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.
7
,
No
.
5
,
Octo
b
e
r
2
0
1
7
,
p
p
.
2
8
9
5
~
2
901
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
7
i
5
.
pp
2
8
9
5
-
2
901
2895
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
JE
C
E
An Ea
r Re
co
g
niti
o
n Metho
d Ba
sed
on
Ro
tatio
n Inv
a
ria
nt
Tra
nsfo
r
m
e
d DC
T
F
a
t
e
m
eh
H
o
ura
li
,
S
o
ra
y
y
a
G
ha
rr
a
v
i
De
p
a
rt
m
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
Esf
a
ra
y
e
n
Un
iv
e
rsity
o
f
Tec
h
n
o
lo
g
y
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
27
,
2
0
1
6
R
ev
i
s
ed
Dec
2
6
,
2
0
1
6
A
cc
ep
ted
Mar
11
,
2
0
1
7
Hu
m
a
n
re
c
o
g
n
it
io
n
sy
ste
m
s
h
a
v
e
g
a
in
e
d
g
r
e
a
t
i
m
p
o
rtan
c
e
re
c
e
n
tl
y
in
a
w
id
e
ra
n
g
e
o
f
a
p
p
li
c
a
ti
o
n
s
li
k
e
a
c
c
e
ss
,
c
o
n
tro
l
,
c
rim
in
a
l
in
v
e
stig
a
ti
o
n
a
n
d
b
o
rd
e
r
se
c
u
rit
y
.
Ear i
s an
e
m
e
r
g
in
g
b
io
m
e
tri
c
w
h
ich
h
a
s ri
c
h
a
n
d
sta
b
le stru
c
tu
re
a
n
d
c
a
n
p
o
ten
ti
a
ll
y
b
e
i
m
p
le
m
e
n
ted
r
e
li
a
b
ly
a
n
d
c
o
st
e
ff
icie
n
tl
y
.
T
h
u
s
h
u
m
a
n
e
a
r
re
c
o
g
n
it
io
n
h
a
s
b
e
e
n
re
se
a
rc
h
e
d
w
id
e
l
y
a
n
d
m
a
d
e
g
re
a
tl
y
p
ro
g
re
ss
.
Hig
h
re
c
o
g
n
it
io
n
ra
tes
w
h
ich
a
re
re
p
o
rt
e
d
in
m
o
st
e
x
isti
n
g
m
e
th
o
d
s
c
a
n
b
e
re
a
c
h
e
d
o
n
ly
u
n
d
e
r
c
lo
se
ly
c
o
n
tro
ll
e
d
c
o
n
d
it
io
n
s.
A
c
tu
a
ll
y
a
slig
h
t
a
m
o
u
n
t
o
f
ro
tatio
n
a
n
d
tran
sla
ti
o
n
w
h
ich
is
i
n
e
sc
a
p
a
b
le
w
o
u
ld
b
e
i
n
ju
r
io
u
s
fo
r
sy
ste
m
p
e
rf
o
r
m
a
n
c
e
.
In
th
is
p
a
p
e
r,
a
m
e
th
o
d
t
h
a
t
u
se
s
a
tran
s
f
o
rm
e
d
t
y
p
e
o
f
DC
T
i
s
im
p
le
m
e
n
ted
to
e
x
trac
t
m
e
a
n
in
g
fu
l
f
e
a
tu
re
s
f
ro
m
e
a
r
i
m
a
g
e
s.
T
h
is
a
lg
o
rit
h
m
is
q
u
it
e
ro
b
u
st
to
e
a
r
ro
tati
o
n
,
t
ra
n
sla
ti
o
n
a
n
d
il
l
u
m
in
a
ti
o
n
.
T
h
e
p
ro
p
o
se
d
m
e
th
o
d
is
e
x
p
e
rime
n
ted
o
n
tw
o
p
o
p
u
lar
d
a
tab
a
se
s,
i.
e
.
UST
B
I
I
a
n
d
IIT
De
lh
i
II,
w
h
ich
a
c
h
iev
e
s
sig
n
i
fica
n
t
im
p
ro
v
e
m
e
n
t
in
th
e
p
e
rf
o
rm
a
n
c
e
in
c
o
m
p
a
riso
n
to
o
th
e
r
m
e
th
o
d
s w
it
h
g
o
o
d
e
ff
icie
n
c
y
b
a
se
d
o
n
L
BP
,
DSIF
T
a
n
d
G
a
b
o
r.
A
lso
b
e
c
a
u
se
o
f
c
o
n
sid
e
rin
g
o
n
ly
i
m
p
o
rtan
t
c
o
e
f
f
icie
n
ts,
th
is
m
e
th
o
d
is
f
a
ste
r
c
o
m
p
a
re
d
to
o
t
h
e
r
m
e
th
o
d
s.
K
ey
w
o
r
d
s
:
Dis
cr
ete
co
s
in
e
tr
a
n
s
f
o
r
m
E
ar
r
ec
o
g
n
itio
n
Featu
r
e
ex
tr
ac
tio
n
I
m
ag
e
p
r
o
ce
s
s
i
n
g
No
r
m
a
lizatio
n
Co
p
y
rig
h
t
©
2
0
1
7
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
:
Fate
m
e
h
Ho
u
r
ali
,
Dep
ar
t
m
en
t o
f
E
lectr
ical
an
d
C
o
m
p
u
ter
E
n
g
in
ee
r
i
n
g
,
E
s
f
ar
a
y
e
n
U
n
i
v
er
s
it
y
o
f
T
ec
h
n
o
lo
g
y
,
A
za
d
eg
a
n
Stre
et,
Ma
d
ar
Sq
u
ar
e,
E
s
f
ar
a
y
e
n
,
No
r
th
K
h
o
r
asan
,
I
r
an
.
E
m
ail:
Ho
u
r
ali@
e
s
f
ar
a
y
e
n
.
ac
.
ir
1.
I
NT
RO
D
UCT
I
O
N
B
io
m
etr
ic
s
is
t
h
e
ev
al
u
atio
n
an
d
an
al
y
s
i
s
o
f
ap
p
ea
r
an
ce
o
r
b
eh
av
io
u
r
al
s
p
ec
i
f
icatio
n
to
r
ec
o
g
n
ize
id
en
tit
y
o
f
a
p
er
s
o
n
.
E
ac
h
id
e
n
ti
f
icatio
n
s
y
s
te
m
o
p
er
ates
i
n
t
w
o
s
ep
ar
ate
m
o
d
es:
r
eg
i
s
tr
ati
o
n
an
d
v
er
i
f
icatio
n
m
o
d
e
s
.
I
n
th
e
r
eg
is
tr
atio
n
m
o
d
e,
n
ec
ess
ar
y
i
n
f
o
r
m
a
tio
n
a
n
d
f
ea
t
u
r
es
ar
e
ex
tr
ac
ted
f
r
o
m
ea
ch
s
u
b
j
ec
t
an
d
r
eser
v
ed
in
d
atab
ase.
T
h
en
th
e
ex
tr
ac
ted
f
ea
t
u
r
es o
f
ea
ch
p
er
s
o
n
ar
e
lab
elled
w
it
h
h
er
o
r
h
i
s
p
er
s
o
n
alit
y
.
I
n
v
er
i
f
icatio
n
m
o
d
e,
th
e
b
i
o
m
e
tr
ic
s
y
s
te
m
ex
tr
ac
t
s
f
ea
t
u
r
es
f
r
o
m
th
e
test
s
u
b
j
ec
t.
Fin
all
y
a
n
ap
p
r
o
p
r
iate
class
i
f
ier
u
s
i
n
g
t
h
e
r
e
g
is
ter
ed
f
ea
t
u
r
es
o
f
t
h
e
clai
m
ed
id
e
n
tit
y
i
n
th
e
d
atab
ase
d
eter
m
i
n
es
t
h
e
id
en
tit
y
o
f
t
h
e
s
u
b
j
ec
t [
1
]
.
T
h
e
p
o
p
u
lar
s
tag
es
d
u
r
i
n
g
r
eg
is
tr
atio
n
an
d
v
er
if
icatio
n
a
r
e
i
m
ag
e
ac
q
u
is
i
tio
n
,
o
b
j
ec
t
d
etec
tio
n
,
p
r
o
ce
s
s
in
g
,
f
ea
t
u
r
e
e
x
tr
ac
tio
n
an
d
clas
s
if
icatio
n
.
