
o
f
In
f
o
rm
a
ti
o
n
,
Co
m
m
u
n
ica
ti
o
n
a
n
d
T
e
c
h
n
o
l
o
g
y
,
GG
S
IP
U
,
I
n
d
ia
2
De
p
a
rt
m
e
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
,
Bh
a
g
wa
n
P
a
rsh
u
ra
m
In
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
,
GG
S
IP
U
,
I
n
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
2
,
2
0
1
8
R
ev
i
s
ed
A
p
r
30
,
2
0
1
8
A
cc
ep
ted
Ma
y
17
,
2
0
1
8
In
e
v
e
r
y
d
a
y
li
fe
,
th
e
re
a
re
situ
a
ti
o
n
s
w
h
e
re
th
e
o
n
ly
w
a
y
to
c
o
m
m
u
n
ica
te
a
re
e
m
o
ti
o
n
s.
EM
OT
ICON
S
a
re
th
e
e
p
it
o
m
e
o
f
th
e
sa
m
e
.
T
h
is
a
sp
e
c
t
o
f
c
o
m
m
u
n
ica
ti
o
n
c
a
n
a
lso
b
e
u
se
d
in
e
m
e
rg
e
n
c
y
situ
a
ti
o
n
s
(terr
o
ri
st
a
tt
a
c
k
s,
h
ij
a
c
k
s)
w
h
e
re
th
e
o
n
ly
w
a
y
to
c
o
m
m
u
n
ica
te
is
b
y
p
e
rf
o
r
m
in
g
so
m
e
e
x
trao
rd
in
a
ry
a
c
ti
o
n
s
o
r
t
h
ro
u
g
h
so
m
e
e
m
o
ti
o
n
s.
I
n
c
o
rp
o
ra
ti
n
g
tec
h
n
o
lo
g
y
to
th
e
a
b
o
v
e
m
e
n
ti
o
n
e
d
c
ircu
m
sta
n
c
e
s
th
e
p
a
p
e
r
p
ro
p
o
se
s
a
n
o
v
e
l
f
r
a
m
e
w
o
rk
o
f
d
e
tec
ti
n
g
a
n
e
m
e
rg
e
n
c
y
situ
a
ti
o
n
b
y
re
tri
e
v
in
g
in
f
o
r
m
a
ti
o
n
f
ro
m
e
m
o
ti
o
n
s
a
n
d
e
y
e
b
li
n
k
s
u
sin
g
se
n
so
r
n
o
d
e
s.
T
h
e
p
ro
p
o
se
d
f
ra
m
e
w
o
rk
c
a
n
b
e
d
e
p
lo
y
e
d
in
p
lac
e
s
(h
o
tels,
b
a
n
k
s,
a
ir
p
o
rts
e
tc.)
w
h
ich
a
re
m
o
re
su
sp
e
c
ted
to
a
tt
a
c
k
s.
T
h
e
f
r
a
m
e
w
o
rk
ta
k
e
s
in
p
u
t
f
ro
m
re
a
l
ti
m
e
p
a
ra
m
e
ters
:
e
y
e
b
li
n
k
s,
e
m
o
ti
o
n
s,
h
e
a
rt
ra
te.
Ba
se
d
o
n
b
e
h
a
v
io
ra
l
c
h
a
n
g
e
s,
b
io
lo
g
ica
l
c
h
a
n
g
e
s
a
n
d
p
h
y
sic
a
l
c
h
a
n
g
e
s
th
e
p
ro
p
o
se
d
f
ra
m
e
w
o
rk
e
x
tra
c
ts
m
e
ti
c
u
lo
u
s
in
f
o
rm
a
ti
o
n
.
T
h
e
p
ro
p
o
se
d
f
ra
m
e
w
o
rk
is
f
u
rth
e
r
v
a
li
d
a
ted
th
r
o
u
g
h
im
p
le
m
e
n
tatio
n
o
f
a
fa
c
ial
e
m
o
ti
o
n
re
c
o
g
n
it
i
o
n
sy
ste
m
th
a
t
su
c
c
e
ss
f
u
ll
y
r
e
c
o
g
n
ize
s
v
a
rio
u
s
h
u
m
a
n
e
m
o
ti
o
n
s.
T
h
e
f
a
c
ial
e
m
o
ti
o
n
re
c
o
g
n
it
io
n
sy
ste
m
o
f
th
e
p
ro
p
o
se
d
f
ra
m
e
w
o
rk
is
c
o
m
p
a
re
d
w
it
h
e
x
isti
n
g
S
V
M
tec
h
n
iq
u
e
in
term
s
o
f
a
c
c
u
ra
c
y
,
train
in
g
a
n
d
tes
ti
n
g
e
rro
r.
Ac
c
u
ra
c
y
w
it
h
th
e
p
ro
p
o
se
d
sy
ste
m
is
in
c
re
a
se
d
to
7
8
.
4
0
%
in
c
o
m
p
a
riso
n
w
it
h
e
x
isti
n
g
S
VM
th
a
t
is
7
5
.
3
7
%
a
n
d
th
e
trai
n
in
g
e
rro
r
is
d
e
c
re
a
se
d
to
0
.
0
0
4
1
0
3
w
h
e
re
a
s
w
it
h
th
e
e
x
isti
n
g
S
V
M
m
e
th
o
d
train
in
g
e
rro
r
is
0
.
0
0
8
9
3
5
.
K
ey
w
o
r
d
:
E
m
o
tico
n
s
E
m
o
tio
n
s
E
y
e
b
lin
k
I
n
f
o
r
m
a
tio
n
R
e
tr
iev
al
Sen
s
o
r
n
e
t
w
o
r
k
s
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
P
u
n
ee
t Si
n
g
h
L
a
m
b
a,
Un
i
v
er
s
it
y
Sc
h
o
o
l o
f
I
n
f
o
r
m
at
io
n
,
C
o
m
m
u
n
icatio
n
a
n
d
T
ec
h
n
o
lo
g
y
,
Gu
r
u
Go
b
in
d
Si
n
g
h
I
n
d
r
ap
r
asth
a
Un
iv
er
s
it
y
,
Secto
r
1
6
C
,
Kak
r
o
la,
Ne
w
De
lh
i,
Del
h
i 1
1
0
0
7
8
,
I
n
d
ia
.
Mo
b
ile:
0
9
8
1
0
9
0
7
1
3
8
E
m
ail:
s
in
g
h
s
.
p
u
n
ee
t
@
g
m
ai
l.c
o
m
.
1.
I
NT
RO
D
UCT
I
O
N
I
n
r
ec
en
t
ti
m
e
s
m
a
n
y
e
v
e
n
ts
o
f
h
o
s
ta
g
es a
n
d
h
ij
ac
k
s
h
av
e
h
a
p
p
en
ed
an
d
th
er
e
i
s
a
n
u
r
g
e
n
t
n
ee
d
o
f
a
n
au
to
m
ated
s
y
s
te
m
w
h
ic
h
ca
n
m
ak
e
t
h
e
co
n
ce
r
n
ed
ag
e
n
cies
a
w
ar
e
o
f
t
h
e
s
itu
a
tio
n
in
ti
m
e
an
d
th
u
s
h
elp
th
e
m
to
tak
e
n
ec
ess
ar
y
ac
tio
n
.
W
ir
eless
Sen
s
o
r
Net
w
o
r
k
s
ca
n
b
e
u
s
ed
in
at
tack
p
r
o
n
e
ar
ea
s
to
ex
a
m
i
n
e
t
h
e
s
i
tu
atio
n
an
d
h
en
ce
w
i
ll h
e
lp
th
e
co
n
ce
r
n
ed
ag
en
c
ies to
tak
e
ap
p
r
o
p
r
iate
ac
tio
n
in
ti
m
e.
T
h
e
2
6
/
1
1
Mu
m
b
ai
attac
k
s
wer
e
a
p
lan
n
ed
s
er
ie
s
o
f
attac
k
s
,
w
h
e
n
a
g
r
o
u
p
o
f
1
0
-
1
2
ter
r
o
r
is
ts
o
f
L
as
h
k
ar
-
e
-
T
aib
a,
ca
r
r
ie
d
o
u
t
a
s
eq
u
en
ce
o
f
co
o
r
d
in
ated
b
o
m
b
i
n
g
a
n
d
s
h
o
o
tin
g
th
a
t
co
n
tin
u
ed
f
o
r
f
o
u
r
d
ay
s
[
1
]
,
[
2
]
.
Den
s
el
y
p
o
p
u
lated
ar
ea
s
lik
e
r
ail
w
a
y
s
tatio
n
an
d
5
s
tar
h
o
tels
(
Ho
tel
T
aj
)
w
er
e
th
eir
m
ai
n
tar
g
ets.
T
h
e
attac
k
s
,
w
h
ic
h
wer
e
h
ea
v
i
l
y
ce
n
s
u
r
ed
g
lo
b
all
y
,
b
eg
an
o
n
2
6
th
No
v
e
m
b
er
an
d
last
ed
u
n
til
2
9
th
No
v
e
m
b
er
2
0
0
8
,
k
illi
n
g
1
6
4
p
eo
p
le
an
d
w
o
u
n
d
i
n
g
at
least
3
0
0
.
A
lo
t
o
f
d
am
a
g
e
h
ad
alr
ea
d
y
b
ee
n
d
o
n
e
ev
e
n
b
ef
o
r
e
t
h
e
s
ec
u
r
it
y
a
g
e
n
cies
c
a
m
e
to
k
n
o
w
w
h
a
t
ac
tu
all
y
h
ad
h
ap
p
en
ed
.
