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.
8
,
No
.
3
,
J
u
n
e
201
8
,
p
p
.
1920
~
1
9
3
2
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
8
i
3
.
p
p
1
9
2
0
-
1932
1920
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
H
erb
Lea
v
es Rec
o
g
nition us
ing
G
r
a
y
Lev
el Co
-
o
ccu
rrence
M
a
trix a
nd
Five
Dista
nce
-
ba
sed
Si
m
ila
rity M
ea
sure
s
R.
Riza
l Is
na
nto
1
,
M
un
a
w
a
r
Ag
us
Riy
a
di
2
, M
uh
a
m
m
a
d
F
a
h
m
i A
w
a
j
3
1,
2
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter
En
g
in
e
e
rin
g
,
Dip
o
n
e
g
o
ro
Un
iv
e
rsit
y
,
In
d
o
n
e
sia
3
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Dip
o
n
e
g
o
ro
U
n
iv
e
rsity
,
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
Ma
r
6
,
2
0
1
8
R
ev
i
s
ed
Ma
y
2
0
,
2
0
1
8
A
cc
ep
ted
Ma
y
2
6
,
2
0
1
8
He
rb
m
e
d
ic
in
a
l
p
ro
d
u
c
ts
d
e
riv
e
d
f
ro
m
p
lan
ts
h
a
v
e
lo
n
g
b
e
e
n
c
o
n
sid
e
re
d
a
s
a
n
a
lt
e
rn
a
ti
v
e
o
p
ti
o
n
f
o
r
trea
ti
n
g
v
a
rio
u
s
d
ise
a
se
s.
In
t
h
is
p
a
p
e
r,
t
h
e
f
e
a
tu
re
e
x
trac
ti
o
n
m
e
th
o
d
u
se
d
is
G
ra
y
Lev
e
l
Co
-
o
c
c
u
rre
n
c
e
M
a
tri
x
(
GL
C
M
),
w
h
il
e
f
o
r
it
s
re
c
o
g
n
it
io
n
u
sin
g
th
e
m
e
tri
c
c
a
lcu
la
ti
o
n
s
o
f
Ch
e
b
y
sh
e
v
,
Cit
y
b
lo
c
k
,
M
in
k
o
w
sk
i
,
Ca
n
b
e
rra
,
a
n
d
Eu
c
li
d
e
a
n
d
istan
c
e
s.
T
h
e
m
e
th
o
d
o
f
d
e
ter
m
in
in
g
th
e
GL
CM
A
n
a
l
y
sis
b
a
se
d
o
n
th
e
tex
tu
re
a
n
a
l
y
sis
re
su
lt
in
g
f
ro
m
th
e
e
x
trac
ti
o
n
o
f
th
is
f
e
a
tu
re
is
An
g
u
lar
S
e
c
o
n
d
M
o
m
e
n
t,
Co
n
tras
t,
In
v
e
rse
Diff
e
r
e
n
t
M
o
m
e
n
t,
En
tr
o
p
y
a
s
w
e
ll
a
s
it
s
Co
rre
latio
n
.
T
h
e
re
c
o
g
n
it
io
n
s
y
ste
m
u
se
d
1
0
lea
f
te
st
i
m
a
g
e
s
w
it
h
GL
CM
m
e
th
o
d
a
n
d
Ca
n
b
e
r
ra
d
istan
c
e
re
su
lt
e
d
in
th
e
h
ig
h
e
st
a
c
c
u
ra
c
y
o
f
9
2
.
0
0
%
.
W
h
il
e
th
e
u
se
o
f
2
0
a
n
d
3
0
tes
t
d
a
ta res
u
lt
e
d
i
n
a
re
c
o
g
n
it
i
o
n
ra
te
o
f
5
0
.
6
7
%
a
n
d
6
0
.
0
0
%
.
K
ey
w
o
r
d
:
C
an
b
er
r
a
d
is
tan
ce
C
h
eb
y
s
h
ev
d
is
tan
ce
C
it
y
-
b
lo
ck
d
i
s
tan
ce
E
u
clid
ea
n
d
is
ta
n
ce
Gr
a
y
-
lev
el
co
o
cc
u
r
r
en
ce
m
atr
i
x
Min
k
o
w
s
k
i d
is
ta
n
ce
Co
p
y
rig
h
t
©
2
0
1
8
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
:
R
.
R
izal
I
s
n
an
to
,
Dep
ar
te
m
en
t o
f
C
o
m
p
u
ter
E
n
g
in
ee
r
i
n
g
,
Dip
o
n
eg
o
r
o
Un
i
v
er
s
it
y
,
J
l.
Pro
f
.
So
ed
ar
to
,
S.H.
,
T
em
b
alan
g
,
Se
m
ar
a
n
g
5
0
2
7
5
,
I
n
d
o
n
esia.
E
m
ail: r
izal_
is
n
a
n
to
@
y
a
h
o
o
.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Her
b
m
ed
ical
p
r
o
d
u
cts
d
er
i
v
e
d
f
r
o
m
p
la
n
ts
h
a
v
e
b
ee
n
tr
ad
it
io
n
all
y
b
eliev
ed
a
s
a
n
al
ter
n
at
iv
e
o
p
tio
n
f
o
r
tr
ea
tin
g
v
ar
io
u
s
d
is
ea
s
es,
in
cl
u
d
in
g
f
o
r
th
e
tr
ea
t
m
e
n
t
o
f
ch
r
o
n
ic
d
is
ea
s
e
s
s
u
c
h
as
ca
n
ce
r
,
h
ea
r
t
d
is
ea
s
e,
h
ep
atitis
,
an
d
al
s
o
k
id
n
e
y
a
n
d
h
ea
r
t
f
ail
u
r
e,
at
lea
s
t
i
n
s
o
m
e
Asi
an
r
eg
io
n
s
.
T
h
e
r
ea
s
o
n
o
f
u
s
i
n
g
h
er
b
al
leav
e
s
is
th
e
lea
f
ca
n
b
e
ea
s
il
y
f
o
u
n
d
an
y
w
h
er
e
an
d
p
r
o
ce
s
s
ed
in
a
n
y
f
o
r
m
.
W
ith
to
d
a
y
's
tec
h
n
o
lo
g
ical
ad
v
an
ce
s
,
t
h
e
d
e
m
an
d
f
o
r
co
m
p
u
ter
ap
p
licatio
n
also
i
n
cr
ea
s
e
s
to
p
r
o
v
id
e
b
en
ef
it
s
to
h
u
m
an
li
f
e.
On
e
o
f
th
e
m
is
t
h
e
n
ee
d
f
o
r
ap
p
licati
o
n
s
t
h
at
ca
n
r
ec
o
g
n
ize
t
h
e
h
er
b
al
leav
es
p
atter
n
.
T
h
e
co
m
p
le
x
it
y
i
n
r
ec
o
g
n
izi
n
g
t
h
e
lea
f
-
b
a
s
ed
p
lan
ts
,
d
u
e
t
o
v
ar
io
u
s
t
y
p
es
a
n
d
t
h
e
d
if
f
e
r
en
t
u
s
es
o
f
h
er
b
al
leav
es,
m
ak
e
s
th
e
h
er
b
leav
es
r
ec
o
g
n
itio
n
d
if
f
ic
u
lt.
Var
io
u
s
f
ea
t
u
r
es
r
elate
d
to
th
e
tex
tu
r
e
o
f
t
h
e
leav
e
s
w
er
e
s
tu
d
ied
an
d
th
e
m
o
s
t
ap
p
r
o
p
r
iate
f
ea
tu
r
e
s
w
er
e
u
s
ed
f
o
r
leaf
i
m
a
g
e
-
b
ased
p
lan
t
cla
s
s
i
f
icatio
n
.
T
h
e
d
ev
elo
p
ed
s
y
s
te
m
co
u
ld
b
e
u
s
ed
to
id
e
n
ti
f
y
m
ed
icin
a
l
p
la
n
ts
f
o
r
p
ar
tic
u
lar
d
is
ea
s
e
s
o
f
h
u
m
a
n
b
ei
n
g
s
.
T
h
e
tex
t
u
r
e
f
ea
t
u
r
es
h
a
v
e
b
ee
n
e
x
tr
ac
ted
w
it
h
u
s
in
g
th
e
G
L
C
M
a
n
d
th
e
P
C
A
alg
o
r
ith
m
s
o
n
t
h
e
3
9
0
im
a
g
es
f
r
o
m
6
5
d
ataset
s
an
d
th
e
n
e
w
le
af
o
r
a
d
ef
ec
t
to
th
e
test
[
1
]
.
T
h
e
P
C
A
m
et
h
o
d
co
m
e
s
o
u
t
to
b
e
m
o
r
e
e
f
f
icie
n
t
co
m
p
a
r
ed
to
th
e
G
L
C
M
m
et
h
o
d
b
y
9
8
.
4
6
%
ac
cu
r
ac
y
.
B
u
t
th
e
ca
lcu
la
tio
n
ti
m
e
i
n
t
h
e
P
C
A
m
et
h
o
d
is
ti
m
e
-
co
n
s
u
m
i
n
g
f
o
r
e
x
a
m
p
le
m
ak
i
n
g
t
h
e
E
ig
en
v
ec
to
r
f
r
o
m
co
n
s
id
er
ed
leav
es
d
ataset
al
m
o
s
t
to
o
k
2
h
o
u
r
s
.
Ho
w
e
v
er
T
h
e
ad
v
an
tag
e
o
f
th
e
m
et
h
o
d
GL
C
M
s
p
ee
d
i
m
a
g
e
r
ec
o
g
n
iti
o
n
in
j
u
s
t 5
s
ec
o
n
d
s
a
n
d
w
ea
k
n
es
s
es G
L
C
M
i
s
v
e
r
y
s
e
n
s
iti
v
e
to
an
y
c
h
an
g
es
f
o
r
i
m
a
g
es s
u
c
h
as
d
ef
o
r
m
in
g
o
r
g
iv
i
n
g
t
h
e
n
e
w
leaf
i
m
a
g
e
as a
t
est.
I
n
p
r
ev
io
u
s
r
esear
ch
,
i
m
a
g
e
-
b
ased
r
etin
a
r
ec
o
g
n
itio
n
u
s
i
n
g
th
e
G
L
C
M
ch
ar
ac
ter
izati
o
n
(
ASM,
C
o
n
tr
ast,
I
DM
,
a
n
d
E
n
tr
o
p
y
)
an
d
th
e
E
u
clid
ea
n
d
is
ta
n
ce
t
h
a
t
p
r
o
d
u
ce
r
etin
al
i
m
a
g
e
r
ec
o
g
n
itio
n
ac
c
u
r
ac
y
r
ate
o
f
6
7
.
7
1
%.
T
h
e
u
s
e
o
f
m
ea
n
an
d
s
tan
d
ar
d
d
ev
ia
tio
n
as
t
h
e
E
u
clid
ea
n
d
is
ta
n
ce
th
r
e
s
h
o
l
d
v
alu
e
in
te
s
ti
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
Herb
Lea
ve
s
R
ec
o
g
n
itio
n
u
s
in
g
Gra
y
Leve
l Co
-
o
cc
u
r
r
en
ce
Ma
tr
ix
a
n
d
F
ive
Di
s
ta
n
ce
-
…
(
R
.
