I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
9
,
No
.
1
,
Ju
ly
2
0
2
5
,
p
p
.
202
~
2
1
3
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
3
9
.i
1
.
pp
202
-
2
1
3
202
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
An ef
ficien
t
seg
m
entatio
n using
ad
a
ptive ra
dia
l bas
is func
tion
neura
l net
wo
rk
for toma
to a
nd ma
ng
o
plant leaf im
a
g
es
J
o
la
k
ula
Aso
k
a
S
m
it
ha
1
,
B
icha
g
a
l Sha
da
k
s
ha
ra
pp
a
1
,
Sh
ee
la
P
a
rv
a
t
hy
2
,
K
ilin
g
a
r
Vee
n
a
2
,
Albert
J
enif
er
2
,
B
a
dd
a
la
Vij
a
y
a
Nirma
la
2
,
Su
bb
ia
h M
ur
ug
a
n
2
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
S
r
i
S
a
i
r
a
m
C
o
l
l
e
g
e
o
f
En
g
i
n
e
e
r
i
n
g
,
A
n
e
k
a
l
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
S
E
R
e
sea
r
c
h
C
e
n
t
r
e
,
S
r
i
S
a
i
r
a
m
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
A
n
e
k
a
l
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
5
,
2
0
2
4
R
ev
is
ed
Feb
5
,
2
0
2
5
Acc
ep
ted
Feb
28
,
2
0
2
5
Ag
ricu
lt
u
re
h
a
s
b
e
c
o
m
e
sim
p
ly
to
fe
e
d
e
v
e
r
-
g
r
o
win
g
p
o
p
u
lat
i
o
n
s.
Th
e
to
m
a
to
is
a
r
g
u
a
b
ly
th
e
m
o
st
we
ll
-
k
n
o
wn
v
e
g
e
tab
le
i
n
a
g
ricu
lt
u
ra
l
a
re
a
s
a
n
d
p
lay
s
a
sig
n
ifi
c
a
n
t
ro
le
in
t
h
e
g
ro
wt
h
o
f
v
e
g
e
ta
b
les
in
o
u
r
d
a
il
y
li
v
e
s.
Ho
we
v
e
r,
b
e
c
a
u
se
th
is
to
m
a
to
h
a
s
m
u
lt
ip
le
d
ise
a
se
s,
ima
g
e
se
g
m
e
n
tatio
n
o
f
th
e
d
ise
a
se
d
lea
f
sh
o
ws
a
k
e
y
r
o
le
in
c
las
sify
in
g
t
h
e
d
ise
a
se
b
y
th
e
lea
f'
s
sy
m
p
to
m
s.
Th
e
re
fo
re
,
i
n
th
is
p
a
p
e
r,
a
n
e
fficie
n
t
p
lan
t
d
ise
a
se
se
g
m
e
n
tatio
n
u
sin
g
a
n
a
d
a
p
ti
v
e
ra
d
ial
b
a
sis
fu
n
c
ti
o
n
n
e
u
ra
l
n
e
two
r
k
(ARBFNN)
c
las
sifier.
Th
e
p
ro
p
o
se
d
ra
d
ial
b
a
sis
fu
n
c
ti
o
n
(RBF
)
n
e
u
ra
l
n
e
two
r
k
is
e
n
h
a
n
c
e
d
b
y
u
sin
g
t
h
e
flo
we
r
p
o
ll
in
a
ti
o
n
a
lg
o
rit
h
m
(F
P
A).
F
irstl
y
,
th
e
n
o
ise
is
d
e
tac
h
e
d
b
y
a
n
a
d
a
p
ti
v
e
m
e
d
ian
fil
ter
a
n
d
h
isto
g
ra
m
e
q
u
a
li
z
a
ti
o
n
.
Th
e
n
,
fr
o
m
e
v
e
ry
lea
f
ima
g
e
,
d
iffere
n
t
k
i
n
d
o
f
c
o
l
o
r
fe
a
tu
re
s
is
e
x
trac
ted
.
Afte
r
t
h
e
e
x
trac
ti
o
n
o
f
fe
a
tu
re
s,
th
o
se
a
re
fe
d
t
o
t
h
e
se
g
m
e
n
tatio
n
p
h
a
se
to
se
c
ti
o
n
t
h
e
d
ise
a
se
se
rv
in
g
fr
o
m
th
e
in
p
u
t
ima
g
e
.
Th
e
e
fficie
n
c
y
o
f
th
e
s
u
g
g
e
ste
d
m
e
th
o
d
is
a
n
a
ly
z
e
d
b
a
se
d
o
n
v
a
rio
u
s
m
e
tri
c
s
a
n
d
o
u
r
tec
h
n
iq
u
e
a
tt
a
in
e
d
a
b
e
tt
e
r
a
c
c
u
ra
c
y
o
f
9
7
.
5
8
%
.
K
ey
w
o
r
d
s
:
C
o
lo
r
f
ea
tu
r
e
Flo
wer
p
o
llin
atio
n
alg
o
r
ith
m
Plan
t d
is
ea
s
e
R
ad
ial
b
asis
f
u
n
ctio
n
n
eu
r
al
n
etwo
r
k
Seg
m
en
tatio
n
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
:
J
o
lak
u
la
Aso
k
a
Sm
ith
a
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
Sri
Sair
a
m
C
o
lleg
e
o
f
E
n
g
in
ee
r
in
g
An
ek
al,
B
en
g
alu
r
u
,
Kar
n
atak
a
,
I
n
d
ia
E
m
ail: sm
ith
a.
as2
0
0
4
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Ab
o
u
t
7
0
%
o
f
th
e
co
u
n
tr
y
'
s
e
co
n
o
m
ic
o
u
tp
u
t
co
m
es
f
r
o
m
a
g
r
icu
ltu
r
e
[
1
]
.
T
h
e
m
ajo
r
ity
o
f
th
e
tim
e,
d
is
ea
s
es
af
f
ec
tin
g
d
if
f
er
en
t
c
r
o
p
s
h
as
an
im
p
ac
t
o
n
th
e
q
u
a
n
tity
an
d
q
u
ality
o
f
ag
r
icu
ltu
r
al
g
o
o
d
s
[
2
]
.
Vital
cr
o
p
p
r
o
ce
s
s
es
lik
e
p
h
o
to
s
y
n
th
esis
,
p
o
llin
atio
n
,
tr
an
s
p
ir
ati
o
n
,
g
er
m
in
atio
n
,
a
n
d
f
er
tili
za
tio
n
ar
e
alter
ed
b
y
th
ese
d
is
ea
s
es
[
3
]
.
So
m
e
o
f
th
ese
illn
ess
e
s
ar
e
b
r
o
u
g
h
t
o
n
b
y
p
ath
o
g
en
s
lik
e
b
ac
ter
ia,
f
u
n
g
i,
an
d
v
ir
u
s
es
[
4
]
.
C
o
n
tin
u
o
u
s
in
s
p
ec
tio
n
s
f
r
o
m
f
ar
m
er
s
o
r
ag
r
ic
u
ltu
r
al
p
r
o
f
ess
io
n
als
wer
e
r
eq
u
ir
ed
f
o
r
tr
a
d
itio
n
al
m
eth
o
d
s
o
f
d
is
ea
s
e
d
etec
tio
n
[
5
]
;
th
o
u
g
h
,
th
ese
m
eth
o
d
s
ar
e
f
r
eq
u
en
tl
y
in
ef
f
icie
n
t,
ex
cl
u
s
iv
e,
an
d
r
eq
u
ir
e
s
ig
n
if
ican
t
lab
o
r
[
6
]
.
B
y
c
r
ea
tin
g
tec
h
n
iq
u
es
f
o
r
p
lan
t
d
is
ea
s
e
d
iag
n
o
s
is
an
d
m
itig
atio
n
,
cr
o
p
p
r
o
d
u
ctiv
ity
ca
n
b
e
in
cr
ea
s
ed
wh
ile
u
s
in
g
f
ewe
r
p
esti
cid
es
[
7
]
.
T
h
er
e
f
o
r
e,
en
s
u
r
in
g
f
o
o
d
s
ec
u
r
ity
r
eq
u
ir
es
d
is
ea
s
e
d
etec
tio
n
[
8
]
.
T
h
e
tr
ad
itio
n
al
tech
n
iq
u
e
o
f
i
d
en
tific
atio
n
o
f
illn
ess
h
as
b
e
en
a
p
h
y
s
ical
in
v
esti
g
atio
n
b
y
p
r
o
d
u
ce
r
s
,
alth
o
u
g
h
th
is
co
u
ld
b
e
tim
e
-
co
n
s
u
m
in
g
an
d
co
s
tly
[
9
]
,
[
1
0
]
.
C
r
o
p
y
iel
d
ca
n
b
e
b
o
o
s
ted
wh
ile
u
s
in
g
f
ewe
r
p
esti
cid
es
b
y
d
ev
elo
p
in
g
m
eth
o
d
s
f
o
r
th
e
id
en
tific
atio
n
an
d
m
itig
atio
n
o
f
p
lan
t d
is
ea
s
es [
1
1
]
.
As a
r
esu
lt,
g
u
ar
an
teein
g
f
o
o
d
s
ec
u
r
ity
n
ec
ess
itates
b
o
th
th
e
d
etec
tio
n
o
f
d
is
ea
s
es
an
d
th
e
cr
ea
tio
n
o
f
im
p
r
o
v
ed
c
r
o
p
v
ar
i
eties
[
1
2
]
.
T
h
e
tr
ad
itio
n
al
tech
n
iq
u
e
o
f
d
is
ea
s
e
d
etec
tio
n
h
as
b
ee
n
m
an
u
al
a
s
s
es
s
m
en
t
b
y
f
ar
m
e
r
s
[
1
3
]
,
b
u
t
th
is
m
ay
b
e
tim
e
-
co
n
s
u
m
in
g
an
d
ex
p
en
s
iv
e
[
1
4
]
.
T
h
e
m
ajo
r
ity
o
f
f
ar
m
er
s
n
o
w
u
s
e
ce
ll
p
h
o
n
es
th
an
k
s
to
r
e
v
o
lu
tio
n
ar
y
c
h
an
g
es
in
in
f
o
r
m
atio
n
i
n
n
o
v
atio
n
[
1
5
]
.
T
h
e
ap
p
licatio
n
o
f
tech
n
ica
l
ad
v
an
ce
m
e
n
t
in
t
h
e
ag
r
icu
ltu
r
al
s
ec
to
r
aid
s
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
n
efficien
t seg
men
ta
tio
n
u
s
in
g
a
d
a
p
tive
r
a
d
ia
l
b
a
s
is
fu
n
ctio
n
n
eu
r
a
l
n
etw
o
r
k
…
(
Jo
la
ku
l
a
A
s
o
ka
S
mith
a
)
203
r
ed
u
cin
g
y
ield
lo
s
s
[
1
6
]
.
Fo
llo
win
g
im
ag
e
r
ec
eip
t,
tech
n
iq
u
es
lik
e
co
n
tr
ast
s
tr
etch
in
g
an
d
n
o
is
e
f
ilter
in
g
ar
e
u
n
d
er
tak
e
n
to
im
p
r
o
v
e
th
e
im
ag
e
[
1
7
]
.
I
n
r
ec
en
t
y
ea
r
s
,
s
ev
er
al
m
eth
o
d
s
wer
e
ex
ec
u
ted
i
n
s
eg
m
en
tin
g
p
lan
t
illn
ess
es
in
leav
es.
Fo
r
id
en
tif
y
in
g
t
h
e
ch
a
r
ac
ter
is
tics
,
s
eg
m
en
tin
g
,
a
n
d
d
iag
n
o
s
in
g
c
r
o
p
d
is
ea
s
e,
s
o
m
e
o
f
t
h
e
ef
f
o
r
ts
u
tili
ze
th
e
Gab
o
r
f
ilter
,
h
o
m
o
g
en
eo
u
s
p
ix
el
co
u
n
ti
n
g
,
f
u
zz
y
C
-
m
ea
n
s
(
FC
M)
,
an
d
ar
tific
ial
n
eu
r
a
l
n
etwo
r
k
s
(
ANN)
.
