I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
24
,
No
.
1
,
Octo
b
er
20
21
,
p
p
.
1
8
9
~
19
7
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
2
4
.i
1
.
pp
1
89
-
19
7
189
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
Face spo
o
fing de
t
ection using
sur
fa
ce and sub
-
surfa
c
e
refle
c
tions
ana
ly
sis
Azim
Z
a
lih
a
Abd Az
iz
1
,
M
o
hd
Rizo
n M
o
ha
m
ed
J
uh
a
ri
2
1
F
a
c
u
lt
y
o
f
I
n
fo
rm
a
ti
c
s a
n
d
Co
m
p
u
ti
n
g
,
Un
i
v
e
rsiti
S
u
lt
a
n
Zain
a
l
A
b
id
i
n
(Un
iS
ZA),
Tere
n
g
g
a
n
u
,
M
a
lay
sia
2
F
a
c
u
lt
y
o
f
E
n
g
i
n
e
e
rin
g
,
Tec
h
n
o
l
o
g
y
a
n
d
Bu
il
t
E
n
v
ir
o
n
m
e
n
t
,
UCS
I
Un
iv
e
rsit
y
,
Ku
a
la L
u
m
p
u
r
,
M
a
l
a
y
sia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
2
0
,
2
0
2
0
R
ev
is
ed
Au
g
9
,
2
0
2
1
Acc
ep
ted
Au
g
2
0
,
2
0
2
1
Re
flec
ti
o
n
b
a
se
d
a
n
a
l
y
sis
h
a
s
b
e
e
n
u
se
d
in
p
re
v
io
u
s
re
se
a
rc
h
fo
r
v
a
rio
u
s
o
b
jec
ti
v
e
s.
M
a
teria
ls
c
las
sifica
ti
o
n
is
o
n
e
o
f
t
h
e
m
.
Ba
sic
a
ll
y
,
e
a
c
h
m
a
teria
l
c
o
n
sists
o
f
two
ty
p
e
s
o
f
re
flec
ti
o
n
s:
su
rfa
c
e
a
n
d
s
u
b
-
su
rfa
c
e
.
T
o
se
p
a
ra
te
th
e
se
two
re
flec
ti
o
n
s,
p
o
lariz
e
d
li
g
h
t
c
o
u
ld
b
e
a
p
p
l
ied
.
P
re
v
io
u
s
ly
,
m
u
l
ti
-
re
flec
ti
o
n
s
c
h
a
ra
c
teristics
we
re
a
n
a
ly
z
e
d
u
sin
g
p
o
lariz
e
d
li
g
h
t
t
o
c
las
sify
o
b
jec
ts
su
c
h
a
s
b
e
twe
e
n
m
e
tals
a
n
d
n
o
n
-
m
e
tals.
Ho
we
v
e
r,
n
o
tri
a
l
h
a
s
b
e
e
n
d
o
n
e
u
sin
g
th
e
sa
m
e
m
e
th
o
d
t
o
d
isti
n
g
u
ish
re
a
l
a
n
d
fa
k
e
fa
c
e
s
th
a
t
c
o
u
l
d
b
e
u
se
d
to
c
o
m
b
a
t
sp
o
o
fin
g
a
tt
e
m
p
ts
in
fa
c
e
b
io
m
e
tri
c
sy
ste
m
.
S
in
c
e
h
u
m
a
n
sk
in
is
m
u
lt
i
lay
e
rs
str
u
c
tu
re
,
it
a
l
so
p
ro
d
u
c
e
s
m
u
l
ti
re
flec
ti
o
n
s.
In
t
h
is
p
a
p
e
r,
d
riv
e
n
b
y
th
e
t
h
e
o
ry
,
s
u
rfa
c
e
a
n
d
su
b
-
su
rfa
c
e
re
flec
ti
o
n
s
o
f
b
o
th
g
e
n
u
in
e
h
u
m
a
n
fa
c
e
a
n
d
p
a
p
e
r
fa
c
e
m
a
s
k
we
re
sta
ti
stica
ll
y
e
x
a
m
in
e
d
.
I
n
a
d
d
i
ti
o
n
,
iP
a
d
d
isp
lay
e
d
fa
c
e
ima
g
e
s
we
re
a
lso
u
se
d
a
s
sp
o
o
fi
n
g
a
tt
e
m
p
ts.
Im
a
g
e
s
o
f
g
e
n
u
i
n
e
a
n
d
sp
o
o
fi
n
g
fa
c
e
s
we
re
c
a
p
tu
re
d
u
si
n
g
p
o
lariz
e
d
li
g
h
t
u
n
d
e
r
two
d
iffere
n
t
p
o
lariz
a
ti
o
n
a
n
g
les
:
0
a
n
d
9
0
d
e
g
re
e
s.
Eac
h
a
n
g
le
c
a
p
tu
r
e
d
ima
g
e
s
with
su
rfa
c
e
a
n
d
s
u
b
-
s
u
rfa
c
e
re
flec
ti
o
n
s,
a
c
c
o
rd
i
n
g
l
y
.
Th
o
se
re
flec
ti
o
n
s
we
re
a
n
a
ly
z
e
d
b
a
se
d
o
n
t
h
e
m
e
a
n
,
s
tan
d
a
rd
d
e
v
iati
o
n
,
sk
e
wn
e
ss
a
n
d
k
u
rt
o
sis.
M
o
d
a
li
t
y
d
istri
b
u
ti
o
n
o
f
e
a
c
h
im
a
g
e
wa
s
a
lso
stu
d
ied
u
sin
g
a
n
o
t
h
e
r
m
e
th
o
d
c
a
ll
e
d
th
e
b
imo
d
a
li
t
y
c
o
e
fficie
n
t
(BC).
F
ro
m
t
h
e
re
su
lt
s,
it
is
n
o
t
p
o
ss
ib
le
t
o
d
isti
n
g
u
is
h
b
e
twe
e
n
g
e
n
u
i
n
e
fa
c
e
a
n
d
p
ri
n
ted
p
h
o
to
s
b
e
c
a
u
se
o
f
th
e
m
u
lt
i
re
flec
ti
o
n
s’
sim
il
a
rit
ies
.
Ho
we
v
e
r,
iP
a
d
d
isp
lay
e
d
fa
c
e
ima
g
e
s
h
a
v
e
b
e
e
n
su
c
c
e
ss
fu
ll
y
id
e
n
ti
fied
a
s s
p
o
o
fin
g
tri
a
ls.
K
ey
w
o
r
d
s
:
Face
b
io
m
etr
ics
Face
s
p
o
o
f
in
g
d
etec
tio
n
Po
lar
ized
im
ag
in
g
s
y
s
tem
Su
b
-
s
u
r
f
ac
e
r
e
f
lectio
n
Su
r
f
ac
e
r
ef
lectio
n
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
:
Azim
Z
alih
a
Ab
d
Aziz
Facu
lty
o
f
I
n
f
o
r
m
atics
an
d
C
o
m
p
u
tin
g
Un
iv
er
s
iti Su
ltan
Z
ain
al
Ab
id
in
(
Un
iSZA)
B
esu
t
C
am
p
u
s
,
2
2
2
0
0
,
T
er
en
g
g
an
u
,
Ma
lay
s
ia
E
m
ail: a
zim
za
lih
a@
u
n
is
za
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
B
io
m
etr
ic
is
an
au
th
en
ticatio
n
tech
n
iq
u
e
th
at
r
el
y
o
n
m
ea
s
u
r
ea
b
le
p
h
y
s
ical
ch
ar
ac
ter
is
tics
th
at
ca
n
b
e
au
to
m
atica
lly
d
etec
te
d
.
T
h
er
e
ar
e
s
ev
er
al
ty
p
es
o
f
b
i
o
m
etr
ic
id
en
tific
atio
n
s
ch
em
es
s
u
ch
as
in
d
iv
id
u
al’
s
u
n
iq
u
e
f
in
g
er
p
r
i
n
ts
,
ir
is
an
al
y
s
is
,
th
e
an
aly
s
is
o
f
f
ac
ial
c
h
ar
ac
ter
is
tics
an
d
also
th
e
an
aly
s
is
o
f
a
p
er
s
o
n
’
s
v
o
ice.
T
h
e
s
elec
tio
n
o
f
an
y
b
io
m
etr
ic
tr
ait
d
ep
en
d
s
o
n
th
e
p
u
r
p
o
s
e
o
f
th
e
ap
p
lic
atio
n
.
As
i
n
f
ac
e
au
th
en
ticatio
n
s
y
s
tem
,
th
e
m
ain
p
u
r
p
o
s
e
o
f
th
e
d
e
v
elo
p
m
e
n
t
is
to
id
en
tify
v
alid
u
s
er
s
f
o
r
s
ec
u
r
ity
r
ea
s
o
n
s
.
I
m
p
lem
en
tatio
n
o
f
f
ac
e
b
i
o
m
etr
ic
s
y
s
tem
ca
n
b
e
wid
el
y
s
e
en
in
air
p
o
r
ts
all
ar
o
u
n
d
th
e
wo
r
ld
.
Ho
wev
e
r
,
th
e
s
y
s
tem
co
u
ld
b
e
ea
s
ily
s
p
o
o
f
ed
b
y
a
n
im
p
o
s
to
r
with
n
o
d
i
f
f
icu
lty
.
T
h
e
im
p
o
s
to
r
ca
n
o
b
tain
a
p
h
o
to
o
f
an
au
th
o
r
ized
p
e
r
s
o
n
,
p
lay
s
a
v
id
eo
,
o
r
d
is
p
lay
a
3
D
m
o
d
el
s
u
c
h
as
a
f
ac
e
m
ask
wh
ich
m
im
ic
s
a
v
alid
in
d
i
v
id
u
al,
in
f
r
o
n
t
o
f
th
e
s
en
s
o
r
to
g
ain
ac
ce
s
s
[
1
]
.
T
h
is
v
u
ln
e
r
ab
ilit
y
o
f
f
ac
e
-
b
ased
s
e
cu
r
ity
s
y
s
t
em
h
as
en
c
o
u
r
ag
e
d
v
ar
io
u
s
s
tu
d
ies an
d
p
r
ev
e
n
tiv
e
m
ea
s
u
r
es a
m
o
n
g
b
io
m
etr
ic
c
o
m
m
u
n
ity
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
2
4
,
No
.
1
,
Octo
b
er
20
21
:
1
8
9
-
19
7
190
Sp
o
o
f
in
g
f
ac
e
co
u
n
ter
m
ea
s
u
r
e
th
at
r
elies
o
n
liv
en
ess
a
n
d
m
o
tio
n
-
b
ased
a
n
aly
s
is
is
n
o
lo
n
g
er
r
elev
an
t.
l
iv
e
n
ess
-
b
ased
an
ti
-
s
p
o
o
f
in
g
d
etec
tio
n
wh
ich
d
et
ec
t
ey
es
a
n
d
lip
s
m
o
v
e
m
en
ts
h
av
e
b
ee
n
tr
ick
ed
with
p
er
f
o
r
ated
m
ask
in
th
e
ey
es
an
d
m
o
u
th
.