R
iz
a
l I
s
n
a
n
to
)
1921
ag
ain
s
t
th
e
e
x
ter
n
al
i
m
ag
e
t
h
a
t
is
n
o
t
co
n
tain
ed
in
t
h
e
d
atab
ase
g
en
er
ate
s
a
lo
w
lev
el
o
f
r
ec
o
g
n
itio
n
ac
cu
r
ac
y
.
T
h
e
d
if
f
er
en
ce
i
n
r
ates
d
u
e
to
th
e
r
es
u
lti
n
g
ac
c
u
r
ac
y
o
f
ca
lc
u
latio
n
s
u
s
i
n
g
GL
C
M
p
ar
a
m
et
er
s
g
r
ea
tl
y
a
f
f
ec
ted
in
i
m
ag
e
p
i
x
el
p
o
s
itio
n
[
2
]
.
T
h
e
ex
p
er
i
m
en
tal
r
es
u
lt
s
s
h
o
w
ed
t
h
at
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
f
ea
t
u
r
es
a
co
m
b
in
at
io
n
o
f
Ga
b
o
r
w
a
v
elet
an
d
Gr
ay
-
L
e
v
el
C
o
-
o
cc
u
r
r
en
c
e
Ma
tr
ix
(
GL
C
M)
an
d
P
r
in
ci
p
al
C
o
m
p
o
n
e
n
t
An
al
y
s
i
s
(
P
C
A
)
alg
o
r
it
h
m
s
g
i
v
es
th
e
b
etter
i
m
a
g
e
cla
s
s
i
f
icatio
n
leav
es
[
3
]
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
ca
n
b
e
u
s
ed
to
in
teg
r
ate
s
i
m
p
le
an
d
co
m
p
le
x
leav
es
f
o
r
p
lan
t id
en
ti
f
icat
io
n
.
GL
C
M
u
s
i
n
g
SVM
f
o
r
lin
ea
r
an
d
q
u
ad
r
at
ic
k
er
n
e
l
f
u
n
ctio
n
s
g
i
v
es
cla
s
s
i
f
icatio
n
ac
cu
r
ac
y
o
f
9
6
.
1
5
%
an
d
1
0
0
%
w
it
h
th
e
ex
ec
u
ti
o
n
ti
m
e
o
f
1
0
.
4
8
4
8
s
ec
o
n
d
s
a
n
d
1
0
.
3
8
3
7
s
ec
o
n
d
s
r
es
p
ec
t
iv
el
y
.
Me
d
ical
p
r
ac
titi
o
n
er
s
b
u
r
d
en
w
ill
b
e
r
ed
u
ce
d
b
y
th
is
m
eth
o
d
w
h
e
n
a
lar
g
e
a
m
o
u
n
t
o
f
d
ata
av
aila
b
le.
Me
an
w
h
ile
,
t
h
e
ac
cu
r
ac
y
is
5
7
.
6
9
%
w
h
e
n
P
C
A
i
s
i
m
p
le
m
e
n
ted
wi
t
h
th
e
s
a
m
e
k
er
n
el
f
u
n
ctio
n
[
4
]
.
GL
C
M
C
h
ar
ac
ter
izat
io
n
co
m
b
in
ed
w
it
h
S
n
ap
Sh
o
t
Me
th
o
d
(
SS
M)
o
f
E
ig
e
n
v
alu
e
s
u
s
in
g
P
r
in
cip
al
C
o
m
p
o
n
e
n
t
An
al
y
s
i
s
(
P
C
A
)
h
as
b
ee
n
u
s
ed
o
n
t
h
e
f
ac
e
id
en
ti
f
icatio
n
s
y
s
te
m
.
G
L
C
M
p
ar
a
m
eter
s
u
s
ed
ar
e
E
n
er
g
y
,
E
n
tr
o
p
y
,
C
o
n
tr
ast
an
d
I
n
v
er
s
e
Di
f
f
er
e
n
ce
.
S
n
ap
Sh
o
t
Me
th
o
d
is
u
s
ed
to
s
elec
t
10
(
ten
)
to
p
im
ag
e
b
ased
o
n
th
e
class
if
ica
tio
n
u
s
in
g
E
u
clid
ea
n
D
is
ta
n
ce
o
n
t
h
e
i
m
ag
e
o
f
th
e
e
x
tr
ac
tio
n
u
s
i
n
g
G
L
C
M.
W
ith
a
n
ac
cu
r
ac
y
o
f
9
8
.
6
%
,
th
i
s
m
et
h
o
d
is
s
u
p
er
io
r
co
m
p
ar
ed
to
th
e
m
et
h
o
d
u
s
i
n
g
W
av
elet
+
P
C
A
=
9
4
.
5
%
o
r
C
u
r
v
elet
+
P
C
A
=
9
6
.
6
% [
5
]
.
Hash
i
n
g
m
e
th
o
d
o
f
f
ea
tu
r
e
e
x
tr
ac
tio
n
w
it
h
G
L
C
M
w
a
s
also
u
s
ed
to
id
en
ti
f
y
t
h
e
ir
is
.
B
y
u
s
i
n
g
th
e
p
ar
am
eter
s
o
f
co
n
tr
as
t,
co
r
r
elatio
n
,
e
n
er
g
y
,
en
tr
o
p
y
,
an
d
h
o
m
o
g
e
n
e
it
y
,
it
is
o
b
tain
ed
t
h
at
th
e
r
ec
o
g
n
itio
n
r
ate
f
o
r
GL
C
M
an
g
le
o
f
0
°,
9
0
°,
4
5
°
an
d
1
3
5
°
ar
e
7
0
.
4
9
%,
5
1
.
4
3
%,
7
6
.
2
4
%,
an
d
7
4
.
8
3
%,
r
e
s
p
ec
tiv
el
y
.
T
h
e
u
s
e
GL
C
M
a
n
g
le
(
0
°
+
4
5
°
+
9
0
°
+
1
3
5
°)
h
av
e
a
h
ig
h
er
le
v
el
o
f
r
ec
o
g
n
it
io
n
i
n
th
e
a
m
o
u
n
t
o
f
8
4
.
1
6
%
co
m
p
ar
ed
to
th
e
u
s
e
o
f
o
n
e
G
L
C
M
a
n
g
le
an
d
GL
C
M
an
g
le
(
4
5
° +
1
3
5
°
)
[
6
]
.
Dig
ital
I
m
a
g
e
P
r
o
ce
s
s
i
n
g
i
m
p
l
e
m
en
tatio
n
w
as
u
s
ed
in
t
h
e
le
af
r
ec
o
g
n
itio
n
p
r
ep
r
o
ce
s
s
in
g
.
T
h
e
i
m
a
g
e
th
at
i
s
m
ea
n
t
h
er
e
is
a
s
till
i
m
ag
e
(
p
h
o
to
)
an
d
th
e
m
o
v
i
n
g
i
m
ag
e
(
v
id
eo
)
co
m
in
g
f
r
o
m
a
w
eb
ca
m
er
a.
W
h
ile
d
ig
ital h
er
e
m
ea
n
s
t
h
at
th
e
i
m
ag
e
p
r
o
ce
s
s
in
g
is
d
o
n
e
d
i
g
itall
y
u
s
in
g
a
co
m
p
u
ter
[
7
]
.
T
h
e
leaf
r
ec
o
g
n
it
io
n
p
r
o
ce
s
s
b
eg
in
s
w
it
h
a
lea
f
i
m
a
g
e
d
ata
p
r
o
ce
s
s
in
g
u
s
in
g
s
e
g
m
e
n
tati
o
n
p
r
o
ce
s
s
w
it
h
t
h
e
in
te
n
s
it
y
n
o
r
m
aliza
ti
o
n
p
r
o
ce
s
s
,
f
ea
tu
r
e
ex
tr
ac
tio
n
w
it
h
Gr
a
y
-
L
e
v
el
C
o
-
o
cc
u
r
r
en
ce
Ma
tr
ix
(
GL
C
M)
an
d
is
id
en
ti
f
ied
u
s
i
n
g
C
h
eb
y
s
h
ev
d
is
ta
n
ce
co
m
p
ar
ed
w
i
th
o
t
h
er
d
is
ta
n
ce
,
i.e
.
C
it
y
-
B
lo
ck
,
Min
k
o
w
s
k
i,
E
u
clid
ea
n
,
as
w
el
l a
s
C
a
n
b
er
r
a
Dis
tan
ce
s
,
s
o
th
e
r
es
u
lti
n
g
th
e
b
est a
m
o
n
g
th
e
f
i
v
e
d
is
ta
n
ce
s
.
I
n
leaf
r
ec
o
g
n
itio
n
s
y
s
te
m
,
t
h
er
e
ar
e
1
7
ty
p
es
o
f
f
ea
tu
r
e
e
x
tr
ac
tio
n
th
at
ar
e
d
o
n
e
co
n
s
i
s
t
o
f
5
b
asic
g
eo
m
etr
ic
f
ea
t
u
r
es
as
w
ell
a
s
1
2
d
ig
ital
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es.
Fi
v
e
b
asic
g
eo
m
etr
y
f
ea
tu
r
e
s
ar
e
th
e
d
ia
m
eter
,
len
g
t
h
,
w
id
t
h
,
o
u
t
s
i
d
e
an
d
leav
e
s
p
er
i
m
eter
[
8
]
,
wh
ile
12
d
ig
i
tal
m
o
r
p
h
o
lo
g
ical
f
ea
t
u
r
es
ar
e
s
m
o
o
th
f
ac
to
r
,
asp
ec
t
r
atio
,
f
o
r
m
f
a
cto
r
,
r
ec
tan
g
u
lar
it
y
,
n
ar
r
o
w
f
ac
to
r
,
th
e
p
er
i
m
eter
r
atio
o
f
t
h
e
d
ia
m
eter
,
t
h
e
p
er
im
e
ter
r
atio
w
it
h
lea
f
’
s
len
g
th
a
n
d
w
id
th
,
a
n
d
5
k
in
d
s
o
f
v
ein
f
ea
t
u
r
e
[
9
]
.
R
es
u
lts
o
f
a
n
e
w
r
esear
c
h
w
i
t
h
a
n
e
w
ap
p
r
o
ac
h
u
s
i
n
g
a
co
m
b
in
at
io
n
o
f
Gr
a
y
-
L
ev
e
l
C
o
-
oc
cu
r
r
en
ce
Ma
tr
ix
,
lacu
n
ar
it
y
w
it
h
S
h
en
f
ea
tu
r
e
s
an
d
B
a
y
esia
n
clas
s
i
f
ier
s
h
o
w
ed
t
h
at
t
h
e
s
y
s
te
m
p
r
o
v
id
es
a
lev
el
o
f
9
7
.
1
9
%
ac
cu
r
ac
y
w
h
e
n
u
s
i
n
g
Flav
ia
d
ataset
[
1
0
]
.
