T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n Co
m
pu
t
ing
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
24
,
No
.
2
,
A
p
r
il
20
26
,
p
p
.
648
~
662
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
KA.
v
24
i
2
.
2
7
3
5
6
648
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Sco
re
-
lev
el biom
e
tric
in
forma
tion f
usio
n wit
h
genera
lized power
mea
n
L
eila
H
ella
l
1
,
Na
ce
ur
-
E
dd
i
n
e
B
o
uk
ez
zo
ula
1
,
M
o
ha
m
ed
C
heniti
2
,
Z
a
hi
d Ak
hta
r
3
1
LI
S
La
b
o
r
a
t
o
r
y
,
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
,
F
a
c
u
l
t
y
o
f
Te
c
h
n
o
l
o
g
y
,
F
e
r
h
a
t
A
b
b
a
s Un
i
v
e
r
si
t
y
,
S
e
t
i
f
,
A
l
g
e
r
i
a
2
I
n
st
i
t
u
t
e
o
f
O
p
t
i
c
s a
n
d
P
r
e
c
i
s
i
o
n
M
e
c
h
a
n
i
c
s,
F
e
r
h
a
t
A
b
b
a
s U
n
i
v
e
r
s
i
t
y
S
e
t
i
f
-
1
,
S
e
t
i
f
,
A
l
g
e
r
i
a
3
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
S
t
a
t
e
U
n
i
v
e
r
si
t
y
o
f
N
e
w
Y
o
r
k
P
o
l
y
t
e
c
h
n
i
c
I
n
st
i
t
u
t
e
,
U
t
i
c
a
,
U
n
i
t
e
d
S
t
a
t
e
s
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
6
,
20
25
R
ev
is
ed
J
an
17
,
2
0
2
6
Acc
ep
ted
J
an
30
,
2
0
2
6
To
o
v
e
rc
o
m
e
th
e
fu
n
d
a
m
e
n
tal
sh
o
rtco
m
in
g
s
o
f
sin
g
le
-
trait
b
io
m
e
tri
c
sy
ste
m
s,
m
u
lt
imo
d
a
l
s
o
lu
t
io
n
s
h
a
v
e
g
a
i
n
e
d
c
o
n
sid
e
ra
b
le
i
n
tere
st.
In
th
is
wo
r
k
,
a
sc
o
re
-
lev
e
l
fu
sio
n
sc
h
e
m
e
fo
r
b
io
m
e
tri
c
a
u
th
e
n
ti
c
a
ti
o
n
is
in
tr
o
d
u
c
e
d
,
wh
e
re
in
fo
rm
a
ti
o
n
fr
o
m
m
u
lt
ip
le
m
o
d
a
l
i
ti
e
s
is
c
o
m
b
in
e
d
u
si
n
g
c
o
n
v
e
n
ti
o
n
a
l
m
e
a
n
o
p
e
ra
to
rs
su
c
h
a
s
a
rit
h
m
e
ti
c
,
h
a
r
m
o
n
ic,
g
e
o
m
e
tri
c
,
a
n
d
q
u
a
d
ra
ti
c
m
e
a
n
s,
with
p
a
rti
c
u
lar
a
tt
e
n
ti
o
n
g
i
v
e
n
to
t
h
e
p
o
we
r
m
e
a
n
f
o
rm
u
lati
o
n
.
T
h
e
p
r
o
p
o
se
d
fra
m
e
wo
rk
in
c
re
a
se
s
sy
ste
m
ro
b
u
stn
e
ss
wh
il
e
p
re
se
rv
in
g
l
o
w
c
o
m
p
u
tat
i
o
n
a
l
c
o
m
p
lex
it
y
a
n
d
re
q
u
iri
n
g
n
o
trai
n
in
g
p
h
a
se
.
P
e
rfo
rm
a
n
c
e
is as
se
ss
e
d
o
n
th
re
e
we
ll
-
k
n
o
w
n
p
u
b
l
ic
d
a
tas
e
ts:
Na
ti
o
n
a
l
I
n
stit
u
te
o
f
S
ta
n
d
a
rd
s
a
n
d
T
e
c
h
n
o
l
o
g
y
(
NIST
)
-
fin
g
e
rp
rin
t,
NIST
-
fa
c
e
,
a
n
d
XM
2
VTS
,
u
sin
g
sta
n
d
a
rd
sc
o
re
n
o
rm
a
li
z
a
ti
o
n
m
e
th
o
d
s
a
n
d
c
o
m
m
o
n
ly
e
m
p
l
o
y
e
d
e
v
a
lu
a
ti
o
n
m
e
t
rics
.
Th
e
e
x
p
e
rime
n
tal
a
n
a
ly
sis
sh
o
ws
th
a
t
th
e
q
u
a
d
ra
ti
c
m
e
a
n
a
tt
a
i
n
s
a
g
e
n
u
i
n
e
a
c
c
e
p
tan
c
e
ra
te
(
G
AR)
o
f
9
1
.
5
0
%
o
n
th
e
NI
S
T
-
fin
g
e
rp
ri
n
t
d
a
tas
e
t,
wh
il
e
t
h
e
p
o
we
r
m
e
a
n
with
α
=
5
a
c
h
iev
e
s
8
2
.
4
0
%
o
n
NIST
-
fa
c
e
.
F
u
rth
e
rm
o
r
e
,
th
e
h
a
lf
to
tal
e
rro
r
ra
te (
HTER)
o
n
XM
2
V
TS
is r
e
d
u
c
e
d
t
o
0
.
0
5
9
.
In
c
o
m
p
a
riso
n
wit
h
lea
rn
in
g
-
b
a
se
d
fu
si
o
n
tec
h
n
iq
u
e
s,
th
e
p
r
o
p
o
se
d
a
p
p
ro
a
c
h
p
r
o
v
i
d
e
s
a
m
o
re
stra
ig
h
tf
o
rwa
rd
,
c
o
m
p
u
tati
o
n
a
ll
y
e
fficie
n
t,
a
n
d
d
e
p
e
n
d
a
b
le
a
lt
e
rn
a
ti
v
e
f
o
r
re
a
l
-
wo
rld
b
io
m
e
tri
c
a
p
p
li
c
a
ti
o
n
s
.
K
ey
w
o
r
d
s
:
B
io
m
etr
ic
au
th
en
ticatio
n
Gen
u
in
e
ac
ce
p
tan
ce
r
ate
Half
to
tal
er
r
o
r
r
ate
Natio
n
al
I
n
s
titu
te
o
f
Stan
d
ar
d
s
an
d
T
ec
h
n
o
lo
g
y
f
in
g
er
p
r
in
t
Sco
r
e
lev
el
f
u
s
io
n
XM
2
VT
S d
atab
ase
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
:
L
eila
Hellal
L
I
S L
ab
o
r
ato
r
y
,
Dep
a
r
tm
en
t o
f
E
lectr
o
n
ics,
Facu
lty
o
f
T
ec
h
n
o
lo
g
y
Fer
h
at
Ab
b
as Un
iv
er
s
ity
Setif
-
1
Setif,
Alg
er
ia
E
m
ail:
leila.
h
ellal@
u
n
iv
-
s
etif
.
d
z
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
r
ap
i
d
g
r
o
wth
o
f
d
ig
ital
id
en
tity
f
r
au
d
h
as
u
r
g
ed
in
d
iv
id
u
als
an
d
o
r
g
a
n
izatio
n
s
to
a
d
o
p
t
r
eliab
le
id
en
tity
m
a
n
ag
em
en
t
s
o
lu
tio
n
s
.
B
io
m
etr
ic
s
y
s
tem
s
au
th
en
ticate
in
d
iv
id
u
als
b
y
ex
a
m
in
in
g
d
is
tin
ctiv
e
p
h
y
s
io
lo
g
ical,
b
e
h
av
io
r
al,
o
r
c
h
em
ical
tr
aits
,
an
d
th
ey
ar
e
n
o
w
co
m
m
o
n
ly
d
e
p
lo
y
e
d
in
ap
p
l
icatio
n
s
,
in
clu
d
in
g
m
o
b
ile
p
latf
o
r
m
s
an
d
b
o
r
d
er
s
ec
u
r
ity
s
y
s
tem
s
[
1
]
.
A
u
n
im
o
d
al
b
io
m
etr
ic
s
y
s
tem
p
er
f
o
r
m
s
id
en
tity
v
er
if
icatio
n
u
s
in
g
o
n
ly
o
n
e
b
io
m
etr
ic
ch
a
r
ac
ter
is
tic,
s
u
ch
as
f
ac
ial
f
ea
t
u
r
es,
f
in
g
er
p
r
i
n
ts
,
o
r
ir
is
p
atte
r
n
s
.
Alth
o
u
g
h
s
u
c
h
s
y
s
tem
s
ca
n
p
er
f
o
r
m
well
u
n
d
er
id
ea
l
co
n
d
itio
n
s
,
d
e
p
en
d
e
n
c
e
o
n
o
n
e
b
io
m
etr
ic
s
o
u
r
ce
e
x
p
o
s
es
th
em
to
s
ev
er
al
wea
k
n
ess
es.
C
o
m
m
o
n
is
s
u
es
in
clu
d
e
lim
ited
p
o
p
u
latio
n
co
v
er
ag
e,
d
eg
r
a
d
ed
o
r
n
o
is
y
s
a
m
p
les,
in
s
u
f
f
icien
t
d
is
cr
im
in
ato
r
y
p
o
wer
,
v
ar
iati
o
n
s
with
in
th
e
s
am
e
in
d
iv
id
u
al,
an
d
in
cr
ea
s
ed
v
u
ln
er
ab
ilit
y
to
s
p
o
o
f
in
g
attac
k
s
[
2
]
.
T
o
a
d
d
r
ess
th
ese
s
h
o
r
tco
m
in
g
s
,
r
esear
ch
h
as
d
e
m
o
n
s
tr
ated
th
at
co
m
b
in
in
g
m
u
ltip
le
b
io
m
etr
ic
tr
aits
lead
s
to
m
o
r
e
r
eliab
le
a
n
d
ac
cu
r
ate
r
ec
o
g
n
itio
n
p
er
f
o
r
m
a
n
ce
[
3
]
.
T
h
ese
s
y
s
tem
s
,
k
n
o
wn
as
m
u
ltimo
d
al
b
io
m
etr
ic
s
y
s
tem
s
,
in
teg
r
ate
in
f
o
r
m
atio
n
f
r
o
m
d
if
f
e
r
en
t
m
o
d
alities
at
d
is
tin
ct
s
tag
es
o
f
p
r
o
ce
s
s
i
n
g
,
eith
er
p
r
io
r
to
m
atch
in
g
(
s
en
s
o
r
-
lev
el
an
d
f
e
atu
r
e
-
lev
el
f
u
s
io
n
)
o
r
af
ter
m
a
tch
in
g
(
s
co
r
e
-
lev
el,
r
an
k
-
lev
el
,
an
d
d
e
cisi
o
n
-
lev
el
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
co
r
e
-
leve
l b
io
metric in
fo
r
ma
tio
n
fu
s
io
n
w
ith
g
en
era
liz
ed
p
o
w
er me
a
n
(
Leila
Hella
l
)
649
f
u
s
io
n
)
[
4
]
,
[
5
]
.
S
e
n
s
o
r
-
l
e
v
e
l
f
u
s
i
o
n
i
s
e
m
p
l
o
y
e
d
w
h
e
n
m
u
lt
i
p
l
e
b
i
o
m
e
t
r
ic
t
r
a
i
ts
a
r
e
c
a
p
t
u
r
e
d
u
s
i
n
g
d
i
f
f
e
r
e
n
t
s
e
n
s
o
r
s
o
r
w
h
e
n
s
e
v
e
r
a
l
i
m
p
r
e
s
s
i
o
n
s
o
f
t
h
e
s
a
m
e
t
r
ai
t
a
r
e
o
b
t
a
in
e
d
w
i
t
h
a
s
i
n
g
l
e
s
e
n
s
o
r
.
F
o
r
ex
a
m
p
l
e
,
C
a
d
e
r
e
t
a
l
.
[
6
]
d
e
m
o
n
s
t
r
a
t
e
d
s
y
s
t
e
m
s
t
h
at
u
s
e
v
a
r
i
o
u
s
f
i
n
g
e
r
p
r
i
n
t
i
m
a
g
e
s
e
g
m
e
n
t
s
t
o
r
e
c
o
n
s
t
r
u
ct
c
o
m
p
le
t
e
f
i
n
g
e
r
p
r
i
n
ts
a
n
d
c
o
m
p
u
t
e
t
h
e
i
r
c
o
r
r
e
s
p
o
n
d
i
n
g
s
c
o
r
e
s
.
Fu
s
io
n
at
th
e
f
ea
tu
r
e
lev
el
is
g
en
er
ally
e
x
p
ec
ted
to
p
r
o
v
id
e
h
ig
h
er
r
ec
o
g
n
itio
n
p
er
f
o
r
m
a
n
ce
[
7
]
.
Ho
wev
er
,
th
is
ap
p
r
o
ac
h
p
o
s
es
s
ig
n
if
ican
t
ch
allen
g
es
d
u
e
t
o
t
h
e
h
eter
o
g
en
eity
an
d
in
co
m
p
atib
ilit
y
am
o
n
g
f
ea
tu
r
e
r
ep
r
esen
tatio
n
s
o
f
d
if
f
e
r
en
t
m
o
d
alities
[
7
]
,
[
8
]
.
Fo
r
in
s
tan
ce
,
co
m
b
in
in
g
ir
is
an
d
f
ac
e
f
ea
tu
r
es
o
f
ten
r
e
q
u
ir
es
co
m
p
lex
n
o
r
m
aliza
tio
n
a
n
d
d
im
en
s
io
n
ality
r
ed
u
ctio
n
tech
n
iq
u
es
to
ad
d
r
ess
th
ese
is
s
u
es
[
9
]
.
On
th
e
o
th
er
h
a
n
d
,
d
ec
is
io
n
-
lev
el
f
u
s
io
n
g
en
er
all
y
y
ield
s
l
o
wer
p
er
f
o
r
m
an
ce
s
in
ce
it c
o
m
b
in
es
o
n
ly
th
e
f
in
al
b
in
ar
y
o
u
tco
m
es
f
r
o
m
in
d
iv
i
d
u
al
m
o
d
alities
.
B
ec
au
s
e
th
e
r
aw
m
atch
in
g
s
co
r
e
s
an
d
i
n
ter
m
ed
iate
in
f
o
r
m
atio
n
ar
e
n
o
lo
n
g
er
a
v
ai
lab
le
at
th
is
s
tag
e,
a
s
ig
n
if
ican
t
am
o
u
n
t
o
f
d
is
cr
im
in
ativ
e
d
at
a
is
lo
s
t,
lead
in
g
to
r
ed
u
ce
d
f
lex
i
b
ilit
y
an
d
ac
cu
r
a
cy
.
Mo
r
eo
v
e
r
,
th
is
ap
p
r
o
ac
h
h
ea
v
ily
d
ep
en
d
s
o
n
th
e
th
r
esh
o
l
d
in
g
s
tr
ateg
ies
an
d
d
ec
is
io
n
r
u
les
a
p
p
lied
to
ea
c
h
class
if
ier
,
wh
ich
ca
n
in
tr
o
d
u
ce
in
co
n
s
is
ten
cies
wh
en
m
o
d
aliti
es
ex
h
ib
it
d
i
f
f
er
en
t
r
eliab
ilit
y
lev
els
[
1
0
]
.