Am
o
n
g
all
p
o
s
s
ib
le
b
io
m
etr
ics,
ea
r
b
ec
a
u
s
e
o
f
its
ad
v
an
tag
e
s
is
co
n
s
id
er
ed
as
a
r
eliab
le
b
io
m
etr
ic.
D
u
r
in
g
h
u
m
a
n
li
f
e
t
h
e
s
tr
u
ct
u
r
e
o
f
e
x
ter
n
al
ea
r
i
s
al
m
o
s
t
co
n
s
ta
n
t.
A
ls
o
f
ac
ial
e
x
p
r
ess
io
n
s
a
n
d
p
ar
tial
f
ac
e
o
cc
l
u
s
io
n
s
u
ch
as
e
x
i
s
ten
ce
o
f
g
las
s
es
a
n
d
h
air
d
o
es
n
o
t
a
f
f
ec
t
th
e
ef
f
icien
c
y
o
f
th
e
ea
r
r
ec
o
g
n
iti
o
n
s
y
s
te
m
.
I
n
ad
d
itio
n
to
t
h
es
e
ad
v
an
ta
g
es
i
f
t
w
o
p
er
s
o
n
s
h
av
e
s
i
m
ilar
f
ac
es,
ce
r
tain
l
y
th
e
ir
ea
r
s
w
ill
b
e
d
if
f
er
en
t.
Fu
r
t
h
er
m
o
r
e
i
f
i
m
a
g
es
o
f
f
ac
e
p
r
o
f
ile
ex
is
t,
E
ar
ca
n
b
e
ex
tr
ac
ted
f
r
o
m
th
e
m
w
h
ic
h
m
ak
e
s
co
s
t lo
w
a
n
d
ti
m
e
ca
n
b
e
s
av
ed
[
2
]
.
I
n
th
e
ea
r
l
y
w
ee
k
s
o
f
th
e
l
if
e
ti
m
e,
th
e
ea
r
ap
p
ea
r
s
w
i
th
t
h
e
s
t
r
u
ctu
r
e
o
f
s
ix
s
ep
ar
ate
h
illo
ck
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
5
,
Octo
b
er
201
7
:
2
8
9
5
–
2
9
0
1
2896
Star
tin
g
f
r
o
m
t
h
e
y
ea
r
1
8
8
2
t
h
er
e
h
av
e
b
ee
n
m
a
n
y
t
h
eo
r
ies
,
s
o
m
e
o
f
th
e
m
p
r
o
p
o
s
ed
b
y
S
t
r
ee
ter
[
3
]
,
Dav
is
[
4
]
,
I
-
C
h
u
an
[
5
]
,
Nix
o
n
[
6
]
,
etc.
P
u
r
s
u
a
n
t
to
th
e
s
e
s
t
u
d
ies,
d
u
e
to
ea
r
’
s
s
p
ec
ial
s
h
ap
e
an
d
ap
p
ea
r
an
ce
,
it
is
u
n
iq
u
e
f
o
r
h
u
m
a
n
r
ec
o
g
n
iti
o
n
.
T
h
e
g
r
o
w
t
h
o
f
ea
r
b
et
w
ee
n
4
m
o
n
t
h
s
to
8
y
ea
r
s
o
ld
is
l
in
ea
r
an
d
af
ter
th
at
i
t
w
ill
b
e
ap
p
r
o
x
i
m
ate
l
y
co
n
s
ta
n
t a
n
d
d
u
r
i
n
g
t
h
e
li
f
eti
m
e
d
o
esn
’
t c
h
an
g
e
[
7
-
1
4
]
.
A
lp
h
o
n
s
e
B
er
till
o
n
w
as
t
h
e
f
ir
s
t
p
er
s
o
n
w
h
o
u
s
ed
ea
r
as
a
b
io
m
etr
ic.
He
u
s
ed
e
ar
’
s
s
h
ap
e
d
ep
en
d
en
ce
o
f
th
e
lo
b
e
an
d
s
alien
ce
an
d
ea
r
h
ei
g
h
t
as
s
o
m
e
o
f
t
h
e
f
ac
to
r
s
f
o
r
ea
r
r
ec
o
g
n
i
tio
n
[
1
5
]
.
A
lf
r
ed
I
an
n
ar
elli
[
7
]
p
r
o
v
id
ed
a
d
atab
ase
u
s
i
n
g
o
v
er
1
0
,
0
0
0
ea
r
im
a
g
es
an
d
b
y
tes
tin
g
,
h
e
i
n
d
icate
d
th
at
ea
r
co
u
ld
b
e
u
s
ed
as
a
r
el
iab
le
b
io
m
etr
ic.
He
d
esig
n
ed
an
ea
r
r
ec
o
g
n
itio
n
s
y
s
te
m
b
ased
o
n
s
ize,
s
h
ap
e
an
d
p
o
s
itio
n
o
f
ea
r
.
A
t
least
in
1
9
9
8
,
an
a
u
to
m
a
ted
ea
r
r
ec
o
g
n
itio
n
s
y
s
te
m
p
r
o
p
o
s
ed
b
y
B
u
r
g
e
a
n
d
B
u
r
g
er
u
s
in
g
Vo
r
o
n
o
i
d
iag
r
a
m
s
[
1
6
]
.
T
o
tally
b
ased
o
n
liter
at
u
r
e,
f
ea
tu
r
e
ex
tr
ac
tio
n
is
d
o
n
e
u
s
i
n
g
v
ar
io
u
s
ap
p
r
o
ac
h
es:
g
eo
m
etr
ical
an
d
g
lo
b
al
ap
p
r
o
ac
h
[
1
7
]
.
So
m
e
m
et
h
o
d
s
w
h
ic
h
ar
e
b
ased
o
n
g
eo
m
e
tr
ical
ap
p
r
o
ac
h
s
u
c
h
a
s
:
p
er
s
p
ec
ti
v
e
m
et
h
o
d
s
[
1
8
]
,
g
eo
m
e
tr
ical
p
ar
a
m
e
ter
s
m
eth
o
d
[
1
9
-
2
0
]
,
g
eo
m
e
tr
ical
s
u
r
f
ac
e
p
r
o
p
e
r
ties
[
2
1
]
,
etc.
,
an
d
s
o
m
e
o
f
th
e
m
ar
e
b
ased
o
n
g
lo
b
al
ap
p
r
o
ac
h
:
f
o
r
ce
f
iel
d
tr
an
s
f
o
r
m
atio
n
[
1
2
]
,
lo
ca
l
b
i
n
ar
y
p
atter
n
[
2
2
]
,
Gab
o
r
f
ea
tu
r
es
[
2
3
]
,
etc.
A
p
ar
t
f
r
o
m
u
s
i
n
g
o
n
l
y
2
D
i
m
a
g
es [
2
4
-
2
6
]
,
[
1
3
]
,
A
s
m
aller
n
u
m
b
er
o
f
r
esear
ch
er
s
h
a
v
e
lo
o
k
ed
at
u
s
i
n
g
3
D
ea
r
s
h
ap
e
[
2
8
-
3
1
]
.
I
n
th
i
s
p
ap
er
a
n
e
w
ea
r
r
ec
o
g
n
itio
n
m
et
h
o
d
b
ased
o
n
a
m
o
d
if
ied
f
o
r
m
o
f
D
C
T
is
p
r
o
p
o
s
ed
w
h
ich
is
r
o
b
u
s
t
to
r
o
tatio
n
,
tr
an
s
latio
n
an
d
ill
u
m
i
n
atio
n
a
n
d
co
m
p
ar
ed
w
it
h
o
t
h
er
m
et
h
o
d
s
.
E
x
p
er
i
m
e
n
tal
r
es
u
lt
s
e
m
p
h
a
s
is
t
h
at
t
h
e
p
r
o
p
o
s
ed
m
eth
o
d
is
s
u
p
er
io
r
to
th
ese
m
et
h
o
d
s
in
u
n
co
n
tr
o
lled
co
n
d
itio
n
s
.
