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.
6
,
No
.
6
,
Dec
em
b
er
201
6
,
p
p
.
3
1
7
4
~
3
1
8
7
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
6i
6
.
1
2
0
4
9
3174
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
Tex
ture
Fusio
n f
o
r Ba
ti
k
Mo
tif
Re
t
riev
a
l Sys
te
m
I
da
Nurha
ida
1
,
H
o
ng
Wei
2
,
Re
mm
y
A.
M
.
Z
e
n
3
,
Ruli M
a
nu
rung
4
,
Ania
t
i M
.
Ary
m
urt
hy
5
1
Co
m
p
u
ter S
c
ien
c
e
F
a
c
u
lt
y
,
Un
iv
e
rsitas
M
e
rc
u
Bu
a
n
a
,
I
n
d
o
n
e
sia
2
S
c
h
o
o
l
o
f
S
y
ste
m
En
g
in
e
e
rin
g
,
Un
iv
e
rsit
y
o
f
Re
a
d
in
g
,
UK
3,
4,
5
Co
m
p
u
ter
S
c
ien
c
e
F
a
c
u
lt
y
,
Un
iv
e
rsitas
In
d
o
n
e
sia
,
In
d
o
n
e
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
2
,
2
0
1
6
R
ev
i
s
ed
No
v
8
,
2
0
1
6
A
cc
ep
ted
No
v
22
,
2
0
1
6
T
h
is
p
a
p
e
r
s
y
ste
m
a
ti
c
a
ll
y
in
v
e
sti
g
a
tes
th
e
e
ff
e
c
t
o
f
i
m
a
g
e
te
x
tu
re
f
e
a
tu
re
s
o
n
b
a
ti
k
m
o
ti
f
r
e
tri
e
v
a
l
p
e
rf
o
r
m
a
n
c
e
.
T
h
e
re
tri
e
v
a
l
p
ro
c
e
ss
u
se
s
a
q
u
e
ry
m
o
ti
f
im
a
g
e
to
f
in
d
m
a
tch
in
g
m
o
ti
f
i
m
a
g
e
s
in
a
d
a
tab
a
se
.
In
th
is
stu
d
y
,
f
e
a
tu
re
f
u
sio
n
o
f
v
a
rio
u
s
ima
g
e
tex
tu
re
fe
a
tu
r
e
s
su
c
h
a
s
G
a
b
o
r,
L
o
g
-
G
a
b
o
r,
G
re
y
L
e
v
e
l
Co
-
Oc
c
u
rre
n
c
e
M
a
tri
c
e
s
(
GL
CM
),
a
n
d
L
o
c
a
l
Bin
a
r
y
P
a
tt
e
rn
(L
BP
)
f
e
a
tu
re
s
a
re
a
tt
e
m
p
ted
in
m
o
ti
f
ima
g
e
re
tri
e
v
a
l.
W
it
h
re
g
a
rd
s
to
p
e
rf
o
r
m
a
n
c
e
e
v
a
lu
a
ti
o
n
,
b
o
t
h
i
n
d
iv
id
u
a
l
f
e
a
tu
re
s
a
n
d
f
u
se
d
f
e
a
tu
re
se
ts
a
r
e
a
p
p
li
e
d
.
Ex
p
e
rime
n
tal
re
su
lt
s
sh
o
w
th
a
t
o
p
ti
m
a
l
fe
a
tu
re
f
u
sio
n
o
u
tp
e
rf
o
rm
s
in
d
iv
id
u
a
l
f
e
a
tu
re
s
in
b
a
ti
k
m
o
ti
f
re
tri
e
v
a
l.
Am
o
n
g
th
e
in
d
iv
id
u
a
l
f
e
a
tu
re
s
tes
ted
,
L
o
g
-
G
a
b
o
r
f
e
a
tu
re
s
p
ro
v
id
e
t
h
e
b
e
st
r
e
su
lt
.
T
h
e
p
r
o
p
o
se
d
a
p
p
ro
a
c
h
is
b
e
st
u
se
d
i
n
a
sc
e
n
a
rio
w
h
e
re
a
q
u
e
ry
i
m
a
g
e
c
o
n
tain
i
n
g
m
u
lt
ip
le
b
a
sic
m
o
ti
f
o
b
jec
ts
is
a
p
p
li
e
d
to
a
d
a
tas
e
t
i
n
w
h
ich
re
tri
e
v
e
d
im
a
g
e
s
a
lso
c
o
n
tain
m
u
lt
ip
le
m
o
ti
f
o
b
jec
ts.
T
h
e
re
tri
e
v
a
l
ra
te
a
c
h
iev
e
s
8
4
.
5
4
%
f
o
r
th
e
ra
n
k
3
p
re
c
isio
n
w
h
e
n
th
e
f
e
a
tu
re
sp
a
c
e
is
f
u
se
d
w
it
h
G
a
b
o
r,
G
L
CM
a
n
d
L
o
g
-
G
a
b
o
r
f
e
a
tu
re
s.
T
h
e
in
v
e
stig
a
ti
o
n
a
lso
sh
o
w
s
th
a
t
th
e
p
ro
p
o
se
d
m
e
th
o
d
d
o
e
s
n
o
t
w
o
rk
w
e
ll
f
o
r
a
re
tri
e
v
a
l
sc
e
n
a
rio
w
h
e
re
th
e
q
u
e
r
y
i
m
a
g
e
c
o
n
tain
s
m
u
lt
ip
le
b
a
sic
m
o
ti
f
o
b
jec
ts
b
e
in
g
a
p
p
l
ied
to
a
d
a
tas
e
t
in
w
h
ich
th
e
re
tri
e
v
e
d
i
m
a
g
e
s
o
n
ly
c
o
n
tain
o
n
e
b
a
sic
m
o
ti
f
o
b
jec
t.
K
ey
w
o
r
d
:
B
atik
B
atik
m
o
ti
f
r
etr
iev
al
s
y
s
te
m
C
o
n
te
n
t
b
ased
i
m
ag
e
r
etr
iev
al
Featu
r
e
f
u
s
io
n
Mo
tif
Co
p
y
rig
h
t
©
2
0
1
6
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
:
I
d
a
Nu
r
h
aid
a,
C
o
m
p
u
ter
Scien
ce
Fac
u
lt
y
,
Un
i
v
er
s
ita
s
Me
r
cu
B
u
a
n
a
,
J
alan
R
a
y
a
Me
r
u
y
a
Selatan
n
o
.
1
,
Kem
b
an
g
a
n
J
ak
ar
ta
B
ar
at
-
1
6
5
5
0
,
I
n
d
o
n
esia.
E
m
ail: id
a.
n
u
r
h
a
id
a@
m
er
cu
b
u
an
a.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
B
atik
is
f
ab
r
ic
o
r
n
a
m
e
n
ted
with
f
ig
u
r
es
o
f
a
s
p
ec
if
ic
n
at
u
r
es
.
B
atik
is
a
f
a
m
o
u
s
c
u
lt
u
r
al
h
er
itag
e
f
r
o
m
I
n
d
o
n
es
ia,
o
r
ig
in
all
y
r
es
er
v
ed
as
an
ar
t
f
o
r
m
f
o
r
J
av
an
ese
r
o
y
alt
y
.
T
h
e
w
o
r
d
b
atik
o
r
ig
in
ate
s
f
r
o
m
t
h
e
J
av
an
ese
tik
a
n
d
m
ea
n
s
to
d
o
t
.
B
atik
i
s
a
m
et
h
o
d
o
f
d
y
ei
n
g
d
esig
n
s
o
n
clo
t
h
b
y
co
ati
n
g
th
e
p
ar
ts
t
h
at
ar
e
n
o
t
to
b
e
d
y
ed
w
it
h
r
e
m
o
v
ab
le
wax
[1
-
2]
.
T
h
e
p
atter
n
s
ar
e
d
r
a
w
n
u
s
in
g
a
s
p
o
o
n
–
lik
e
to
o
l
(
ca
n
ti
n
g
)
,
s
ta
m
p
in
g
b
lo
ck
,
o
r
b
r
u
s
h
.
B
ee
s
w
ax
(
ma
la
m
)
p
atter
n
s
ar
e
ap
p
lied
to
f
ab
r
ic
d
u
r
in
g
an
i
m
m
er
s
io
n
p
r
o
ce
s
s
.
T
h
e
p
ar
ts
co
v
er
ed
in
w
ax
r
e
s
is
t
t
h
e
d
y
e
an
d
r
e
m
ai
n
in
th
e
o
r
i
g
in
al
co
l
o
u
r
.
T
h
e
w
ax
i
s
r
e
m
o
v
ed
af
ter
th
e
d
y
ei
n
g
p
r
o
ce
s
s
.
T
h
e
o
r
n
am
e
n
tatio
n
s
o
n
a
p
iece
o
f
b
atik
ar
r
an
g
ed
s
o
as
a
to
t
al
co
m
p
o
s
it
io
n
ca
lled
m
o
ti
f
[
3
]
.
T
h
e
w
a
x
i
n
g
a
n
d
d
y
ei
n
g
p
r
o
ce
s
s
ca
n
b
e
r
ep
ea
ted
to
cr
ea
te
m
o
r
e
m
o
tifs a
n
d
d
e
s
ig
n
s
.
B
atik
p
la
y
s
a
m
ea
n
i
n
g
f
u
l
r
o
le
in
I
n
d
o
n
e
s
ia
’
s
c
u
lt
u
r
al
h
er
ita
g
e.
B
atik
g
i
v
es
s
i
g
n
i
f
ica
n
t
co
n
t
r
ib
u
tio
n
to
g
lo
b
al
ar
t
as
w
ell.
T
h
e
p
h
ilo
s
o
p
h
y
o
f
b
ati
k
is
r
ela
ted
to
th
e
cu
lt
u
r
al
id
en
ti
t
y
o
f
t
h
e
I
n
d
o
n
e
s
ian
p
eo
p
le
an
d
t
h
e
s
y
m
b
o
lic
m
ea
n
i
n
g
s
t
h
at
e
x
p
r
es
s
t
h
eir
cr
ea
tiv
it
y
a
n
d
s
p
ir
it
u
al
it
y
.
W
h
ile
v
ar
io
u
s
o
th
er
co
u
n
t
r
ies
also
h
a
v
e
t
h
eir
o
w
n
v
ar
ieties
o
f
b
atik
,
UNE
S
C
O
h
as
d
esi
g
n
ated
I
n
d
o
n
es
ia
n
b
ati
k
a
s
a
Ma
s
ter
p
iece
o
f
Or
al
an
d
I
n
ta
n
g
ib
le
Her
itag
e
o
f
H
u
m
a
n
it
y
in
2
0
0
9
.
I
n
d
o
n
esia’
s
B
atik
p
att
er
n
s
ar
e
class
if
ie
d
as
g
eo
m
etr
ic
p
atter
n
s
an
d
n
o
n
-
g
eo
m
etr
ic
p
atter
n
s
.
T
h
er
e
ar
e
s
ev
er
al
g
r
o
u
p
s
o
f
g
eo
m
etr
ic
p
atter
n
s
n
a
m
el
y
ce
p
lo
k,
ka
w
u
n
g
,
p
a
r
a
n
g
,
leren
g
,
an
d
n
itik
,
w
h
ile
v
ar
iet
y
d
esig
n
s
o
f
n
o
n
-
g
eo
m
etr
ic
p
atter
n
s
co
n
s
is
t
o
f
f
o
u
r
class
e
s
n
a
m
el
y
Lu
n
g
-
lu
n
g
a
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
Textu
r
e
F
u
s
io
n
fo
r
B
a
tik
Mo
tif R
etri
ev
a
l S
ystem
(
I
d
a
N
u
r
h
a
id
a
)
3175
an
d
S
eme
n
,
P
a
g
e
r
s
a
r
i,
Ta
p
la
k
Meja
an
d
W
a
ya
n
g
.
E
a
c
h
g
r
o
u
p
h
a
v
e
its
o
w
n
v
ar
iat
io
n
s
an
d
d
is
ti
n
cti
v
e
f
ea
t
u
r
es.
Sa
m
p
le
s
o
f
r
ea
l i
m
a
g
es b
ati
k
m
o
ti
f
s
f
r
o
m
g
eo
m
etr
ic
p
atter
n
s
ca
n
b
e
s
ee
n
in
Fig
u
r
e
1
.
T
h
e
cr
af
t
o
f
b
atik
m
u
s
t
b
e
co
n
s
er
v
ed
an
d
p
r
eser
v
ed
.