I
n
s
co
r
e
-
lev
el,
also
k
n
o
wn
as
m
ea
s
u
r
em
en
t
-
lev
el,
f
u
s
io
n
,
th
e
m
atch
in
g
s
co
r
es
p
r
o
d
u
ce
d
b
y
m
u
ltip
le
b
io
m
etr
ic
m
o
d
alities
ar
e
a
g
g
r
e
g
ated
in
to
a
u
n
if
ie
d
s
co
r
e.
T
h
is
co
m
b
i
n
ed
v
alu
e
is
s
u
b
s
eq
u
en
t
ly
ev
alu
ated
b
y
th
e
v
er
if
icatio
n
o
r
id
e
n
tific
atio
n
m
o
d
u
le
to
d
eter
m
in
e
th
e
f
in
al
au
th
en
ticatio
n
o
u
tc
o
m
e.
T
h
is
m
e
th
o
d
is
wid
ely
u
s
ed
as
it
o
f
f
er
s
an
ef
f
ec
tiv
e
tr
a
d
e
-
o
f
f
b
etwe
en
c
o
m
p
u
tatio
n
al
ef
f
icien
cy
an
d
r
e
co
g
n
itio
n
ac
cu
r
ac
y
wh
en
co
m
p
ar
ed
with
f
ea
tu
r
e
a
n
d
d
ec
is
io
n
-
lev
el
f
u
s
io
n
.
I
ts
f
lex
ib
ilit
y
allo
w
s
in
teg
r
atio
n
o
f
h
eter
o
g
en
eo
u
s
b
io
m
etr
ic
s
o
u
r
ce
s
with
o
u
t
th
e
n
ee
d
f
o
r
co
m
p
lex
f
ea
tu
r
e
alig
n
m
e
n
t
o
r
r
etr
ain
i
n
g
o
f
m
o
d
ality
-
s
p
ec
if
ic
m
o
d
els
,
m
ak
in
g
it
s
u
itab
le
f
o
r
r
ea
l
-
wo
r
ld
d
e
p
lo
y
m
en
ts
.
R
ec
en
t
s
tu
d
ies
co
n
tin
u
e
to
s
h
o
w
th
at
s
co
r
e
-
lev
el
f
u
s
io
n
r
em
ain
s
a
v
er
y
s
tr
o
n
g
p
r
ac
tical
s
tr
a
teg
y
in
m
u
ltimo
d
al
b
io
m
et
r
ic
s
y
s
tem
s
,
esp
ec
ially
wh
en
co
m
b
in
in
g
d
is
p
ar
ate
m
o
d
alities
o
r
m
atch
er
s
wh
er
e
r
aw
f
ea
t
u
r
es
ar
e
in
co
m
p
atib
le
o
r
u
n
av
aila
b
le.
Fo
r
ex
am
p
le,
th
e
q
u
ality
-
g
u
id
ed
m
ix
tu
r
e
o
f
s
co
r
e
-
f
u
s
io
n
e
x
p
er
ts
(
QM
E
)
f
r
am
ewo
r
k
ac
h
iev
e
d
s
tate
-
of
-
th
e
-
ar
t
w
h
o
le
-
b
o
d
y
r
ec
o
g
n
iti
o
n
b
y
em
p
l
o
y
in
g
a
lear
n
ed
s
co
r
e
-
f
u
s
io
n
s
tr
ateg
y
t
h
at
ef
f
ec
tiv
ely
m
an
a
g
ed
v
a
r
ia
b
ilit
y
in
s
co
r
e
d
is
tr
ib
u
tio
n
s
an
d
d
ata
q
u
ality
ac
r
o
s
s
m
o
d
alities
,
o
u
tp
e
r
f
o
r
m
in
g
b
as
elin
e
m
eth
o
d
s
[
1
1
]
.
I
n
co
n
tr
ast,
r
ec
en
t
f
ea
tu
r
e
-
lev
el
f
u
s
io
n
a
p
p
r
o
ac
h
es,
s
u
ch
as
Vek
ar
iy
a
et
a
l
.
[
1
2
]
em
p
lo
y
i
n
g
a
b
in
ar
y
ch
im
p
-
o
p
tim
ized
ad
ap
tiv
e
k
er
n
el
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
e
(
B
C
O
-
AKSVM
)
,
h
av
e
d
em
o
n
s
tr
ated
ex
ce
llen
t
ac
cu
r
ac
y
an
d
lo
w
eq
u
al
er
r
o
r
r
ate
(
E
E
R
)
,
s
o
m
etim
es
s
u
r
p
ass
in
g
s
co
r
e
-
lev
el
m
eth
o
d
s
.
Ho
wev
er
,
th
es
e
m
eth
o
d
s
co
m
e
with
in
cr
ea
s
ed
co
m
p
u
tatio
n
al
c
o
s
t
an
d
co
m
p
lex
ity
in
f
ea
tu
r
e
ex
tr
ac
tio
n
a
n
d
d
im
en
s
io
n
ality
r
ed
u
ctio
n
.
An
o
t
h
er
c
o
m
p
a
r
ativ
e
wo
r
k
[
1
3
]
c
o
m
b
in
es
b
o
th
f
ea
t
u
r
e
an
d
s
co
r
e
-
lev
el
f
u
s
io
n
to
g
ai
n
r
o
b
u
s
tn
ess
,
s
h
o
win
g
th
at
wh
ile
f
ea
tu
r
e
f
u
s
io
n
b
o
o
s
ts
in
f
o
r
m
atio
n
c
o
n
ten
t,
ad
d
in
g
s
co
r
e
-
lev
el
f
u
s
io
n
h
el
p
s
s
tab
ilize
p
er
f
o
r
m
an
ce
ac
r
o
s
s
d
atasets
.
T
h
u
s
,
r
e
ce
n
t
ev
id
en
ce
s
u
p
p
o
r
ts
th
e
id
ea
th
at
s
co
r
e
-
lev
el
f
u
s
io
n
o
f
f
er
s
a
v
er
y
f
av
o
r
ab
le
tr
ad
e
-
o
f
f
(
m
o
d
u
lar
ity
,
lo
wer
d
ata
s
h
ar
in
g
d
em
a
n
d
s
,
ea
s
ier
i
m
p
lem
en
tatio
n
)
.
I
n
g
en
er
al,
f
u
s
io
n
tech
n
iq
u
es
f
all
in
to
th
r
ee
m
ain
g
r
o
u
p
s
:
tr
an
s
f
o
r
m
atio
n
-
b
ased
,
d
en
s
ity
-
b
ased
,
an
d
class
if
ier
-
b
ased
m
eth
o
d
s
[
1
4
]
.
T
r
an
s
f
o
r
m
atio
n
-
b
ased
a
p
p
r
o
a
ch
es
r
escale
r
aw
m
atch
in
g
s
c
o
r
es
o
n
to
a
u
n
if
ied
d
o
m
ain
th
r
o
u
g
h
th
e
u
s
e
o
f
n
o
r
m
aliza
tio
n
m
eth
o
d
s
,
in
cl
u
d
in
g
m
in
–
m
ax
s
ca
lin
g
an
d
z
-
s
co
r
e,
tan
h
f
u
n
ctio
n
s
,
o
r
d
o
u
b
le
-
s
ig
m
o
id
m
ap
p
in
g
s
[
1
5
]
.
T
h
e
n
o
r
m
alize
d
s
co
r
es
ar
e
th
en
co
m
b
i
n
ed
u
s
in
g
r
u
les
s
u
c
h
as
s
u
m
,
weig
h
ted
s
u
m
,
m
in
,
m
ax
,
tr
ia
n
g
u
lar
n
o
r
m
s
,
o
r
p
r
o
d
u
ct
[
1
6
]
–
[
1
9
]
.
Den
s
ity
-
b
ased
f
u
s
io
n
u
s
es
esti
m
ated
p
r
o
b
a
b
ilit
y
d
is
tr
ib
u
tio
n
s
o
f
g
en
u
in
e
an
d
im
p
o
s
to
r
s
co
r
es
to
ca
lcu
late
lik
elih
o
o
d
r
atio
s
[
2
0
]
,
[
2
1
]
.
C
lass
if
ier
-
b
ased
m
eth
o
d
s
tr
ea
t
co
m
b
in
e
d
s
co
r
es
as
f
ea
tu
r
e
v
ec
to
r
s
an
d
class
if
y
th
em
as
g
en
u
in
e
o
r
im
p
o
s
to
r
u
s
in
g
m
o
d
els
s
u
ch
as
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
(
SVM)
[
2
2
]
,
[
2
3
]
,
h
id
d
e
n
M
ar
k
o
v
m
o
d
els
(
HM
M)
[
2
4
]
,
m
u
ltil
ay
er
p
e
r
ce
p
tr
o
n
s
(
ML
P)
[
2
5
]
,
o
r
B
ay
esian
class
if
ier
s
with
G
au
s
s
ian
m
ix
tu
r
e
m
o
d
els
(
GM
Ms)
[
2
6
]
.
B
u
ild
in
g
o
n
th
ese
ca
teg
o
r
ies,
r
esear
ch
ef
f
o
r
ts
h
av
e
co
n
tin
u
ally
ex
p
lo
r
e
d
n
ew
s
tr
ateg
ies
to
en
h
an
ce
s
co
r
e
f
u
s
io
n
p
er
f
o
r
m
an
ce
.
E
ar
ly
ap
p
r
o
ac
h
es
r
elied
m
ain
ly
o
n
s
im
p
le
co
m
b
in
atio
n
r
u
les,
s
u
ch
as
th
e
weig
h
ted
s
u
m
[
2
7
]
,
wh
ile
later
m
eth
o
d
s
em
p
lo
y
ed
m
o
r
e
ad
v
an
ce
d
tech
n
iq
u
es
lik
e
ar
tific
ial
n
eu
r
al
n
et
wo
r
k
s
(
ANN)
[
2
8
]
an
d
f
u
zz
y
lo
g
ic
s
y
s
tem
s
(
FLS)
[
2
9
]
.
N
u
m
er
o
u
s
s
tu
d
ies
h
av
e
co
n
f
ir
m
ed
t
h
e
ef
f
ec
tiv
en
ess
o
f
th
ese
ap
p
r
o
ac
h
es.
Fo
r
in
s
tan
ce
,
Am
m
o
u
r
et
a
l
.
[
3
0
]
r
ep
o
r
ted
th
at
co
m
b
in
i
n
g
f
ac
e
an
d
ir
is
s
co
r
es
u
s
in
g
a
weig
h
ted
s
u
m
s
ig
n
if
ican
tly
im
p
r
o
v
es
id
en
tifi
ca
tio
n
ac
cu
r
ac
y
.
Similar
ly
,
Al
h
ar
b
i
an
d
Als
h
an
b
ar
i
[
3
1
]
f
o
u
n
d
th
at
f
u
s
in
g
f
ac
e
an
d
v
o
ice
m
o
d
alities
at
th
e
s
co
r
e
lev
el
r
ed
u
ce
s
th
e
E
E
R
.
Ho
w
ev
er
,
th
ese
i
m
p
r
o
v
em
en
ts
ar
e
o
f
ten
ac
c
o
m
p
an
ie
d
b
y
h
ig
h
er
co
m
p
u
tatio
n
al
d
em
a
n
d
s
an
d
g
r
ea
ter
s
en
s
itiv
ity
to
o
u
tlier
s
.
W
eig
h
ted
f
u
s
io
n
m
et
h
o
d
s
r
eq
u
ir
e
d
ataset
-
s
p
ec
if
ic
p
ar
am
et
er
tu
n
in
g
,
wh
i
ch
lim
its
g
en
er
aliza
tio
n
,
wh
er
e
as
m
ax
f
u
s
io
n
,
alth
o
u
g
h
ef
f
icie
n
t,
ca
n
b
e
u
n
s
tab
le
d
u
e
to
its
r
elian
ce
o
n
th
e
h
ig
h
e
s
t sco
r
e,
wh
ich
m
ay
b
e
n
o
is
y
.
R
ec
en
t
s
tu
d
ies
h
av
e
in
cr
ea
s
in
g
ly
em
p
h
asized
s
co
r
e
-
lev
el
f
u
s
io
n
with
in
d
ee
p
lear
n
in
g
f
r
am
ewo
r
k
s
,
p
ar
ticu
lar
ly
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN
)
-
b
ased
ar
ch
itectu
r
es,
wh
ich
ef
f
ec
tiv
ely
lear
n
cr
o
s
s
-
m
o
d
al
r
elatio
n
s
h
ip
s
an
d
p
er
f
o
r
m
n
o
n
-
lin
ea
r
f
u
s
io
n
.
Fo
r
in
s
tan
ce
,
R
iaz
et
a
l
.
[
3
2
]
d
em
o
n
s
tr
ated
th
at
f
u
s
in
g
C
NN
em
b
ed
d
in
g
s
with
au
x
iliar
y
f
a
cial
m
ar
k
s
at
th
e
s
co
r
e
lev
el
im
p
r
o
v
e
d
f
ac
e
v
er
if
icatio
n
ac
c
u
r
ac
y
o
n
t
h
e
I
J
B
-
A
d
ataset.
Similar
ly
,
Sh
in
d
e
an
d
Kay
te
[
3
3
]
s
h
o
we
d
th
at
m
u
lti
m
o
d
al
C
NN
f
u
s
io
n
o
f
f
ac
e
an
d
f
in
g
e
r
p
r
in
t
f
ea
tu
r
es
u
s
in
g
a
weig
h
ted
s
co
r
e
co
m
b
i
n
atio
n
o
u
t
p
er
f
o
r
m
ed
u
n
im
o
d
a
l
r
ec
o
g
n
itio
n
.
M
o
r
eo
v
er
,
Far
h
ad
ip
o
u
r
et
a
l
.
[
3
4
]
f
u
r
th
er
co
n
f
ir
m
e
d
th
e
ef
f
ec
tiv
en
ess
o
f
s
co
r
e
-
lev
el
f
u
s
io
n
b
y
in
teg
r
atin
g
Face
Net
an
d
Vo
iceN
et
em
b
ed
d
in
g
s
,
ac
h
iev
in
g
s
u
p
er
io
r
m
u
ltimo
d
a
l
r
ec
o
g
n
itio
n
co
m
p
ar
ed
to
s
in
g
le
-
m
o
d
ality
s
y
s
tem
s
.
C
o
n
s
eq
u
en
tly
,
r
esear
ch
er
s
ar
e
r
ev
is
itin
g
tr
ad
itio
n
al
f
u
s
io
n
o
p
er
ato
r
s
an
d
s
tatis
tical
f
r
am
ewo
r
k
s
th
at
o
f
f
er
co
m
p
etitiv
e
ac
cu
r
ac
y
with
lo
wer
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
6
4
8
-
6
6
2
650
co
m
p
u
tatio
n
al
co
s
ts
,
h
ig
h
lig
h
t
in
g
th
e
n
ee
d
f
o
r
a
s
im
p
le
y
et
e
f
f
icien
t f
u
s
io
n
s
tr
ateg
y
th
at
b
al
an
ce
s
p
er
f
o
r
m
an
ce
an
d
co
m
p
lex
ity
.