W
h
ile
f
o
r
m
o
tio
n
-
b
ased
co
u
n
ter
m
ea
s
u
r
e,
it
ca
n
b
e
ea
s
ily
ch
allen
g
ed
b
y
p
lacin
g
a
r
ec
o
r
d
ed
v
id
eo
in
f
r
o
n
t
o
f
th
e
ca
m
er
a.
An
o
th
er
f
ac
e
a
n
ti
-
s
p
o
o
f
i
n
g
m
eth
o
d
wh
ich
is
b
ased
o
n
r
e
f
le
ctan
ce
p
r
o
p
er
tie
s
h
as
d
r
iv
e
n
m
o
r
e
s
tu
d
ies
to
w
ar
d
s
th
is
an
al
y
s
is
.
Sin
ce
f
ac
e
m
ask
s
ar
e
m
a
d
e
o
f
v
ar
io
u
s
ty
p
es
o
f
m
ater
ials
,
c
h
ar
ac
ter
is
tics
o
f
ea
ch
m
ater
ial
wh
ich
p
r
o
v
id
e
v
ital
in
f
o
r
m
ati
o
n
ca
n
b
e
u
s
ed
as
class
if
icatio
n
p
ar
am
eter
s
.
As
in
[
2
]
,
a
r
ef
lecta
n
ce
-
ba
s
ed
a
n
aly
s
is
o
n
tex
tu
r
e
im
a
g
es
f
r
o
m
a
2
D+
3
D
f
ac
e
m
ask
was
in
tr
o
d
u
ce
d
.
I
m
a
g
es
o
f
3
D
f
ac
e
m
ask
wer
e
d
ec
o
m
p
o
s
ed
in
to
illu
m
in
atio
n
an
d
r
ef
le
ctan
ce
co
m
p
o
n
en
ts
.
T
h
e
r
esu
lts
r
ep
o
r
te
d
th
at
3
D
f
a
ce
m
ask
r
ef
lecta
n
ce
c
o
m
p
o
n
e
n
t is h
ig
h
er
th
a
n
th
e
r
ea
l
f
ac
e.
A
n
o
t
h
e
r
s
t
u
d
y
p
r
es
e
n
ts
m
u
l
ti
s
p
e
c
t
r
a
l
r
e
f
l
e
c
t
a
n
ce
u
s
i
n
g
m
u
l
t
is
p
e
c
t
r
a
l
li
g
h
t
i
n
g
t
o
d
i
s
c
r
i
m
i
n
a
t
e
b
e
t
w
e
e
n
a
g
e
n
u
i
n
e
f
a
c
e
,
a
p
h
o
t
o
f
a
c
e
a
n
d
v
a
r
i
e
t
y
o
f
f
a
c
e
m
a
s
k
s
m
a
d
e
o
f
p
l
a
s
t
ic
,
s
il
i
c
a
g
el
,
p
a
p
e
r
,
p
l
a
s
t
e
r
a
n
d
s
p
o
n
g
e
[
3
]
.
D
i
f
f
e
r
e
n
t
w
a
v
e
l
e
n
g
t
h
s
w
e
r
e
t
est
e
d
o
n
e
a
c
h
m
at
e
r
i
al
a
n
d
f
r
o
m
th
e
a
n
a
l
y
s
i
s
,
b
o
t
h
w
a
v
e
le
n
g
t
h
s
o
f
8
5
0
n
m
a
n
d
1
4
5
0
n
m
p
r
o
d
u
c
e
d
t
h
e
m
o
s
t
d
i
s
c
r
i
m
in
a
t
i
v
e
v
a
l
u
es
.
I
n
a
n
o
t
h
e
r
s
t
u
d
y
[
4
]
,
t
w
o
d
i
s
c
r
i
m
i
n
a
t
i
v
e
w
a
v
e
l
en
g
t
h
s
(
6
8
5
a
n
d
8
5
0
n
m
)
w
e
r
e
s
e
l
e
c
te
d
a
f
t
e
r
e
x
a
m
in
i
n
g
t
h
e
a
l
b
e
d
o
c
u
r
v
e
s
o
f
t
h
e
f
o
r
e
h
e
a
d
r
e
g
i
o
n
o
f
f
a
c
i
a
l
s
k
i
n
a
n
d
m
a
s
k
m
a
t
e
r
i
a
ls
.
A
t
t
h
es
e
t
w
o
li
g
h
t
w
a
v
el
e
n
g
t
h
s
,
g
r
e
a
t
e
r
r
e
f
l
ec
t
a
n
c
e
c
o
n
t
r
a
s
t
h
as
b
e
e
n
a
c
h
i
e
v
e
d
b
e
tw
e
e
n
a
g
e
n
u
i
n
e
f
a
c
i
a
l
s
k
i
n
a
n
d
2
D
i
m
a
g
e
s
.
I
n
o
d
e
r
t
o
f
i
n
d
t
h
e
p
r
e
s
e
n
c
e
o
f
h
u
m
a
n
s
k
i
n
,
a
m
e
t
h
o
d
b
a
s
e
d
o
n
s
p
e
c
t
r
o
s
c
o
p
y
w
a
s
e
m
p
l
o
y
e
d
[
5
]
.
T
h
e
p
r
o
c
e
s
s
i
n
g
m
e
t
h
o
d
w
a
s
d
iv
i
d
e
d
i
n
t
o
t
w
o
r
e
g
i
o
n
s
k
n
o
w
n
a
s
v
is
i
b
l
e
l
i
g
h
t
(
V
IS
)
a
n
d
n
e
a
r
-
i
n
f
r
a
r
e
d
(
N
I
R
)
.
I
n
V
I
S
,
Y
C
b
C
r
c
o
l
o
u
r
s
p
a
ce
wa
s
u
s
e
d
t
o
s
p
e
ci
f
y
p
i
x
e
l
w
h
o
s
e
c
o
l
o
u
r
i
s
s
i
m
i
la
r
to
s
k
i
n
.
W
h
i
l
e
i
n
N
I
R
,
t
h
e
r
e
f
l
e
c
t
a
n
c
e
d
i
f
f
e
r
e
n
ce
i
n
t
w
o
d
if
f
e
r
e
n
t
w
a
v
e
l
e
n
g
t
h
s
o
f
e
a
c
h
i
m
a
g
e
w
a
s
u
s
e
d
t
o
d
e
t
e
r
m
i
n
e
h
u
m
a
n
s
k
i
n
.
Ov
er
th
e
last
2
0
y
ea
r
s
,
class
if
y
in
g
m
ater
ials
h
av
e
alr
ea
d
y
b
ee
n
d
o
n
e
b
y
u
s
in
g
p
o
lar
i
ze
d
lig
h
t.
Var
io
u
s
ex
p
er
im
e
n
ts
wer
e
d
o
n
e
to
p
r
o
o
f
th
at
r
e
f
lectio
n
p
r
o
p
e
r
ties
o
f
ea
ch
o
b
ject
m
a
y
b
e
u
s
ed
as c
lass
if
icatio
n
m
eth
o
d
[
6
]
.
I
n
ad
d
itio
n
,
d
if
f
u
s
e
an
d
s
p
ec
u
lar
r
ef
lectio
n
s
ca
n
b
e
s
ep
ar
ated
u
s
in
g
p
o
lar
ize
d
lig
h
t
b
y
ad
ju
s
tin
g
th
e
p
o
lar
izatio
n
a
n
g
les
.
A
y
e
ar
later
,
a
p
o
lar
i
z
atio
n
b
ased
m
eth
o
d
was
in
tr
o
d
u
ce
d
to
d
is
cr
im
in
ate
b
etwe
en
m
etal
an
d
d
ielec
tr
ic
s
u
r
f
ac
es
b
ased
u
p
o
n
Fre
s
n
el
r
ef
lecta
n
ce
th
eo
r
y
[
7
]
.
Acc
o
r
d
i
n
g
to
th
is
t
h
eo
r
y
,
th
e
s
p
ec
u
la
r
co
m
p
o
n
en
t o
f
r
ef
lectio
n
is
ass
u
m
ed
to
b
e
h
ig
h
er
th
a
n
th
e
d
i
f
f
u
s
e
r
ef
lectio
n
c
o
m
p
o
n
en
t.
P
o
p
u
lar
ity
o
f
p
o
lar
izatio
n
-
b
a
s
ed
tech
n
iq
u
es
h
as
in
cr
ea
s
ed
s
in
ce
th
en
,
n
o
tab
ly
in
d
if
f
e
r
en
tiatin
g
m
ater
ials
.
I
n
2
0
1
1
,
a
n
im
a
g
e
s
en
s
o
r
ca
lled
co
m
p
lem
e
n
tar
y
m
etal
-
o
x
id
e
-
s
em
ico
n
d
u
cto
r
(
C
MO
S
)
was
d
ev
elo
p
e
d
to
id
en
tify
m
etal
a
n
d
d
ielec
tr
ic
s
u
r
f
ac
es
[
8
].
D
e
g
r
ee
o
f
p
o
lar
izatio
n
f
o
r
p
last
ic
s
u
r
f
ac
e
was
h
ig
h
e
r
co
m
p
ar
ed
to
alu
m
in
i
u
m
s
u
r
f
ac
e
.
An
o
th
er
s
tu
d
y
u
s
in
g
p
o
lar
ized
lig
h
t
h
as
b
ee
n
d
o
n
e
to
class
if
y
am
o
n
g
tr
an
s
p
ar
en
t
an
d
o
p
aq
u
e
o
b
jects
[
9
]
.
M
eth
o
d
o
lo
g
ies
s
u
ch
a
s
Sto
k
es
d
eg
r
ee
o
f
p
o
lar
izati
o
n
an
d
p
o
lar
izatio
n
Fre
s
n
el
r
atio
wer
e
p
r
esen
te
d
.
T
h
e
r
esu
lts
s
h
o
wed
th
at
t
h
e
d
eg
r
ee
o
f
p
o
lar
izatio
n
of
tr
an
s
p
ar
en
t
o
b
ject
was
h
ig
h
er
th
a
n
o
p
aq
u
e
o
b
jects f
o
r
m
o
s
t o
f
in
cid
e
n
t a
n
g
les.
I
n
o
d
er
to
d
if
f
er
en
tiate
g
en
u
i
n
e
h
u
m
an
s
k
in
an
d
o
th
e
r
m
ater
ials
,
p
r
o
p
er
ty
o
f
h
u
m
an
r
ea
l
s
k
in
ca
n
b
e
m
an
ip
u
lated
.
Fo
r
in
s
tan
ce
,
h
u
m
an
s
k
in
tex
tu
r
e
co
n
s
is
ts
o
f
th
r
ee
lay
er
s
s
t
r
u
ctu
r
e
wh
ich
ar
e
k
n
o
wn
as
th
e
ep
id
er
m
is
lay
er
,
th
e
d
er
m
is
lay
er
an
d
t
h
e
f
at
lay
er
[
10
]
.
T
h
er
ef
o
r
e,
s
k
in
p
r
o
d
u
ce
s
m
o
r
e
t
h
an
o
n
e
r
ef
lectio
n
wh
ich
ar
e
n
am
ed
as
s
u
r
f
ac
e
a
n
d
s
u
b
-
s
u
r
f
ac
e
r
ef
lectio
n
s
.