T
o
th
at
en
d
,
o
u
r
m
ai
n
g
o
al
is
t
h
e
d
o
cu
m
en
tatio
n
o
f
b
atik
i
n
a
lar
g
e
d
ig
ita
l
r
ep
o
s
ito
r
y
s
y
s
te
m
.
B
atik
Mo
ti
f
R
etr
iev
al
S
y
s
te
m
(
B
MRS
)
w
il
l
r
etr
iev
e
i
m
a
g
e
s
w
h
ic
h
ar
e
v
is
u
all
y
s
i
m
i
lar
to
th
e
q
u
er
y
i
m
a
g
e
in
a
d
atab
ase.
B
MRS
s
y
s
te
m
s
p
er
f
o
r
m
f
ea
t
u
r
e
ex
tr
ac
tio
n
a
s
a
p
r
ep
r
o
ce
s
s
in
g
s
tep
.
T
o
r
etr
ie
v
e
s
p
ec
if
ic
b
atik
m
o
ti
f
s
f
r
o
m
th
is
r
ep
o
s
ito
r
y
,
a
u
s
er
ca
n
p
lace
a
q
u
er
y
to
th
e
s
y
s
te
m
,
w
h
ich
th
e
n
a
u
to
m
a
ticall
y
p
r
o
v
id
es
r
elate
d
i
n
f
o
r
m
atio
n
.
P
r
o
ce
s
s
i
n
g
o
f
q
u
er
y
i
m
a
g
e
i
n
v
o
l
v
e
s
ex
tr
ac
tio
n
o
f
i
m
a
g
e
f
ea
tu
r
e
s
a
n
d
s
ea
r
ch
i
n
th
e
v
i
s
u
al
f
ea
t
u
r
e
s
p
ac
e
f
o
r
s
i
m
ilar
i
m
a
g
es.
T
h
e
To
p
-
N
m
o
s
t si
m
i
lar
b
atik
m
o
ti
f
i
m
a
g
es
ar
e
r
etr
iev
ed
an
d
p
r
esen
ted
to
th
e
u
s
er
.
B
MRS
s
y
s
te
m
s
p
er
f
o
r
m
f
ea
t
u
r
e
ex
tr
ac
tio
n
as
a
p
r
ep
r
o
ce
s
s
in
g
s
tep
.
O
n
ce
o
b
tain
ed
,
i
m
ag
e
f
ea
tu
r
e
s
ac
t
a
s
i
n
p
u
t
s
to
s
u
b
s
eq
u
en
t
i
m
a
g
e
an
al
y
s
i
s
ta
s
k
s
a
s
s
i
m
ilar
it
y
est
i
m
a
tio
n
.
Ho
w
e
v
er
,
co
n
te
n
t
-
b
ased
r
etr
iev
al
s
y
s
te
m
s
h
a
v
e
li
m
itatio
n
s
b
et
w
ee
n
t
h
e
h
u
m
a
n
r
ep
r
esen
tatio
n
o
f
a
n
i
m
ag
e
a
n
d
th
e
lo
w
lev
e
l
f
ea
t
u
r
es
s
to
r
ed
in
th
e
d
atab
ase,
o
f
ten
ca
lled
th
e
Se
m
an
t
ic
Gap
[4
-
6]
.
T
h
e
r
e
d
u
ctio
n
o
f
t
h
e
s
e
m
an
t
ic
g
ap
an
d
h
o
w
to
ac
h
iev
e
ac
c
u
r
ate
r
etr
iev
al
r
es
u
l
ts
ar
e
a
ch
allen
g
i
n
g
p
r
o
b
lem
in
C
o
n
te
n
t
B
ased
I
m
ag
e
R
etr
iev
a
l
(
C
B
I
R
)
s
y
s
te
m
s
.
T
h
e
k
e
y
p
r
o
b
le
m
in
a
b
atik
m
o
ti
f
r
etr
iev
a
l
s
y
s
te
m
is
t
h
e
n
at
u
r
e
o
f
o
b
j
ec
t
in
b
ati
k
m
o
ti
f
,
w
h
ic
h
co
n
s
is
t
s
o
f
g
eo
m
etr
ic
o
r
n
a
m
e
n
ts
w
it
h
cir
cle
s
h
ap
e
li
k
e
f
lo
w
er
o
r
s
tar
,
ellip
s
e,
an
d
s
er
ies
o
f
s
u
c
h
o
b
j
ec
ts
in
th
e
f
o
r
m
o
f
d
o
ts
a
n
d
s
m
a
ll
li
n
es
i
n
p
ar
allel
p
o
s
iti
o
n
.
T
h
e
o
r
n
a
m
e
n
ts
ar
e
p
lace
d
i
n
d
ec
o
r
ativ
e
ar
ea
w
it
h
v
ar
iatio
n
s
i
n
p
o
s
itio
n
,
s
ca
le,
a
n
d
r
o
tatio
n
.
I
t
is
o
u
r
h
y
p
o
th
e
s
is
th
at
th
e
s
e
f
ea
t
u
r
es
co
m
p
le
m
e
n
t
ea
c
h
o
t
h
er
i
n
r
ep
r
ese
n
ti
n
g
th
e
b
atik
m
o
tif
p
r
o
p
er
ties
in
an
i
m
ag
e
f
o
r
m
at
.
E
x
p
er
i
m
e
n
tal
r
es
u
lt
s
h
av
e
s
h
o
w
n
th
at
t
h
e
f
u
s
io
n
o
f
f
ea
tu
r
es
e
x
tr
ac
ted
b
y
d
if
f
er
en
t
m
et
h
o
d
s
i
n
cr
ea
s
es
r
ec
o
g
n
itio
n
r
ate
in
t
h
e
b
atik
m
o
tif
r
etr
iev
al
s
y
s
te
m
.
(
a)
T
r
u
n
tu
m
So
g
an
o
f
C
ep
lo
k
class
(
b
)
Kaw
u
n
g
P
icis
o
f
Ka
w
u
n
g
cla
s
s
(
c)
L
er
en
g
Ud
a
n
L
ir
is
o
f
L
er
e
n
g
class
(
d
)
P
ar
an
g
R
u
s
a
k
o
f
P
ar
an
g
class
(
e)
Nitik
R
a
n
d
u
o
f
Nitik
cla
s
s
Fig
u
r
e
1
.
B
atik
m
o
ti
f
f
r
o
m
g
e
o
m
e
tr
ic
p
atter
n
s
.
C
ep
lo
k
m
o
t
if
h
as r
ep
etitiv
e
g
eo
m
e
tr
ic
o
r
n
am
en
ts
b
ased
o
n
cir
cu
lar
s
h
ap
es,
s
tar
s
,
s
q
u
ar
e
s
,
cu
b
es a
n
d
o
th
er
g
eo
m
etr
ic
li
n
es.
K
a
w
u
n
g
m
o
tif
i
s
th
e
o
ld
est
k
n
o
w
n
b
atik
p
atter
n
.
K
a
w
u
n
g
m
o
tif
co
n
s
is
t
s
o
f
th
e
r
ep
etitio
n
o
f
cir
cles o
r
ellip
tical
s
h
ap
es t
h
at
t
h
at
to
u
c
h
o
r
o
v
er
lap
.
Ler
en
g
m
o
tif
h
as d
ia
g
o
n
al
r
o
w
s
o
f
p
atter
n
s
in
b
et
w
ee
n
f
i
lle
d
w
it
h
s
m
all
p
atter
n
s
.
P
a
r
a
n
g
m
o
ti
f
co
n
s
i
s
ts
o
f
s
o
m
e
p
a
r
allel
lin
e
s
i
n
d
iag
o
n
al
f
o
r
m
f
illed
w
it
h
s
m
all
o
r
n
a
m
e
n
ts
.
N
itik
m
o
t
if
i
s
cr
ea
ted
w
it
h
s
m
all
d
o
ts
a
n
d
d
ash
es i
m
it
at
in
g
t
h
e
o
r
ig
i
n
al
w
o
v
en
f
ab
r
ic
(
R
ef
er
e
n
ce
: B
ati
k
Mu
s
eu
m
J
ak
ar
ta,
I
n
d
o
n
e
s
ia
an
d
Sak
u
n
d
r
ia
C
o
llectio
n
)
.
T
h
is
p
ap
er
b
u
ild
s
o
n
ea
r
lier
w
o
r
k
[
7
]
in
w
h
ic
h
a
co
m
p
ar
is
o
n
w
as
co
n
d
u
cted
b
ased
o
n
a
s
in
g
le
tex
t
u
r
e
f
ea
t
u
r
e
in
t
h
e
d
o
m
ai
n
o
f
a
b
atik
i
m
a
g
e
d
atab
ase.
I
t
s
tu
d
ied
b
atik
m
o
ti
f
id
en
ti
f
ic
atio
n
in
co
m
p
ar
i
n
g
w
it
h
o
th
er
lab
eled
b
atik
m
o
ti
f
in
th
e
d
atab
ase.
T
h
e
h
ig
h
est
p
er
f
o
r
m
a
n
ce
o
f
clas
s
i
f
icatio
n
ac
c
u
r
ac
y
ac
h
iev
ed
n
ea
r
l
y
8
0
%
u
s
in
g
Gr
e
y
L
e
v
e
l
C
o
-
o
cc
u
r
r
en
ce
Ma
tr
i
x
f
ea
t
u
r
es.
Sh
ap
e
s
i
m
ilar
it
y
an
d
tex
t
u
r
e
ch
ar
ac
ter
i
s
tics
h
av
e
also
b
ee
n
e
m
p
lo
y
ed
f
o
r
b
atik
i
m
ag
e
r
etr
iev
al
[
8
]
.
T
h
is
r
esear
ch
u
tili
ze
d
ed
g
e
d
etec
tio
n
an
d
s
h
ap
e
in
v
ar
ia
n
t
m
o
m
e
n
t
as
a
f
ea
t
u
r
e
.
A
th
r
es
h
o
ld
in
g
ap
p
r
o
ac
h
is
u
s
ed
to
r
etr
iev
e
th
e
i
m
a
g
es
b
a
s
ed
o
n
th
e
v
al
u
e
o
f
th
e
h
i
g
h
e
s
t
-
g
r
ad
e
r
ep
r
esen
tati
o
n
o
n
ea
c
h
i
m
a
g
e
q
u
er
y
.
T
h
e
b
est
p
er
f
o
r
m
a
n
ce
ac
h
ie
v
ed
a
p
r
ec
is
io
n
a
n
d
r
ec
all
o
f
7
0
% a
n
d
7
5
% r
esp
ec
ti
v
el
y
.
I
n
a
n
o
th
er
s
t
u
d
y
o
f
b
atik
i
m
a
g
e
r
etr
iev
a
l,
s
i
m
ilar
v
al
u
es o
f
7
4
% a
n
d
8
9
%
w
er
e
o
b
tain
ed
[
9
]
.
T
h
is
s
tu
d
y
ap
p
lied
ed
g
e
f
ea
tu
r
e
o
r
ien
tatio
n
co
m
b
i
n
ed
w
it
h
m
icr
o
s
tr
u
c
t
u
r
e
d
escr
ip
to
r
f
o
r
en
h
a
n
ci
n
g
r
etr
iev
a
l
p
er
f
o
r
m
a
n
ce
.
R
a
n
g
k
u
ti,
et
al
[
1
0
]
r
ep
o
r
ted
u
s
i
n
g
C
an
n
y
ed
g
e
d
etec
tio
n
to
a
n
in
p
u
t
i
m
a
g
e,
w
a
v
elet
s
as
tex
t
u
r
e
f
ea
t
u
r
es
an
d
in
v
ar
ia
n
t
m
o
m
en
t
as
s
h
ap
e
f
ea
t
u
r
es
m
et
h
o
d
.
T
h
e
p
er
f
o
r
m
a
n
ce
r
es
u
lts
ac
h
iev
ed
o
p
ti
m
a
l p
r
ec
is
io
n
o
f
av
er
ag
e
9
0
%
-
92
%.
I
n
th
is
ar
ticle,
w
e
f
ir
s
t
s
h
o
w
t
h
at
co
m
b
in
in
g
m
u
lt
ip
le
tex
tu
r
e
d
escr
ip
tio
n
m
eth
o
d
s
s
ig
n
if
ican
t
l
y
i
m
p
r
o
v
es
t
h
e
p
er
f
o
r
m
an
ce
co
m
p
ar
ed
to
u
s
i
n
g
t
h
e
s
i
n
g
le
tex
tu
r
e
m
et
h
o
d
alo
n
e
f
o
r
B
atik
Mo
tif
R
etr
ie
v
al.