So
if
th
er
e
w
o
u
ld
h
av
e
b
ee
n
s
u
c
h
au
to
m
ated
s
y
s
te
m
s
,
a
lo
t
w
o
u
l
d
h
av
e
b
ee
n
s
a
v
ed
in
ter
m
s
o
f
liv
es a
n
d
p
u
b
lic
p
r
o
p
er
ty
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
24
33
–
2
4
4
1
2434
An
o
th
er
s
u
c
h
e
v
en
t
o
f
h
o
s
ta
g
e
i
n
clu
d
es
h
ij
ac
k
i
n
g
o
f
I
n
d
ian
A
ir
li
n
e
F
lig
h
t
8
1
4
en
r
o
u
te
f
r
o
m
T
r
i
b
h
u
v
a
n
I
n
ter
n
atio
n
al
A
ir
p
o
r
t,
Nep
al
to
I
n
d
ir
a
Ga
n
d
h
i
A
ir
p
o
r
t,
I
n
d
ia
o
n
Frid
a
y
,
2
4
Dec
e
m
b
er
1
9
9
9
.
T
h
e
m
o
s
t
r
ec
en
t
Ur
i
attac
k
w
as
an
in
h
u
m
an
attac
k
b
y
h
ea
v
il
y
ar
m
ed
ter
r
o
r
is
ts
in
Ur
i
ar
ea
o
f
B
ar
am
u
lla
d
is
tr
ict
,
J
am
m
u
an
d
Ka
s
h
m
ir
.
I
t
i
s
co
n
s
id
er
ed
as
"
th
e
d
ea
d
lies
t
att
ac
k
o
n
s
ec
u
r
it
y
f
o
r
ce
s
i
n
last
t
w
o
d
ec
ad
es”.
N
o
ter
r
o
r
is
t g
r
o
u
p
h
as c
lai
m
ed
r
es
p
o
n
s
ib
ilit
y
f
o
r
th
e
s
a
m
e.
I
n
t
h
is
p
ap
er
w
e
p
r
o
p
o
s
e
a
n
o
v
el
f
r
a
m
e
w
o
r
k
f
o
r
d
etec
tin
g
an
e
m
er
g
en
c
y
s
it
u
atio
n
w
h
ic
h
u
s
e
s
t
h
e
in
telli
g
e
n
t
in
f
o
r
m
a
tio
n
r
etr
iev
ed
f
r
o
m
p
u
ls
e
r
ate
an
d
e
y
e
b
lin
k
s
[
3
]
u
s
i
n
g
w
ir
ele
s
s
s
e
n
s
o
r
n
o
d
es
as
w
ell
a
s
b
eh
av
io
r
al
p
atter
n
s
[
4
]
,
[
5
]
ex
h
ib
ited
b
y
h
u
m
a
n
s
in
r
esp
o
n
s
e
to
an
y
ch
a
n
g
es
in
th
e
f
o
r
m
er
p
ar
am
eter
s
.
T
o
th
e
b
est
o
f
o
u
r
k
n
o
w
led
g
e,
s
u
ch
s
ec
u
r
it
y
f
r
a
m
e
w
o
r
k
th
at
co
n
f
ir
m
s
e
m
er
g
e
n
c
y
o
r
in
tr
u
s
io
n
d
e
tectio
n
o
n
all
g
i
v
en
p
ar
am
eter
s
d
o
es n
o
t e
x
i
s
t.
T
h
e
p
ar
am
eter
s
i
n
r
esp
o
n
s
e
ab
n
o
r
m
al
s
i
tu
at
io
n
d
is
c
u
s
s
ed
h
er
e
ar
e:
a.
E
y
e
b
lin
k
p
atter
n
,
b.
Hu
m
an
E
m
o
tio
n
s
,
c.
Hea
r
t r
ate
Fo
r
ea
ch
p
ar
am
eter
w
e
in
te
n
t
to
an
al
y
ze
t
h
e
b
eh
av
io
r
al
(
em
o
tio
n
)
,
b
io
lo
g
ical
(
p
u
ls
e
r
ate)
an
d
p
h
y
s
ical
(
e
y
e
b
lin
k
)
p
atter
n
ch
an
g
es
as
m
ea
s
u
r
ed
b
y
t
h
e
s
y
s
te
m
s
d
ep
lo
y
ed
w
h
ic
h
ca
n
i
n
f
u
tu
r
e
b
e
u
s
ed
to
in
s
ta
ll
an
e
m
er
g
e
n
c
y
o
r
i
n
tr
u
s
i
o
n
w
ar
n
i
n
g
s
y
s
te
m
.
Au
t
h
o
r
in
[
6
]
h
av
e
p
r
ese
n
ted
a
n
o
v
e
l
ap
p
r
o
ac
h
to
d
etec
t
w
h
et
h
e
r
th
e
e
y
e
s
i
n
a
n
i
m
a
g
e
(
s
till
)
ar
e
clo
s
ed
o
r
n
o
t
an
d
th
is
ap
p
r
o
ac
h
h
a
s
w
id
e
r
an
g
in
g
ap
p
lic
atio
n
s
i
n
f
ac
ial
ex
p
r
es
s
io
n
r
ec
o
g
n
itio
n
,
f
at
ig
u
e
d
etec
tio
n
,
an
d
s
o
o
n
.
A
s
i
m
il
ar
an
al
y
s
is
i
n
[
7
]
h
as
b
ee
n
d
i
s
cu
s
s
ed
to
d
etec
t
e
y
e
b
li
n
k
s
i
n
a
v
id
eo
s
eq
u
en
ce
fr
o
m
a
s
t
an
d
ar
d
ca
m
er
a.
E
m
o
t
io
n
is
also
o
n
e
o
f
th
e
p
ar
a
m
et
er
s
w
h
ich
p
la
y
a
n
i
m
p
o
r
tan
t
p
ar
t
in
o
u
r
p
r
o
p
o
s
e
d
f
r
a
m
e
w
o
r
k
.
I
n
[
8
]
a
r
eg
io
n
al
h
id
d
en
Ma
r
k
o
v
m
o
d
el
(
R
HM
M)
is
p
r
o
p
o
s
ed
f
o
r
r
ec
o
g
n
iz
in
g
f
ac
ial
e
x
p
r
ess
io
n
s
in
v
id
eo
s
eq
u
e
n
ce
s
.
Facial
ac
tio
n
u
n
i
ts
ar
e
d
escr
ib
ed
f
o
r
e
y
eb
r
o
w
s
,
e
y
e
s
an
d
m
o
u
th
r
e
g
io
n
s
b
y
R
HM
M
s
.
P
r
o
p
o
s
ed
tech
n
iq
u
e
h
a
s
o
u
tp
er
f
o
r
m
ed
o
th
er
ex
is
t
in
g
m
e
t
h
o
d
s
w
h
e
n
test
ed
w
it
h
ex
te
n
d
ed
C
o
h
n
-
Ka
n
ad
e
d
atab
ase.
Au
th
o
r
i
n
[
9
]
p
r
ese
n
ts
a
n
Au
to
m
at
ic
Face
An
al
y
s
i
s
(
A
F
A
)
s
y
s
te
m
to
i
n
s
p
ec
t
f
ac
ial
e
x
p
r
ess
io
n
s
tak
i
n
g
in
to
co
n
s
id
er
atio
n
b
o
th
p
er
m
an
e
n
t
f
ac
ial
f
ea
tu
r
es
(
b
r
o
w
s
,
e
y
es,
m
o
u
t
h
)
an
d
m
o
m
e
n
tar
y
f
ac
ia
l
f
ea
tu
r
e
s
(
d
ee
p
en
in
g
o
f
f
ac
ial
f
u
r
r
o
w
s
)
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
Sen
s
o
r
No
d
es
as
k
n
o
w
n
f
o
r
t
h
e
m
o
n
i
to
r
in
g
ab
ilit
ies
s
u
ch
a
s
ac
cu
r
ac
y
a
n
d
s
e
n
s
it
iv
i
t
y
ar
e
u
s
ed
as
t
h
e
b
aselin
e
f
o
r
th
e
p
r
o
p
o
s
ed
f
r
am
e
w
o
r
k
.
Fi
g
u
r
e
1
d
escr
ib
es th
e
p
r
o
p
o
s
ed
Fra
m
e
w
o
r
k
.
T
h
e
in
telli
g
en
t
in
f
o
r
m
a
tio
n
c
o
llected
f
r
o
m
s
ta
n
d
alo
n
e
p
ar
a
m
eter
s
,
w
i
ll
ac
t
as
in
p
u
t
to
th
e
s
y
s
te
m
w
h
ic
h
af
ter
p
er
f
o
r
m
i
n
g
ca
lc
u
l
atio
n
s
w
i
ll d
ec
id
e
w
h
et
h
er
it
h
as c
r
o
s
s
ed
th
e
t
h
r
es
h
o
ld
o
r
n
o
t.
Up
o
n
cr
o
s
s
in
g
t
h
e
th
r
es
h
o
ld
an
alar
m
o
r
b
u
zz
er
o
r
s
o
m
e
m
e
s
s
a
g
e
w
ill
b
e
tr
an
s
m
i
tted
to
th
e
co
n
ce
r
n
ed
au
th
o
r
ities
.
T
h
e
s
en
s
o
r
s
n
o
d
es
w
i
ll
b
e
d
ep
lo
y
ed
to
ex
tr
ac
t
th
e
p
r
o
f
o
u
n
d
in
f
o
r
m
atio
n
f
r
o
m
th
e
g
iv
e
n
p
ar
a
m
eter
s
,
t
h
u
s
ca
n
b
e
u
s
ed
as
fo
r
er
u
n
n
er
f
o
r
p
r
o
v
o
k
in
g
t
h
e
a
lar
m
m
e
s
s
a
g
es.
a.
E
y
e
b
lin
k
T
h
is
attr
ib
u
te
is
o
f
u
t
m
o
s
t
i
m
p
o
r
tan
ce
as
f
ar
as
o
u
r
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
is
co
n
ce
r
n
ed
.