I
n
r
e
s
p
o
n
s
e
t
o
t
h
es
e
c
h
a
ll
e
n
g
e
s
,
m
e
a
n
o
p
e
r
a
t
o
r
s
e
m
e
r
g
e
as
a
s
i
m
p
l
e
y
e
t
p
o
w
e
r
f
u
l
al
t
e
r
n
a
ti
v
e
f
o
r
s
c
o
r
e
-
l
e
v
e
l
f
u
s
i
o
n
.
C
l
as
s
i
c
a
l
m
e
a
n
s
;
a
r
i
t
h
m
e
t
i
c
,
g
e
o
m
e
t
r
ic
,
q
u
a
d
r
a
t
ic
,
a
n
d
h
a
r
m
o
n
i
c
,
s
e
r
v
e
as
f
u
n
d
a
m
e
n
t
a
l
s
t
at
i
s
t
i
c
al
t
o
o
l
s
c
a
p
a
b
l
e
o
f
s
u
m
m
a
r
i
zi
n
g
c
o
m
p
l
e
x
d
a
t
a
w
h
i
l
e
m
i
t
i
g
at
i
n
g
t
h
e
e
f
f
e
c
t
s
o
f
n
o
i
s
e
a
n
d
o
u
t
l
i
er
s
[
3
5
]
,
[
3
6
]
.
T
h
e
i
r
s
i
m
p
l
e
m
a
t
h
e
m
at
i
c
al
s
t
r
u
c
t
u
r
e
a
n
d
c
l
ea
r
i
n
te
r
p
r
e
t
a
ti
o
n
m
a
k
e
m
e
a
n
-
b
a
s
e
d
f
u
s
i
o
n
m
e
t
h
o
d
s
u
s
e
f
u
l
f
o
r
d
i
f
f
e
r
e
n
t
b
i
o
m
e
t
r
i
c
s
et
u
p
s
,
s
u
c
h
as
m
u
l
t
i
-
s
e
n
s
o
r
,
m
u
l
t
i
-
i
n
s
ta
n
c
e
,
a
n
d
h
y
b
r
i
d
s
y
s
t
e
m
s
.
T
h
e
g
e
n
e
r
a
l
i
ze
d
v
e
r
s
i
o
n
s
,
li
k
e
p
o
w
e
r
m
e
a
n
s
[
3
5
]
,
m
a
k
e
t
h
e
s
e
m
e
t
h
o
d
s
e
v
e
n
m
o
r
e
f
l
e
x
i
b
l
e
b
y
a
l
l
o
w
i
n
g
t
h
e
f
u
s
i
o
n
p
r
o
c
e
s
s
t
o
a
d
a
p
t
a
n
d
b
a
l
a
n
c
e
b
e
t
w
e
e
n
a
c
c
u
r
a
c
y
a
n
d
c
o
m
p
u
t
a
t
i
o
n
a
l
co
s
t
.
U
n
l
i
k
e
m
a
n
y
a
d
v
a
n
c
e
d
f
u
s
i
o
n
t
e
c
h
n
i
q
u
e
s
,
m
ea
n
-
b
a
s
e
d
m
e
t
h
o
d
s
d
o
n
o
t
n
e
e
d
t
r
a
i
n
i
n
g
o
r
p
a
r
a
m
e
t
e
r
a
d
j
u
s
t
m
en
t
,
m
a
k
i
n
g
t
h
e
m
s
u
it
a
b
l
e
,
e
f
f
i
cie
n
t
,
a
n
d
r
e
l
i
a
b
l
e
f
o
r
r
e
al
-
w
o
r
l
d
b
i
o
m
e
t
r
i
c
s
y
s
t
e
m
s
.
T
h
is
s
tu
d
y
e
x
p
lo
r
es
th
e
p
o
ten
tial
o
f
m
ea
n
o
p
er
ato
r
s
f
o
r
s
co
r
e
-
lev
el
f
u
s
io
n
in
m
u
ltimo
d
al
b
io
m
etr
ic
s
y
s
tem
s
.
Usi
n
g
th
r
ee
d
atasets
:
Natio
n
al
I
n
s
titu
te
o
f
Stan
d
a
r
d
s
an
d
T
ec
h
n
o
lo
g
y
(
NI
ST
)
f
i
n
g
e
r
p
r
in
t,
NI
ST
f
ac
es,
an
d
XM
2
VT
S,
we
s
y
s
tem
atica
lly
ev
alu
ate
m
u
ltip
le
m
ea
n
f
o
r
m
u
latio
n
s
in
m
u
lti
-
alg
o
r
ith
m
,
m
u
lti
-
in
s
tan
ce
,
an
d
m
u
ltimo
d
al
s
ce
n
ar
io
s
.
Ou
r
c
o
n
tr
ib
u
tio
n
s
ar
e
th
r
ee
f
o
ld
:
f
i
r
s
t,
we
d
em
o
n
s
tr
ate
th
e
ef
f
ec
tiv
e
n
ess
o
f
m
ea
n
-
b
ased
o
p
er
ato
r
s
in
en
h
a
n
cin
g
r
ec
o
g
n
itio
n
p
er
f
o
r
m
an
ce
w
h
ile
m
in
im
izin
g
co
m
p
lex
ity
;
s
ec
o
n
d
,
we
p
r
o
v
id
e
a
co
m
p
ar
ativ
e
a
n
aly
s
is
o
f
d
if
f
e
r
en
t
m
ea
n
f
o
r
m
u
latio
n
s
ac
r
o
s
s
m
u
ltip
le
d
atasets
;
an
d
f
in
ally
,
we
h
ig
h
lig
h
t
th
ei
r
r
esil
ien
ce
to
n
o
is
e
an
d
o
u
tlie
r
s
,
p
o
s
itio
n
in
g
th
em
as
a
p
r
a
ctica
l
s
o
lu
tio
n
f
o
r
s
ca
lab
le,
r
ea
l
-
wo
r
ld
b
io
m
etr
ic
au
th
en
ticatio
n
.
2.
P
RO
P
O
SE
D
M
E
T
H
O
D
T
h
is
wo
r
k
s
e
e
k
s
t
o
s
t
r
e
n
g
t
h
en
b
o
t
h
t
h
e
p
e
r
f
o
r
m
a
n
c
e
a
n
d
d
e
p
e
n
d
a
b
i
l
i
t
y
o
f
b
i
o
m
e
t
r
i
c
r
e
c
o
g
n
i
t
i
o
n
b
y
i
n
t
e
g
r
a
t
i
n
g
s
c
o
r
e
s
f
r
o
m
d
i
f
f
e
r
en
t
m
o
d
a
l
i
ti
e
s
i
n
a
s
i
m
p
l
e
,
a
d
a
p
t
a
b
l
e
,
a
n
d
t
r
a
i
n
i
n
g
-
f
r
e
e
m
a
n
n
e
r
.
I
n
s
t
e
a
d
o
f
r
e
l
y
i
n
g
o
n
c
o
m
p
l
e
x
f
u
s
i
o
n
a
l
g
o
r
i
t
h
m
s
,
t
h
e
p
r
o
p
o
s
e
d
m
e
t
h
o
d
u
s
e
s
a
m
a
t
h
e
m
a
t
i
c
a
l
o
p
e
r
a
t
o
r
t
h
a
t
ca
n
e
a
s
i
l
y
a
d
j
u
s
t
i
ts
b
e
h
a
v
i
o
r
t
o
d
i
f
f
e
r
e
n
t
b
i
o
m
e
t
r
ic
s
c
e
n
a
r
i
o
s
.
B
y
c
h
a
n
g
i
n
g
a
s
in
g
l
e
p
a
r
a
m
e
t
e
r
,
t
h
e
s
a
m
e
f
o
r
m
u
l
a
t
i
o
n
c
a
n
a
c
t
l
i
k
e
s
e
v
e
r
a
l
w
e
ll
-
k
n
o
w
n
f
u
s
i
o
n
r
u
l
e
s
,
a
l
l
o
w
i
n
g
t
h
e
s
y
s
t
e
m
t
o
a
d
a
p
t
w
h
e
n
o
n
e
b
i
o
m
e
t
r
i
c
s
o
u
r
c
e
i
s
n
o
i
s
y
o
r
l
ess
r
eliab
le.
T
h
is
co
n
ce
p
t
is
b
ased
o
n
Ko
l
m
o
g
o
r
o
v
’
s
a
x
io
m
atic
th
e
o
r
y
o
f
th
e
m
ea
n
[
3
5
]
,
[
3
7
]
,
wh
ich
p
r
o
v
id
es
a
u
n
if
ied
f
o
u
n
d
atio
n
f
o
r
d
esig
n
in
g
an
d
an
aly
zin
g
v
ar
i
o
u
s
m
ea
n
f
u
n
ctio
n
s
u
s
ed
in
s
co
r
e
-
lev
el
f
u
s
io
n
.
L
et
is
an
in
p
u
t
v
ec
to
r
d
e
f
in
e
d
as:
=
(
1
,
⋯
,
)
.
A
r
eg
u
lar
m
ea
n
is
a
f
u
n
ctio
n
:
→
th
at
s
atis
f
ies th
e
f
o
llo
win
g
f
u
n
d
a
m
en
tal
p
r
o
p
e
r
ties
:
−
(
)
is
co
n
tin
u
o
u
s
an
d
m
o
n
o
to
n
ica
lly
in
cr
ea
s
in
g
with
r
esp
ec
t to
i
ts
ar
g
u
m
en
ts
−
(
)
is
s
y
m
m
etr
ic,
m
ea
n
in
g
its
v
al
u
e
is
in
v
ar
ian
t u
n
d
er
a
n
y
p
er
m
u
tatio
n
o
f
t
h
e
in
p
u
t e
lem
en
ts
−
T
h
e
m
ea
n
o
f
id
e
n
tical
v
alu
es r
etu
r
n
s
th
e
s
am
e
v
alu
e,
i.e
.
,
f
o
r
r
ep
ea
ted
in
p
u
ts
,
(
)
=
−
T
h
e
m
ea
n
o
f
a
co
m
b
in
ed
d
ata
s
et
r
em
ain
s
u
n
ch
an
g
e
d
wh
en
a
s
u
b
s
et
o
f
v
alu
es
is
r
ep
lace
d
b
y
th
eir
o
wn
m
ea
n
=
(
1
,
⋯
1
)
:
(
)
=
(
,
⋯
,
1
+
1
,
⋯
,
)
(
1
)
−
T
h
e
f
u
n
ctio
n
is
id
em
p
o
ten
t,
m
ea
n
in
g
th
at
a
p
p
ly
in
g
th
e
m
e
an
to
id
en
tical
i
n
p
u
t
v
alu
es
r
etu
r
n
s
th
e
s
am
e
v
alu
e
:
∀
∈
[
0
,
1
]
;
(
,
)
=
(
2
)
−
Ass
o
ciativ
ity
is
a
s
tr
o
n
g
er
q
u
a
lity
th
an
b
is
y
m
m
etr
y
:
[
0
,
1
]
4
;
[
(
,
)
,
(
,
)
]
=
[
(
,
)
,
(
,
)
]
(
3
)
A
m
ea
n
th
at
s
ati
s
f
ies
th
i
s
r
eq
u
ir
em
en
t,
wh
ile
r
em
ain
in
g
co
n
tin
u
o
u
s
an
d
s
tr
ictly
in
cr
ea
s
in
g
,
i
s
ty
p
ically
r
ep
r
esen
ted
b
y
th
e
f
o
llo
win
g
m
ath
em
atica
l f
o
r
m
:
∀
(
,
)
∈
[
0
,
1
]
2
,
(
,
)
=
−
1
[
(
)
+
(
)
2
]
(
4
)
Ko
lm
o
g
o
r
o
v
s
h
o
we
d
th
at
if
th
ese
co
n
d
itio
n
s
h
o
l
d
,
th
e
f
u
n
cti
o
n
(
)
ca
n
b
e
r
e
p
r
esen
ted
as
(
5
)
[
3
5
]
:
(
)
=
−
1
(
1
∑
(
)
=
1
)
(
5
)
w
h
er
e
is
th
e
co
n
tin
u
o
u
s
m
o
n
o
to
n
e,
−
1
is
th
e
in
v
er
s
e
f
u
n
ctio
n
,
a
n
d
is
th
e
v
ec
to
r
:
=
(
1
,
⋯
,
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
co
r
e
-
leve
l b
io
metric in
fo
r
ma
tio
n
fu
s
io
n
w
ith
g
en
era
liz
ed
p
o
w
er me
a
n
(
Leila
Hella
l
)
651
T
h
e
f
u
n
ctio
n
ad
ju
s
ts
th
e
d
y
n
am
ic
v
alu
es
p
r
io
r
to
th
ei
r
in
teg
r
atio
n
.
T
h
e
m
o
d
if
ied
v
alu
es
ar
e
s
u
b
s
eq
u
en
tly
co
m
b
in
ed
u
s
in
g
a
s
im
p
le
ar
ith
m
etic
av
er
ag
e
,
a
s
d
em
o
n
s
tr
ated
in
th
e
f
o
llo
win
g
ex
am
p
les
[
3
5
]
:
−
Ar
ith
m
etic
m
ea
n
: w
h
en
th
e
tr
a
n
s
f
o
r
m
atio
n
f
u
n
ctio
n
s
atis
f
ies
(
)
=
lead
s
to
:
=
1
∑
=
1
(
6
)
−
Fo
r
th
e
g
eo
m
etr
ic
m
ea
n
,
s
ettin
g
(
)
=
(
)
f
o
r
>
0
lead
s
to
:
l
o
g
(
)
=
(
∏
=
1
)
1
(
7
)
−
Fo
r
th
e
h
ar
m
o
n
ic
m
ea
n
,
s
ettin
g
(
)
=
1
with
>
0
r
esu
lts
in
:
1
(
)
=
∑
−
1
=
1
(
8
)
As s
h
o
wn
in
(
5
)
also
e
n
co
m
p
a
s
s
es o
th
er
ca
s
es,
in
clu
d
in
g
th
e
p
o
wer
m
ea
n
.
Fo
r
∈
[
0
,
∞
]
,
we
h
av
e
[
3
5
]
:
(
)
=
{
1
∑
=
1
}
1
(
)
=
(
,
⋯
,
1
+
1
,
⋯
,
)
(
9
)
T
h
e
Ko
lm
o
g
o
r
o
v
ax
io
m
atic
f
r
am
ewo
r
k
en
s
u
r
es
th
at
t
h
e
m
ea
n
o
p
er
ato
r
s
s
atis
f
y
im
p
o
r
tan
t
m
ath
em
atica
l
p
r
o
p
e
r
ties
s
u
ch
as
m
o
n
o
to
n
icity
,
c
o
n
tin
u
ity
,
an
d
s
y
m
m
etr
y
,
wh
ich
m
a
k
e
s
th
em
s
u
itab
le
f
o
r
r
eliab
ly
co
m
b
in
in
g
d
if
f
e
r
en
t
b
io
m
etr
ic
s
co
r
es.
B
ased
o
n
th
is
f
r
am
ewo
r
k
,
t
h
e
g
e
n
er
a
lized
p
o
wer
m
ea
n
in
tr
o
d
u
ce
s
a
p
ar
am
eter
(
)
th
at
co
n
tr
o
ls
h
o
w
th
e
s
co
r
es
ar
e
co
m
b
in
ed
.