P
C
A
m
e
th
o
d
s
u
cc
es
s
f
u
l
l
y
r
e
co
g
n
izes
9
8
%
to
class
if
y
1
3
t
y
p
es
o
f
p
lan
t
s
w
i
th
6
5
n
e
w
o
r
d
ef
ec
t
lea
v
es
a
s
tes
t
i
m
ag
e
s
o
f
th
e
alg
o
r
it
h
m
i
s
tr
ain
ed
b
y
3
9
0
leav
es.
G
L
C
M
m
et
h
o
d
g
iv
e
s
o
n
l
y
7
8
% a
cc
u
r
ac
y
w
it
h
th
e
s
a
m
e
o
b
j
ec
t [
1
1
]
.
T
h
e
r
ec
o
g
n
itio
n
a
n
d
id
en
ti
f
y
i
n
g
p
la
n
ts
u
s
in
g
s
h
ap
e,
v
e
in
,
c
o
lo
r
,
tex
tu
r
e
f
ea
t
u
r
es,
co
m
b
i
n
ed
w
it
h
t
h
e
ap
p
ar
en
t
m
o
v
e
m
en
t
o
f
Z
er
n
i
k
e.
R
ad
ial
b
as
i
s
p
r
o
b
ab
ilis
tic
n
eu
r
al
n
et
w
o
r
k
(
R
B
F
NN)
h
as
b
ee
n
u
s
ed
as
a
class
i
f
ier
.
T
o
tr
ain
R
B
FNN
u
s
in
g
d
o
u
b
le
s
ta
g
e
tr
ain
i
n
g
alg
o
r
ith
m
th
at
s
ig
n
i
f
ica
n
tl
y
i
m
p
r
o
v
es
th
e
p
er
f
o
r
m
an
c
e
o
f
th
e
clas
s
i
f
ier
.
9
5
.
1
2
% a
cc
u
r
ac
y
r
ate
f
o
r
th
e
s
i
m
u
latio
n
r
es
u
lts
o
n
a
d
ataset
Fla
v
ia
leav
e
s
[
1
2
]
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
M
a
t
er
ia
ls
I
n
t
h
is
r
e
s
ea
r
ch
,
t
h
er
e
ar
e
3
0
t
y
p
e
s
o
f
h
er
b
lea
f
t
h
at
h
ad
b
ee
n
tr
ai
n
ed
an
d
te
s
ted
.
A
ll
leav
e
s
o
b
s
er
v
ed
ar
e
th
e
w
ell
-
k
n
o
w
n
h
er
b
leav
e
s
b
y
I
n
d
o
n
es
ian
p
eo
p
le.
T
h
e
3
0
ty
p
e
s
o
f
h
er
b
lea
f
,
b
o
th
in
b
i
n
o
m
ial
(
s
cien
tific
)
as
w
ell
as
i
n
lo
ca
l
(
I
n
d
o
n
esia
n
)
n
a
m
es
ca
n
b
e
s
h
o
w
n
i
n
T
ab
le
2
.
T
h
e
tab
le
co
m
p
leted
w
it
h
i
m
a
g
es
o
f
a
ll
h
er
b
leaf
o
b
s
er
v
ed
ca
n
b
e
s
ee
n
i
n
A
ttac
h
m
e
n
t.
W
h
ile,
s
i
x
s
a
m
p
l
es
o
f
h
er
b
leaf
i
m
a
g
e
w
it
h
u
n
iq
u
e
tex
t
u
r
e
ca
n
b
e
s
h
o
w
n
in
F
ig
u
r
e
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
3
,
J
u
n
e
201
8
:
1
9
2
0
–
1932
1922
T
ab
le
1
.
L
is
t
o
f
3
0
T
y
p
es
o
f
H
er
b
L
ea
f
Ob
s
er
v
ed
No
B
i
n
o
mi
a
l
(
S
c
i
e
n
t
i
f
i
c
)
n
a
me
L
o
c
a
l
n
a
me
1
P
l
u
c
h
e
a
i
n
d
i
c
a
Be
l
u
n
t
a
s
2
F
i
c
u
s
b
e
n
j
a
mi
n
a
Be
ri
n
g
i
n
3
A
n
r
e
d
e
r
a
c
o
r
d
i
f
o
l
i
a
Bi
n
a
h
o
n
g
4
R
i
c
i
n
u
s
c
o
mm
u
n
i
s
J
a
r
a
k
5
H
e
mi
g
r
a
p
h
i
s
a
l
t
e
r
n
a
t
e
S
a
m
b
a
n
g
D
a
ra
h
6
A
l
p
i
n
a
g
a
l
a
n
g
a
l
L
e
n
g
k
u
a
s
7
S
t
r
o
b
i
l
a
n
t
h
e
s c
r
i
s
p
u
s
K
e
j
i
B
e
l
i
n
g
8
P
o
l
y
sci
a
s sc
u
t
e
l
l
a
r
i
a
Ma
n
g
k
o
k
a
n
9
M
o
r
i
n
d
a
c
i
t
r
i
f
o
l
i
a
Me
n
g
k
u
d
u
10
M
u
ss
a
e
n
d
a
p
u
b
e
sce
n
s
N
u
s
a
I
n
d
a
h
11
S
t
a
c
h
y
t
a
r
p
h
e
t
a
j
a
m
a
i
c
e
n
si
s
Pe
c
u
t
K
u
d
a
12
C
a
r
i
c
a
p
a
p
a
y
a
Pe
p
a
y
a
13
S
y
z
y
g
i
u
m
p
o
l
y
a
n
t
h
u
m
S
a
l
a
m
14
P
i
p
e
r
o
r
n
a
t
u
m
S
i
r
i
h
M
e
ra
h
15
A
n
n
o
n
a
m
u
r
i
c
a
t
a
S
i
rs
a
k
16
P
si
d
i
u
m
g
u
a
j
a
v
a
J
a
m
b
u
Me
ra
h
17
Z
a
mi
o
c
u
l
c
a
s
z
a
mi
f
o
l
i
a
D
o
l
a
r
18
I
x
o
r
a
j
a
v
a
n
i
c
a
S
o
k
a
19
L
o
b
e
l
i
a
i
n
f
l
a
t
a
L
o
b
e
l
i
a
20
P
i
p
e
r
b
e
t
l
e
S
i
r
i
h
H
i
j
a
u
21
C
i
t
r
u
s
h
y
st
r
i
x
J
e
r
u
k
Pu
r
u
t
22
S
a
u
r
o
p
u
s
a
n
d
r
o
g
y
n
u
s
K
a
t
u
k
23
C
l
i
n
a
c
a
n
t
h
u
s
n
u
t
a
n
s L
i
n
d
a
u
D
a
n
d
a
n
g
G
e
n
d
is
24
C
u
r
c
u
m
a
l
o
n
g
a
K
u
n
i
r
25
C
a
t
h
a
r
a
n
t
h
u
s
r
o
se
u
s
I
T
a
p
a
k
D
a
r
a
Me
r
a
h
26
A
v
e
r
r
h
o
a
b
i
l
i
mb
i
Be
l
i
m
b
i
n
g
Wu
l
u
h
27
A
n
n
o
n
a
s
q
u
a
mo
sa
S
ri
k
a
y
a
28
C
a
t
h
a
r
a
n
t
h
u
s
r
o
se
u
s
I
I
T
a
p
a
k
D
a
r
a
P
u
t
i
h
29
B
o
e
se
n
b
e
r
g
i
a
r
o
t
u
n
d
a
K
u
n
c
i
30
P
a
n
d
a
n
u
s
a
m
a
r
y
l
l
i
f
o
l
i
u
s
Pa
n
d
a
n
(
a
)
R
i
c
i
n
u
s
c
o
mm
u
n
i
s
(
b
)
S
t
a
c
h
y
t
a
r
p
h
e
t
a
j
a
ma
i
c
e
n
s
i
s
(
c
)
C
a
r
i
c
a
p
a
p
a
y
a
(
d
)
P
i
p
e
r
o
r
n
a
t
u
m
(
e
)
O
b
e
l
i
a
i
n
f
l
a
t
a
(
f
)
C
a
t
h
a
r
a
n
t
h
u
s
r
o
se
u
s
I
I
Fig
u
r
e
1
.
Six
s
a
m
p
les o
f
h
er
b
leaf
o
b
s
er
v
ed
T
h
er
e
ar
e
3
(
th
r
ee
)
s
ce
n
ar
io
s
in
t
h
is
r
esear
c
h
to
o
b
s
er
v
e
t
h
e
h
er
b
leaf
r
ec
o
g
n
itio
n
r
ate,
i.e
.
u
s
i
n
g
1
0
,
2
0
,
an
d
3
0
t
y
p
es
o
f
lea
f
r
esp
ec
ti
v
el
y
.
W
h
e
n
w
e
u
s
e
1
0
an
d
2
0
t
y
p
e
s
o
f
lea
v
e,
t
h
e
n
th
e
t
y
p
es
l
ea
f
u
s
ed
in
tr
ain
i
n
g
an
d
test
i
n
g
p
r
o
ce
s
s
ar
e
lis
ted
i
n
T
ab
le
2
.
T
ab
le
2
.
L
is
t o
f
1
0
an
d
2
0
T
y
p
es o
f
L
ea
f
u
s
ed
in
T
esti
n
g
L
i
st
o
f
1
0
t
y
p
e
s o
f
l
e
a
f
u
se
d
i
n
t
e
st
i
n
g
No
B
i
n
o
mi
a
l
(
S
c
i
e
n
t
i
f
i
c
)
n
a
me
L
o
c
a
l
n
a
me
1
P
l
u
c
h
e
a
i
n
d
i
c
a
Be
l
u
n
t
a
s
2
F
i
c
u
s
b
e
n
j
a
mi
n
a
Be
ri
n
g
i
n
3
A
n
r
e
d
e
r
a
c
o
r
d
i
f
o
l
i
a
Bi
n
a
h
o
n
g
4
R
i
c
i
n
u
s
c
o
mm
u
n
i
s
J
a
r
a
k
5
S
t
r
o
b
i
l
a
n
t
h
e
s c
r
i
s
p
u
s
K
e
j
i
B
e
l
i
n
g
6
A
l
p
i
n
a
g
a
l
a
n
g
a
l
L
e
n
g
k
u
a
s
7
P
o
l
y
sci
a
s sc
u
t
e
l
l
a
r
i
a
Ma
n
g
k
o
k
a
n
8
M
o
r
i
n
d
a
c
i
t
r
i
f
o
l
i
a
M
e
n
g
k
u
d
u
9
M
u
ss
a
e
n
d
a
p
u
b
e
sce
n
s
N
u
s
a
I
n
d
a
h
10
S
t
a
c
h
y
t
a
r
p
h
e
t
a
j
a
m
a
i
c
e
n
si
s
Pe
c
u
t
K
u
d
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
Herb
Lea
ve
s
R
ec
o
g
n
itio
n
u
s
in
g
Gra
y
Leve
l Co
-
o
cc
u
r
r
en
ce
Ma
tr
ix
a
n
d
F
ive
Di
s
ta
n
ce
-
…
(
R
.