T
h
e
p
r
o
p
o
s
ed
wo
r
k
d
ev
el
o
p
s
an
ad
ap
tiv
e
r
ad
ial
b
asis
f
u
n
ctio
n
n
eu
r
al
n
etwo
r
k
(
AR
B
FNN
)
ap
p
r
o
ac
h
f
o
r
th
e
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
.
Her
e,
s
o
m
e
r
ec
en
t
liter
atu
r
e
th
at
f
o
cu
s
ed
r
esear
ch
o
n
t
o
m
ato
leaf
d
is
ea
s
e,
as
well
as
s
o
m
e
o
th
er
leaf
d
is
ea
s
e
d
et
ec
tio
n
is
r
ev
iewe
d
.
R
esid
u
al
n
etwo
r
k
(
R
esNet
)
a
n
d
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
-
d
ee
p
n
eu
r
al
n
etwo
r
k
(
L
STM
–
DNN
)
b
ased
p
lan
t
le
af
m
u
lti
-
d
is
ea
s
e
clas
s
if
icatio
n
.
FC
M
clu
s
ter
in
g
alg
o
r
ith
m
was
u
s
ed
f
o
r
im
ag
e
s
eg
m
en
tatio
n
[
1
8
]
.
T
h
e
s
eg
m
en
ted
im
ag
e
was
u
s
ed
f
o
r
m
u
lti
-
d
is
ea
s
e
clas
s
if
icatio
n
p
r
o
ce
s
s
.
T
h
e
R
esNet1
5
0
was
u
s
ed
f
o
r
class
if
icatio
n
p
r
o
ce
s
s
.
Plan
t
leaf
d
is
ea
s
e
cla
s
s
i
f
icatio
n
an
d
d
a
m
ag
e
d
etec
tio
n
ap
p
r
o
ac
h
in
itially
,
th
ey
id
en
tifie
d
wh
ich
ty
p
e
o
f
d
is
ea
s
e
af
f
ec
ted
in
th
e
in
p
u
t
im
ag
e
u
s
in
g
Den
s
eNe
t.
T
h
is
m
o
d
el
was
tr
ai
n
ed
u
s
in
g
p
h
o
to
s
th
at
wer
e
d
iv
id
e
d
in
to
ca
teg
o
r
ies
b
ased
o
n
th
eir
n
atu
r
e,
s
u
ch
as
h
ea
lth
y
an
d
d
if
f
er
en
t
ty
p
es
o
f
s
ick
n
ess
[
1
9
]
.
T
h
en
,
n
ew
leaf
im
ag
es
wer
e
test
ed
u
s
in
g
th
i
s
m
o
d
el
[
2
0
]
.
An
im
ag
e
o
f
p
l
an
t
illn
ess
d
iv
is
io
n
ty
p
ical
d
ep
en
d
in
g
o
n
a
n
im
p
r
o
v
e
d
p
u
l
se
-
co
u
p
led
n
eu
r
al
n
etwo
r
k
(
PC
NN)
u
s
in
g
a
s
h
u
f
f
led
f
r
o
g
-
leap
in
g
alg
o
r
ith
m
(
SF
L
A)
[
2
1
]
.
I
t
ca
n
s
u
cc
ess
f
u
lly
ex
tr
ac
t
lesi
o
n
p
ictu
r
es
th
r
o
u
g
h
th
e
s
u
r
r
o
u
n
d
in
g
ar
ea
,
lay
in
g
th
e
g
r
o
u
n
d
wo
r
k
f
o
r
later
d
is
ea
s
e
id
en
tific
atio
n
.
Dr
y
b
ea
n
leaf
d
is
ea
s
e
s
eg
m
en
tatio
n
u
s
in
g
U
-
Net
m
ec
h
an
is
m
in
itially
,
th
e
f
o
u
n
d
th
e
d
is
ea
s
e
n
am
e
f
r
o
m
t
h
e
s
eg
m
en
ted
p
a
r
t.
T
h
en
,
ag
ai
n
,
th
ey
class
if
y
th
e
d
is
ea
s
es
p
r
esen
t
in
th
e
r
aw
in
p
u
t
im
a
g
es
[
2
2
]
.
Utilizin
g
co
m
p
u
ter
v
is
io
n
t
ec
h
n
iq
u
es
with
a
s
u
p
er
p
i
x
el
clu
s
ter
an
d
h
y
b
r
id
n
eu
r
al
n
etwo
r
k
,
a
d
iag
n
o
s
tic
m
eth
o
d
m
ay
b
e
au
to
m
ated
.
Dif
f
er
en
t
alg
o
r
ith
m
s
ar
e
u
s
ed
to
ev
alu
ate
f
ea
tu
r
e
ev
alu
atio
n
s
f
o
r
c
o
lo
r
,
s
h
ap
e,
a
n
d
tex
tu
r
e
[
2
3
]
.
Fin
ally
,
t
h
e
p
h
o
to
s
wer
e
d
iv
id
ed
i
n
to
th
r
ee
g
r
o
u
p
s
u
s
in
g
s
ev
en
d
if
f
er
en
t
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es.
A
leaf
d
is
ea
s
e
th
at
ca
n
b
e
id
en
tifie
d
f
r
o
m
p
h
o
to
s
o
f
th
e
p
lan
t,
a
n
d
th
en
th
e
s
y
s
tem
'
s
ac
cu
r
ac
y
ca
n
b
e
in
cr
ea
s
ed
u
s
in
g
m
ac
h
i
n
e
lear
n
in
g
o
r
d
ee
p
lea
r
n
in
g
m
eth
o
d
s
[
2
4
]
.
T
h
e
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
e
two
r
k
s
(
C
NN
s
)
alg
o
r
ith
m
[
2
5
]
-
[
2
7
]
is
ap
p
lied
with
ca
m
er
as
o
n
i
n
d
ep
en
d
en
t
r
o
b
o
tic
p
latf
o
r
m
s
to
d
is
tin
g
u
is
h
cr
o
p
a
n
d
wee
d
ty
p
es.
2.
P
RO
P
O
SE
D
M
E
CH
AN
I
S
M
T
h
e
s
u
g
g
ested
m
eth
o
d
o
l
o
g
y
'
s
p
r
im
ar
y
g
o
al
is
to
s
ep
ar
ate
t
h
e
in
f
ec
ted
a
r
ea
f
r
o
m
th
e
ca
p
tu
r
ed
p
lan
t
leaf
.
Mo
s
t
d
is
ea
s
es
o
f
th
e
to
m
ato
p
lan
t
ca
n
b
e
d
etec
ted
at
an
ea
r
ly
s
tag
e
b
y
attac
k
in
g
th
e
leav
es
f
ir
s
t.
T
h
e
in
ev
itab
le
d
am
a
g
e
ca
n
b
e
p
r
e
v
en
ted
b
y
s
p
o
ttin
g
illn
ess
es
i
n
th
e
p
la
n
ts
as
s
o
o
n
as
p
o
s
s
ib
le.
Fo
r
th
e
im
ag
e
class
if
icatio
n
p
r
o
ce
s
s
,
th
e
s
eg
m
en
tatio
n
s
tag
e
is
s
ig
n
if
ican
tly
u
n
i
q
u
e.
T
h
r
o
u
g
h
o
u
t
th
e
s
eg
m
e
n
tatio
n
p
r
o
ce
d
u
r
e,
th
e
illn
ess
p
o
r
tio
n
s
ar
e
s
eg
m
en
ted
.
Fo
r
th
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
,
an
AR
B
FNN
c
lass
if
ier
is
u
tili
ze
d
.
T
h
e
s
u
g
g
ested
m
eth
o
d
in
v
o
l
v
es th
r
ee
s
tag
es: p
r
e
-
p
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
e
x
tr
ac
tio
n
,
a
n
d
s
eg
m
en
tatio
n
.
2
.
1
.
P
re
-
pro
ce
s
s
ing
Fo
r
th
e
s
eg
m
en
tatio
n
p
h
ase,
p
r
e
-
p
r
o
ce
s
s
in
g
is
a
cr
u
cial
s
tep
.
Sin
ce
th
er
e
will b
e
s
ig
n
if
ican
t
d
is
to
r
tio
n
in
th
e
co
llected
p
h
o
to
s
,
th
e
s
eg
m
en
tatio
n
o
u
tco
m
es
will
b
e
im
p
ac
ted
.
So
,
b
ef
o
r
e
b
ein
g
s
en
t
to
th
e
ad
ap
tiv
e
r
ad
ial
b
asis
f
u
n
ctio
n
(
RBF
)
n
eu
r
al
n
etwo
r
k
,
d
is
to
r
ted
d
ata
ar
e
p
r
e
-
p
r
o
ce
s
s
ed
.
Fo
r
p
r
e
-
p
r
o
ce
s
s
in
g
,
we
u
s
e
an
ad
ap
tiv
e
m
ed
ia
n
f
ilter
an
d
h
is
to
g
r
am
eq
u
aliza
tio
n
.
I
n
itially
,
we
ap
p
ly
an
a
d
ap
tiv
e
m
e
d
ian
f
ilter
f
o
llo
wed
th
at
h
is
to
g
r
am
e
q
u
aliza
tio
n
.
T
h
e
a
d
ap
tiv
e
m
e
d
ian
f
ilter
is
em
p
l
o
y
ed
to
elim
in
ate
th
e
d
is
to
r
tio
n
f
r
o
m
th
e
v
is
io
n
.
T
h
e
p
ix
el
in
ten
s
ities
f
r
eq
u
e
n
c
y
is
ch
an
g
ed
t
o
th
e
m
ed
ian
p
ix
el
v
alu
e
i
n
a
ce
r
tain
n
ei
g
h
b
o
r
h
o
o
d
wh
ile
th
is
f
ilter
k
ee
p
s
th
e
im
ag
e'
s
es
s
e
n
tial
ch
ar
ac
ter
is
tics
.
Av
o
id
in
g
d
is
to
r
tin
g
th
e
im
ag
e'
s
b
o
u
n
d
ar
ies,
th
is
f
ilter
r
ed
u
ce
s
d
is
to
r
tio
n
in
a
p
ictu
r
e.
T
h
e
av
er
ag
e
is
d
eter
m
in
ed
b
y
ar
r
an
g
in
g
ea
ch
o
f
th
e
n
ei
g
h
b
o
r
in
g
win
d
o
w'
s
p
ix
el
elem
en
ts
in
an
ar
ith
m
e
tic
s
eq
u
en
ce
a
n
d
s
u
b
s
titu
tin
g
th
e
p
ix
el
u
n
d
e
r
co
n
s
id
er
atio
n
u
s
in
g
th
e
ce
n
ter
(
m
ed
ian
)
p
i
x
el
v
alu
e.
Fig
u
r
e
1
ex
p
lain
s
a
g
r
ap
h
ica
l
r
e
p
r
esen
t
atio
n
o
f
th
e
m
ed
ia
n
f
ilter
an
d
s
am
p
le
r
ed
,
g
r
ee
n
,
an
d
b
lu
e
(
R
GB
)
v
alu
e
is
g
iv
en
in
F
ig
u
r
e
2
.
In
(
1
)
r
e
p
r
es
en
ts
th
e
m
ath
em
atica
l
ex
p
r
es
s
io
n
o
f
th
e
m
e
d
ian
f
ilter
ed
im
ag
e
J
(
x
,
y
)
o
f
th
e
p
ictu
r
e
K(
u
,
v
)
g
iv
e
n
in
(
1
)
.
(
,
)
=
(
,
)
∈
{
(
,
)
}
(
1
)
Utilizin
g
h
is
to
g
r
am
eq
u
aliza
tio
n
,
th
e
p
ictu
r
e
is
im
p
r
o
v
e
d
a
f
t
er
th
e
d
is
to
r
tio
n
f
iltra
tio
n
.
C
h
an
g
in
g
th
e
in
ten
s
ity
d
is
tr
ib
u
tio
n
is
a
s
tr
ateg
y
f
o
r
b
o
o
s
tin
g
co
n
tr
ast.
C
o
n
s
id
er
in
g
th
at
J
(
x
,
y
)
is
an
im
ag
e
th
at
h
as
b
ee
n
d
eliv
er
ed
a
n
d
is
r
e
p
r
esen
ted
a
s
an
m
r
b
y
m
c
m
atr
ix
with
in
teg
er
im
ag
e
i
n
ten
s
ity
th
at
v
a
r
ies
f
r
o
m
0
to
L
-
1.