T
h
e
r
est
o
f
th
i
s
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
w
s
:
i
n
S
ec
tio
n
2
d
atab
ases
,
p
r
e
-
p
r
o
ce
s
s
i
n
g
a
n
d
n
o
r
m
aliza
t
io
n
i
s
d
es
cr
ib
ed
,
S
ec
tio
n
3
ad
d
r
ess
e
s
tr
a
n
s
f
o
r
m
e
d
DC
T
b
ased
ea
r
r
ec
o
g
n
itio
n
,
E
x
p
er
i
m
e
n
t
s
a
n
d
R
es
u
lts
ar
e
d
ep
icted
in
Sectio
n
4
an
d
at
th
e
la
s
t in
Sectio
n
5
C
o
n
cl
u
s
io
n
is
p
r
esen
ted
.
2.
D
AT
A
B
AS
E
S,
P
RE
-
P
RO
C
E
SS
I
N
G
AND
NO
RM
AL
I
Z
AT
I
O
N
T
h
is
p
ap
er
p
e
r
f
o
r
m
s
th
e
e
x
p
er
i
m
en
ts
o
n
U
ST
B
I
I
[
3
2
]
a
n
d
I
I
T
Delh
i
I
I
[
3
3
]
d
atab
as
es.
A
l
l
th
e
i
m
a
g
es
i
n
t
h
ese
d
atab
ase
s
h
av
e
b
ee
n
ca
p
t
u
r
ed
b
y
h
i
g
h
r
eso
lu
tio
n
ca
m
er
as
a
n
d
u
n
d
e
r
d
if
f
er
e
n
t
li
g
h
tin
g
co
n
d
itio
n
s
w
it
h
v
ar
io
u
s
v
ie
w
i
n
g
an
g
les.
UST
B
I
I
d
atab
ase
co
n
tain
s
r
ig
h
t
ea
r
i
m
a
g
es
f
r
o
m
7
7
s
u
b
j
ec
ts
w
ith
4
i
m
a
g
es
f
r
o
m
ea
c
h
s
u
b
j
ec
t,
3
0
8
im
a
g
es
i
n
to
tal.
T
h
e
I
I
T
D
I
I
ea
r
im
ag
e
d
atab
ase
co
n
tai
n
s
7
9
3
g
r
ay
-
s
ca
le
ea
r
i
m
a
g
es o
f
2
2
1
s
u
b
j
ec
ts
.
At
least th
r
ee
ea
r
i
m
a
g
es a
r
e
ac
q
u
ir
e
d
f
r
o
m
ea
ch
s
u
b
j
ec
t.
Sa
m
p
le
p
ict
u
r
es
o
f
th
e
s
e
d
at
ab
ases
ar
e
s
h
o
w
n
i
n
Fi
g
u
r
e
1
an
d
Fig
u
r
e
2
.
E
ac
h
r
o
w
i
n
f
i
g
u
r
es
is
r
elate
d
to
ea
r
im
ag
e
s
o
f
o
n
e
p
e
r
s
o
n
.
Fig
u
r
e
1
.
E
x
a
m
p
le
i
m
ag
e
s
o
f
e
ar
UST
B
I
I
d
atab
ase
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2088
-
8708
A
n
E
a
r
R
ec
o
g
n
itio
n
Meth
o
d
B
a
s
ed
o
n
R
o
ta
tio
n
I
n
va
r
ia
n
t Tr
a
n
s
fo
r
med
DC
T
(
F
a
teme
h
Ho
u
r
a
li
)
2897
Fig
u
r
e
2
.
T
y
p
ical
i
m
a
g
e
s
a
m
p
l
es f
r
o
m
ea
r
I
I
T
D
I
I
d
atab
ase
Fo
r
th
e
m
et
h
o
d
s
w
h
ich
ar
e
u
s
ed
f
o
r
ea
r
r
ec
o
g
n
itio
n
in
t
h
i
s
p
ap
er
w
e
as
s
u
m
e
t
h
at
t
h
e
r
eg
i
o
n
o
f
ea
r
h
as
b
ee
n
s
p
ec
if
ied
b
y
ea
r
d
etec
to
r
.
B
y
u
s
in
g
t
h
e
C
a
n
n
y
ed
g
e
d
etec
to
r
,
ed
g
e
m
ap
o
f
th
e
ea
r
i
m
a
g
e
is
o
b
tain
ed
,
th
en
its
m
as
s
ce
n
ter
w
h
ic
h
p
r
esen
t
s
th
e
ce
n
ter
o
f
a
cir
c
u
lar
r
eg
io
n
co
n
tain
in
g
t
h
e
ea
r
is
co
m
p
u
ted
.
T
r
an
s
latio
n
in
d
e
p
en
d
en
ce
o
f
r
ec
o
g
n
itio
n
p
r
o
ce
s
s
if
m
i
s
ali
g
n
m
e
n
t
v
ia
d
etec
to
r
o
cc
u
r
s
is
ac
h
ie
v
ed
b
y
ca
lc
u
lati
n
g
m
as
s
ce
n
ter
o
f
th
e
ed
g
e
m
ap
o
f
ea
r
i
m
ag
e
a
s
f
o
llo
w
s
[
3
4
]
:
=
1
×
∑
∑
(
,
)
=
1
=
1
(1
)
=
1
×
∑
∑
(
,
)
=
1
=
1
(2
)
W
h
er
e
M
an
d
N
ar
e
n
u
m
b
er
o
f
r
o
w
s
a
n
d
co
lu
m
n
s
o
f
th
e
i
n
p
u
t
i
m
a
g
e
an
d
I
is
th
e
ea
r
i
m
ag
e.
B
y
m
ea
n
s
o
f
t
h
e
ca
lc
u
lated
m
a
s
s
ce
n
ter
a
n
d
a
s
u
itab
le
r
ad
iu
s
wh
ich
e
x
p
er
i
m
e
n
tall
y
i
s
lar
g
e
e
n
o
u
g
h
f
o
r
t
h
e
f
u
t
u
r
e
s
tag
e
s
,
th
e
r
eg
io
n
co
n
tain
in
g
th
e
ea
r
h
as
b
ee
n
cr
o
p
p
ed
.
T
h
en
tr
an
s
itio
n
f
r
o
m
C
ar
tesi
an
to
p
o
lar
s
p
ac
e
is
p
er
f
o
r
m
ed
u
s
i
n
g
th
e
f
o
llo
w
in
g
eq
u
atio
n
s
.
=
√
(
−
)
2
+
(
−
)
2
(
3
)
=
2
(
−
−
)
T
h
e
r
ea
s
o
n
o
f
t
h
i
s
s
p
ac
e
c
h
a
n
g
i
n
g
i
s
f
o
llo
w
in
g
i
n
t
h
e
n
e
x
t
s
ec
tio
n
.
Af
ter
th
i
s
,
f
o
r
d
ec
r
ea
s
in
g
t
h
e
lig
h
ti
n
g
v
ar
iatio
n
b
et
w
ee
n
e
ar
im
a
g
es,
h
is
to
g
r
a
m
eq
u
aliz
in
g
u
s
i
n
g
lin
ea
r
co
n
tr
ast
e
x
p
an
d
in
g
f
u
n
ctio
n
i
s
p
er
f
o
r
m
ed
o
n
p
o
lar
m
ap
o
f
e
ar
i
m
a
g
e.
Fo
r
GFD
a
n
d
tr
a
n
s
f
o
r
m
ed
D
C
T
b
ased
m
et
h
o
d
s
,
all
s
tep
s
w
h
ic
h
ar
e
d
escr
ib
ed
f
o
r
n
o
r
m
aliza
tio
n
ar
e
p
er
f
o
r
m
ed
.
Fig
u
r
e
3
s
h
o
w
s
t
h
e
ea
r
i
m
ag
e
th
a
t sp
ec
i
f
ied
b
y
d
etec
to
r
,
ed
g
e
m
ap
o
f
it a
n
d
th
e
lo
ca
tio
n
o
f
it
s
m
a
s
s
ce
n
ter
.
Fig
u
r
e
4
d
is
p
la
y
s
h
is
to
g
r
a
m
eq
u
al
ized
p
o
lar
m
ap
o
f
th
e
ea
r
i
m
a
g
e.
