T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
18
,
No
.
4
,
A
u
g
u
s
t
2020
,
p
p
.
1
8
9
2
~
189
6
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Ke
m
e
n
r
is
te
k
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
KA
.
v
1
8
i4
.
1
1
1
7
6
1892
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
A
princi
pa
l co
m
p
o
nent
a
na
ly
sis
-
ba
sed featu
re di
m
e
n
sio
na
lity
reduction s
che
m
e
f
o
r cont
ent
-
ba
sed i
m
a
g
e
retr
i
ev
a
l sy
ste
m
O
luw
o
le
A.
Adeg
bo
la
1
,
I
s
m
a
il A.
Adey
e
m
o
2
,
F
o
la
s
a
de
A.
Se
m
ire
3
,
Seg
un
I
.
P
o
po
o
la
4
,
Adere
m
i A
.
At
a
y
er
o
5
1,
2,
3
De
p
a
rtm
e
n
t
o
f
El
e
c
tro
n
ic an
d
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
L
a
d
o
k
e
Ak
in
to
la Un
iv
e
rsity
o
f
T
e
c
h
n
o
lo
g
y
,
Nig
e
ria
4,
5
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
In
f
o
r
m
a
ti
o
n
En
g
in
e
e
rin
g
,
C
o
v
e
n
a
n
t
Un
iv
e
rsity
,
Nig
e
ria
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
1
3
,
2
0
1
8
R
ev
i
s
ed
Ma
r
2
3
,
2
0
2
0
A
cc
ep
ted
A
p
r
3
,
2
0
2
0
I
n
c
o
n
t
e
n
t
-
b
a
s
e
d
im
a
g
e
re
t
r
i
e
v
a
l
(
C
B
I
R
)
sy
s
t
e
m
,
o
n
e
a
p
p
r
o
a
c
h
o
f
im
a
g
e
r
e
p
r
e
se
n
t
a
t
i
o
n
i
s
t
o
e
m
p
l
o
y
c
o
m
b
i
n
a
t
i
o
n
o
f
l
o
w
-
l
e
v
e
l
v
i
s
u
a
l
f
e
a
t
u
r
e
s
c
a
s
c
a
d
e
d
t
o
g
e
t
h
e
r
i
n
t
o
a
f
l
a
t
v
e
c
t
o
r
.
W
h
i
le
t
h
i
s
p
r
e
se
n
t
s
m
o
re
d
e
s
c
r
i
p
t
iv
e
i
n
f
o
rm
a
t
i
o
n
,
i
t
h
o
w
e
v
e
r
p
o
s
e
s
se
r
i
o
u
s
c
h
a
l
l
e
n
g
e
s
i
n
t
e
rm
s
o
f
h
i
g
h
d
i
m
e
n
s
i
o
n
a
l
i
ty
a
n
d
h
i
g
h
c
o
m
p
u
t
a
t
i
o
n
a
l
c
o
s
t
o
f
f
e
a
t
u
r
e
e
x
t
ra
c
t
i
o
n
a
l
g
o
r
i
t
h
m
s
t
o
d
e
p
l
o
y
m
e
n
t
o
f
CB
I
R
o
n
p
l
a
t
f
o
rm
s
(
d
e
v
i
c
e
s
)
w
i
t
h
l
im
i
te
d
c
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
s
t
o
r
a
g
e
re
s
o
u
rc
e
s
.
He
n
c
e
,
i
n
t
h
i
s
w
o
rk
a
f
e
a
t
u
r
e
d
i
m
e
n
s
i
o
n
a
l
i
ty
re
d
u
c
t
i
o
n
t
e
c
h
n
i
q
u
e
b
a
s
e
d
o
n
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
a
ly
s
i
s
(
P
C
A
)
i
s
im
p
l
e
m
e
n
t
e
d
.
E
a
c
h
im
a
g
e
i
n
a
d
a
t
a
b
a
s
e
i
s
i
n
d
e
x
e
d
u
s
i
n
g
1
7
4
-
d
i
m
e
n
s
i
o
n
a
l
f
e
a
t
u
r
e
v
e
c
t
o
r
c
o
m
p
r
is
i
n
g
o
f
5
4
-
d
i
m
e
n
s
i
o
n
a
l
c
o
l
o
u
r
m
o
m
e
n
t
s
(
C
M
5
4
)
,
3
2
-
b
i
n
H
S
V
-
h
i
s
t
o
g
r
a
m
(
H
I
S
T
3
2
)
,
4
8
-
d
im
e
n
s
i
o
n
a
l
g
a
b
o
r
w
a
v
e
l
e
t
(
G
W
4
8
)
a
n
d
4
0
-
d
i
m
e
n
s
i
o
n
a
l
w
a
v
e
l
e
t
m
o
m
e
n
t
s
(
M
W
4
0
)
.
T
h
e
P
CA
s
c
h
e
m
e
w
a
s
i
n
c
o
r
p
o
r
a
t
e
d
i
n
t
o
a
C
B
I
R
sy
s
t
e
m
t
h
a
t
u
t
i
l
i
z
e
d
t
h
e
e
n
t
i
re
f
e
a
t
u
r
e
v
e
c
t
o
r
s
p
a
c
e
.
T
h
e
k
-
l
a
r
g
e
s
t
e
i
g
e
n
v
a
l
u
e
s
t
h
a
t
y
i
e
l
d
e
d
a
n
o
t
m
o
r
e
t
h
a
n
5
%
d
e
g
r
a
d
a
t
i
o
n
i
n
m
e
a
n
p
re
c
i
s
i
o
n
w
e
r
e
r
e
ta
i
n
e
d
f
o
r
d
im
e
n
s
i
o
n
a
l
i
ty
r
e
d
u
c
t
i
o
n
.
T
h
r
e
e
im
a
g
e
d
a
ta
b
a
se
s
(
D
B
1
0
,
D
B
2
0
a
n
d
D
B
1
0
0
)
w
e
r
e
u
s
e
d
f
o
r
t
e
s
t
i
n
g
.
T
h
e
re
s
u
l
t
o
b
t
a
i
n
e
d
s
h
o
w
e
d
t
h
a
t
w
i
t
h
8
0
%
re
d
u
c
t
i
o
n
i
n
f
e
a
t
u
r
e
d
im
e
n
s
i
o
n
s
,
t
o
l
e
r
a
b
l
e
l
o
s
s
o
f
3
.
4
5
,
4
.
3
9
a
n
d
7
.
4
0
%
i
n
m
e
a
n
p
re
c
i
s
i
o
n
v
a
l
u
e
w
e
re
a
c
h
i
e
v
e
d
o
n
D
B
1
0
,
D
B
2
0
a
n
d
D
B
1
0
0
.
K
ey
w
o
r
d
s
:
C
o
n
te
n
t
-
b
ased
i
m
a
g
e
r
etr
iev
al
s
y
s
te
m
F
e
a
tu
r
e
d
im
en
s
i
o
n
a
li
ty
r
e
d
u
c
t
io
n
L
o
w
-
lev
el
v
i
s
u
al
f
ea
t
u
r
e
P
r
in
cip
al
co
m
p
o
n
e
n
t a
n
al
y
s
is
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
:
Seg
u
n
I
.
P
o
p
o
o
la
,
Dep
ar
t
m
en
t o
f
E
lectr
ical
an
d
I
n
f
o
r
m
atio
n
E
n
g
in
ee
r
i
n
g
,
C
o
v
e
n
an
t U
n
i
v
er
s
it
y
,
P
.
M.
B
.
1
0
2
3
,
Ota,
Nig
er
ia
.
E
m
ail:
s
e
g
u
n
.
p
o
p
o
o
la@
co
v
e
n
an
tu
n
i
v
er
s
it
y
.
ed
u
.
n
g
1.
I
NT
RO
D
UCT
I
O
N
O
n
e
o
f
t
h
e
c
h
al
l
en
g
es
o
f
r
e
l
e
v
a
n
c
e
f
ee
d
b
a
c
k
(
R
F
)
in
im
ag
e
r
e
t
r
i
ev
a
l
is
t
h
e
i
n
h
e
r
e
n
t
‘
cu
r
s
e
o
f
d
im
en
s
i
o
n
al
i
ty
’
o
c
c
as
i
o
n
e
d
b
y
s
m
al
l
s
am
p
l
e
s
i
z
e
w
it
h
h
ig
h
f
e
a
tu
r
e
d
im
en
s
i
o
n
.