T
h
eo
r
itically
,
th
ese
co
u
ld
b
e
u
s
e
d
as
a
clas
s
if
icatio
n
f
ea
tu
r
e
b
etwe
en
h
u
m
an
s
k
in
a
n
d
f
ak
e
f
ac
e.
T
h
u
s
,
p
o
lar
ize
d
l
ig
h
t
co
u
ld
b
e
u
s
ed
f
o
r
th
is
p
u
r
p
o
s
e.
Pre
v
io
u
s
ly
,
a
s
tu
d
y
u
s
in
g
p
o
lar
ized
lig
h
t
was
im
p
lem
en
ted
to
co
m
p
ar
e
b
etwe
en
v
is
ib
le
lig
h
t
im
ag
es
a
n
d
p
o
lar
ized
im
a
g
es
o
f
h
u
m
an
s
k
in
with
m
u
ltip
le
d
is
ea
s
es
[
1
1
]
.
B
ase
d
o
n
t
h
e
s
t
u
d
y
,
p
o
la
r
izatio
n
im
a
g
es
v
is
u
alize
d
th
e
d
is
r
u
p
tio
n
o
n
s
k
in
s
u
r
f
ac
e
s
u
ch
as sk
in
ca
n
ce
r
b
ased
o
n
th
e
s
u
r
f
ac
e
an
d
s
u
b
-
s
u
r
f
ac
e
r
e
f
lectio
n
s
an
aly
s
is
.
A
p
o
lar
izatio
n
im
ag
in
g
d
e
v
ice
was
d
em
o
n
s
tr
ated
to
m
o
n
ito
r
lig
h
ts
tr
an
s
m
itti
n
g
th
r
o
u
g
h
h
ea
lth
y
b
r
ea
s
t
tis
s
u
e
an
d
b
r
ea
s
t
t
is
s
u
e
with
ca
n
ce
r
[
1
2
]
.
I
n
an
o
th
er
s
tu
d
y
,
Sto
k
es
v
alu
es
o
f
im
ag
es
o
b
tain
ed
u
n
d
e
r
p
ar
allel
p
o
lar
ized
lig
h
t
ar
e
r
el
ated
to
h
u
m
a
n
s
k
in
ch
ar
ac
te
r
is
tics
[
1
3
]
.
Th
u
s
,
th
e
f
in
d
in
g
w
as
u
s
ed
to
ev
alu
ate
d
er
m
ato
lo
g
ic
d
is
ea
s
es
.
I
n
2
0
1
2
,
th
e
ad
v
an
tag
es
o
f
p
o
lar
iz
ed
lig
h
t
wer
e
m
an
ip
u
lated
t
o
class
if
y
s
k
in
in
to
s
ev
er
al
ag
e
g
r
o
u
p
s
[
1
4
]
.
As th
e
r
esu
lts
,
s
u
r
f
ac
e
r
ef
lectio
n
o
f
y
o
u
n
g
er
s
k
in
is
h
ig
h
er
co
m
p
ar
ed
with
o
ld
er
s
k
i
n
.
I
n
v
iew
o
f
s
ev
er
al
p
r
ev
io
u
s
wo
r
k
s
th
at
h
as
b
ee
n
m
e
n
tio
n
ed
s
o
f
ar
,
it
is
u
n
d
en
iab
le
th
at
p
o
lar
izatio
n
-
b
ased
tech
n
iq
u
e
ca
n
b
e
im
p
le
m
en
ted
as
o
n
e
o
f
a
n
ti
-
s
p
o
o
f
in
g
tech
n
iq
u
es
in
f
ac
e
r
ec
o
g
n
iti
o
n
s
y
s
tem
s
.
T
h
u
s
,
in
th
is
p
ap
er
,
m
u
lti
-
r
e
f
lectio
n
s
o
f
r
ea
l
f
ac
e,
p
r
in
ted
p
h
o
t
o
an
d
iPad
d
is
p
lay
ed
im
ag
es
wer
e
s
ep
ar
ated
an
d
s
tatis
t
ically
an
aly
s
ed
.
Th
is
p
a
p
er
is
o
r
g
a
n
ized
i
n
s
ev
er
al
s
ec
tio
n
s
.
S
ec
tio
n
2
lis
t
s
s
ev
er
al
th
eo
r
ies
t
h
at
d
r
iv
e
to
war
d
th
is
r
esear
ch
.
I
n
s
ec
tio
n
3
,
r
esear
ch
m
eth
o
d
is
p
r
esen
ted
.
I
n
s
ec
tio
n
4
,
ex
p
e
r
i
m
en
tal
r
esu
lts
ar
e
d
is
cu
s
s
ed
.
Fin
ally
,
co
n
clu
s
io
n
an
d
f
u
t
u
r
e
wo
r
k
ar
e
p
r
o
v
id
ed
i
n
s
ec
tio
n
5.
2.
RE
S
E
ARCH
T
H
E
O
RY
2
.
1
.
H
um
a
n sk
in’s str
uct
ure
T
o
d
etec
t
f
ak
e
f
ac
es
in
a
f
a
ce
b
io
m
etr
ic
s
y
s
tem
,
it
is
v
i
tal
to
u
n
d
er
s
tan
d
th
e
ch
ar
ac
t
er
is
tics
o
f
g
en
u
in
e
h
u
m
a
n
s
k
in
.
On
e
o
f
th
em
is
s
k
in
s
u
r
f
a
ce
r
ef
lectio
n
s
.
B
asically
,
h
u
m
an
s
k
in
s
tr
u
ct
u
r
e
h
as
th
r
ee
lay
er
s
wh
ich
ar
e
ep
id
er
m
is
,
d
e
r
m
is
an
d
f
at
lay
er
s
[
1
5
]
.
T
h
e
o
u
t
m
o
s
t
lay
er
is
ca
lled
th
e
ep
id
er
m
is
th
at
co
n
tain
s
m
elan
in
wh
ich
ac
t
as
ab
s
o
r
p
tio
n
an
d
s
ca
tter
in
g
ag
en
ts
.
Sk
in
co
lo
u
r
co
n
tr
ib
u
tes
to
th
e
a
m
o
u
n
t
o
f
m
elan
in
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
F
a
ce
s
p
o
o
fin
g
d
etec
tio
n
u
s
in
g
s
u
r
fa
ce
a
n
d
s
u
b
-
s
u
r
fa
ce
r
eflec
tio
n
s
a
n
a
lysi
s
(
A
z
im
Za
lih
a
A
b
d
A
z
iz
)
191
Dar
k
s
k
in
h
as
m
o
r
e
am
o
u
n
t
o
f
m
elan
in
co
m
p
ar
e
d
to
f
air
s
k
in
co
lo
u
r
[
1
6
]
.
Der
m
is
,
th
e
s
ec
o
n
d
lay
er
is
a
th
ick
lay
er
u
n
d
er
n
ea
t
h
th
e
e
p
id
er
m
i
s
.
T
h
e
d
ee
p
est
lay
er
is
m
ad
e
o
f
f
at
a
n
d
c
o
n
n
ec
tiv
e
tis
s
u
es.
Fig
u
r
e
1
illu
s
tr
ates
th
e
s
tr
u
ctu
r
e
o
f
s
k
in
lay
er
s
wh
ich
was d
ep
icted
f
r
o
m
[
1
7
]
.
Fig
u
r
e
1
.
Hu
m
an
s
k
in
’
s
s
tr
u
ct
u
r
e
As
ca
n
b
e
s
ee
n
in
Fig
u
r
e
1
,
s
k
in
co
n
tain
s
o
f
m
u
ltil
ay
er
s
s
tr
u
ctu
r
es
.
B
ec
au
s
e
o
f
th
is
,
s
k
in
p
r
o
d
u
ce
s
m
o
r
e
th
a
n
o
n
e
r
e
f
lectio
n
.
W
h
en
n
o
r
m
al
lig
h
t
b
ea
m
s
tr
ik
e
s
o
n
s
k
in
s
u
r
f
ac
e,
p
ar
t
o
f
th
e
in
cid
en
ce
lig
h
t
is
r
ef
lecte
d
at
th
e
air
-
s
k
in
s
u
r
f
a
ce
.
T
h
is
h
ap
p
en
s
d
u
e
to
t
h
e
d
if
f
er
en
t
i
n
in
d
e
x
o
f
r
e
f
r
ac
tio
n
o
f
ea
ch
m
e
d
iu
m
.
Fig
u
r
e
2
s
h
o
ws
th
e
r
e
f
lectio
n
s
o
f
n
o
r
m
al
lig
h
t
b
ea
m
wh
e
n
it
s
tr
ik
es
o
n
s
k
in
s
u
r
f
ac
e.
Fro
m
Fig
u
r
e
2
,
it
ca
n
b
e
s
ee
n
th
at
p
ar
ticles
l
ik
e
m
elan
in
,
co
llag
en
an
d
h
ae
m
o
g
lo
b
in
ac
t a
s
s
ca
tter
in
g
an
d
ab
s
o
r
p
tio
n
ag
en
ts
o
n
th
e
lig
h
t
wav
e
th
at
p
en
etr
ates th
r
o
u
g
h
t
h
e
lay
er
s
.
Fig
u
r
e
2
.
R
ef
lectio
n
s
wh
en
lig
h
ts
h
it h
u
m
an
s
k
i
n
s
u
r
f
ac
e
2
.
2
.
P
rint
ed
ph
o
t
o
pa
per’
s
s
t
ruct
ure
Ph
o
to
attac
k
s
ar
e
u
s
u
ally
h
ap
p
en
ed
wh
en
a
p
h
o
to
g
r
ap
h
o
f
a
l
eg
itima
te
u
s
er
is
p
r
esen
ted
in
f
r
o
n
t
o
f
a
f
ac
e
b
io
m
etr
ic
s
y
s
tem
as
attem
p
ts
to
g
ain
ac
ce
s
s
.
I
m
p
o
s
ter
s
u
s
ed
p
ap
er
m
ask
as
s
p
o
o
f
in
g
attac
k
s
b
ec
au
s
e
o
f
s
o
m
e
b
asic
r
ea
s
o
n
s
.
First,
it
is
ea
s
y
to
g
et
a
p
h
o
to
o
f
an
y
leg
itima
te
u
s
er
s
eith
er
b
y
d
o
wn
l
o
ad
in
g
f
r
o
m
r
elate
d
s
o
cial
m
ed
ia
web
s
ite
s
o
r
ap
p
licatio
n
s
.
Nex
t,
p
h
o
to
s
ca
n
also
b
e
tak
en
b
y
an
im
p
o
s
to
r
u
s
in
g
d
ig
ital
o
r
v
id
eo
ca
m
er
a.
T
h
o
s
e
p
h
o
to
s
ca
n
b
e
ea
s
ily
p
r
in
ted
an
d
u
s
ed
a
s
s
p
o
o
f
in
g
attem
p
ts
.
B
ec
au
s
e
o
f
lo
w
co
s
t
an
d
af
f
o
r
d
a
b
le,
p
a
p
er
m
ask
is
ac
tu
ally
a
p
o
p
u
lar
ch
o
ice
as sp
o
o
f
i
n
g
attac
k
s
am
o
n
g
im
p
o
s
ter
s
.