L
s
tan
d
s
f
o
r
th
e
n
u
m
b
er
o
f
in
ten
s
ity
v
alu
es,
wh
ic
h
is
o
f
ten
in
t
h
e
r
a
n
g
e
o
f
2
5
6
.
Usi
n
g
a
b
in
f
o
r
ea
c
h
co
n
ce
iv
ab
le
in
ten
s
ity
as
in
(
2
)
let
th
e
n
o
r
m
alize
d
h
is
to
g
r
a
m
o
f
J
(
x
,
y
)
b
e
d
e
n
o
ted
b
y
h
a
v
in
g
a
b
in
f
o
r
e
v
er
y
p
o
s
s
ib
le
in
ten
s
ity
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
1
,
Ju
ly
20
25
:
202
-
2
1
3
204
=
ℎ
=
0
,
1
,
.
.
.
,
−
1
(
2
)
Fig
u
r
e
1
.
Gr
a
p
h
ical
r
ep
r
esen
tatio
n
o
f
th
e
m
ed
ian
f
ilter
Fig
u
r
e
2
.
C
o
lo
r
in
f
o
r
m
atio
n
at
tain
ed
th
r
o
u
g
h
t
h
e
R
GB
co
lo
r
m
o
d
el
2
.
2
.
F
e
a
t
ure
ex
t
r
a
ct
io
n
Her
e,
f
o
u
r
ty
p
es
o
f
c
o
lo
r
m
o
d
els
ar
e
ex
p
r
ess
ed
f
r
o
m
e
v
er
y
im
ag
e.
R
GB
co
lo
r
m
o
d
el
co
n
s
is
t
o
f
r
ed
(
R
)
,
g
r
ee
n
(
G)
,
an
d
b
l
u
e
(
B
)
.
On
th
e
in
p
u
t
im
ag
e,
we
r
an
d
o
m
ly
ch
o
o
s
e
twen
ty
p
ar
a
m
eter
s
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
.
T
en
o
f
th
e
twen
ty
p
o
in
ts
ar
e
ch
o
s
en
f
r
o
m
th
e
h
ea
lth
y
leaf
s
ec
tio
n
,
a
n
d
ten
p
o
in
ts
ar
e
ch
o
s
en
th
r
o
u
g
h
th
e
s
ick
ar
ea
.
W
e
m
ak
e
th
e
m
ask
o
n
ea
ch
p
o
in
t
af
t
er
ch
o
o
s
in
g
it.
E
ig
h
t
n
eig
h
b
o
r
in
g
p
ix
els
s
u
r
r
o
u
n
d
th
e
o
n
e
ce
n
tr
al
p
ix
el
th
at
m
a
k
es
u
p
th
e
m
ask
(
3
×
3
m
ask
s
)
.
T
h
en
we
r
ec
o
r
d
th
e
s
tan
d
ar
d
s
R
,
G,
a
n
d
B
f
o
r
ea
ch
p
o
in
t,
y
ield
i
n
g
a
to
tal
o
f
2
7
v
alu
es
[
1
5
]
.
E
v
er
y
p
o
i
n
t
in
F
ig
u
r
e
2
h
as
2
7
r
esu
lts
.
As
a
r
esu
lt,
2
7
0
d
ata
(
2
7
v
alu
es m
u
ltip
lied
b
y
1
0
p
h
o
to
s
)
ar
e
co
llected
th
r
o
u
g
h
th
e
ill ar
ea
as we
ll a
s
2
7
0
p
iece
s
o
f
in
f
o
r
m
atio
n
th
r
o
u
g
h
th
e
n
o
r
m
al
ar
ea
f
o
r
e
v
e
r
y
tr
ai
n
in
g
s
am
p
le.
T
h
e
R
,
G,
an
d
B
co
m
p
o
n
en
ts
ar
e
th
e
n
ex
tr
ac
te
d
f
o
r
ev
er
y
im
ag
e
.
L
ig
h
tin
g
v
a
r
iab
les h
av
e
a
n
im
p
ac
t o
n
R
GB
p
h
o
to
g
r
ap
h
s
[
1
6
]
.
T
h
e
f
o
llo
win
g
e
x
p
lain
s
h
o
w
to
ca
lcu
late
r
atio
s
.
=
+
+
(
3
)
=
+
+
(
4
)
=
+
+
(
5
)
T
h
is
to
m
ato
leaf
h
as
a
m
i
n
o
r
am
o
u
n
t
o
f
t
h
e
c
o
lo
r
b
l
u
e.
W
e,
t
h
er
ef
o
r
e,
d
is
r
eg
ar
d
th
e
b
lu
e
ch
ar
ac
ter
is
tics
.
Ad
d
itio
n
ally
,
we
h
av
e
in
clu
d
ed
a
d
d
itio
n
al
co
lo
r
m
o
d
els
f
o
r
f
ea
tu
r
es
s
u
c
h
as
HSV,
Y
C
b
C
r
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
n
efficien
t seg
men
ta
tio
n
u
s
in
g
a
d
a
p
tive
r
a
d
ia
l
b
a
s
is
fu
n
ctio
n
n
eu
r
a
l
n
etw
o
r
k
…
(
Jo
la
ku
l
a
A
s
o
ka
S
mith
a
)
205
an
d
YI
Q.
HSV
co
lo
r
m
o
d
el:
th
e
HSV
m
o
d
el
v
iews
f
o
r
h
u
e
(
H)
,
s
atu
r
atio
n
(
S)
,
an
d
v
alu
e
(
V)
.
Du
e
to
illu
m
in
atio
n
s
itu
atio
n
s
,
th
e
S
an
d
V
a
r
e
af
f
ec
tin
g
th
e
s
eg
m
en
tatio
n
ac
cu
r
ac
y
.
So
,
we
elim
in
ate
th
e
f
ea
tu
r
es
v
alu
e
S
an
d
V
f
r
o
m
th
e
f
ea
tu
r
e
v
ec
to
r
an
d
we
o
n
ly
c
o
n
s
id
er
th
e
h
u
e
f
ea
t
u
r
e
f
r
o
m
th
e
H
SV
m
o
d
el.
I
n
th
is
f
ea
tu
r
e
m
o
d
el,
we
g
et
n
i
n
e
v
a
lu
es.
YC
b
C
r
co
lo
r
m
o
d
el:
I
n
t
h
e
YC
b
C
r
clas
s
ical
,
Y
o
p
in
io
n
s
f
o
r
lu
m
in
an
ce
th
at
is
d
is
co
n
n
ec
ted
s
in
ce
th
e
two
ch
r
o
m
in
a
n
ce
m
ec
h
an
is
m
s
(
C
b
,
C
r
)
in
th
is
co
lo
r
s
p
ac
e.
T
h
e
leaf
im
ag
es
in
th
is
co
lo
r
s
am
p
le
h
av
e
a
wea
k
b
l
u
e
v
alu
e
in
t
h
e
leaf
r
an
g
e
,
m
ak
in
g
th
e
m
s
u
itab
le
f
o
r
leaf
s
eg
m
en
tatio
n
.
T
h
e
co
n
v
er
s
io
n
o
f
th
e
R
GB
co
lo
r
s
p
ac
e
to
th
e
YC
b
C
r
co
lo
r
s
p
ac
e
is
ap
p
r
o
v
ed
an
d
av
ailab
le
u
tili
zin
g
in
(
6
)
.
[
]
=
[
0
.
2990
.
5870
.
114
0
.
596
−
0
.
275
−
0
.
321
0
.
212
−
0
.
523
−
0
.
311
]
[
]
(
6
)
T
h
e
“Y”
elem
en
t
is
af
f
ec
ted
b
y
th
e
lig
h
tin
g
co
n
d
itio
n
[
1
5
]
.
T
h
er
e
f
o
r
e,
we
co
n
s
id
er
b
o
th
"C
b
"
an
d
"C
r
"
as
u
ltima
te
ch
ar
ac
ter
is
ti
cs
v
ec
to
r
s
,
elim
in
atin
g
th
e
"
Y"
ess
en
tial
s
.
YI
Q
co
lo
r
m
o
d
el:
s
im
ilar
Y
C
b
C
r
,
lu
m
in
an
ce
an
d
ch
r
o
m
in
a
n
ce
a
r
e
s
ep
ar
ated
in
YI
Q,
wh
er
e
"
Y"
is
lu
m
in
an
ce
an
d
"I
"
an
d
"Q"
is
ch
r
o
m
in
an
ce
m
ec
h
an
is
m
s
.
Fo
r
in
s
tan
ce
,
th
e
"Q"
ch
an
n
el
is
p
ar
ticu
lar
l
y
s
u
cc
ess
f
u
l
f
o
r
b
o
o
s
tin
g
lea
v
es,
wh
ile
ch
an
n
el
"I
"
is
q
u
ite
f
o
r
leaf
d
if
f
e
r
en
tiatio
n
.
In
(
7
)
is
ap
p
lied
to
c
o
n
v
e
r
t th
e
R
GB
co
lo
r
in
ter
p
lan
etar
y
to
t
h
e
YI
Q
co
lo
r
s
p
ac
e.
[
]
=
[
0
.
2990
.
5870
.
114
0
.
596
−
0
.
274
−
0
.
322
0
.
211
−
0
.
5230
.
312
]
[
]
(
7
)
T
h
e
Y
elem
e
n
t
is
im
p
ac
ted
b
y
th
e
le
v
els
o
f
lig
h
t
o
n
ce
th
e
YI
Q
d
etac
h
es
t
h
e
lu
m
i
n
o
u
s
ele
m
en
ts
"Y"
f
r
o
m
th
e
"I
"
an
d
"Q"
ch
r
o
m
i
n
s
,
h
en
ce
th
e
u
ltima
te
ch
ar
ac
ter
is
tic
m
u
s
t
b
e
elim
in
ated
f
r
o
m
th
e
v
ec
to
r
.
T
o
s
u
p
p
o
r
t
to
u
ltima
te
ch
ar
ac
ter
i
s
tics
v
ec
to
r
in
d
icatin
g
th
e
h
ea
lth
y
leaf
/d
is
ea
s
ed
leaf
,
elem
e
n
ts
"I
"
an
d
"Q"
ar
e
tak
en
.
No
wad
ay
s
,
th
is
ch
ar
ac
ter
is
tics
v
ec
to
r
co
n
s
is
t
s
o
f
R
,
G,
H,
C
b
,
C
r
,
I
,
an
d
Q.
I
n
d
iv
id
u
ally
co
m
p
o
n
en
ts
h
av
e
n
in
e
v
alu
es,
s
o
th
e
f
ea
tu
r
e
v
ec
to
r
s
ize
is
6
3
.
2
.
3
.
Seg
m
ent
a
t
i
o
n us
ing
AR
B
F
NN
Su
b
s
eq
u
en
tly
f
ea
tu
r
e
ex
tr
a
ctio
n
p
r
o
ce
s
s
;
th
e
ex
tr
ac
ted
attr
ib
u
tes
ar
e
f
ed
to
th
e
in
p
u
t
o
f
th
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
.
Fo
r
s
eg
m
en
tatio
n
,
in
th
is
p
ap
er
,
AR
B
NN
is
p
r
esen
ted
.
AR
B
NN
i
s
a
co
m
b
in
atio
n
o
f
a
f
lo
wer
p
o
llin
atio
n
alg
o
r
ith
m
(
FP
A)
an
d
an
R
B
F
n
eu
r
al
n
etwo
r
k
.
Her
e,
th
e
weig
h
t
v
alu
es
p
r
esen
t
in
th
e
R
B
F
n
eu
r
al
n
etwo
r
k
a
r
e
o
p
tim
ally
d
esig
n
ated
b
y
u
tili
zin
g
th
e
A
FP
alg
o
r
ith
m
.
T
h
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
co
n
tain
s
two
p
h
ases
n
am
ely
,
tr
ain
in
g
a
n
d
test
in
g
.
Fo
r
tr
ain
i
n
g
,
8
0
%
o
f
im
ag
es
ar
e
u
s
ed
a
n
d
2
0
%
o
f
im
ag
es
ar
e
u
s
e
d
f
o
r
th
e
test
in
g
p
r
o
ce
s
s
.