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
(
a)
T
h
e
E
ar
im
a
g
e
lo
ca
ted
b
y
d
etec
to
r
,
(
b
)
I
ts
C
an
n
y
ed
g
e
m
ap
,
(
c)
Ma
s
s
ce
n
ter
o
f
it
Fig
u
r
e
4
.
P
o
lar
m
ap
o
f
th
e
ea
r
af
ter
h
i
s
to
g
r
a
m
eq
u
aliz
at
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
5
,
Octo
b
er
201
7
:
2
8
9
5
–
2
9
0
1
2898
3.
T
RANSF
O
RM
E
D
DC
T
B
ASE
D
E
AR
R
E
CO
G
NI
T
I
O
N
T
h
e
p
r
esen
ted
m
e
th
o
d
f
o
r
ea
r
r
ec
o
g
n
itio
n
i
n
th
is
p
ap
er
is
b
ased
o
n
a
m
o
d
if
ied
f
o
r
m
o
f
d
is
cr
ete
co
s
in
e
tr
an
s
f
o
r
m
.
T
h
is
m
et
h
o
d
w
it
h
d
i
v
id
i
n
g
t
h
e
i
m
a
g
e
in
to
d
if
f
er
en
t
co
m
p
o
n
en
t
s
in
ter
m
s
o
f
v
is
u
al
i
m
p
o
r
tan
ce
h
elp
s
to
i
m
a
g
e
co
m
p
r
e
s
s
i
n
g
.
I
n
f
ac
t
b
y
d
is
ca
r
d
i
n
g
t
h
e
le
s
s
i
m
p
o
r
tan
t
D
C
T
C
o
ef
f
icien
ts
r
elate
d
to
th
e
co
n
s
id
er
ed
i
m
a
g
e,
v
o
lu
m
e
an
d
s
p
ee
d
ca
lcu
latio
n
s
ca
n
b
e
d
e
cr
ea
s
ed
esp
ec
ially
f
o
r
au
d
io
an
d
im
a
g
e
co
m
p
r
es
s
i
n
g
ap
p
licatio
n
s
.
B
y
u
s
i
n
g
m
o
d
i
f
ied
f
o
r
m
o
f
D
C
T
(
tr
an
s
f
o
r
m
ed
D
C
T
)
o
n
th
e
p
o
lar
m
ap
o
f
i
m
a
g
e
as
f
o
llo
ws
it
w
ill
b
e
a
r
o
tatio
n
in
v
ar
ia
n
t
o
p
er
ato
r
.
I
t
ca
n
b
e
d
o
n
e
w
ith
i
g
n
o
r
in
g
th
e
p
h
ase
in
f
o
r
m
atio
n
o
f
th
e
co
ef
f
icien
ts
an
d
o
n
l
y
co
n
s
id
er
in
g
th
e
ir
m
a
g
n
i
tu
d
es.
B
u
t
s
lig
h
tl
y
tr
an
s
latio
n
s
i
g
n
if
ican
tl
y
in
f
l
u
e
n
ce
s
s
y
s
te
m
p
er
f
o
r
m
a
n
ce
s
.
E
v
en
i
f
m
is
p
lace
m
e
n
t
o
cc
u
r
s
v
ia
ea
r
d
etec
to
r
,
it
ca
n
b
e
r
ei
m
b
u
r
s
ed
w
ith
ca
lcu
la
tin
g
t
h
e
m
a
s
s
ce
n
ter
.
T
h
er
ef
o
r
tr
an
s
f
o
r
m
ed
DC
T
is
ap
p
lied
o
n
th
e
p
o
lar
m
ap
o
f
n
o
r
m
alize
d
ea
r
im
a
g
e.
T
h
e
f
o
llo
w
i
n
g
eq
u
atio
n
is
r
elate
d
to
g
en
er
al
2
D
D
C
T
f
o
r
an
N
b
y
M
i
m
a
g
e:
11
00
..
(
,
)
4
(
,
)
c
o
s
(2
1
)
c
o
s
(2
1
)
(4
)
2
.
2
.
NM
ij
uv
F
u
v
f
i
j
i
j
NM
(
4
)
W
h
er
e
f
(
i,j
)
is
th
e
in
ten
s
it
y
o
f
th
e
p
ix
els
i
n
th
e
i
-
t
h
r
o
w
an
d
j
-
th
co
lu
m
n
o
f
i
m
a
g
e
an
d
F(u
,
v
)
ar
e
th
e
co
ef
f
icie
n
t
s
o
f
DC
T
in
r
o
w
u
an
d
co
lu
m
n
v
o
f
D
C
T
m
at
r
ix
.
Fo
r
m
o
s
t
ca
s
e
s
co
m
p
r
ess
io
n
is
ac
h
ie
v
ed
b
y
ig
n
o
r
i
n
g
th
e
s
m
all
co
e
f
f
icien
ts
w
h
ich
ar
e
ap
p
ea
r
ed
in
u
p
p
er
lef
t
co
r
n
er
o
f
th
e
DC
T
m
atr
i
x
s
o
t
h
at
t
h
es
e
ch
an
g
es o
n
au
d
io
o
r
i
m
ag
e
i
s
n
o
t sen
s
ib
le.
W
e
u
s
e
a
tr
an
s
f
o
r
m
ed
t
y
p
e
o
f
D
C
T
b
y
th
e
f
o
llo
w
i
n
g
eq
u
atio
n
:
11
00
..
ˆ
(
,
)
4
(
,
)
c
o
s
(2
1
)
s
i
n
(2
1
)
(
5
)
2
.
2
.
NM
ij
uv
F
u
v
f
i
j
i
j
NM
(
5
)
B
y
co
m
p
o
s
in
g
t
h
ese
co
ef
f
icie
n
ts
w
it
h
DC
T
co
ef
f
icien
ts
(
i
n
eq
u
atio
n
(
4
)
)
as
f
o
llo
w
s
i
f
b
o
th
o
f
t
h
e
m
ap
p
lied
o
n
th
e
p
o
lar
m
ap
o
f
n
o
r
m
al
ized
ea
r
i
m
a
g
e
i
n
w
h
ic
h
i
an
d
j
r
elate
d
to
p
h
ase
a
n
d
p
h
ase
r
esp
ec
ti
v
el
y
,
th
e
p
h
ase
i
n
f
o
r
m
atio
n
w
ill b
e
d
is
ca
r
d
ed
.
T
h
en
ex
tr
ac
ted
f
ea
t
u
r
es i
n
th
i
s
m
e
th
o
d
ar
e
r
o
tatio
n
in
v
ar
ia
n
t.
11
22
00
.
ˆ
(
,
)
(
,
)
(
,
)
4
(
,
)
c
o
s
(2
1
)
(
6
)
2.
NM
ij
u
F
u
v
F
u
v
F
u
v
f
i
j
i
N
(
6
)
T
h
e
in
p
u
t
i
m
ag
e
is
d
ec
o
m
p
o
s
ed
to
8
×8
s
u
b
im
a
g
e
s
an
d
a
f
t
er
ap
p
ly
i
n
g
t
h
i
s
o
p
er
ato
r
o
n
ea
ch
ar
r
a
y
,
th
e
r
esu
lted
co
ef
f
icien
ts
ar
e
co
m
p
o
s
ed
.
4.
E
XP
E
R
I
M
E
NT
S AN
D
RE
S
UL
T
S
I
n
th
is
p
r
o
ce
d
u
r
e,
w
e
p
er
f
o
r
m
t
w
o
ex
p
er
i
m
en
t
s
to
e
v
alu
ate
t
h
e
p
r
o
p
o
s
ed
m
eth
o
d
.
A
l
l
t
h
e
ex
p
er
i
m
e
n
ts
ar
e
ca
r
r
ied
o
u
t
o
n
t
w
o
p
o
p
u
lar
ea
r
d
atab
ases
i.
e.
UST
B
s
u
b
s
et
I
I
an
d
I
I
T
De
lh
i
s
u
b
s
et
I
I
.