Similar
as
o
th
er
m
ater
ials
,
p
a
p
er
also
h
as
its
o
wn
p
h
y
s
ical
p
r
o
p
er
ties
.
I
t
is
n
ec
ess
ar
y
to
s
tu
d
y
t
h
e
p
r
o
p
er
ties
o
f
a
p
iece
o
f
p
a
p
er
b
ef
o
r
e
p
r
o
ce
ed
in
g
in
ex
am
in
at
io
n
o
f
p
o
lar
ize
d
r
ef
lectio
n
b
et
wee
n
r
ea
l
f
ac
e
an
d
p
h
o
to
attac
k
s
.
B
asically
,
p
a
p
e
r
is
m
ad
e
o
f
ce
llu
l
o
s
e
f
ib
r
es
,
a
co
m
p
le
x
ca
r
b
o
h
y
d
r
ate
co
n
s
is
tin
g
o
f
m
o
r
e
th
a
n
3
,
0
0
0
g
lu
c
o
s
e
u
n
its
[
18
]
.
Pap
er
h
as
f
o
u
r
im
p
o
r
ta
n
t
o
p
tical
p
r
o
p
er
ties
wh
ich
ar
e
b
r
ig
h
tn
e
s
s
,
co
lo
u
r
,
o
p
ac
ity
,
an
d
g
lo
s
s
[
19
]
.
Acc
o
r
d
in
g
to
[
2
0
]
,
m
atte
p
a
p
er
h
as
lo
w
r
ef
le
ctio
n
p
r
o
p
er
ty
.
Mo
s
t
o
f
lig
h
t
s
th
at
h
it
m
atte
p
ap
er
is
ab
s
o
r
b
ed
,
o
n
ly
a
s
m
all
am
o
u
n
t
is
r
ef
lecte
d
.
T
h
e
ab
s
o
r
b
e
d
lig
h
t
is
d
if
f
u
s
ed
b
y
th
e
p
ap
e
r
b
ef
o
r
e
r
e
-
e
m
er
g
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
2
4
,
No
.
1
,
Octo
b
er
20
21
:
1
8
9
-
19
7
192
to
th
e
ai
r
.
A
m
o
u
n
t
o
f
th
e
r
ef
le
cted
lig
h
t
is
also
af
f
ec
ted
b
y
t
h
e
in
k
s
p
r
in
ted
o
n
th
e
p
a
p
er
.
I
n
an
in
v
esti
g
atio
n
,
lig
h
t w
h
ich
s
tr
ik
es o
n
th
e
b
lac
k
d
o
ts
is
ab
s
o
r
b
ed
wh
ile
th
e
r
e
m
ain
in
g
is
d
if
f
u
s
ed
b
y
th
e
p
a
p
er
[
2
1
]
.
G
e
n
e
r
a
l
l
y
,
i
t
s
e
e
m
s
t
h
a
t
h
u
m
a
n
s
k
i
n
a
n
d
c
o
l
o
u
r
e
d
p
r
i
n
t
e
d
p
a
p
e
r
h
a
v
e
s
i
m
i
l
a
r
r
e
f
l
e
c
t
i
o
n
p
r
o
p
e
r
t
i
e
s
.
B
o
t
h
p
r
o
d
u
c
e
m
o
r
e
t
h
a
n
o
n
e
r
e
f
l
e
c
t
i
o
n
:
s
u
r
f
a
c
e
a
n
d
s
u
b
s
u
r
f
a
c
e
(
d
i
f
f
u
s
e
)
r
e
f
l
e
c
t
i
o
n
s
.
B
e
c
a
u
s
e
o
f
t
h
i
s
,
i
t
c
a
n
b
e
a
s
s
u
m
e
d
t
h
a
t
t
h
e
a
n
a
l
y
s
i
s
t
o
d
i
f
f
e
r
e
n
t
i
a
t
e
b
e
t
w
e
e
n
r
e
a
l
f
a
c
e
a
n
d
p
r
i
n
t
e
d
p
h
o
t
o
w
i
l
l
e
x
p
e
r
i
e
n
c
e
a
m
o
r
e
i
n
t
e
n
s
e
c
h
a
l
l
e
n
g
e
.
2
.
3
.
iP
a
d scre
en
dis
pla
y
ed
i
m
a
g
e
iPad
s
cr
ee
n
is
a
f
lat
d
is
p
lay
p
an
el
o
r
b
asically
k
n
o
wn
as
a
liq
u
id
-
cr
y
s
tal
d
is
p
lay
(
L
C
D)
.
L
iq
u
id
cr
y
s
tal
d
o
es
n
o
t
e
m
it
lig
h
t
d
ir
ec
tly
,
in
s
tead
it
u
s
e
b
ac
k
li
g
h
t
o
r
r
ef
lecto
r
to
p
r
o
d
u
ce
i
m
ag
es
in
co
lo
r
o
r
m
o
n
o
c
h
r
o
m
e.
T
h
e
b
ac
k
lig
h
t
em
itted
f
r
o
m
t
h
e
s
cr
ee
n
is
p
o
lar
ized
.
L
ig
h
t
wav
es
in
ter
f
er
e
n
ce
m
ig
h
t
o
cc
u
r
if
two
o
r
m
o
r
e
lig
h
t
wav
es
m
e
et.
Po
lar
ized
lig
h
t
em
itted
f
r
o
m
iPad
s
cr
ee
n
s
h
o
u
ld
b
e
p
er
p
en
d
icu
la
r
to
t
h
e
p
o
lar
izatio
n
ax
is
o
f
lig
h
t
s
o
u
r
ce
.
T
h
u
s
,
th
e
tr
av
elin
g
p
er
p
en
d
icu
lar
lig
h
t
wav
e
is
s
to
p
p
ed
b
y
a
p
o
lar
izer
f
r
o
m
tr
an
s
m
itti
n
g
th
r
o
u
g
h
.
T
h
e
ca
m
er
a
ca
p
tu
r
ed
im
ag
e
at
9
0
0
p
o
lar
izatio
n
an
g
le
with
o
u
t a
n
y
lig
h
t tr
an
s
m
itted
.
T
h
is
is
a
s
ig
n
if
ican
t
p
o
s
itiv
e
co
r
r
elatio
n
b
etwe
en
lig
h
ts
em
itted
f
r
o
m
iPad
s
cr
ee
n
an
d
th
e
p
o
la
r
izatio
n
an
g
le.
T
h
e
d
if
f
er
en
ce
s
b
etwe
en
r
ea
l
f
ac
e
an
d
iPad
d
is
p
lay
ed
im
ag
e
at
9
0
0
p
o
lar
izatio
n
an
g
le
ca
n
b
e
u
s
ed
to
d
is
cr
im
in
ate
r
ea
l f
ac
e
an
d
th
e
iPad
attac
k
.
3.
RE
S
E
ARCH
M
E
T
H
O
D
3
.
1
.
T
he
po
la
riza
t
io
n i
m
a
g
i
ng
s
y
s
t
em
Sin
ce
th
er
e
is
n
o
p
o
lar
ized
i
m
ag
es
d
ataset
th
at
is
p
u
b
licly
av
ailab
le,
p
o
lar
ize
d
im
ag
es
o
f
h
u
m
a
n
f
ac
es
an
d
p
r
i
n
ted
p
h
o
to
s
u
s
e
d
in
th
is
r
esear
ch
wer
e
s
elf
-
co
llected
.
T
h
e
d
ata
wer
e
co
l
lecte
d
u
s
in
g
o
wn
-
de
v
elo
p
e
d
im
ag
in
g
s
y
s
tem
s
im
ilar
to
an
im
ag
in
g
s
y
s
tem
in
tr
o
d
u
ce
d
in
[
2
2
]
-
[
2
4
]
.
Fig
u
r
e
3
illu
s
tr
ates
th
e
ex
p
er
im
en
tal
s
etu
p
u
s
ed
in
to
ca
p
tu
r
e
im
ag
es in
t
h
is
s
tu
d
y
.
Fig
u
r
e
3
.
T
h
e
p
o
la
r
izatio
n
im
a
g
in
g
s
etu
p
I
n
Fig
u
r
e
3
,
two
tab
le
lig
h
ts
wer
e
p
lace
d
8
0
cm
ea
c
h
to
th
e
r
ig
h
t
an
d
lef
t
o
f
th
e
ca
m
er
a.
T
h
e
lig
h
ts
wer
e
p
lace
d
to
lig
h
t
u
p
th
e
ex
p
er
im
en
tal
s
u
b
jects
at
an
an
g
le
o
f
4
5
0
.
E
ac
h
o
f
th
em
was
co
u
p
led
with
a
lin
ea
r
p
o
lar
izer
s
h
ee
t,
lab
elled
as
P
2
an
d
P
3
.
B
o
th
p
o
lar
izer
s
wer
e
alig
n
ed
in
p
a
r
allel
p
o
s
itio
n
t
o
ea
ch
o
th
e
r
at
th
e
s
am
e
p
o
lar
izatio
n
ax
is
.
T
h
is
s
etu
p
will
let
th
e
tab
le
lig
h
ts
wav
e
tr
av
elled
in
th
e
s
am
e
d
ir
ec
tio
n
to
war
d
s
th
e
s
u
b
ject.
T
h
e
ca
m
er
a
u
s
ed
in
th
is
ex
p
er
im
en
t
was
Sam
s
u
n
g
Fu
ll
HD
ca
m
er
a
,
p
lace
d
8
0
cm
in
f
r
o
n
t
o
f
th
e
s
u
b
ject.
A
lin
ea
r
p
o
lar
ize
r
m
a
d
e
o
f
g
lass
,
P
1
,
co
u
p
led
with
a
n
an
g
le
r
o
tato
r
,
was
p
lace
d
in
f
r
o
n
t
o
f
th
e
ca
m
e
r
a
len
s
.
An
an
g
le
r
o
tato
r
was u
s
e
d
to
ad
ju
s
t th
e
p
o
lar
izatio
n
a
n
g
les d
u
r
in
g
t
h
e
d
ataset
r
ec
o
r
d
i
n
g
p
r
o
ce
s
s
.
3
.
2
.
Su
rf
a
ce
a
nd
s
ub
-
s
urf
a
c
e
im
a
g
es’
da
t
a
s
et
Po
lar
ize
d
im
ag
e
ca
p
tu
r
e
d
u
n
d
er
p
ar
allel
p
o
lar
izatio
n
co
n
s
is
ts
b
o
th
s
u
r
f
ac
e
an
d
s
u
b
s
u
r
f
ac
e
r
ef
lectio
n
s
[
1
4
]
.
I
m
a
g
e
with
o
n
ly
s
u
b
s
u
r
f
ac
e
r
ef
lectio
n
ca
n
b
e
r
ec
o
r
d
ed
u
n
d
er
9
0
0
p
o
lar
iz
atio
n
an
g
le.
Fo
r
th
is
,
p
o
lar
ized
im
a
g
e
u
n
d
er
9
0
0
p
o
lar
izatio
n
an
g
le
is
s
u
b
tr
ac
te
d
f
r
o
m
p
o
la
r
ized
im
ag
e
r
ec
o
r
d
ed
u
n
d
er
p
ar
allel
p
o
lar
izatio
n
.