I
n
itially
,
th
e
tr
ain
in
g
p
r
o
ce
s
s
is
d
o
n
e.
T
h
r
o
u
g
h
o
u
t
th
e
tr
ai
n
in
g
d
ev
elo
p
m
en
t,
th
e
ex
tr
ac
ted
f
ea
tu
r
es
ar
e
tr
ain
ed
in
AR
B
NN
an
d
th
e
f
in
al
tr
ain
ed
R
B
F
s
tr
u
ctu
r
e
is
d
ep
o
s
ited
.
T
h
en
,
th
r
o
u
g
h
o
u
t
th
e
test
in
g
d
ev
elo
p
m
en
t,
b
ase
d
o
n
th
e
tr
a
in
ed
c
o
n
s
tr
u
ctio
n
th
e
s
eg
m
en
tatio
n
is
ca
r
r
ie
d
o
u
t
.
T
r
ain
in
g
p
r
o
ce
s
s
:
f
o
r
t
h
e
tr
ai
n
in
g
p
r
o
ce
s
s
,
AR
B
F
n
eu
r
al
n
etwo
r
k
is
u
tili
ze
d
.
L
et
u
s
ass
u
m
e
1
,
2
,
.
.
.
,
d
e
n
o
tes
as
a
v
ec
to
r
o
f
in
p
u
t
n
o
d
es
1
,
2
,
.
.
.
,
r
ep
r
esen
ts
a
v
ec
t
o
r
o
f
h
id
d
en
lay
er
n
o
d
es
an
d
1
,
2
,
.
.
.
,
r
ep
r
esen
ts
a
v
ec
to
r
o
f
o
u
tp
u
t
lay
e
r
n
o
d
es.
Ad
d
itio
n
ally
,
th
e
weig
h
t
a
m
o
n
g
th
e
h
id
d
en
an
d
i
n
p
u
t
is
r
e
p
r
esen
ted
as
ℎ
an
d
th
e
weig
h
t
b
etwe
en
th
e
h
id
d
en
a
n
d
o
u
tp
u
t
lay
er
s
.
T
h
e
f
o
llo
w
in
g
d
escr
ib
es th
e
tr
ain
in
g
p
r
o
ce
d
u
r
e.
−
Step
1
:
i
n
itially
,
th
e
ex
tr
ac
te
d
f
ea
tu
r
es
(
1
,
2
,
.
.
.
,
63
)
ar
e
f
ed
t
o
th
e
i
n
p
u
t
lay
er
o
f
R
B
FNN.
T
h
e
ex
tr
ac
ted
f
ea
tu
r
es
an
d
i
n
p
u
t
la
y
er
n
e
u
r
o
n
s
a
r
e
id
le.
Fo
llo
win
g
in
p
u
t,
th
e
weig
h
t
q
u
a
n
tity
c
o
r
r
elatin
g
t
o
th
e
in
p
u
t
an
d
h
id
d
en
n
eu
r
o
n
is
c
o
m
p
o
u
n
d
ed
b
y
th
e
f
ea
t
u
r
e
v
a
lu
es.
T
h
e
h
id
d
en
lay
er
'
s
in
p
u
t,
as
s
p
ec
if
ied
in
(
8
)
,
r
ec
ei
v
es th
e
p
r
o
d
u
ce
d
o
u
t
p
u
t is g
iv
en
b
elo
w.
W
h
er
e
r
ep
r
esen
t th
e
b
ias v
alu
e
o
f
th
e
h
i
d
d
en
lay
e
r
.
=
+
∑
ℎ
=
1
(
8
)
−
Step
2
:
t
h
en
,
th
e
o
b
tain
ed
is
f
ed
to
th
e
g
au
s
s
ian
R
B
F
ac
ti
v
atio
n
f
u
n
ctio
n
.
T
h
is
f
u
n
ctio
n
m
o
n
o
to
n
o
u
s
r
ed
u
ce
s
th
e
d
etac
h
m
en
t
t
h
r
o
u
g
h
th
e
ce
n
te
r
.
I
t
is
d
escr
ib
e
d
b
y
its
ce
n
ter
:
=
1
,
2
,
.
.
.
,
an
d
co
v
ar
ian
ce
m
atr
ices
=
2
.
T
h
e
i
n
p
u
t
v
ec
to
r
with
ℎ
a
h
id
d
en
u
n
it
is
g
i
v
en
i
n
(
9
)
.
W
h
er
e,
r
ep
r
esen
t
th
e
m
ax
im
u
m
d
is
tan
ce
an
d
is
a
n
em
p
ir
ical
s
ca
le
f
ac
to
r
th
at
is
u
s
ed
to
r
eg
u
late
th
e
s
m
o
o
th
n
ess
o
f
th
e
m
ap
p
in
g
f
u
n
ctio
n
.
T
h
e
(
1
0
)
is
s
u
b
s
titu
ted
to
(
9
)
a
n
d
th
e
o
u
tp
u
t la
y
er
is
g
iv
en
b
elo
w.
(
)
=
(
−
‖
−
‖
2
2
2
)
(
9
)
2
=
2
2
(
1
0
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
1
,
Ju
ly
20
25
:
202
-
2
1
3
206
(
)
=
(
−
‖
−
‖
2
2
)
(
1
1
)
−
Step
3
:
f
o
llo
win
g
th
e
ac
tiv
atio
n
co
m
p
u
tatio
n
,
we
d
eter
m
in
e
th
e
R
B
FNN
's
o
u
tco
m
e.
T
h
e
o
u
tp
u
t
lay
er
is
p
r
o
v
id
e
d
th
e
ac
tiv
atio
n
f
u
n
cti
o
n
'
s
r
esu
lt
in
th
is
ca
s
e.
T
h
e
(
1
2
)
d
escr
ib
es
th
e
o
u
tp
u
t
f
u
n
c
tio
n
.
W
h
er
e
d
en
o
tes
th
e
o
u
tp
u
t
la
y
er
s
.
T
h
e
(
1
3
)
t
o
d
eter
m
in
e
th
e
n
etwo
r
k
'
s
lear
n
in
g
er
r
o
r
.
W
h
er
e,
n
r
e
p
r
esen
ts
th
e
tr
ain
in
g
in
f
o
r
m
ati
o
n
,
r
ep
r
esen
t
th
e
tar
g
et
an
d
th
e
o
u
tp
u
t
v
alu
e.
Ad
ju
s
tin
g
t
h
e
weig
h
t
v
alu
es
is
n
ec
ess
ar
y
to
lo
wer
th
e
er
r
o
r
v
al
u
e.
I
n
th
is
p
ap
er
,
we
em
p
l
o
y
ed
t
h
e
FP
alg
o
r
ith
m
to
d
o
th
is
.
=
+
∑
=
1
(
)
,
=
1
,
2
,
.
.
.
,
(
1
2
)
=
1
2
∑
√
(
−
)
−
1
=
0
(
1
3
)
2
.
3
.
1
.
O
ptim
a
l
weig
ht
s
elec
t
io
n us
ing
F
P
a
lg
o
rit
hm
T
o
r
eg
u
late
th
e
s
tan
d
a
r
d
s
o
f
t
h
e
R
B
F
n
eu
r
al
n
etwo
r
k
,
th
e
FP
alg
o
r
ith
m
is
u
tili
ze
d
.
De
p
en
d
in
g
o
n
h
o
w
f
lo
wer
s
p
o
llin
ate
o
n
e
a
n
o
th
er
,
an
alg
o
r
ith
m
f
o
r
f
lo
wer
p
o
llin
atio
n
was
cr
ea
ted
.
Gen
er
ally
,
f
lo
wer
p
o
llin
atio
n
in
v
o
lv
es
th
e
m
o
v
e
m
en
t
o
f
p
o
llen
,
wh
ic
h
is
co
m
m
o
n
ly
ass
o
ciate
d
with
p
o
llin
ato
r
s
lik
e
b
ir
d
s
.
Sin
ce
s
o
m
e
f
lo
wer
s
m
ay
r
ec
r
u
it
an
d
r
ely
o
n
s
p
ec
if
ic
k
in
d
s
o
f
i
n
s
ec
ts
o
r
b
ir
d
s
f
o
r
s
u
cc
ess
f
u
l
p
o
llin
atio
n
,
s
o
m
e
f
lo
wer
s
an
d
i
n
s
ec
ts
h
av
e
a
v
e
r
y
s
k
illed
f
lo
wer
-
p
o
llin
ato
r
co
o
p
er
atio
n
.
T
h
e
m
ajo
r
ity
o
f
p
l
an
t
s
p
ec
ies
r
ely
o
n
b
io
tic
p
o
llin
atio
n
,
i
n
wh
ich
p
o
llin
ato
r
s
s
p
r
ea
d
p
o
lle
n
.
T
h
e
r
est
o
f
p
o
llin
atio
n
o
b
s
er
v
es
a
n
ab
io
tic
p
h
ase
th
at
d
o
es
n
o
t
d
em
a
n
d
a
n
y
p
o
llin
at
o
r
s
,
lik
e
g
r
ass
;
s
u
ch
f
l
o
wer
in
g
p
lan
ts
'
p
o
llin
atio
n
wo
r
k
is
s
u
p
p
o
r
ted
b
y
win
d
an
d
d
if
f
u
s
io
n
.
On
th
e
o
th
er
h
a
n
d
,
s
elf
-
p
o
llin
atio
n
o
r
cr
o
s
s
-
p
o
llin
atio
n
ca
n
b
e
u
s
ed
f
o
r
p
o
llin
atio
n
.
Self
-
p
o
llin
atio
n
is
wh
en
o
n
e
f
lo
wer
is
p
o
llin
at
ed
b
y
t
h
e
p
o
llen
o
f
an
o
th
e
r
f
l
o
wer
o
n
th
e
s
am
e
p
lan
t
o
r
a
n
o
th
er
f
lo
wer
.
C
r
o
s
s
-
p
o
llin
atio
n
is
th
e
p
r
o
ce
s
s
o
f
p
o
llin
atin
g
a
b
l
o
o
m
f
r
o
m
a
d
if
f
er
en
t
p
lan
t.
T
h
e
FPA
ca
n
ea
s
ily
s
o
lv
e
lo
w
-
d
im
en
s
io
n
al
u
n
i
-
m
o
d
al
o
p
ti
m
izatio
n
is
s
u
es
an
d
f
all
o
n
lo
ca
l
o
p
tim
u
m
.
W
h
en
tak
i
n
g
ca
r
e
o
f
th
e
h
ig
h
d
im
en
s
io
n
al
an
d
m
u
lti
-
m
o
d
u
l
ar
en
h
an
ce
m
en
t
is
s
u
es,
we
ca
n
f
in
d
th
at
th
e
s
o
lu
tio
n
g
o
t
b
y
FP
A
ar
e
s
u
f
f
icien
tly
b
ad
.
T
o
en
h
an
ce
th
e
g
lo
b
al
s
e
ar
ch
in
g
an
d
lo
ca
l
s
ea
r
ch
in
g
ab
ilit
ies,
th
e
f
ir
ef
ly
alg
o
r
ith
m
o
p
er
atio
n
is
in
clu
d
ed
in
th
e
lo
ca
l p
o
llin
atio
n
.
Stag
e
1
:
s
o
lu
tio
n
en
co
d
in
g
:
t
h
is
alg
o
r
ith
m
'
s
p
r
im
ar
y
g
o
al
i
s
to
d
eter
m
in
e
th
e
id
ea
l
weig
h
t
r
atio
am
o
n
g
th
e
h
id
d
en
in
p
u
t
an
d
h
id
d
en
o
u
tp
u
t
lay
er
s
.
A
f
u
n
d
am
e
n
tal
co
m
p
o
n
en
t
o
f
o
p
tim
izatio
n
is
cr
ea
tin
g
an
ea
r
ly
s
o
lu
tio
n
.
On
ly
af
ter
th
e
s
o
lu
ti
o
n
h
as
b
ee
n
d
eter
m
in
ed
ca
n
th
e
alg
o
r
ith
m
b
e
a
d
v
an
ce
d
.
I
n
i
tially
,
th
e
s
o
lu
tio
n
s
o
r
f
l
o
wer
s
(
)
ar
e
a
r
b
itra
r
ily
in
it
ialized
.
T
h
e
in
itial
s
o
lu
tio
n
is
g
iv
en
in
(
1
4
)
.