R
iz
a
l I
s
n
a
n
to
)
1923
L
i
st
o
f
2
0
t
y
p
e
s o
f
l
e
a
f
u
se
d
i
n
t
e
st
i
n
g
No
B
i
n
o
mi
a
l
(
S
c
i
e
n
t
i
f
i
c
)
n
a
me
L
o
c
a
l
n
a
me
1
1
P
l
u
c
h
e
a
i
n
d
i
c
a
Be
l
u
n
t
a
s
1
2
R
i
c
i
n
u
s
c
o
mm
u
n
i
s
J
a
r
a
k
1
3
A
l
p
i
n
a
g
a
l
a
n
g
a
l
L
e
n
g
k
u
a
s
1
4
P
o
l
y
sci
a
s sc
u
t
e
l
l
a
r
i
a
Ma
n
g
k
o
k
a
n
1
5
M
o
r
i
n
d
a
c
i
t
r
i
f
o
l
i
a
Me
n
g
k
u
d
u
1
6
M
u
ss
a
e
n
d
a
p
u
b
e
sce
n
s
N
u
s
a
I
n
d
a
h
1
7
S
t
a
c
h
y
t
a
r
p
h
e
t
a
j
a
m
a
i
c
e
n
si
s
Pe
c
u
t
K
u
d
a
1
8
C
a
r
i
c
a
p
a
p
a
y
a
Pe
p
a
y
a
1
9
H
e
mi
g
r
a
p
h
i
s
a
l
t
e
r
n
a
t
e
S
a
m
b
a
n
g
D
a
ra
h
2
0
P
i
p
e
r
o
r
n
a
t
u
m
S
i
r
i
h
M
e
ra
h
21
A
n
n
o
n
a
m
u
r
i
c
a
t
a
S
i
rs
a
k
22
P
si
d
i
u
m
g
u
a
j
a
v
a
J
a
m
b
u
Me
ra
h
23
Z
a
mi
o
c
u
l
c
a
s
z
a
mi
f
o
l
i
a
D
o
l
a
r
24
I
x
o
r
a
j
a
v
a
n
i
c
a
S
o
k
a
25
L
o
b
e
l
i
a
i
n
f
l
a
t
a
L
o
b
e
l
i
a
26
P
i
p
e
r
b
e
t
l
e
S
i
r
i
h
H
i
j
a
u
27
C
i
t
r
u
s
h
y
st
r
i
x
J
e
r
u
k
Pu
r
u
t
28
S
a
u
r
o
p
u
s
a
n
d
r
o
g
y
n
u
s
K
a
t
u
k
29
C
l
i
n
a
c
a
n
t
h
u
s
n
u
t
a
n
s L
i
n
d
a
u
D
a
n
d
a
n
g
G
e
n
d
i
s
30
C
a
t
h
a
r
a
n
t
h
u
s
r
o
se
u
s
I
I
T
a
p
a
k
D
a
r
a
P
u
t
i
h
2
.
2.
Sy
s
t
e
m
des
ig
n
I
n
th
is
h
er
b
id
en
ti
f
icatio
n
s
y
s
te
m
,
t
h
er
e
ar
e
t
w
o
m
ai
n
p
r
o
ce
s
s
es
i.e
.
tr
ain
i
n
g
a
n
d
r
ec
o
g
n
itio
n
p
r
o
ce
s
s
es
.
T
h
e
f
lo
w
d
iag
r
a
m
o
f
th
e
id
en
t
if
icatio
n
s
y
s
te
m
p
r
o
ce
s
s
,
in
g
e
n
er
al,
ca
n
b
e
ill
u
s
tr
ated
in
Fi
g
u
r
e
2
.
T
h
e
f
ig
u
r
e
s
h
o
w
s
th
at
id
e
n
ti
f
i
ca
tio
n
s
y
s
te
m
co
n
s
is
t
s
o
f
t
w
o
m
ain
p
r
o
ce
s
s
e
s
n
a
m
el
y
tr
ain
i
n
g
a
n
d
r
ec
o
g
n
it
io
n
p
r
o
ce
s
s
es.
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
s
er
v
e
s
to
r
eg
is
ter
an
d
s
to
r
e
th
e
i
m
a
g
e
f
ea
t
u
r
es
o
f
th
e
lea
v
es
i
n
d
atab
ase
th
at
w
il
l
b
e
u
s
ed
as
a
co
m
p
ar
is
o
n
to
d
eter
m
i
n
e
w
h
et
h
er
th
e
i
m
ag
e
o
f
leav
es
r
ec
o
g
n
izab
le
o
r
n
o
t
o
n
th
e
r
ec
o
g
n
itio
n
p
r
o
ce
s
s
.
W
h
ile
i
n
t
h
e
r
ec
o
g
n
it
io
n
p
r
o
c
ess
,
t
h
e
f
ea
tu
r
e
s
f
r
o
m
th
e
tr
ai
n
i
n
g
p
r
o
ce
s
s
ar
e
co
m
p
ar
ed
w
ith
t
h
e
tes
t
i
m
a
g
es
to
o
b
tai
n
t
h
e
r
ec
o
g
n
iti
o
n
r
esu
lts
.
T
h
e
m
et
h
o
d
s
o
f
i
m
ag
e
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
e
x
t
r
ac
tio
n
,
as
w
ell
as
5
(
f
iv
e)
d
is
ta
n
ce
-
b
ased
s
i
m
ilar
it
y
m
ea
s
u
r
e
s
w
ill b
e
ex
p
lai
n
ed
in
3
(
th
r
ee
)
s
ep
ar
ate
s
ec
tio
n
s
.
Fig
u
r
e
2.
Flo
w
c
h
ar
t o
f
s
y
s
te
m
2
.
2.
I
m
a
g
e
prepro
ce
s
s
i
ng
T
h
e
f
ir
s
t
p
r
o
ce
s
s
i
n
i
m
a
g
e
p
r
ep
r
o
ce
s
s
in
g
is
c
o
n
v
er
s
io
n
o
f
co
lo
r
im
a
g
es
in
to
g
r
a
y
-
le
v
el
i
m
a
g
es
to
m
ak
e
t
h
e
p
r
o
g
r
a
m
m
i
n
g
p
r
o
ce
s
s
ea
s
ier
b
ec
au
s
e
th
e
n
u
m
b
er
o
f
b
its
in
t
h
e
g
r
a
y
-
s
ca
le
i
m
a
g
e
(8
-
b
it)
is
less
t
h
an
th
e
n
u
m
b
er
o
f
b
it
s
in
t
h
e
R
GB
i
m
ag
e
(
2
4
-
b
it)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
3
,
J
u
n
e
201
8
:
1
9
2
0
–
1932
1924
T
h
e
f
o
llo
w
i
n
g
eq
u
atio
n
i
s
u
s
e
d
f
o
r
co
n
v
er
tin
g
t
h
e
R
GB
i
m
a
g
e
in
to
a
g
r
a
y
-
s
ca
le
i
m
a
g
e
[
1
3
]
.
Gr
a
y
-
s
ca
le
=
0
,
2
9
9
*
R
+
0
,
5
8
7
*
G
+
0
,
1
1
4
*
B
(
1
)
W
h
er
e
:
Gr
a
y
-
s
ca
le:
v
alu
e
o
f
lev
el
s
o
f
g
r
a
y
a
f
ter
co
n
v
er
s
io
n
R
:
m
a
tr
ix
v
alu
e
o
f
r
ed
co
m
p
o
n
en
t
G
:
m
atr
i
x
v
al
u
e
o
f
g
r
ee
n
co
m
p
o
n
en
t
B
: m
a
tr
ix
v
alu
e
o
f
b
lu
e
co
m
p
o
n
en
t
T
h
e
s
ec
o
n
d
p
r
o
ce
s
s
i
n
i
m
a
g
e
p
r
ep
r
o
ce
s
s
in
g
is
i
m
a
g
e
s
e
g
m
e
n
tat
io
n
to
s
ep
ar
ate
o
b
j
ec
t
f
r
o
m
t
h
e
b
ac
k
g
r
o
u
n
d
.
W
h
e
n
t
h
e
i
m
a
g
e
is
to
o
b
r
ig
h
t
a
s
w
el
l
as
to
o
d
ar
k
,
th
e
h
i
s
to
g
r
a
m
eq
u
al
izati
o
n
is
r
eq
u
ir
ed
.
T
h
e
h
is
to
g
r
a
m
eq
u
al
izatio
n
is
to
o
b
tain
a
h
is
to
g
r
a
m
b
y
eq
u
aliz
in
g
th
e
g
r
a
y
s
ca
le
i
n
te
n
s
it
y
v
al
u
e
s
in
an
i
m
ag
e.
T
h
e
o
b
j
ec
tiv
e
i
s
to
o
b
tai
n
a
d
is
tr
ib
u
tio
n
o
f
t
h
e
h
is
t
o
g
r
a
m
w
it
h
eq
u
alize
d
i
n
te
n
s
it
y
s
o
t
h
at
ea
ch
d
eg
r
ee
o
f
g
r
a
y
h
as
a
n
u
m
b
er
o
f
p
ix
el
s
th
a
t a
r
e
r
e
lativ
el
y
eq
u
al.
T
h
e
eq
u
atio
n
f
o
r
ca
lcu
lati
n
g
h
is
to
g
r
a
m
eq
u
aliza
tio
n
o
n
an
i
m
ag
e
w
i
th
k
-
b
its
g
r
a
y
-
s
ca
le
is
:
(
(
)
)
(
2
)
W
h
er
e:
C
i
:
cu
m
u
la
tiv
e
d
i
s
tr
ib
u
tio
n
o
f
i
t
h
g
r
a
y
-
s
ca
le
v
al
u
e
o
f
t
h
e
o
r
ig
i
n
al
i
m
ag
e
K
o
:
g
r
a
y
-
s
ca
le
as a
r
es
u
lt o
f
h
i
s
to
g
r
a
m
eq
u
aliza
tio
n
w
:
i
m
a
g
e
w
id
t
h
h
:
im
a
g
e
h
eig
h
t
2
.
3
.
F
ea
t
ure
ex
t
ra
ct
io
n
Gr
a
y
L
e
v
el
C
o
o
cu
r
r
en
ce
Ma
t
r
ix
(
GL
C
M
)
is
u
s
ed
to
ex
tr
ac
t
th
e
f
ea
tu
r
es
o
f
h
er
b
leaf
i
m
ag
es.
T
h
e
r
esu
lt
o
f
G
L
C
M
is
s
o
m
e
p
air
o
f
p
ix
els
t
h
at
ea
c
h
ha
s
a
ce
r
tain
in
te
n
s
it
y
v
al
u
e.
T
h
e
p
ix
el
p
air
p
atter
n
is
s
p
ac
ed
b
y
d
is
tan
ce
d
,
an
d
d
ir
ec
tio
n
θ
.
T
h
e
d
is
tan
ce
is
ex
p
r
ess
ed
i
n
p
ix
els
an
d
t
h
e
o
r
ien
tatio
n
an
g
le
is
e
x
p
r
ess
ed
in
d
eg
r
ee
s
.
A
d
j
ac
en
c
y
o
f
p
ix
el
s
i
n
GL
C
M
f
ea
tu
r
e
ex
tr
ac
tio
n
m
eth
o
d
ca
n
b
e
illu
s
tr
ated
in
a
f
o
u
r
-
d
ir
ec
tio
n
s
w
i
th
in
ter
v
a
ls
o
f
4
5
°,
i.e
.