T
h
e
id
ea
is
th
at
t
h
e
th
eo
r
etica
l
p
r
o
p
er
ties
o
f
ea
ch
f
ea
tu
r
e
ar
e
ex
a
m
i
n
ed
in
o
r
d
er
to
s
elec
t
th
e
r
o
b
u
s
t
an
d
r
eliab
le
f
ea
t
u
r
es
f
o
r
f
u
s
io
n
.
W
e
p
ar
ticu
lar
l
y
f
o
cu
s
o
n
d
is
cu
s
s
i
n
g
a
n
d
ev
alu
a
tin
g
h
o
w
t
h
e
s
in
g
le
f
ea
tu
r
e
an
d
f
u
s
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.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
3
1
7
4
–
3
187
3176
f
ea
t
u
r
es
p
er
f
o
r
m
in
th
e
r
etr
ie
v
al
o
f
b
atik
m
o
ti
f
.
Fi
n
all
y
,
w
e
p
r
o
v
id
e
a
co
m
p
r
eh
e
n
s
i
v
e
e
v
alu
atio
n
o
f
b
o
th
m
u
ltip
le
f
ea
t
u
r
es
a
n
d
s
in
g
le
f
ea
tu
r
e
f
o
r
th
e
tas
k
o
f
i
m
a
g
e
r
etr
iev
al
b
ati
k
m
o
ti
f
.
T
h
is
w
ill
co
n
tr
ib
u
te
to
w
ar
d
s
th
e
d
ev
elo
p
m
e
n
t
o
f
a
d
ig
ita
l
i
m
ag
e
r
ep
o
s
ito
r
y
s
y
s
te
m
o
f
b
atik
m
o
ti
f
s
f
o
r
cu
ltu
r
al
h
e
r
itag
e
p
r
eser
v
atio
n
s
o
lu
tio
n
s
.
I
n
th
i
s
p
ap
e
r
,
w
e
co
n
d
u
ct
s
y
s
t
e
m
atica
l
a
n
al
y
s
i
s
to
i
n
v
e
s
ti
g
at
e
th
e
i
n
f
lu
e
n
ce
o
f
i
m
ag
e
tex
t
u
r
e
f
ea
tu
r
e
f
u
s
io
n
to
t
h
e
p
er
f
o
r
m
a
n
ce
o
n
a
b
atik
m
o
ti
f
r
etr
iev
al
s
y
s
te
m
.
T
h
e
p
ap
e
r
is
o
r
g
an
ized
as
f
o
l
lo
w
s
.
I
n
Sectio
n
2
,
th
e
alg
o
r
it
h
m
o
f
f
ea
t
u
r
e
ex
tr
ac
tio
n
m
et
h
o
d
s
,
f
u
s
io
n
m
et
h
o
d
,
an
d
s
i
m
ilar
it
y
d
is
tan
ce
ar
e
ela
b
o
r
ated
.
I
n
Sectio
n
3
,
th
e
p
er
f
o
r
m
an
ce
o
f
f
ea
t
u
r
e
f
u
s
io
n
u
s
i
n
g
s
i
m
ilar
it
y
d
i
s
tan
ce
i
s
test
ed
a
n
d
co
m
p
ar
ed
in
a
s
er
ies
o
f
ex
p
er
i
m
e
n
ts
.
Fi
n
all
y
,
t
h
e
r
es
u
l
t
o
f
C
B
I
R
e
x
p
er
i
m
en
t
s
ar
e
d
i
s
cu
s
s
ed
in
Sect
io
n
4
an
d
co
n
cl
u
s
io
n
s
ar
e
g
iv
e
n
i
n
Sectio
n
5
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
F
ea
t
ure
E
x
t
ra
ct
io
n M
e
t
ho
ds
a
nd
P
er
f
o
r
m
a
nce
E
v
a
lua
t
io
n
I
n
th
is
p
ap
er
,
w
e
u
s
e
v
ar
io
u
s
tex
tu
r
e
f
ea
t
u
r
es,
i.e
.
Gab
o
r
f
ilter
s
,
L
o
g
Gab
o
r
f
il
ter
s
,
G
r
ey
L
e
v
el
Co
-
o
cc
u
r
r
en
ce
Ma
tr
ices,
a
n
d
L
o
ca
l
B
i
n
ar
y
P
atter
n
s
;
a
n
d
an
al
y
ze
th
e
ir
co
m
b
i
n
atio
n
in
b
atik
m
o
ti
f
i
m
ag
e
r
etr
iev
al
ap
p
licatio
n
s
.
T
h
e
s
p
atial
lo
ca
lit
y
,
o
r
ien
tatio
n
s
ele
ctiv
it
y
,
an
d
f
r
eq
u
e
n
c
y
ar
e
ca
p
tu
r
ed
as
th
e
m
ai
n
ch
ar
ac
ter
is
tic
s
f
o
r
r
ep
r
esen
tati
o
n
o
f
s
alie
n
t v
is
u
al
p
r
o
p
er
ties
[
1
1
-
12]
.
2
.
1
.
1
.
G
a
bo
r
F
ilte
r
Gab
o
r
f
ilter
s
ar
e
u
s
ed
to
m
o
d
el
th
e
s
p
atia
l
s
u
m
m
atio
n
p
r
o
p
er
ties
o
f
s
i
m
p
le
ce
ll
in
th
e
v
is
u
al
co
r
tex
[
1
3
-
14]
.
Fil
ter
in
g
o
p
er
atio
n
is
co
n
d
u
cted
b
y
i
m
ag
e
co
n
v
o
lu
tio
n
o
f
a
n
o
r
ig
i
n
al
i
m
ag
e
w
it
h
a
G
ab
o
r
f
ilter
to
g
e
n
er
ate
a
n
e
w
i
m
ag
e
.
T
h
e
n
u
m
b
er
o
f
n
e
w
i
m
a
g
es
is
co
r
r
elate
d
to
th
e
n
u
m
b
er
o
f
f
ilter
s
u
s
ed
.
A
2
D
Gau
s
s
ia
n
en
v
elo
p
is
m
o
d
u
late
d
a
2
D
Gab
o
r
f
ilter
in
co
m
p
l
ex
s
i
n
u
s
o
id
al
w
a
v
e.
T
h
e
2
D
Gab
o
r
f
ilter
s
ca
n
b
e
ca
teg
o
r
ized
in
to
t
w
o
co
m
p
o
n
en
t
s
:
a
r
ea
l
p
ar
t
as
s
y
m
m
etr
ic
co
m
p
o
n
en
t
an
d
a
n
i
m
a
g
in
ar
y
p
ar
t
as
t
h
e
as
y
m
m
etr
ic
co
m
p
o
n
e
n
t.
T
h
e
2
D
Gab
o
r
f
u
n
ct
io
n
ca
n
b
e
m
a
th
e
m
atica
ll
y
f
o
r
m
u
lated
as
[
1
4
]
:
(
)
(
)
(
1
)
w
h
er
e
:
I
n
eq
u
atio
n
(
1
)
,
f
is
t
h
e
f
r
eq
u
e
n
c
y
o
f
s
i
n
u
s
o
id
al
w
a
v
e,
r
ep
r
esen
t
s
th
e
an
ti
-
clo
c
k
w
is
e
r
o
tat
io
n
o
f
t
h
e
Gau
s
s
ia
n
en
v
elo
p
e
an
d
th
e
s
i
n
u
s
o
id
,
d
en
o
tes
th
e
s
m
o
o
th
in
g
p
ar
a
m
e
ter
s
o
f
t
h
e
Gau
s
s
i
an
en
v
elo
p
e
,
an
d
in
d
icate
s
t
h
e
o
r
th
o
g
o
n
al
to
th
e
d
ir
ec
tio
n
o
f
th
e
w
a
v
e
,
r
esp
ec
tiv
el
y
.
T
h
e
to
tal
n
u
m
b
er
o
f
f
r
eq
u
en
c
i
es
an
d
t
h
e
to
tal
n
u
m
b
er
o
f
o
r
ien
tatio
n
s
o
f
th
e
Gab
o
r
f
ilter
s
ar
e
d
eter
m
in
ed
to
d
es
ig
n
Gab
o
r
f
ilter
b
an
k
.
T
h
e
co
m
b
i
n
atio
n
o
f
f
r
eq
u
e
n
c
y
an
d
o
r
ien
tatio
n
g
en
er
ates
t
h
e
Gab
o
r
f
ilter
b
an
k
[
1
3
]
.
R
esear
ch
b
y
C
lau
s
i
[
1
4
]
s
elec
ted
h
ig
h
est
f
r
eq
u
en
c
y
√
,
f
o
u
r
n
u
m
b
er
o
f
f
r
eq
u
en
c
y
(
2
2
.
6
3
,
1
1
.
3
1
,
5
.
6
6
,
an
d
2
.
8
3
p
ix
el
p
er
c
y
cle)
an
d
s
i
x
o
r
ien
t
atio
n
s
(0
0
,
30
0
,
60
0
,
90
0
,
1
2
0
0
,
an
d
1
5
0
0
)
to
f
ilter
ea
ch
te
s
t i
m
ag
e.
T
h
ese
f
i
lter
s
ar
e
p
u
r
p
o
s
ed
to
w
ell
-
lo
ca
lized
m
ea
s
u
r
e
o
f
th
e
lo
ca
l
i
n
f
o
r
m
atio
n
.
2
.
1
.
2
.
G
re
y
L
ev
el
Co
-
o
cc
urre
nce
M
a
t
rice
s
Gr
e
y
L
e
v
el
C
o
-
o
cc
u
r
r
en
ce
M
atr
ix
(
GL
C
M)
is
a
co
m
m
o
n
m
eth
o
d
u
s
ed
f
o
r
an
al
y
zi
n
g
i
m
a
g
e
tex
tu
r
e
s
.
T
h
e
b
asic
id
ea
f
r
o
m
t
h
i
s
m
e
th
o
d
is
to
ex
tr
ac
t
h
o
m
o
g
e
n
e
o
u
s
ch
ar
ac
ter
i
s
tics
f
r
o
m
i
m
a
g
e
tex
tu
r
e.
G
L
C
M
p
r
o
d
u
ce
s
f
ea
t
u
r
es
w
h
ic
h
d
esc
r
ib
e
w
ell
t
h
e
r
elatio
n
s
h
ip
o
f
a
d
j
ac
en
cy
a
m
o
n
g
p
ix
el
s
i
n
a
n
tex
t
u
r
e
i
m
ag
e
[
1
5
]
.
T
h
e
s
ec
o
n
d
o
r
d
er
s
tatis
tics
ar
e
ac
cu
m
u
lated
i
n
to
a
s
e
t
o
f
2
D
m
atr
ice
s
,
(
)
ea
ch
o
f
w
h
ich
m
e
asu
r
es
t
h
e
s
p
atial
d
ep
en
d
en
c
y
o
f
t
w
o
g
r
a
y
lev
e
ls
,
i
an
d
j
,
g
iv
e
n
a
d
is
p
l
ac
e
m
en
t
v
ec
to
r
(
)
(
)
[
1
3
]
.
T
h
e
n
u
m
b
er
o
f
o
cc
u
r
r
en
ce
s
(
f
r
eq
u
en
cie
s
)
o
f
i
a
n
d
j
,
s
ep
ar
ated
b
y
d
i
s
tan
ce
d
,
co
n
tr
ib
u
tes
t
h
e
(
i,j
)
en
tr
y
in
t
h
e
co
-
o
cc
u
r
r
en
ce
m
atr
ix
(
)
.
A
co
-
o
cc
u
r
r
en
ce
m
a
tr
ix
i
s
g
i
v
e
n
as:
(
)
‖
,
(
(
)
(
)
)
(
)
(
)
-
‖
(
)
(
)
(
)
(
)
(
2
)
w
h
er
e:
(
)
= N
u
m
b
er
o
f
o
cc
u
r
r
en
ce
s
o
f
th
e
p
air
o
f
g
r
e
y
le
v
els
i
a
n
d
j
(
)
an
d
(
)
=
co
o
r
d
in
ates o
f
p
ix
el
s
i
n
t
w
o
p
o
s
itio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
Textu
r
e
F
u
s
io
n
fo
r
B
a
tik
Mo
tif R
etri
ev
a
l S
ystem
(
I
d
a
N
u
r
h
a
id
a
)
3177
=
d
is
p
lace
m
en
t
v
ec
to
r
=
d
is
tan
ce
=
o
r
ien
tatio
n
‖
‖
=
C
ar
d
in
alit
y
o
f
a
s
et
T
ex
tu
r
e
f
ea
t
u
r
es,
s
u
c
h
a
s
h
o
m
o
g
e
n
eit
y
,
e
n
er
g
y
,
en
tr
o
p
y
,
c
o
n
tr
ast,
a
n
d
co
r
r
elatio
n
,
ar
e
t
h
en
d
er
iv
ed
f
r
o
m
t
h
e
co
-
o
cc
u
r
r
en
ce
m
atr
i
x
.
a.