T
h
e
r
ef
o
r
e
,
f
o
r
R
F
t
e
ch
n
i
q
u
es
w
h
ic
h
a
r
e
b
a
s
e
d
o
n
t
r
a
in
in
g
c
l
a
s
s
if
i
e
r
u
s
i
n
g
f
e
e
d
b
a
ck
e
x
am
p
l
e
s
,
th
e
c
u
r
s
e
o
f
d
im
en
s
i
o
n
a
li
ty
c
an
d
e
t
e
r
i
o
r
at
e
t
h
e
c
la
s
s
if
ie
r
p
e
r
f
o
r
m
an
c
e
,
th
er
e
b
y
l
e
a
d
in
g
t
o
p
o
o
r
r
e
t
r
i
e
v
al
r
e
s
u
l
ts
.
T
o
m
it
ig
a
t
e
th
i
s
p
r
o
b
l
em
,
a
t
e
ch
n
i
q
u
e
th
a
t
r
e
l
i
e
s
o
n
th
e
p
r
o
p
e
r
t
ie
s
o
f
th
e
f
e
e
d
b
a
ck
ex
am
p
l
es
f
o
r
s
el
e
c
tin
g
a
l
o
w
e
r
d
im
en
s
i
o
n
a
l
f
e
a
tu
r
e
,
t
h
at
w
il
l
s
e
r
v
e
a
s
g
o
o
d
r
e
p
r
e
s
en
t
at
iv
e
f
o
r
c
l
ass
if
ic
a
t
i
o
n
c
an
b
e
em
p
l
o
y
e
d
.
I
n
th
is
w
ay
,
a
s
ig
n
if
i
ca
n
t
d
im
en
s
i
o
n
a
l
i
ty
r
e
d
u
c
ti
o
n
c
a
n
b
e
a
c
h
i
ev
e
d
b
y
r
em
o
v
in
g
i
r
r
e
l
ev
an
t
o
r
r
e
d
u
n
d
an
t
f
e
a
tu
r
es
,
t
h
u
s
l
e
a
d
in
g
t
o
a
s
ig
n
if
i
c
an
t
d
e
c
r
ea
s
e
in
t
r
a
in
in
g
t
im
e
a
n
d
m
em
o
r
y
c
o
m
p
l
ex
i
t
i
es
,
an
d
b
e
t
t
e
r
cl
a
s
s
if
i
e
r
p
e
r
f
o
r
m
a
n
ce
[
1
,
2
]
.
A
p
p
r
o
a
ch
e
s
f
o
r
f
e
a
tu
r
e
d
im
en
s
i
o
n
a
li
ty
r
e
d
u
c
ti
o
n
h
a
v
e
b
e
en
g
r
o
u
p
e
d
in
t
o
tw
o
[
3
]
:
(
a
)
t
h
o
s
e
th
a
t
in
v
o
l
v
es
l
in
e
a
r
o
r
n
o
n
l
in
e
a
r
m
a
p
p
in
g
f
r
o
m
t
h
e
o
r
ig
in
al
f
e
at
u
r
e
s
p
a
ce
t
o
a
n
e
w
o
n
e
o
f
l
o
w
e
r
d
im
en
s
i
o
n
a
l
ity
.
N
o
t
a
b
le
am
o
n
g
th
es
e
a
r
e
l
in
e
a
r
d
i
s
c
r
im
i
n
a
n
t
an
a
ly
s
i
s
[
4
]
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
A
p
r
in
cip
a
l c
o
mp
o
n
en
t
a
n
a
lys
is
-
b
a
s
ed
fea
tu
r
e
d
imen
s
io
n
a
lit
y
r
ed
u
ctio
n
s
ch
eme
…
(
Olu
w
o
le
A
.
A
d
eg
b
o
la
)
1893
a
n
d
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
al
y
s
i
s
[
1
,
5
-
7
]
;
(
b
)
t
h
o
s
e
th
a
t
d
ir
e
c
t
ly
r
e
d
u
c
e
th
e
n
u
m
b
e
r
o
f
th
e
o
r
ig
in
a
l
f
e
a
tu
r
es
b
y
s
el
e
c
tin
g
a
s
u
b
s
e
t
o
f
th
em
t
h
at
s
t
i
ll
r
e
t
a
in
s
s
u
f
f
ic
i
en
t
i
n
f
o
r
m
a
ti
o
n
f
o
r
cl
a
s
s
if
ic
a
t
i
o
n
.
I
n
g
e
n
e
r
al
,
a
p
p
r
o
a
c
h
es
i
n
th
i
s
ca
t
eg
o
r
y
c
an
b
e
g
r
o
u
p
e
d
i
n
t
o
tw
o
n
am
e
ly
:
f
i
l
t
e
r
m
eth
o
d
s
an
d
w
r
a
p
p
e
r
m
e
th
o
d
s
[
8
]
.
T
h
e
f
ilter
m
et
h
o
d
s
ar
e
g
en
er
all
y
n
o
t
cla
s
s
i
f
ier
d
ep
en
d
en
t
as
t
h
e
y
ac
q
u
ir
e
n
o
f
ee
d
b
ac
k
f
r
o
m
th
e
clas
s
i
f
ier
s
,
b
u
t
d
ep
en
d
o
n
in
d
ir
ec
t
as
s
es
s
m
e
n
t
s
li
k
e
d
is
tan
ce
m
ea
s
u
r
e
to
esti
m
ate
cla
s
s
i
f
icatio
n
p
er
f
o
r
m
a
n
ce
o
n
t
h
e
o
th
er
h
a
n
d
,
th
e
w
r
ap
p
er
m
eth
o
d
s
ar
e
cl
ass
i
f
ier
s
d
ep
en
d
en
t
a
n
d
ar
e
k
n
o
w
n
to
y
ield
b
etter
class
i
f
icatio
n
p
er
f
o
r
m
an
ce
[
8
,
9
]
.
Ma
n
y
f
ea
t
u
r
es
s
elec
t
io
n
m
et
h
o
d
s
f
o
r
class
i
f
icatio
n
h
a
v
e
b
ee
n
p
r
o
p
o
s
ed
in
th
e
l
iter
atu
r
e,
[
1
0
]
w
ith
m
a
n
y
e
x
p
er
i
m
e
n
tal
r
es
u
lt
s
in
f
av
o
u
r
o
f
th
e
w
r
ap
p
er
m
et
h
o
d
s
[
8
,
1
1
,
1
2
]
.
Ho
w
e
v
er
,
in
s
p
ite
o
f
g
o
o
d
class
if
icatio
n
p
er
f
o
r
m
an
ce
,
th
e
w
r
ap
p
er
m
et
h
o
d
s
h
av
e
li
m
ited
ap
p
licatio
n
d
u
e
to
h
ig
h
co
m
p
u
tatio
n
al
co
m
p
le
x
it
y
,
e
s
p
ec
iall
y
w
h
e
n
ap
p
lied
to
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
(
SVM)
cla
s
s
i
f
ier
s
.
P
C
A
is
a
d
i
m
e
n
s
io
n
alit
y
r
e
d
u
ctio
n
tec
h
n
iq
u
e
th
at
tr
an
s
f
o
r
m
s
t
h
e
o
r
ig
i
n
al
s
et
o
f
f
ea
tu
r
es
i
n
to
a
s
m
al
ler
s
u
b
s
e
t th
a
t a
cc
o
u
n
t f
o
r
as
m
u
c
h
o
f
t
h
e
to
tal
v
ar
iati
o
n
in
t
h
e
d
ata
as p
o
s
s
ib
le
[
1
3
]
.
I
t is
w
id
el
y
u
s
ed
i
n
th
e
ar
ea
o
f
p
atter
n
r
ec
o
g
n
iti
o
n
,
co
m
p
u
ter
v
is
io
n
a
n
d
s
i
g
n
al
p
r
o
ce
s
s
i
n
g
[
7
]
.