T
h
is
allo
ws
o
n
e
f
o
r
m
u
la
to
r
ep
r
esen
t
s
ev
er
al
well
-
k
n
o
w
n
f
u
s
io
n
r
u
l
es.
W
h
en
(
)
is
s
m
all
o
r
n
eg
ati
v
e,
th
e
lo
west
s
co
r
e
h
as
m
o
r
e
in
f
lu
e
n
ce
o
n
th
e
f
in
al
r
esu
lt,
wh
ich
is
u
s
ef
u
l
wh
en
o
n
e
b
io
m
etr
ic
s
o
u
r
ce
m
ay
b
e
wea
k
o
r
n
o
is
y
.
W
h
e
n
(
)
is
p
o
s
itiv
e
an
d
in
cr
ea
s
es,
th
e
h
ig
h
er
s
co
r
es
g
ain
m
o
r
e
in
f
lu
en
ce
,
im
p
r
o
v
in
g
p
er
f
o
r
m
an
ce
wh
en
o
n
e
m
o
d
ality
is
s
tr
o
n
g
an
d
h
ig
h
ly
r
eliab
le.
T
h
er
ef
o
r
e,
b
y
ad
ju
s
tin
g
(
)
,
th
e
p
o
we
r
m
ea
n
c
an
f
lex
ib
l
y
ad
a
p
t
th
e
f
u
s
io
n
b
e
h
av
io
r
to
d
if
f
er
e
n
t
b
io
m
etr
ic
co
n
d
itio
n
s
[
3
5
]
,
[
3
7
]
.
As
s
h
o
wn
in
T
ab
le
1
,
th
e
g
en
er
alize
d
m
ea
n
ca
n
s
m
o
o
t
h
ly
tr
an
s
itio
n
f
r
o
m
m
in
im
u
m
to
m
ax
im
u
m
b
eh
a
v
io
r
,
d
e
p
en
d
in
g
o
n
th
e
ch
o
s
en
v
alu
e
o
f
(
).
T
ab
le
1
.
I
llu
s
tr
atio
n
s
o
f
v
ar
i
o
u
s
m
ea
n
f
u
n
ctio
n
s
co
r
r
esp
o
n
d
in
g
to
d
if
f
er
en
t v
al
u
es o
f
(
)
[
3
5
]
a
M
(
x
,
y
)
C
o
mm
e
n
t
-
∞
mi
n
(
x
,
y
)
Li
mi
t
v
a
l
u
e
-
1
2
+
H
a
r
mo
n
i
c
me
a
n
0
(
)
1
2
G
e
o
me
t
r
i
c
m
e
a
n
+1
+
2
A
r
i
t
h
m
e
t
i
c
mea
n
+2
√
2
+
2
2
Q
u
a
d
r
a
t
i
c
mea
n
+∞
max
(
x
,
y
)
Li
mi
t
v
a
l
u
e
T
o
ev
alu
ate
th
e
b
en
ef
it
o
f
em
p
h
asizin
g
h
ig
h
-
q
u
ality
in
f
o
r
m
at
io
n
,
th
is
s
tu
d
y
also
in
v
esti
g
ates
th
e
f
u
s
io
n
p
er
f
o
r
m
an
ce
u
s
in
g
a
p
o
wer
m
e
an
with
(
=
5
)
.
T
h
is
co
n
f
i
g
u
r
atio
n
in
cr
ea
s
es
th
e
in
f
lu
en
ce
o
f
s
tr
o
n
g
an
d
r
eliab
le
s
co
r
es wh
ile
s
u
p
p
r
ess
in
g
th
e
e
f
f
ec
t o
f
u
n
ce
r
tain
o
r
n
o
is
y
o
n
e
s
[
3
8
]
.
I
n
p
r
ac
tical
b
io
m
etr
ic
s
ce
n
ar
io
s
,
wh
er
e
th
e
q
u
ality
o
f
f
in
g
er
p
r
i
n
t
o
r
f
ac
e
s
am
p
les
m
ay
f
lu
ct
u
ate.
T
h
is
s
tr
ateg
y
en
h
an
ce
s
r
ec
o
g
n
itio
n
ac
cu
r
ac
y
with
o
u
t
in
cr
ea
s
in
g
co
m
p
u
tatio
n
al
co
m
p
lex
ity
,
t
h
er
eb
y
m
ai
n
tain
in
g
th
e
s
im
p
licity
an
d
ef
f
icien
cy
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
.
T
h
e
m
ea
n
o
p
er
at
o
r
is
also
p
r
ac
tical
f
r
o
m
a
c
o
m
p
u
tatio
n
al
p
o
in
t
o
f
v
iew.
Sin
ce
it
d
o
es
n
o
t
r
eq
u
ir
e
lear
n
in
g
,
weig
h
tin
g
,
o
r
co
m
p
le
x
co
m
p
a
r
is
o
n
s
,
it
ca
n
b
e
ca
lc
u
lated
v
er
y
q
u
ick
ly
.
T
h
is
m
ak
es
it
s
u
itab
le
f
o
r
r
ea
l
-
tim
e
b
io
m
etr
ic
s
y
s
tem
s
wh
er
e
f
ast d
ec
is
io
n
-
m
ak
in
g
is
n
ee
d
e
d
[
3
8
]
.
I
n
ad
d
itio
n
,
th
e
f
u
s
io
n
ca
n
b
e
p
r
o
ce
s
s
ed
in
p
ar
allel
f
o
r
m
u
ltip
le
u
s
er
s
o
r
m
u
ltip
le
s
co
r
e
in
p
u
ts
,
wh
ich
h
elp
s
m
ain
tain
h
i
g
h
s
y
s
tem
p
er
f
o
r
m
an
ce
ev
en
wh
en
th
e
n
u
m
b
er
o
f
u
s
er
s
in
cr
ea
s
es.
Fig
u
r
e
1
o
u
tlin
es
th
e
p
r
o
p
o
s
ed
s
co
r
e
-
lev
el
f
u
s
io
n
p
r
o
ce
s
s
u
s
in
g
m
ea
n
o
p
er
ato
r
s
.
T
h
e
s
y
s
tem
b
eg
in
s
b
y
ca
p
tu
r
in
g
b
i
o
m
etr
ic
d
ata,
s
u
ch
as
a
f
in
g
er
p
r
in
t
an
d
a
f
ac
e
im
ag
e,
f
r
o
m
th
e
s
am
e
in
d
iv
id
u
al.
E
ac
h
b
io
m
etr
i
c
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
6
4
8
-
6
6
2
652
s
o
u
r
ce
is
p
r
o
ce
s
s
ed
s
ep
ar
ately
to
e
x
tr
ac
t
d
is
cr
im
in
ativ
e
f
ea
tu
r
es.
T
h
ese
f
ea
tu
r
es
ar
e
t
h
en
co
m
p
ar
ed
with
th
e
co
r
r
esp
o
n
d
in
g
en
r
o
lled
tem
p
la
tes
in
th
e
d
atab
ase
to
g
en
er
ate
m
atch
in
g
s
co
r
es
[
1
]
,
[
4
]
,
[
5
]
.
B
ec
au
s
e
th
e
s
co
r
es
m
ay
co
m
e
f
r
o
m
d
if
f
er
en
t
s
y
s
tem
s
an
d
m
a
y
n
o
t sh
ar
e
th
e
s
am
e
n
u
m
er
ical
s
ca
le,
a
n
o
r
m
aliza
tio
n
s
tep
is
ap
p
lied
to
p
lace
th
em
with
in
a
u
n
if
ie
d
r
an
g
e
[
1
5
]
,
[
1
6
]
,
[
3
9
]
.
Af
ter
n
o
r
m
aliza
tio
n
,
t
h
e
s
co
r
es
ar
e
co
m
b
in
ed
u
s
in
g
th
e
g
en
er
alize
d
m
ea
n
o
p
er
ato
r
to
o
b
tain
a
s
in
g
le
f
u
s
ed
s
co
r
e
th
at
r
ep
r
esen
ts
th
e
f
in
al
m
atch
in
g
co
n
f
i
d
en
ce
.
T
h
is
f
u
s
ed
s
co
r
e
is
th
en
ev
alu
ated
ag
ain
s
t
a
p
r
ed
ef
in
e
d
th
r
esh
o
ld
.
I
f
it
m
ee
ts
o
r
ex
ce
ed
s
th
is
t
h
r
esh
o
ld
,
th
e
u
s
er
is
g
r
an
ted
ac
ce
s
s
;
o
th
er
wis
e,
th
e
attem
p
t
is
d
en
ied
.
B
y
in
co
r
p
o
r
atin
g
in
f
o
r
m
atio
n
f
r
o
m
m
u
ltip
l
e
b
io
m
etr
ic
s
o
u
r
ce
s
in
th
is
s
tr
u
ctu
r
ed
way
,
th
e
s
y
s
tem
en
h
an
ce
s
d
ec
is
io
n
ac
cu
r
ac
y
an
d
r
o
b
u
s
tn
ess
.
Fig
u
r
e
1
.
Ov
e
r
v
iew
o
f
th
e
p
r
o
p
o
s
ed
s
co
r
e
-
lev
el
f
u
s
io
n
f
r
am
ewo
r
k
u
s
in
g
m
ea
n
o
p
er
ato
r
s
Po
wer
m
ea
n
f
u
s
io
n
,
also
ca
lled
th
e
g
en
e
r
alize
d
m
ea
n
,
h
as
r
ec
en
tly
b
ec
o
m
e
p
o
p
u
lar
b
ec
a
u
s
e
it
ca
n
f
lex
ib
ly
c
o
m
b
in
e
d
if
f
er
en
t
k
i
n
d
s
o
f
in
f
o
r
m
atio
n
.
B
y
a
d
ju
s
tin
g
o
n
e
p
ar
am
eter
,
th
e
o
p
er
at
o
r
ca
n
ch
a
n
g
e
h
o
w
s
tr
o
n
g
ly
it
r
ea
cts
to
h
ig
h
o
r
l
o
w
in
p
u
t
v
alu
es,
m
a
k
in
g
it
u
s
ef
u
l
in
m
an
y
ap
p
licatio
n
s
.
Fo
r
ex
am
p
le,
in
d
ee
p
lear
n
in
g
,
it
is
u
s
ed
in
t
h
e
g
en
e
r
alize
d
m
ea
n
(
GeM
)
p
o
o
lin
g
tech
n
iq
u
e
to
b
alan
ce
th
e
b
eh
av
io
r
o
f
av
e
r
ag
e
an
d
m
ax
p
o
o
lin
g
,
wh
ich
h
elp
s
im
p
r
o
v
e
f
ea
t
u
r
e
r
ep
r
esen
tatio
n
in
im
ag
e
r
etr
iev
al
task
s
[
4
0
]
,
[
4
1
]
.
I
n
im
ag
e
p
r
o
ce
s
s
in
g
,
p
o
wer
m
ea
n
f
u
s
io
n
h
as
b
ee
n
s
u
cc
ess
f
u
lly
ap
p
lied
in
m
u
lti
-
ex
p
o
s
u
r
e,
m
u
lti
-
f
o
cu
s
,
an
d
m
ed
ical
im
ag
e
f
u
s
io
n
,
wh
e
r
e
c
o
n
tr
o
lli
n
g
th
e
p
o
wer
p
a
r
am
eter
h
elp
s
ac
h
iev
e
b
etter
co
n
tr
ast
an
d
d
etail
[
4
2
]
.
T
h
ese
s
tr
en
g
th
s
m
ak
e
th
e
p
o
wer
m
e
an
a
p
r
o
m
is
in
g
c
h
o
ice
f
o
r
b
io
m
etr
ic
s
y
s
tem
s
,
wh
er
e
in
f
o
r
m
atio
n
f
r
o
m
d
if
f
er
e
n
t
s
o
u
r
ce
s
m
u
s
t
b
e
co
m
b
in
e
d
ef
f
icien
tly
.
I
ts
ad
ap
tab
le
s
tr
u
ctu
r
e
is
p
ar
ticu
lar
ly
s
u
itab
le
f
o
r
s
co
r
e
-
lev
el
f
u
s
io
n
,
h
elp
in
g
to
p
r
o
d
u
ce
a
s
in
g
le,
r
e
liab
le
s
co
r
e
th
at
im
p
r
o
v
es th
e
f
in
al
r
ec
o
g
n
itio
n
d
ec
is
io
n
.
3.
M
E
T
H
O
D
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
d
atasets
,
ex
p
er
im
en
tal
s
etu
p
,
an
d
ev
alu
atio
n
m
etr
ics
u
s
ed
to
ass
es
s
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
f
u
s
io
n
ap
p
r
o
ac
h
.
T
h
e
aim
is
to
g
u
ar
a
n
tee
th
at
th
e
r
esu
lts
ar
e
r
eliab
le,
u
n
b
iased
,
an
d
r
ep
r
o
d
u
cib
le.
3
.
1
.
Da
t
a
ba
s
e
a
nd
ex
perim
e
nta
l pro
t
o
co
ls
E
v
alu
atio
n
o
f
t
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
was
co
n
d
u
cted
u
s
i
n
g
th
r
ee
b
io
m
etr
ic
d
atasets
:
th
e
NI
ST
f
in
g
er
p
r
in
t
d
atab
ase,
t
h
e
NI
ST
f
ac
e
d
atab
ase,
a
n
d
th
e
XM
2
V
T
S
m
u
ltimo
d
al
d
ata
b
ase
.
T
h
es
e
d
atasets
co
n
tain
f
in
g
er
p
r
in
t
an
d
f
ac
e
s
am
p
les
f
r
o
m
a
lar
g
e
n
u
m
b
er
o
f
u
s
er
s
,
allo
win
g
u
s
to
ev
al
u
ate
b
o
t
h
s
in
g
le
-
m
o
d
ality
an
d
m
u
ltimo
d
al
f
u
s
io
n
p
er
f
o
r
m
a
n
c
e
u
n
d
er
d
if
f
e
r
en
t c
o
n
d
itio
n
s
.
3.
1
.
1
.
NIS
T
-
f
ing
er
print
da
t
a
ba
s
e
T
h
e
NI
ST
f
in
g
er
p
r
in
t
d
atab
a
s
e
is
a
well
-
k
n
o
wn
an
d
p
u
b
licly
av
ailab
le
r
eso
u
r
ce
th
a
t
in
clu
d
es
f
in
g
er
p
r
in
t
s
am
p
les
f
r
o
m
m
an
y
in
d
iv
id
u
als.
I
t
p
r
o
v
id
es
two
f
in
g
er
p
r
in
t
s
co
r
es
f
o
r
ea
ch
p
er
s
o
n
,
o
n
e
f
r
o
m
th
e
r
ig
h
t
in
d
e
x
f
i
n
g
er
a
n
d
o
n
e
f
r
o
m
th
e
lef
t
in
d
ex
f
in
g
er
.
S
in
ce
ea
ch
u
s
er
h
as
two
s
am
p
les,
a
g
en
u
in
e
co
m
p
a
r
is
o
n
is
m
ad
e
b
y
m
atch
in
g
th
e
in
d
e
x
f
i
n
g
e
r
s
o
f
b
o
t
h
t
h
e
r
i
g
h
t
a
n
d
l
e
f
t
h
a
n
d
s
o
f
t
h
e
s
a
m
e
p
e
r
s
o
n
,
r
e
s
u
lt
i
n
g
i
n
6
,
0
0
0
g
e
n
u
i
n
e
s
c
o
r
e
s
(
o
n
e
p
e
r
u
s
e
r
)
.