I
n
th
is
s
tu
d
y
,
p
o
lar
ized
im
ag
e
ca
p
tu
r
ed
u
n
d
er
9
0
0
p
o
lar
izatio
n
an
g
le
is
d
e
n
o
ted
as
I
m
g
90
wh
ile
p
o
lar
ized
im
a
g
e
ca
p
tu
r
ed
u
n
d
er
p
ar
allel
p
o
la
r
izatio
n
is
lab
el
led
as
I
m
g
0
.
An
y
im
ag
e
with
s
u
r
f
ac
e
r
e
f
lecti
o
n
is
k
n
o
wn
as
S
1
wh
ich
is
o
n
e
o
f
t
h
e
Sto
k
es
co
m
p
o
n
en
ts
.
Fig
u
r
e
4
lis
ted
s
am
p
les
o
f
I
m
g
90
an
d
S
1
o
f
th
e
r
ea
l
an
d
f
ak
e
f
ac
es
av
ailab
le
in
th
e
s
elf
-
co
llected
d
ataset.
T
h
er
e
ar
e
two
n
ew
d
atasets
,
n
am
ely
as
f
ac
e
-
p
h
o
to
-
s
p
o
o
f
(
FaPs
)
d
ataset
an
d
Face
-
iPad
d
ataset
.
I
n
FaPs
d
ataset,
th
er
e
ar
e
3
7
g
e
n
u
in
e
f
ac
es
wer
e
r
ec
o
r
d
e
d
r
an
d
o
m
ly
s
elec
ted
am
o
n
g
th
e
f
ac
u
lty
m
em
b
er
s
.
T
h
ey
wer
e
th
e
n
g
e
n
er
ally
class
if
ied
in
to
th
r
ee
g
r
o
u
p
s
:
2
6
Asi
an
;
4
B
lack
;
an
d
7
C
au
ca
s
ian
.
Me
an
wh
ile,
p
r
in
ted
p
h
o
to
s
o
f
e
ac
h
s
u
b
ject
wer
e
ad
d
ed
to
th
e
FaPs
d
atase
t.
I
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
F
a
ce
s
p
o
o
fin
g
d
etec
tio
n
u
s
in
g
s
u
r
fa
ce
a
n
d
s
u
b
-
s
u
r
fa
ce
r
eflec
tio
n
s
a
n
a
lysi
s
(
A
z
im
Za
lih
a
A
b
d
A
z
iz
)
193
an
o
th
er
d
ataset
ca
lled
Face
-
iPad
d
ataset,
th
er
e
ar
e
f
iv
e
f
em
ales
an
d
th
r
ee
m
ales
o
f
two
s
k
in
co
lo
u
r
s
:
Asi
an
an
d
C
au
ca
s
ian
.
I
m
ag
es
o
f
ea
c
h
s
u
b
ject
wer
e
ca
p
tu
r
e
d
a
n
d
d
ig
itally
s
to
r
ed
t
o
b
e
later
d
is
p
lay
ed
as
s
p
o
o
f
in
g
attem
p
t.
As
s
h
o
wn
in
Fi
g
u
r
e
4
,
it
is
q
u
ite
clea
r
t
h
at
S
1
im
ag
es
b
e
twee
n
r
ea
l
f
ac
e
,
p
r
in
ted
p
h
o
t
o
an
d
iPad
d
is
p
lay
ed
im
ag
es
ar
e
s
ig
n
if
i
ca
n
tly
d
if
f
er
e
n
t.
Ho
wev
er
,
t
h
e
s
u
b
s
u
r
f
ac
e
im
a
g
es
r
ep
r
es
en
ted
b
y
I
m
g
90
ar
e
r
elativ
ely
s
im
ilar
b
etwe
en
th
e
r
ea
l
f
ac
e
an
d
p
r
in
ted
p
h
o
t
o
.
Fo
r
I
m
g
90
o
f
th
e
iPad
d
is
p
lay
ed
f
ac
e
s
h
o
ws
to
tal
d
if
f
er
en
t c
o
m
p
ar
e
d
to
I
m
g
90
o
f
th
e
g
en
u
in
e
f
ac
e.
T
h
ese
im
ag
es we
r
e
th
en
s
tatis
tically
an
aly
s
ed
.
Fig
u
r
e
4
.
T
h
e
s
u
r
f
ac
e
(
S1
)
an
d
s
u
b
s
u
r
f
ac
e
(
I
m
g
9
0
)
im
ag
es
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
R
esu
lts
o
f
th
is
r
esear
ch
is
p
r
e
s
en
ted
an
d
d
is
cu
s
s
ed
in
s
ev
er
al
s
u
b
s
ec
tio
n
s
.
Sin
ce
th
e
m
ai
n
o
b
jectiv
e
is
to
s
tu
d
y
s
u
r
f
ac
e
r
ef
lectio
n
p
r
o
p
er
ties
b
y
u
s
in
g
s
tatis
tical
m
ea
s
u
r
es,
ea
ch
m
ea
s
u
r
e
is
ex
p
lain
ed
in
d
etails
in
s
u
b
s
ec
tio
n
ac
co
r
d
i
n
g
ly
.
4.
1
.
M
ea
n,
s
t
a
nd
a
rd
dev
ia
t
i
o
n,
s
k
ewness
a
nd
k
urt
o
s
is
I
n
t
h
i
s
s
e
ct
i
o
n
,
r
e
f
l
e
ct
i
o
n
p
r
o
p
e
r
t
i
e
s
f
o
r
b
o
t
h
S
1
a
n
d
I
m
g
90
i
m
a
g
e
s
o
f
e
a
c
h
g
e
n
u
i
n
e
a
n
d
f
a
k
e
f
a
c
e
s
w
e
r
e
s
t
a
ti
s
t
i
c
al
l
y
a
n
a
l
y
s
e
d
.
Fi
g
u
r
e
s
5
a
n
d
6
p
r
e
s
e
n
t
t
h
e
s
t
a
t
is
t
i
ca
l
r
e
s
u
l
t
s
o
f
S
1
a
n
d
I
m
g
90
i
m
a
g
e
s
o
f
e
a
c
h
s
u
b
j
e
c
t
,
r
e
s
p
e
c
t
i
v
el
y
.
A
s
c
a
n
b
e
s
ee
n
i
n
F
i
g
u
r
e
5
,
t
h
e
r
e
i
s
a
s
i
g
n
i
f
i
c
a
n
t
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
r
e
a
l
f
a
c
es
a
n
d
i
P
a
d
d
i
s
p
l
a
y
ed
f
a
c
e
s
,
m
e
as
u
r
e
d
b
y
t
h
e
m
e
a
n
an
d
s
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
.
I
n
c
o
n
t
r
a
s
t
,
s
k
e
w
n
es
s
a
n
d
k
u
r
t
o
s
is
o
f
th
e
s
e
tw
o
s
u
b
j
e
ct
s
a
r
e
s
i
m
il
a
r
t
o
e
a
c
h
o
t
h
e
r
.
S
u
r
p
r
i
s
in
g
l
y
,
n
o
n
e
o
f
t
h
e
s
e
m
e
a
s
u
r
e
s
s
h
o
w
s
s
i
g
n
i
f
i
c
a
n
t
d
i
f
f
e
r
e
n
t
b
e
tw
e
e
n
r
e
a
l
f
a
c
e
s
a
n
d
p
r
i
n
t
e
d
p
h
o
t
o
f
a
c
e
s
.
D
at
a
i
n
F
ig
u
r
e
5
c
a
n
b
e
c
o
m
p
a
r
e
d
w
i
t
h
s
ta
t
i
s
t
i
c
al
d
a
t
a
o
f
s
u
b
s
u
r
f
a
c
e
i
m
ag
e
s
i
n
Fi
g
u
r
e
6
.
Fig
u
r
e
5
.
Su
r
f
ac
e
im
ag
e,
S1
o
f
r
ea
l a
n
d
f
ak
e
f
ac
es’
s
tatis
tical
an
aly
s
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
2
4
,
No
.
1
,
Octo
b
er
20
21
:
1
8
9
-
19
7
194
F
r
o
m
F
i
g
u
r
e
6
,
i
t
c
a
n
b
e
s
ee
n
t
h
a
t
m
e
a
n
a
n
d
s
t
a
n
d
a
r
d
d
e
v
i
at
i
o
n
s
c
o
r
es
b
e
t
w
ee
n
r
e
a
l
f
a
c
es
a
n
d
i
P
a
d
f
a
c
e
s
a
r
e
s
i
g
n
i
f
i
c
a
n
tl
y
d
i
f
f
e
r
e
n
t
.
D
e
s
p
i
te
a
s
l
i
g
h
t
m
e
a
n
i
n
ten
s
i
t
y
d
i
f
f
e
r
e
n
c
e
o
f
I
m
g
90
b
e
t
w
e
e
n
r
e
a
l
a
n
d
p
h
o
t
o
f
a
c
e
s
,
n
o
o
t
h
e
r
d
i
f
f
e
r
e
n
ce
s
w
er
e
f
o
u
n
d
i
n
t
h
e
s
t
a
n
d
a
r
d
d
e
v
iat
i
o
n
,
s
k
e
w
n
ess
a
n
d
k
u
r
t
o
s
is
m
e
a
s
u
r
e
s
.
T
o
f
u
r
t
h
e
r
e
x
a
m
i
n
e
t
h
e
d
is
t
r
i
b
u
ti
o
n
o
f
S
1
a
n
d
I
m
g
90
f
o
r
t
h
e
r
e
al
f
a
c
es
an
d
t
h
e
t
w
o
f
a
k
e
f
a
ce
s
,
h
i
s
t
o
g
r
a
m
s
o
f
e
a
c
h
s
u
b
j
e
ct
w
a
s
p
l
o
tt
e
d
a
n
d
p
r
e
s
e
n
t
e
d
i
n
Fi
g
u
r
e
7
.
A
s
d
i
s
p
l
a
y
e
d
i
n
Fi
g
u
r
e
7
,
S
1
i
m
a
g
e
s
f
o
r
t
h
e
r
e
a
l
f
a
ce
a
n
d
p
r
i
n
t
e
d
p
h
o
t
o
f
a
c
e
a
r
e
ass
u
m
e
d
t
o
h
a
v
e
u
n
im
o
d
a
l
d
i
s
t
r
i
b
u
t
i
o
n
s
w
h
e
r
e
as
iPa
d
d
i
s
p
l
a
y
e
d
f
a
c
e
i
m
a
g
e
s
h
a
s
b
i
m
o
d
a
l
d
i
s
t
r
i
b
u
t
i
o
n
.
I
n
c
o
n
t
r
a
s
t
,
I
m
g
90
f
o
r
t
h
e
i
P
a
d
f
a
c
e
s
h
o
ws
u
n
i
m
o
d
a
l
d
is
t
r
i
b
u
tio
n
w
h
i
l
e
t
h
e
r
e
al
a
n
d
p
h
o
t
o
f
a
ce
s
i
n
d
i
c
at
e
b
i
m
o
d
a
l
d
i
s
t
r
i
b
u
ti
o
n
s
.
Fig
u
r
e
6
.