T
h
e
s
o
lu
tio
n
i
s
co
n
s
is
tin
g
o
f
weig
h
t v
alu
es.
n
F
F
F
i
S
...,
,
2
,
1
=
(
1
4
)
Step
2
:
f
itn
ess
ca
lcu
latio
n
:
ut
il
izin
g
f
itn
ess
v
alu
e,
th
e
s
o
lu
tio
n
'
s
u
s
ef
u
ln
ess
is
ev
alu
ated
.
W
e
ass
es
s
th
e
f
itn
ess
o
f
ea
c
h
p
r
o
p
o
s
al
f
o
llo
win
g
in
itiatio
n
.
T
h
e
f
itn
ess
f
u
n
ctio
n
i
s
u
s
ed
to
d
ef
in
e
s
eg
m
en
tatio
n
r
esu
lts
.
T
h
e
id
ea
l
ap
p
r
o
ac
h
is
r
eg
ar
d
ed
as h
av
i
n
g
th
e
h
ig
h
est f
itn
ess
v
alu
e.
In
(
1
5
)
c
o
n
tain
s
th
e
f
itn
ess
f
u
n
ctio
n
.
=
(
)
(
1
5
)
=
+
+
+
+
(
1
6
)
Step
3
:
u
p
d
atio
n
u
s
in
g
E
FP
A:
a
f
ter
th
e
f
itn
ess
ca
lcu
latio
n
,
e
ac
h
f
lo
wer
is
u
p
d
ated
with
th
e
h
elp
o
f
E
FP
A.
I
n
FP
A,
two
ty
p
es
o
f
p
o
llin
atio
n
ar
e
av
ailab
le
n
a
m
ely
,
g
l
o
b
al
p
o
llin
atio
n
an
d
lo
ca
l
p
o
llin
atio
n
.
T
o
en
h
a
n
ce
th
e
p
er
f
o
r
m
an
ce
o
f
FP
A,
lo
ca
l
p
o
llin
atio
n
is
r
ep
lace
d
with
th
e
h
elp
o
f
th
e
f
ir
ef
ly
al
g
o
r
ith
m
.
Glo
b
al
p
o
llin
atio
n
ca
n
b
e
r
e
p
r
esen
ted
m
ath
e
m
atica
lly
as (
1
7
)
.
+
1
=
+
(
)
(
∗
−
)
(
1
7
)
W
h
er
e,
is
th
e
p
o
llen
o
r
r
eso
l
u
tio
n
v
ec
to
r
at
em
p
h
asis
,
an
d
is
th
e
p
r
esen
t
b
est
s
o
lu
tio
n
f
o
u
n
d
am
o
n
g
all
s
o
lu
tio
n
s
at
th
e
p
r
esen
t
ag
e/cy
cle.
T
h
e
p
r
o
g
r
ess
io
n
s
ize
is
co
n
s
tr
ain
ed
b
y
a
s
ca
lin
g
o
p
er
ato
r
.
Her
e,
(
)
m
ea
n
s
th
e
L
ev
y
f
lig
h
t
-
b
ased
ad
v
an
ce
s
ize
th
at
r
elat
es
to
th
e
ca
p
ac
ity
o
f
p
o
llin
ati
o
n
.
T
h
e
l
o
ca
l
p
o
llin
atio
n
an
d
f
lo
wer
co
n
s
i
s
ten
cy
ca
n
b
e
d
em
o
n
s
tr
ated
to
as
b
elo
w.
I
n
th
at
o
v
er
h
ea
d
eq
u
atio
n
,
+
1
d
em
o
n
s
tr
ates
th
e
f
r
esh
u
p
d
at
ed
s
o
lu
tio
n
,
illu
s
tr
ates
th
e
p
r
esen
t
ℎ
s
o
lu
tio
n
an
d
p
er
f
o
r
m
s
th
e
ℎ
s
o
lu
tio
n
.
Mo
r
e
o
v
er
,
s
h
o
ws th
e
ar
b
itra
r
y
f
ac
to
r
a
n
d
is
a
r
an
d
o
m
n
u
m
b
er
an
d
is
th
e
co
n
s
tan
t v
alu
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
n
efficien
t seg
men
ta
tio
n
u
s
in
g
a
d
a
p
tive
r
a
d
ia
l
b
a
s
is
fu
n
ctio
n
n
eu
r
a
l
n
etw
o
r
k
…
(
Jo
la
ku
l
a
A
s
o
ka
S
mith
a
)
207
+
1
=
−
0
−
̈
2
(
−
)
+
(
1
8
)
2
.
3
.
2
.
Seg
m
ent
a
t
io
n
I
m
ag
es
ca
n
b
e
s
eg
m
en
te
d
u
s
in
g
a
m
eth
o
d
k
n
o
wn
as
class
if
icatio
n
.
T
h
e
im
ag
e'
s
p
ix
els
ea
ch
r
ep
r
esen
t
a
p
atter
n
th
at
ca
n
b
e
class
ed
.
W
e
em
p
lo
y
f
r
esh
,
u
n
tr
ain
ed
p
h
o
to
s
f
o
r
test
in
g
p
u
r
p
o
s
es.
W
e
u
s
ed
ev
er
y
p
ix
el
in
th
e
test
ed
im
ag
e
as
w
ell
a
s
its
s
u
r
r
o
u
n
d
in
g
p
i
x
els,
ju
s
t
lik
e
d
u
r
in
g
tr
ai
n
in
g
.
T
h
e
test
in
g
p
r
o
ce
d
u
r
e
m
a
k
es
u
s
e
o
f
th
e
g
en
er
ated
R
B
F
n
eu
r
al
n
etwo
r
k
.
A
tr
ain
ed
E
R
B
F
cla
s
s
if
ier
u
s
e
s
th
e
attr
ib
u
tes
to
i
d
en
tify
th
e
class
to
wh
ich
th
e
s
am
p
les co
r
r
esp
o
n
d
to
ch
ar
ac
ter
ize
th
e
im
ag
e.
I
t
will b
e
ca
teg
o
r
ized
as a
d
am
a
g
ed
class
if
th
e
v
alu
e
is
lar
g
er
th
an
0
.
5
; e
ls
e,
it will b
e
ca
teg
o
r
ized
as a
b
ac
k
g
r
o
u
n
d
class
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
NS
T
h
e
s
u
g
g
ested
leaf
d
is
ea
s
e
s
eg
m
en
tatio
n
m
o
d
el
is
s
im
u
la
ted
in
t
h
e
p
latf
o
r
m
o
f
Py
th
o
n
with
th
e
s
y
s
tem
h
av
in
g
an
I
n
tel
C
o
r
e
i5
p
r
o
ce
s
s
o
r
,
a
n
d
o
n
a
co
m
p
u
t
er
with
6
GB
o
f
m
em
o
r
y
u
s
in
g
th
e
W
in
d
o
ws
1
0
OS.
I
n
th
is
wo
r
k
,
two
ty
p
es
o
f
p
la
n
t
leav
es
ar
e
u
s
ed
f
o
r
e
x
p
er
im
en
tatio
n
n
am
el
y
,
to
m
ato
an
d
m
an
g
o
.
T
h
e
im
ag
es
ar
e
co
llected
f
r
o
m
th
e
p
lan
t
v
illag
e
d
ataset.
T
h
is
d
ataset
in
clu
d
es
s
ev
er
al
k
in
d
s
o
f
d
is
ea
s
es
in
to
m
ato
an
d
p
lan
t
leav
es.
I
n
th
e
tr
ain
in
g
d
ataset,
1
,
0
0
0
im
ag
es
ar
e
in
clu
d
ed
f
o
r
ea
ch
d
is
ea
s
e.
As
w
ell,
1
0
0
im
ag
es
ar
e
in
clu
d
ed
f
o
r
ea
c
h
d
is
ea
s
e
in
th
e
test
in
g
d
ataset.
Fig
u
r
e
3
s
h
o
ws
s
eg
m
en
tatio
n
o
u
t
p
u
t
o
f
to
m
ato
leaf
.
T
h
is
f
ig
u
r
e
co
n
tain
s
4
co
m
p
o
n
en
ts
s
u
ch
as:
o
r
ig
in
al
im
ag
e,
n
o
is
e
r
em
o
v
ed
im
a
g
e,
co
n
tr
ast
en
h
an
ce
m
en
t
im
ag
e,
s
eg
m
en
ted
o
u
tp
u
t.
Fin
ally
,
th
e
p
r
o
p
o
s
ed
m
o
d
el
d
ec
id
es
th
e
h
ea
lth
y
o
r
s
p
ec
if
ie
d
d
is
ea
s
e
n
am
e.
T
h
is
f
ig
u
r
e
s
h
o
ws
th
e
f
iv
e
d
if
f
er
e
n
t
leav
es
s
am
p
les.
T
h
e
o
r
ig
in
al
lea
v
es
im
ag
e
n
o
is
e
is
r
em
o
v
e
d
th
en
g
o
t
o
c
o
n
tr
ast
en
h
an
ce
m
e
n
t
im
ag
e.
Nex
t,
p
r
o
v
id
e
th
e
s
eg
m
en
t
o
u
tp
u
t
an
d
th
is
o
u
tp
u
t
d
ec
id
es
b
ac
ter
ial
s
p
o
t,
E
ar
ly
_
b
lig
h
t,
L
ate
_
b
lig
h
t.
Fig
u
r
e
4
s
h
o
ws
th
e
s
eg
m
en
tatio
n
o
u
tp
u
t
o
f
m
an
g
o
leaf
.
T
h
i
s
f
ig
u
r
e
co
n
tain
s
5
co
m
p
o
n
en
t
s
s
u
ch
as
:
o
r
ig
in
al
im
ag
e,
n
o
is
e
r
em
o
v
e
d
im
ag
e,
co
n
tr
ast
en
h
an
ce
m
e
n
t
im
ag
e,
s
eg
m
en
ted
o
u
tp
u
t.
Fin
ally
,
th
e
p
r
o
p
o
s
ed
m
o
d
el
d
ec
id
es
th
e
d
is
ea
s
e
n
am
e
o
r
h
ea
lth
y
.
An
t
h
r
ac
n
o
s
e.
T
h
is
f
ig
u
r
e
s
h
o
ws
th
e
f
iv
e
d
if
f
e
r
en
t
leav
es
s
a
m
p
les.
T
h
e
o
r
i
g
in
al
leav
es
im
ag
e
n
o
is
e
is
r
em
o
v
ed
th
en
g
o
to
c
o
n
tr
ast
en
h
a
n
ce
m
en
t
im
a
g
e.
Nex
t,
p
r
o
v
id
e
th
e
s
eg
m
en
t o
u
tp
u
t a
n
d
t
h
is
o
u
tp
u
t d
ec
id
es An
th
r
ac
n
o
s
e
d
is
ea
s
e.
Fig
u
r
e
3
.
Seg
m
e
n
tatio
n
o
u
tp
u
t
o
f
to
m
ato
leaf
3
.
1
.
P
er
f
o
r
m
a
nce
a
na
ly
s
is
o
f
t
o
ma
t
o
le
a
f
s
eg
m
ent
a
t
io
n
T
h
is
s
ec
tio
n
ex
p
lain
s
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
is
m
ea
s
u
r
ed
ag
ain
s
t
th
e
c
u
r
r
en
tly
u
s
ed
s
eg
m
en
tatio
n
m
eth
o
d
s
,
s
u
ch
as
tr
a
d
itio
n
al
R
B
F,
FC
M,
an
d
r
eg
io
n
e
x
p
an
d
in
g
(
R
G)
.
W
e
co
n
cl
u
d
ed
f
r
o
m
th
is
in
v
esti
g
atio
n
th
at
th
e
s
eg
m
en
tatio
n
s
y
s
tem
d
ep
en
d
i
n
g
o
n
th
e
s
u
g
g
ested
m
eth
o
d
o
l
o
g
y
y
ield
s
m
o
r
e
ac
c
u
r
ate
o
u
tc
o
m
es
th
an
th
e
R
B
F,
F
C
M,
an
d
R
G.