0
°,
4
5
°,
9
0
°,
an
d
1
3
5
°.
So
m
e
s
tep
s
in
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
w
it
h
G
L
C
M
m
et
h
o
d
ar
e
Gr
ay
-
le
v
el
p
i
x
el
q
u
a
n
tizatio
n
to
a
m
atr
i
x
,
to
cr
ea
te
a
GL
C
M
m
atr
i
x
w
o
r
k
in
g
ar
ea
,
to
d
eter
m
i
n
e
th
e
s
p
atial
r
elatio
n
s
h
ip
b
et
w
ee
n
th
e
r
ef
er
en
ce
p
ix
el
an
d
n
eig
h
b
o
r
in
g
p
ix
e
l f
o
r
t
h
e
v
alu
e
d
an
d
th
e
a
n
g
le
θ
,
to
co
m
p
u
t
e
th
e
n
u
m
b
er
o
f
p
air
s
o
f
p
ix
e
ls
th
at
h
a
v
e
th
e
s
a
m
e
in
te
n
s
it
y
a
n
d
in
co
r
p
o
r
at
in
g
t
h
e
m
i
n
to
th
e
G
L
C
M
m
a
tr
ix
w
o
r
k
in
g
ar
ea
,
r
esu
lt
in
g
i
n
a
co
-
o
cc
u
r
r
en
ce
m
a
tr
ix
,
to
co
n
v
er
t
t
h
e
co
-
o
cc
u
r
r
en
ce
m
a
tr
ix
i
n
to
s
y
m
m
etr
ic
al
m
atr
i
x
b
y
ad
d
in
g
t
h
e
m
atr
i
x
to
it
s
tr
a
n
s
p
o
s
e,
an
d
f
i
n
all
y
,
to
n
o
r
m
aliz
e
t
h
e
GL
C
M
s
y
m
m
etr
ical
m
atr
ix
i
n
to
a
f
o
r
m
o
f
p
r
o
b
a
b
ilit
y
n
u
m
b
er
s
.
T
h
er
e
ar
e
f
iv
e
GL
C
M
p
ar
a
m
et
er
s
to
ca
lcu
late
th
e
s
ec
o
n
d
o
r
d
er
s
t
atis
tical
c
h
ar
ac
ter
is
tic
s
o
f
i
m
a
g
e:
a.
ASM
(
An
g
u
lar
Seco
n
d
Mo
m
e
n
t)
ASM
o
r
en
er
g
y
is
u
s
ed
to
m
e
asu
r
e
th
e
co
n
ce
n
tr
atio
n
o
f
p
ai
r
s
o
f
p
ix
els
w
ith
p
ar
ticu
lar
g
r
a
y
i
n
te
n
s
i
t
y
in
th
e
m
atr
i
x
G
L
C
M.
ASM
v
alu
e
w
o
u
ld
b
e
g
r
ea
ter
if
t
h
e
v
ar
iatio
n
i
n
t
h
e
in
te
n
s
it
y
o
f
t
h
e
i
m
a
g
e
d
ec
r
ea
s
es.
Fu
n
ctio
n
to
ca
lc
u
late
A
SM
i
s
s
h
o
w
n
b
y
th
e
f
o
llo
w
i
n
g
eq
u
ati
o
n
:
∑
∑
(
(
)
)
(
3
)
b.
C
o
n
tr
as
t
C
o
n
tr
ast
i
s
a
f
ea
t
u
r
e
th
at
is
u
s
ed
to
m
ea
s
u
r
e
th
e
d
if
f
er
e
n
ce
in
in
te
n
s
it
y
o
r
v
ar
iatio
n
s
o
f
g
r
a
y
p
i
x
els
i
n
th
e
i
m
ag
e.
T
h
e
f
o
llo
w
i
n
g
eq
u
a
tio
n
is
u
s
ed
to
m
ea
s
u
r
e
th
e
co
n
tr
ast o
f
a
n
i
m
ag
e.
∑
{
∑
(
)
|
|
}
(
4
)
c.
I
DM
(
I
n
v
er
s
e
Di
f
f
er
en
t M
o
m
e
n
t)
I
DM
r
ep
r
esen
ts
a
lo
ca
l
h
o
m
o
g
en
eit
y
i
n
th
e
i
m
ag
e
th
a
t
h
as
s
i
m
ilar
s
h
ad
es
o
f
g
r
a
y
i
n
t
h
e
co
-
o
cc
u
r
r
en
ce
m
atr
i
x
.
I
DM
v
a
lu
e
w
il
l
b
e
g
r
ea
ter
w
h
e
n
co
u
p
le
s
o
f
p
ix
el
s
t
h
at
h
a
v
e
t
h
e
eli
g
i
b
le
in
ten
s
it
y
o
f
co
-
o
cc
u
r
r
en
ce
m
atr
i
x
ar
e
c
o
n
ce
n
t
r
ated
in
a
f
e
w
co
o
r
d
in
ates a
n
d
w
il
l s
h
r
in
k
w
h
en
s
p
r
ea
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
Herb
Lea
ve
s
R
ec
o
g
n
itio
n
u
s
in
g
Gra
y
Leve
l Co
-
o
cc
u
r
r
en
ce
Ma
tr
ix
a
n
d
F
ive
Di
s
ta
n
ce
-
…
(
R
.
R
iz
a
l I
s
n
a
n
to
)
1925
∑
∑
(
(
)
)
(
)
(
5
)
d.
E
n
tr
o
p
y
E
n
tr
o
p
y
o
n
G
L
C
M
m
ea
s
u
r
e
s
t
h
e
d
is
o
r
d
er
d
is
tr
ib
u
tio
n
o
f
g
r
a
y
lev
e
ls
o
f
a
n
i
m
a
g
e
o
n
a
co
-
o
cc
u
r
r
en
ce
m
atr
i
x
.
E
n
tr
o
p
y
is
h
ig
h
w
h
en
t
h
e
e
le
m
e
n
t
s
o
f
G
L
C
M
h
a
v
e
r
elativ
el
y
eq
u
al
v
al
u
es.
∑
∑
(
(
)
(
(
)
)
)
(
6
)
e.
C
o
r
r
ela
t
i
on
C
o
r
r
elatio
n
is
a
m
ea
s
u
r
e
o
f
lin
ea
r
d
ep
en
d
en
ce
b
et
w
ee
n
t
h
e
v
al
u
es
o
f
g
r
a
y
lev
e
ls
in
t
h
e
i
m
a
g
e.
C
o
r
r
elatio
n
f
u
n
c
tio
n
ca
n
b
e
s
ee
n
in
t
h
e
eq
u
atio
n
:
∑
∑
(
)
(
(
)
)
(
7
)
2
.
4
.
Si
m
ila
rit
y
m
ea
s
ure
s
us
i
ng
5
(
f
iv
e)
dis
t
a
nces
Af
ter
t
h
e
f
ea
t
u
r
e
e
x
tr
ac
tio
n
p
r
o
ce
s
s
,
s
i
m
ilar
it
y
test
s
b
et
w
e
en
te
s
ti
n
g
i
m
ag
e
f
ea
tu
r
e
s
a
n
d
r
eg
is
ter
ed
i
m
a
g
e
f
ea
tu
r
e
s
.
T
h
er
e
ar
e
5
(
f
iv
e)
d
i
s
tan
ce
s
i
m
p
le
m
en
ted
,
th
o
s
e
ar
e
C
h
eb
y
s
h
ev
,
C
it
y
b
lo
ck
,
Min
k
o
w
s
k
i,
C
an
b
er
r
a,
an
d
E
u
clid
ea
n
d
i
s
ta
n
ce
s
.
C
it
y
-
b
lo
ck
d
i
s
tan
ce
i
s
d
ef
i
n
ed
as f
o
llo
w
s
:
(
8
)
W
h
er
e
v
1
an
d
v
2
ar
e
t
w
o
v
ec
to
r
s
w
h
o
s
e
d
i
s
ta
n
ce
s
w
i
ll
b
e
ca
l
cu
lated
an
d
N
d
en
o
te
s
t
h
e
le
n
g
th
o
f
t
h
e
v
ec
to
r
.
I
f
th
e
v
ec
to
r
h
a
s
t
w
o
v
al
u
es,
cit
y
-
b
lo
ck
d
i
s
tan
ce
ca
n
b
e
i
m
ag
i
n
ed
as
a
h
o
r
izo
n
tal
p
l
u
s
v
er
tica
l
d
is
ta
n
ce
f
r
o
m
t
h
e
f
ir
s
t
v
ec
to
r
to
th
e
s
ec
o
n
d
v
ec
t
o
r
,
w
h
ic
h
is
ill
u
s
t
r
ated
in
Fi
g
u
r
e
3
(
a)
.
Dis
ta
n
ce
b
o
x
ch
es
s
o
r
also
k
n
o
w
n
b
y
th
e
n
a
m
e
o
f
C
h
eb
y
s
h
e
v
d
is
ta
n
ce
is
d
ef
i
n
ed
as
f
o
llo
w
s
.
(
)
(
|
(
)
-
(
)
|
)
(9
)
T
h
e
eq
u
atio
n
ab
o
v
e
ill
u
s
tr
ate
s
th
at
v
1
an
d
v
2
ar
e
t
w
o
v
ec
to
r
s
t
h
at
d
is
ta
n
ce
w
ill
b
e
ca
lcu
la
ted
an
d
N
d
en
o
tes
t
h
e
le
n
g
th
o
f
th
e
v
ec
t
o
r
.
I
f
th
e
v
ec
to
r
h
as
t
w
o
v
alu
es,
th
e
d
is
ta
n
ce
ca
n
b
e
e
n
v
is
i
o
n
ed
as
t
h
e
lo
n
g
e
s
t
d
is
tan
ce
b
et
w
ee
n
th
e
h
o
r
izo
n
t
al
d
is
tan
ce
a
n
d
v
er
tical
d
is
ta
n
c
e
,
w
h
ich
i
s
d
ep
icted
in
Fi
g
ur
e
3
(
b
)
.
(
a)
C
it
y
-
b
lo
ck
Dis
tan
ce
il
lu
s
tr
atio
n
(
b
)
C
h
eb
y
s
h
ev
D
is
ta
n
ce
ill
u
s
tr
atio
n
Fig
u
r
e
3
.
C
o
m
p
ar
is
o
n
b
et
w
ee
n
C
it
y
-
b
lo
ck
an
d
C
h
eb
y
s
h
ev
d
is
tan
ce
s
Min
k
o
w
s
k
i d
is
ta
n
ce
is
d
ef
i
n
ed
as f
o
llo
w
s
.
(
1
0
)
I
f
th
e
v
al
u
e
p
=
1
,
th
en
t
h
e
eq
u
atio
n
b
ec
o
m
es
t
h
e
C
it
y
-
b
lo
ck
d
is
tan
ce
.