I
n
all
m
et
h
o
d
s
P
r
in
cip
le
C
o
m
p
o
n
en
t
An
al
y
s
i
s
(
P
C
A
)
is
ad
o
p
ted
to
en
h
an
ce
t
h
e
o
r
t
h
o
g
o
n
al
it
y
b
etw
ee
n
t
h
e
f
ea
t
u
r
es
’
co
m
p
o
n
e
n
t
s
a
n
d
8
5
%
o
f
t
h
e
t
o
tal
d
ata
ar
e
u
s
ed
f
o
r
tr
ain
in
g
an
d
t
h
e
r
est
is
i
n
te
n
d
ed
f
o
r
t
esti
n
g
.
W
e
e
v
al
u
ate
th
e
d
is
ti
n
cti
v
en
e
s
s
o
f
t
h
e
p
r
o
p
o
s
e
d
m
eth
o
d
w
it
h
co
m
p
ar
is
o
n
to
th
e
o
th
er
p
r
ev
io
u
s
m
et
h
o
d
s
w
i
th
g
o
o
d
p
er
f
o
r
m
a
n
ce
s
u
c
h
as t
h
e
L
P
B
[
3
5
]
,
DSI
F
T
[
3
6
]
an
d
Gab
o
r
[
3
7
]
.
I
n
f
ir
s
t e
x
p
er
i
m
e
n
t,
s
y
s
te
m
e
v
alu
atio
n
is
ca
r
r
ied
o
u
t b
ased
o
n
t
h
e
m
et
h
o
d
w
h
ic
h
is
d
e
f
i
n
ed
in
[
3
8
]
.
I
n
th
is
m
et
h
o
d
w
h
ich
is
n
a
m
ed
C
u
m
u
lati
v
e
Ma
tch
Sco
r
e,
p
er
ce
n
tag
e
o
f
co
r
r
ec
t
r
ec
o
g
n
itio
n
o
f
th
e
test
s
e
t
is
p
lo
tted
f
o
r
v
ar
io
u
s
r
an
k
s
a
s
i
n
d
icate
d
in
F
ig
u
r
e
4
an
d
Fi
g
u
r
e
5
.
I
n
th
ese
ca
s
e
s
,
w
h
e
n
f
ea
tu
r
e
m
atr
i
x
i
s
b
u
ilt,
r
ec
o
g
n
itio
n
is
d
o
n
e
b
y
m
ea
s
u
r
in
g
th
e
E
u
clid
ea
n
d
is
ta
n
ce
b
et
w
ee
n
f
ea
t
u
r
e
v
ec
to
r
s
an
d
ea
ch
clas
s
d
eter
m
in
ed
v
ec
to
r
.
I
n
Fig
u
r
e
5
th
e
o
p
er
atio
n
o
f
s
y
s
te
m
s
o
n
UST
B
I
I
ea
r
d
atab
ase
ar
e
co
m
p
ar
ed
.
A
s
t
h
is
f
i
g
u
r
e
s
h
o
ws
th
e
tr
a
n
s
f
o
r
m
ed
D
C
T
s
tar
ts
m
u
c
h
b
etter
th
a
n
o
t
h
er
m
eth
o
d
s
(
9
2
.
3
%
at
r
an
k
1
)
.
T
r
an
s
f
o
r
m
ed
D
C
T
r
ea
ch
es
1
0
0
%
r
ec
o
g
n
it
io
n
r
ate
at
r
an
k
6
o
n
th
e
f
ir
s
t
d
atab
ase
w
h
ile
L
B
P
,
DSI
FT
an
d
Ga
b
o
r
b
ased
m
et
h
o
d
s
r
esp
ec
tiv
el
y
ac
h
ie
v
e
at
r
an
k
1
5
,
1
3
an
d
1
5
.
Fig
u
r
e
6
p
r
esen
ts
s
y
s
te
m
s
’
p
er
f
o
r
m
an
ce
o
n
I
I
T
D
I
I
d
atab
ase.
A
ll
m
et
h
o
d
s
ar
e
n
ea
r
l
y
i
d
en
tical
i
n
s
tar
t,
b
u
t
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
h
a
s
a
b
etter
s
tar
t
(
9
8
.
8
%
at
r
a
n
k
1
)
an
d
r
ea
ch
es
1
0
0
%
r
ec
o
g
n
itio
n
r
ate
i
n
lo
w
er
r
an
k
.
T
h
er
ef
o
r
e
th
e
tr
an
s
f
o
r
m
ed
DC
T
b
ased
m
et
h
o
d
r
esu
lt
i
n
b
etter
p
er
f
o
r
m
a
n
ce
co
m
p
ar
i
n
g
w
it
h
o
th
er
m
et
h
o
d
s
o
n
t
w
o
p
u
b
licl
y
ea
r
d
atab
ases
esp
ec
iall
y
f
o
r
r
o
tated
p
r
o
b
es a
n
d
u
n
d
er
d
i
f
f
er
e
n
t lig
h
t c
o
n
d
itio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2088
-
8708
A
n
E
a
r
R
ec
o
g
n
itio
n
Meth
o
d
B
a
s
ed
o
n
R
o
ta
tio
n
I
n
va
r
ia
n
t Tr
a
n
s
fo
r
med
DC
T
(
F
a
teme
h
Ho
u
r
a
li
)
2899
Fig
u
r
e
5
.
UST
B
I
I
C
MSs
Fig
u
r
e
6
.
I
I
T
D
I
I
C
MSs
I
n
s
ec
o
n
d
ex
p
er
i
m
e
n
t,
w
e
ev
alu
a
te
th
e
p
er
f
o
r
m
a
n
ce
o
f
o
u
r
p
r
o
p
o
s
ed
m
eth
o
d
in
ter
m
s
o
f
ea
r
r
ec
o
g
n
itio
n
.
Ag
ain
,
w
e
co
m
p
a
r
e
th
e
p
er
f
o
r
m
an
ce
o
f
o
u
r
p
r
o
p
o
s
ed
m
eth
o
d
w
it
h
o
t
h
er
m
et
h
o
d
s
o
n
t
w
o
d
ef
i
n
ed
d
atab
ases
.
T
h
e
r
esu
lt
s
o
f
ap
p
ly
in
g
m
et
h
o
d
s
u
s
i
n
g
f
o
u
r
d
if
f
er
en
t c
la
s
s
i
f
ier
s
,
w
h
ic
h
ar
e
K
-
n
e
ar
est n
ei
g
h
b
o
r
s
(
K
-
NN)
,
L
in
ea
r
Su
p
p
o
r
t
Vec
to
r
Ma
ch
i
n
e
(
L
-
SV
M)
,
R
ad
ial
B
asis
F
u
n
ctio
n
SV
M
(
R
B
F
-
SVM)
,
an
d
Sp
ar
s
e
R
ep
r
esen
tat
io
n
C
la
s
s
i
f
ier
(
S
R
C
)
o
n
t
w
o
d
atab
ases
ar
e
ca
lcu
lated
an
d
av
er
ag
ed
.
T
ab
l
e
1
s
u
m
m
ar
izes
th
e
co
m
p
ar
is
o
n
r
es
u
lt
s
o
f
t
h
e
p
r
o
p
o
s
ed
m
eth
o
d
w
i
th
o
th
er
m
et
h
o
d
s
.
A
s
s
h
o
w
n
in
T
ab
le
1
,
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
s
till
p
r
o
d
u
ce
s
t
h
e
b
est
id
en
tific
atio
n
r
ate
r
eg
ar
d
less
o
f
a
n
y
class
if
icatio
n
tech
n
i
q
u
es
u
s
e
d
.
Ov
er
all,
t
h
e
b
e
s
t
id
en
ti
f
icatio
n
r
e
s
u
l
t
(
9
8
.
2
%)
is
ac
h
iev
ed
f
o
r
th
e
p
r
o
p
o
s
ed
m
e
th
o
d
u
s
i
n
g
K
-
NN
cla
s
s
i
f
ier
,
wh
er
e
K
is
b
ased
o
n
o
n
e
n
ea
r
est
n
eig
h
b
o
r
.
T
ab
le
1
.
E
ar
R
ec
o
g
n
itio
n
R
e
s
u
lts
f
o
r
Var
io
u
s
Me
t
h
o
d
s
M
e
t
h
o
d
C
l
a
ssi
f
i
e
r
s
K
-
NN
L
-
S
V
M
RBF
-
S
V
M
S
R
C
L
B
P
[
3
3
]
9
3
.