Av
er
ag
e
e
y
e
b
lin
k
i
n
g
r
ate
is
ar
o
u
n
d
1
2
-
1
9
p
er
m
i
n
u
te.
(
I
n
ten
tio
n
al
E
y
e
b
lin
k
s
>
2
5
A
b
n
o
r
m
al)
.
P
r
o
p
o
s
ed
f
r
am
e
w
o
r
k
u
s
es t
h
e
s
en
s
o
r
p
lace
d
in
th
e
li
n
e
o
f
s
ig
h
t o
f
t
h
e
e
y
e
w
h
ic
h
co
n
t
in
u
o
u
s
l
y
s
tu
d
ie
s
th
e
b
lin
k
p
atter
n
.
Se
n
s
o
r
s
u
s
ed
w
ill d
r
a
w
a
cu
r
v
e
to
in
d
icate
a
ch
an
g
e
in
e
y
e
b
lin
k
[
1
0
]
,
[
1
1
]
r
ate.
b.
Hu
m
an
e
m
o
tio
n
s
E
m
o
tio
n
i
s
a
co
g
n
i
tiv
e
s
tate
t
h
at
is
g
e
n
er
ated
i
m
p
u
l
s
i
v
el
y
r
ath
er
th
a
n
t
h
r
o
u
g
h
r
esp
o
n
s
iv
e
ef
f
o
r
t
a
n
d
is
ac
co
m
p
an
ied
b
y
p
h
y
s
io
lo
g
ical
c
h
an
g
es
w
h
ic
h
ar
e
f
a
m
iliar
to
th
e
o
u
ter
w
o
r
ld
.
So
m
e
e
m
o
t
io
n
s
w
h
ic
h
h
u
m
a
n
s
f
ac
e
in
e
v
er
y
d
a
y
l
if
e
ar
e
w
o
r
r
y
,
s
u
r
p
r
is
e,
co
n
f
u
s
e,
h
ap
p
y
etc.
Hu
m
a
n
v
i
s
io
n
ca
n
p
r
ac
tice
em
o
tio
n
as
co
u
p
led
w
it
h
p
er
s
o
n
alit
y
,
te
m
p
er
a
m
e
n
t,
an
d
m
o
o
d
.
C
o
m
p
u
ter
Vis
io
n
tr
ie
s
to
f
o
llo
w
t
h
e
h
u
m
an
v
i
s
io
n
b
y
an
al
y
z
in
g
d
i
g
ital
i
m
a
g
e
as
in
p
u
t.
Se
n
s
o
r
s
d
ep
lo
y
ed
w
il
l
co
n
tin
u
o
u
s
l
y
s
t
u
d
y
t
h
e
h
u
m
a
n
e
m
o
tio
n
s
.
A
d
atab
a
s
e
o
f
e
m
o
t
io
n
s
w
it
h
an
i
n
s
tan
ce
o
f
ti
m
e
w
ill b
e
m
ai
n
tai
n
ed
w
it
h
s
en
s
o
r
n
o
d
es.
c.
Hea
r
t r
ate
Hea
r
t
r
ate
is
th
e
n
u
m
b
er
o
f
ti
m
es
a
h
ea
r
t
b
ea
ts
i
n
a
m
i
n
u
te
.
Hea
r
t
r
ate
[
1
2
]
v
ar
ies
f
r
o
m
p
er
s
o
n
to
p
er
s
o
n
d
ep
en
d
in
g
o
n
o
n
e
’
s
a
g
e,
b
o
d
y
s
ize,
f
it
n
ess
a
n
d
h
ea
r
t
co
n
d
iti
o
n
.
T
h
e
s
a
m
e
al
s
o
d
ep
en
d
s
o
n
w
h
et
h
er
th
e
p
er
s
o
n
is
r
u
n
n
in
g
,
s
tan
d
i
n
g
,
s
itti
n
g
,
m
o
v
in
g
o
r
i
s
u
n
d
er
i
n
f
lu
en
ce
o
f
s
o
m
e
m
ed
icatio
n
.
E
m
o
tio
n
s
ca
n
al
s
o
in
cr
ea
s
e
t
h
e
h
ea
r
t
r
ate.
A
h
ea
lth
y
h
ea
r
t
b
ea
ts
6
0
to
1
0
0
ti
m
es
in
a
m
i
n
u
te
a
n
d
ca
n
s
p
i
k
e
u
p
d
u
e
to
s
tr
es
s
,
o
v
e
r
ex
er
tio
n
o
r
n
er
v
o
u
s
n
es
s
.
Ou
r
p
r
o
p
o
s
ed
f
r
am
e
w
o
r
k
w
i
ll
co
n
tin
u
o
u
s
l
y
m
o
n
ito
r
th
e
h
ea
r
t r
ate.
T
h
e
p
r
o
p
o
s
ed
f
r
am
e
w
o
r
k
u
n
f
o
ld
s
th
e
e
n
tire
s
tr
u
ct
u
r
e
o
f
t
h
e
p
r
o
ce
s
s
co
m
b
i
n
ed
to
g
et
h
er
.
S
m
ar
t
m
o
te
s
(
n
o
d
es)
w
ill
b
e
i
n
s
ta
lled
as
p
er
th
e
r
eq
u
ir
e
m
e
n
ts
.
T
h
r
ee
t
y
p
es
o
f
s
e
n
s
o
r
s
(
e
y
e
b
li
n
k
,
e
m
o
tio
n
d
etec
tio
n
a
n
d
h
ea
r
t
r
ate
s
e
n
s
o
r
s
)
w
ill
b
e
d
ep
lo
y
ed
.
T
h
e
s
en
s
o
r
s
w
il
l
wo
r
k
in
d
ep
en
d
en
t
l
y
a
n
d
w
i
ll
t
r
an
s
f
er
th
e
s
en
s
ed
attr
ib
u
tes
to
th
e
s
m
ar
t
s
er
v
er
v
ia
s
m
ar
t
g
ate
w
h
er
e
t
h
e
ac
t
u
al
f
ea
t
u
r
e
s
elec
tio
n
w
ill
ta
k
e
p
lace
.
T
h
en
a
f
ter
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
I
n
fo
r
ma
tio
n
R
etri
ev
a
l fro
m
E
mo
tio
n
s
a
n
d
E
ye
B
lin
ks w
ith
h
elp
o
f S
en
s
o
r
…
(
P
u
n
ee
t S
in
g
h
La
mb
a
)
2435
s
o
m
e
o
b
s
er
v
at
io
n
,
ca
lc
u
latio
n
an
d
co
m
p
ar
is
o
n
s
,
t
h
e
f
in
a
l
v
er
d
ict
(
w
h
e
th
er
t
h
e
t
h
r
es
h
o
ld
is
r
ea
ch
ed
o
r
n
o
t
)
w
il
l
b
e
ta
k
e
n
.
I
f
t
h
r
es
h
o
ld
is
cr
o
s
s
ed
,
alar
m
w
ill
b
e
r
aised
.
Fo
r
ex
a
m
p
le,
if
t
h
e
e
y
e
b
li
n
k
s
en
s
o
r
s
e
n
s
e
s
a
n
u
n
u
s
u
al
b
li
n
k
p
atter
n
o
r
n
u
m
b
er
,
in
ad
d
itio
n
to
a
Fear
o
r
Su
r
p
r
is
e
em
o
tio
n
d
etec
ted
b
y
t
h
e
e
m
o
tio
n
d
etec
tio
n
s
en
s
o
r
o
r
an
ab
n
o
r
m
al
h
ea
r
t
r
ate,
th
e
s
itu
at
io
n
is
a
n
a
la
r
m
in
g
o
n
e.
B
lin
k
p
atter
n
a
n
d
em
o
tio
n
s
ca
n
b
e
ca
lcu
lated
b
y
s
til
l
o
r
m
o
v
i
n
g
i
m
a
g
es
v
ia
s
p
ec
ialized
cc
t
v
’
s
.
Fo
r
h
ea
r
t
r
ate
co
n
ce
r
n
ed
p
er
s
o
n
s
(
b
ar
ten
d
er
s
i
n
h
o
tels
,
b
an
k
m
a
n
a
g
er
s
,
f
l
ig
h
t a
tten
d
an
t
s
)
w
ill b
e
eq
u
ip
p
ed
w
i
th
w
ir
eles
s
h
ea
r
t r
ate
s
e
n
s
o
r
s
o
r
s
tr
a
p
s
.
Fig
u
r
e
1
.
P
r
o
p
o
s
ed
F
r
am
e
w
o
r
k
f
o
r
E
x
i
g
e
n
c
y
d
etec
tio
n
T
h
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
w
o
r
k
s
in
t
w
o
p
h
ase
s
.
T
h
e
f
ir
s
t
p
h
ase
(
an
al
y
s
is
p
h
ase)
tak
e
s
i
n
p
u
t
f
r
o
m
e
y
e
b
lin
k
s
,
e
m
o
tio
n
s
a
n
d
h
ea
r
t
r
at
e.
T
h
e
s
ec
o
n
d
p
h
ase
co
n
s
tit
u
t
es
th
e
cla
s
s
i
f
ica
tio
n
p
h
a
s
e.
T
h
e
r
ea
ctio
n
s
co
u
ld
b
e
ch
an
g
e
i
n
b
eh
a
v
io
r
o
r
e
m
o
tio
n
s
o
r
p
h
y
s
io
lo
g
ical
o
v
er
g
iv
e
n
t
i
m
e.