Se
v
er
al
o
p
ti
m
alit
y
p
r
o
p
er
ties
o
f
P
C
A
h
av
e
b
ee
n
id
en
ti
f
ied
n
a
m
e
l
y
:
v
ar
ian
ce
o
f
e
x
tr
ac
ted
f
ea
tu
r
es
i
s
m
a
x
i
m
ized
;
th
e
e
x
tr
ac
ted
f
ea
tu
r
es
ar
e
u
n
co
r
r
elate
d
;
f
i
n
d
s
b
est
li
n
ea
r
ap
p
r
o
x
im
a
tio
n
i
n
th
e
m
ea
n
-
s
q
u
ar
e
s
en
s
e
an
d
m
ax
i
m
ize
s
in
f
o
r
m
at
io
n
co
n
tai
n
ed
i
n
th
e
e
x
tr
ac
ted
f
ea
t
u
r
e
[
1
4
]
.
T
h
e
s
e
p
r
o
p
e
r
t
i
es
o
f
P
C
A
h
av
e
a
t
t
r
a
c
t
e
d
r
es
e
a
r
ch
o
n
PC
A
-
b
as
e
d
v
a
r
i
a
b
l
e
s
e
le
c
t
i
o
n
m
e
th
o
d
s
[
7
,
1
3
-
1
8
]
an
d
h
a
s
b
ee
n
ap
p
lied
to
r
elev
an
ce
f
ee
d
b
ac
k
in
b
o
t
h
d
o
cu
m
en
t
a
n
d
i
m
a
g
e
r
etr
ie
v
al
s
y
s
te
m
s
[
1
,
5
,
6
]
.
I
n
[
1
]
,
a
n
o
v
el
P
C
A
-
b
ased
f
ea
tu
r
e
d
i
m
en
s
io
n
al
it
y
r
ed
u
ctio
n
s
c
h
e
m
e
(
o
r
ap
p
r
o
ac
h
)
w
a
s
p
r
o
p
o
s
ed
f
o
r
th
e
R
F
f
r
a
m
e
w
o
r
k
w
it
h
a
v
ie
w
to
ca
p
tu
r
in
g
t
h
e
s
u
b
j
ec
tiv
e
clas
s
i
m
p
lied
i
n
t
h
e
p
o
s
iti
v
e
ex
a
m
p
les
.
Si
m
i
lar
l
y
,
th
e
w
o
r
k
s
o
f
C
o
x
,
et
a
l,
[
1
9
]
an
d
Va
s
co
n
ce
lo
s
&
L
ip
p
m
a
n
[
2
0
]
,
e
m
p
lo
y
ed
B
a
y
e
s
ian
l
ea
r
n
in
g
to
i
n
te
g
r
ate
u
s
er
’
s
f
ee
d
b
ac
k
f
o
r
u
p
d
atin
g
i
m
ag
e
p
r
o
b
ab
ilit
y
d
is
tr
ib
u
tio
n
an
d
s
u
b
s
eq
u
en
tl
y
r
e
-
r
an
k
i
m
a
g
es i
n
th
e
d
atab
ase.
I
t
w
a
s
r
ep
o
r
ted
th
at
t
h
e
s
c
h
e
m
e
(
o
r
ap
p
r
o
ac
h
)
r
ed
u
ce
d
th
e
av
er
ag
e
r
etr
iev
al
ti
m
e
a
n
d
s
ig
n
if
ican
t
l
y
r
ed
u
ce
d
s
to
r
ag
e
s
p
ac
e
u
tili
za
ti
o
n
.
Ho
w
e
v
er
,
th
e
p
r
ec
is
io
n
m
ea
s
u
r
e
in
to
p
2
0
r
etr
iev
al
r
esu
lts
in
f
o
u
r
f
ee
d
b
ac
k
iter
atio
n
s
w
a
s
45
%
.
T
h
is
m
a
y
b
e
d
u
e
to
th
e
f
ai
lu
r
e
o
f
B
a
y
esia
n
cla
s
s
i
f
ier
s
to
u
s
e
t
h
e
f
e
w
a
v
ailab
le
i
m
a
g
e
s
a
m
p
les
g
ath
er
ed
o
v
er
th
e
f
ee
d
b
ac
k
iter
atio
n
s
to
es
ti
m
ate
t
h
e
cla
s
s
p
r
o
b
ab
ilit
y
d
i
s
tr
ib
u
ti
o
n
.
I
t
w
as
s
tated
b
y
Yin
,
B
h
a
n
u
,
C
h
a
n
g
an
d
Do
n
g
[
2
1
]
th
at
o
n
e
o
f
t
h
e
s
h
o
r
tco
m
in
g
s
o
f
t
h
e
B
a
y
e
s
ian
ap
p
r
o
ac
h
i
s
t
h
at
it
r
eq
u
ir
e
s
m
o
r
e
f
ee
d
b
ac
k
iter
atio
n
s
to
g
a
th
er
m
o
r
e
s
a
m
p
le
s
,
w
h
ic
h
is
n
o
t
al
w
a
y
s
av
a
ilab
le
in
r
ea
l
ti
m
e
r
etr
i
ev
al
s
y
s
te
m
s
,
to
ef
f
ec
ti
v
el
y
esti
m
ate
t
h
e
p
r
o
b
ab
ilit
y
d
is
tr
ib
u
tio
n
o
f
t
h
e
i
m
a
g
e
s
a
m
p
les
.
I
n
o
th
er
to
ad
d
r
ess
t
h
e
co
m
p
u
tatio
n
a
l
co
m
p
le
x
it
y
is
s
u
e,
a
SVM
-
b
ased
tec
h
n
iq
u
e
,
ter
m
ed
f
ilter
ed
an
d
s
u
p
p
o
r
ted
s
eq
u
en
tial
f
o
r
w
ar
d
s
ea
r
c
h
,
w
as
p
r
o
p
o
s
ed
f
ea
t
u
r
e
s
elec
tio
n
[
3
]
.
T
h
e
tech
n
iq
u
e
i
n
te
g
r
ates
th
e
f
ilter
a
n
d
w
r
ap
p
er
p
ar
ts
i
n
to
o
n
e
s
c
h
e
m
e
b
y
le
v
er
ag
i
n
g
o
n
t
h
eir
u
n
iq
u
e
s
tr
e
n
g
th
s
.
R
es
u
lts
o
f
e
x
p
er
i
m
en
ta
l
o
n
b
o
th
s
y
n
t
h
etic
a
n
d
r
ea
l
d
a
ta
s
h
o
w
ed
ef
f
ec
ti
v
e
n
ess
o
f
t
h
e
m
e
th
o
d
r
eg
ar
d
in
g
clas
s
if
ica
tio
n
ac
c
u
r
ac
y
.
Ho
w
e
v
er
,
g
i
v
en
t
h
e
f
ac
t
t
h
at
m
u
c
h
s
m
aller
d
ata
,
co
m
p
ar
ed
to
w
h
at
o
b
tain
s
in
C
B
I
R
s
y
s
te
m
,
w
as
u
s
ed
to
ev
alu
a
te
th
e
s
y
s
te
m
,
an
a
v
er
ag
e
r
u
n
-
ti
m
e
o
f
1
6
.
2
3
s
ec
o
n
d
s
w
as
r
ec
o
r
d
ed
Su
ch
a
len
g
th
y
r
u
n
ti
m
e
is
n
o
t
ac
ce
p
tab
le
f
o
r
C
B
I
R
s
y
s
te
m
with
R
F f
r
a
m
e
w
o
r
k
.
2.
M
AT
E
RIAL
S AN
D
M
E
T
H
O
DS
2
.
1
.
F
e
a
t
ure
ex
t
ra
ct
io
n
Featu
r
e
e
x
tr
ac
tio
n
is
o
n
e
v
er
y
cr
u
cia
l
tas
k
in
C
B
I
R
ap
p
li
ca
tio
n
,
an
d
i
t
is
t
h
e
co
r
e
o
f
an
y
s
u
c
h
s
y
s
te
m
[
4
]
.
T
h
e
ex
tr
ac
tio
n
o
f
s
u
itab
le
f
ea
tu
r
e
s
f
r
o
m
th
e
i
m
a
g
es
i
n
f
lu
e
n
ce
s
to
a
g
r
ea
t
ex
ten
t
th
e
c
h
o
ice
o
f
th
e
in
d
e
x
i
n
g
s
tr
u
c
t
u
r
e
an
d
th
e
q
u
er
y
p
r
o
ce
s
s
in
g
u
n
it.