5
%
8
9
.
5
%
9
1
.
8
%
8
8
.
7
%
D
S
I
F
T
[
3
4
]
7
9
.
7
%
7
6
.
2
%
7
7
.
3
%
7
5
.
1
%
G
a
b
o
r
[
3
5
]
8
5
.
6
%
8
2
.
9
%
8
3
.
2
%
8
0
.
1
%
T
r
a
n
sf
o
r
me
d
D
C
T
9
8
.
2
%
9
6
.
7
%
9
7
.
5
%
9
3
.
5
%
5.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
ad
d
r
ess
es
th
e
ea
r
r
ec
o
g
n
itio
n
p
r
o
b
le
m
d
ep
en
d
i
n
g
o
n
a
p
r
o
p
o
s
ed
tr
an
s
f
o
r
m
ed
DC
T
b
ased
m
et
h
o
d
w
ith
co
m
p
ar
in
g
th
r
ee
ex
i
s
ti
n
g
m
et
h
o
d
s
i
n
v
ar
ia
n
t
co
n
d
itio
n
s
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
r
o
tatio
n
an
d
tr
an
s
latio
n
i
n
v
ar
ian
t
o
p
er
ato
r
an
d
h
a
s
ac
ce
p
tab
le
p
er
f
o
r
m
a
n
ce
u
n
d
er
v
ar
io
u
s
li
g
h
te
n
i
n
g
c
o
n
d
itio
n
s
.
B
ec
au
s
e
o
f
th
ese
s
p
ec
ial
f
ea
t
u
r
es,
th
is
m
et
h
o
d
is
s
u
itab
le
f
o
r
f
u
l
l
y
a
u
to
m
at
ic
r
ec
o
g
n
it
io
n
s
y
s
te
m
s
.
T
h
e
s
en
s
iti
v
it
y
o
f
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
to
tr
an
s
latio
n
s
h
as
b
ee
n
o
m
itted
b
y
ca
lcu
lati
n
g
th
e
m
as
s
ce
n
ter
o
f
th
e
p
r
o
c
ess
ed
ea
r
i
m
a
g
e,
b
ef
o
r
e
ap
p
l
y
i
n
g
tr
an
s
f
o
r
m
ed
d
is
cr
ete
co
s
in
e
tr
an
s
f
o
r
m
.
A
ls
o
b
ec
au
s
e
o
f
co
n
s
id
er
in
g
o
n
l
y
i
m
p
o
r
tan
t
co
ef
f
icie
n
t
s
,
th
i
s
m
eth
o
d
is
f
a
s
ter
co
m
p
ar
ed
to
o
th
er
m
et
h
o
d
s
.
E
x
p
er
i
m
en
t
r
esu
l
ts
o
n
t
w
o
d
atab
ases
(
UST
B
I
I
an
d
I
I
T
D
I
I
)
s
h
o
w
th
a
t
p
r
o
p
o
s
ed
m
et
h
o
d
is
s
u
p
er
io
r
w
i
th
r
esp
ec
t
to
o
th
er
m
et
h
o
d
s
s
u
c
h
as
L
B
P
,
DSI
FT
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
5
,
Octo
b
er
201
7
:
2
8
9
5
–
2
9
0
1
2900
Gab
o
r
in
th
e
co
n
d
itio
n
o
f
i
m
ag
e
ac
q
u
ir
in
g
w
h
e
n
t
h
er
e
is
r
o
tatio
n
an
d
tr
an
s
lat
io
n
in
ea
r
i
m
ag
e
s
an
d
u
n
d
er
v
ar
io
u
s
li
g
h
te
n
in
g
co
n
d
itio
n
s
.
RE
F
E
R
E
NC
E
S
[1
]
P
.
Ya
n
,
"
Ear
Bio
m
e
tri
c
s
in
Hu
m
a
n
Id
e
n
t
if
ica
ti
o
n
,
"
Ph
D
th
e
sis
,
U
n
ive
rs
it
y
o
f
N
o
tre
Da
me
,
N
o
tre
Da
me
,
IN,
US
A
,
Ju
ly
2
0
0
6
.
[2
]
S
.
Zo
k
a
e
e
a
n
d
K.
F
a
e
z
,
"
Hu
m
a
n
Id
e
n
ti
f
ica
t
io
n
Ba
se
d
o
n
ECG
a
n
d
P
a
lm
p
rin
t,
"
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
trica
l
a
n
d
Co
m
p
u
ter
E
n
g
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
2
,
p
p
.
2
6
1
-
2
6
6
,
2
0
1
2
.
[3
]
G
.
L
.
S
tree
ter,
"
De
v
e
lo
p
m
e
n
t
o
f
t
h
e
A
u
ricle
in
th
e
Hu
m
a
n
Em
b
r
y
o
,
"
Ca
rn
e
g
ie
In
stit
u
ti
o
n
o
f
W
a
sh
in
g
to
n
p
u
b
li
c
a
ti
o
n
,
Co
n
tri
b
u
ti
o
n
s
to
e
mb
ry
o
lo
g
y
,
v
o
l.
1
4
,
p
p
.
1
1
1
-
1
1
7
,
1
9
2
2
.
[4
]
J.
Da
v
is,
"
S
u
rg
ica
l
e
m
b
r
y
o
lo
g
y
,
"
1
st ed
.
Ne
w
Y
o
rk
,
N
Y
:
S
p
rin
g
e
r
-
Ver
la
g
,
1
9
8
7
.
[5
]
F
.
W
o
o
d
-
J
o
n
e
s
a
n
d
W
.
I
-
Ch
u
a
n
,
"
T
h
e
De
v
e
lo
p
m
e
n
t
o
f
th
e
Ex
tern
a
l
Ear
,
"
J
o
u
rn
a
l
o
f
A
n
a
t
o
my
,
v
o
l.
6
8
,
p
p
.
5
2
5
-
5
3
3
,
1
9
3
4
.
[6
]
B.
A
rb
a
b
-
Zav
a
r
a
n
d
M
.
S
.
Nix
o
n
,
"
On
G
u
id
e
d
M
o
d
e
l
-
b
a
se
d
A
n
a
ly
sis
f
o
r
Ear
Bio
m
e
tri
c
s
,
"
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
Vi
sio
n
a
n
d
Im
a
g
e
Un
d
e
rs
ta
n
d
i
n
g
,
v
o
l.
1
1
5
,
p
p
.
4
8
7
–
5
0
2
,
2
0
1
1
.
[7
]
A
.
Ia
n
n
a
re
ll
i,
"
Ear Id
e
n
ti
f
ica
ti
o
n
,
"
Pa
ra
m
o
n
t
Pu
b
li
sh
i
n
g
Co
mp
a
n
y
,
Fre
mo
n
t,
C
a
li
f
o
rn
i
a
,
1
9
8
9
.
[8
]
M
.
Bu
rg
e
a
n
d
W
.
Bu
rg
e
r,
"
Ear
Bio
m
e
tri
c
s
in
Co
m
p
u
ter
V
isio
n
,
"
in
1
5
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fe
re
n
c
e
o
f
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
2
0
0
0
.
ICPR
2
0
0
0
.
I
EE
E,
2
0
0
0
,
p
p
.
8
2
2
–
8
2
6
.
[9
]
B.
Bh
a
n
u
a
n
d
H.
C
h
e
n
,
"
Hu
m
a
n
Ear
Re
c
o
g
n
it
io
n
in
3
D,"
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
P
a
tt
e
rn
A
n
a
lys
is
a
n
d
M
a
c
h
in
e
In
telli
g
e
n
c
e
,
v
o
l.
2
9
,
p
p
.
7
1
8
-
7
3
7
,
2
0
0
7
.
[1
0
]
M
.
Bu
rg
e
a
n
d
W
.
Bu
rg
e
r,
"
Ear
Bio
m
e
tri
c
s
,
"
Bi
o
me
trics
:
Per
so
n
a
l
I
d
e
n
ti
f
ica
ti
o
n
i
n
Ne
two
rk
e
d
S
o
c
iety
,
Klu
we
r
Aca
d
e
mic
,
1
9
9
9
.
[1
1
]
K.
Ch
a
n
g
,
e
t
a
l.