I
n
t
h
i
s
p
ap
er
a
P
er
s
p
i
ca
cio
u
s
I
n
f
o
r
m
a
tio
n
R
etr
iev
al
tech
n
iq
u
e
w
h
ich
s
e
n
s
e
s
s
er
io
u
s
d
iv
er
s
io
n
o
f
p
ar
am
eter
s
f
r
o
m
b
aseli
n
e
v
a
lu
e
s
v
ia
s
e
n
s
o
r
s
s
p
an
n
ed
ac
r
o
s
s
tar
g
et
zo
n
e
s
is
p
r
esen
t
ed
.
T
h
e
d
ata
is
tr
an
s
m
itted
to
a
b
ase
s
tatio
n
w
h
ic
h
is
f
u
r
t
h
er
p
r
o
ce
s
s
ed
at
m
ai
n
s
er
v
er
.
T
h
e
m
o
m
en
t
m
aj
o
r
d
ef
lectio
n
f
r
o
m
n
o
r
m
al
i
s
o
b
s
er
v
ed
u
s
in
g
t
h
e
i
n
f
o
r
m
atio
n
g
a
t
h
er
ed
,
s
en
s
o
r
s
ar
e
tr
ain
ed
to
is
s
u
e
alar
m
s
ig
n
als
as in
w
h
en
s
u
c
h
ab
n
o
r
m
a
l a
cti
v
it
y
is
s
e
n
s
ed
.
2
.
1
.
I
m
ple
m
ent
ing
info
r
m
a
t
io
n r
et
riev
a
l t
ec
hn
iqu
e
Fig
u
r
e
2
s
h
o
w
s
h
o
w
m
et
h
o
d
ically
P
er
s
p
icac
io
u
s
I
n
f
o
r
m
atio
n
R
etr
ie
v
al
T
ec
h
n
iq
u
e
(
P
I
R
T
)
is
i
m
p
le
m
en
ted
at
s
er
v
er
s
w
h
er
e
m
e
s
s
a
g
es
s
tati
n
g
u
n
u
s
u
al
h
u
m
a
n
e
m
o
t
io
n
,
p
u
ls
e
r
ate
an
d
b
lin
k
p
atter
n
ar
e
r
ec
eiv
ed
at
b
ase
s
tatio
n
.
I
t
is
i
m
p
o
r
ta
n
t
to
k
n
o
w
th
at
i
n
p
u
t
v
alu
e
s
w
o
u
ld
b
e
g
ath
er
ed
f
r
o
m
s
o
m
e
tr
ain
ed
s
ta
f
f
(
air
h
o
s
tes
s
i
n
air
p
la
n
es,
b
ar
te
n
d
er
s
i
n
h
o
tels
,
s
o
m
e
o
f
f
icial
s
i
n
b
an
k
s
etc)
h
av
i
n
g
t
h
e
k
n
o
w
led
g
e
o
f
h
o
w
to
r
ea
ct
in
an
e
m
er
g
e
n
c
y
s
it
u
atio
n
.
T
h
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
s
h
o
u
ld
b
e
d
ep
lo
y
ed
in
ar
ea
s
w
h
ic
h
ar
e
i
n
d
ir
ec
t
li
n
e
o
f
s
ig
h
t
o
f
t
h
e
co
n
ce
r
n
ed
p
er
s
o
n
s
a
n
d
w
ill
h
elp
g
en
er
ate
alar
m
s
i
g
n
als
a
s
s
o
o
n
as
c
h
an
g
e
i
n
b
eh
a
v
io
r
al
o
r
b
lin
k
p
atter
n
s
is
s
en
s
ed
.
Me
s
s
ag
e
s
ac
q
u
ir
ed
ar
e
f
u
r
t
h
er
r
ef
in
ed
in
o
r
d
er
to
ex
ce
r
p
t
in
telli
g
e
n
t
in
f
o
r
m
atio
n
f
r
o
m
it.
Fi
g
u
r
e
2
p
r
esen
ts
a
n
o
v
el
p
r
o
g
n
o
s
tic
al
g
o
r
ith
m
w
h
ich
tak
e
s
as
i
n
p
u
t
s
en
s
ed
d
ata
f
r
o
m
t
h
e
p
r
e
ce
d
in
g
s
ta
g
es
a
n
d
f
i
g
u
r
e
o
u
t
ce
r
tain
co
n
d
itio
n
s
o
n
t
h
e
d
ata
r
ec
eiv
ed
.
Var
io
u
s
s
ta
g
es o
f
P
I
R
T
ar
e
:
a.
Z
1
: E
y
e
b
li
n
k
T
h
is
s
tan
d
alo
n
e
co
n
d
it
io
n
ev
a
lu
ates
t
h
e
B
lin
k
p
atter
n
(
B
P
)
a
n
d
h
en
ce
t
h
e
n
u
m
b
er
s
o
f
B
lin
k
s
i
n
s
a
y
1
m
i
n
u
te.
Av
er
ag
e
e
y
e
b
lin
k
i
n
g
r
ate
is
ar
o
u
n
d
1
2
-
1
9
p
er
m
in
u
te.
So
an
u
n
u
s
u
a
l
n
u
m
b
er
/
p
atter
n
w
i
ll
b
e
th
e
in
d
icatio
n
t
h
at
s
o
m
e
th
i
n
g
ab
n
o
r
m
al
i
s
th
er
e.
Fo
r
a
p
r
ec
is
e
d
ec
is
io
n
w
e
ar
e
tak
i
n
g
it
2
5
B
lin
k
s
p
er
m
i
n
u
te.
T
h
e
v
alu
e
s
tated
w
h
ich
ap
p
r
o
x
i
m
a
tel
y
ac
co
u
n
ts
to
2
5
is
th
e
lo
w
est
d
ef
a
u
lt
v
a
lu
e.
As
n
o
s
i
g
n
s
o
f
ch
an
g
e
i
n
b
lin
k
p
atter
n
ar
e
o
b
s
er
v
ed
b
el
o
w
v
alu
e
-
No
alar
m
w
ill
b
e
r
aise
d
.
Fo
r
g
ettin
g
t
h
e
B
R
M
(
b
lin
k
r
ate
p
er
m
i
n
u
te)
,
f
r
a
m
e
s
f
r
o
m
th
e
v
id
eo
s
eq
u
e
n
ce
ca
n
b
e
ex
tr
ac
ted
an
d
w
ill
b
e
f
u
r
t
h
er
p
r
o
ce
s
s
ed
f
o
r
d
etec
tin
g
f
ac
e
a
n
d
h
e
n
ce
th
e
e
y
e
p
o
r
tio
n
o
f
th
e
i
m
a
g
e.
Fro
m
th
e
e
y
e
la
n
d
m
ar
k
s
[
7
]
d
etec
ted
f
r
o
m
t
h
e
i
m
ag
e
,
t
h
e
e
y
e
asp
ec
t r
atio
(
E
A
R
)
th
at
is
u
s
ed
as
a
n
ap
p
r
o
x
i
m
ate
o
f
th
e
e
y
e
o
p
en
i
n
g
/clo
s
i
n
g
s
t
ate
ca
n
b
e
ca
lcu
lated
.
T
h
e
ey
e
asp
ec
t
r
atio
(
E
A
R
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
24
33
–
2
4
4
1
2436
is
co
m
p
u
ted
b
et
w
ee
n
h
e
ig
h
t a
n
d
w
id
th
o
f
th
e
e
y
e.
C
h
ar
ac
ter
is
tic
r
atio
o
f
t
h
e
o
p
en
e
y
e
h
as
a
s
m
all
d
is
cr
ep
an
c
y
a
m
o
n
g
h
u
m
a
n
s
a
n
d
it
i
s
f
u
ll
y
u
n
if
o
r
m
to
a
co
n
s
tan
t
s
ca
li
n
g
o
f
th
e
i
m
a
g
e.
As
b
lin
k
i
n
g
i
s
p
r
ac
ticed
b
y
b
o
th
e
y
es
s
i
m
u
lta
n
e
o
u
s
l
y
,
t
h
e
E
AR
o
f
th
e
p
air
is
av
er
a
g
ed
.
T
h
e
E
AR
is
p
r
i
m
ar
il
y
co
n
s
ta
n
t
f
o
r
an
o
p
en
e
y
e
a
n
d
is
ap
p
r
o
x
im
a
tel
y
ze
r
o
f
o
r
a
clo
s
e
d
ey
e
.
P
a
tter
n
fo
r
co
n
d
itio
n
Z1
:
{
Z
1
: c
alcu
late
B
R
M
if
(
B
R
M<
2
5
)
n
o
aler
t ;
E
ls
e
p
r
o
ce
ed
to
Z
2
(
2
nd
co
n
d
itio
n
)
}
Fig
u
r
e
2
.
P
er
s
p
icac
io
u
s
I
n
f
o
r
m
atio
n
R
etr
ie
v
al
T
ec
h
n
iq
u
e
f
o
r
ex
ig
e
n
c
y
d
etec
tio
n
b.
Z
2
: H
ea
r
t
r
ate
T
h
is
co
n
d
itio
n
w
ill
b
e
ev
a
lu
at
ed
o
n
l
y
if
co
n
d
itio
n
Z
1
is
tr
u
e
an
d
w
ill
ev
a
lu
ate
w
h
et
h
er
th
e
Hea
r
t
r
ate
(
HR
)
o
r
p
u
ls
e
h
as
cr
o
s
s
ed
t
h
e
n
o
r
m
al
d
e
f
i
n
ed
r
ate.
I
f
it
h
a
s
,
alar
m
w
il
l
b
e
g
e
n
er
ated
;
o
th
er
w
i
s
e
w
e
p
r
o
ce
ed
to
ch
ec
k
t
h
e
n
e
x
t c
o
n
d
it
io
n
w
h
ic
h
is
g
etti
n
g
O
R
ed
w
it
h
th
e
c
u
r
r
en
t c
o
n
d
itio
n
.