Su
b
s
u
r
f
ac
e
im
ag
e,
I
m
g
9
0
o
f
r
ea
l a
n
d
f
a
k
e
f
ac
es’
s
tatis
tical
an
aly
s
i
s
Fig
u
r
e
7
.
His
to
g
r
am
s
o
f
th
e
S
1
an
d
I
m
g
9
0
im
ag
es
4
.
2
.
T
he
bim
o
da
lity
co
ef
f
ici
ent
(
B
C)
T
o
m
ea
s
u
r
e
m
o
d
ality
o
f
a
d
i
s
tr
ib
u
tio
n
,
th
e
b
im
o
d
ality
co
ef
f
icien
t
(
B
C
)
was
u
s
ed
as
t
h
e
s
ec
o
n
d
em
asu
r
e
in
th
is
s
tu
d
y
.
Valu
es
o
f
th
e
B
C
r
an
g
ed
f
r
o
m
0
t
o
1
.
An
y
em
p
ir
ical
v
alu
es
th
at
u
s
m
o
r
e
th
a
n
0
.
5
5
5
s
u
g
g
esti
n
g
b
im
o
d
al
d
is
tr
ib
u
ti
o
n
[
SAS
I
n
s
titu
te,
(
1
9
8
9
)
]
.
Oth
er
wis
e,
th
e
m
o
d
ality
o
f
d
is
tr
ib
u
tio
n
is
class
if
ied
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
F
a
ce
s
p
o
o
fin
g
d
etec
tio
n
u
s
in
g
s
u
r
fa
ce
a
n
d
s
u
b
-
s
u
r
fa
ce
r
eflec
tio
n
s
a
n
a
lysi
s
(
A
z
im
Za
lih
a
A
b
d
A
z
iz
)
195
as
u
n
im
o
d
al.
T
h
e
b
im
o
d
ality
co
ef
f
icien
t
alg
o
r
ith
m
was
m
ea
s
u
r
ed
u
s
in
g
(
1
)
as
p
r
o
p
o
s
ed
b
y
[
2
5
]
.
Mo
d
ality
o
f
d
is
tr
ib
u
tio
n
s
f
o
r
all
S
1
an
d
I
m
g
90
im
ag
es in
th
e
FaPs
an
d
Face
-
iPad
d
atasets
wer
e
m
ea
s
u
r
ed
an
d
t
h
e
r
esu
lts
ar
e
co
m
p
ar
ed
in
Fig
u
r
e
8
.
(
1
)
W
h
er
e
=
s
k
ewn
ess
(
x
,
0
)
;
=
k
u
r
to
s
is
(
x
,
0
)
–
3
; a
n
d
n
=
s
am
p
l
e
s
ize
.
Fig
u
r
e
8
.
S1
a
n
d
I
m
g
9
0
p
r
e
d
ic
tio
n
s
co
r
es b
ased
o
n
th
e
B
C
Fro
m
th
e
d
ata
in
Fig
u
r
e
8
,
to
t
al
n
u
m
b
er
o
f
f
ak
e
f
ac
es
wer
e
4
5
co
n
s
is
tin
g
o
f
3
7
p
r
in
te
d
p
h
o
to
s
an
d
8
iPad
d
is
p
lay
ed
f
ac
es.
Ov
er
all,
m
ajo
r
ity
o
f
th
e
f
ak
e
f
ac
es
wer
e
s
h
o
wn
to
h
av
e
u
n
im
o
d
al
d
is
tr
ib
u
tio
n
s
an
d
h
av
e
b
ee
n
id
en
tifie
d
as
g
en
u
in
e
f
a
ce
s
.
T
r
u
e
p
o
s
itiv
e
r
at
e
(
T
PR
)
an
d
f
alse
p
o
s
itiv
e
r
ate
(
FP
R
)
wer
e
ca
lcu
lated
b
y
u
s
in
g
(
2
)
a
n
d
th
e
a
cc
u
r
ac
y
r
ate
was
m
ea
s
u
r
ed
u
s
in
g
(
3
)
.
T
ab
le
1
p
r
o
v
id
es
th
e
d
etec
tio
n
r
ates
o
f
th
e
B
C
alg
o
r
ith
m
f
o
r
b
o
th
S
1
a
n
d
I
m
g
90
d
is
tr
ib
u
tio
n
s
.
As
s
h
o
w
n
in
T
ab
le
1
,
th
e
ac
cu
r
ac
y
r
ates
o
f
th
e
B
C
f
o
r
b
o
th
S
1
an
d
I
m
g
90
d
is
tr
ib
u
tio
n
s
ar
e
6
2
.
2
0
%
in
wh
ich
m
o
r
e
th
an
h
alf
o
f
th
e
f
ak
e
f
ac
es’
d
is
tr
ib
u
tio
n
s
wer
e
lab
elled
as
u
n
im
o
d
al.
B
ased
o
n
t
h
e
an
aly
s
is
r
esu
lts
,
s
u
r
f
ac
e
an
d
s
u
b
s
u
r
f
ac
e
d
is
tr
ib
u
tio
n
s
m
o
d
a
lity
u
s
in
g
th
e
B
C
alg
o
r
ith
m
,
s
ee
m
n
o
t
p
o
s
s
ib
le
to
b
e
u
s
ed
as o
n
e
o
f
th
e
f
ac
e
s
p
o
o
f
in
g
d
etec
tio
n
p
ar
am
eter
s
.
(
)
=
∑
(
)
/
∑
(
)
=
∑
(
)
/
∑
(
2
)
(
)
=
TP
+
TN
TP
+
FN
+
FP
+
TN
(
3
)
T
ab
le
1
.
Dete
ctio
n
r
ates o
f
th
e
B
C
f
o
r
S1
an
d
I
m
g
9
0
TPR
FPR
A
c
c
u
r
a
c
y
(
A
C
)
i
n
p
e
r
c
e
n
t
a
g
e
(
%)
S
1
9
1
.
8
9
6
2
.
2
2
6
2
.
2
0
I
mg
90
8
9
.
1
9
6
0
.
0
0
6
2
.
2
0
5.
CO
NCLU
SI
O
N
As
s
tated
in
th
e
I
n
tr
o
d
u
ctio
n
,
m
u
lti
r
ef
lectio
n
s
p
r
o
p
er
ties
o
f
r
ea
l
f
ac
e
an
d
f
ak
e
f
ac
es
co
u
ld
b
e
m
an
ip
u
lated
as
o
n
e
class
if
icatio
n
m
eth
o
d
in
s
p
o
o
f
in
g
f
ac
e
d
etec
tio
n
.
Su
r
f
ac
e
an
d
s
u
b
-
s
u
r
f
ac
e
r
ef
lectio
n
o
f
ea
ch
m
ater
ial
wer
e
d
if
f
er
en
ti
ated
u
s
in
g
p
o
lar
ized
li
g
h
t.
H
o
wev
er
,
th
e
f
in
d
in
g
s
s
h
o
wed
th
at
r
ea
l
f
ac
es
an
d
p
r
in
ted
p
h
o
to
f
ac
es
co
n
s
is
t
s
i
m
ilar
r
ef
lectio
n
s
p
r
o
p
er
ties
.
Du
e
to
th
ese
s
im
ilar
ities
,
it
is
d
i
f
f
icu
lt
to
d
is
tin
g
u
is
h
b
etwe
en
r
ea
l
f
ac
es
a
n
d
p
r
in
ted
p
h
o
to
f
ac
es
b
ased
o
n
S
1
a
n
d
I
m
g
90
im
ag
es.
On
th
e
o
th
e
r
h
an
d
,
iPad
attac
k
s
ca
n
b
e
ea
s
ily
d
etec
ted
o
n
ly
b
y
th
e
m
ea
n
v
alu
es
o
f
S
1
an
d
I
m
g
90
.
Mo
r
eo
v
er
,
d
en
s
ity
o
f
th
e
iPad
d
is
tr
ib
u
tio
n
s
m
o
d
e
was
also
s
ig
n
if
ican
tly
d
if
f
er
en
t
f
r
o
m
th
e
g
en
u
in
e
f
ac
es.
T
h
e
r
e
s
u
l
ts
o
f
t
h
i
s
s
t
u
d
y
w
il
l
n
o
w
b
e
c
o
m
p
a
r
e
d
t
o
t
h
e
f
i
n
d
i
n
g
s
o
f
p
r
e
v
i
o
u
s
w
o
r
k
s
:
c
l
a
s
s
i
f
i
ca
t
i
o
n
b
e
t
w
e
e
n
p
l
as
ti
c
an
d
a
l
u
m
i
n
i
u
m
o
b
j
e
c
t
;
a
n
d
c
l
a
s
s
i
f
i
c
a
ti
o
n
b
e
t
w
e
e
n
t
r
a
n
s
p
a
r
e
n
t
a
n
d
o
p
a
q
u
e
o
b
j
e
c
ts
.
B
o
t
h
r
es
u
l
ts
m
e
as
u
r
e
d
t
h
e
d
eg
r
e
e
o
f
p
o
l
a
r
i
z
a
t
i
o
n
(
DO
P
)
v
a
lu
e
s
o
f
t
h
e
m
a
t
e
r
i
a
ls
.
S
e
v
e
r
a
l
p
o
s
s
i
b
le
e
x
p
l
a
n
a
ti
o
n
s
c
a
n
b
e
l
is
t
e
d
f
o
r
t
h
es
e
r
es
u
l
ts
.
T
h
e
p
o
l
a
r
i
z
a
t
i
o
n
o
f
t
h
e
r
e
f
l
e
c
te
d
l
i
g
h
t
d
e
p
e
n
d
s
o
n
t
h
e
p
r
o
p
e
r
t
i
es
o
f
t
h
e
o
b
je
c
t
s
u
r
f
a
c
e
.
I
n
a
c
c
o
r
d
a
n
c
e
t
o
t
h
e
s
u
r
f
a
c
e
p
r
o
p
e
r
t
i
es
,
t
h
e
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
t
r
a
n
s
p
a
r
e
n
t
a
n
d
o
p
a
q
u
e
o
b
j
e
c
t
s
i
s
p
h
y
s
ic
a
ll
y
o
b
v
i
o
u
s
.
T
r
a
n
s
p
a
r
e
n
t
o
b
j
e
c
t
s
u
c
h
a
s
g
l
a
s
s
n
o
t
o
n
l
y
a
l
l
o
w
s
li
g
h
t
t
o
t
r
a
n
s
m
i
t
b
u
t
a
l
s
o
r
e
f
l
e
c
ts
li
g
h
t
.
T
h
e
r
e
f
l
e
c
tio
n
s
f
r
o
m
a
t
r
a
n
s
p
a
r
e
n
t
o
b
j
ec
t
h
a
v
e
b
e
e
n
r
e
p
o
r
t
e
d
t
o
b
e
m
o
s
tl
y
s
p
e
c
u
l
a
r
w
h
il
e
a
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
2
4
,
No
.
1
,
Octo
b
er
20
21
:
1
8
9
-
19
7
196
o
p
a
q
u
e
o
b
j
e
c
t
’
s
r
e
f
l
e
c
ti
o
n
s
a
r
e
d
i
f
f
u
s
e
d
.