,
"
Co
m
p
a
riso
n
a
n
d
Co
m
b
in
a
ti
o
n
o
f
Ear
a
n
d
F
a
c
e
Im
a
g
e
s
in
A
p
p
e
a
ra
n
c
e
-
b
a
se
d
Bi
o
m
e
tri
c
s
,
"
IEE
E
T
ra
n
sa
c
ti
o
n
o
n
P
a
tt
e
rn
A
n
a
lys
is
a
n
d
M
a
c
h
i
n
e
In
telli
g
e
n
c
e
,
v
o
l.
2
5
,
p
p
.
1
1
6
0
–
1
1
6
5
,
2
0
0
3
.
[1
2
]
D.
Hu
rley
,
e
t
a
l.
,
"
F
o
rc
e
F
iel
d
En
e
rg
y
F
u
n
c
ti
o
n
a
l
f
o
r
Im
a
g
e
F
e
a
tu
re
Ex
trac
ti
o
n
,
"
J
o
u
rn
a
l
o
f
Ima
g
e
a
n
d
Vi
sio
n
Co
mp
u
t
in
g
,
v
o
l
.
2
0
,
p
p
.
4
2
9
-
4
3
2
,
2
0
0
2
.
[1
3
]
N.
Da
m
e
r
a
n
d
B.
F
ü
h
re
r,
"
Ear
Re
c
o
g
n
it
io
n
Us
in
g
M
u
l
ti
-
S
c
a
le
H
isto
g
ra
m
o
f
Orie
n
ted
G
ra
d
ie
n
ts,
"
in
Ei
g
h
t
h
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
In
telli
g
e
n
t
In
fo
rm
a
t
io
n
Hid
in
g
a
n
d
M
u
lt
ime
d
i
a
S
i
g
n
a
l
Pro
c
e
ss
in
g
,
2
0
1
2
.
IIH
-
M
S
P2
0
1
2
.
IEE
E
,
2
0
1
2
,
p
p
.
2
1
-
2
4
.
[1
4
]
B.
V
ict
o
r
,
e
t
a
l
.
,
"
A
n
Ev
a
lu
a
ti
o
n
o
f
F
a
c
e
a
n
d
Ear
Bio
m
e
tri
c
s
,
"
in
1
6
t
h
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
f
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
2
0
0
2
.
ICPR
2
0
0
2
.
I
EE
E
,
2
0
0
2
,
p
p
.
4
2
9
-
4
3
2
.
[1
5
]
A
.
Be
rti
ll
o
n
,
"
L
a
p
h
o
t
o
g
ra
p
h
ie
ju
d
icia
ire:
a
v
e
c
u
n
a
p
p
e
n
d
ice
su
r
la
c
las
sif
ica
ti
o
n
e
t
l’i
d
e
n
ti
f
ica
ti
o
n
a
n
th
ro
p
o
m
´
e
tri
q
u
e
s,"
Pa
ris: Ga
u
t
h
ier
-
Vi
ll
a
rs
,
1
9
1
4
.
[1
6
]
M
.
Bu
rg
e
a
n
d
W
.
Bu
rg
e
r,
"
U
sin
g
Ear
Bio
m
e
tri
c
s
f
o
r
P
a
ss
iv
e
Id
e
n
ti
f
ica
ti
o
n
,
"
in
Pro
c
e
e
d
in
g
s
o
f
1
4
th
i
n
ter
n
a
t
io
n
a
l
c
o
n
fer
e
n
c
e
o
n
in
f
o
rm
a
ti
o
n
se
c
u
rity,
1
9
9
8
.
S
EC
1
9
9
8
.
IEE
E,
1
9
9
8
,
p
p
.
1
3
9
–
1
4
8
.
[1
7
]
A
.
A
b
a
z
a
,
e
t
a
l.
,
"
A
S
u
rv
e
y
on
Ear Bi
o
m
e
tri
c
s
,"
ACM
Co
mp
u
ti
n
g
S
u
rv
e
y
s,
v
o
l.
4
5
,
p
p
.
1
-
2
2
,
2
0
1
3
.
[1
8
]
M
.
Ch
o
ra
s,
"
Ear
b
i
o
m
e
tri
c
s
b
a
se
d
o
n
g
e
o
m
e
tri
c
a
l
fe
a
tu
re
e
x
trac
ti
o
n
,
"
El
e
c
tro
n
ic
L
e
tt
e
rs
o
n
C
o
mp
u
ter
Vi
sio
n
a
n
d
Ima
g
e
An
a
lys
is
(
J
o
u
rn
a
l
EL
CVI
A
),
v
o
l.
5
,
p
p
.
8
4
-
9
5
,
2
0
0
5
.
[1
9
]
I.
Om
a
ra
,
e
t
a
l.
,
"
A
No
v
e
l
G
e
o
m
e
tri
c
F
e
a
tu
re
Ex
trac
ti
o
n
M
e
th
o
d
f
o
r
Ear
Re
c
o
g
n
i
ti
o
n
,
"
In
ter
n
a
t
io
n
a
l
j
o
u
r
n
a
l
o
f
e
x
p
e
rt sy
ste
ms
wit
h
a
p
p
li
c
a
ti
o
n
s,
v
o
l.
6
5
,
p
p
.
1
2
7
-
1
3
5
,
2
0
1
6
.
[2
0
]
A
.
S
a
b
e
t
A
n
w
a
r
,
e
t
a
l.
,
"
Hu
m
a
n
Ear
Re
c
o
g
n
it
io
n
Us
in
g
G
e
o
m
e
tri
c
a
l
F
e
a
tu
re
s
Ex
trac
ti
o
n
,
"
in
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
C
o
mm
u
n
ica
ti
o
n
s,
ma
n
a
g
e
me
n
t,
a
n
d
In
f
o
rm
a
ti
o
n
t
e
c
h
n
o
l
o
g
y
,
2
0
1
5
.
ICCM
IT
2
0
1
5
.
El
se
v
ier
,
2
0
1
5
,
p
p
.
5
2
9
–
5
3
7
.
[2
1
]
I.
Iy
a
p
p
a
n
G
a
n
d
S
.
P
ra
k
a
sh
,
"
F
a
lse
M
a
p
p
e
d
F
e
a
tu
re
Re
m
o
v
a
l
in
S
p
in
Im
a
g
e
s
Ba
s
e
d
3
D
Ear
Re
c
o
g
n
it
io
n
,
"
in
3
rd
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
S
ig
n
a
l
Pro
c
e
ss
in
g
a
n
d
In
teg
r
a
ted
Ne
t
wo
rk
s,
2
0
1
6
.
S
PIN
2
0
1
6
.
IE
EE
,
2
0
1
6
,
p
p
.
1
1
-
1
2
.
[2
2
]
Z.
q
.
W
a
n
g
a
n
d
X.d
.
Ya
n
,
"
M
u
l
ti
-
S
c
a
le
F
e
a
tu
re
Ex
trac
ti
o
n
A
lg
o
rit
h
m
o
f
Ear
Im
a
g
e
,
"
in
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
El
e
c
tric In
fo
rm
a
ti
o
n
a
n
d
C
o
n
tro
l
En
g
i
n
e
e
rin
g
,
2
0
1
1
.
ICEIC
E
2
0
1
1
.
IEE
E,
2
0
1
1
,
p
p
.
5
2
8
-
5
3
0
.
[2
3
]
Y.
L
i,
e
t
a
l.
,
"
Ear
Re
c
o
g
n
it
io
n
w
it
h
Oc
c
lu
sio
n
V
ia
Disc
rim
in
a
ti
o
n
Dic
ti
o
n
a
ry
a
n
d
Oc
c
lu
si
o
n
Di
c
ti
o
n
a
ry
Ba
se
d
S
p
a
rse
Re
p
re
se
n
tatio
n
,
"
J
o
u
rn
a
l
o
f
Ne
u
ro
Co
mp
u
ti
n
g
(
El
se
v
ier
),
v
o
l.
1
7
1
,
p
p
.
5
4
0
-
5
5
0
,
2
0
1
6
.
[2
4
]
Zh
.
L
i
a
n
d
H.