T
h
e
co
m
p
ar
is
o
n
o
f
p
r
ec
is
io
n
,
r
ec
al
l,
F
-
m
ea
s
u
r
e,
ac
cu
r
ac
y
,
J
ac
ca
r
d
co
ef
f
icien
t
(
J
C
)
,
an
d
Dice
co
ef
f
icien
t
(
DC
)
o
f
d
if
f
er
en
t
s
eg
m
en
tatio
n
m
o
d
e
ls
s
u
ch
as
R
B
F,
FC
M
an
d
R
G
is
illu
s
tr
ated
in
F
i
g
u
r
es 5
to
1
0
.
B
ased
o
n
th
e
af
o
r
em
en
tio
n
ed
ex
p
lan
atio
n
s
:
(
i)
t
h
e
FC
M
a
p
p
r
o
ac
h
d
e
g
r
ad
es
wh
e
n
in
itia
lizin
g
th
e
clu
s
ter
ce
n
ter
a
n
d
ca
lc
u
latin
g
th
e
n
u
m
b
e
r
o
f
clu
s
ter
s
;
(
ii)
th
e
R
G
ap
p
r
o
ac
h
s
u
f
f
er
s
f
r
o
m
o
v
er
-
s
eg
m
en
tatio
n
an
d
tak
es
a
lo
n
g
tim
e
;
a
n
d
(
iii
)
th
e
s
ta
n
d
ar
d
R
B
F
ap
p
lies
th
e
s
am
e
weig
h
t
to
ea
ch
attr
ib
u
t
e,
wh
ich
m
ay
h
av
e
an
im
p
ac
t
o
n
s
eg
m
en
tatio
n
a
cc
u
r
ac
y
.
As
illu
s
tr
ated
in
th
e
f
ig
u
r
e,
R
B
F,
FC
M,
an
d
R
G
attain
ed
9
5
.
0
8
%,
9
3
.
9
7
%,
an
d
9
1
.
7
3
%
o
f
p
r
e
cisi
o
n
co
r
r
esp
o
n
d
in
g
ly
.
B
u
t
th
e
s
u
g
g
ested
m
eth
o
d
attain
ed
th
e
m
ax
im
u
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
1
,
Ju
ly
20
25
:
202
-
2
1
3
208
p
r
ec
is
io
n
i.e
.
,
9
6
.
7
8
%.
T
h
e
r
elativ
e
in
v
esti
g
atio
n
o
f
th
e
r
ec
all
o
f
d
is
s
im
ilar
s
eg
m
en
tatio
n
s
im
u
latio
n
s
is
d
ep
icted
in
F
ig
u
r
e
6
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
-
b
ased
s
eg
m
en
tatio
n
m
o
d
el
attain
ed
9
6
.
9
3
%
o
f
th
e
r
ec
all.
T
h
e
F
-
m
ea
s
u
r
e
o
f
d
if
f
er
en
t
s
eg
m
e
n
tatio
n
m
et
h
o
d
s
is
d
ep
icted
in
F
ig
u
r
e
6
.
C
o
m
p
a
r
ed
to
R
B
F,
FC
M,
an
d
R
G,
th
e
F
-
m
ea
s
u
r
e
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
is
in
cr
ea
s
ed
to
9
7
.
0
6
%.
Fig
u
r
e
7
d
em
o
n
s
tr
ates
th
e
ass
es
s
m
en
t
o
f
th
e
ac
cu
r
ac
y
o
f
v
ar
i
o
u
s
s
eg
m
en
t
atio
n
ap
p
r
o
ac
h
es.
B
ec
au
s
e
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
-
b
ased
s
eg
m
en
tat
io
n
,
th
e
ac
cu
r
ac
y
is
in
c
r
ea
s
ed
to
9
7
.
5
8
%
wh
ile
th
e
ex
is
tin
g
m
o
d
els
R
B
F,
FC
M,
an
d
R
G
attain
9
4
.
9
1
%,
9
3
.
0
8
%,
an
d
9
0
.
7
2
%
c
o
r
r
esp
o
n
d
in
g
ly
.
Fig
u
r
e
8
illu
s
tr
ates
an
ass
ess
m
en
t
o
f
th
e
JC
o
f
v
ar
io
u
s
s
eg
m
en
tat
io
n
ap
p
r
o
ac
h
es.
As
illu
s
tr
ated
in
th
e
f
ig
u
r
e,
c
o
m
p
ar
ed
t
o
FC
M
an
d
R
G
-
b
ased
s
eg
m
en
tatio
n
m
o
d
els,
t
h
e
co
n
v
e
n
tio
n
al
R
B
F
m
o
d
el
attain
ed
th
e
h
ig
h
est
JC
i.e
.
,
9
2
.
0
5
%.
T
h
u
s
,
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
attain
ed
9
4
.
8
3
%
o
f
th
e
JC
.
T
h
e
co
m
p
a
r
ativ
e
an
aly
s
is
o
f
th
e
DC
i
s
d
ep
icted
in
F
ig
u
r
e
9
.
As
p
o
r
tr
ay
e
d
in
th
e
F
ig
u
r
e
1
0
,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
-
b
ased
s
eg
m
en
tatio
n
attain
ed
9
2
.
8
4
% o
f
th
e
DC
.
Fig
u
r
e
4
.
A
n
t
h
r
ac
n
o
s
e
d
is
ea
s
e
s
eg
m
en
tatio
n
o
u
tp
u
t
Fig
u
r
e
5
.
Pre
cisi
o
n
o
f
d
if
f
er
en
t seg
m
en
tatio
n
Fig
u
r
e
6
.
R
ec
all
o
f
d
if
f
er
e
n
t seg
m
en
tatio
n
Fig
u
r
e
7
.
F
-
m
ea
s
u
r
e
o
f
d
if
f
er
e
n
t seg
m
en
tatio
n
Fig
u
r
e
8
.
Acc
u
r
ac
y
o
f
d
if
f
er
en
t seg
m
en
tatio
n
88
90
92
94
96
98
P
r
opos
e
d
meth
od
E
xis
ti
ng
RBF
E
xis
ti
ng
F
C
M
E
xis
ti
ng
RG
Pr
e
c
i
si
o
n
S
e
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
85
90
95
100
P
ropos
e
d
m
e
t
h
od
E
xi
s
t
i
ng
RBF
E
xi
s
t
i
ng
F
CM
E
xi
s
t
i
ng
RG
R
e
c
a
l
l
Seg
m
e
nt
a
t
i
o
n
m
o
del
80
90
1
0
0
P
ropos
e
d
m
e
t
h
od
E
xi
s
t
i
ng
RBF
E
xi
s
t
i
ng
F
CM
E
xi
s
t
i
ng
RG
F
-
m
e
a
su
r
e
S
e
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
s
86
88
90
92
94
96
98
1
0
0
P
r
o
p
o
sed
me
t
h
o
d
E
x
i
st
i
n
g
R
B
F
Ex
i
st
i
n
g
F
C
M
Ex
i
st
i
n
g
RG
A
c
c
u
r
a
c
y
S
e
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
n
efficien
t seg
men
ta
tio
n
u
s
in
g
a
d
a
p
tive
r
a
d
ia
l
b
a
s
is
fu
n
ctio
n
n
eu
r
a
l
n
etw
o
r
k
…
(
Jo
la
ku
l
a
A
s
o
ka
S
mith
a
)
209
Fig
u
r
e 9
.
JC
o
f
d
if
f
er
en
t
s
eg
m
en
tatio
n
Fig
u
r
e
1
0
.
DC
o
f
d
if
f
e
r
en
t seg
m
en
tatio
n
3
.
2
.
P
er
f
o
r
m
a
nce
a
na
ly
s
is
o
f
m
a
ng
o
lea
f
s
eg
m
ent
a
t
io
n
Fig
u
r
e
1
1
r
e
p
r
esen
ts
th
e
co
m
p
ar
is
o
n
o
f
p
r
ec
is
io
n
with
d
if
f
er
en
t
s
eg
m
en
tatio
n
m
o
d
els
s
u
ch
as
R
B
F,
FC
M
,
an
d
R
G.
As
illu
s
tr
ated
in
th
e
f
ig
u
r
e,
R
B
F,
FC
M,
an
d
R
G
attain
ed
9
4
.
0
8
%,
9
2
.
9
7
%,
an
d
9
0
.
8
6
%
o
f
p
r
ec
is
io
n
co
r
r
esp
o
n
d
i
n
g
ly
.
B
u
t
th
e
an
ticip
ated
m
eth
o
d
attain
ed
th
e
m
a
x
im
u
m
p
r
ec
is
io
n
i.e
.
,
9
5
.
9
7
%.
T
h
e
q
u
alif
ied
in
v
esti
g
atio
n
o
f
t
h
e
r
ec
all
o
f
v
ar
io
u
s
s
eg
m
en
tatio
n
ap
p
r
o
ac
h
es is
d
ep
icted
in
F
ig
u
r
e
1
2
.
As
d
ep
icted
in
th
e
f
ig
u
r
e,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
-
b
ased
s
eg
m
en
tatio
n
m
o
d
el
attain
ed
9
5
.
9
3
%
o
f
th
e
r
ec
all.
T
h
e
F
-
m
ea
s
u
r
e
o
f
v
ar
i
o
u
s
s
eg
m
en
tatio
n
r
ep
r
esen
tati
o
n
s
is
d
ep
icted
in
F
ig
u
r
e
1
3
.
C
o
m
p
ar
e
d
to
R
B
F
,
FC
M,
an
d
R
G,
th
e
F
-
m
ea
s
u
r
e
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
is
in
cr
ea
s
ed
to
9
6
.
0
6
%.
Fig
u
r
e
1
4
d
em
o
n
s
tr
ates
th
e
ass
es
s
m
en
t
o
f
th
e
ac
cu
r
ac
y
o
f
v
a
r
io
u
s
s
eg
m
e
n
tatio
n
s
im
u
latio
n
s
.
B
ec
au
s
e
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
-
b
ased
s
eg
m
en
tatio
n
m
o
d
el,
th
e
ac
c
u
r
ac
y
is
im
p
r
o
v
ed
to
9
6
.
5
8
%
t
h
o
u
g
h
th
e
e
x
is
tin
g
s
im
u
latio
n
s
R
B
F,
FC
M,
an
d
R
G
attain
9
3
.
9
1
%,
9
2
.
0
8
%,
a
n
d
8
9
.
7
2
%
co
r
r
esp
o
n
d
in
g
ly
.
Fig
u
r
e
1
5
illu
s
tr
ates
an
ass
ess
m
en
t
o
f
th
e
JC
o
f
v
ar
io
u
s
s
eg
m
en
tatio
n
ap
p
r
o
ac
h
es.
As
illu
s
tr
ated
in
th
e
f
ig
u
r
e,
co
m
p
ar
e
d
to
FC
M
an
d
R
G
-
b
ased
s
eg
m
en
tatio
n
m
o
d
els,
th
e
co
n
v
e
n
tio
n
al
R
B
F
m
o
d
el
attain
ed
th
e
h
ig
h
est
JC
i.e
.
,
9
0
.
0
5
%.
T
h
u
s
,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
attain
ed
9
3
.
8
3
%
o
f
th
e
JC
.
T
h
e
co
m
p
ar
ativ
e
an
al
y
s
is
o
f
th
e
DC
o
f
d
if
f
er
en
t
s
eg
m
en
tatio
n
m
o
d
els
is
d
ep
icted
in
F
ig
u
r
e
1
6
.
As
d
e
p
icted
i
n
t
h
e
f
ig
u
r
e,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
-
b
ased
s
eg
m
e
n
tatio
n
m
o
d
el
attain
ed
9
1
.
8
4
%
o
f
th
e
DC
.
Fig
u
r
e
1
1
.
Pre
cisi
o
n
o
f
d
if
f
er
e
n
t seg
m
en
tatio
n
Fig
u
r
e
1
2
.
R
ec
all
o
f
d
if
f
e
r
en
t
s
eg
m
en
tatio
n
Fig
u
r
e
1
3
.
F
-
m
ea
s
u
r
e
o
f
d
if
f
e
r
en
t seg
m
en
tatio
n
Fig
u
r
e
1
4
.