As
b
ee
n
d
is
cu
s
s
ed
in
s
ec
tio
n
2
,
th
e
iPad
s
cr
ee
n
was
m
ad
e
o
f
g
la
ss
wh
ich
is
an
in
s
u
lato
r
.
Prin
ted
p
h
o
to
s
wer
e
o
b
v
io
u
s
ly
in
s
u
l
ato
r
s
.
Ap
ar
t
f
r
o
m
t
h
at,
th
e
th
r
ee
m
ater
ials
wer
e
o
p
aq
u
e
o
b
jects
alth
o
u
g
h
t
h
e
i
Pad
s
cr
ee
n
was
m
ad
e
o
f
g
las
s
.
An
o
th
er
s
im
ilar
ity
b
etwe
en
th
e
r
ea
l
f
ac
e
an
d
p
r
in
ted
p
h
o
to
is
th
at
b
o
t
h
o
f
th
em
p
r
o
d
u
ce
two
ty
p
es
o
f
r
e
f
lectio
n
s
:
s
p
ec
u
lar
an
d
d
if
f
u
s
e.
I
n
c
o
n
tr
ast,
iPad
d
o
es
n
o
t
r
ef
lect
lig
h
t
b
u
t
em
its
its
o
wn
lig
h
t.
Desp
ite
th
e
s
e
s
im
ilar
ities
,
s
tati
s
tically
,
p
r
in
ted
p
h
o
to
s
we
r
e
u
n
ab
le
to
b
e
d
is
tin
g
u
is
h
ed
f
r
o
m
th
e
g
en
u
in
e
f
ac
es b
y
u
s
in
g
b
o
th
s
u
r
f
ac
e
an
d
s
u
b
-
s
u
r
f
ac
e
r
ef
lectio
n
.
RE
F
E
R
E
NC
E
S
[1
]
G
.
P
a
n
,
Z.
W
u
a
n
d
L.
S
u
n
,
“
Li
v
e
n
e
ss
De
tec
ti
o
n
fo
r
F
a
c
e
Re
c
o
g
n
it
i
o
n
,
”
Re
c
e
n
t
A
d
v
a
n
c
e
s
i
n
F
a
c
e
Re
c
o
g
n
it
i
o
n
,
Kre
sim
ir
De
lac
,
M
islav
Grg
ic
a
n
d
M
a
rian
S
tew
a
rt
Ba
rtl
e
tt
,
In
tec
h
O
p
e
n
,
2
0
0
8
.
A
v
a
il
a
b
le:
h
tt
p
s:/
/www
.
in
tec
h
o
p
e
n
.
c
o
m
/b
o
o
k
s/re
c
e
n
t_
a
d
v
a
n
c
e
s_
in
_
fa
c
e
_
re
c
o
g
n
it
i
o
n
/l
iv
e
n
e
ss
_
d
e
tec
ti
o
n
_
f
o
r_
fa
c
e
_
re
c
o
g
n
it
i
o
n
d
o
i:
1
0
.
5
7
7
2
/6
3
9
7
.
[2
]
N.
Ko
s
e
a
n
d
J.
L.
Du
g
e
lay
,
“
Re
flec
tan
c
e
An
a
ly
sis
Ba
se
d
Co
u
n
term
e
a
su
re
Tec
h
n
iq
u
e
to
De
t
e
c
t
F
a
c
e
M
a
s
k
Attac
k
s,”
in
18
th
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
Dig
it
a
l
S
ig
n
a
l
Pro
c
e
ss
in
g
IE
EE
,
J
u
ly
2
0
1
3
,
p
p
.
1
-
6
,
d
o
i:
1
0
.
1
1
0
9
/ICD
S
P
.
2
0
1
3
.
6
6
2
2
7
0
4
.
[3
]
Y.
Kim
,
J.
Na
,
S
.
Y
o
o
n
,
a
n
d
J.
Yi,
“
M
a
sk
e
d
fa
k
e
fa
c
e
d
e
tec
ti
o
n
u
si
n
g
ra
d
ian
c
e
m
e
a
su
re
m
e
n
ts,”
J
o
u
rn
a
l
o
f
th
e
Op
ti
c
a
l
S
o
c
iety
o
f
Ame
ric
a
,
v
o
l
.
2
6
,
n
o
.
4
,
p
p
.
7
6
0
-
7
6
6
,
2
0
0
9
,
d
o
i:
1
0
.
1
3
6
4
/JOSAA.2
6
.
0
0
0
7
6
0
.
[4
]
Z.
Zh
a
n
g
,
D.
Yi,
Z.
Lei
a
n
d
S
.
Z
.
Li
,
“
F
a
c
e
li
v
e
n
e
ss
d
e
tec
ti
o
n
b
y
le
a
rn
in
g
m
u
lt
isp
e
c
tral
re
flec
tan
c
e
d
istri
b
u
t
io
n
s,”
in
Au
to
m
a
ti
c
F
a
c
e
a
n
d
Ge
stu
re
Rec
o
g
n
it
io
n
W
o
rk
sh
o
p
s
,
2
0
1
1
,
p
p
.
4
3
6
-
4
4
1
,
d
o
i:
1
0
.
1
1
0
9
/
F
G
.
2
0
1
1
.
5
7
7
1
4
3
8
.
[5
]
Y.
Ka
n
z
a
wa
,
Y.
Kim
u
ra
a
n
d
T.
Na
it
o
,
“
Hu
m
a
n
S
k
i
n
De
tec
ti
o
n
b
y
Visib
le
a
n
d
Ne
a
r
-
in
fra
re
d
Im
a
g
in
g
,
”
i
n
IA
PR
Co
n
fer
e
n
c
e
o
n
M
a
c
h
i
n
e
Vi
sio
n
A
p
p
li
c
a
ti
o
n
s,
v
o
l
.
1
2
,
2
0
1
1
,
p
p
.
1
4
-
2
2
.
[6
]
L.
B.
W
o
lf,
“
Us
in
g
p
o
lariz
a
ti
o
n
t
o
se
p
a
ra
te
re
flec
ti
o
n
c
o
m
p
o
n
e
n
ts,
”
in
Co
mp
u
ter
S
o
c
iety
Co
n
fer
e
n
c
e
o
n
Co
m
p
u
ter
Vi
sio
n
a
n
d
P
a
tt
e
rn
Rec
o
g
n
it
i
o
n
(
CVP
R)
IEE
E
,
p
p
.
3
6
3
-
3
6
9
,
1
9
8
9
,
d
o
i:
1
0
.
1
1
0
9
/CVP
R.
1
9
8
9
.
3
7
8
7
3
.
[7
]
L.
B.
W
o
lf,
"
P
o
lariz
a
ti
o
n
-
b
a
se
d
M
a
teria
l
Clas
sifica
ti
o
n
fr
o
m
S
p
e
c
u
lar
Re
flec
ti
o
n
,
"
i
n
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Pa
tt
e
rn
A
n
a
lys
is a
n
d
M
a
c
h
i
n
e
In
t
e
ll
ig
e
n
c
e
1
2
.
1
1
,
p
p
.
1
0
5
9
-
1
0
7
1
,
1
9
9
0
,
d
o
i:
1
0
.
1
1
0
9
/3
4
.
6
1
7
0
5
.
[8
]
M
.
S
a
rk
a
r
e
t
a
l
.
,
“
In
teg
ra
ted
P
o
l
a
riza
ti
o
n
A
n
a
ly
z
in
g
CM
OS
Im
a
g
e
S
e
n
s
o
r
f
o
r
M
a
teria
l
Clas
sific
a
ti
o
n
,
”
in
IEE
E
S
e
n
so
rs
J
o
u
rn
a
l
,
v
o
l.
1
1
,
n
o
.
8
,
p
p
.
1
6
9
2
-
1
7
0
3
,
2
0
1
1
,
d
o
i:
1
0
.
1
1
0
9
/JS
EN.
2
0
1
0
.
2
0
9
5
0
0
3
.
[9
]
A.
M
a
h
e
n
d
r
u
a
n
d
M
.
S
a
rk
a
r,
“
Bi
o
-
in
sp
ired
o
b
jec
t
c
las
sifica
ti
o
n
u
sin
g
p
o
lariz
a
ti
o
n
ima
g
i
n
g
,
”
i
n
S
ixth
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
e
n
si
n
g
T
e
c
h
n
o
l
o
g
y
(ICS
T
)
IEE
E,
p
p
.
2
0
7
-
2
1
2
,
2
0
1
2
,
d
o
i:
1
0
.
1
1
0
9
/IC
S
e
n
sT.
2
0
1
2
.
6
4
6
1
6
7
2
.
[1
0
]
C.
S
o
-
Li
n
g
a
n
d
L.
Li
n
g
,
“
A
M
u
lt
i
-
lay
e
re
d
Re
flec
ti
o
n
M
o
d
e
l
o
f
Na
tu
ra
l
Hu
m
a
n
S
k
in
,
”
i
n
Co
m
p
u
ter
Gr
a
p
h
ics
In
ter
n
a
t
io
n
a
l
I
EE
E
,
2
0
0
1
,
d
o
i:
1
0
.
1
1
0
9
/CG
I.
2
0
0
1
.
9
3
4
6
8
1
.
[1
1
]
S
.
L
.
Ja
c
q
u
e
s,
J.
C.
Ra
m
e
ll
a
-
Ro
m
a
n
a
n
d
K.
Lee
,
“
Im
a
g
i
n
g
S
k
in
P
a
th
o
l
o
g
y
wit
h
P
o
lariz
e
d
Li
g
h
t,
”
in
J
o
u
r
n
a
l
o
f
Bi
o
me
d
ica
l
Op
ti
c
s
7
,
p
p
.
3
2
9
-
3
4
0
,
2
0
0
2
,
d
o
i:
1
0
.
1
1
1
7
/1
.
1
4
8
4
4
9
8
.
[1
2
]
Bin
Zh
o
u
,
Jia
n
h
u
a
Xu
a
n
,
Ho
n
g
z
h
i
Zh
a
o
,
G
lo
ria
J.
Ch
e
p
k
o
,
M
a
tt
h
e
w
T.
F
re
e
d
m
a
n
a
n
d
Ke
v
in
Yin
g
y
i
n
Zo
u
,
“
P
o
lariz
a
ti
o
n
Im
a
g
i
n
g
fo
r
Bre
a
st
Ca
n
c
e
r
Dia
g
n
o
sis
u
si
n
g
Tex
t
u
re
An
a
ly
sis
a
n
d
S
VM,
”
i
n
L
if
e
S
c
ien
c
e
S
y
ste
ms
a
n
d
Ap
p
li
c
a
ti
o
n
s (L
I
S
A)
W
o
rk
sh
o
p
I
E
EE
,
p
p
.
2
1
7
-
2
2
0
,
2
0
0
7
,
d
o
i
:
1
0
.
1
1
0
9
/L
S
S
A.2
0
0
7
.
4
4
0
0
9
2
3
.