T
o
g
e
n
e
r
a
t
e
i
m
p
o
s
t
o
r
s
c
o
r
es
,
f
in
g
e
r
p
r
i
n
t
s
f
r
o
m
d
i
f
f
e
r
e
n
t
u
s
e
r
s
a
r
e
c
o
m
p
a
r
e
d
.
E
a
c
h
u
s
e
r
’
s
f
i
n
g
e
r
p
r
i
n
t
is
m
at
c
h
e
d
ag
a
i
n
s
t
t
h
o
s
e
o
f
5
,
9
9
9
o
t
h
e
r
u
s
e
r
s
,
p
r
o
d
u
c
i
n
g
a
t
o
t
a
l
o
f
3
5
,
9
9
4
,
0
0
0
i
m
p
o
s
t
o
r
s
c
o
r
e
s
.
O
v
e
r
a
l
l
,
t
h
e
d
at
a
s
e
t
c
o
m
p
r
is
e
s
6
,
0
0
0
g
en
u
i
n
e
m
atch
s
co
r
es a
n
d
3
5
,
9
9
4
,
0
0
0
im
p
o
s
to
r
m
atch
s
co
r
es
[
4
3
]
,
[
4
4
]
.
3
.
1
.
2
.
NIS
T
-
f
a
ce
da
t
a
ba
s
e
T
h
e
NI
ST
-
Face
d
atab
ase
is
an
im
p
o
r
tan
t
r
eso
u
r
ce
f
o
r
f
ac
ial
r
ec
o
g
n
itio
n
r
esear
c
h
,
p
r
o
v
id
in
g
d
etailed
s
im
ilar
ity
s
co
r
e
v
ec
to
r
s
f
o
r
in
-
d
ep
th
an
al
y
s
is
.
I
t
co
n
tain
s
two
d
is
tin
ct
s
ets,
ea
ch
co
m
p
r
is
in
g
6
,
0
0
0
s
am
p
les
co
llected
f
r
o
m
3
,
0
0
0
i
n
d
iv
id
u
al
u
s
er
s
.
Similar
i
ty
s
co
r
es
we
r
e
g
e
n
er
ated
u
s
in
g
two
d
i
f
f
er
e
n
t
f
ac
e
r
ec
o
g
n
itio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
co
r
e
-
leve
l b
io
metric in
fo
r
ma
tio
n
fu
s
io
n
w
ith
g
en
era
liz
ed
p
o
w
er me
a
n
(
Leila
Hella
l
)
653
alg
o
r
ith
m
s
,
r
ef
er
r
ed
to
as
m
atc
h
er
C
an
d
m
atch
er
G.
Ov
e
r
all,
th
e
d
atab
ase
in
clu
d
es
6
,
0
0
0
g
e
n
u
in
e
m
atch
s
co
r
es,
co
r
r
esp
o
n
d
in
g
t
o
o
n
e
c
o
m
p
a
r
is
o
n
p
er
s
u
b
ject
p
air
,
an
d
1
7
,
9
9
4
,
0
0
0
im
p
o
s
to
r
m
atc
h
s
co
r
es,
r
esu
ltin
g
f
r
o
m
co
m
p
ar
is
o
n
s
o
f
ea
ch
s
am
p
le
with
th
e
r
em
ain
in
g
2
,
9
9
9
s
u
b
j
ec
ts
[
4
3
]
,
[
4
4
]
.
T
o
en
s
u
r
e
u
n
if
o
r
m
ity
an
d
f
ac
ilit
ate
f
air
co
m
p
ar
is
o
n
b
etwe
en
th
e
two
m
atch
er
s
,
tan
h
-
esti
m
ato
r
n
o
r
m
aliza
tio
n
was
ap
p
lied
to
th
eir
r
esp
ec
tiv
e
s
im
ilar
ity
s
co
r
es.
T
h
is
n
o
r
m
aliza
tio
n
s
tep
en
h
a
n
ce
s
th
e
co
n
s
is
ten
cy
an
d
r
eliab
ilit
y
o
f
th
e
p
er
f
o
r
m
an
ce
ev
alu
atio
n
f
o
r
b
o
th
f
ac
e
-
m
atc
h
in
g
s
y
s
tem
s
.
3
.
1
.
3
.
T
he
XM
2
VT
S da
t
a
s
et
a
nd
t
he
ev
a
lua
t
io
n pro
t
o
c
o
ls
Fo
r
b
o
th
s
p
ea
k
er
a
n
d
f
ac
e
v
e
r
if
icatio
n
task
s
,
th
e
XM
2
VT
S
d
atab
ase
was
u
tili
ze
d
ac
co
r
d
in
g
to
th
e
p
r
o
to
co
l
d
escr
ib
ed
in
[
4
5
]
.
T
h
e
d
ataset
in
clu
d
es
2
9
5
in
d
i
v
id
u
als,
d
iv
id
e
d
in
to
2
0
0
g
en
u
i
n
e
s
u
b
jects,
7
0
test
im
p
o
s
to
r
s
,
an
d
2
5
ev
alu
atio
n
i
m
p
o
s
to
r
s
.
Fo
r
f
ac
e
an
aly
s
is
,
th
r
ee
f
ea
tu
r
e
ty
p
es
wer
e
em
p
l
o
y
ed
:
f
ac
e
h
is
to
g
r
am
(
FH)
,
d
is
cr
ete
co
s
in
e
tr
an
s
f
o
r
m
s
m
all
(
DC
T
s
)
(
s
m
all
DC
T
f
ea
tu
r
es f
r
o
m
4
0
×3
2
im
ag
es),
an
d
DC
T
b
(
lar
g
er
DC
T
f
ea
tu
r
es
f
r
o
m
8
0
×6
4
im
ag
es).
Fo
r
s
p
ee
ch
,
f
ea
tu
r
es
in
clu
d
e
d
lin
ea
r
f
ilter
-
b
an
k
ce
p
s
tr
al
co
ef
f
icien
ts
(
L
FC
C
)
,
p
h
ase
au
to
-
c
o
r
r
elatio
n
m
el
f
i
lter
-
b
an
k
ce
p
s
tr
al
co
ef
f
icien
t
s
(
PAC
-
MF
C
C
)
,
an
d
s
p
ec
tr
al
s
u
b
b
an
d
ce
n
tr
o
i
d
(
SSC
)
,
all
co
m
b
in
ed
with
tem
p
o
r
al
d
e
r
iv
ativ
es
to
im
p
r
o
v
e
p
er
f
o
r
m
an
ce
.
T
o
en
s
u
r
e
s
y
s
tem
atic
tr
ain
in
g
an
d
ev
alu
atio
n
,
th
e
d
atab
ase
was d
iv
id
ed
in
to
t
h
r
ee
s
u
b
s
ets:
−
L
P
t
r
ain
:
t
h
is
s
et
tr
ain
s
clien
t m
o
d
els b
y
lear
n
in
g
p
atter
n
s
f
r
o
m
g
en
u
in
e
u
s
er
d
ata.
−
L
P
e
v
al:
t
h
is
s
et
d
eter
m
in
es
d
ec
is
io
n
th
r
esh
o
ld
s
,
h
elp
in
g
class
if
ier
s
d
is
tin
g
u
is
h
b
etwe
en
g
en
u
in
e
u
s
er
s
an
d
im
p
o
s
to
r
s
.
−
L
P
t
est:
t
h
is
f
in
al
s
et
as
s
es
s
es
o
v
er
all
s
y
s
tem
p
er
f
o
r
m
an
ce
u
s
in
g
u
n
s
ee
n
d
ata.
T
h
e
tr
ain
in
g
an
d
ev
al
u
atio
n
p
r
o
ce
d
u
r
es
ar
e
d
ef
in
e
d
b
y
two
p
r
o
to
co
ls
:
L
au
s
an
n
e
p
r
o
t
o
co
l
I
(
L
P1
)
an
d
L
au
s
an
n
e
p
r
o
t
o
co
l
I
I
(
L
P2
)
.
B
o
th
p
r
o
to
co
ls
u
s
e
th
e
s
am
e
test
s
et
[
4
5
]
,
[
4
6
]
,
en
s
u
r
i
n
g
f
air
p
er
f
o
r
m
an
ce
co
m
p
ar
is
o
n
.
L
P1
p
r
o
v
id
es
eig
h
t
s
am
p
les
p
er
u
s
er
,
wh
er
e
as
L
P2
in
clu
d
es
f
iv
e
T
ab
le
2
.
T
o
p
r
eser
v
e
d
ata
in
teg
r
ity
,
o
n
e
f
au
lty
s
p
ee
c
h
f
il
e
an
d
7
0
test
im
p
o
s
to
r
e
n
tr
ies
wer
e
r
em
o
v
ed
,
y
ield
in
g
a
to
tal
o
f
1
1
1
,
8
0
0
im
p
o
s
to
r
s
co
r
es
[
4
7
]
.
T
ab
le
2
.
Ov
e
r
v
iew
o
f
t
h
e
L
au
s
an
n
e
p
r
o
to
co
ls
f
o
r
th
e
XM
2
V
T
S d
atab
ase
[
4
7
]
D
a
t
a
s
e
t
s
La
u
s
a
n
n
e
p
r
o
t
o
c
o
l
s
F
u
si
o
n
p
r
o
t
o
c
o
l
LP1
LP2
Tr
a
i
n
c
l
i
e
n
t
a
c
c
e
sses
3
4
-
Ev
a
l
c
l
i
e
n
t
a
c
c
e
ss
e
s
6
0
0
(
3
×
2
0
0
)
4
0
0
(
2
×
2
0
0
)
F
u
si
o
n
d
e
v
Ev
a
l
i
mp
o
s
t
o
r
a
c
c
e
ss
e
s
4
0
.
0
0
0
(
2
5
×
8
×
2
0
0
)
F
u
si
o
n
d
e
v
Te
st
c
l
i
e
n
t
a
c
c
e
s
ses
4
0
0
(
2
×
2
0
0
)
F
u
si
o
n
e
v
a
Te
st
i
m
p
o
st
o
r
a
c
c
e
s
ses
1
1
2
.
0
0
0
(
7
0
×
8
×
2
0
0
)
F
u
si
o
n
e
v
a
Fo
r
class
if
icatio
n
,
two
m
ac
h
in
e
lear
n
in
g
m
o
d
els
ar
e
em
p
lo
y
ed
:
GM
Ms
an
d
ML
Ps
.
GM
Ms
ar
e
p
r
o
b
a
b
ilis
tic
m
o
d
els
th
at
h
an
d
le
v
ar
iab
le
-
len
g
th
f
ea
tu
r
e
s
eq
u
en
ce
s
b
y
f
ir
s
t
tr
ain
in
g
a
u
n
iv
er
s
al
b
ac
k
g
r
o
u
n
d
m
o
d
el
with
th
e
ex
p
ec
tatio
n
-
m
ax
im
izatio
n
(
E
M)
alg
o
r
ith
m
,
t
h
en
ad
ap
tin
g
it
to
ea
ch
u
s
er
t
h
r
o
u
g
h
m
ax
im
u
m
-
a
-
p
o
s
ter
io
r
i
(
MA
P)
a
d
ap
tatio
n
[
4
8
]
.
I
n
co
n
tr
ast,
ML
Ps
ar
e
tr
ain
e
d
with
s
to
ch
asti
c
er
r
o
r
b
ac
k
p
r
o
p
ag
atio
n
o
n
f
ix
ed
-
len
g
th
f
ea
tu
r
e
v
ec
to
r
s
to
d
is
tin
g
u
is
h
g
e
n
u
in
e
u
s
er
s
f
r
o
m
im
p
o
s
to
r
s
[
4
9
]
.
B
o
th
m
o
d
els
o
u
tp
u
t
a
m
atch
s
co
r
e
r
ep
r
esen
tin
g
th
e
lik
elih
o
o
d
th
at
th
e
in
p
u
t
m
atch
es
th
e
claim
ed
id
e
n
tity
.
T
h
is
s
co
r
e
is
c
o
m
p
ar
ed
ag
ain
s
t
a
d
ec
is
io
n
th
r
esh
o
ld
,
ca
lib
r
ate
d
with
th
e
L
P
e
v
al
u
atio
n
s
et,
to
b
alan
ce
th
e
f
alse
ac
ce
p
tan
ce
r
ate
(
FAR
)
an
d
f
alse
r
ejec
tio
n
r
ate
(
FR
R
)
.
H
ig
h
s
c
o
r
es
in
d
icate
g
en
u
in
e
m
atch
es
,
wh
ile
lo
w
s
co
r
es
s
u
g
g
est
im
p
o
s
to
r
s
.
GM
Ms
ar
e
p
ar
ticu
lar
ly
ef
f
ec
tiv
e
f
o
r
v
ar
ia
b
le
-
len
g
th
s
p
ee
ch
d
ata,
wh
e
r
ea
s
ML
Ps
ex
ce
l
at
ca
p
tu
r
in
g
d
is
cr
im
in
ativ
e
p
atter
n
s
in
f
ix
ed
-
s
ize
f
ac
e
f
ea
tu
r
es
[
5
0
]
.
C
o
m
b
in
in
g
th
eir
s
tr
en
g
th
s
en
h
an
ce
s
s
co
r
e
ac
cu
r
ac
y
,
v
e
r
if
ic
atio
n
p
er
f
o
r
m
an
ce
,
an
d
r
o
b
u
s
tn
ess
ag
ain
s
t im
p
o
s
to
r
attem
p
ts
.
3
.
2
.
No
r
m
a
liza
t
io
n a
nd
ev
a
l
ua
t
io
n m
et
rics
B
ef
o
r
e
ap
p
ly
in
g
th
e
m
ea
n
-
b
a
s
ed
f
u
s
io
n
r
u
le,
th
e
m
atch
in
g
s
co
r
es
f
r
o
m
d
if
f
er
e
n
t
b
io
m
etr
ic
s
y
s
tem
s
m
u
s
t
b
e
n
o
r
m
alize
d
to
en
s
u
r
e
th
ey
ar
e
co
m
p
ar
ab
le.
I
n
th
is
s
tu
d
y
,
m
i
n
-
m
ax
n
o
r
m
aliza
tio
n
i
s
ap
p
lied
d
u
e
to
its
s
im
p
licity
an
d
e
f
f
ec
tiv
en
ess
w
h
en
th
e
s
co
r
e
r
an
g
e
is
k
n
o
wn
[
1
5
]
,
[
3
9
]
.
T
h
e
tr
an
s
f
o
r
m
atio
n
m
ap
s
all
s
co
r
es
in
t
o
th
e
in
ter
v
al
[
0
,
1]
,
as e
x
p
r
ess
e
d
in
:
′
=
−
−
(1
0
)
w
h
er
e
′
th
e
n
o
r
m
alize
d
s
co
r
e
,
is
th
e
o
r
ig
in
al
s
co
r
e;
an
d
ar
e
th
e
m
in
im
u
m
a
n
d
m
a
x
im
u
m
s
co
r
es,
r
esp
ec
tiv
ely
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
6
4
8
-
6
6
2
654
Af
ter
n
o
r
m
aliza
tio
n
an
d
f
u
s
io
n
,
s
y
s
tem
p
er
f
o
r
m
an
ce
is
ev
alu
a
ted
u
s
in
g
s
tan
d
ar
d
b
io
m
etr
ic
v
er
if
icatio
n
m
etr
ics.