C
o
n
tr
ast :
∑
∑
(
)
(
)
(
3
)
w
h
er
e:
k
=
th
e
n
u
m
b
er
o
f
r
o
w
s
o
r
co
lu
m
n
s
T
h
e
p
r
o
b
a
b
ilit
y
o
f
t
w
o
p
ix
els
(
)
in
d
icate
d
th
e
s
ep
ar
atio
n
o
f
t
w
o
p
ix
els
w
it
h
d
if
f
er
e
n
t
g
r
e
y
le
v
el
i
an
d
j
[
6
]
.
C
o
n
tr
ast
m
ea
s
u
r
es lo
ca
l i
n
ten
s
i
t
y
o
f
i
n
v
ar
ian
ce
.
b.
Ho
m
o
g
e
n
eit
y
o
r
An
g
u
lar
Seco
n
d
Mo
m
en
t (
ASM)
:
∑
∑
(
)
(
4
)
T
h
e
h
o
m
o
g
e
n
eit
y
o
f
an
i
m
a
g
e
is
m
ea
s
u
r
ed
b
y
u
s
in
g
An
g
u
lar
Seco
n
d
Mo
m
en
t.
T
h
e
s
u
m
o
f
s
q
u
ar
es
w
ill
b
e
h
ig
h
i
f
h
o
m
o
g
e
n
eo
u
s
s
ce
n
e
co
n
tai
n
s
o
n
l
y
a
f
e
w
g
r
a
y
lev
el
s
b
u
t r
elati
v
el
y
h
i
g
h
v
al
u
es o
f
(
)
.
c.
I
n
v
er
s
e
Di
f
f
er
en
ce
Mo
m
en
t (
I
DM
)
:
∑
∑
(
)
(
)
(
5
)
T
h
e
h
o
m
o
g
e
n
eit
y
o
f
t
h
e
i
m
a
g
e
a
f
f
ec
ted
th
e
I
DM
’
s
v
alu
e.
I
DM
v
al
u
e
w
i
ll
b
e
lo
w
i
f
t
h
e
i
m
a
g
e
is
n
o
t
h
o
m
o
g
en
y
.
Ot
h
er
w
is
e,
f
o
r
h
o
m
o
g
e
n
eo
u
s
i
m
a
g
e,
t
h
e
I
DM
v
alu
e
is
h
i
g
h
.
C
o
r
r
elatio
n
C
o
r
r
elatio
n
is
a
m
ea
s
u
r
e
o
f
g
r
a
y
le
v
el
lin
ea
r
d
ep
en
d
en
ce
b
et
w
ee
n
th
e
p
ix
el
s
at
th
e
s
p
e
cif
ied
p
o
s
itio
n
s
r
elativ
e
to
ea
ch
o
th
er
.
∑
∑
(
)
(
)
(
)
(
6
)
w
h
er
e
ar
e
m
ea
n
v
al
u
e
o
f
r
o
w
i
an
d
co
l
u
m
n
j
a
n
d
ar
e
s
tan
d
a
r
d
d
ev
iatio
n
o
f
r
o
w
i
a
n
d
co
lu
m
n
j
.
2
.
1
.
3
.
L
o
g
G
a
bo
r
F
ilte
r
Field
p
r
o
p
o
s
ed
L
o
g
-
Gab
o
r
f
il
ter
s
as
a
m
o
d
i
f
icatio
n
to
t
h
e
b
asic
Gab
o
r
f
u
n
ct
io
n
[
1
6
]
(
Fi
eld
1
9
8
7
)
.
T
h
e
s
in
g
u
lar
it
y
o
f
th
e
lo
g
f
u
n
ct
io
n
L
o
g
Gab
o
r
f
ilter
s
b
asicall
y
ar
e
d
ef
i
n
ed
in
th
e
f
r
e
q
u
en
c
y
d
o
m
ai
n
as
Gau
s
s
ia
n
f
u
n
ctio
n
s
t
h
at
s
h
i
f
t
f
r
o
m
t
h
e
o
r
ig
in
[
1
7
]
.
Gab
o
r
f
ilter
s
p
r
esen
t
a
li
m
itatio
n
in
th
e
b
an
d
w
id
th
w
h
er
e
o
n
l
y
b
an
d
w
id
th
o
f
1
o
ctav
e
m
ax
i
m
u
m
co
u
ld
b
e
d
esi
g
n
ed
[
1
8
-
20]
.
L
o
g
Gab
o
r
co
n
s
is
ts
o
f
a
lo
g
ar
ith
m
i
c
tr
an
s
f
o
r
m
a
tio
n
in
th
e
Gab
o
r
d
o
m
a
in
w
h
ic
h
eli
m
i
n
ate
s
th
e
D
C
-
co
m
p
o
n
en
t
allo
ca
ted
in
m
e
d
iu
m
a
n
d
h
ig
h
-
p
ass
f
ilter
s
.
T
h
e
f
r
eq
u
e
n
c
y
r
esp
o
n
s
e
is
a
Ga
u
s
s
ian
o
n
a
lo
g
f
r
eq
u
en
c
y
a
x
i
s
.
T
h
e
co
m
p
ar
is
o
n
b
et
w
ee
n
Gab
o
r
an
d
L
o
g
Gab
o
r
f
u
n
ct
io
n
s
ca
n
b
e
s
e
en
in
Fig
u
r
e
2
.
(
a)
Gab
o
r
F
u
n
ctio
n
(
b
)
L
o
g
Gab
o
r
F
u
n
ctio
n
Fig
u
r
e
2
.
C
o
m
p
ar
is
o
n
o
f
Gab
o
r
an
d
L
o
g
Gab
o
r
Fu
n
ctio
n
s
[
1
6
]
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.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
3
1
7
4
–
3
187
3178
T
h
e
f
r
eq
u
en
c
y
r
e
s
p
o
n
s
e
o
f
L
o
g
Gab
o
r
f
ilter
ca
n
b
e
d
ef
i
n
ed
as
:
(
)
(
*
(
)
+
*
(
)
+
)
(
7
)
w
h
er
e
is
a
ce
n
ter
f
r
eq
u
en
c
y
o
f
a
f
il
ter
an
d
is
a
s
ca
li
n
g
f
ac
to
r
o
f
a
r
ad
ian
b
an
d
w
id
t
h
[
1
6
]
.
2
.
1
.
4
.
Lo
ca
l B
ina
ry
P
a
t
t
er
n (
L
B
P
)
T
h
e
L
B
P
o
p
er
ato
r
is
o
n
e
o
f
t
h
e
b
es
t
p
er
f
o
r
m
i
n
g
te
x
tu
r
e
d
escr
ip
to
r
s
an
d
it
h
a
s
b
ee
n
w
id
el
y
u
s
ed
i
n
v
ar
io
u
s
ap
p
licatio
n
s
.
T
h
e
co
m
p
lex
i
t
y
ti
m
e
w
ill
b
e
m
i
n
i
m
i
ze
d
w
h
eth
er
p
r
ep
r
o
ce
s
s
in
g
a
n
d
L
B
P
ar
e
ap
p
ly
in
g
r
esp
ec
tiv
el
y
[
2
1
]
.
T
h
e
L
o
ca
l
B
in
ar
y
P
atter
n
(
L
B
P
)
o
p
er
a
to
r
u
tili
ze
s
th
e
ce
n
ter
v
al
u
e
as
a
r
ef
er
en
ce
in
a
3
×3
p
ix
el
n
eig
h
b
o
r
h
o
o
d
[
2
2
]
.
T
h
e
th
r
esh
o
ld
v
a
lu
e
i
s
f
r
o
m
t
h
e
ce
n
ter
p
ix
el
w
h
ile
t
h
e
p
i
x
el
v
alu
e
o
f
a
n
eig
h
b
o
r
is
m
ar
k
ed
as
“
0
”
w
h
e
n
it i
s
b
el
o
w
t
h
e
t
h
r
es
h
o
ld
an
d
“
1
”
o
th
e
r
w
i
s
e.
A
b
in
ar
y
n
u
m
b
er
is
f
o
r
m
ed
to
ch
ar
ac
ter
ize
th
e
lo
ca
l
tex
t
u
r
e
(
s
ee
Fig
u
r
e
3
)
.
T
h
en
,
s
u
b
tr
ac
tin
g
th
e
av
er
ag
e
g
r
e
y
le
v
els
b
elo
w
t
h
e
ce
n
ter
p
ix
el
f
r
o
m
t
h
e
g
r
e
y
le
v
el
ab
o
v
e
o
r
eq
u
al
to
th
e
ce
n
ter
p
ix
el
w
ill r
es
u
lt i
n
C
o
n
tr
ast (
C
)
.
(
a)
E
x
a
m
p
le
(
b
)
T
h
r
esh
o
ld
er
(
c)
W
eig
h
ts
(
d
)
L
B
P
C
alcu
-
latio
n
Fig
u
r
e
3
.
An
E
x
a
m
p
le
o
f
L
o
ca
l B
in
ar
y
P
atter
n
C
alc
u
latio
n
A
b
i
n
ar
y
n
u
m
b
er
is
r
ep
r
esen
te
d
as f
o
llo
w
:
(
)
∑
(
)
(
)
,
(
8
)
w
h
er
e:
=
th
e
g
r
a
y
lev
el
o
f
th
e
ce
n
ter
p
ix
el
o
f
a
lo
ca
l n
ei
g
h
b
o
r
h
o
o
d
=
th
e
g
r
a
y
lev
el
s
o
f
N
e
v
en
l
y
s
p
ac
ed
p
ix
els o
n
a
cir
cle
o
f
r
ad
iu
s
R
.
2
.
2
.
F
ea
t
ure
F
us
io
n
Featu
r
e
f
u
s
io
n
i
n
te
g
r
ates
i
n
f
o
r
m
atio
n
f
r
o
m
all
a
v
ailab
le
f
ea
tu
r
es
i
n
to
a
u
n
i
f
ied
r
ep
r
esen
ta
tio
n
[
2
3
]
.
Data
f
u
s
io
n
ca
n
b
e
co
n
d
u
cte
d
at
th
r
ee
d
is
tin
ct
lev
e
ls
,
i.e
.
f
ea
tu
r
e
le
v
el
f
u
s
io
n
,
m
atc
h
i
n
g
/
s
co
r
e
lev
el,
an
d
d
ec
i
s
io
n
lev
el
[
2
4
]
.
Featu
r
e
lev
el
f
u
s
io
n
is
p
er
f
o
r
m
ed
b
y
co
n
ca
ten
ati
n
g
th
e
f
ea
tu
r
e
s
r
esu
lti
n
g
f
r
o
m
f
ea
tu
r
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
.
I
t
i
s
m
o
r
e
co
m
p
licated
s
in
ce
a
s
e
t
o
f
f
ea
t
u
r
es
p
r
o
b
ab
ly
h
a
v
e
d
if
f
er
en
t
d
i
m
e
n
s
io
n
.
Ho
w
e
v
er
,
f
ea
t
u
r
e
lev
el
f
u
s
io
n
h
as
b
ee
n
co
n
s
id
er
ed
p
r
ef
er
ab
le
b
ec
au
s
e
f
u
s
ed
f
ea
t
u
r
es
m
a
y
co
n
tain
ad
d
itio
n
al
d
is
tin
ct
i
n
f
o
r
m
atio
n
to
i
n
d
iv
id
u
al
f
ea
t
u
r
es.
On
t
h
e
o
t
h
er
h
a
n
d
,
it
is
p
o
s
s
ib
le
to
s
elec
t
f
ea
t
u
r
es
f
o
r
el
i
m
in
at
in
g
r
ed
u
n
d
an
t i
n
f
o
r
m
atio
n
f
r
o
m
a
f
ea
t
u
r
e
s
et.
I
m
p
le
m
e
n
tatio
n
f
o
r
f
u
s
io
n
at
m
atch
i
n
g
/s
co
r
e
lev
el
i
s
m
o
s
t
f
r
eq
u
en
tl
y
u
s
ed
co
m
p
ar
ed
to
f
ea
tu
r
e
f
u
s
io
n
i
n
th
e
lo
w
er
lev
el.