I
n
v
i
e
w
o
f
t
h
is
,
v
ar
io
u
s
m
eth
o
d
s
o
f
f
ea
tu
r
e
ex
tr
ac
tio
n
to
ex
tr
ac
t
v
ar
io
u
s
t
y
p
es
o
f
v
i
s
u
al
co
n
te
n
ts
f
r
o
m
th
e
i
m
ag
e
s
h
av
e
b
ee
n
d
e
v
elo
p
ed
an
d
a
r
e
b
ein
g
i
m
p
r
o
v
ed
u
p
o
n
o
v
er
ti
m
e
[
2
2
,
2
3
]
.
T
h
r
e
e
g
e
n
er
ic
d
o
m
ai
n
i
m
a
g
e
d
ata
b
ases
(
DB
1
0
,
DB
2
0
an
d
DB
1
0
0
)
w
er
e
e
m
p
lo
y
ed
w
it
h
ea
ch
i
m
ag
e
d
atab
ase
in
d
ex
ed
u
s
in
g
t
w
o
co
lo
u
r
m
o
d
els
(
C
M5
4
an
d
HI
S
T
3
2
)
an
d
t
w
o
tex
t
u
r
e
m
o
d
els
(
GW
5
4
an
d
W
M4
0
)
.
A
d
eg
b
o
l
a,
A
b
o
r
is
ad
e,
P
o
p
o
o
la
an
d
A
ta
y
er
o
[
2
4
]
p
r
esen
ts
d
etailed
d
es
cr
ip
tio
n
o
f
v
ar
io
u
s
i
m
a
g
e
d
atab
ase
an
d
f
ea
t
u
r
e
ex
tr
ac
tio
n
m
o
d
els.
2
.2
.
F
e
a
t
ure
s
elec
t
io
n
m
o
del
I
n
a
g
e
n
er
ic
s
y
s
te
m
,
it
i
s
ex
t
r
e
m
el
y
d
i
f
f
icu
lt
to
k
n
o
w
t
h
e
p
ar
ticu
lar
f
ea
t
u
r
e
m
o
d
el
(
s
)
t
o
b
e
u
s
ed
to
u
n
iq
u
el
y
id
en
ti
f
y
ce
r
tai
n
g
r
o
u
p
s
o
f
i
m
ag
e
s
.
T
h
er
ef
o
r
e,
a
co
m
b
in
a
tio
n
o
f
s
ev
er
al
i
m
a
g
e
f
ea
t
u
r
e
m
o
d
el
s
is
u
s
u
all
y
e
m
p
lo
y
ed
w
it
h
t
h
e
ass
u
m
p
tio
n
t
h
at
at
least
o
n
e
w
il
l
h
av
e
th
e
ab
il
it
y
to
ca
p
tu
r
e
th
e
u
n
iq
u
e
id
e
n
tit
y
o
f
th
e
tar
g
eted
i
m
ag
e
s
.
T
h
is
ap
p
r
o
ac
h
p
o
s
es
s
ev
er
al
ch
alle
n
g
e
s
.
First,
b
ec
au
s
e
t
h
e
i
m
a
g
e
f
ea
tu
r
e
s
ar
e
ca
s
ca
d
ed
as
a
f
lat
v
ec
to
r
,
s
u
c
h
ar
r
an
g
e
m
e
n
t
m
a
y
i
n
cr
ea
s
e
th
e
c
h
an
ce
s
o
f
d
ilu
tin
g
th
e
f
ea
tu
r
e
co
m
p
o
n
e
n
t
th
at
u
n
iq
u
el
y
id
en
ti
f
ie
s
t
h
e
tar
g
eted
i
m
a
g
e
g
r
o
u
p
.
T
h
i
s
m
a
y
also
lead
to
w
h
at
i
s
k
n
o
w
n
as
cu
r
s
e
o
f
d
imen
s
io
n
a
lity
in
C
B
I
R
s
y
s
te
m
t
h
at
e
m
p
lo
y
s
m
ac
h
i
n
e
lear
n
i
n
g
tec
h
n
iq
u
es
f
o
r
r
elev
an
ce
f
ee
d
b
ac
k
.
C
o
s
t
o
f
f
ea
tu
r
e
e
x
tr
ac
tio
n
al
g
o
r
ith
m
i
s
an
o
t
h
er
is
s
u
e
w
h
ich
m
a
y
b
ec
o
m
e
p
r
o
h
ib
iti
v
e
as
t
h
e
n
u
m
b
er
o
f
f
ea
t
u
r
e
d
escr
ip
to
r
s
in
cr
ea
s
es.
I
n
v
ie
w
o
f
t
h
i
s
,
i
n
clu
d
i
n
g
to
o
m
an
y
f
ea
tu
r
es
i
s
o
b
v
io
u
s
l
y
n
o
t
f
ea
s
i
b
le
f
o
r
ap
p
licatio
n
in
v
o
l
v
i
n
g
h
u
m
a
n
-
m
ac
h
i
n
e
i
n
ter
ac
tio
n
.
Si
n
ce
s
u
c
h
s
y
s
te
m
i
s
e
x
p
ec
ted
to
b
e
f
ast
e
n
o
u
g
h
f
o
r
s
m
o
o
t
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
18
,
No
.
4
,
A
u
g
u
s
t 2
0
2
0
:
1
8
9
2
-
1
8
9
6
1894
in
ter
ac
tio
n
,
t
h
e
s
elec
t
io
n
o
f
m
o
s
t
ap
p
r
o
p
r
iate
f
ea
tu
r
es
to
r
eld
u
ce
co
m
p
u
tat
io
n
al
b
u
r
d
en
b
ec
o
m
e
s
i
m
p
er
ati
v
e
an
d
to
ac
h
iev
e
t
h
i
s
,
a
p
r
o
ce
d
u
r
e
th
at
u
s
es
Pr
in
cip
al
C
o
m
p
o
n
e
n
t
An
al
y
s
is
i
s
e
m
p
lo
y
ed
in
t
h
i
s
w
o
r
k
.
Ass
u
m
e
a
b
in
ar
y
clas
s
i
f
ica
tio
n
p
r
o
b
lem
,
g
i
v
en
a
s
et
o
f
lab
el
tr
ain
in
g
d
ata
{
(
,
)
=
1
,
|
≠
0
}
w
h
er
e
s
a
m
p
le
∈
ℝ
an
d
∈
{
−
1
,
1
}
.
L
et
=
{
1
,
2
,
…
.
,
}
(
1)
b
e
th
e
s
et
o
f
all
f
ea
t
u
r
es u
n
d
er
ex
a
m
i
n
atio
n
,
an
d
let
=
{
(
,
)
|
=
1
,
2
,
…
.
,
}
=
{
[
1
,
2
,
…
.
,
]
,
|
=
1
,
2
,
…
.
,
}
(
2)
d
en
o
te
th
e
tr
ain
i
n
g
s
et
co
n
tai
n
in
g
N
tr
ai
n
i
n
g
p
air
s
,
w
h
er
e
is
th
e
n
u
m
er
ical
v
a
lu
e
o
f
f
ea
t
u
r
e
f
o
r
th
e
i
th
tr
ain
i
n
g
s
a
m
p
le.
T
h
e
g
o
al
o
f
d
i
m
en
s
io
n
al
it
y
r
ed
u
cti
o
n
is
to
f
in
d
a
m
i
n
i
m
al
s
et
o
f
f
ea
tu
r
e
s
=
{
1
,
2
,
…
.
,
}
to
r
ep
r
esen
t th
e
i
n
p
u
t
v
ec
to
r
X
in
a
lo
w
er
d
i
m
e
n
s
io
n
al
s
p
ac
e
as
=
{
1
,
2
,
…
.
,
}
(
3)
w
h
er
e
<
,
w
h
ile
t
h
e
clas
s
i
f
icatio
n
o
b
tain
ed
in
t
h
e
lo
w
-
d
i
m
e
n
s
i
o
n
al
s
p
ac
e
s
till
y
ield
s
t
h
e
d
esir
ed
ac
cu
r
ac
y
.
2
.
3
.
P
rincipa
l c
o
m
po
nent
a
n
a
ly
s
is
P
C
A
is
a
s
tatis
t
ical
p
r
io
ce
d
u
r
e
f
o
r
h
ig
h
d
i
m
e
n
s
io
n
alit
y
r
ed
u
c
tio
n
o
f
f
ea
t
u
r
e
s
p
ac
e.