T
h
e
FAR
m
ea
s
u
r
es
h
o
w
o
f
ten
an
im
p
o
s
to
r
is
in
c
o
r
r
ec
tly
ac
ce
p
ted
as
a
g
en
u
in
e
u
s
er
,
wh
er
ea
s
th
e
FR
R
m
ea
s
u
r
es h
o
w
o
f
ten
a
g
en
u
in
e
u
s
er
is
in
co
r
r
ec
tly
r
ejec
ted
[
5
1
]
,
[
5
2
]
.
T
h
ey
a
r
e
ca
lcu
lated
a
s
f
o
llo
ws:
=
(
1
1
)
=
(
1
2
)
Fals
e
ac
ce
p
tan
ce
(
FA)
an
d
f
alse
r
ejec
tio
n
(
FR
)
r
ep
r
esen
t
f
alse
ac
ce
p
tan
ce
an
d
f
alse
r
ejec
tio
n
,
r
esp
ec
tiv
ely
,
wh
ile
n
u
m
b
er
o
f
im
p
o
s
to
r
ac
ce
s
s
es
(
NI
)
an
d
th
e
n
u
m
b
er
o
f
clien
t
ac
ce
s
s
es
(
NC
)
co
r
r
esp
o
n
d
to
th
e
to
tal
clien
t a
n
d
im
p
o
s
to
r
a
cc
ess
es
[
5
2
]
,
[
5
3
]
.
I
n
p
r
ac
tical
ter
m
s
,
a
s
ec
u
r
e
b
io
m
etr
ic
s
y
s
tem
s
h
o
u
ld
m
ain
tain
a
v
er
y
lo
w
FAR
to
r
ed
u
ce
th
e
r
is
k
o
f
u
n
au
th
o
r
ized
ac
ce
s
s
,
wh
ile
also
ac
h
iev
in
g
a
lo
w
FR
R
to
en
s
u
r
e
u
s
er
co
n
v
en
ie
n
ce
an
d
s
y
s
tem
u
s
ab
ilit
y
.
T
o
f
u
r
th
er
ass
ess
p
er
f
o
r
m
an
ce
,
th
e
g
en
u
in
e
ac
ce
p
tan
ce
r
ate
(
GAR)
r
ep
r
esen
ts
th
e
f
r
ac
tio
n
o
f
g
en
u
in
e
u
s
er
s
co
r
r
ec
tly
ac
ce
p
ted
:
=
1
−
(
1
3
)
Fo
r
a
co
m
p
r
eh
e
n
s
iv
e
p
er
f
o
r
m
a
n
ce
co
m
p
ar
is
o
n
,
GAR
an
d
FAR
cu
r
v
es
ar
e
p
l
o
tted
as
r
ec
eiv
e
r
o
p
er
ati
n
g
ch
ar
ac
ter
is
ti
c
(
R
OC
)
cu
r
v
es.
A
R
OC
cu
r
v
e
p
o
s
itio
n
ed
n
e
ar
er
to
t
h
e
u
p
p
er
-
lef
t
co
r
n
er
s
ig
n
if
ies
a
h
ig
h
er
-
p
er
f
o
r
m
in
g
s
y
s
tem
with
a
lo
wer
r
ate
o
f
ac
ce
p
tan
ce
er
r
o
r
s
[
5
2
]
,
[
5
4
]
.
T
h
e
d
ec
is
io
n
th
r
esh
o
l
d
p
lay
s
a
k
ey
r
o
le
in
d
eter
m
in
in
g
th
e
b
alan
ce
b
etw
ee
n
s
ec
u
r
ity
(
lo
w
FAR
)
an
d
ac
ce
s
s
ib
ilit
y
(
lo
w
F
R
R
)
.
L
o
w
er
in
g
th
e
th
r
esh
o
l
d
in
cr
ea
s
es
co
n
v
e
n
ien
ce
b
u
t
m
a
y
wea
k
en
s
ec
u
r
ity
,
wh
ile
i
n
cr
ea
s
in
g
it
lead
s
t
o
s
tr
o
n
g
er
s
ec
u
r
ity
at
th
e
c
o
s
t
o
f
m
o
r
e
r
ejec
tio
n
s
.
B
y
a
n
aly
zin
g
th
ese
cu
r
v
es,
th
e
p
r
o
p
o
s
ed
f
u
s
io
n
s
tr
ateg
y
d
em
o
n
s
tr
ates
its
ab
ilit
y
to
im
p
r
o
v
e
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
i
n
d
iv
id
u
al
m
o
d
alities
,
p
r
o
v
id
in
g
e
n
h
a
n
ce
d
d
is
cr
im
in
atio
n
ca
p
a
b
ilit
y
with
o
u
t
in
c
r
ea
s
in
g
s
y
s
tem
co
m
p
lex
ity
.
Fo
r
th
e
X
M2
VT
S
m
u
ltimo
d
al
d
atab
ase,
th
e
s
y
s
tem
’
s
p
er
f
o
r
m
an
ce
is
ass
es
s
ed
u
s
in
g
th
e
h
alf
to
tal
er
r
o
r
r
ate
(
HT
E
R
)
,
wh
ich
is
th
e
p
r
im
ar
y
m
etr
ic
s
p
ec
if
ied
in
th
e
XM
2
VT
S
b
en
ch
m
ar
k
p
r
o
to
c
o
l,
d
ef
in
ed
as
(
1
4
)
.
=
+
2
(
1
4
)
A
lo
wer
HT
E
R
v
alu
e
r
ef
lects
a
m
o
r
e
ac
cu
r
ate
an
d
s
ec
u
r
es
s
y
s
tem
,
as
it
in
d
icate
s
b
o
th
f
ewe
r
f
alse
ac
ce
p
tan
ce
s
an
d
f
ewe
r
f
alse
r
ejec
tio
n
s
[
4
5
]
,
[
4
7
]
.
T
h
is
m
ak
es
HT
E
R
p
ar
ticu
lar
ly
s
u
ita
b
le
f
o
r
m
u
ltimo
d
al
b
io
m
etr
ic
ev
alu
atio
n
,
wh
er
e
a
s
tr
o
n
g
b
alan
ce
b
etwe
en
u
s
ab
il
ity
an
d
s
ec
u
r
ity
is
r
e
q
u
ir
ed
.
T
o
ev
alu
ate
wh
eth
er
th
e
p
er
f
o
r
m
an
ce
im
p
r
o
v
em
e
n
t
ac
h
ie
v
ed
b
y
th
e
p
r
o
p
o
s
ed
f
u
s
io
n
m
eth
o
d
is
s
tatis
t
ically
s
ig
n
if
ican
t,
a
p
air
ed
t
-
test
is
co
n
d
u
cted
b
etwe
en
th
e
b
est
in
d
iv
id
u
al
m
atch
er
a
n
d
th
e
f
u
s
ed
s
co
r
es
[
5
5
]
,
[
5
6
]
.
T
h
is
test
ex
am
in
es
wh
eth
er
th
e
m
ea
n
d
if
f
e
r
en
ce
b
etwe
en
th
e
two
s
ets
o
f
g
en
u
i
n
e
s
co
r
es
is
g
r
ea
ter
th
an
ze
r
o
.
T
h
e
t
-
s
tatis
tic
is
ca
l
cu
lated
u
s
in
g
th
e
(
1
5
)
.
=
̅
/
√
(
1
5
)
w
h
er
e
̅
r
ep
r
esen
ts
th
e
m
ea
n
d
i
f
f
er
en
ce
b
etwe
en
p
air
ed
g
en
u
i
n
e
s
co
r
es,
d
en
o
tes
th
e
s
tan
d
a
r
d
d
e
v
iatio
n
o
f
th
ese
d
if
f
er
en
ce
s
,
an
d
is
th
e
to
tal
n
u
m
b
e
r
o
f
p
air
ed
g
en
u
in
e
co
m
p
ar
is
o
n
s
[
5
6
]
.
P
-
v
alu
es
b
elo
w
0
.
0
5
ar
e
r
eg
ar
d
e
d
as
s
tatis
tically
s
ig
n
i
f
ican
t,
s
u
g
g
esti
n
g
th
at
th
e
o
b
s
er
v
ed
p
er
f
o
r
m
a
n
ce
g
ai
n
s
ar
is
e
f
r
o
m
th
e
f
u
s
io
n
s
tr
ateg
y
an
d
n
o
t f
r
o
m
r
a
n
d
o
m
ef
f
ec
ts
.
4.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S AN
D
D
I
SC
USS
I
O
N
T
h
is
s
ec
tio
n
r
ep
o
r
ts
th
e
e
x
p
er
i
m
en
tal
ass
ess
m
en
t
o
f
th
e
p
r
o
p
o
s
ed
m
ea
n
-
b
ased
s
co
r
e
f
u
s
io
n
ap
p
r
o
ac
h
.
T
h
e
f
u
s
io
n
p
er
f
o
r
m
a
n
ce
is
an
a
ly
ze
d
u
s
in
g
s
tan
d
ar
d
b
io
m
etr
i
c
m
etr
ics,
an
d
th
e
im
p
ac
t
o
f
th
e
p
o
wer
p
ar
am
eter
(
)
o
n
r
ec
o
g
n
itio
n
ac
cu
r
ac
y
is
a
s
s
es
s
ed
.
T
h
e
f
o
llo
win
g
s
u
b
s
ec
tio
n
s
d
etail
th
e
r
esu
lts
o
b
tain
e
d
o
n
ea
c
h
d
ataset.
4
.
1
.
P
er
f
o
r
m
a
nce
o
f
m
ea
n
o
pera
t
o
r
-
ba
s
ed
f
us
io
n o
n NIS
T
f
ing
er
print
da
t
a
ba
s
e
T
h
e
NI
ST
f
in
g
er
p
r
in
t
d
ata
b
ase
is
u
s
ed
f
ir
s
t
to
ev
alu
ate
t
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
p
r
o
p
o
s
ed
s
co
r
e
-
lev
el
f
u
s
io
n
a
p
p
r
o
ac
h
in
a
m
u
lti
-
in
s
tan
ce
b
io
m
etr
ic
s
ce
n
ar
io
.
Sin
c
e
th
e
le
f
t
an
d
r
ig
h
t
f
in
g
e
r
p
r
in
t
s
o
f
t
h
e
s
am
e
u
s
er
m
ay
v
a
r
y
in
q
u
ality
,
r
el
y
in
g
o
n
o
n
l
y
o
n
e
in
s
tan
ce
ca
n
lea
d
t
o
r
ed
u
ce
d
a
cc
u
r
ac
y
o
r
in
co
n
s
is
ten
t
d
ec
is
io
n
s
.
B
y
co
m
b
in
in
g
th
e
m
atch
in
g
s
co
r
es
th
r
o
u
g
h
t
h
e
g
en
e
r
alize
d
m
ea
n
o
p
e
r
ato
r
,
th
e
s
y
s
tem
c
an
b
en
ef
it
f
r
o
m
th
e
s
tr
en
g
th
s
o
f
b
o
th
f
i
n
g
er
p
r
in
t
in
s
tan
ce
s
an
d
o
b
tain
a
m
o
r
e
r
elia
b
le
id
en
tific
atio
n
o
u
tco
m
e
.
T
h
e
f
o
llo
win
g
a
n
aly
s
is
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
co
r
e
-
leve
l b
io
metric in
fo
r
ma
tio
n
fu
s
io
n
w
ith
g
en
era
liz
ed
p
o
w
er me
a
n
(
Leila
Hella
l
)
655
co
m
p
ar
es
th
e
p
er
f
o
r
m
an
ce
o
f
in
d
iv
id
u
al
in
s
tan
ce
s
ag
ain
s
t
s
ev
er
al
m
ea
n
-
b
ased
f
u
s
io
n
r
u
les
to
h
ig
h
lig
h
t
th
e
b
en
ef
its
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
.
Fig
u
r
e
2
p
r
esen
ts
t
h
e
R
OC
cu
r
v
es
d
er
iv
ed
f
r
o
m
th
e
NI
ST
f
i
n
g
er
p
r
i
n
t
d
ata
b
ase
f
o
r
b
o
th
th
e
lef
t
an
d
r
ig
h
t
in
d
ex
f
in
g
e
r
s
,
alo
n
g
with
th
e
p
r
o
p
o
s
ed
s
co
r
e
-
lev
el
f
u
s
io
n
u
s
in
g
m
ea
n
o
p
er
at
o
r
s
.
I
n
d
i
v
id
u
ally
,
th
e
lef
t
an
d
r
ig
h
t
in
d
ex
f
in
g
er
s
ac
h
iev
e
GARs
o
f
7
5
.
5
%
an
d
8
3
%,
r
e
s
p
ec
tiv
ely
,
at
a
FA
R
o
f
0
.
0
1
%,
h
ig
h
lig
h
tin
g
th
e
p
er
f
o
r
m
an
ce
v
a
r
iatio
n
b
etwe
e
n
th
e
two
b
i
o
m
etr
ic
tr
a
its
.
W
h
en
th
e
m
ea
n
o
p
er
ato
r
-
b
ase
d
f
u
s
io
n
is
ap
p
lied
,
th
e
o
v
er
all
r
ec
o
g
n
itio
n
im
p
r
o
v
es
s
ig
n
if
ic
an
tly
.
T
h
e
r
esu
lts
s
h
o
w
th
at
th
e
ar
ith
m
eti
c
m
ea
n
r
ea
ch
es
a
GAR
o
f
9
1
.
2
0
%,
wh
ile
t
h
e
h
ar
m
o
n
ic
m
ea
n
y
ield
s
8
5
.
5
0
%
at
th
e
s
am
e
s
ec
u
r
ity
th
r
esh
o
ld
.
T
h
e
b
est
p
er
f
o
r
m
a
n
ce
is
ac
h
iev
ed
u
s
in
g
th
e
q
u
ad
r
atic
m
ea
n
wit
h
a
GAR
o
f
9
1
.
5
0
%,
f
o
llo
wed
clo
s
ely
b
y
th
e
p
o
wer
m
ea
n
with
(
=
5
)
at
9
0
.
9
0
%.
T
h
ese
im
p
r
o
v
em
en
ts
co
n
f
ir
m
th
at
s
co
r
e
-
lev
el
f
u
s
io
n
ef
f
ec
tiv
el
y
ex
p
l
o
its
co
m
p
lem
en
tar
y
i
n
f
o
r
m
atio
n
in
m
u
lti
-
in
s
tan
ce
f
in
g
er
p
r
in
ts
.
Fig
u
r
e
2
.
R
OC
cu
r
v
es o
f
i
n
d
iv
id
u
al
m
atch
er
s
an
d
m
ea
n
-
b
ased
s
co
r
e
-
lev
el
f
u
s
io
n
tech
n
i
q
u
e
s
T
h
e
R
OC
cu
r
v
es
also
s
h
if
t
n
o
ticea
b
ly
to
war
d
th
e
u
p
p
er
-
lef
t
co
r
n
er
co
m
p
a
r
ed
to
th
e
in
d
iv
id
u
al
m
o
d
alities
,
in
d
icatin
g
s
tr
o
n
g
er
d
is
cr
im
in
atio
n
ca
p
ab
ilit
y
an
d
r
ed
u
ce
d
ac
ce
p
tan
ce
e
r
r
o
r
s
.