Acc
u
r
ac
y
o
f
d
if
f
er
e
n
t seg
m
en
tatio
n
80
90
1
0
0
P
r
opos
e
d
meth
od
E
xis
ti
ng
RBF
E
xis
ti
ng
F
C
M
E
xis
ti
ng
RG
J
a
c
c
a
r
d
c
o
e
fficie
n
t
S
e
g
m
e
n
ta
tio
n
m
o
d
e
ls
80
90
100
P
r
opos
e
d
m
e
thod
E
xi
s
ti
ng
R
B
F
E
xi
s
ti
ng
F
C
M
E
xi
s
ti
ng
RG
D
i
ce
co
ef
f
i
ci
ent
S
e
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
s
80
90
1
0
0
Pre
c
isio
n
S
e
g
m
e
n
ta
tio
n
m
o
d
e
ls
80
90
1
0
0
P
r
opos
e
d
meth
od
E
xis
ti
ng
RBF
E
xis
ti
ng
F
C
M
E
xis
ti
ng
RG
R
e
c
a
l
l
S
e
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
s
85
90
95
1
0
0
P
r
opos
e
d
method
E
xi
s
ti
ng
R
B
F
E
xi
s
ti
ng
F
C
M
E
xi
s
ti
ng
RG
F
-
m
e
a
su
r
e
S
e
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
s
86
88
90
92
94
96
98
P
r
opos
e
d
me
th
od
E
xis
ti
ng
RBF
E
xis
ti
ng
F
C
M
E
xis
ti
ng
R
G
A
c
c
u
r
a
c
y
S
e
g
m
e
n
d
a
t
i
o
n
m
o
d
e
l
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
9
,
No
.
1
,
Ju
ly
20
25
:
202
-
2
1
3
210
Fig
u
r
e
1
5
.
JC
o
f
d
if
f
er
e
n
t seg
m
en
tatio
n
Fig
u
r
e
1
6
.
DC
o
f
d
if
f
e
r
en
t seg
m
en
tatio
n
3
.
3
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
o
f
prev
io
us
wo
rk
s
T
h
is
s
ec
tio
n
d
ep
icts
th
e
co
m
p
ar
ativ
e
in
v
esti
g
atio
n
o
f
th
e
s
u
g
g
ested
wo
r
k
with
th
e
p
r
ev
i
o
u
s
wo
r
k
lik
e
p
r
ec
is
io
n
,
r
ec
all,
F
-
m
ea
s
u
r
e
,
ac
cu
r
ac
y
,
J
C
,
an
d
DC
.
A
g
a
r
w
a
l
e
t
a
l
.
[
9
]
h
ad
o
b
tai
n
ab
le
a
d
ee
p
lear
n
in
g
C
NN
alg
o
r
ith
m
f
o
r
to
m
at
o
leaf
s
eg
m
en
tatio
n
.
As
a
r
esu
lt
o
f
th
is
alg
o
r
ith
m
'
s
in
cr
ea
s
ed
u
s
ag
e
o
f
c
o
m
p
u
tatio
n
s
to
d
eter
m
in
e
th
e
v
alid
ity
in
d
ex
,
th
e
ac
cu
r
ac
y
o
f
t
h
e
wo
r
k
is
lo
wer
th
a
n
th
at
o
f
t
h
e
p
la
n
n
ed
wo
r
k
b
y
2
.
7
1
%.
Nev
er
th
eless
,
th
e
ac
cu
r
ac
y
o
f
th
e
wo
r
k
was
in
cr
ea
s
ed
to
0
.
9
7
%.
An
e
n
h
an
ce
d
R
B
FNN
is
u
s
ed
f
o
r
p
la
n
t
leaf
s
eg
m
en
tatio
n
.
T
h
e
J
C
o
f
th
e
wo
r
k
was
d
ec
r
ea
s
ed
to
1
5
%
m
o
r
e
th
an
th
at
o
f
t
h
e
p
r
o
p
o
s
e
d
wo
r
k
.
C
h
o
u
h
a
n
et
al
.
[
2
3
]
h
ad
p
r
esen
ted
a
s
eg
m
en
tatio
n
m
eth
o
d
u
s
in
g
Su
p
e
r
p
ix
el
clu
s
ter
an
d
h
y
b
r
id
n
e
u
r
al
n
etwo
r
k
.
Nam
ely
,
co
m
p
ar
ed
to
th
e
p
r
o
p
o
s
ed
wo
r
k
,
th
e
ac
cu
r
ac
y
o
f
th
e
wo
r
k
d
e
cr
ea
s
ed
to
4
.
6
%
an
d
th
e
s
p
ec
if
icity
is
9
5
.
3
4
%.
A
m
in
-
m
ax
h
u
e
h
is
to
g
r
am
an
d
K
-
m
ea
n
clu
s
ter
in
g
is
u
s
ed
f
o
r
to
m
ato
leaf
s
eg
m
en
tatio
n
.
T
h
e
J
C
o
f
th
e
wo
r
k
is
d
ec
r
ea
s
ed
to
1
7
% m
o
r
e
t
h
an
t
h
e
p
r
o
p
o
s
ed
wo
r
k
.
4.
CO
NCLU
SI
O
N
Plan
ts
ar
e
a
s
ig
n
if
ican
t o
r
ig
in
o
f
f
o
o
d
f
o
r
th
e
wo
r
ld
p
o
p
u
lati
o
n
.
Plan
t d
is
ea
s
es lea
d
to
p
r
o
d
u
ctio
n
lo
s
s
th
at
ca
n
b
e
tack
led
with
u
n
i
n
ter
r
u
p
ted
o
b
s
er
v
i
n
g
.
T
h
is
r
esear
ch
p
r
esen
ts
an
ad
a
p
tiv
e
R
B
F
n
eu
r
al
n
etwo
r
k
class
if
ier
-
b
ased
s
eg
m
en
tatio
n
alg
o
r
ith
m
t
o
en
h
a
n
ce
th
e
s
eg
m
en
tatio
n
ac
cu
r
ac
y
o
f
d
is
ea
s
ed
to
m
ato
leav
es
an
d
m
an
g
o
leav
es.
Usi
n
g
an
ad
a
p
tiv
e
m
ed
ian
f
ilter
a
n
d
h
is
to
g
r
am
eq
u
aliza
tio
n
,
n
o
is
e
th
at
was
p
r
esen
t
in
t
h
e
tr
ain
in
g
an
d
test
in
g
p
h
o
to
s
was
elim
in
ated
d
u
r
in
g
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
h
e
s
u
g
g
ested
s
eg
m
en
tatio
n
d
esig
n
ad
ap
tab
le
R
B
F n
eu
r
al
n
etwo
r
k
'
s
weig
h
t v
alu
es we
r
e
th
en
en
h
an
ce
d
u
tili
zin
g
th
e
FPA
,
em
p
lo
y
in
g
th
e
o
b
tain
e
d
co
lo
r
f
ea
tu
r
es
o
f
ea
c
h
p
ix
el
a
s
in
p
u
t.
I
n
ter
m
s
o
f
ac
c
u
r
ac
y
,
J
C
,
an
d
DC
,
th
e
p
er
f
o
r
m
an
c
e
o
f
th
e
s
u
g
g
ested
s
eg
m
en
tatio
n
m
o
d
el
h
as
b
ee
n
ex
am
in
ed
.
As
d
ep
icted
in
th
e
r
esu
lts
,
th
e
p
r
o
p
o
s
ed
s
eg
m
e
n
tatio
n
m
o
d
el
h
as
attain
ed
9
7
.
5
8
%
o
f
ac
cu
r
ac
y
.
T
h
e
p
r
o
p
o
s
ed
m
ec
h
an
is
m
o
f
f
er
s
s
u
itab
le
in
s
ig
h
ts
,
tr
ea
tm
en
ts
,
d
is
ea
s
e
av
o
id
an
ce
,
a
n
d
lead
in
g
in
e
n
h
an
ce
d
cr
o
p
y
ield
s
.
I
n
f
u
t
u
r
e
,
we
u
s
e
an
ar
tific
ial
in
tellig
en
ce
with
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
i
n
to
m
ato
an
d
m
a
n
g
o
p
la
n
t le
af
im
ag
es.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Fu
n
d
in
g
in
f
o
r
m
ati
o
n
is
n
o
t a
v
ailab
le.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
J
o
lak
u
la
Aso
k
a
Sm
ith
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
B
ich
ag
al
Sh
ad
ak
s
h
ar
ap
p
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Sh
ee
la
Par
v
ath
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Kilin
g
ar
Vee
n
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
Alb
er
t Jen
if
er
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
B
ad
d
ala
Vijay
a
Nir
m
ala
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
80
82
84
86
88
90
92
94
P
r
opos
e
d
me
tho
d
E
xi
s
ti
ng
R
B
F
E
xi
s
ti
ng
FC
M
E
xi
s
ti
ng
R
G
S
e
r
ies
1
93,
83748
90,
05756
87,
02383
85,
08263
J
a
c
c
a
r
d
c
o
e
ffiec
ien
t
80
82
84
86
88
90
92
P
r
opos
e
d
method
E
xi
s
ti
ng
R
B
F
E
xi
s
ti
ng
F
C
M
E
xi
s
ti
ng
RG
S
e
r
ies
1
91,
84627
89,
027234
86,
902742
84,
02387
D
i
c
e
c
o
e
f
f
i
c
i
e
n
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
n
efficien
t seg
men
ta
tio
n
u
s
in
g
a
d
a
p
tive
r
a
d
ia
l
b
a
s
is
fu
n
ctio
n
n
eu
r
a
l
n
etw
o
r
k
…
(
Jo
la
ku
l
a
A
s
o
ka
S
mith
a
)
211
C
:
C
o
n
c
e
p
tu
a
li
z
a
ti
o
n
M
:
M
e
th
o
d
o
l
o
g
y
So
:
So
ftwa
re
Va
:
Va
li
d
a
ti
o
n
Fo
:
Fo
rm
a
l
a
n
a
ly
sis
I
:
I
n
v
e
stig
a
ti
o
n
R
:
R
e
so
u
rc
e
s
D
:
D
a
ta Cu
ra
ti
o
n
O
:
Wr
it
in
g
-
O
ri
g
in
a
l
Dra
ft
E
:
Wr
it
in
g
-
Re
v
iew
&
E
d
it
i
n
g
Vi
:
Vi
su
a
li
z
a
ti
o
n
Su
:
Su
p
e
rv
isi
o
n
P
:
P
ro
jec
t
a
d
m
in
istrati
o
n
Fu
:
Fu
n
d
in
g
a
c
q
u
isi
ti
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
h
av
e
n
o
co
n
f
lict o
f
in
ter
est r
elev
an
t to
t
h
is
p
ap
er
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
aila
b
le
o
n
r
eq
u
est
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
au
th
o
r
,
[
J
.
A.
S
]
.
T
h
e
d
ata,
wh
ich
co
n
tain
in
f
o
r
m
atio
n
t
h
at
co
u
ld
co
m
p
r
o
m
is
e
th
e
p
r
iv
ac
y
o
f
r
esear
ch
p
ar
ticip
an
ts
,
ar
e
n
o
t p
u
b
licly
a
v
ailab
le
d
u
e
to
ce
r
tain
r
estrictio
n
s
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
F
.
D
a
n
i
l
e
v
i
c
z
a
n
d
P
.
E.
B
a
y
e
r
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
i
ma
g
e
a
n
a
l
y
si
s
:
l
e
a
f
d
i
s
e
a
s
e
s
e
g
m
e
n
t
a
t
i
o
n
,
”
i
n
M
e
t
h
o
d
s
i
n
M
o
l
e
c
u
l
a
r
Bi
o
l
o
g
y
,
v
o
l
.
2
4
4
3
,
S
p
r
i
n
g
e
r
U
S
,
2
0
2
2
,
p
p
.
4
2
9
–
4
4
9
.
[
2
]
J.
S
u
j
i
t
h
r
a
a
n
d
M
.
F
.
U
k
r
i
t
,
“
C
R
U
N
-
b
a
se
d
l
e
a
f
d
i
s
e
a
se
s
e
g
m
e
n
t
a
t
i
o
n
a
n
d
m
o
r
p
h
o
l
o
g
i
c
a
l
-
b
a
se
d
st
a
g
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
Ma
t
h
e
m
a
t
i
c
a
l
Pro
b
l
e
m
s i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
0
2
2
,
p
p
.
1
–
1
3
,
Ju
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
2
5
4
6
8
7
3
.
[
3
]
S
.
K
.
P
.