Fo
r
th
e
v
al
u
e
p
=
2
,
it
d
(
v
1
,
v
2
)
=
K
=
1
N
å
v
1
(
k
)
-
v
2
(
k
)
d
(
v
1
,
v
2
)
=
k
=
1
N
å
v
1
(
k
)
-
v
2
(
k
)
p
p
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
3
,
J
u
n
e
201
8
:
1
9
2
0
–
1932
1926
b
ec
o
m
e
s
th
e
E
u
clid
ea
n
d
is
ta
n
ce
,
w
h
ich
ca
n
b
e
s
tated
as th
e
f
o
r
m
u
la:
(
1
1
)
W
h
ile
C
a
n
b
er
r
a
d
is
tan
ce
ca
n
b
e
ex
p
r
ess
ed
as:
(
1
2
)
R
ef
er
r
i
n
g
to
t
h
e
e
x
p
lan
at
io
n
u
s
es
lea
f
p
atter
n
a
s
it
s
c
h
ar
ac
te
r
is
tic,
t
h
is
r
esear
ch
is
to
d
esi
g
n
a
s
y
s
te
m
th
at
ca
n
r
ec
o
g
n
ize
t
h
e
t
y
p
es
o
f
h
er
b
al
leav
e
s
.
T
h
e
leaf
i
m
a
g
e
r
ec
o
g
n
itio
n
p
r
o
ce
s
s
b
eg
i
n
s
w
it
h
a
n
i
m
a
g
e
d
ata
p
r
o
ce
s
s
in
g
u
s
in
g
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
w
it
h
t
h
e
i
n
te
n
s
it
y
n
o
r
m
aliza
tio
n
p
r
o
ce
s
s
,
f
ea
t
u
r
e
ex
tr
ac
tio
n
w
it
h
Gr
a
y
-
L
e
v
el
C
o
-
o
cc
u
r
r
en
ce
Ma
tr
ix
(
GL
C
M)
m
eth
o
d
an
d
i
ts
id
e
n
ti
f
icatio
n
u
s
i
n
g
d
i
s
ta
n
ce
-
b
ased
s
i
m
ilar
it
y
m
ea
s
u
r
es.
T
h
er
ef
o
r
e,
th
is
r
esear
ch
a
i
m
s
to
co
m
p
ar
e
th
e
r
ec
o
g
n
itio
n
m
eth
o
d
u
s
in
g
d
is
ta
n
ce
-
b
ase
d
s
i
m
ilar
it
y
m
ea
s
u
r
es
,
i.e
.
C
h
eb
y
s
h
e
v
,
C
i
t
y
-
B
lo
ck
,
M
in
k
o
w
s
k
i,
C
a
n
b
er
r
a,
an
d
E
u
clid
ea
n
d
is
tan
ce
s
.
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
3
.
1
.
G
L
CM
f
ea
t
ure
ex
t
ra
ct
io
n
T
h
er
e
ar
e
f
iv
e
ch
ar
ac
ter
is
tic
s
as
th
e
r
es
u
lt
s
o
f
G
L
C
M
f
ea
t
u
r
e
ex
tr
ac
tio
n
,
i.e
.
ASM,
co
n
tr
ast,
I
DM
,
en
tr
o
p
y
,
an
d
co
r
r
elatio
n
.
T
h
es
e
f
i
v
e
p
ar
a
m
eter
s
ca
n
b
e
u
s
ed
as
d
is
tin
g
u
is
h
i
n
g
f
ea
tu
r
e
s
f
o
r
ea
ch
leaf
i
m
ag
e
o
f
m
ed
icin
a
l
p
lan
t
s
.
T
h
e
ex
a
m
p
les
f
r
o
m
G
L
C
M
c
h
ar
ac
ter
is
ti
c
v
al
u
e
o
f
th
e
test
i
m
a
g
es
b
y
t
h
e
t
w
o
t
y
p
es
o
f
leav
es
,
i.e
P
lu
c
h
ea
i
n
d
i
ca
an
d
Ficu
s
b
en
j
a
m
i
n
o
f
m
ed
ici
n
al
p
lan
ts
at
a
d
i
s
tan
ce
o
f
1
p
ix
el
an
d
a
n
o
r
ien
ta
tio
n
an
g
le
o
f
0
° ca
n
b
e
s
ee
n
i
n
T
ab
le
3
.
T
ab
le
3
.
Sam
p
les o
f
GL
C
M
V
alu
e
o
f
T
est I
m
ag
e
at
An
g
le
0
0
an
d
Dis
ta
n
ce
1
P
ix
el
I
mag
e
N
a
me
A
S
M
C
o
n
t
r
a
s
t
I
D
M
En
t
r
o
p
y
C
o
r
r
e
l
a
t
i
on
P
l
u
c
h
e
a
i
n
d
i
c
a
(
1
)
0
.
0
0
1
2
1
2
9
.
6
6
0
2
0
.
2
4
2
6
7
.
9
7
9
5
4
.
2
5
e
-
04
P
l
u
c
h
e
a
i
n
d
i
c
a
(
2
)
0
.
0
0
2
7
1
0
2
.
3
7
3
9
0
.
2
9
2
9
7
.
6
7
5
7
3
.
7
8
e
-
04
F
i
c
u
s
b
e
n
j
a
mi
n
(
1
)
0
.
0
0
3
3
9
5
.
2
2
1
5
0
.
3
6
2
7
.
6
0
9
7
2
.
0
3
e
-
04
F
i
c
u
s
b
e
n
j
a
mi
n
(
2
)
0
.
0
0
2
6
1
2
3
.
8
4
6
2
0
.
3
3
7
7
.
7
6
5
5
1
.
8
9
e
-
04
3
.
2
.
T
ra
ini
ng
i
m
a
g
e
re
co
g
nitio
n
T
r
ain
in
g
i
m
a
g
e
r
ec
o
g
n
itio
n
i
s
co
n
d
u
cted
o
n
th
e
i
m
a
g
e
t
h
at
h
as
p
r
ev
io
u
s
l
y
e
x
is
ted
i
n
d
atab
ase.
T
esti
n
g
w
it
h
t
h
is
tr
ai
n
in
g
i
m
a
g
e
ai
m
s
to
v
alid
ate
w
h
et
h
er
t
h
e
r
esu
l
tin
g
C
h
eb
y
s
h
e
v
d
is
tan
c
e
is
ze
r
o
o
r
n
o
t.
T
h
e
T
ab
le
4
s
h
o
w
s
t
h
r
ee
s
a
m
p
l
es o
f
leaf
a
s
test
i
n
g
r
esu
lts
o
f
tr
ai
n
in
g
i
m
a
g
e.
T
ab
le
4
.
Sam
p
le
s
o
f
T
r
ain
in
g
I
m
ag
e
T
est
T
e
st
D
a
t
a
N
a
me
C
h
e
b
y
sh
e
v
d
i
s
t
a
n
c
e
R
e
c
o
g
n
i
z
e
d
a
s
Ex
p
l
a
n
a
t
i
o
n
R
i
c
i
n
u
s
c
o
mm
u
n
i
s_
0
1
2
0
R
i
c
i
n
u
s
c
o
mm
u
n
i
s
T
r
u
e
S
a
u
r
o
p
u
s
a
n
d
r
o
g
y
n
u
s_
0
6
4
0
S
a
u
r
o
p
u
s
a
n
d
r
o
g
y
n
u
s
T
r
u
e
Z
a
mi
o
c
u
l
c
a
s
z
a
mi
f
o
l
i
a
_
0
5
0
0
Z
a
mi
o
c
u
l
c
a
s
z
a
mi
f
o
l
i
a
T
r
u
e
T
ab
le
4
is
t
h
e
r
es
u
lt
o
f
C
h
eb
y
s
h
ev
d
is
ta
n
ce
i
n
tr
ai
n
i
n
g
i
m
ag
e
tes
tin
g
is
ze
r
o
as
th
e
r
es
u
lt
o
f
t
h
e
d
if
f
er
e
n
ce
b
et
w
ee
n
th
e
f
ea
t
u
r
e
v
ec
to
r
o
f
th
e
tr
ai
n
i
n
g
i
m
ag
e
b
ein
g
tes
ted
an
d
th
e
tr
a
in
i
n
g
i
m
a
g
e
i
n
t
h
e
d
atab
ase.
T
h
u
s
,
it
is
k
n
o
w
n
t
h
at
th
e
r
ec
o
g
n
itio
n
p
r
o
g
r
a
m
w
o
r
k
s
p
r
o
p
er
ly
b
ec
au
s
e
t
h
e
tr
ain
i
m
a
g
e
b
ein
g
te
s
ted
is
e
x
ac
tl
y
t
h
e
s
a
m
e
as
t
h
e
e
x
is
ti
n
g
i
m
ag
e
in
th
e
d
atab
ase.
I
f
t
h
e
tr
ai
n
i
m
a
g
e
te
s
ti
n
g
d
o
es
n
o
t
p
r
o
d
u
ce
th
e
C
h
eb
y
s
h
ev
d
is
tan
ce
o
f
ze
r
o
,
it
m
ea
n
s
th
a
t
th
e
r
ec
o
g
n
itio
n
p
r
o
g
r
a
m
d
o
es n
o
t
w
o
r
k
p
r
o
p
er
ly
.
3
.
3
.
T
est
i
m
a
g
e
re
co
g
nitio
n
T
h
e
test
i
m
a
g
e
r
ec
o
g
n
itio
n
o
f
t
h
e
h
er
b
leav
e
s
i
s
p
er
f
o
r
m
e
d
to
f
i
n
d
th
e
p
er
ce
n
ta
g
e
o
f
s
u
cc
es
s
f
u
l
r
ec
o
g
n
itio
n
f
r
o
m
3
0
t
y
p
es
o
f
leav
es
o
f
m
ed
ici
n
al
p
lan
t
s
.
Fi
v
e
i
m
a
g
es
f
r
o
m
e
ac
h
t
y
p
e
o
f
leav
es
ar
e
ta
k
en
a
s
th
e
te
s
t
i
m
a
g
es
t
h
at
ar
e
n
o
t
g
o
in
g
t
h
r
o
u
g
h
t
h
e
tr
ai
n
i
n
g
s
t
ag
e,
t
h
en
test
ed
ag
ai
n
s
t
5
v
a
r
iatio
n
s
o
f
G
L
C
M
o
r
ien
tatio
n
an
g
le
a
n
d
5
-
p
ix
el
d
is
tan
ce
s
.
T
o
ca
lcu
late
th
e
p
er
ce
n
tag
e
o
f
t
h
e
s
u
cc
ess
r
ate
o
f
r
ec
o
g
n
it
io
n
(
p
)
,
it
d
(
v
1
,
v
2
)
=
k
=
1
N
å
v
1
(
k
)
-
v
2
(
k
)
2
d
(
v
1
,
v
2
)
=
v
1
(
k
)
-
v
2
(
k
)
v
1
(
k
)
+
v
2
(
k
)
k
=
1
n
å
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
Herb
Lea
ve
s
R
ec
o
g
n
itio
n
u
s
in
g
Gra
y
Leve
l Co
-
o
cc
u
r
r
en
ce
Ma
tr
ix
a
n
d
F
ive
Di
s
ta
n
ce
-
…
(
R
.
R
iz
a
l I
s
n
a
n
to
)
1927
ca
n
u
s
e
t
h
e
E
q
u
atio
n
(
1
3
)
.