[1
3
]
Eu
i
J
o
n
g
Ba
e
,
S
o
o
Ho
n
g
S
e
o
,
Y
o
u
n
g
Ch
u
l
K
y
e
a
n
d
H
y
o
Hy
u
n
Ah
n
,
“
A
Qu
a
n
ti
tativ
e
As
se
ss
m
e
n
t
o
f
t
h
e
Hu
m
a
n
S
k
in
S
u
rfa
c
e
u
si
n
g
P
o
lariz
e
d
Li
g
h
t
Dig
i
tal
P
h
o
to
g
ra
p
h
y
a
n
d
it
s
D
e
rm
a
to
lo
g
ic
S
i
g
n
if
ica
n
c
e
,
”
in
S
k
i
n
Res
e
a
rc
h
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
1
6
,
n
o
.
3
,
p
p
.
2
7
0
-
2
7
4
,
2
0
1
0
,
d
o
i:
1
0
.
1
1
1
1
/j
.
1
6
0
0
-
0
8
4
6
.
2
0
1
0
.
0
0
4
3
9
.
x
.
[1
4
]
A.
M
a
tsu
b
a
ra
,
“
Diffe
re
n
c
e
s in
th
e
S
u
rfa
c
e
a
n
d
S
u
b
s
u
rfa
c
e
Re
flec
ti
o
n
Ch
a
ra
c
teristics
o
f
F
a
c
ial
S
k
in
b
y
Ag
e
G
ro
u
p
,
”
in
S
k
i
n
Res
e
a
rc
h
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
8
,
n
o
.
1
,
p
p
.
2
9
-
3
5
,
2
0
1
2
,
d
o
i:
1
0
.
1
1
1
1
/
j.
1
6
0
0
-
0
8
4
6
.
2
0
1
1
.
0
0
5
3
7
.
x
.
[1
5
]
Li
n
g
Li
,
L
in
g
Ng
a
n
d
Ca
rm
e
n
S
o
,
“
A
P
h
y
sic
a
ll
y
-
b
a
se
d
Hu
m
a
n
S
k
in
Re
flec
ti
o
n
M
o
d
e
l
,
”
in
W
o
rl
d
S
c
ien
ti
fi
c
a
n
d
En
g
i
n
e
e
rin
g
Aca
d
e
my
a
n
d
S
o
c
iety
(W
S
EA
S
)
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
,
2
0
0
9
.
[1
6
]
Zaid
i,
Z.
,
“
Ch
a
ra
c
teristics
o
f
sk
i
n
c
o
l
o
u
r,
”
in
T
h
e
J
o
u
rn
a
l
o
f
t
h
e
P
a
k
ista
n
M
e
d
ica
l
Asso
c
i
a
ti
o
n
6
6
(
J
PM
A),
v
o
l.
6
6
,
n
o
.
8
,
p
.
9
1
4
,
2
0
1
6
.
[1
7
]
S
k
in
S
tru
c
t
u
re
,
2
0
2
0
.
A
v
a
il
a
b
le
fro
m
h
tt
p
s://
ww
w.d
re
a
m
sti
m
e
.
c
o
m
/h
u
m
a
n
-
sk
in
-
stru
c
tu
re
-
v
e
c
to
r
-
il
lu
stra
ti
o
n
-
iso
late
d
-
b
a
c
k
g
ro
u
n
d
-
h
u
m
a
n
-
sk
i
n
-
stru
c
tu
re
-
v
e
c
to
r
-
i
ll
u
stra
ti
o
n
-
is
o
lat
e
d
-
wh
it
e
-
b
a
c
k
g
ro
u
n
d
-
ima
g
e
1
2
8
7
3
9
8
8
7
.
[1
8
]
Brit
a
n
n
ica
,
T.
E
.
o
.
E
.
,
“
P
a
p
e
r,
”
A
c
c
e
ss
e
d
:
F
e
b
ru
a
ry
5,
2
0
1
7
.
[
On
li
n
e
]
.
Av
a
il
a
b
le:
h
tt
p
s:/
/www
.
b
rit
a
n
n
ica
.
c
o
m
/t
e
c
h
n
o
lo
g
y
/
p
a
p
e
r
[1
9
]
Brit
t
,
K.
W.
,
“
P
a
p
e
rm
a
k
in
g
.
”
En
c
y
c
lo
p
æd
ia
Br
it
a
n
n
ica
,
En
c
y
c
l
o
p
æd
ia
Brit
a
n
n
ica
,
I
n
c
.
,
Av
a
il
a
b
le:
ww
w.b
rit
a
n
n
ica
.
c
o
m
/t
e
c
h
n
o
l
o
g
y
/
p
a
p
e
rm
a
k
in
g
[2
0
]
M
a
tt
.
,
“
T
h
e
Diffe
re
n
c
e
Be
twe
e
n
M
a
tt
a
n
d
S
e
m
i
G
lo
ss
F
in
i
sh
P
h
o
t
o
P
a
p
e
r,
”
2
0
1
2
.
[
On
li
n
e
].
Av
a
il
a
b
le:
h
tt
p
:
//
ww
w.p
h
o
t
o
p
a
p
e
r
d
irec
t.
c
o
m
/b
lo
g
/?p
=
4
0
5
[2
1
]
J.
Yu
le
a
n
d
W.
Nie
lse
n
,
“
Th
e
p
e
n
e
tratio
n
o
f
l
ig
h
t
in
t
o
p
a
p
e
r
a
n
d
it
s
e
ffe
c
t
o
n
h
a
lft
o
n
e
re
p
ro
d
u
c
ti
o
n
,
”
in
:
Pro
c
.
T
AGA
,
p
p
.
6
5
-
7
6
,
1
9
5
1
.
[2
2
]
A.
Z.
A.
Az
iz,
W.
Ho
n
g
a
n
d
J.
F
e
rry
m
a
n
,
J
.
,
“
F
a
c
e
An
ti
-
s
p
o
o
fin
g
C
o
u
n
term
e
a
s
u
re
:
Eff
icie
n
t
2
D
M
a
teria
ls
Clas
sifica
ti
o
n
u
si
n
g
P
o
lariz
a
ti
o
n
Im
a
g
in
g
,
”
i
n
IEE
E
2
0
1
7
5
th
I
n
ter
n
a
ti
o
n
a
l
W
o
rk
sh
o
p
o
n
Bi
o
me
trics
a
n
d
F
o
re
n
sic
s
(IW
BF
),
p
p
.
1
-
6
,
2
0
1
7
,
d
o
i:
1
0
.
1
1
0
9
/IW
B
F
.
2
0
1
7
.
7
9
3
5
1
0
5
.
[2
3
]
A.
Z.
A.
Az
iz,
“
Eff
e
c
ts
o
f
Visib
le
a
n
d
Ne
a
r
In
fra
re
d
P
o
lariz
e
d
Li
g
h
ts
o
n
S
p
o
o
fin
g
F
a
c
e
De
tec
ti
o
n
,
”
in
J
o
u
rn
a
l
o
f
Co
mp
u
ter
S
c
ien
c
e
,
v
o
l.
1
5
.
p
p
.
2
8
8
-
3
0
1
,
2
0
1
9
,
d
o
i:
1
0
.
3
8
4
4
/j
c
ss
p
.
2
0
1
9
.
2
8
8
.
3
0
1
.
[2
4
]
A.
Z.
A.
Az
iz
a
n
d
W.
Ho
n
g
,
“
P
o
lariz
a
ti
o
n
Im
a
g
i
n
g
f
o
r
F
a
c
e
S
p
o
o
fin
g
De
tec
ti
o
n
:
I
d
e
n
ti
f
ica
ti
o
n
o
f
Blac
k
Et
h
n
ica
l
G
ro
u
p
,
”
in
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
C
o
mp
u
ta
ti
o
n
a
l
A
p
p
r
o
a
c
h
in
S
ma
rt
S
y
ste
ms
De
sig
n
a
n
d
Ap
p
li
c
a
ti
o
n
s
(ICAS
S
DA
),
p
p
.
1
-
6
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/ICAS
S
DA
.
2
0
1
8
.
8
4
7
7
6
0
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
F
a
ce
s
p
o
o
fin
g
d
etec
tio
n
u
s
in
g
s
u
r
fa
ce
a
n
d
s
u
b
-
s
u
r
fa
ce
r
eflec
tio
n
s
a
n
a
lysi
s
(
A
z
im
Za
lih
a
A
b
d
A
z
iz
)
197
[2
5
]
R.
P
fister,
K.A.
S
c
h
wa
rz
,
M
.
Ja
n
c
z
y
k
,
R.
Da
le,
J.B.
F
re
e
m
a
n
,
"
G
o
o
d
T
h
i
n
g
s
P
e
a
k
in
P
a
irs:
a
No
te
o
n
th
e
Bim
o
d
a
li
ty
Co
e
fficie
n
t
,
"
F
ro
n
ti
e
r
s in
P
s
y
c
h
o
l
o
g
y
,
v
o
l.
4
,
p
p
.
7
0
0
,
2
0
1
3
,
d
o
i:
1
0
.
3
3
8
9
/fp
s
y
g
.
2
0
1
3
.
0
0
7
0
0
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Az
im
Za
li
h
a
A
b
d
Az
iz
is
c
u
rre
n
t
ly
a
lec
tu
re
r
i
n
F
a
c
u
lt
y
o
f
I
n
f
o
rm
a
ti
c
s
&
Co
m
p
u
ti
n
g
,
Un
i
v
e
rsiti
S
u
lt
a
n
Zain
a
l
Ab
id
in
(
Un
iS
ZA),
Tere
n
g
g
a
n
u
M
a
lay
sia
.
S
h
e
r
e
c
e
iv
e
d
h
e
r
P
h
D
(Co
m
p
u
te
r
S
c
ien
c
e
)
fro
m
Un
iv
e
rsit
y
o
f
Re
a
d
in
g
,
Un
it
e
d
Kin
g
d
o
m
in
2
0
1
7
.
S
h
e
d
id
h
e
r
first
d
e
g
re
e
i
n
Co
m
p
u
ter
S
c
ien
c
e
in
2
0
0
0
a
t
F
a
c
u
lt
y
o
f
C
o
m
p
u
ter S
c
ien
c
e
,
Un
i
v
e
r
siti
Tek
n
o
l
o
g
i
M
a
lay
sia
.
Pro
f.
M
o
h
d
Riz
o
n
J
u
h
a
r
i
g
ra
d
u
a
ted
Do
c
to
r
o
f
En
g
i
n
e
e
rin
g
a
t
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
In
tel
li
g
e
n
c
e
S
y
ste
m
in
Oi
ta
Un
iv
e
rsity
,
Ja
p
a
n
.
He
wa
s
a
P
ro
fe
ss
o
r
o
f
B
io
m
e
d
ica
l
Tec
h
n
o
l
o
g
y
i
n
Ki
n
g
S
a
u
d
U
n
iv
e
rsity
,
S
a
u
d
i
Ara
b
ia
i
n
2
0
1
0
.
He
se
rv
e
d
a
t
M
I
EC,
F
u
z
h
o
u
Un
iv
e
rsity
,
Ch
i
n
a
in
2
0
1
9
.
C
u
rre
n
tl
y
,
h
e
is
a
P
ro
fe
ss
o
r
a
t
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
,
UCSI
U
n
iv
e
rsit
y
,
M
a
lay
sia
.
M
IET
,
M
IE
M
m
e
m
b
e
r.
Evaluation Warning : The document was created with Spire.PDF for Python.