F
u
s
i
o
n
at
d
ec
is
io
n
lev
el
g
a
th
er
s
i
n
f
o
r
m
atio
n
af
ter
a
d
ec
is
io
n
is
t
ak
en
b
y
a
m
a
tch
er
b
ased
o
n
it
s
d
eli
v
er
ed
i
n
p
u
t
[
2
5
]
.
T
h
e
f
in
al
d
ec
is
io
n
i
s
m
ad
e
b
y
a
m
aj
o
r
ity
v
o
te
s
c
h
e
m
e,
b
eh
av
io
r
k
n
o
w
led
g
e
s
p
a
ce
,
w
ei
g
h
ted
v
o
ti
n
g
,
a
n
d
A
N
D
r
u
le
a
n
d
OR
r
u
le.
Fu
s
io
n
at
t
h
e
h
i
g
h
er
lev
el
m
a
y
d
ec
r
ea
s
e
th
e
r
ec
o
g
n
itio
n
p
er
f
o
r
m
a
n
ce
s
i
n
ce
s
o
m
e
i
n
f
o
r
m
atio
n
w
i
ll b
e
lo
s
t in
t
h
e
co
u
r
s
e
o
f
f
u
s
io
n
p
r
o
ce
s
s
.
2
.
3
.
F
ea
t
ure
M
a
t
ching
Featu
r
e
m
atc
h
i
n
g
is
a
f
u
n
d
a
m
en
tal
p
r
o
b
le
m
in
co
m
p
u
ter
v
is
io
n
,
a
n
d
p
la
y
s
a
cr
itical
r
o
le
in
m
an
y
task
s
s
u
c
h
as
o
b
j
ec
t
r
ec
o
g
n
itio
n
an
d
lo
ca
lizatio
n
[
2
6
]
.
A
s
i
m
i
lar
it
y
m
ea
s
u
r
e
f
o
r
co
n
ten
t
-
b
as
ed
r
etr
iev
al
s
h
o
u
ld
b
e
ef
f
ic
ien
t
e
n
o
u
g
h
to
m
a
tch
s
i
m
ilar
i
m
a
g
es
a
s
w
ell
a
s
b
ein
g
ab
le
to
d
is
cr
i
m
i
n
ate
d
is
s
i
m
ila
r
o
n
es
[
6
]
.
Featu
r
e
v
ec
to
r
s
u
s
u
all
y
ex
i
s
t
in
a
v
er
y
h
ig
h
-
d
i
m
en
s
io
n
al
s
p
ac
e.
T
h
e
p
r
o
b
lem
o
f
m
atc
h
in
g
ca
n
b
e
d
ef
in
ed
as
estab
lis
h
in
g
a
m
ap
p
in
g
b
et
wee
n
f
ea
t
u
r
es
i
n
o
n
e
i
m
a
g
e
a
n
d
s
i
m
ilar
f
ea
tu
r
es
i
n
a
n
o
th
e
r
i
m
ag
e.
S
i
m
ilar
it
y
m
ea
s
u
r
e
o
n
t
h
is
r
esear
c
h
w
as
co
n
d
u
cted
u
s
in
g
a
E
u
cl
id
ian
d
is
tan
ce
f
u
n
ctio
n
.
T
h
e
E
u
cli
d
ea
n
d
is
tan
ce
i
s
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
Textu
r
e
F
u
s
io
n
fo
r
B
a
tik
Mo
tif R
etri
ev
a
l S
ystem
(
I
d
a
N
u
r
h
a
id
a
)
3179
d
is
tan
ce
f
u
n
ct
io
n
m
o
s
t
w
id
el
y
u
s
ed
to
m
ea
s
u
r
e
t
h
e
d
is
ta
n
ce
of
t
w
o
v
ec
to
r
s
.
I
f
t
h
e
t
w
o
v
e
cto
r
s
ar
e
v
ec
to
r
s
A
an
d
B
,
w
h
er
e:
[
]
[
]
(
9
)
T
h
e
E
u
clid
ian
Dis
tan
ce
i
s
d
ef
i
n
ed
as:
√
∑
(
)
(
1
0
)
2
.
4
.
P
er
f
o
r
m
a
nce
E
v
a
lua
t
io
n
P
r
ec
is
io
n
an
d
R
ec
all
ar
e
t
w
o
in
d
icato
r
s
o
f
t
h
e
co
r
r
ec
tn
es
s
r
etr
iev
al
r
es
u
lt
[2
7
-
28]
.
R
ec
all
m
ea
n
s
a
r
atio
b
et
w
ee
n
t
h
e
n
u
m
b
er
o
f
co
r
r
ec
tly
r
etr
ie
v
ed
i
m
ag
e
b
y
th
e
s
y
s
te
m
an
d
t
h
e
n
u
m
b
er
o
f
all
i
m
a
g
es
in
th
e
d
atab
ase
w
h
ich
h
av
e
t
h
e
s
a
m
e
class
w
ith
t
h
e
q
u
er
y
.
I
n
o
th
er
w
o
r
d
s
,
r
ec
all
d
escr
ib
es
th
e
s
y
s
te
m
’
s
ca
p
ab
ilit
y
to
r
etr
iev
e
r
elev
an
t
i
m
a
g
es,
w
h
er
ea
s
p
r
ec
is
io
n
is
t
h
e
n
u
m
b
er
o
f
r
elev
a
n
t
i
m
ag
e
s
f
o
u
n
d
co
m
p
ar
ed
to
th
e
n
u
m
b
er
o
f
all
i
m
a
g
es
f
o
u
n
d
in
a
q
u
er
y
p
r
o
ce
s
s
.
P
r
ec
is
io
n
d
ep
icts
th
e
ab
ilit
y
o
f
t
h
e
s
y
s
te
m
to
r
ej
ec
t
th
e
ir
r
elev
an
t
i
m
a
g
e.
A
p
er
f
ec
t
p
r
ec
is
io
n
s
co
r
e
o
f
1
.
0
m
ea
n
s
th
at
ev
er
y
r
es
u
lt
r
etr
ie
v
ed
b
y
a
s
ea
r
ch
w
a
s
r
elev
a
n
t
w
h
er
ea
s
a
p
er
f
ec
t
r
ec
all
s
co
r
e
o
f
1
.
0
m
ea
n
s
th
a
t
all
r
ele
v
an
t
i
m
a
g
es
w
er
e
r
etr
ie
v
ed
b
y
th
e
s
ea
r
ch
.
R
ec
al
l
an
d
p
r
ec
is
io
n
ar
e
in
v
er
s
el
y
r
elate
d
an
d
ca
n
b
e
d
ef
in
ed
as:
2
.
5
.
P
ro
po
s
ed
M
et
ho
d
I
n
th
i
s
p
ap
er
,
w
e
p
r
o
p
o
s
e
a
C
B
I
R
ap
p
r
o
ac
h
f
o
r
b
atik
m
o
tif
r
etr
iev
al
s
y
s
te
m
t
h
at
b
u
i
ld
s
o
n
tex
tu
r
e
f
ea
t
u
r
e
f
u
s
io
n
t
h
r
o
u
g
h
a
n
e
f
f
ec
tiv
e
s
e
lectio
n
f
ea
t
u
r
e.
F
u
r
t
h
er
m
o
r
e,
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
w
as
e
x
te
n
s
iv
el
y
ev
alu
a
ted
o
n
th
e
i
m
ag
e
d
ata
s
et
o
f
b
ati
k
.
I
t
s
h
o
u
ld
b
e
n
o
te
d
th
at
w
e
u
s
e
th
e
s
a
m
e
p
ar
a
m
eter
u
s
ed
b
y
[
7
]
f
o
r
GL
C
M
an
d
Gab
o
r
f
ea
t
u
r
e
ex
t
r
ac
ti
o
n
.
T
h
e
p
u
r
p
o
s
e
is
to
co
m
p
ar
e
o
u
r
r
esu
lt
s
w
it
h
p
r
ev
io
u
s
w
o
r
k
.
N
u
r
h
a
id
a,
et.
al
[
7
]
d
is
co
v
er
ed
th
at
u
s
in
g
s
in
g
le
GL
C
M
f
ea
t
u
r
e
w
ith
r
a
w
i
m
ag
e
i
n
p
u
t
g
a
v
e
th
e
b
est
ac
cu
r
ac
y
f
o
r
b
atik
m
o
ti
f
class
i
f
icatio
n
.
Ho
w
e
v
er
,
w
e
w
ill
u
s
e
th
e
f
ea
t
u
r
e
f
o
r
o
u
r
B
atik
Mo
tif
r
etr
iev
al
s
y
s
t
e
m
tas
k
.
Fig
u
r
e
4
d
ep
icts
th
e
f
r
a
m
e
w
o
r
k
o
f
t
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
f
o
r
b
atik
m
o
ti
f
r
etr
iev
al
s
y
s
te
m
,
w
h
ic
h
m
ai
n
l
y
in
cl
u
d
es
o
f
f
ea
t
u
r
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
,
d
ata
n
o
r
m
aliza
t
io
n
,
f
u
s
io
n
at
t
h
e
f
ea
t
u
r
e
lev
el,
a
n
d
th
e
m
ea
s
u
r
e
m
en
t
o
f
s
i
m
ilar
it
y
m
atc
h
in
g
b
et
w
ee
n
q
u
er
y
i
m
ag
es
ag
ai
n
s
t
i
m
ag
e
s
in
th
e
d
atab
ase.
T
h
e
f
r
a
m
e
w
o
r
k
o
f
f
ea
t
u
r
e
ex
tr
ac
tio
n
f
u
s
io
n
is
s
h
o
w
n
i
n
Fi
g
u
r
e
4
.
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.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
3
1
7
4
–
3
187
3180
Fig
u
r
e
4
.
Fra
m
e
w
o
r
k
o
f
f
ea
t
u
r
e
ex
tr
ac
tio
n
f
u
s
io
n
f
o
r
B
atik
Mo
tif
R
etr
ie
v
al
S
y
s
te
m
3.
E
XP
E
R
I
M
E
NT
A
L
DE
SI
G
N
T
he
d
ataset
co
n
s
is
t
s
o
f
2
1
0
te
m
p
lates
(
1
5
0
x
1
5
0
p
ix
els)
m
o
ti
f
s
o
f
b
ati
k
i
m
a
g
es
f
r
o
m
5
class
es
o
f
g
eo
m
etr
ic
p
atter
n
.
T
h
e
i
m
ag
e
s
w
er
e
f
ir
s
t
ly
g
r
o
u
p
ed
in
to
5
class
es
b
ased
o
n
th
e
n
at
u
r
e
b
atik
m
o
ti
f
s
.
T
h
e
5
class
es
ar
e
ce
p
lo
k,
ka
w
u
n
g
,
le
r
en
g
,
n
itik
,
an
d
p
a
r
a
n
g
.
T
h
ese
tem
p
late
s
w
er
e
cr
o
p
p
e
d
in
to
p
r
im
iti
v
e
p
atter
n
s
.
T
h
e
q
u
er
y
d
ataset
i
s
g
e
n
er
ated
f
r
o
m
ea
ch
te
m
p
late
b
ased
o
n
tr
an
s
latio
n
,
s
ca
le
,
an
d
r
o
tatio
n
.
T
h
e
to
tal
n
u
m
b
er
o
f
g
e
n
er
atin
g
q
u
er
y
i
m
ag
e
s
i
s
3
,
1
6
5
in
J
P
G
f
o
r
m
at
a
n
d
t
h
e
y
h
a
v
e
a
s
ize
o
f
4
5
0
x
4
5
0
p
ix
els.
T
h
e
q
u
er
y
i
m
a
g
es
ar
e
g
e
n
er
ated
b
ased
o
n
th
e
r
ele
v
an
t
te
m
p
late
w
it
h
a
v
ar
iatio
n
o
f
tr
a
n
s
la
tio
n
,
s
ca
le,
an
d
r
o
tatio
n
.
T
h
er
e
is
a
n
u
m
b
er
o
f
te
m
p
lates i
n
ea
ch
q
u
er
y
i
m
a
g
e
r
an
g
in
g
f
r
o
m
1
to
9
o
b
j
ec
ts
.
E
x
a
m
p
les ar
e
s
h
o
w
n
i
n
Fi
g
u
r
e
5
.
I
n
th
is
r
esear
c
h
,
w
e
u
s
e
Gab
o
r
f
ilter
,
G
L
C
M,
L
o
g
Gab
o
r
f
il
ter
,
an
d
L
o
ca
l
B
i
n
ar
y
P
atter
n
as
tex
tu
r
e
f
ea
t
u
r
es.