T
ab
le
5
is
an
ex
a
m
p
le
o
f
th
e
test
r
esu
lt
s
o
n
3
(
th
r
ee
)
v
ar
iatio
n
d
is
tan
ce
-
b
ased
s
i
m
ilar
it
y
m
ea
s
u
r
es
o
f
an
g
le
9
0
°
an
d
d
is
tan
ce
o
f
1
p
ix
el.
(
1
3
)
T
ab
le
5
.
Sam
p
les o
f
T
esti
n
g
R
esu
lt o
f
T
est
I
m
a
g
e
s
w
i
th
A
n
g
le
9
0
0
,
D
=
1
T
e
st
D
a
t
a
N
a
me
C
h
e
b
y
sh
e
v
C
i
t
y
b
l
o
c
k
M
i
n
k
o
w
sk
i
P
l
u
c
h
e
a
i
n
d
i
c
a
(
1
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
P
l
u
c
h
e
a
i
n
d
i
c
a
(
2
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
P
l
u
c
h
e
a
i
n
d
i
c
a
(
3
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
P
l
u
c
h
e
a
i
n
d
i
c
a
(
4
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
P
l
u
c
h
e
a
i
n
d
i
c
a
(
5
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
F
i
c
u
s B
e
n
j
a
m
i
n
a
(
1
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
F
i
c
u
s B
e
n
j
a
m
i
n
a
(
2
)
F
a
l
s
e
T
r
u
e
T
r
u
e
F
i
c
u
s B
e
n
j
a
m
i
n
a
(
3
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
F
i
c
u
s B
e
n
j
a
m
i
n
a
(
4
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
F
i
c
u
s B
e
n
j
a
m
i
n
a
(
5
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
A
n
r
e
d
e
r
a
C
o
r
d
i
f
o
l
i
a
(
1
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
A
n
r
e
d
e
r
a
C
o
r
d
i
f
o
l
i
a
(
2
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
A
n
r
e
d
e
r
a
C
o
r
d
i
f
o
l
i
a
(
3
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
A
n
r
e
d
e
r
a
C
o
r
d
i
f
o
l
i
a
(
4
)
T
r
u
e
F
a
l
s
e
F
a
l
s
e
A
n
r
e
d
e
r
a
C
o
r
d
i
f
o
l
i
a
(
5
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
R
i
c
i
n
u
s
C
o
mm
u
n
i
s (
1
)
F
a
l
s
e
F
a
l
s
e
T
r
u
e
R
i
c
i
n
u
s
C
o
mm
u
n
i
s (
2
)
F
a
l
s
e
T
r
u
e
T
r
u
e
R
i
c
i
n
u
s
C
o
mm
u
n
i
s (
3
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
R
i
c
i
n
u
s
C
o
mm
u
n
i
s (
4
)
F
a
l
s
e
T
r
u
e
T
r
u
e
R
i
c
i
n
u
s
C
o
mm
u
n
i
s (
5
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
H
e
mi
g
r
a
p
h
i
s A
l
t
e
r
n
a
t
a
(
1
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
H
e
mi
g
r
a
p
h
i
s A
l
t
e
r
n
a
t
a
(
2
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
H
e
mi
g
r
a
p
h
i
s A
l
t
e
r
n
a
t
a
(
3
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
H
e
mi
g
r
a
p
h
i
s A
l
t
e
r
n
a
t
a
(
4
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
H
e
mi
g
r
a
p
h
i
s A
l
t
e
r
n
a
t
a
(
5
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
A
l
p
i
n
i
a
G
a
l
a
n
g
a
(
1
)
T
r
u
e
T
r
u
e
T
r
u
e
A
l
p
i
n
i
a
G
a
l
a
n
g
a
(
2
)
T
r
u
e
T
r
u
e
T
r
u
e
A
l
p
i
n
i
a
G
a
l
a
n
g
a
(
3
)
T
r
u
e
T
r
u
e
T
r
u
e
A
l
p
i
n
i
a
G
a
l
a
n
g
a
(
4
)
T
r
u
e
T
r
u
e
T
r
u
e
A
l
p
i
n
i
a
G
a
l
a
n
g
a
(
5
)
F
a
l
s
e
F
a
l
s
e
F
a
l
s
e
R
es
u
lts
o
f
r
ec
o
g
n
itio
n
r
ate
th
r
o
u
g
h
o
u
t
t
h
e
test
i
n
g
t
h
at
h
as
b
ee
n
co
n
d
u
cted
o
n
1
5
0
test
im
ag
es
w
it
h
all
v
ar
iatio
n
s
ar
e
s
h
o
w
n
in
T
ab
le
6
.
T
ab
le
6
.
R
ec
o
g
n
itio
n
P
er
ce
n
ta
g
e
o
f
A
ll Va
r
iatio
n
s
o
f
GL
C
M
P
ar
am
eter
s
(
C
h
eb
y
s
h
e
v
)
O
r
i
e
n
t
a
t
i
on
A
n
g
l
e
P
i
x
e
l
D
i
st
a
n
c
e
d
=
1
d
=
2
d
=
3
d
=
4
d
=
5
A
l
l
A
n
g
l
e
s
6
.
0
0
5
.
3
3
6
.
0
0
8
.
0
0
4
.
0
0
A
n
g
l
e
0
1
7
.
3
3
1
4
.
6
7
1
3
.
3
3
1
2
.
0
0
1
3
.
3
3
A
n
g
l
e
45
1
4
.
0
0
1
8
.
0
0
1
9
.
3
3
1
3
.
3
3
1
4
.
6
7
A
n
g
l
e
90
1
8
.
0
0
1
8
.
6
7
1
6
.
6
7
1
5
.
3
3
1
6
.
0
0
A
n
g
l
e
135
1
3
.
3
3
8
.
0
0
8
.
0
0
8
.
0
0
8
.
6
7
Fig
u
r
e
5
a
n
d
T
ab
le
6
illu
s
tr
at
e
th
e
r
ec
o
g
n
i
tio
n
s
u
cc
e
s
s
r
ates
g
r
ap
h
o
f
1
5
0
test
i
m
a
g
e
s
f
r
o
m
3
0
t
y
p
e
s
o
f
h
er
b
lea
f
u
s
i
n
g
C
h
eb
y
s
h
e
v
d
is
ta
n
ce
m
ea
s
u
r
e
m
e
n
t.
Af
t
er
test
in
g
b
e
d
o
n
e,
t
h
is
r
ec
o
g
n
i
tio
n
s
y
s
te
m
w
it
h
GL
C
M
m
et
h
o
d
an
d
C
h
eb
y
s
h
e
v
d
is
ta
n
ce
p
r
o
d
u
ce
t
h
e
h
ig
h
es
t
ac
cu
r
ac
y
r
ate
o
f
1
9
.
3
3
%
w
h
ich
i
s
at
a
d
i
s
ta
n
c
e
o
f
3
p
ix
els
an
d
u
s
i
n
g
a
s
in
g
le
o
r
ien
tatio
n
an
g
le
o
f
4
5
°
.
W
h
i
le
th
e
lo
w
est
ac
c
u
r
ac
y
r
ate
is
4
.
0
0
%
w
h
e
n
u
s
in
g
an
o
r
ien
tatio
n
all
an
g
le
w
it
h
i
n
5
n
eig
h
b
o
r
in
g
p
i
x
els.
T
ab
le
7
an
d
Fig
u
r
e
6
d
ep
ict
t
h
e
r
ec
o
g
n
itio
n
s
u
c
ce
s
s
r
ate
s
g
r
ap
h
o
f
1
5
0
test
i
m
ag
e
s
u
s
in
g
C
it
y
-
b
lo
c
k
d
is
tan
ce
m
ea
s
u
r
e
m
en
t.
Af
ter
test
i
n
g
b
e
d
o
n
e,
th
i
s
r
ec
o
g
n
it
io
n
s
y
s
te
m
w
it
h
G
L
C
M
m
et
h
o
d
an
d
C
it
y
b
lo
ck
d
is
tan
ce
p
r
o
d
u
ce
th
e
h
ig
h
est
ac
cu
r
ac
y
r
ate
o
f
1
8
.
6
7
%
w
h
ich
i
s
at
a
d
is
ta
n
ce
o
f
1
p
ix
els
a
n
d
a
s
in
g
le
or
ien
tatio
n
a
n
g
le
o
f
0
°.
W
h
ile
th
e
lo
w
est
ac
cu
r
ac
y
r
ate
i
s
4
.
0
0
%
w
h
e
n
u
s
i
n
g
a
n
o
r
ien
tatio
n
all
an
g
le
w
it
h
i
n
5
n
eig
h
b
o
r
in
g
p
ix
el
s
an
d
u
s
i
n
g
a
n
o
r
ien
tatio
n
all
an
g
le.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
3
,
J
u
n
e
201
8
:
1
9
2
0
–
1932
1928
Fig
u
r
e
5
.
Su
cc
es
s
R
ate
s
Gr
ap
h
o
f
T
est I
m
a
g
e
R
ec
o
g
n
itio
n
u
s
i
n
g
C
h
eb
y
s
h
e
v
d
is
tan
ce
Fig
u
r
e
6
.
Su
cc
es
s
R
ate
s
Gr
ap
h
o
f
T
est I
m
a
g
e
R
ec
o
g
n
itio
n
u
s
i
n
g
C
i
t
y
-
b
lo
ck
T
ab
l
e
7
.
R
ec
o
g
n
itio
n
P
er
ce
n
ta
g
e
o
f
A
ll Va
r
iatio
n
s
o
f
GL
C
M
P
ar
am
eter
s
(
C
i
t
y
-
b
lo
ck
)
O
r
i
e
n
t
a
t
i
on
A
n
g
l
e
P
i
x
e
l
D
i
st
a
n
c
e
d
=
1
d
=
2
d
=
3
d
=
4
d
=
5
A
l
l
A
n
g
l
e
s
6
.
0
0
5
.
3
3
6
.
0
0
8
.
0
0
4
.
0
0
A
n
g
l
e
0
1
8
.
6
7
9
.
3
3
1
2
.
0
0
1
3
.
3
3
1
0
.
6
7
A
n
g
l
e
45
1
4
.
0
0
1
0
.
0
0
1
4
.
6
7
1
0
.
0
0
1
0
.
0
0
A
n
g
l
e
90
1
6
.
0
0
1
6
.
0
0
1
0
.
6
7
1
1
.
3
3
1
4
.
0
0
A
n
g
l
e
135
1
2
.
6
7
1
4
.
6
7
1
4
.
0
0
1
0
.
0
0
1
1
.
3
3
T
ab
le
8
.
R
ec
o
g
n
itio
n
P
er
ce
n
ta
g
e
o
f
A
ll Va
r
iatio
n
s
o
f
GL
C
M
P
ar
am
eter
s
(
Mi
n
k
o
w
s
k
i)
O
r
i
e
n
t
a
t
i
on
A
n
g
l
e
P
i
x
e
l
D
i
st
a
n
c
e
d
=
1
d
=
2
d
=
3
d
=
4
d
=
5
A
l
l
A
n
g
l
e
s
6
.
0
0
5
.
3
3
6
.
0
0
8
.
0
0
4
.
0
0
A
n
g
l
e
0
1
6
.
6
7
1
1
.