P
a
tter
n
fo
r
co
n
d
itio
n
Z2
:
{
I
f
(
Z
1
)
{
Z
2
: c
alc
u
late
HR
//NH
R
N
o
r
m
al
Hea
r
t
R
ate
if
(
HR
>N
HR
)
g
en
er
ate
A
ler
t
s
ig
n
al
s
;
else
p
r
o
ce
e
d
to
Z
3
;
}}
c.
Z
3
: Fac
ial
e
m
o
tio
n
T
h
is
co
n
d
itio
n
ev
al
u
ate
s
th
e
e
m
o
tio
n
d
etec
ted
.
I
f
th
e
r
e
s
u
lta
n
t
e
m
o
tio
n
d
etec
t
ed
is
FEAR
o
r
SUR
P
R
I
SE
,
alar
m
s
i
g
n
al
w
i
ll
b
e
g
en
er
ated
(
o
n
l
y
i
f
co
n
d
iti
o
n
Z
1
s
tan
d
s
tr
u
e)
el
s
e
w
e
will
p
r
o
ce
ed
to
f
ir
s
t
co
n
d
itio
n
ag
ai
n
.
I
t
is
i
m
p
o
r
tan
t
to
n
o
te
th
a
t
ei
th
er
C
o
n
d
i
tio
n
2
(
Z
2
)
,
3
(
Z
3
)
o
r
b
o
th
s
h
o
u
ld
b
e
test
ed
to
g
en
er
ate
w
ar
n
in
g
m
es
s
ag
e
s
(
g
iv
en
Z
1
is
tr
u
e)
.
P
a
tter
n
fo
r
co
n
d
itio
n
Z3
:
{
If
(
Z
1
)
{Z
3
: D
etec
t E
m
o
t
io
n
Prognos
ti
c
Al
go
rit
hm
1.
Obta
in the
blink pa
t
te
rn
from
t
he
sensors
(1
m
i
n
dura
ti
on)
.
2.
For
a
giv
en
p
a
tt
ern
sa
y
num
ber
of
bl
inks b
e
BP
2.
1
If
BP
<= 25
2.
1
.
1
No
al
er
t
g
ene
ra
te
d
-
>
Go
to
Step
1
.
2.
2
W
hil
e
2
5
<
BP
2.
2
.
1
Chang
e in
Hea
rt
ra
te (HR)
and
Emotions (
EM)
are
sens
ed
2
.
2
.
1.
1
if
(
(HR
>
NH
R)
or
(
EM=
(Fea
r
or
Surprise)))
Go t
o
Step 3
E
lse
Go
to S
te
p
1
//
NH
R=
nor
m
al
hea
r
t
r
ate
3.
Gen
era
t
e
Al
er
t
signa
ls
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
I
n
fo
r
ma
tio
n
R
etri
ev
a
l fro
m
E
mo
tio
n
s
a
n
d
E
ye
B
lin
ks w
ith
h
elp
o
f S
en
s
o
r
…
(
P
u
n
ee
t S
in
g
h
La
mb
a
)
2437
if
(
E
m
o
tio
n
==
(
Fear
o
r
Su
r
p
r
is
e)
)
g
en
er
ate
A
ler
t sig
n
als
;
else
Saf
e
s
tate;
}}
2
.
2
.
F
a
ce
t
v
a
lue a
na
ly
s
is
T
h
e
f
ac
et
ev
al
u
atio
n
is
d
o
n
e
o
n
th
e
s
en
s
ed
d
ata
to
f
u
r
th
er
f
i
g
u
r
e
o
u
t
co
n
d
itio
n
s
a
n
d
g
e
n
er
ate
aler
t
s
ig
n
al
s
w
h
e
n
e
v
er
co
n
d
itio
n
Z
1
,
Z
2
o
r
Z
3
ev
al
u
ates
to
tr
u
e/
f
alse
i
n
ce
r
tai
n
p
er
m
u
tatio
n
s
o
r
co
m
b
i
n
atio
n
.
B
ased
o
n
ab
o
v
e
an
al
y
s
is
w
e
v
alid
ate
a
f
r
a
m
e
w
o
r
k
th
at
ca
n
h
atch
aler
t
s
i
g
n
als
b
ased
o
n
m
e
s
s
a
g
es
r
ec
eiv
ed
by
th
e
s
e
n
s
o
r
s
d
ep
lo
y
ed
.
Fi
g
u
r
e
3
s
h
o
w
s
s
o
m
e
ca
u
s
e
e
f
f
ec
t
g
r
ap
h
s
w
h
ich
s
tate
t
h
at
co
n
d
itio
n
Z
1
in
co
n
j
u
n
ctio
n
w
it
h
Z
2
o
r
Z
3
o
r
b
o
th
is
a
n
ec
ess
ar
y
p
r
er
eq
u
is
ite
f
o
r
th
e
f
r
am
e
w
o
r
k
to
is
s
u
e
w
ar
n
i
n
g
s
ig
n
al.
Fig
u
r
e
3
.
C
au
s
e
g
r
ap
h
b
ased
o
n
co
n
d
itio
n
E
v
al
u
ated
2
.
3
E
m
o
t
io
n
d
et
ec
t
io
n
T
h
e
p
a
p
er
p
r
esen
ts
a
d
etailed
an
al
y
s
is
o
f
o
n
e
o
f
t
h
e
p
ar
am
et
er
s
o
f
th
e
p
r
o
p
o
s
ed
f
r
am
e
w
o
r
k
:
Hu
m
a
n
E
m
o
tio
n
s
[
1
3
]
,
[
1
4
]
.
A
Facia
l
E
m
o
tio
n
R
ec
o
g
n
itio
n
S
y
s
te
m
i
s
p
r
o
p
o
s
ed
th
at
w
o
u
ld
r
ec
o
g
n
ize
7
e
m
o
tio
n
s
lab
eled
Neu
tr
al,
An
g
r
y
,
Dis
g
u
s
t,
Fear
,
Hap
p
y
,
Sad
an
d
Su
r
p
r
is
e
u
s
i
n
g
Ma
tlab
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
w
i
ll
ex
tr
ac
t
e
m
o
tio
n
s
o
n
t
h
e
b
asis
o
f
s
o
m
e
u
n
iq
u
e
f
ea
t
u
r
es.
Fi
g
u
r
e
4
p
r
esen
ts
a
n
o
v
el
al
g
o
r
ith
m
f
o
r
d
etec
tin
g
e
m
o
tio
n
s
f
r
o
m
a
n
i
m
a
g
e.
T
h
e
i
m
a
g
e
to
b
e
test
ed
ca
n
b
e
tak
en
f
r
o
m
a
s
m
ar
t
p
h
o
n
e
u
s
in
g
an
ap
p
I
P
W
eb
ca
m
w
h
ic
h
tu
r
n
s
t
h
e
an
d
r
o
id
p
h
o
n
e
in
to
w
ir
eless
ca
m
er
a
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
er
f
o
r
m
an
ce
o
f
p
r
o
p
o
s
ed
Facial
e
m
o
tio
n
r
ec
o
g
n
iti
o
n
s
y
s
te
m
is
e
v
alu
a
ted
an
d
T
a
b
le
1
s
h
o
w
ca
s
e
s
t
h
e
co
m
p
ar
is
o
n
b
et
w
ee
n
th
e
p
r
o
p
o
s
ed
m
eth
o
d
a
n
d
ex
is
ti
n
g
SVM
d
u
r
in
g
t
h
e
t
r
ain
in
g
an
d
te
s
ti
n
g
p
h
ase.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
i
s
co
m
p
ar
ed
w
it
h
e
x
is
tin
g
SV
M
(
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e)
t
ec
h
n
iq
u
e
i
n
ter
m
s
o
f
ac
cu
r
ac
y
,
tr
ai
n
i
n
g
an
d
te
s
ti
n
g
er
r
o
r
.
A
cc
u
r
ac
y
a
n
d
tr
ai
n
i
n
g
e
r
r
o
r
co
m
p
ar
is
o
n
b
et
w
ee
n
S
V
M
an
d
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
s
h
o
w
n
in
Fi
g
u
r
e
5
.
A
cc
u
r
ac
y
w
ith
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
i
n
cr
ea
s
ed
to
7
8
.
4
0
%
in
co
m
p
ar
is
o
n
w
i
th
ex
is
t
in
g
SVM
t
h
at
is
7
5
.
3
7
%.
T
r
ain
in
g
er
r
o
r
f
o
r
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
d
ec
r
ea
s
ed
to
0
.
0
0
4
1
0
3
w
h
ea
r
as
w
it
h
th
e
ex
i
s
ti
n
g
SVM
m
eth
o
d
tr
ai
n
in
g
er
r
o
r
is
0
.
0
0
8
9
3
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
24
33
–
2
4
4
1
2438
Fig
u
r
e
4
.
A
l
g
o
r
ith
m
f
o
r
d
etec
tin
g
e
m
o
tio
n
s
Fig
u
r
e
5
.
A
cc
u
r
ac
y
an
d
tr
ai
n
i
n
g
er
r
o
r
co
m
p
ar
is
o
n
b
et
w
ee
n
S
VM
an
d
th
e
P
r
o
p
o
s
ed
Me
th
o
d
Fig
u
r
e
6
s
h
o
w
s
t
h
e
i
m
a
g
e
w
h
i
ch
s
y
s
te
m
ac
ce
p
ts
as
i
n
p
u
t
a
n
d
Fig
u
r
e
7
d
ep
icts
th
e
r
es
u
lta
n
t
e
m
o
tio
n
(
n
eu
tr
al)
b
ein
g
d
etec
ted
.
T
w
o
d
if
f
er
e
n
t
r
esu
ltan
t
e
m
o
tio
n
s
(
s
u
r
p
r
is
e,
h
ap
p
y
)
ar
e
also
s
h
o
w
n
i
n
Fi
g
u
r
e
8
an
d
Fig
u
r
e
9
.