T
h
e
s
u
p
er
i
o
r
p
er
f
o
r
m
an
ce
o
f
th
e
q
u
ad
r
atic
an
d
p
o
wer
m
ea
n
s
ca
n
b
e
ex
p
l
ain
ed
b
y
t
h
eir
a
b
ilit
y
to
g
iv
e
m
o
r
e
weig
h
t
to
h
ig
h
-
co
n
f
id
en
ce
s
co
r
es,
w
h
ile
m
in
i
m
izin
g
th
e
n
eg
ativ
e
in
f
l
u
en
ce
o
f
u
n
ce
r
tain
m
atc
h
es,
an
o
b
s
er
v
atio
n
c
o
n
s
is
ten
t
with
p
r
io
r
f
i
n
d
in
g
s
in
m
u
ltim
o
d
al
f
u
s
io
n
s
tu
d
ies
[
1
0
]
,
[
3
8
]
,
[
4
4
]
.
T
ab
le
3
r
ein
f
o
r
ce
s
th
e
s
e
o
b
s
er
v
atio
n
s
b
y
d
em
o
n
s
tr
atin
g
t
h
at
th
e
q
u
ad
r
a
tic
an
d
h
ig
h
-
p
o
wer
m
ea
n
o
p
e
r
ato
r
s
co
n
s
is
ten
tly
ac
h
iev
e
lo
wer
er
r
o
r
r
ates
th
a
n
th
e
u
n
im
o
d
al
f
in
g
e
r
p
r
in
t sy
s
te
m
s
.
T
ab
le
3
.
Per
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
o
f
v
a
r
io
u
s
f
u
s
io
n
ap
p
r
o
ac
h
es o
n
th
e
NI
ST
-
f
in
g
er
p
r
in
t d
atab
ase
S
c
o
r
e
-
l
e
v
e
l
f
u
s
i
o
n
met
h
o
d
s
G
A
R
(
%)
M
a
x
r
u
l
e
[
5
7
]
9
0
.
3
0
M
i
n
r
u
l
e
[
5
7
]
7
9
.
6
0
S
V
M
[
5
8
]
9
1
.
4
0
Li
k
e
l
i
h
o
o
d
r
a
t
i
o
[
5
8
]
9
1
.
4
0
En
t
r
o
p
y
-
w
i
t
h
-
F
r
a
n
k
=
0
.
01
[
5
9
]
8
7
.
7
7
En
t
r
o
p
y
-
w
i
t
h
-
H
a
mac
h
e
r
=
0
.
01
[
5
9
]
8
5
.
7
7
S
-
su
m
u
si
n
g
ma
x
r
u
l
e
[
4
4
]
9
0
.
7
5
S
-
su
m
u
si
n
g
p
r
o
b
a
b
i
l
i
s
t
i
c
-
n
o
r
m
[
4
4
]
8
9
.
0
0
S
-
su
m
u
si
n
g
H
a
m
a
c
h
e
r
-
n
o
r
m
[
4
4
]
7
5
.
5
0
S
-
su
m
u
si
n
g
Y
a
g
e
r
t
-
n
o
r
m w
i
t
h
=
10
.
3
[
4
4
]
9
0
.
0
0
S
-
su
m
u
si
n
g
S
c
h
w
e
i
z
e
r
a
n
d
S
k
l
a
r
t
-
n
o
r
m w
i
t
h
=
0
.
9
[
4
4
]
8
9
.
0
0
W
e
i
g
h
t
e
d
q
u
a
d
r
a
t
i
c
a
r
i
t
h
m
e
t
i
c
mea
n
(
W
Q
A
M
)
u
s
i
n
g
c
o
s(
s)
w
i
t
h
=
11
[
5
8
]
9
1
.
6
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
a
r
i
t
h
me
t
i
c
m
e
a
n
9
1
.
2
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
h
a
r
m
o
n
i
c
me
a
n
8
5
.
5
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
q
u
a
d
r
a
t
i
c
mea
n
9
1
.
5
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
m
e
a
n
r
u
l
e
(
=
5
)
9
0
.
9
0
T
h
e
r
esu
lts
d
em
o
n
s
tr
ate
th
at
ad
ju
s
tin
g
th
e
p
ar
am
eter
(
)
allo
ws
th
e
g
en
er
alize
d
m
ea
n
o
p
er
ato
r
to
p
r
o
g
r
ess
iv
ely
en
h
a
n
ce
r
ec
o
g
n
itio
n
p
e
r
f
o
r
m
an
ce
.
T
h
e
b
est
r
e
s
u
lt
is
o
b
tain
ed
u
s
in
g
th
e
q
u
a
d
r
atic
m
ea
n
,
wh
ic
h
r
ea
ch
es
a
GAR
o
f
9
1
.
5
0
%
at
a
FAR
o
f
0
.
0
1
%,
o
u
t
p
er
f
o
r
m
in
g
th
e
in
d
iv
id
u
al
f
in
g
e
r
p
r
in
t
in
s
tan
ce
s
an
d
th
e
o
th
e
r
test
ed
o
p
er
ato
r
s
.
As p
r
esen
ted
in
T
ab
le
3
,
th
is
p
er
f
o
r
m
an
ce
a
ls
o
ex
ce
ed
s
th
e
r
esu
lts
r
ep
o
r
ted
in
ea
r
lier
s
tu
d
ies,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
6
4
8
-
6
6
2
656
co
n
f
ir
m
in
g
th
e
e
f
f
ec
tiv
en
ess
o
f
th
e
p
r
o
p
o
s
ed
f
u
s
io
n
s
tr
ateg
y
in
m
u
lti
-
in
s
tan
ce
f
in
g
e
r
p
r
in
t
r
ec
o
g
n
itio
n
.
T
h
ese
o
b
s
er
v
atio
n
s
alig
n
with
th
e
f
in
d
in
g
s
o
f
Ab
d
e
r
r
ah
m
a
n
e
et
a
l.
[
6
0
]
,
s
u
p
p
o
r
tin
g
th
e
claim
th
at
em
p
h
asizin
g
h
ig
h
-
co
n
f
id
en
ce
s
co
r
es
lead
s
to
m
o
r
e
s
tab
le
an
d
ac
cu
r
ate
d
ec
is
io
n
s
.
Ad
d
itio
n
ally
,
th
e
m
eth
o
d
is
co
m
p
u
tatio
n
ally
lig
h
tweig
h
t,
av
o
id
in
g
co
m
p
lex
n
o
n
-
lin
ea
r
f
u
n
ctio
n
s
an
d
r
eq
u
i
r
in
g
n
o
tr
ai
n
in
g
p
h
ase,
wh
ich
m
ak
es
it
ap
p
r
o
p
r
iate
f
o
r
u
s
e
in
r
ea
l
-
tim
e
b
io
m
etr
ic
ap
p
licatio
n
s
.
Ov
er
all,
th
ese
r
e
s
u
lts
s
h
o
w
th
at
th
e
p
r
o
p
o
s
ed
m
ea
n
o
p
er
ato
r
-
b
ased
f
u
s
io
n
p
r
o
v
id
es a
p
r
ac
tical
an
d
ef
f
icien
t a
p
p
r
o
ac
h
t
o
im
p
r
o
v
i
n
g
r
ec
o
g
n
itio
n
r
eliab
ilit
y
an
d
s
y
s
tem
s
ec
u
r
ity
.
4
.
2
.
E
v
a
lua
t
i
o
n o
f
m
ea
n
-
ba
s
ed
f
us
io
n m
et
ho
ds
o
n t
he
NI
ST
f
a
ce
da
t
a
ba
s
e
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
p
e
r
f
o
r
m
an
ce
ev
al
u
atio
n
o
f
th
e
p
r
o
p
o
s
ed
s
co
r
e
-
lev
el
f
u
s
io
n
m
et
h
o
d
u
s
in
g
th
e
NI
ST
f
ac
e
d
atab
ase
in
a
m
u
lti
-
alg
o
r
ith
m
b
io
m
etr
ic
s
etu
p
.
Fig
u
r
e
3
s
h
o
ws
th
e
R
O
C
cu
r
v
es
o
b
tain
ed
f
r
o
m
two
in
d
ep
en
d
en
t
f
ac
e
m
atch
er
s
,
r
e
f
er
r
ed
t
o
as
m
atch
er
C
an
d
m
atch
er
G.
At
a
s
tr
ict
s
ec
u
r
ity
t
h
r
esh
o
ld
o
f
FAR
=
0
.
0
1
%,
th
e
m
atch
er
s
ac
h
iev
e
GARs
o
f
6
3
.
0
0
%
an
d
7
2
.
1
0
%,
r
esp
ec
tiv
ely
.
T
h
is
n
o
ticea
b
le
p
er
f
o
r
m
an
ce
g
ap
in
d
icate
s
th
at
ea
ch
m
atch
er
c
ap
tu
r
es
d
if
f
e
r
en
t
d
is
cr
im
in
ativ
e
f
ac
ial
in
f
o
r
m
atio
n
,
m
ak
in
g
th
e
f
u
s
io
n
o
f
th
ei
r
s
co
r
es h
ig
h
ly
b
e
n
ef
icial.
Fig
u
r
e
3
.
R
OC
cu
r
v
es o
f
s
in
g
le
m
atch
er
s
an
d
m
ea
n
-
b
ased
s
co
r
e
-
lev
el
f
u
s
io
n
tech
n
i
q
u
es
T
o
ass
ess
th
e
ad
ap
ta
b
ilit
y
o
f
th
e
p
r
o
p
o
s
ed
f
u
s
io
n
s
tr
ateg
y
,
s
ix
s
co
r
e
c
o
m
b
in
atio
n
s
ett
in
g
s
wer
e
ev
alu
ated
b
y
v
ar
y
in
g
t
h
e
p
o
w
er
p
ar
am
eter
(
)
.
As
r
ep
o
r
ted
in
T
ab
le
4
,
in
cr
ea
s
in
g
co
n
s
is
ten
tly
en
h
a
n
ce
s
r
ec
o
g
n
itio
n
ac
cu
r
ac
y
.
T
h
is
co
n
f
ir
m
s
th
at
th
e
g
e
n
er
alize
d
p
o
wer
m
ea
n
e
f
f
ec
tiv
ely
b
alan
ce
s
s
co
r
e
co
n
tr
ib
u
tio
n
ac
co
r
d
in
g
to
th
e
r
eliab
ilit
y
o
f
ea
ch
m
atch
er
.
T
ab
le
4
.
C
o
m
p
a
r
is
o
n
o
f
v
ar
io
u
s
f
u
s
io
n
m
eth
o
d
s
o
n
t
h
e
NI
S
T
f
ac
e
d
atab
ase
S
c
o
r
e
-
l
e
v
e
l
f
u
s
i
o
n
met
h
o
d
s
G
A
R
(
%)
M
a
x
r
u
l
e
[
5
7
]
7
1
.
5
0
M
i
n
r
u
l
e
[
5
7
]
7
3
.
1
5
H
a
mac
h
e
r
t
-
n
o
r
m
[
5
7
]
7
5
.
9
0
S
V
M
[
5
8
]
7
7
.
3
0
Li
k
e
l
i
h
o
o
d
r
a
t
i
o
[
5
8
]
7
7
.
2
0
S
-
su
m
u
si
n
g
ma
x
r
u
l
e
[
4
4
]
7
6
.
0
0
S
-
su
m
u
si
n
g
H
a
m
a
c
h
e
r
t
-
n
o
r
m
[
4
4
]
7
5
.
7
5
A
sy
m
-
A
O
2
u
si
n
g
A
c
z
e
l
-
A
l
l
s
i
n
a
w
i
t
h
=
0
.
7
a
n
d
=
3
[
6
0
]
7
6
.
2
0
A
sy
m
-
A
O
2
u
si
n
g
A
l
g
e
b
r
a
i
c
p
r
o
d
u
c
t
w
i
t
h
=
1
.
2
[
6
0
]
7
6
.
3
3
A
sy
m
-
A
O
2
u
si
n
g
H
a
m
a
c
h
e
r
w
i
t
h
=
10
[
6
0
]
7
6
.
1
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
a
r
i
t
h
me
t
i
c
m
e
a
n
7
8
.
7
5
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
q
u
a
d
r
a
t
i
c
mea
n
8
0
.
6
3
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
h
a
r
m
o
n
i
c
me
a
n
7
5
.
1
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
g
e
o
me
t
r
i
c
me
a
n
7
7
.
5
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
m
e
a
n
r
u
l
e
(
=
3
)
8
1
.
7
0
P
r
o
p
o
se
d
s
y
st
e
m w
i
t
h
m
e
a
n
r
u
l
e
(
=
5
)
8
2
.
4
0
A
m
o
n
g
a
l
l
t
es
t
e
d
c
o
n
f
i
g
u
r
a
ti
o
n
s
,
(
=
5
)
a
c
h
i
e
v
e
d
t
h
e
h
i
g
h
e
s
t
p
e
r
f
o
r
m
a
n
c
e
,
r
e
a
c
h
i
n
g
a
GA
R
o
f
8
2
.
4
0
%
a
t
F
AR
=
0
.
0
1
%
.
T
h
i
s
i
m
p
r
o
v
e
m
e
n
t
d
e
m
o
n
s
t
r
a
te
s
t
h
a
t
as
s
i
g
n
i
n
g
s
t
r
o
n
g
e
r
i
n
f
l
u
e
n
c
e
t
o
h
i
g
h
-
c
o
n
f
i
d
e
n
c
e
m
a
t
c
h
es
l
e
a
d
s
t
o
m
o
r
e
d
i
s
c
r
i
m
i
n
a
ti
v
e
a
n
d
r
e
l
i
a
b
l
e
f
u
s
i
o
n
o
u
t
c
o
m
es
.
C
o
m
p
a
r
e
d
t
o
t
r
a
d
i
t
i
o
n
a
l
r
u
les
s
u
c
h
a
s
m
a
x
[
5
7
]
,
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
S
co
r
e
-
leve
l b
io
metric in
fo
r
ma
tio
n
fu
s
io
n
w
ith
g
en
era
liz
ed
p
o
w
er me
a
n
(
Leila
Hella
l
)
657
as
well
as
m
ac
h
in
e
lear
n
in
g
-
b
ased
m
eth
o
d
s
lik
e
SVM
o
r
li
k
elih
o
o
d
r
atio
[
4
8
]
,
[
5
8
]
.
T
h
e
p
r
o
p
o
s
ed
o
p
e
r
ato
r
o
f
f
er
s
a
b
etter
tr
ad
e
-
o
f
f
b
etwe
en
p
er
f
o
r
m
an
ce
,
s
im
p
licity
,
an
d
co
m
p
u
tatio
n
al
ef
f
icie
n
cy
.
T
h
er
ef
o
r
e
,
th
e
p
o
we
r
m
ea
n
p
r
o
v
id
es a
p
r
ac
tical
an
d
ef
f
ec
tiv
e
s
o
lu
tio
n
f
o
r
e
n
h
an
cin
g
s
co
r
e
-
lev
el
f
u
s
io
n
in
f
ac
e
b
i
o
m
etr
ics.