3
3
1
3
.
3
3
1
5
.
3
3
1
0
.
6
7
A
n
g
l
e
45
1
5
.
3
3
1
2
.
6
7
1
4
.
0
0
1
0
.
6
7
1
0
.
0
0
A
n
g
l
e
90
1
8
.
6
7
1
8
.
0
0
1
1
.
3
3
1
2
.
6
7
1
4
.
6
7
A
n
g
l
e
135
1
4
.
0
0
1
6
.
6
7
1
4
.
6
7
1
1
.
3
3
1
0
.
6
7
T
ab
le
8
an
d
Fig
u
r
e
7
d
ep
ict
th
e
r
ec
o
g
n
itio
n
s
u
cc
es
s
r
ates
g
r
a
p
h
o
f
1
5
0
test
im
a
g
es
u
s
i
n
g
M
in
k
o
w
s
k
i
d
is
tan
ce
m
ea
s
u
r
e
m
en
t.
Af
ter
test
i
n
g
b
e
d
o
n
e,
th
is
r
ec
o
g
n
it
i
o
n
s
y
s
te
m
w
it
h
G
L
C
M
m
et
h
o
d
an
d
Min
k
o
w
s
k
i
d
is
tan
ce
p
r
o
d
u
ce
th
e
h
ig
h
est
ac
cu
r
ac
y
r
ate
o
f
1
8
.
6
7
%
w
h
ich
i
s
at
a
d
is
ta
n
ce
o
f
1
p
ix
els
a
n
d
a
s
in
g
le
o
r
ien
tatio
n
a
n
g
le
o
f
9
0
°.
W
h
il
e
th
e
lo
w
es
t a
cc
u
r
ac
y
r
ate
is
4
.
0
0
%
w
h
en
u
s
in
g
a
n
o
r
ien
tat
io
n
all
a
n
g
le
w
it
h
i
n
5
n
eig
h
b
o
r
in
g
p
ix
el
s
an
d
u
s
i
n
g
a
n
o
r
ien
tatio
n
an
g
le.
Fig
u
r
e
7
.
Su
cc
es
s
R
ate
s
Gr
ap
h
o
f
T
est I
m
a
g
e
R
ec
o
g
n
itio
n
Mi
n
k
o
w
s
k
i
Fig
u
r
e
8
.
Gr
ap
h
o
f
Su
cc
e
s
s
R
a
te
I
n
tr
o
d
u
ctio
n
o
f
1
0
,
2
0
an
d
3
0
L
ea
f
T
y
p
es
0.0
0
5.0
0
10.
00
15.
00
20.
00
25.
00
d=
1
d=
2
d=
3
d=
4
d=
5
A
l
l
Ang
l
e
s
A
ng
l
e
0
A
ng
l
e
4
5
A
ng
l
e
9
0
A
ng
l
e
1
35
0.0
0
5.0
0
10.
00
15.
00
20.
00
d=
1
d=
2
d=
3
d=
4
d=
5
A
l
l
Ang
l
e
A
ng
l
e
0
A
ng
l
e
4
5
A
ng
l
e
9
0
A
ng
l
e
1
35
0.0
0%
5.0
0%
10.
00%
15.
00%
20.
00%
d=
1
d=
2
d=
3
d=
4
d=
5
A
l
l
Ang
l
e
A
ng
l
e
0
A
ng
l
e
4
5
A
ng
l
e
9
0
A
ng
l
e
1
35
0.0
0%
20.
00%
40.
00%
60.
00%
80.
00%
100
.00%
1
0
T
es
t
D
at
a
2
0
T
es
t
D
at
a
3
0
T
es
t
D
at
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
Herb
Lea
ve
s
R
ec
o
g
n
itio
n
u
s
in
g
Gra
y
Leve
l Co
-
o
cc
u
r
r
en
ce
Ma
tr
ix
a
n
d
F
ive
Di
s
ta
n
ce
-
…
(
R
.
R
iz
a
l I
s
n
a
n
to
)
1929
Fig
u
r
e
8
d
ep
icts
a
g
r
ap
h
o
f
t
h
e
s
u
cc
e
s
s
r
ate
o
f
1
0
,
2
0
an
d
3
0
leaf
i
m
ag
e
s
.
Fro
m
t
h
e
te
s
t
u
s
i
n
g
1
0
(
ten
)
test
i
m
ag
e
d
ata
w
i
th
v
ar
io
u
s
d
is
ta
n
ce
s
u
s
ed
f
o
r
id
en
ti
f
icatio
n
,
it
ca
n
b
e
s
h
o
w
n
th
at
test
in
g
u
s
in
g
C
an
b
er
r
a
d
is
tan
ce
i
s
th
e
b
est
w
it
h
r
ec
o
g
n
itio
n
r
ate
9
2
%.
T
h
er
e
ar
e
s
o
m
e
m
o
s
t
p
r
o
b
ab
l
y
r
ea
s
o
n
s
th
at
i
n
f
l
u
en
ce
t
h
e
r
e
co
g
n
itio
n
r
ates
ar
e
n
o
t
to
o
g
o
o
d
r
esu
lts
.
F
ir
s
t,
s
o
m
e
o
b
j
ec
ts
ar
e
to
o
s
m
all
i
f
co
m
p
ar
ed
to
its
b
ac
k
g
r
o
u
n
d
,
f
o
r
e
x
a
m
p
le:
A
lp
in
a
g
alan
g
a,
C
li
n
ac
at
h
u
s
m
u
tan
s
L
i
n
d
au
,
a
n
d
P
an
d
an
u
s
a
m
ar
y
lli
f
o
liu
s
.
T
h
is
s
it
u
atio
n
ten
d
s
to
g
i
v
e
f
al
s
e
r
ec
o
g
n
itio
n
r
es
u
lt
s
.
Fi
g
u
r
e
9
d
ep
icts
s
o
m
e
leav
es
w
i
th
it
s
o
b
j
ec
t
is
to
o
s
m
all
i
f
co
m
p
ar
ed
to
o
its
b
ac
k
g
r
o
u
n
d
.
S
ec
o
n
d
,
s
o
m
e
leav
e
s
h
a
v
e
al
m
o
s
t
s
i
m
ilar
te
x
t
u
r
e
as
t
h
e
o
th
er
s
.
T
h
is
s
i
m
ilar
it
y
te
n
d
s
to
g
i
v
e
f
alse
r
ec
o
g
n
itio
n
r
es
u
lts
to
o
.
Fig
u
r
e
10
s
h
o
w
s
ei
g
h
t
p
air
s
o
f
d
if
f
er
e
n
t l
ea
v
es
w
it
h
s
i
m
i
lar
tex
t
u
r
e.
(
a
)
A
l
p
i
n
a
g
a
l
a
n
g
a
(
b
)
C
l
i
n
a
c
a
t
h
u
s
mu
t
a
n
s
L
i
n
d
a
u
(
b
)
P
a
n
d
a
n
u
s
a
mary
l
l
i
f
o
l
i
u
s
Fig
u
r
e
9
.
Sa
m
p
le
o
f
s
o
m
e
leav
es th
at
ar
e
to
o
s
m
al
l c
o
m
p
ar
ed
to
its
b
ac
k
g
r
o
u
n
d
C
a
t
h
a
n
a
n
t
h
u
s ro
se
u
s I
a
n
d
A
n
n
o
n
a
s
q
u
a
mo
sal
F
i
c
u
s
b
e
n
j
a
mi
n
a
a
n
d
S
a
u
r
o
p
u
s
a
n
d
r
o
g
y
n
o
u
s
A
v
e
r
r
h
o
a
b
i
l
i
mb
i
a
n
d
B
o
e
se
n
b
e
r
g
i
a
r
o
t
u
n
d
a
S
t
r
o
b
i
l
a
n
t
h
e
s c
r
i
s
p
u
s a
n
d
S
y
z
y
g
i
u
m
p
o
l
y
a
n
t
h
u
m
A
n
n
o
n
a
m
u
r
i
c
a
t
a
a
n
d
M
o
r
i
n
d
a
c
i
t
r
i
f
o
l
i
a
A
l
p
i
n
a
g
a
l
a
n
g
a
a
n
d
P
a
n
d
a
n
u
s
a
m
a
r
y
l
l
i
f
o
l
i
u
s
M
u
ss
a
e
n
d
a
p
u
b
e
sce
n
s
a
n
d
P
si
d
i
u
m g
u
a
j
a
v
a
C
l
i
n
a
c
a
t
h
u
s
mu
t
a
n
s L
i
n
d
a
u
a
n
d
A
l
p
i
n
a
g
a
l
a
n
g
a
l
Fig
u
r
e
1
0
.
E
ig
h
t p
air
o
f
leav
e
s
w
it
h
s
i
m
i
lar
tex
t
u
r
e
No
t
all
leaf
i
m
a
g
es
ca
n
b
e
w
el
l
r
ec
o
g
n
ized
b
y
t
h
e
ap
p
licatio
n
b
ec
au
s
e
th
e
r
ec
o
g
n
itio
n
p
r
o
ce
s
s
o
f
t
h
e
leaf
i
m
a
g
e
to
b
e
r
ec
o
g
n
ized
s
till
d
ep
en
d
s
o
n
t
h
e
n
u
m
b
er
o
f
leaf
test
i
m
a
g
e
s
.
No
r
m
all
y
,
t
h
e
m
o
r
e
d
ata
test
e
d
th
e
te
n
d
en
c
y
o
f
r
ec
o
g
n
it
io
n
r
esu
lt
s
to
i
n
cr
ea
s
e.
Ho
w
ev
er
,
w
h
e
n
w
e
ad
d
m
o
r
e
d
ata
to
b
e
test
ed
w
h
ich
i
ts
tex
t
u
r
e
is
s
i
m
ilar
to
o
t
h
er
t
y
p
e
s
o
f
lea
f
,
t
h
e
r
ec
o
g
n
it
io
n
w
ill
b
e
f
alse.
T
h
er
ef
o
r
e,
in
t
h
is
r
es
ea
r
ch
,
t
h
e
ad
d
itio
n
o
f
test
d
ata
ev
e
n
lo
w
er
s
t
h
e
r
ec
o
g
n
itio
n
r
ate.
Als
o
,
th
e
c
h
a
r
ac
ter
is
tics
o
f
t
h
e
lea
v
es
ca
n
b
e
s
ee
n
f
r
o
m
o
t
h
er
asp
ec
ts
s
u
c
h
as
co
lo
r
,
s
ize,
s
h
ap
e,
tex
t
u
r
e,
an
d
o
d
o
r
f
o
r
w
h
ic
h
t
h
e
y
ar
e
n
o
t
b
e
in
v
o
l
v
ed
in
t
h
is
r
esear
c
h
.
T
h
er
ef
o
r
e,
s
o
m
e
r
esear
ch
in
v
o
lv
in
g
th
e
o
t
h
er
asp
ec
ts
m
e
n
tio
n
ed
ab
o
v
e
to
r
ec
o
g
n
ize
th
e
h
er
b
leav
es
ar
e
s
u
g
g
e
s
ted
to
p
r
o
d
u
ce
th
e
b
est r
ec
o
g
n
itio
n
r
esu
l
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.