I
t
u
s
es
o
r
th
o
g
o
n
al
tr
an
s
f
o
r
m
atio
n
to
d
ec
o
r
r
elate
a
s
et
o
f
co
r
r
elate
d
f
ea
t
u
r
e
s
p
ac
e
to
en
h
a
n
ce
v
ar
ian
c
e
b
y
e
m
p
h
az
is
i
n
g
th
e
d
ir
ec
tio
n
s
o
f
p
r
in
cip
al
v
ar
iatio
n
o
f
d
ataset
[
2
5
]
.
C
o
n
s
id
er
a
s
et
o
f
d
-
d
im
en
s
io
n
a
l
v
ec
to
r
s
{
=
[
1
,
…
,
]
}
w
it
h
d
is
tr
ib
u
tio
n
ce
n
tr
ed
at
th
e
o
r
ig
in
,
(
)
=
0
.
T
h
e
co
v
ar
ian
ce
is
o
b
tain
ed
u
s
i
n
g
(
4
)
=
{
(
−
)
(
−
)
}
=
{
}
,
(
4)
w
h
er
e
is
t
h
e
ex
p
ec
tatio
n
o
p
er
ato
r
.
T
h
e
p
ar
am
eter
s
ca
n
b
e
ar
r
an
g
ed
to
f
o
r
m
t
h
e
×
co
v
ar
ian
c
e
m
atr
ix
=
{
(
−
)
(
−
)
}
=
{
}
(
5)
Ass
u
m
in
g
(
)
≠
0
,
th
en
b
y
ap
p
l
y
in
g
ei
g
en
v
ec
to
r
d
ec
o
m
p
o
s
it
i
o
n
,
ca
n
b
e
d
ec
o
m
p
o
s
ed
in
t
o
th
e
p
r
o
d
u
ct
o
f
th
r
ee
m
atr
ices:
=
Λ
−
1
(
6)
w
h
er
e,
Λ
=
dia
g
{
λ
1
,
…
,
λ
d
}
is
th
e
E
ig
e
n
v
al
u
e
m
a
tr
ix
.
=
[
w
1
,
…
,
w
d
]
T
f
o
r
m
s
a
s
et
o
f
o
r
th
o
n
o
r
m
al
b
asis
v
ec
to
r
s
ca
lled
E
ig
e
n
v
ec
to
r
s
.
Fo
r
d
i
m
en
s
io
n
al
it
y
r
ed
u
ctio
n
,
o
n
l
y
th
e
s
et
o
f
o
r
th
o
n
o
r
m
a
l
b
ases
v
ec
to
r
s
r
es
u
lti
n
g
f
r
o
m
t
h
e
k
-
lar
g
est
E
ig
en
v
al
u
es
ar
e
r
etai
n
ed
.
T
h
is
w
i
ll
r
es
u
lt
i
n
to
s
i
g
n
if
ica
n
t
f
ea
tu
r
e
d
i
m
en
s
io
n
al
it
y
r
e
d
u
ctio
n
.
No
r
m
al
l
y
,
th
e
k
-
lar
g
est
E
i
g
e
n
v
al
u
es
t
h
a
t
c
o
n
s
tit
u
tes
9
5
%
o
f
t
h
e
to
tal
E
ig
en
v
al
u
es
ar
e
r
etain
ed
f
o
r
d
i
m
en
s
io
n
ali
t
y
r
ed
u
ctio
n
.
Ho
w
e
v
er
,
th
is
w
o
r
k
e
m
p
lo
y
ed
p
r
ec
is
io
n
/r
ec
all
g
r
ap
h
to
d
ete
r
m
i
n
e
th
e
d
i
m
en
s
i
o
n
o
f
f
ea
tu
r
e
to
b
e
r
etain
ed
.
T
h
is
is
a
m
o
r
e
o
b
j
ec
tiv
e
c
h
o
ice,
s
i
n
ce
t
h
e
r
es
u
lt
i
n
g
lo
w
er
d
i
m
e
n
s
io
n
al
f
ea
t
u
r
e
v
ec
to
r
s
ar
e
u
s
ed
f
o
r
d
is
tan
ce
(
s
i
m
ilar
it
y
)
m
ea
s
u
r
e
m
en
t
i
n
i
m
a
g
e
r
etr
iev
al
s
y
s
te
m
w
i
th
r
ele
v
an
ce
f
ee
d
b
ac
k
.
C
o
n
s
eq
u
e
n
tl
y
,
th
e
n
u
m
b
er
o
f
f
ea
tu
r
e
d
i
m
en
s
io
n
r
eta
in
ed
i
s
b
ased
o
n
a
5
%
m
a
x
i
m
u
m
lo
s
s
co
n
s
tr
ain
t
i
m
p
o
s
ed
o
n
th
e
p
r
ec
is
io
n
/r
ec
all
g
r
ap
h
.
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
C
o
m
b
in
a
t
i
o
n
o
f
v
is
u
a
l
d
es
c
r
i
p
t
o
r
s
r
e
s
u
lt
s
t
o
in
c
r
e
as
e
in
th
e
d
im
en
s
i
o
n
o
f
th
e
r
esu
l
t
in
g
f
e
a
tu
r
e
v
e
c
t
o
r
.
N
o
r
m
al
ly
,
th
e
r
e
s
u
lt
in
g
f
e
at
u
r
e
m
o
d
e
l
,
w
h
i
ch
i
s
th
e
c
o
n
c
at
en
at
i
o
n
o
f
in
d
iv
i
d
u
a
l
f
e
a
tu
r
e
v
e
ct
o
r
s
,
c
o
u
l
d
h
av
e
v
e
r
y
h
ig
h
d
im
en
s
i
o
n
s
an
d
th
u
s
in
c
r
e
a
s
e
t
h
e
l
at
en
cy
o
f
R
F
s
ch
em
e
e
v
en
o
n
a
m
e
d
iu
m
-
s
i
z
e
im
ag
e
d
a
t
a
b
a
s
e
.
H
en
c
e
,
in
o
r
d
e
r
t
o
m
i
t
ig
at
e
th
e
cu
r
s
e
o
f
d
im
e
n
s
i
o
n
a
li
ty
p
r
o
b
l
em
a
s
s
o
c
i
a
t
e
d
w
i
th
m
ac
h
in
e
le
a
r
n
in
g
b
a
s
e
d
R
F
s
ch
em
e
,
r
e
d
u
c
in
g
th
e
d
im
en
s
i
o
n
s
o
f
f
e
at
u
r
e
v
e
ct
o
r
s
m
ay
b
e
n
e
ce
s
s
a
r
y
.
I
n
th
is
s
tu
d
y
,
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
a
ly
s
i
s
(
PC
A
)
i
s
in
te
g
r
at
e
d
t
o
th
e
d
ev
el
o
p
e
d
O
C
-
S
V
M
R
F
f
o
r
th
e
p
u
r
p
o
s
e
o
f
f
e
a
tu
r
e
v
ec
t
o
r
d
im
en
s
i
o
n
al
i
ty
r
e
d
u
c
ti
o
n
.
A
cr
iter
io
n
o
f
5
%
m
a
x
i
m
u
m
d
e
g
r
ad
ati
o
n
in
m
ea
n
p
r
ec
is
io
n
v
al
u
e
w
as
u
s
ed
to
d
eter
m
i
n
e
th
e
d
i
m
e
n
s
io
n
o
f
f
ea
t
u
r
e
v
ec
to
r
t
o
k
ee
p
.
T
h
e
ef
f
ec
t
o
f
f
e
atu
r
e
v
ec
to
r
d
i
m
en
s
io
n
al
it
y
r
ed
u
ctio
n
is
s
h
o
w
n
in
Fig
u
r
e
1
.
T
h
e
m
a
x
i
m
u
m
m
ea
n
p
r
ec
is
io
n
v
alu
e
s
obt
a
i
n
ed
o
n
DB
1
0
,
DB
2
0
an
d
DB
1
0
0
w
er
e
0
.
9
0
6
7
,
0
.
7
2
6
6
an
d
0
.
7
2
7
5
r
esp
ec
tiv
el
y
,
f
o
r
8
0
%
r
e
d
u
ctio
n
in
f
ea
t
u
r
e
v
ec
to
r
d
im
en
s
io
n
.