T
r
a
in
in
g
er
r
o
r
is
s
h
o
w
n
i
n
Fi
g
u
r
e
1
0
w
h
i
c
h
r
ed
u
ce
s
w
ith
ea
c
h
iter
atio
n
.
A
l
g
o
r
i
t
h
m
f
o
r
E
m
o
t
i
o
n
D
e
t
e
c
t
i
o
n
P
r
e
p
r
o
c
e
ss t
h
e
d
a
t
a
b
a
se
(
c
o
h
n
-
k
a
n
a
d
e
)
C
r
e
a
t
e
t
r
a
i
n
d
i
r
e
c
t
o
r
y
mk
d
i
r
(
'T
r
a
i
n
i
n
g
_
d
a
t
a
\
0
')
;
..
mk
d
i
r
(
'T
r
a
i
n
i
n
g
_
d
a
t
a
\
5
')
;
mk
d
i
r
(
'T
r
a
i
n
i
n
g
_
d
a
t
a
\
6
')
;
f
o
l
d
e
r
0
:
=
a
l
l
n
e
u
t
r
a
l
i
mag
e
s
f
o
l
d
e
r
1
:
=
a
l
l
a
n
g
r
y
i
mag
e
s
.
f
o
l
d
e
r
6
:
=
a
l
l
su
r
p
r
i
se
i
mag
e
s
T
e
st
d
i
r
e
c
t
o
r
y
<
=
(
2
5
-
3
0
%
i
mag
e
s
o
f
t
r
a
i
n
d
i
r
e
c
t
o
r
y
)
T
r
a
i
n
i
n
g
a
n
d
t
e
st
i
n
g
d
a
t
a
ma
t
r
i
x
f
o
r
c
o
n
c
e
r
n
e
d
d
a
t
a
b
a
se
i
s
p
r
e
p
a
r
e
d
u
si
n
g
l
o
c
a
l
b
i
n
a
r
y
p
a
t
t
e
r
n
.
f
o
r
i
=
1
:
n
o
_
f
i
l
e
s
f
i
l
e
n
a
me
=
f
i
l
e
s{
i
};
I
m =
i
mr
e
a
d
(
f
i
l
e
n
a
me
)
;
C
a
l
l
n
e
x
t
t
w
o
st
e
p
s
f
o
r
f
a
c
e
d
e
t
e
c
t
i
o
n
,
c
r
o
p
p
i
n
g
(
e
y
e
,
n
o
se
a
n
d
mo
u
t
h
)
a
n
d
l
o
c
a
l
b
i
n
a
r
y
h
i
st
o
g
r
a
m [
1
5
]
.
t
r
a
i
n
X
(
i
,
:
)
=
l
b
p
H
i
st
;
t
r
a
i
n
Y
(
i
,
s
t
r
2
d
o
u
b
l
e
(
f
i
l
e
n
a
me
(
1
9
)
)
+
1
)
=
1
;
P
r
o
c
e
e
d
t
o
T
ra
i
n
i
n
g
s
t
e
p
.
D
e
t
e
c
t
f
a
c
e
s
f
r
o
m
t
h
e
i
m
a
g
e
s
u
s
i
n
g
v
o
i
l
a
j
o
n
e
s
a
l
g
o
r
i
t
h
m
[
1
6
]
[
1
7
]
.
[
X
,
Y
,
h
e
i
g
h
t
,
w
i
d
t
h
]
=
v
i
o
l
a
Jo
n
e
s(
I
mag
e
)
;
T
h
e
t
a
r
g
e
t
f
a
c
e
w
i
l
l
b
e
d
i
v
i
d
e
d
i
n
t
o
t
h
r
e
e
r
e
g
i
o
n
s
(
e
y
e
,
n
o
se
a
n
d
mo
u
t
h
)
c
r
o
p
L
=
i
mcro
p
(
I
mag
e
,
[
X
,
Y
,
h
e
i
g
h
t
,
w
i
d
t
h
]
)
;
c
r
o
p
L
=
i
mr
e
si
z
e
(
c
r
o
p
L
,
[
4
0
,
4
0
]
)
;
L
1
=
c
r
o
p
L
(
1
0
:
2
0
,
1
0
:
3
0
)
;
/
/
e
y
e
L
2
=
c
r
o
p
L
(
2
0
:
3
0
,
1
0
:
3
0
)
;
/
/
n
o
se
L
3
=
c
r
o
p
L
(
3
0
:
4
0
,
1
0
:
3
0
)
;
/
/
mo
u
t
h
l
b
p
H
1
=
l
b
p
(
L
1
)
;
/
/
l
o
c
a
l
b
i
n
a
r
y
p
a
t
t
e
r
n
(
e
y
e
)
l
b
p
H
2
=
l
b
p
(
L
2
)
;
/
/
l
o
c
a
l
b
i
n
a
r
y
p
a
t
t
e
r
n
(
n
o
se
)
l
b
p
H
3
=
l
b
p
(
L
3
)
;
/
/
l
o
c
a
l
b
i
n
a
r
y
p
a
t
t
e
r
n
(
mo
u
t
h
)
l
b
p
H
i
s
t
=
[
l
b
p
H
1
,
l
b
p
H
2
,
l
b
p
H
3
]
;
/
/
L
B
P
/
/
A
h
i
st
o
g
r
a
m
o
f
t
h
e
l
a
b
e
l
e
d
i
mag
e
f
l
(
x
,
y
)
c
a
n
b
e
d
e
f
i
n
e
d
a
s
H
i
=
∑
x
,
y
I
{
f
l
(
x
,
y
)
=
i
}
,
i
=
0
,
.
.
.
,
n
–
1
i
n
w
h
i
c
h
n
i
s
t
h
e
n
u
mb
e
r
o
f
d
i
f
f
e
r
e
n
t
l
a
b
e
l
s
p
r
o
d
u
c
e
d
b
y
t
h
e
L
B
P
o
p
e
r
a
t
o
r
a
n
d
I
{
A
} =
1
,
A
i
s t
r
u
e
=
0
,
A
i
s fal
se
.
T
h
i
s
h
i
s
t
o
g
r
a
m
c
o
n
t
a
i
n
s
i
n
f
o
r
ma
t
i
o
n
a
b
o
u
t
t
h
e
d
i
s
t
r
i
b
u
t
i
o
n
o
f
t
h
e
l
o
c
a
l
mi
c
r
o
p
a
t
t
e
r
n
s,
s
u
c
h
a
s
e
d
g
e
s,
sp
o
t
s
a
n
d
f
l
a
t
a
r
e
a
s,
o
v
e
r
t
h
e
w
h
o
l
e
i
mag
e
.
(
T
ra
i
n
i
n
g
s
t
e
p
)
T
r
a
i
n
mat
r
i
c
e
s
a
r
e
t
h
e
n
u
se
d
t
o
t
r
a
i
n
t
h
e
c
l
a
ss
i
f
i
e
r
(
n
e
u
r
a
l
n
e
t
w
o
r
k
)
.
N
e
t
w
o
r
k
i
s
t
r
a
i
n
e
d
b
y
a
p
p
l
y
i
n
g
t
r
a
i
n
i
n
g
d
a
t
a
a
n
d
t
r
a
i
n
i
n
g
p
a
r
a
me
t
e
r
s
(
l
e
a
r
n
i
n
g
r
a
t
e
p
a
r
a
me
t
e
r
(
e
t
a
=
0
.
0
1
)
a
n
d
n
u
m
b
e
r
o
f
e
p
o
c
h
s
(
n
E
p
o
c
h
s
=
5
0
0
)
,
h
i
d
d
e
n
l
a
y
e
r
s
(
H
=
1
6
)
)
a
s
i
n
p
u
t
w
h
i
c
h
r
e
su
l
t
s
i
n
u
p
d
a
t
e
d
w
e
i
g
h
t
s
o
f
t
h
e
i
n
p
u
t
a
n
d
h
i
d
d
e
n
l
a
y
e
r
w
i
t
h
mi
n
i
m
u
m e
r
r
o
r
.
T
o
c
o
mp
l
e
t
e
t
h
e
e
mo
t
i
o
n
d
e
t
e
c
t
i
o
n
,
p
r
a
c
t
i
c
e
t
h
e
p
r
o
p
o
se
d
sy
st
e
m
t
a
k
e
s
a
s
i
n
p
u
t
u
p
d
a
t
e
d
w
e
i
g
h
t
s
w
a
n
d
v
(
b
a
c
k
w
a
r
d
p
a
ss)
f
r
o
m
p
r
e
v
i
o
u
s
st
e
p
a
n
d
t
e
st
i
n
g
p
o
i
n
t
s
X
(
L
B
P
f
e
a
t
u
r
e
s
o
f
i
mag
e
t
o
b
e
t
e
st
e
d
)
a
n
d
c
o
n
c
l
u
d
e
s
w
i
t
h
t
h
e
f
a
m
i
l
y
o
f
t
h
e
d
a
t
a
a
s a
n
t
i
c
i
p
a
t
e
d
b
y
t
h
e
n
e
t
w
o
r
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
I
n
fo
r
ma
tio
n
R
etri
ev
a
l fro
m
E
mo
tio
n
s
a
n
d
E
ye
B
lin
ks w
ith
h
elp
o
f S
en
s
o
r
…
(
P
u
n
ee
t S
in
g
h
La
mb
a
)
2439
T
ab
le
1
.
C
o
m
p
ar
is
o
n
b
et
w
ee
n
P
r
o
p
o
s
ed
Me
th
o
d
an
d
SVM
Fig
u
r
e
6
.
I
n
p
u
t i
m
a
g
e
Fig
u
r
e
7
.
E
m
o
tio
n
1
Fig
u
r
e
8
.