Fo
r
th
e
Gab
o
r
f
ilter
b
an
k
,
w
e
c
h
o
o
s
e
6
v
alu
e
s
o
f
o
r
ien
tatio
n
:
0
0
,
3
0
0
,
6
0
0
,
9
0
0
,
1
2
0
0
,
an
d
1
5
0
0
an
d
4
f
r
eq
u
en
c
ies
ar
e
s
elec
ted
:
2
2
.
6
3
,
1
1
.
3
1
,
5
.
6
6
,
an
d
2
.
8
3
f
o
llo
w
i
n
g
[
1
4
]
,
[
2
9
]
.
GL
C
M
w
a
s
co
m
p
u
ted
o
v
e
r
d
is
tan
ce
d
=
1
an
d
4
d
ir
ec
tio
n
s
(
0
0
,
4
5
0
,
9
0
0
,
an
d
1
3
5
0
)
.
F
o
r
L
o
g
Gab
o
r
,
f
o
llo
w
i
n
g
[
1
6
]
,
w
e
ch
o
o
s
e
f
o
u
r
n
u
m
b
er
s
o
f
w
a
v
elet
s
ca
le
s
a
n
d
s
ix
n
u
m
b
er
s
o
f
f
ilter
o
r
ien
t
atio
n
s
.
L
o
ca
l
B
i
n
ar
y
P
atter
n
(
L
B
P
)
f
ea
tu
r
es
ar
e
ex
tr
ac
ted
u
s
i
n
g
i
n
v
ar
ia
n
t
p
ar
a
m
eter
s
et
to
1
w
h
ic
h
m
ea
n
s
ev
er
y
L
B
P
lab
el
i
s
co
m
p
u
ted
u
s
in
g
th
e
m
in
i
m
al
ch
ain
b
y
r
o
tati
n
g
n
ei
g
h
b
o
r
in
g
p
ix
els
.
I
n
o
r
d
er
to
g
et
th
e
b
est r
etr
iev
al
r
esu
lt,
t
h
e
t
w
o
t
y
p
es o
f
d
ata
s
et
u
s
ed
as a
q
u
er
y
ar
e
ex
p
lai
n
ed
b
elo
w
:
Data
s
et
1
:
I
m
ag
e
d
ataset
w
h
ic
h
co
n
tai
n
ed
o
n
e
b
asi
c
m
o
tif
o
b
j
ec
t te
m
p
late
f
r
o
m
a
p
ar
ticu
lar
clas
s
.
Data
s
et
2
:
I
m
ag
e
d
ataset
w
h
ic
h
co
n
tai
n
ed
m
u
ltip
le
o
b
j
ec
ts
I
n
th
i
s
r
esear
ch
w
e
atte
m
p
t t
h
r
ee
s
ce
n
ar
io
s
as
f
o
llo
w
s
:
Scen
ar
io
1
: O
n
e
i
m
ag
e
co
n
tai
n
i
n
g
o
n
e
o
b
j
ec
t is selected
a
s
a
q
u
er
y
i
m
ag
e,
w
h
ile
th
e
r
est
i
m
a
g
es
i
n
d
atase
t
1
ar
e
u
s
ed
as te
m
p
late
i
m
ag
e
s
.
Scen
ar
io
2
:
On
e
i
m
a
g
e
co
n
tai
n
in
g
m
u
l
t
ip
le
o
b
j
ec
ts
is
s
elec
ted
f
r
o
m
th
e
d
ataset
2
,
w
h
i
le
th
e
i
m
ag
es
in
d
ataset
1
ar
e
u
s
ed
as t
e
m
p
late
i
m
a
g
es.
Scen
ar
io
3
:
On
e
i
m
a
g
e
co
n
tain
in
g
m
u
ltip
l
e
o
b
j
ec
ts
is
s
elec
ted
f
r
o
m
th
e
d
ataset
2
,
w
h
ile
t
h
e
r
est
i
m
ag
e
in
d
ataset
2
w
ill ar
e
as te
m
p
late
i
m
ag
e
s
.
w
h
er
e:
T
em
p
late
i
m
a
g
e
s
:
i
m
a
g
es i
n
d
atab
ase
Qu
er
y
i
m
a
g
es
:
i
m
a
g
es
r
e
f
er
r
in
g
to
u
s
er
’
s
tas
k
o
f
s
ea
r
ch
in
g
f
o
r
b
atik
m
o
ti
f
o
b
j
ec
ts
th
at
ar
e
r
elev
a
n
t
to
a
q
u
er
y
w
it
h
i
n
i
m
a
g
es i
n
d
atab
ase
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
Textu
r
e
F
u
s
io
n
fo
r
B
a
tik
Mo
tif R
etri
ev
a
l S
ystem
(
I
d
a
N
u
r
h
a
id
a
)
3181
(
a)
T
em
p
late
I
m
a
g
e
002
-
C
P
K0
0
3
-
B
lib
ar
1
1
.
p
n
g
w
it
h
S
ca
le
1
.
0
.
(
b
)
Qu
er
y
I
m
ag
e
w
i
th
3
O
b
j
ec
ts
,
S
ca
le
0
.
5
an
d
R
o
tatio
n
0
D
eg
r
ee
w
it
h
R
an
d
o
m
P
o
s
itio
n
(
c)
Qu
er
y
d
ata
w
i
th
9
o
b
j
ec
ts
,
Sc
ale
0
.
5
an
d
Ro
tatio
n
0
De
g
r
ee
w
it
h
R
eg
u
lar
P
a
tter
n
Fig
u
r
e
5
.
E
x
p
er
i
m
en
t
Data
s
e
t
Gen
e
r
ated
f
r
o
m
T
e
m
p
late
I
m
a
ge
E
ac
h
i
m
a
g
e
i
n
a
d
ataset
i
s
u
s
ed
as
a
q
u
er
y
i
m
a
g
e
a
n
d
te
m
p
late
i
m
a
g
e
alter
n
atel
y
.
T
h
e
p
u
r
p
o
s
e
o
f
th
ese
s
ce
n
ar
io
s
is
to
f
i
n
d
o
u
t
w
h
et
h
er
th
e
f
ea
t
u
r
e
f
u
s
io
n
w
i
ll
i
m
p
r
o
v
e
r
etr
ie
v
al
r
ate
i
f
t
h
e
n
u
m
b
er
o
f
m
o
ti
f
o
b
j
ec
ts
is
ad
d
ed
to
te
m
p
late
a
n
d
q
u
er
y
d
ata
in
cr
e
m
en
tall
y
.
Fu
r
t
h
er
m
o
r
e,
w
e
i
n
v
e
s
ti
g
ate
w
h
et
h
er
th
e
s
y
s
te
m
ca
n
d
is
co
v
er
t
h
e
p
r
i
m
iti
v
e
m
o
tif
t
h
at
is
co
n
tai
n
ed
in
q
u
e
r
y
i
m
ag
e
a
g
ai
n
s
t
te
m
p
late
d
atase
t
.
W
e
f
ir
s
t
r
ep
r
esen
t
th
e
i
m
ag
e
s
in
t
h
e
g
r
e
y
s
ca
le
a
n
d
th
e
n
e
m
p
lo
y
t
h
e
f
ea
tu
r
e
e
x
tr
ac
tio
n
m
et
h
o
d
s
to
g
ai
n
th
e
f
e
at
u
re
v
e
cto
r
f
o
r
e
ach
imag
e
.
In
p
art
icu
l
ar,
t
h
e
im
ag
e
re
t
rie
v
al
p
ro
ce
s
s
is
b
as
e
d
o
n
f
u
s
io
n
p
ro
ce
s
s
w
h
ich
u
t
il
ize
d
f
e
at
u
re
s
re
s
u
lt
e
d
f
ro
m
f
e
a
t
u
re
e
x
t
r
act
io
n
p
ro
ce
s
s
e
s
re
s
p
e
cti
v
e
ly
.
T
h
e
f
e
at
u
re
v
e
cto
r
is
ach
ie
v
e
d
as
f
o
ll
o
w
s
:
{
}
,
{
}
{
}
{
}
{
}
-
w
h
er
e:
=
f
ea
tu
r
e
v
ec
to
r
r
esu
lted
f
r
o
m
ea
ch
f
ea
t
u
r
e
ex
tr
ac
tio
n
m
et
h
o
d
E
ac
h
t
y
p
e
o
f
f
ea
t
u
r
es
r
ep
r
ese
n
ts
s
o
m
e
i
m
ag
e
ch
ar
ac
ter
is
tic
s
o
th
a
t
t
h
e
f
u
s
io
n
p
r
o
ce
s
s
w
i
ll
in
teg
r
ate
i
m
a
g
e
ch
ar
ac
ter
i
s
tics
to
f
u
l
f
il
l
i
m
a
g
e
r
etr
iev
a
l.
Sp
ec
i
f
icall
y
,
th
e
f
ea
t
u
r
e
v
ec
to
r
is
r
ep
r
es
en
ted
as
a
s
i
n
g
le
v
ec
to
r
.
T
h
er
ef
o
r
e,
w
e
ap
p
l
y
t
h
e
n
o
r
m
aliza
tio
n
p
r
o
ce
s
s
w
h
ic
h
tr
an
s
f
o
r
m
s
all
f
ea
t
u
r
es
in
to
a
co
m
p
ar
ab
le
r
an
g
e
o
f
v
al
u
es.
I
n
o
t
h
er
w
o
r
d
s
,
th
is
n
o
r
m
aliza
tio
n
p
r
o
ce
s
s
ch
an
g
es
t
h
e
r
an
g
e
o
f
p
i
x
el
in
te
n
s
it
y
v
al
u
es
b
et
w
ee
n
[
0
–
1
]
.
T
h
e
q
u
er
y
d
at
aset
an
d
te
m
p
late
d
ataset
f
r
o
m
t
h
e
d
atab
ases
ar
e
u
s
ed
b
y
th
e
s
y
s
te
m
to
o
b
tain
th
e
b
est
1
,
3
,
5
,
7
,
4
2
an
d
6
3
2
m
atch
ed
i
m
ag
e
s
o
r
d
er
ed
b
y
t
h
eir
s
i
m
ilar
it
y
,
i.e
.
E
u
cli
d
ea
n
d
is
tan
ce
.
T
h
e
T
o
p
-
4
2
an
d
T
o
p
-
6
3
2
ar
e
th
e
a
v
er
ag
e
n
u
m
b
er
o
f
te
m
p
late
i
m
ag
es
an
d
q
u
er
y
i
m
ag
e
s
w
h
ic
h
d
is
tr
ib
u
ted
i
n
ea
c
h
class
co
n
s
ec
u
ti
v
el
y
.
T
h
e
ai
m
is
to
ac
h
ie
v
e
t
h
e
b
est
p
er
f
o
r
m
an
ce
o
f
t
h
e
s
y
s
te
m
t
h
at
ca
n
r
etr
iev
e
u
p
to
t
h
o
s
e
n
u
m
b
er
s
o
f
i
m
a
g
es.
T
h
e
s
i
m
il
ar
it
y
ca
lcu
latio
n
o
f
a
s
in
g
le
f
e
atu
r
e
v
ec
to
r
is
as
f
o
llo
w
s
:
1)
C
alcu
late
th
e
a
v
er
ag
e
v
al
u
e
o
f
th
e
f
ea
t
u
r
e
v
ec
to
r
an
d
v
ar
ian
c
e
.
2)
C
alcu
late
E
u
clid
ia
n
d
is
ta
n
ce
(
)
w
h
er
e
(
i≠
j
)
b
et
w
ee
n
q
u
er
y
i
m
ag
e
an
d
te
m
p
la
te
i
m
a
g
es i
n
d
a
tab
ase
(
)
√
∑
(
)
w
h
er
e:
r
=
co
r
r
esp
o
n
d
in
g
d
escr
ip
to
r
o
f
f
ea
t
u
r
e
3)
Fo
r
g
iv
e
n
q
u
er
y
i
m
a
g
e
q
,
ca
lcu
late
E
u
c
lid
ian
d
is
ta
n
ce
w
it
h
a
ll i
m
a
g
e
in
te
m
p
late
d
atab
ase
Dis
t
(
1
,
q
)
,
Dis
t
(
2
,
q
)
,
Dis
t
(
3
,
q
)
,
…
Dis
t
(
m
total
,
q
)
w
h
er
e:
m
total
=
to
tal
te
m
p
late
i
m
ag
e
s
i
n
th
e
d
atab
ase
T
h
e
p
er
f
o
r
m
an
ce
o
f
s
u
cc
ess
f
u
l r
etr
iev
al
i
m
a
g
e
is
e
v
al
u
ated
b
y
t
h
e
t
w
o
m
ea
s
u
r
es,
p
r
ec
is
io
n
,
an
d
r
ec
all.