W
h
ile
a
r
ed
u
ctio
n
o
f
f
ea
t
u
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
A
p
r
in
cip
a
l c
o
mp
o
n
en
t
a
n
a
lys
is
-
b
a
s
ed
fea
tu
r
e
d
imen
s
io
n
a
lit
y
r
ed
u
ctio
n
s
ch
eme
…
(
Olu
w
o
le
A
.
A
d
eg
b
o
la
)
1895
d
i
m
en
s
io
n
b
y
8
3
%
f
o
r
DB
1
0
,
DB
2
0
an
d
DB
1
0
0
r
esu
lted
in
to
m
ea
n
p
r
ec
is
io
n
v
a
lu
e
s
o
f
0
.
6
9
3
3
,
0
.
5
0
9
3
an
d
0
.
3
6
5
7
r
esp
ec
tiv
el
y
.
Fig
u
r
e
2
s
h
o
w
s
th
e
co
m
p
ar
is
o
n
b
et
w
ee
n
t
h
e
O
C
-
SVM
R
F
th
at
u
s
ed
t
h
e
w
h
o
le
1
74
-
d
i
m
en
s
io
n
a
l
f
ea
t
u
r
e
(
ST
D)
an
d
th
e
OC
-
SVM
R
F
w
i
th
P
C
A
th
at
u
s
ed
3
5
-
d
i
m
en
s
io
n
a
l
f
ea
t
u
r
es
(
P
C
A
)
.
T
h
e
m
a
x
i
m
u
m
m
ea
n
p
r
ec
is
io
n
v
al
u
es
o
f
0
.
9
4
0
0
,
0
.
7
6
0
0
a
n
d
0
.
7
8
6
0
w
er
e
ac
h
ie
v
ed
o
n
th
e
DB
1
0
,
DB
2
0
an
d
DB
1
0
0
r
esp
ec
tiv
el
y
f
o
r
t
h
e
ST
D
.
T
h
e
m
ax
i
m
u
m
m
ea
n
p
r
ec
is
io
n
ac
h
ie
v
ed
w
it
h
P
C
A
o
n
t
h
e
DB
1
0
,
DB
2
0
an
d
DB
1
0
0
w
er
e
0
.
9
0
6
7
,
0
.
7
2
6
6
an
d
0
.
7
2
7
5
r
esp
ec
tiv
el
y
.
T
h
u
s
a
n
8
0
%
r
ed
u
ctio
n
i
n
f
ea
t
u
r
e
d
i
m
e
n
s
io
n
,
y
ield
ed
to
ler
ab
le
d
eg
r
ad
atio
n
o
f
3
.
5
4
%,
4
.
3
9
%
an
d
7
.
4
%
in
m
ax
i
m
u
m
m
ea
n
p
r
e
cisi
o
n
p
er
f
o
r
m
a
n
ce
o
n
DB
1
0
,
DB
2
0
an
d
DB
1
0
0
r
esp
ec
tiv
el
y
.
D
a
t
a
b
a
se
D
B
1
0
D
B
2
0
D
B
1
0
0
%
r
e
d
u
c
t
i
o
n
80%
8
3
%
80%
8
3
%
80%
8
3
%
M
e
a
n
P
r
e
c
i
si
o
n
0
.
9
0
6
7
0
.
6
9
3
3
0
.
7
2
6
6
0
.
5
0
9
3
0
.
7
2
7
5
0
.
3
6
5
7
Fig
u
r
e
1
.
Me
an
p
r
ec
is
io
n
r
es
u
l
t
o
f
th
e
O
C
-
SVM
R
F
w
it
h
P
C
A
o
f
d
i
f
f
er
en
t d
i
m
en
s
io
n
al
it
y
r
ed
u
ctio
n
D
a
t
a
b
a
se
D
B
1
0
D
B
2
0
D
B
1
0
0
%
r
e
d
u
c
t
i
o
n
S
T
D
P
C
A
S
T
D
P
C
A
S
T
D
P
C
A
M
e
a
n
P
r
e
c
i
si
o
n
0
.
9
4
0
0
0
.
9
0
6
7
0
.
7
6
0
0
0
.
7
2
6
6
0
.
7
8
6
0
0
.
7
2
7
5
Fig
u
r
e
2
.
Me
an
p
r
ec
is
io
n
r
es
u
l
t
o
f
th
e
O
C
-
SVM
r
elev
a
n
ce
f
e
ed
b
ac
k
w
it
h
P
C
A
u
tili
zi
n
g
8
0
% d
i
m
e
n
s
io
n
alit
y
r
ed
u
ctio
n
4.
CO
NCLU
SI
O
N
I
n
C
B
I
R
s
y
s
te
m
d
e
s
ig
n
ed
f
o
r
g
e
n
er
ic
i
m
a
g
e
d
atab
ases
,
it
is
g
en
er
al
p
r
ac
tice
to
r
ep
r
esen
t
i
m
a
g
e
s
u
s
i
n
g
co
m
b
i
n
atio
n
o
f
s
e
v
er
al
d
if
f
er
e
n
t
i
m
a
g
e
f
ea
t
u
r
es
w
i
th
a
v
ie
w
to
ca
p
tu
r
i
n
g
ex
tr
a
i
n
f
o
r
m
at
io
n
t
h
at
m
a
y
i
m
p
r
o
v
e
r
etr
iev
al
ac
cu
r
ac
y
.
T
h
is
u
s
u
all
y
r
es
u
lt
s
in
h
i
g
h
d
i
m
en
s
io
n
a
lit
y
o
f
v
is
u
al
f
ea
tu
r
e
v
ec
to
r
s
f
o
r
C
B
I
R
s
y
s
te
m
w
i
th
clas
s
i
f
ier
-
b
ased
r
elev
an
ce
f
ee
d
b
ac
k
s
ch
e
m
e.
I
n
th
is
p
ap
er
,
th
e
is
s
u
e
o
f
cu
r
s
e
o
f
d
im
e
n
s
io
n
alit
y
is
ad
d
r
ess
ed
u
s
in
g
a
P
C
A
-
b
a
s
ed
f
ea
tu
r
e
s
elec
tio
n
ap
p
r
o
ac
h
.
T
h
e
f
ea
tu
r
e
s
elec
tio
n
m
o
d
el
w
as
i
n
c
o
r
p
o
r
ated
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
18
,
No
.
4
,
A
u
g
u
s
t 2
0
2
0
:
1
8
9
2
-
1
8
9
6
1896
in
to
an
ex
i
s
is
tin
g
OC
-
SVM
R
F
r
etr
iev
al
s
y
s
te
m
.
T
h
e
f
in
d
in
g
s
r
ev
ea
led
th
a
t
b
y
allo
w
i
n
g
a
5
%
lo
s
s
to
ler
an
ce
in
m
ea
n
p
r
ec
is
o
n
,
it
w
a
s
p
o
s
s
ib
le
to
ac
h
ie
v
e
8
0
%
r
ed
u
ctio
n
i
n
f
ea
tu
r
e
v
ec
to
r
d
i
m
e
n
s
io
n
alit
y
,
w
h
ile
a
tte
m
p
t
to
in
cr
ea
s
e
th
e
p
er
ce
n
ta
g
e
r
ed
u
c
tio
n
o
f
f
ea
t
u
r
e
v
ec
to
r
d
i
m
e
n
s
io
n
r
es
u
lted
in
to
p
o
o
r
r
etr
iev
al
r
esu
lt
s
.
Ref
er
ence
s
[1
]
Z
.
S
u
,
S
.
L
i
,
a
n
d
H
.
Z
h
a
n
g
,
"
E
x
tr
a
c
t
i
o
n
o
f
f
e
a
t
u
r
e
s
u
b
s
p
a
c
e
s
f
o
r
c
o
n
t
e
n
t
-
b
a
s
e
d
r
e
t
r
i
e
v
a
l
u
s
i
n
g
r
e
le
v
a
n
c
e
f
e
e
d
b
a
c
k
,
"
P
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
n
i
n
t
h
A
C
M
i
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
r
e
n
c
e
o
n
M
u
l
t
i
me
d
i
a
,
p
p
.
9
8
-
1
0
6
,
2
0
0
1
.
[2
]
A
.