E
m
o
tio
n
2
Fig
u
r
e
9
.
E
m
o
tio
n
3
Fig
u
r
e
1
0
.
T
r
ain
in
g
er
r
o
r
S
V
M
(
T
r
a
i
n
i
n
g
)
P
r
o
p
o
se
d
M
e
t
h
o
d
(
T
r
a
i
n
i
n
g
)
S
V
M
(
Te
st
i
n
g
)
P
r
o
p
o
se
d
M
e
t
h
o
d
(
Te
st
i
n
g
)
T
o
t
a
l
N
u
mb
e
r
o
f
I
n
st
a
n
c
e
s
3
7
0
3
7
0
2
6
4
2
6
4
C
o
r
r
e
c
t
l
y
C
l
a
ssi
f
i
e
d
I
n
st
a
n
c
e
s
3
7
0
3
6
9
1
9
9
2
0
7
I
n
c
o
r
r
e
c
t
l
y
C
l
a
ssi
f
i
e
d
I
n
st
a
n
c
e
s
0
1
65
57
M
e
a
n
a
b
so
l
u
t
e
e
r
r
o
r
0
.
2
0
4
1
0
.
0
0
2
5
0
.
2
0
9
8
0
.
0
7
2
2
K
a
p
p
a
s
t
a
t
i
s
t
i
c
1
0
.
9
9
6
2
0
.
6
3
2
1
0
.
6
7
9
1
R
o
o
t
me
a
n
sq
u
a
r
e
d
e
r
r
o
r
0
.
3
0
0
6
0
.
0
2
2
9
0
.
3
0
9
9
0
.
2
2
8
1
R
e
l
a
t
i
v
e
a
b
so
l
u
t
e
e
r
r
o
r
1
0
0
.
4
0
4
8
%
1
.
2
5
2
9
%
1
0
6
.
4
6
3
7
%
3
6
.
6
3
1
9
%
R
o
o
t
r
e
l
a
t
i
v
e
sq
u
a
r
e
d
e
r
r
o
r
9
4
.
4
7
4
3
%
7
.
1
9
4
6
%
1
0
0
.
5
2
0
6
%
7
3
.
9
9
4
%
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
:
24
33
–
2
4
4
1
2440
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
u
n
d
er
lin
es
m
ea
s
u
r
ab
le
p
ar
a
m
eter
s
t
h
at
ca
n
b
e
p
er
ce
iv
ed
u
s
i
n
g
s
e
n
s
o
r
n
et
w
o
r
k
s
in
ter
m
s
o
f
in
tell
ig
e
n
t
in
f
o
r
m
a
tio
n
m
es
s
ag
e
s
.
Su
c
h
p
ar
a
m
eter
s
in
c
lu
d
e
ey
e
b
lin
k
p
atter
n
,
e
m
o
tio
n
s
an
d
h
ea
r
t
r
ate,
o
f
w
h
ic
h
b
lin
k
p
atter
n
is
g
i
v
en
t
h
e
u
t
m
o
s
t
i
m
p
o
r
tan
ce
.
R
e
s
u
ltan
t
m
es
s
a
g
es
ar
e
t
h
en
p
er
co
lated
an
d
co
n
d
itio
n
al
l
y
tag
g
ed
to
ex
ce
r
p
t
in
telli
g
e
n
t
in
f
o
r
m
atio
n
,
th
u
s
g
en
er
a
ti
n
g
ex
ig
e
n
c
y
aler
t.
A
l
s
o
it
h
ig
h
li
g
h
ts
b
o
th
th
e
s
u
f
f
icie
n
t
an
d
n
ec
es
s
ar
y
co
n
d
itio
n
s
r
eq
u
ir
ed
f
o
r
alar
m
to
b
e
r
aised
.
P
a
p
er
p
r
o
p
o
s
es
an
in
telli
g
en
t
f
r
a
m
e
w
o
r
k
alo
n
g
w
it
h
P
er
s
p
icac
io
u
s
I
n
f
o
r
m
at
io
n
R
et
r
iev
al
T
ec
h
n
iq
u
e
an
d
th
e
p
r
o
g
n
o
s
tic
alg
o
r
it
h
m
f
o
r
d
etec
tin
g
th
e
ex
ig
e
n
c
y
.
T
h
e
p
r
o
p
o
s
ed
Facial
E
m
o
tio
n
R
ec
o
g
n
itio
n
S
y
s
te
m
r
ec
o
g
n
ize
s
7
em
o
tio
n
s
lab
eled
Neu
tr
al
,
An
g
r
y
,
Dis
g
u
s
t,
Fear
,
Hap
p
y
,
Sad
an
d
S
u
r
p
r
is
e.
T
h
e
alg
o
r
ith
m
is
v
alid
ated
th
r
o
u
g
h
M
A
T
L
A
B
an
d
r
esu
l
ts
d
ep
ict
th
e
ca
teg
o
r
izatio
n
o
f
f
ac
es a
s
n
e
u
tr
al
,
s
u
r
p
r
i
s
e
an
d
h
ap
p
y
d
ep
en
d
in
g
o
n
t
h
e
ex
p
r
ess
io
n
.
RE
F
E
R
E
NC
E
S
[1
]
M
a
g
n
ier,
M
a
rk
;
S
h
a
rm
a
,
S
u
b
h
a
sh
(2
7
No
v
e
m
b
e
r
2
0
0
8
)
,
"
In
d
ia
terr
o
rist
a
tt
a
c
k
s
lea
v
e
a
t
lea
st
1
0
1
d
e
a
d
in
M
u
m
b
a
i,
"
L
o
s A
n
g
e
les
T
i
m
e
s
.
p
.
A1
.
Re
tri
e
v
e
d
2
8
N
o
v
e
m
b
e
r
2
0
0
8
.
[2
]
F
ried
m
a
n
,
T
h
o
m
a
s
(1
7
F
e
b
ru
a
ry
2
0
0
9
)
,
"
No
W
a
y
,
No
Ho
w
,
No
t
He
re
,
"
T
h
e
Ne
w
Yo
rk
T
i
m
e
s
.
A
r
c
h
iv
e
d
f
ro
m
th
e
o
rig
in
a
l
o
n
2
9
A
u
g
u
st 2
0
1
2
.
Re
tri
e
v
e
d
1
7
M
a
y
2
0
1
0
.
[3
]
M
r.
Ra
e
e
s
A
h
m
a
d
,
P
r
o
f
.
J.N.
B
o
ro
le
,
“
Dro
w
s
y
Driv
e
r
Id
e
n
ti
f
ic
a
ti
o
n
Us
in
g
Ey
e
Bli
n
k
d
e
tec
ti
o
n
”
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
In
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
ies
,
2
0
1
5
,
v
o
l.
6
,
n
o
.
1
,
p
p
.
2
7
0
-
2
7
4
.
[4
]
Ku
srin
i
K
u
sri
n
i,
M
.
De
d
i
Isk
a
n
d
a
r,
F
e
rry
W
a
h
y
u
W
ib
o
w
o
,
“
M
u
lt
i
F
e
a
tu
re
s
Co
n
ten
t
-
Ba
se
d
Im
a
g
e
Re
tri
e
v
a
l
Us
in
g
Clu
ste
rin
g
A
n
d
De
c
isio
n
T
r
e
e
A
l
g
o
rit
h
m
”
,
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
ti
o
n
,
Co
mp
u
ti
n
g
,
E
lec
tro
n
ics
a
n
d
Co
n
tro
l
),
2
0
1
6
,
v
o
l.
14
,
n
o
.
2
,
p
p
.
1
4
8
0
-
1
4
9
2
.
[5
]
Jin
h
u
a
Z
h
a
n
g
,
Da
n
iel
S
c
h
o
lt
e
n
,
“
A
F
a
c
e
Re
c
o
g
n
it
io
n
A
lg
o
rit
h
m
Ba
se
d
o
n
Im
p
ro
v
e
d
Co
n
to
u
rlet
T
ra
n
s
f
o
r
m
a
nd
P
ri
n
c
ip
le
Co
m
p
o
n
e
n
t
A
n
a
l
y
sis
”
,
T
EL
KOM
NIKA
(
T
e
le
c
o
mm
u
n
ica
t
io
n
,
C
o
mp
u
t
in
g
,
El
e
c
tro
n
ics
a
n
d
Co
n
tro
l
),
2
0
1
6
,
v
o
l.
14
,
n
o
.
2A
,
p
p
.
1
1
4
-
1
1
9
.
[6
]
F
e
n
g
y
i
S
o
n
g
,
X
iao
y
a
n
g
Tan
,
X
u
e
L
iu
,
S
o
n
g
c
a
n
Ch
e
n
,
“
Ey
e
s
c
lo
se
n
e
ss
d
e
tec
ti
o
n
f
ro
m
stil
l
ima
g
e
s
w
it
h
m
u
lt
i
-
sc
a
le
h
isto
g
ra
m
s o
f
p
rin
c
ip
a
l
o
rien
ted
g
ra
d
ien
ts
”
,
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
El
se
v
ier
,
2014
,
v
o
l.
47
,
n
o
.
9
,
p
p
.
2
8
2
5
-
2
8
3
8
.
[7
]
T
e
re
z
a
S
o
u
k
u
p
o
v
a
a
n
d
Ja
n
Ce
c
h
,
“
Re
a
l
-
T
i
m
e
E
y
e
Bli
n
k
De
tec
ti
o
n
u
si
n
g
F
a
c
ial
L
a
n
d
m
a
r
k
s
”
,
Pro
c
e
e
d
in
g
s
o
f
t
h
e
2
1
st C
o
mp
u
ter
Vi
sio
n
W
in
ter
W
o
rk
sh
o
p
Ri
ms
k
e
T
o
p
li
c
e
,
S
lo
v
e
n
i
a
,
2
0
1
6
.