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.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
3
1
7
4
–
3
187
3182
4.
RE
SU
L
T
AND
ANA
L
YS
I
S
T
ab
le
1
c
o
m
p
ar
es
t
h
e
p
r
ec
is
io
n
a
n
d
r
ec
all
r
esu
lts
o
b
tai
n
ed
f
r
o
m
th
e
s
y
s
te
m
o
n
to
p
-
N
s
i
m
i
lar
i
m
a
g
e
s
f
r
o
m
ea
c
h
c
lass
.
B
ased
o
n
th
e
ex
p
er
i
m
e
n
tal
r
es
u
lts
u
s
i
n
g
s
ce
n
ar
io
1
,
all
s
i
n
g
le
-
f
ea
t
u
r
e
an
d
f
ea
tu
r
e
f
u
s
io
n
m
et
h
o
d
s
ar
e
ab
le
to
r
etr
iev
e
th
e
s
a
m
e
i
m
a
g
e
f
r
o
m
th
e
i
m
a
g
e
d
atab
ase.
T
h
e
f
u
s
io
n
o
f
Gab
o
r
,
GL
C
M,
L
B
P
,
an
d
L
o
g
Gab
o
r
f
ea
t
u
r
es
g
ai
n
ed
th
e
h
i
g
h
est
r
etr
iev
al
p
er
f
o
r
m
a
n
ce
f
o
r
th
e
to
p
3
r
e
tr
iev
al
im
ag
e
s
,
w
h
er
ea
s
t
h
e
f
u
s
io
n
o
f
Gab
o
r
,
L
o
g
Gab
o
r
,
an
d
GL
C
M
f
ea
tu
r
e
s
ac
h
ie
v
es
t
h
e
h
i
g
h
est
v
al
u
e
f
o
r
t
h
e
to
p
5
,
7
,
an
d
4
2
.
T
h
e
b
est
p
r
ec
is
io
n
is
7
3
.
9
3
%
f
o
r
th
e
to
p
3
an
d
th
e
b
est
r
ec
all
is
3
2
.
5
3
%
f
o
r
to
p
4
2
.
T
h
e
b
est
p
r
ec
is
io
n
an
d
r
ec
all
u
s
in
g
a
s
i
n
g
le
f
ea
tu
r
e
is
o
b
tain
ed
u
s
in
g
L
o
g
Gab
o
r
f
ea
-
tu
r
e
w
it
h
p
r
ec
is
io
n
o
f
7
2
.
8
3
%
f
o
r
th
e
to
p
3
an
d
r
ec
all
o
f
3
1
.
8
8
% f
o
r
th
e
to
p
4
2
.
T
ab
le
1
.
P
r
ec
is
io
n
an
d
Re
ca
ll
o
f
th
e
B
atik
Mo
ti
f
R
etr
iev
al
S
y
s
te
m
f
o
r
Te
m
p
late
to
T
em
p
la
te
Scen
ar
io
1
F
e
a
t
u
r
e
Ex
t
r
a
c
t
i
o
n
M
e
t
h
o
d
T
O
P
1
3
5
7
4
2
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
G
a
b
o
r
1
0
0
%
2
.
3
7
%
6
9
.
9
8
%
4
.
5
3
%
5
9
.
0
5
%
6
.
0
2
%
5
4
.
1
6
%
7
.
4
8
%
3
5
.
4
7
%
2
6
.
2
3
%
G
L
C
M
1
0
0
%
2
.
3
7
%
5
8
.
6
1
%
3
.
9
0
%
4
8
.
9
1
%
5
.
2
1
%
4
4
.
4
1
%
6
.
4
1
%
3
4
.
0
4
%
2
6
.
6
2
%
L
B
P
1
0
0
%
2
.
3
7
%
5
7
.
5
0
%
3
.
5
5
%
4
8
.
3
4
%
4
.
8
6
%
4
2
.
9
9
%
5
.
8
2
%
3
1
.
7
9
%
2
2
.
8
6
%
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
2
.
8
3
%
4
.
5
0
%
6
3
.
5
1
%
6
.
3
3
%
5
8
.
9
0
%
7
.
9
4
%
4
3
.
5
5
%
3
1
.
8
8
%
G
a
b
o
r
G
L
C
M
1
0
0
%
2
.
3
7
%
6
5
.
5
6
%
4
.
3
0
%
5
5
.
4
5
%
5
.
8
4
%
5
0
.
6
4
%
7
.
2
4
%
3
6
.
4
0
%
2
8
.
1
5
%
G
a
b
o
r
L
B
P
1
0
0
%
2
.
3
7
%
6
4
.
6
1
%
4
.
1
6
%
5
5
.
1
7
%
5
.
6
4
%
5
0
.
8
5
%
7
.
0
1
%
3
5
.
1
4
%
2
6
.
1
1
%
G
a
b
o
r
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
2
.
8
3
%
4
.
5
2
%
6
4
.
2
7
%
6
.
4
5
%
5
9
.
3
8
%
8
.
0
9
%
4
4
.
0
1
%
3
2
.
4
8
%
G
L
C
M
L
B
P
1
0
0
%
2
.
3
7
%
6
0
.
6
6
%
3
.
9
9
%
5
2
.
2
3
%
5
.
4
1
%
4
7
.
1
2
%
6
.
6
3
%
3
4
.
1
8
%
2
6
.
2
1
%
G
L
C
M
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
2
.
9
9
%
4
.
5
3
%
6
3
.
4
1
%
6
.
3
6
%
5
8
.
3
6
%
7
.
9
7
%
4
3
.
4
2
%
3
2
.
1
9
%
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
3
.
4
6
%
4
.
5
1
%
6
3
.
8
9
%
6
.
3
0
%
5
9
.
4
4
%
8
.
0
4
%
4
3
.
7
5
%
3
1
.
9
8
%
G
a
b
o
r
G
L
C
M
L
B
P
1
0
0
%
2
.
3
7
%
6
6
.
3
5
%
4
.
3
3
%
5
5
.
5
5
%
5
.
7
8
%
5
0
.
9
8
%
7
.
2
9
%
3
6
.
7
7
%
2
8
.
2
3
%
G
a
b
o
r
G
L
C
M
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
3
.
1
4
%
4
.
5
7
%
6
3
.
6
0
%
6
.
4
2
%
5
8
.
9
7
%
8
.
0
7
%
4
3
.
5
5
%
3
2
.
4
7
%
G
a
b
o
r
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
3
.
1
4
%
4
.
5
7
%
6
4
.
9
3
%
6
.
4
9
%
6
0
.
1
9
%
8
.
2
1
%
4
4
.
1
8
%
3
2
.
5
3
%
G
L
C
M
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
3
.
4
6
%
4
.
5
6
%
6
3
.
6
0
%
6
.
3
7
%
5
8
.
9
0
%
8
.
0
8
%
4
3
.
4
7
%
3
2
.
0
8
%
G
a
b
o
r
G
L
C
M
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
3
.
9
3
%
4
.
6
1
%
6
3
.
9
8
%
6
.
4
6
%
5
9
.
2
4
%
8
.
1
4
%
4
3
.
7
3
%
3
2
.
4
8
%
Fro
m
Fi
g
u
r
e
6
,
it
is
ap
p
ar
en
t
t
h
at
th
e
f
ir
s
t
r
an
k
o
f
th
e
r
etr
ie
v
ed
im
a
g
e
s
f
r
o
m
t
h
e
te
m
p
late
d
atab
ase
is
ex
ac
tl
y
t
h
e
s
a
m
e
i
m
a
g
e
a
s
a
q
u
er
y
i
m
a
g
e.
A
s
s
ee
n
i
n
T
ab
l
e
2
,
th
e
f
ea
tu
r
e
f
u
s
io
n
p
er
f
o
r
m
an
ce
o
u
tp
er
f
o
r
m
s
s
in
g
le
-
f
ea
tu
r
e
p
er
f
o
r
m
a
n
ce
.
Featu
r
e
f
u
s
io
n
f
r
o
m
G
L
C
M
an
d
L
o
g
Gab
o
r
ac
h
iev
es
th
e
b
est
p
r
ec
is
io
n
o
f
5
4
.
6
8
%,
4
6
.
1
5
%,
4
3
.
7
1
%
f
o
r
to
p
1
,
3
,
an
d
5
,
r
esp
ec
tiv
e
l
y
.
Ho
w
e
v
er
,
r
etr
ie
v
al
o
f
t
h
e
i
m
ag
e
f
o
r
th
e
to
p
4
2
o
b
tain
s
f
r
o
m
f
ea
t
u
r
e
f
u
s
io
n
o
f
L
B
P
an
d
L
o
g
Gab
o
r
at
3
6
.
4
0
%.
Fu
r
t
h
er
m
o
r
e,
r
etr
iev
al
r
ate
f
r
o
m
f
ea
tu
r
e
f
u
s
io
n
o
u
t
p
er
f
o
r
m
s
t
h
e
b
est
s
in
g
le
f
e
atu
r
e
p
er
f
o
r
m
an
ce
f
r
o
m
L
o
g
Gab
o
r
f
ea
tu
r
e
w
h
ic
h
s
h
o
w
s
p
r
ec
is
io
n
at
5
3
.
6
1
%,
4
6
.
0
4
%,
4
3
.
4
5
%,
an
d
3
6
.
0
8
%
f
o
r
to
p
1
,
3
,
5
,
an
d
4
2
,
r
e
s
p
ec
tiv
el
y
.
I
n
t
h
i
s
s
ce
n
ar
io
,
th
e
r
esu
lt
s
o
f
r
ec
all
ar
e
s
ig
n
i
f
ica
n
tl
y
lo
w
er
t
h
an
r
ec
all
v
al
u
es
f
r
o
m
s
ce
n
ar
io
1
.
I
t
is
d
u
e
to
i
n
s
u
f
f
ic
ien
t
in
f
o
r
m
a
tio
n
lo
ca
ted
in
te
m
p
late
d
ataset
co
m
p
ar
ed
to
th
e
g
r
ea
t
er
n
u
m
b
er
s
o
f
m
o
ti
f
co
n
tai
n
e
d
in
th
e
q
u
er
y
i
m
a
g
e
d
ataset.
I
t
m
ea
n
s
t
h
e
q
u
er
y
i
m
a
g
es
h
a
v
e
m
o
r
e
co
m
p
l
icate
d
ch
ar
ac
ter
is
tic.
T
h
er
ef
o
r
e,
it
is
h
ar
d
to
f
i
n
d
t
h
e
r
ele
v
an
t
i
m
ag
e
i
n
th
e
te
m
p
late
d
atab
ase.
T
h
e
r
etr
iev
ed
i
m
a
g
e
s
b
ase
d
o
n
s
ce
n
ar
io
2
is
s
h
o
w
n
in
Fi
g
u
r
e
7.
T
ab
le
2
.
P
r
ec
is
io
n
an
d
r
ec
all
o
f
th
e
b
ati
k
m
o
t
if
r
etr
ie
v
al
s
y
s
t
e
m
f
o
r
q
u
er
y
to
te
m
p
late
s
ce
n
ar
io
2
F
e
a
t
u
r
e
Ex
t
r
a
c
t
i
o
n
M
e
t
h
o
d
T
O
P
1
3
5
7
42
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
G
a
b
o
r
2
9
.
8
4
%
0
.
6
1
%
2
9
.
3
3
%
1
.
8
1
%
2
9
.
2
5
%
2
.
9
6
%
2
9
.
2
8
%
4
.
1
2
%
2
8
.
9
5
%
2
1
.
9
3
%
G
L
C
M
3
2
.
5
9
%
0
.
6
1
%
3
1
.
4
6
%
1
.
7
8
%
3
1
.
1
9
%
2
.
9
6
%
3
0
.
3
7
%
4
.
0
7
%
2
8
.
9
1
%
2
2
.
3
9
%
L
B
P
1
8
.
4
3
%
0
.
4
5
%
1
5
.
0
0
%
1
.
3
8
%
2
1
.
8
0
%
2
.
3
2
%
2
7
.
0
8
%
3
.
2
5
%
3
0
.
5
6
%
1
9
.