M
a
ra
k
a
k
is
,
N
.
G
a
l
a
ts
a
n
o
s
,
A
.
L
ik
a
s
,
a
n
d
A
.
S
t
a
f
y
l
o
p
a
t
i
s
,
"
R
e
l
e
v
a
n
c
e
f
e
e
d
b
a
c
k
f
o
r
c
o
n
t
e
n
t
-
b
a
s
e
d
im
a
g
e
re
t
r
i
e
v
a
l
u
s
i
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s
a
n
d
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
,
"
A
r
t
i
f
ic
i
a
l
Ne
u
r
a
l
N
e
t
w
o
rk
s
–
I
C
A
N
N
2
0
0
9
,
S
p
r
i
n
g
e
r
,
v
o
l
.
5768
,
p
p
.
9
4
2
-
9
5
1
,
2
0
0
9
.
[3
]
Y
.
L
i
u
a
n
d
Y
.
F
.
Z
h
e
n
g
,
"
F
S
_
S
F
S
:
A
n
o
v
e
l
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s
,
"
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
,
v
o
l
.
3
9
,
n
o
.
7
,
p
p
.
1
3
3
3
-
1
3
4
5
,
2
0
0
6
.
[4
]
K.
P
.
C
h
u
n
g
,
"
I
n
t
e
l
l
i
g
e
n
t
c
o
n
t
e
n
t
-
b
a
s
e
d
im
a
g
e
re
t
r
i
e
v
a
l
f
ra
m
e
w
o
rk
b
a
se
d
o
n
s
e
m
i
-
a
u
t
o
m
a
te
d
l
e
a
r
n
i
n
g
a
n
d
h
i
s
t
o
r
i
c
p
r
o
f
i
l
e
s
,
"
T
h
e
s
i
s
,
M
u
r
d
o
c
h
U
n
i
v
e
rs
i
ty
,
2
0
0
7
.
[5
]
Y
.
L
u
,
I
.
C
o
h
e
n
,
X.
S
.
Z
h
o
u
,
a
n
d
Q
.
T
ia
n
,
"
F
e
a
t
u
r
e
se
l
e
c
t
i
o
n
u
s
i
n
g
p
r
i
n
c
i
p
a
l
f
e
a
t
u
r
e
a
n
a
ly
s
i
s
,
"
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
1
5
t
h
A
C
M
i
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
r
e
n
c
e
o
n
M
u
l
t
i
m
e
d
i
a
,
p
p
.
3
0
1
-
304
,
2
0
0
7
.
[6
]
R
.
T
a
v
o
l
i
a
n
d
F
.
M
a
h
m
o
u
d
i
,
"
P
CA
-
b
a
se
d
r
e
le
v
a
n
c
e
f
e
e
d
b
a
c
k
i
n
d
o
c
u
m
e
n
t
im
a
g
e
r
e
t
r
ie
v
a
l
,
"
I
J
CS
I
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
I
s
s
u
e
s
,
v
o
l
.
9
,
n
o
.
2
,
p
p
.
4
9
7
-
5
0
2
,
J
u
l
y
2012.
[7
]
M
.
S
u
g
a
n
t
h
y
a
n
d
P
.
R
a
m
a
m
o
o
r
th
y
,
"
P
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
a
ly
s
is
b
a
se
d
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
,
m
o
r
p
h
o
l
o
g
i
c
a
l
e
d
g
e
d
e
t
e
c
t
i
o
n
a
n
d
l
o
c
a
l
i
z
a
t
i
o
n
f
o
r
f
a
s
t
ir
i
s
r
e
c
o
g
n
i
t
i
o
n
,
"
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
sc
i
e
n
c
e
,
v
o
l
.
8
,
n
o
.
9
,
p
p
.
1
4
2
8
-
1433
,
2
0
1
2
.
[8
]
R
.
K
o
h
a
v
i
a
n
d
G
.
H
.
J
o
h
n
,
"
W
r
a
p
p
e
r
s
f
o
r
f
e
a
t
u
r
e
s
u
b
s
e
t
s
e
l
e
c
t
i
o
n
,
"
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
9
7
,
n
o
.
1
-
2
,
p
p
.
2
7
3
-
3
2
4
,
D
e
c
e
m
b
e
r
1997.
[9
]
F
.
A
l
o
n
s
o
A
t
i
e
n
z
a
,
J
.
L
.
R
o
j
o
Á
lv
a
re
z
,
A
.
R
o
sa
d
o
M
u
ñ
o
z
,
J
.
J
.
Vi
n
a
g
r
e
,
A
.
G
a
rc
í
a
A
l
b
e
r
o
l
a
,
a
n
d
G
.
C
a
m
p
s
Va
l
l
s
,
"
F
e
a
t
u
r
e
s
e
le
c
t
i
o
n
u
s
i
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s
a
n
d
b
o
o
t
s
t
r
a
p
m
e
th
o
d
s
f
o
r
v
e
n
t
r
i
c
u
l
a
r
f
i
b
r
i
l
l
a
t
i
o
n
d
e
t
e
c
t
i
o
n
,
"
E
x
p
e
r
t
S
y
s
t
e
ms
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
9
,
n
o
.
2
,
p
p
.
1
9
5
6
-
1
9
6
7
,
2
0
1
2
.
[1
0
]
I
.
G
u
y
o
n
,
J
.
W
e
s
t
o
n
,
S
.
B
a
r
n
h
i
l
l
,
a
n
d
V.
Va
p
n
i
k
,
"
G
e
n
e
se
l
e
c
t
i
o
n
f
o
r
c
a
n
c
e
r
c
la
s
s
if
ic
a
t
i
o
n
u
s
i
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s
,
"
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
v
o
l
.
4
6
,
p
p
.
3
8
9
-
4
2
2
,
2
0
0
2
.
[1
1
]
B
.
H
.
J
u
a
n
g
a
n
d
S
.
K
a
t
a
g
i
r
i
,
"
D
i
sc
r
im
i
n
a
t
iv
e
le
a
r
n
i
n
g
f
o
r
m
i
n
im
u
m
e
r
r
o
r
c
l
a
ss
if
i
c
a
t
i
o
n
(
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
)
,
"
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
s
i
g
n
a
l
p
r
o
c
e
s
s
i
n
g
,
v
o
l
.
4
0
,
n
o
.
1
2
,
p
p
.
3
0
4
3
-
3
0
5
4
,
1
9
9
2
.
[1
2
]
H
.
W
a
ta
n
a
b
e
,
T
.
Y
a
m
a
g
u
c
h
i
,
a
n
d
S
.
K
a
t
a
g
i
r
i
,
"
D
i
sc
r
im
i
n
a
t
iv
e
m
e
t
r
i
c
d
e
s
ig
n
f
o
r
r
o
b
u
s
t
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
,
"
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
S
i
g
n
a
l
P
r
o
c
e
s
s
i
n
g
,
v
o
l
.
4
5
,
n
o
.
1
1
,
p
p
.
2
6
5
5
-
2
6
6
2
,
1
9
9
7
.
[1
3
]
I.
Jo
ll
if
f
e
,
"
P
r
in
c
ip
a
l
c
o
m
p
o
n
e
n
t
a
n
a
ly
sis,
"
In
ter
n
a
ti
o
n
a
l
e
n
c
y
c
lo
p
e
d
i
a
o
f
sta
ti
stica
l
sc
ie
n
c
e
,
S
p
rin
g
e
r
,
p
p
.
1
0
9
4
-
1
0
9
6
,
2
0
1
1
[1
4
]
A
.
T
sy
m
b
a
l
,
S
.
P
u
u
r
o
n
e
n
,
M
.
P
e
c
h
e
n
i
z
k
iy
,
M
.
B
a
u
m
g
a
r
t
e
n
,
a
n
d
D
.
W
.
P
a
t
t
e
r
s
o
n
,
"
E
i
g
e
n
v
e
c
t
o
r
-
B
a
s
e
d
F
e
a
t
u
r
e
E
x
t
r
a
c
t
i
o
n
f
o
r
C
la
s
s
if
ic
a
t
i
o
n
,
"
F
L
AI
RS
-
0
2
P
r
o
c
e
e
d
i
n
g
s
,
p
p
.
3
5
4
-
3
5
8
,
2002
.
[1
5
]
W
.
J
.