[8
]
Y.
S
u
n
a
n
d
A
.
N.
A
k
a
n
su
,
“
F
a
c
ia
l
e
x
p
re
ss
io
n
re
c
o
g
n
it
io
n
w
it
h
re
g
io
n
a
l
h
id
d
e
n
M
a
rk
o
v
m
o
d
e
ls
”
,
EL
ECT
RONIC
S
L
ET
T
ER
S
,
2
0
1
4
,
v
o
l.
50
,
n
o
.
9
,
p
p
.
6
7
1
-
6
7
3
.
[9
]
Yin
g
-
li
T
ian
,
T
a
k
e
o
Ka
n
a
d
e
,
a
n
d
Je
ff
re
y
F
.
Co
h
n
,
“
Re
c
o
g
n
izi
n
g
A
c
ti
o
n
Un
it
s
f
o
r
F
a
c
ial
Ex
p
re
s
sio
n
A
n
a
ly
sis
”
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Pa
t
ter
n
An
a
lys
is
a
n
d
M
a
c
h
in
e
In
telli
g
e
n
c
e
,
2
0
0
1
,
v
o
l.
23
,
n
o
.
2
,
p
p
.
97
-
1
1
5
.
[1
0
]
J.
Ce
c
h
,
V
.
F
ra
n
c
,
J.
M
a
tas
,
“
A
3
D
a
p
p
r
o
a
c
h
t
o
f
a
c
ial
lan
d
m
a
rk
s
:
De
tec
ti
o
n
,
re
f
in
e
m
e
n
t,
a
n
d
trac
k
in
g
”
,
In
Pro
c
.
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
P
a
tt
e
rn
Rec
o
g
n
it
io
n
,
2
0
1
4
,
p
p
.
2
1
7
3
-
2
1
7
8
.
[1
1
]
Re
h
a
b
F
.
A
b
d
e
l
-
Ka
d
e
rn
,
Ra
n
d
a
A
tt
a
,
S
h
e
re
n
El
-
S
h
a
k
h
a
b
e
,
“
A
n
e
ff
icie
n
t
e
y
e
d
e
te
c
ti
o
n
a
n
d
trac
k
in
g
s
y
ste
m
b
a
se
d
o
n
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iz
a
ti
o
n
a
n
d
a
d
a
p
ti
v
e
b
l
o
c
k
-
m
a
tch
in
g
se
a
rc
h
a
lg
o
rit
h
m
”
,
En
g
in
e
e
rin
g
Ap
p
li
c
a
ti
o
n
s
o
f
Arti
fi
c
ia
l
I
n
telli
g
e
n
c
e
,
2
0
1
4
,
v
o
l.
31
,
p
p
.
90
-
1
0
0
.
[1
2
]
H.
Ch
u
d
u
c
,
K.
Ng
u
y
e
n
p
h
a
n
,
a
n
d
D.
Ng
u
y
e
n
,
“
A
Re
v
ie
w
o
f
He
a
rt
Ra
te
V
a
riab
il
i
ty
a
n
d
it
s
A
p
p
li
c
a
ti
o
n
s
”
,
A
PCB
EE
Pro
c
e
d
ia
E
L
S
E
VIE
R
,
2
0
1
3
,
v
o
l.
7
,
p
p
.
80
-
85.
[1
3
]
Ch
in
g
Ye
e
Yo
n
g
,
R
u
b
i
ta
S
u
d
irm
a
n
,
Kim
M
e
y
Ch
e
w
,
“
Co
lo
u
r
P
e
r
c
e
p
ti
o
n
o
n
F
a
c
ial
Ex
p
re
ss
io
n
to
w
a
rd
s
Em
o
ti
o
n
”
,
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
t
io
n
,
Co
mp
u
ti
n
g
,
El
e
c
tro
n
ics
a
n
d
Co
n
tro
l
)
,
2
0
1
2
,
v
o
l.
10
,
n
o
.
4
,
p
p
.
7
2
3
-
7
3
3
.
[1
4
]
Yu
tai
W
a
n
g
,
X
in
g
h
a
i
Ya
n
g
,
Jin
g
Zo
u
,
“
Re
se
a
rc
h
o
f
E
m
o
ti
o
n
Re
c
o
g
n
it
io
n
Ba
se
d
o
n
S
p
e
e
c
h
a
n
d
F
a
c
ial
Ex
p
re
ss
io
n
.
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
t
io
n
,
Co
mp
u
ti
n
g
,
El
e
c
tro
n
ics
a
n
d
Co
n
tro
l
)
,
2
0
1
3
,
v
o
l.
11
,
n
o
.
1
,
p
p
.
83
-
9
0
.
[1
5
]
T
i
m
o
A
h
o
n
e
n
,
A
b
d
e
n
o
u
r
Ha
d
id
,
a
n
d
M
a
tt
i
P
ietik
a
in
e
n
,
“
F
a
c
e
R
e
c
o
g
n
i
ti
o
n
w
it
h
L
o
c
a
l
Bin
a
ry
P
a
tt
e
rn
s
”
,
ECCV
2
0
0
4
,
L
NCS
3
0
2
1
,
2
0
0
4
,
p
p
.
4
6
9
-
4
8
1
.
[1
6
]
P
a
u
l
V
io
la
a
n
d
M
ich
a
e
l
J.
Jo
n
e
s
,
“
Ro
b
u
st
Re
a
l
-
T
i
m
e
F
a
c
e
De
tec
t
io
n
”
,
I
n
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
mp
u
ter
Vi
si
o
n
,
2
0
0
4
,
v
o
l.
57
,
n
o
.
2
,
p
p
.
1
3
7
-
1
5
4
.
[1
7
]
Yi
-
Qin
g
W
a
n
g
,
“
A
n
A
n
a
l
y
sis
o
f
th
e
V
i
o
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
in
e
,
2
0
1
4
,
v
o
l.
4
,
p
p
.
1
2
8
-
1
4
8
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Pu
n
e
e
t
S
in
g
h
L
a
m
b
a
re
c
e
iv
e
d
th
e
M
a
ste
r
d
e
g
re
e
(M
.
tec
h
)
in
I
n
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
f
ro
m
Un
iv
e
rsit
y
S
c
h
o
o
l
o
f
In
f
o
r
m
a
ti
o
n
A
n
d
Co
m
m
u
n
ica
ti
o
n
T
e
c
h
n
o
lo
g
y
,
G
u
ru
Go
b
i
n
d
S
i
n
g
h
In
d
ra
p
ra
st
h
a
Un
iv
e
rs
it
y
in
2
0
1
3
.
He
is
a
P
h
D
c
a
n
d
i
d
a
te
in
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
a
t
USICT
,
GG
S
IP
U.
His
re
se
a
r
c
h
in
tere
sts
a
re
in
t
h
e
a
re
a
s
o
f
S
e
n
so
r
Ne
tw
o
rk
s
a
n
d
A
d
h
o
c
Ne
tw
o
rk
s.
He
h
a
s
m
o
re
th
a
n
8
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
ri
e
n
c
e
a
n
d
is
c
u
rre
n
t
ly
w
o
rk
in
g
a
s
As
si
sta
n
t
P
ro
f
e
ss
o
r
in
Bh
a
ra
ti
V
i
d
y
a
p
e
e
th
’s
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
Ne
w
De
lh
i,
In
d
ia
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
I
n
fo
r
ma
tio
n
R
etri
ev
a
l fro
m
E
mo
tio
n
s
a
n
d
E
ye
B
lin
ks w
ith
h
elp
o
f S
en
s
o
r
…
(
P
u
n
ee
t S
in
g
h
La
mb
a
)
2441
Dr
.
De
e
p
a
li
V
i
r
m
a
n
i
re
c
e
iv
e
d
th
e
d
o
c
to
ra
l
d
e
g
re
e
in
C
o
m
p
u
ter
S
c
ien
c
e
f
ro
m
Ne
taji
S
u
b
h
a
s
I
n
stit
u
t
e
o
f
T
e
c
h
n
o
lo
g
y
,
Ne
w
De
lh
i
in
2
0
1
2
.
He
r
re
se
a
rc
h
in
tere
sts
a
re
in
th
e
a
re
a
s
o
f
S
e
n
so
r
Ne
tw
o
rk
s,
Da
ta
M
in
i
n
g
a
n
d
S
e
c
u
ri
ty
.
S
h
e
h
a
s
p
u
b
li
s
h
e
d
m
o
re
th
a
n
6
0
p
a
p
e
rs
i
n
re
f
e
rre
d
In
tern
a
ti
o
n
a
l
jo
u
rn
a
ls
a
n
d
c
o
n
f
e
re
n
c
e
s.
S
h
e
is
b
ra
n
c
h
c
o
u
n
s
e
lo
r
IEE
E
stu
d
e
n
t
c
h
a
p
ter an
d
CS
I
stu
d
e
n
t
b
ra
n
c
h
a
n
d
h
a
s
m
o
re
th
a
n
1
5
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
S
h
e
is
c
u
rre
n
tl
y
w
o
rk
in
g
a
s
As
s
o
c
iate
P
ro
f
e
ss
o
r
(He
a
d
o
f
Co
m
p
u
ter
S
c
ien
c
e
De
p
a
rtme
n
t)
in
B
h
a
g
wa
n
P
a
rsh
u
ra
m
In
stit
u
te
o
f
T
e
c
h
n
o
l
o
g
y
a
p
p
ro
v
e
d
b
y
G
u
ru
G
o
b
in
d
S
in
g
h
In
d
ra
p
ra
sth
a
U
n
iv
e
rsity
,
Ne
w
D
e
lh
i,
I
n
d
ia.
Evaluation Warning : The document was created with Spire.PDF for Python.