A
d
e
li
,
"
Im
a
g
e
S
e
t
Ba
se
d
Ear
Re
c
o
g
n
it
i
o
n
Us
i
n
g
No
v
e
l
Dic
ti
o
n
a
ry
L
e
a
rn
in
g
a
n
d
C
las
sif
ica
ti
o
n
S
c
h
e
m
e
,
"
J
o
u
rn
a
l
o
f
E
n
g
in
e
e
rin
g
Ap
p
li
c
a
ti
o
n
s
o
f
Art
if
icia
l
In
tell
ig
e
n
c
e
,
v
o
l.
5
5
,
p
p
.
3
7
–
4
6
,
2
0
1
6
.
[2
5
]
A
.
Ba
sit
a
n
d
M
.
S
h
o
a
ib
,
"
A
Hu
m
a
n
Ear
Re
c
o
g
n
it
io
n
M
e
th
o
d
Us
in
g
No
n
l
in
e
a
r
Cu
rv
e
let
F
e
a
tu
re
S
u
b
s
p
a
c
e
,
"
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
M
a
th
e
ma
ti
c
s,
v
o
l
.
9
1
,
p
p
.
6
1
6
-
6
2
4
,
2
0
1
4
.
[2
6
]
A
.
Tariq
a
n
d
M
.
Us
m
a
n
A
k
ra
m
,
"
P
e
rso
n
a
l
Id
e
n
t
i
f
ica
ti
o
n
Us
in
g
Ea
r
Re
c
o
g
n
i
ti
o
n
,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
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
0
,
p
p
.
3
2
1
-
3
2
6
,
2
0
1
2
.
[2
7
]
A
.
Be
n
z
a
o
u
i,
e
t
a
l
.
,
"
Ear
+
,
"
J
o
u
r
n
a
l
o
f
E
lec
tro
n
ic Im
a
g
i
n
g
,
v
o
l.
2
3
,
p
p
.
1
2
1
-
1
3
0
,
2
0
1
4
.
[2
8
]
S
.
M
.
S
.
Isla
m
,
e
t
a
l.
,
"
M
u
lt
i
b
io
m
e
tri
c
Hu
m
a
n
Re
c
o
g
n
it
io
n
Us
in
g
3
D
Ear
a
n
d
F
a
c
e
F
e
a
tu
re
s
,
"
J
o
u
rn
a
l
o
f
P
a
tt
e
rn
Rec
o
g
n
it
io
n
,
v
o
l.
4
6
,
p
p
.
6
1
3
–
6
2
7
,
2
0
1
3
.
[2
9
]
G
.
D.
T
r
e
,
e
t
a
l.
,
"
Hu
m
a
n
Ce
n
t
ric
Re
c
o
g
n
it
io
n
o
f
3
D
Ear
M
o
d
e
ls,
"
In
ter
n
a
t
io
n
a
l
J
o
u
r
n
a
l
o
f
Co
mp
u
ta
t
io
n
a
l
In
telli
g
e
n
c
e
S
y
ste
ms
,
v
o
l.
9
,
p
p
.
2
9
6
-
3
1
0
,
2
0
1
6
.
[3
0
]
X
.
S
u
n
e
t
a
l.
,
"
3
D
E
a
r
Re
c
o
g
n
it
io
n
Us
i
n
g
L
o
c
a
l
S
a
li
e
n
c
e
a
n
d
P
r
in
c
ip
a
l
M
a
n
if
o
ld
,
"
J
o
u
r
n
a
l
o
f
Gr
a
p
h
ica
l
M
o
d
e
ls
(
El
se
v
i
e
r),
v
o
l.
7
6
,
p
p
.
4
0
2
-
4
1
2
,
2
0
1
4
.
[3
1
]
H.
Zen
g
e
t
a
l.
,
"
L
o
c
a
l
F
e
a
tu
re
D
e
sc
rip
to
r
Ba
se
d
Ra
p
id
3
D
Ear
R
e
c
o
g
n
it
io
n
,
"
in
33
rd
Ch
in
e
s
Co
n
t
ro
l
Co
n
fer
e
n
c
e
,
2
0
1
4
.
CCC
2
0
1
4
.
IE
EE
,
2
0
1
4
,
p
p
.
4
9
4
2
-
4
9
4
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2088
-
8708
A
n
E
a
r
R
ec
o
g
n
itio
n
Meth
o
d
B
a
s
ed
o
n
R
o
ta
tio
n
I
n
va
r
ia
n
t Tr
a
n
s
fo
r
med
DC
T
(
F
a
teme
h
Ho
u
r
a
li
)
2901
[3
2
]
h
tt
p
:
//
ww
w
1
.
u
stb
.
e
d
u
.
c
n
/res
b
/en
/i
n
d
e
x
.
h
tm
[3
3
]
h
tt
p
:
//
ww
w
4
.
c
o
m
p
.
p
o
ly
u
.
e
d
u
.
h
k
/~
c
sa
ja
y
k
r/II
T
D/D
a
tab
a
se
_
Ear.h
tm
[3
4
]
A
.
F
.
A
b
a
te
,
e
t
a
l.
,
"
Ear
Re
c
o
g
n
it
io
n
b
y
m
e
a
n
s
o
f
a
Ro
tatio
n
I
n
v
a
rian
t
De
sc
rip
to
r
,
"
in
1
8
th
I
n
ter
n
a
t
i
o
n
a
l
Co
n
fer
e
n
c
e
o
n
P
a
tt
e
rn
Rec
o
g
n
it
i
o
n
2
0
0
6
.
IC
PR
2
0
0
6
.
IEE
E,
2
0
0
6
,
p
p
.
4
3
7
-
4
4
0
.
[3
5
]
A
.
Be
n
z
a
o
u
i
,
e
t
a
l.
,
"
Ear
De
sc
rip
ti
o
n
+
EL
BP
a
n
d
W
a
v
e
lets
,
"
in
Pro
c
e
e
d
in
g
s
o
f
t
h
e
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Ap
p
li
e
d
Res
e
a
rc
h
i
n
C
o
mp
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
,
2
0
1
5
,
IC
AR
2
0
1
5
.
IEE
E,
2
0
1
5
,
p
p
.
1
-
6.
[3
6
]
A
.
M
o
ra
les
,
e
t
a
l.
,
"
A
n
a
l
y
sis
o
f
Lo
c
a
l
De
sc
rip
to
rs
F
e
a
tu
re
s
a
n
d
it
s
Ro
b
u
st
n
e
ss
A
p
p
li
e
d
to
Ear
Re
c
o
g
n
it
io
n
,
"
i
n
4
7
t
h
IEE
E
In
ter
n
a
ti
o
n
a
l
C
a
rn
a
h
a
n
C
o
n
fer
e
n
c
e
o
n
S
e
c
u
rity T
e
c
h
n
o
l
o
g
y
,
2
0
1
4
.
ICCS
T
2
0
1
4
.
IEE
E,
2
0
1
4
,
p
p
.
1
-
5.
[3
7
]
A
.
M
e
ra
o
u
m
ia
,
e
t
a
l.
,
"
A
n
A
u
to
m
a
ted
Ear
Id
e
n
t
if
ica
ti
o
n
S
y
ste
m
Us
in
g
G
a
b
o
r
F
il
ter
Re
sp
o
n
se
s
,
"
i
n
Pro
c
e
e
d
i
n
g
s
o
f
th
e
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Ne
w
Circ
u
it
s
a
n
d
S
y
ste
ms
,
2
0
1
5
.
NEW
CAS
2
0
1
5
.
T
h
irtee
n
A
n
n
u
a
l
IEE
E
,
2
0
1
5
,
p
p
.
1
-
4.
[3
8
]
P
.
J.
P
h
il
li
p
s,
e
t
a
l.
,
"
T
h
e
F
ERE
T
Ev
a
lu
a
ti
o
n
M
e
th
o
d
o
l
o
g
y
f
o
r
F
a
c
e
Re
c
o
g
n
it
io
n
A
lg
o
rit
h
m
,
"
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
P
a
tt
e
r
n
A
n
a
l
y
sis a
n
d
M
a
c
h
in
e
I
n
t
e
ll
ig
e
n
c
e
,
v
o
l.
2
2
,
p
p
.
1
0
9
0
-
1
1
0
4
,
2
0
0
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.