K
rz
a
n
o
w
sk
i
,
"
S
e
l
e
c
t
i
o
n
o
f
v
a
r
i
a
b
l
e
s
t
o
p
r
e
se
rv
e
m
u
l
t
iv
a
r
i
a
te
d
a
t
a
s
t
r
u
c
t
u
r
e
,
u
s
i
n
g
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
s
,
"
A
p
p
l
i
e
d
S
t
a
t
i
s
t
i
c
s
,
v
o
l
.
3
6
,
n
o
.
1
,
p
p
.
2
2
-
3
3
,
1
9
8
7
.
[1
6
]
W
.
J
.
K
r
z
a
n
o
w
s
k
i
,
"
A
s
t
o
p
p
i
n
g
r
u
l
e
f
o
r
s
tr
u
c
t
u
r
e
-
p
r
e
s
e
rv
i
n
g
v
a
r
i
a
b
l
e
s
e
l
e
c
t
i
o
n
,
"
S
t
a
t
i
s
t
i
c
s
a
n
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
6
,
p
p
.
5
1
-
5
6
,
1
9
9
6
.
[1
7
]
M
.
M
a
n
s
o
r
,
S
.
Ya
a
c
o
b
,
M
.
Ha
r
i
h
a
r
a
n
,
S
.
Ba
s
a
h
,
S
.
A
.
J
a
m
i
l
,
M
.
M
.
K
h
i
d
i
r
,
e
t
a
l
.
,
"
F
u
z
z
y
k
-
N
N
a
n
d
k
-
N
N
A
lg
o
r
i
t
h
m
f
o
r
f
a
s
t
I
n
f
a
n
t
C
u
e
s
De
t
e
c
t
i
o
n
,
"
W
o
r
l
d
C
o
n
g
r
e
s
s
o
n
M
e
d
i
c
a
l
P
h
y
s
i
c
s
a
n
d
B
i
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
M
a
y
2
6
-
3
1
,
2
0
1
2
,
B
e
i
j
i
n
g
,
C
h
i
n
a
,
2
0
1
3
,
p
p
.
1
2
6
0
-
1
2
6
3
.
[1
8
]
F
.
S
o
n
g
,
Z
.
G
u
o
,
a
n
d
D
.
M
e
i
,
"
F
e
a
t
u
r
e
se
l
e
c
t
i
o
n
u
s
i
n
g
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
a
ly
s
is
,
"
2
0
1
0
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
y
s
t
e
m
S
c
i
e
n
c
e
,
E
n
g
i
n
e
e
r
i
n
g
D
e
s
i
g
n
a
n
d
M
a
n
u
f
a
c
t
u
r
i
n
g
I
n
f
o
r
m
a
t
i
z
a
t
i
o
n
,
p
p
.
2
7
-
30
,
2
0
1
0
.
[1
9
]
I
.
J
.
C
o
x
,
M
.
L
.
M
i
l
l
e
r
,
T
.
P
.
M
i
n
k
a
,
T
.
V.
P
a
p
a
t
h
o
m
a
s
,
a
n
d
P
.
N
.
Y
i
a
n
i
l
o
s
,
"
T
h
e
B
a
y
e
s
i
a
n
im
a
g
e
re
t
r
i
e
v
a
l
sy
s
tem
,
P
i
c
H
u
n
t
e
r
:
t
h
e
o
r
y
,
im
p
lem
e
n
t
a
t
i
o
n
,
a
n
d
p
s
y
c
h
o
p
h
y
s
i
c
a
l
e
x
p
e
r
im
e
n
t
s
,
"
I
E
E
E
t
r
a
n
s
a
c
t
i
o
n
s
o
n
i
m
a
g
e
p
r
o
c
e
ss
i
n
g
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
2
0
-
3
7
,
2
0
0
0
.
[2
0
]
N
.
Va
s
c
o
n
c
e
l
o
s
a
n
d
A
.
L
i
p
p
m
a
n
,
"
L
e
a
r
n
i
n
g
f
r
o
m
u
se
r
f
e
e
d
b
a
c
k
i
n
i
m
a
g
e
re
t
r
i
e
v
a
l
sy
s
t
e
m
s
,
"
A
d
v
a
n
c
e
s
i
n
n
e
u
r
a
l
i
n
f
o
r
m
a
t
i
o
n
p
r
o
c
e
s
s
i
n
g
s
y
s
t
e
m
s
,
p
p
.
9
7
7
-
986
,
2
0
0
0
.
[2
1
]
P.
Y
.
Y
i
n
,
B
.
B
h
a
n
u
,
K
.
C
.
C
h
a
n
g
,
a
n
d
A
.
D
o
n
g
,
"
I
n
t
e
g
r
a
t
i
n
g
r
e
le
v
a
n
c
e
f
e
e
d
b
a
c
k
t
e
c
h
n
i
q
u
e
s
f
o
r
im
a
g
e
r
e
t
r
i
e
v
a
l
u
s
i
n
g
r
e
i
n
f
o
rc
e
m
e
n
t
l
e
a
r
n
i
n
g
,
"
I
E
E
E
t
r
a
n
s
a
c
t
i
o
n
s
o
n
p
a
t
t
e
r
n
a
n
a
l
y
s
i
s
a
n
d
m
a
c
h
i
n
e
i
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
2
7
,
n
o
.
1
0
,
p
p
.
1
5
3
6
-
1
5
5
1
,
2
0
0
5
.
[2
2
]
Y
.
R
u
i
,
T
.
S
.
H
u
a
n
g
,
a
n
d
S
.
-
F
.
C
h
a
n
g
,
"
Im
a
g
e
r
e
tr
i
e
v
a
l
:
C
u
r
re
n
t
t
e
c
h
n
i
q
u
e
s
,
p
r
o
m
i
s
i
n
g
d
i
r
e
c
t
i
o
n
s
,
a
n
d
o
p
e
n
i
s
s
u
e
s
,
"
J
o
u
r
n
a
l
o
f
v
i
s
u
a
l
c
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
i
m
a
g
e
r
e
p
r
e
s
e
n
t
a
t
i
o
n
,
v
o
l
.
1
0
,
n
o
.
1
,
p
p
.
3
9
-
6
2
,
1
9
9
9
.
[2
3
]
F
.
L
o
n
g
,
H
.
Z
h
a
n
g
,
a
n
d
D
.
D
.
F
e
n
g
,
"
F
u
n
d
a
m
e
n
t
a
ls
o
f
c
o
n
t
e
n
t
-
b
a
s
e
d
im
a
g
e
r
e
t
r
ie
v
a
l
,
"
M
u
l
t
im
e
d
i
a
I
n
f
o
rm
a
t
i
o
n
R
e
t
r
i
e
v
a
l
a
n
d
M
a
n
a
g
e
m
e
n
t
,
S
p
r
i
n
g
e
r
,
p
p
.
1
-
26
,
2
0
0
3
.
[2
4
]
O
.
A
.
A
d
e
g
b
o
l
a
,
D
.
O
.
A
b
o
r
i
s
a
d
e
,
S
.
I
.
P
o
p
o
o
l
a
,
a
n
d
A
.
A
.
A
tay
e
r
o
,
"
P
e
r
f
o
rm
a
n
c
e
Ev
a
l
u
a
t
i
o
n
o
f
Vi
s
u
a
l
D
e
sc
r
i
p
t
o
r
s
f
o
r
Im
a
g
e
I
n
d
e
x
i
n
g
i
n
C
o
n
t
e
n
t
B
a
s
e
d
Im
a
g
e
R
e
tr
i
e
v
a
l
S
y
s
tem
s
,
"
I
C
C
S
A
2
0
1
8
:
C
o
m
p
u
t
a
t
i
o
n
a
l
S
c
i
e
n
c
e
a
n
d
I
t
s
A
p
p
l
i
c
a
t
i
o
n
s
–
I
C
C
S
A
2
0
1
8
,
p
p
.
5
3
9
-
5
4
9
,
2
0
1
8
.
[2
5
]
K
.
I
.
D
i
a
m
a
n
t
a
r
a
s
a
n
d
S
.
Y
.
K
u
n
g
,
P
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
s
:
t
h
e
o
ry
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s
:
J
o
h
n
W
i
l
e
y
&
S
o
n
s
,
I
n
c
.
,
1
9
9
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.