T
h
e
co
m
p
ar
is
o
n
b
etwe
en
th
e
p
r
o
p
o
s
ed
s
co
r
e
-
le
v
el
f
u
s
io
n
s
t
r
ateg
y
a
n
d
in
d
iv
id
u
al
b
i
o
m
etr
i
c
m
atch
er
s
s
h
o
ws
a
clea
r
im
p
r
o
v
em
en
t
i
n
r
ec
o
g
n
itio
n
p
er
f
o
r
m
an
ce
.
Acr
o
s
s
all
test
ed
v
alu
es
o
f
(
)
,
th
e
g
en
er
alize
d
m
ea
n
co
n
s
is
ten
tly
s
u
r
p
ass
es
b
o
th
u
n
im
o
d
al
s
y
s
tem
s
an
d
co
n
v
en
tio
n
al
s
co
r
e
f
u
s
io
n
tec
h
n
iq
u
es.
T
h
is
ad
v
an
tag
e
ar
is
es
f
r
o
m
its
s
im
p
licity
,
r
o
b
u
s
tn
ess
,
an
d
lo
w
co
m
p
u
tatio
n
al
co
s
t.
Un
lik
e
weig
h
ted
ap
p
r
o
ac
h
es,
d
e
cisi
o
n
-
lev
el
f
u
s
io
n
,
o
r
m
ac
h
in
e
lea
r
n
in
g
-
b
ased
m
e
th
o
d
s
[
4
8
]
,
[
6
1
]
,
[
6
2
]
,
th
e
p
r
o
p
o
s
ed
o
p
er
ato
r
d
o
es
n
o
t
r
eq
u
ir
e
tr
ain
in
g
,
p
a
r
am
eter
o
p
tim
izatio
n
,
o
r
c
o
m
p
lex
s
co
r
e
tr
an
s
f
o
r
m
atio
n
.
I
ts
b
u
ilt
-
in
r
esis
tan
ce
to
o
u
tlier
s
r
ed
u
ce
s
th
e
in
f
lu
en
ce
o
f
n
o
is
y
b
io
m
etr
ic
s
am
p
les,
wh
ile
s
till
r
etain
in
g
th
e
m
o
s
t
in
f
o
r
m
ativ
e
ch
ar
ac
ter
is
tics
o
f
ea
c
h
m
atch
er
.
As
a
r
esu
lt,
t
h
e
p
r
o
p
o
s
ed
m
eth
o
d
is
well
ad
ap
t
ed
to
p
r
ac
tical
s
ce
n
ar
io
s
,
esp
ec
ially
wh
er
e
co
m
p
u
tatio
n
al
ca
p
ab
ilit
ies
ar
e
lim
ited
o
r
b
io
m
et
r
ic
d
ata
q
u
ality
v
ar
ie
s
.
T
o
an
aly
ze
th
e
b
eh
a
v
io
r
o
f
th
e
m
atch
s
co
r
es
o
n
th
e
NI
ST
F
ac
e
d
atab
ase,
d
escr
ip
tiv
e
s
tati
s
tics
wer
e
co
m
p
u
ted
f
o
r
b
o
th
g
en
u
in
e
an
d
im
p
o
s
to
r
d
is
tr
ib
u
tio
n
s
b
ef
o
r
e
f
u
s
io
n
an
d
af
ter
ap
p
ly
in
g
th
e
m
ea
n
-
r
u
le
s
tr
ateg
ies
(
ar
ith
m
etic,
g
e
o
m
etr
ic,
q
u
a
d
r
a
tic,
an
d
h
ar
m
o
n
i
c
m
ea
n
s
)
.
T
h
e
r
esu
lts
s
h
o
w
th
at
th
e
m
ea
n
-
b
ased
f
u
s
io
n
im
p
r
o
v
es
class
s
ep
ar
ab
ilit
y
b
y
s
lig
h
tly
r
aisi
n
g
th
e
av
er
a
g
e
g
e
n
u
in
e
s
co
r
es
an
d
lo
wer
in
g
th
e
av
e
r
ag
e
im
p
o
s
to
r
s
co
r
es.
A
s
u
m
m
ar
y
o
f
th
e
d
etailed
s
tatis
tical
r
esu
lts
f
o
r
ea
ch
f
u
s
io
n
r
u
l
e
is
p
r
o
v
id
e
d
in
T
ab
le
5
.
T
ab
le
5
.
Gen
u
in
e
v
s
im
p
o
s
to
r
s
co
r
e
s
tatis
tic
s
f
o
r
d
if
f
er
e
n
t
m
ean
-
r
u
le
f
u
s
io
n
m
eth
o
d
s
M
e
a
n
r
u
l
e
G
e
n
u
i
n
e
mea
n
(
)
G
e
n
u
i
n
e
s
t
d
(
)
I
mp
o
st
o
r
me
a
n
(
)
I
mp
o
st
o
r
s
t
d
(
)
A
r
i
t
h
m
e
t
i
c
mea
n
0
.
7
9
0
9
0
.
0
5
6
7
0
.
6
2
5
2
0
.
0
784
G
e
o
me
t
r
i
c
m
e
a
n
0
.
7
8
2
7
0
.
0
6
1
3
0
.
5
9
0
7
0
.
0
9
5
4
H
a
r
mo
n
i
c
M
e
a
n
0
.
7
6
6
0
0
.
0
8
0
4
0
.
5
1
9
1
0
.
1
0
7
2
Q
u
a
d
r
a
t
i
c
mea
n
0
.
7
8
0
7
0
.
0
5
6
9
0
.
6
1
5
3
0
.
0
7
6
8
Ov
er
all,
d
esp
ite
s
o
m
e
d
if
f
er
en
ce
s
in
s
co
r
e
d
is
p
er
s
io
n
,
all
m
ea
n
o
p
er
ato
r
s
co
n
tr
ib
u
te
to
b
etter
class
s
ep
ar
ab
ilit
y
co
m
p
ar
ed
to
th
e
u
n
f
u
s
ed
b
aselin
e,
wh
e
r
e
g
e
n
u
i
n
e
an
d
im
p
o
s
to
r
av
e
r
ag
e
s
co
r
es
wer
e
0
.
6
8
3
6
an
d
0
.
3
9
4
0
,
r
esp
ec
tiv
ely
.
T
o
s
tatis
tically
co
m
p
ar
e
th
e
im
p
r
o
v
em
en
ts
,
p
air
ed
t
-
test
s
wer
e
ap
p
lie
d
alo
n
g
with
ef
f
ec
t
s
ize
an
d
9
5
%
co
n
f
i
d
en
ce
in
ter
v
als
[
5
5
]
,
[
5
6
]
.
T
h
e
an
al
y
s
is
r
ev
ea
led
a
s
tatis
tically
s
ig
n
if
ican
t
d
if
f
er
en
ce
b
etwe
en
th
e
o
p
er
ato
r
s
.
Fo
r
th
e
g
en
u
i
n
e
s
co
r
es,
th
e
test
y
ield
ed
(
2
9
9
9
)
=
1
1
1
.
5
7
,
<
0
.
0
0
0
0
1
,
wh
ile
f
o
r
im
p
o
s
to
r
s
co
r
es
it
g
av
e
(
8
,
9
9
6
,
9
9
9
)
=
1
1
5
7
3
.
1
4
,
<
0
.
0
0
0
0
1
.
T
h
e
co
n
f
id
en
ce
i
n
ter
v
als
also
in
d
icate
s
tab
le
im
p
o
s
to
r
s
.
B
ec
au
s
e
th
e
d
ataset
is
ex
ten
s
iv
e,
ex
tr
em
ely
s
m
all
p
-
v
alu
es
ar
e
ex
p
ec
ted
ev
en
with
s
m
all
n
u
m
er
ical
d
if
f
er
en
ce
s
.
T
h
er
ef
o
r
e,
ad
d
itio
n
al
p
er
f
o
r
m
an
ce
in
d
icato
r
s
,
s
u
ch
as
th
e
R
OC
cu
r
v
e,
wer
e
in
clu
d
ed
t
o
b
etter
r
ef
lect
th
e
p
r
ac
tical
im
p
r
o
v
em
e
n
t
ac
h
ie
v
ed
b
y
f
u
s
io
n
.
Ov
er
all,
th
ese
f
i
n
d
in
g
s
d
em
o
n
s
tr
ate
th
at
m
ea
n
-
b
ased
s
co
r
e
f
u
s
io
n
en
h
an
ce
s
d
is
cr
im
in
atio
n
ca
p
a
b
ilit
y
an
d
im
p
r
o
v
es
im
p
o
s
to
r
r
ejec
tio
n
,
lead
in
g
t
o
a
m
o
r
e
r
eliab
le
b
io
m
etr
ic
v
er
if
icatio
n
s
y
s
tem
u
n
d
er
r
ea
l
-
wo
r
ld
co
n
d
itio
n
s
.
4
.
3
.
P
er
f
o
r
m
a
nce
o
f
m
ea
n
o
pera
t
o
r
-
b
a
s
ed
f
us
io
n o
n XM
2
VT
S
d
a
t
a
ba
s
e
T
h
e
XM
2
VT
S
ex
p
er
im
en
ts
wer
e
p
er
f
o
r
m
ed
f
o
llo
win
g
th
e
s
tan
d
ar
d
L
au
s
an
n
e
p
r
o
t
o
co
ls
L
P1
an
d
L
P2
,
wh
er
e
th
e
s
y
s
tem
is
d
iv
id
ed
in
to
eig
h
t
b
aselin
e
c
o
n
f
i
g
u
r
a
t
i
o
n
s
i
n
L
P
1
a
n
d
f
i
v
e
b
a
s
el
i
n
e
c
o
n
f
i
g
u
r
a
t
i
o
n
s
i
n
L
P
2
.
E
a
c
h
b
a
s
el
i
n
e
c
o
n
s
is
ts
o
f
a
b
in
a
r
y
r
e
c
o
g
n
i
t
i
o
n
c
o
m
p
o
n
e
n
t
f
o
r
m
e
d
b
y
c
o
m
b
i
n
i
n
g
a
s
p
e
c
i
f
ic
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
m
e
t
h
o
d
w
i
t
h
a
n
a
p
p
r
o
p
r
i
a
t
e
cl
as
s
i
f
i
e
r
.
T
o
ass
e
s
s
t
h
e
a
d
a
p
t
a
b
il
i
ty
o
f
t
h
e
p
r
o
p
o
s
e
d
f
u
s
i
o
n
m
e
t
h
o
d
o
l
o
g
y
,
t
h
r
e
e
s
c
o
r
e
-
l
e
v
e
l
f
u
s
i
o
n
s
t
r
a
t
e
g
i
es
w
e
r
e
a
p
p
l
i
e
d
a
c
r
o
s
s
d
i
f
f
e
r
e
n
t
b
a
s
e
l
i
n
e
c
o
m
b
i
n
a
t
i
o
n
s
,
as
d
es
c
r
i
b
e
d
in
[
4
5
]
.
T
ab
les
6
an
d
7
p
r
esen
t
th
e
r
esu
lts
o
f
3
2
f
u
s
io
n
co
n
f
ig
u
r
atio
n
s
f
o
r
L
P1
a
n
d
L
P2
,
r
esp
ec
tiv
ely
.
Su
b
s
ec
tio
n
(
a
)
r
ep
o
r
ts
m
u
ltimo
d
al
f
u
s
io
n
ac
r
o
s
s
f
ac
e
a
n
d
s
p
ee
c
h
m
o
d
alities
,
wh
ile
s
u
b
s
ec
tio
n
s
(
b
)
a
n
d
(
c)
f
o
cu
s
o
n
class
if
ier
-
b
ased
f
u
s
io
n
an
d
f
u
s
io
n
ac
r
o
s
s
h
eter
o
g
en
e
o
u
s
f
ea
tu
r
e
s
ets.
T
h
is
ex
p
er
im
e
n
tal
s
etu
p
allo
ws
a
co
m
p
r
eh
en
s
iv
e
ev
alu
atio
n
o
f
h
o
w
m
ea
n
-
b
ased
s
co
r
e
f
u
s
io
n
e
n
h
a
n
ce
s
r
ec
o
g
n
itio
n
p
er
f
o
r
m
a
n
ce
u
n
d
er
d
if
f
er
e
n
t b
io
m
etr
ic
co
n
d
itio
n
s
.
I
n
th
is
ex
p
er
im
en
t,
th
r
ee
m
ea
n
o
p
e
r
ato
r
s
wer
e
ev
alu
ate
d
f
o
r
s
co
r
e
-
le
v
el
f
u
s
io
n
:
th
e
h
ar
m
o
n
ic,
g
eo
m
etr
ic,
a
n
d
q
u
ad
r
atic
m
ea
n
s
.
T
h
e
f
u
s
io
n
p
er
f
o
r
m
an
ce
w
as
q
u
an
tifie
d
u
s
in
g
th
e
HT
E
R
,
an
d
t
h
e
r
esu
lts
wer
e
co
m
p
ar
ed
with
th
o
s
e
r
ep
o
r
ted
in
[
4
8
]
to
ass
ess
th
e
b
en
ef
it
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
.
T
ab
les
5
an
d
6
p
r
esen
t
th
e
HT
E
R
v
alu
es o
b
tain
ed
f
o
r
L
P
1
an
d
L
P2
,
r
esp
ec
tiv
ely
.
T
o
an
aly
ze
th
e
in
f
l
u
en
ce
o
f
th
e
p
o
wer
p
ar
am
eter
,
ad
d
itio
n
al
ex
p
er
im
en
ts
wer
e
c
o
n
d
u
cted
u
s
in
g
th
r
ee
v
alu
es
o
f
(
−1
,
0
,
an
d
2
)
,
r
ep
r
esen
tin
g
d
if
f
er
e
n
t
f
u
s
io
n
b
e
h
a
v
io
r
s
r
an
g
in
g
f
r
o
m
co
n
s
er
v
ativ
e
to
co
n
f
i
d
en
ce
-
f
o
cu
s
ed
weig
h
tin
g
.
T
h
ese
r
esu
lts
h
ig
h
lig
h
t
h
o
w
ad
ju
s
tin
g
(
)
d
ir
ec
tly
af
f
ec
ts
m
u
ltimo
d
al
s
co
r
e
ag
g
r
eg
atio
n
a
n
d
h
elp
s
id
en
tify
th
e
m
o
s
t
ef
f
ec
ti
v
e
co
n
f
ig
u
r
atio
n
.
T
o
g
eth
e
r
,
th
ese
an
aly
s
es
p
r
o
v
id
e
a
c
o
m
p
r
eh
en
s
iv
e
ev
alu
atio
n
o
f
th
e
p
r
o
p
o
s
ed
f
u
s
io
n
s
tr
ateg
y
o
n
th
e
XM
2
VT
S d
atab
ase.
I
n
o
u
r
ex
p
e
r
im
en
ts
,
th
e
p
r
o
p
o
s
ed
m
ea
n
-
b
ased
f
u
s
io
n
f
u
r
th
e
r
r
ed
u
ce
d
t
h
e
er
r
o
r
r
ates
f
o
r
th
ese
s
tr
o
n
g
b
aselin
e
m
atch
er
s
.
T
h
e
b
est
i
m
p
r
o
v
e
m
en
t
was
ac
h
iev
ed
u
s
in
g
th
e
q
u
ad
r
atic
m
ea
n
(
=
2
)
,
wh
e
r
e
th
e
HT
E
R
Evaluation Warning : The document was created with Spire.PDF for Python.