9
1
%
L
o
g
-
G
a
b
o
r
5
3
.
5
1
%
1
.
1
5
%
4
6
.
0
4
%
2
.
7
6
%
4
3
.
4
5
%
4
.
2
5
%
4
1
.
5
7
%
5
.
5
5
%
3
6
.
0
8
%
2
5
.
2
4
%
G
a
b
o
r
G
L
C
M
3
2
.
2
4
%
0
.
6
5
%
3
1
.
2
3
%
1
.
9
9
%
3
0
.
4
2
%
3
.
1
8
%
2
9
.
9
6
%
4
.
4
3
%
2
8
.
6
2
%
2
3
.
1
7
%
G
a
b
o
r
L
B
P
2
0
.
1
3
%
0
.
5
1
%
3
0
.
0
1
%
1
.
5
6
%
3
3
.
1
2
%
2
.
5
7
%
3
3
.
6
1
%
3
.
5
6
%
3
2
.
8
7
%
2
0
.
7
3
%
G
a
b
o
r
L
o
g
-
G
a
b
o
r
5
2
.
1
5
%
1
.
1
4
%
4
4
.
9
9
%
2
.
7
6
%
4
2
.
4
9
%
4
.
2
4
%
4
0
.
8
5
%
5
.
5
2
%
3
5
.
8
1
%
2
5
.
4
9
%
G
L
C
M
L
B
P
3
1
.
4
5
%
0
.
5
9
%
3
3
.
0
2
%
1
.
6
9
%
3
4
.
0
6
%
2
.
7
5
%
3
2
.
2
7
%
3
.
7
6
%
3
0
.
2
5
%
2
0
.
6
5
%
G
L
C
M
L
o
g
-
G
a
b
o
r
5
4
.
6
8
%
1
.
1
8
%
4
6
.
1
5
%
2
.
8
0
%
4
3
.
7
1
%
4
.
3
2
%
4
1
.
2
7
%
5
.
6
1
%
3
5
.
7
9
%
2
5
.
3
3
%
L
B
P
L
o
g
-
G
a
b
o
r
4
8
.
9
6
%
0
.
9
5
%
4
4
.
2
3
%
2
.
4
4
%
4
2
.
6
6
%
3
.
8
7
%
4
1
.
5
6
%
5
.
1
5
%
3
6
.
4
0
%
2
4
.
9
0
%
G
a
b
o
r
G
L
C
M
L
B
P
2
9
.
9
0
%
0
.
6
1
%
3
2
.
8
0
%
1
.
7
7
%
3
3
.
4
0
%
2
.
8
6
%
3
3
.
4
6
%
3
.
8
9
%
3
2
.
3
2
%
2
1
.
6
2
%
G
a
b
o
r
G
L
C
M
L
o
g
-
G
a
b
o
r
5
3
.
4
1
%
1
.
1
7
%
4
5
.
1
5
%
2
.
7
7
%
4
2
.
5
4
%
4
.
3
1
%
4
0
.
6
2
%
5
.
6
0
%
3
5
.
6
4
%
2
5
.
5
8
%
G
a
b
o
r
L
B
P
L
o
g
-
G
a
b
o
r
4
8
.
7
0
%
0
.
9
3
%
4
4
.
5
8
%
2
.
4
7
%
4
2
.
8
4
%
3
.
9
2
%
4
1
.
5
8
%
5
.
1
5
%
3
6
.
2
2
%
2
5
.
0
8
%
G
L
C
M
L
B
P
L
o
g
-
G
a
b
o
r
5
1
.
1
4
%
1
.
0
0
%
4
5
.
0
0
%
2
.
5
5
%
4
3
.
0
2
%
3
.
9
5
%
4
1
.
1
1
%
5
.
1
8
%
3
6
.
1
8
%
2
4
.
9
2
%
G
a
b
o
r
G
L
C
M
L
B
P
L
o
g
-
G
a
b
o
r
4
9
.
9
4
%
0
.
9
9
%
4
5
.
1
1
%
2
.
5
5
%
4
2
.
5
2
%
3
.
9
6
%
4
1
.
2
6
%
5
.
2
4
%
3
6
.
1
2
%
2
5
.
1
3
%
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
Textu
r
e
F
u
s
io
n
fo
r
B
a
tik
Mo
tif R
etri
ev
a
l S
ystem
(
I
d
a
N
u
r
h
a
id
a
)
3183
Fig
u
r
e
6
.
T
h
e
r
etr
iev
ed
im
a
g
e
s
at
to
p
5
b
ased
o
n
1
m
o
ti
f
te
m
p
late
as a
q
u
er
y
i
m
a
g
e
ag
a
in
s
t to
te
m
p
late
i
m
a
g
es i
n
d
atab
ase
(
s
ce
n
ar
io
1
)
Fig
u
r
e
7.
T
h
e
r
etr
iev
ed
im
a
g
e
s
b
ased
o
n
q
u
er
y
i
m
ag
e
at
to
p
5
th
at
co
n
tai
n
ed
m
o
r
e
th
a
n
1
m
o
ti
f
as a
q
u
er
y
i
m
a
g
e
ag
ai
n
s
t to
te
m
p
late
i
m
a
g
es i
n
th
e
d
atab
ase
(
s
ce
n
ar
io
2
)
T
ab
le
3
illu
s
tr
ates
s
o
m
e
c
h
ar
a
cter
is
tics
o
f
i
m
a
g
e
r
etr
iev
al
r
esu
lt
s
f
r
o
m
t
h
e
s
y
s
te
m
f
o
r
s
ce
n
ar
io
3
.
A
p
o
s
itiv
e
co
r
r
elatio
n
w
a
s
f
o
u
n
d
b
et
w
ee
n
an
in
cr
ea
s
e
in
th
e
n
u
m
b
er
o
f
m
o
ti
f
s
in
q
u
er
y
i
m
ag
e
s
a
n
d
te
m
p
lat
e
i
m
a
g
e
in
p
r
ec
is
io
n
.
Fu
r
t
h
er
m
o
r
e,
it
ca
n
b
e
s
ee
n
f
r
o
m
T
ab
le
3
th
at
f
ea
t
u
r
e
f
u
s
io
n
o
u
tp
er
f
o
r
m
s
s
i
n
g
le
f
ea
tu
r
es
f
o
r
p
r
ec
is
io
n
at
3
,
5
,
an
d
7
.
T
h
e
b
es
t
p
r
ec
is
io
n
i
s
8
4
.
3
6
%
ac
h
iev
ed
f
r
o
m
f
ea
t
u
r
e
f
u
s
io
n
o
f
Gab
o
r
,
GL
C
M,
an
d
L
o
g
Ga
b
o
r
f
o
r
to
p
3
;
w
h
i
le
th
e
s
i
n
g
le
f
ea
t
u
r
e
ac
h
iev
e
s
8
3
.
2
8
%
f
r
o
m
L
o
g
Ga
b
o
r
as
th
e
b
est.
Me
an
w
h
ile,
r
ec
all
g
ets
0
.
3
8
%,
0
.
5
6
%,
0
.
7
1
%,
an
d
2
3
.
8
8
%
f
o
r
to
p
3
,
5
,
7
,
an
d
6
3
2
c
o
n
s
e
cu
ti
v
el
y
.
T
h
ese
r
esu
lts
ar
e
li
k
el
y
to
b
e
r
elate
d
to
ad
eq
u
ate
in
f
o
r
m
atio
n
f
o
u
n
d
i
n
b
o
th
q
u
er
y
i
m
a
g
e
an
d
i
m
a
g
e
in
th
e
d
atab
ase
to
f
in
d
t
h
e
s
i
m
ilar
i
m
a
g
e.
T
h
e
r
etr
iev
ed
i
m
a
g
es b
ased
o
n
s
ce
n
ar
io
3
ca
n
b
e
s
ee
n
in
F
ig
u
r
e
8
.
T
ab
le
3
.
P
r
ec
is
io
n
an
d
r
ec
all
o
f
th
e
b
ati
k
m
o
t
if
r
etr
ie
v
al
s
y
s
t
e
m
f
o
r
q
u
er
y
to
q
u
er
y
(
s
ce
n
ar
i
o
3
)
F
e
a
t
u
r
e
Ex
t
r
a
c
t
i
o
n
M
e
t
h
o
d
T
O
P
1
3
5
7
6
3
2
(
a
v
e
r
a
g
e
)
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
P
r
e
c
R
e
c
G
a
b
o
r
1
0
0
%
2
.
3
7
%
7
4
.
9
4
%
0
.
3
3
%
6
4
.
6
7
%
0
.
4
6
%
5
9
.
0
2
%
0
.
5
8
%
3
0
.
3
8
%
2
1
.
3
6
%
G
L
C
M
1
0
0
%
2
.
3
7
%
6
7
.
8
0
%
0
.
3
0
%
5
8
.
0
9
%
0
.
4
1
%
5
1
.
8
7
%
0
.
5
0
%
2
9
.
7
3
%
2
1
.
4
4
%
L
B
P
1
0
0
%
2
.
3
7
%
5
9
.
0
6
%
0
.
2
5
%
4
8
.
9
3
%
0
.
3
3
%
4
4
.
5
1
%
0
.
4
1
%
2
8
.
8
3
%
1
9
.
9
9
%
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
3
.
2
8
%
0
.
3
8
%
7
5
.
3
4
%
0
.
5
5
%
6
9
.
5
6
%
0
.
6
9
%
3
5
.
3
6
%
2
3
.
7
9
%
G
a
b
o
r
G
L
C
M
1
0
0
%
2
.
3
7
%
7
5
.
0
0
%
0
.
3
4
%
6
4
.
8
5
%
0
.
4
7
%
5
8
.
3
8
%
0
.
5
8
%
3
0
.
2
0
%
2
1
.
9
2
%
G
a
b
o
r
L
B
P
1
0
0
%
2
.
3
7
%
7
0
.
6
6
%
0
.
3
1
%
5
9
.
7
2
%
0
.
4
1
%
5
4
.
2
0
%
0
.
5
1
%
2
9
.
6
8
%
2
0
.
8
3
%
G
a
b
o
r
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
3
.
8
7
%
0
.
3
8
%
7
5
.
8
0
%
0
.
5
6
%
7
0
.
0
7
%
0
.
7
0
%
3
5
.
3
0
%
2
3
.
8
8
%
G
L
C
M
L
B
P
1
0
0
%
2
.
3
7
%
6
4
.
3
6
%
0
.
2
8
%
5
4
.
0
0
%
0
.
3
8
%
4
8
.
9
8
%
0
.
4
7
%
2
9
.
1
4
%
2
1
.
0
2
%
G
L
C
M
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
3
.
9
0
%
0
.
3
8
%
7
6
.
0
0
%
0
.
5
5
%
7
0
.
3
3
%
0
.
6
9
%
3
5
.
2
0
%
2
3
.
8
2
%
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
2
.
4
9
%
0
.
3
7
%
7
3
.
8
4
%
0
.
5
3
%
6
8
.
2
1
%
0
.
6
7
%
3
4
.
0
5
%
2
3
.
2
0
%
G
a
b
o
r
G
L
C
M
L
B
P
1
0
0
%
2
.
3
7
%
6
9
.
9
0
%
0
.
3
1
%
5
9
.
0
6
%
0
.
4
2
%
5
3
.
8
1
%
0
.
5
2
%
2
9
.
4
4
%
2
1
.
2
9
%
G
a
b
o
r
G
L
C
M
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
4
.
3
6
%
0
.
3
8
%
7
6
.
1
6
%
0
.
5
6
%
7
0
.
8
1
%
0
.
7
0
%
3
5
.
0
7
%
2
3
.
8
8
%
G
a
b
o
r
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
3
.
0
7
%
0
.
3
7
%
7
4
.
7
5
%
0
.
5
4
%
6
8
.
9
4
%
0
.
6
8
%
3
4
.
0
7
%
2
3
.
2
9
%
G
L
C
M
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
2
.
9
9
%
0
.
3
7
%
7
4
.
9
4
%
0
.
5
4
%
6
9
.
1
7
%
0
.
6
8
%
3
3
.
8
5
%
2
3
.
2
2
%
G
a
b
o
r
G
L
C
M
L
B
P
L
o
g
-
G
a
b
o
r
1
0
0
%
2
.
3
7
%
8
3
.
5
4
%
0
.
3
8
%
7
5
.
4
3
%
0
.
5
5
%
6
9
.
6
7
%
0
.
6
9
%
3
3
.
8
2
%
2
3
.
2
9
%
Evaluation Warning : The document was created with Spire.PDF for Python.