K
u
mar
,
M
.
G
.
S
u
m
i
t
h
r
a
,
a
n
d
N
.
S
a
r
a
n
y
a
,
“
A
r
t
i
f
i
c
i
a
l
b
e
e
c
o
l
o
n
y
-
b
a
se
d
f
u
z
z
y
c
m
e
a
n
s
(
A
B
C
-
F
C
M
)
se
g
me
n
t
a
t
i
o
n
a
l
g
o
r
i
t
h
m
a
n
d
d
i
me
n
si
o
n
a
l
i
t
y
r
e
d
u
c
t
i
o
n
f
o
r
l
e
a
f
d
i
se
a
se
d
e
t
e
c
t
i
o
n
i
n
b
i
o
i
n
f
o
r
mat
i
c
s
,
”
T
h
e
J
o
u
rn
a
l
o
f
S
u
p
e
rc
o
m
p
u
t
i
n
g
,
v
o
l
.
7
5
,
n
o
.
1
2
,
p
p
.
8
2
9
3
–
8
3
1
1
,
D
e
c
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
2
2
7
-
0
1
9
-
0
2
9
9
9
-
z.
[
4
]
K
.
J
a
y
a
p
r
a
k
a
sh
a
n
d
S
.
P
.
B
a
l
a
m
u
r
u
g
a
n
,
“
D
e
s
i
g
n
o
f
o
p
t
i
ma
l
mu
l
t
i
l
e
v
e
l
t
h
r
e
sh
o
l
d
i
n
g
b
a
se
d
se
g
m
e
n
t
a
t
i
o
n
w
i
t
h
A
l
e
x
N
e
t
m
o
d
e
l
f
o
r
p
l
a
n
t
l
e
a
f
d
i
sea
s
e
d
i
a
g
n
o
si
s
,
”
i
n
2
0
2
2
4
t
h
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
m
a
rt
S
y
s
t
e
m
s
a
n
d
I
n
v
e
n
t
i
v
e
T
e
c
h
n
o
l
o
g
y
(
I
C
S
S
I
T
)
,
Ja
n
.
2
0
2
2
,
p
p
.
1
4
7
3
–
1
4
7
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
S
I
T5
3
2
6
4
.
2
0
2
2
.
9
7
1
6
2
3
3
.
[
5
]
Y
.
M
.
A
.
A
l
g
a
n
i
,
O
.
J.
M
.
C
a
r
o
,
L.
M
.
R
.
B
r
a
v
o
,
C
.
K
a
u
r
,
M
.
S
.
A
l
A
n
sari
,
a
n
d
B
.
K
.
B
a
l
a
,
“
Le
a
f
d
i
se
a
se
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
o
p
t
i
mi
z
e
d
d
e
e
p
l
e
a
r
n
i
n
g
,
”
Me
a
s
u
reme
n
t
:
S
e
n
so
rs
,
v
o
l
.
2
5
,
p
.
1
0
0
6
4
3
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
a
se
n
.
2
0
2
2
.
1
0
0
6
4
3
.
[
6
]
S
.
P
.
M
a
n
i
k
a
n
d
a
n
,
S
.
R
.
N
a
r
a
n
i
,
S
.
K
a
r
t
h
i
k
e
y
a
n
,
a
n
d
N
.
M
o
h
a
n
k
u
m
a
r
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
s
k
i
n
me
l
a
n
o
ma
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
d
e
r
m
o
sc
o
p
i
c
i
ma
g
e
s
i
n
d
i
f
f
e
r
e
n
t
c
o
l
o
r
sp
a
c
e
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
E
l
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
ri
n
g
(
I
J
EEC
S
)
,
v
o
l
.
1
5
,
n
o
.
1
,
p
p
.
3
1
9
–
3
2
7
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
5
i
1
.
p
p
3
1
9
-
3
2
7
.
[
7
]
O
.
M
z
o
u
g
h
i
a
n
d
I
.
Y
a
h
i
a
o
u
i
,
"
D
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
se
g
me
n
t
a
t
i
o
n
f
o
r
d
i
s
e
a
se
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
"
Ec
o
l
o
g
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
7
5
,
p
p
.
1
0
2
0
0
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
c
o
i
n
f
.
2
0
2
3
.
1
0
2
0
0
0
.
[
8
]
R
.
S
a
t
h
y
a
,
S
.
S
e
n
t
h
i
l
v
a
d
i
v
u
,
S
.
A
n
a
n
t
h
i
,
V
.
C
.
B
h
a
r
a
t
h
i
,
a
n
d
G
.
R
e
v
a
t
h
y
,
“
V
i
si
o
n
b
a
se
d
p
l
a
n
t
l
e
a
f
d
i
se
a
se
d
e
t
e
c
t
i
o
n
a
n
d
r
e
c
o
g
n
i
t
i
o
n
m
o
d
e
l
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
i
n
7
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
El
e
c
t
ro
n
i
c
s,
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
Ae
ro
s
p
a
c
e
T
e
c
h
n
o
l
o
g
y
,
I
C
E
C
A
2
0
2
3
-
Pro
c
e
e
d
i
n
g
s
,
N
o
v
.
2
0
2
3
,
p
p
.
4
5
8
–
4
6
4
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
EC
A
5
8
5
2
9
.
2
0
2
3
.
1
0
3
9
5
6
2
0
.
[
9
]
A
.
A
g
a
r
w
a
l
,
D
.
K
.
A
g
a
r
w
a
l
,
T
.
K
.
S
h
a
r
m
a
,
V
.
K
.
Y
a
d
a
v
,
a
n
d
M
.
S
r
i
v
a
s
t
a
v
a
,
“
P
l
a
n
t
s
d
i
s
e
a
s
e
d
e
t
e
c
t
i
o
n
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
p
l
a
n
t
d
i
s
e
a
s
e
C
N
N
m
o
d
e
l
s
f
o
r
i
m
a
g
e
s
e
g
m
e
n
t
a
t
i
o
n
,
”
F
L
O
R
A
A
N
D
F
A
U
N
A
,
v
o
l
.
2
8
,
n
o
.
1
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
4
5
1
/
f
l
o
r
a
f
a
u
n
a
.
v
2
8
i
1
p
p
7
7
-
85.
[
1
0
]
M
.
P
r
a
b
u
a
n
d
B
.
J.
C
h
e
l
l
i
a
h
,
“
A
n
i
n
t
e
l
l
i
g
e
n
t
a
p
p
r
o
a
c
h
u
s
i
n
g
b
o
o
s
t
e
d
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
b
a
s
e
d
a
r
i
t
h
m
e
t
i
c
o
p
t
i
m
i
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
a
c
c
u
r
a
t
e
d
e
t
e
c
t
i
o
n
o
f
p
l
a
n
t
l
e
a
f
d
i
s
e
a
se
,
”
P
a
t
t
e
r
n
A
n
a
l
y
si
s a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
6
,
n
o
.
1
,
p
p
.
3
6
7
–
3
7
9
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
0
4
4
-
022
-
0
1
0
8
6
-
z.
[
1
1
]
I
.
M
a
smo
u
d
i
a
n
d
R
.
L
g
h
o
u
l
,
“
A
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
a
p
p
r
o
a
c
h
f
o
r
p
l
a
n
t
l
e
a
f
s
e
g
m
e
n
t
a
t
i
o
n
a
n
d
d
i
se
a
se
c
l
a
ss
i
f
i
c
a
t
i
o
n
i
n
smar
t
a
g
r
i
c
u
l
t
u
r
e
,
”
i
n
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
,
S
p
r
i
n
g
e
r
I
n
t
e
r
n
a
t
i
o
n
a
l
P
u
b
l
i
s
h
i
n
g
,
2
0
2
1
,
p
p
.
1
0
4
4
–
1
0
5
5
.
[
1
2
]
L.
C
a
o
,
H
.
L
i
,
H
.
Y
u
,
G
.
C
h
e
n
,
a
n
d
H
.
W
a
n
g
,
“
P
l
a
n
t
l
e
a
f
se
g
m
e
n
t
a
t
i
o
n
a
n
d
p
h
e
n
o
t
y
p
i
c
a
n
a
l
y
s
i
s
b
a
se
d
o
n
f
u
l
l
y
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
Ap
p
l
i
e
d
En
g
i
n
e
e
ri
n
g
i
n
A
g
ri
c
u
l
t
u
re
,
v
o
l
.
3
7
,
n
o
.
5
,
p
p
.
9
2
9
–
9
4
0
,
2
0
2
1
,
d
o
i
:
1
0
.
1
3
0
3
1
/
a
e
a
.
1
4
4
9
5
.
[
1
3
]
R
.
S
a
l
e
e
m
,
J.
H
.
S
h
a
h
,
M
.
S
h
a
r
i
f
,
M
.
Y
a
smin
,
H
.
S
.
Y
o
n
g
,
a
n
d
J.
C
h
a
,
“
M
a
n
g
o
l
e
a
f
d
i
s
e
a
se
r
e
c
o
g
n
i
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
n
o
v
e
l
s
e
g
me
n
t
a
t
i
o
n
a
n
d
v
e
i
n
p
a
t
t
e
r
n
t
e
c
h
n
i
q
u
e
,
”
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
(
S
w
i
t
zerl
a
n
d
)
,
v
o
l
.
1
1
,
n
o
.
2
4
,
p
.
1
1
9
0
1
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
1
2
4
1
1
9
0
1
.
[
1
4
]
F
.
G
.
W
a
l
d
a
mi
c
h
a
e
l
,
T.
G
.
D
e
b
e
l
e
e
,
a
n
d
Y
.
M
.
A
y
a
n
o
,
“
C
o
f
f
e
e
d
i
s
e
a
s
e
d
e
t
e
c
t
i
o
n
u
s
i
n
g
a
r
o
b
u
s
t
H
S
V
c
o
l
o
r
-
b
a
sed
se
g
me
n
t
a
t
i
o
n
a
n
d
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
f
o
r
u
s
e
o
n
smar
t
p
h
o
n
e
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
,
v
o
l
.
3
7
,
n
o
.
8
,
p
p
.
4
9
6
7
–
4
9
9
3
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
2
/
i
n
t
.
2
2
7
4
7
.
[
1
5
]
L.
G
.
D
i
v
y
a
n
t
h
,
A
.
A
h
ma
d
,
a
n
d
D
.
S
a
r
a
sw
a
t
,
“
A
t
w
o
-
st
a
g
e
d
e
e
p
-
l
e
a
r
n
i
n
g
b
a
s
e
d
se
g
me
n
t
a
t
i
o
n
m
o
d
e
l
f
o
r
c
r
o
p
d
i
sea
s
e
q
u
a
n
t
i
f
i
c
a
t
i
o
n
b
a
se
d
o
n
c
o
r
n
f
i
e
l
d
i
ma
g
e
r
y
,
”
S
m
a
rt
A
g
ri
c
u
l
t
u
r
a
l
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
3
,
p
.
1
0
0
1
0
8
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
t
e
c
h
.
2
0
2
2
.
1
0
0
1
0
8
.
[
1
6
]
V
.
S
i
n
g
h
,
“
S
u
n
f
l
o
w
e
r
l
e
a
f
d
i
sea
s
e
s
d
e
t
e
c
t
i
o
n
u
si
n
g
i
ma
g
e
s
e
g
m
e
n
t
a
t
i
o
n
b
a
s
e
d
o
n
p
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
,
”
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
i
n
A
g
r
i
c
u
l
t
u
r
e
,
v
o
l
.
3
,
p
p
.
6
2
–
6
8
,
S
e
p
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
i
i
a
.
2
0
1
9
.
0
9
.
0
0
2
.
[
1
7
]
K
.
P
.
P
a
n
i
g
r
a
h
i
,
H
.
D
a
s,
A
.
K
.
S
a
h
o
o
,
a
n
d
S
.
C
.
M
o
h
a
r
a
n
a
,
“
M
a
i
z
e
l
e
a
f
d
i
sea
se
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,”
i
n
Pro
g
ress
i
n
C
o
m
p
u
t
i
n
g
,
A
n
a
l
y
t
i
c
s
a
n
d
N
e
t
w
o
rk
i
n
g
,
v
o
l
.
1
1
1
9
,
S
p
r
i
n
g
e
r
S
i
n
g
a
p
o
r
e
,
2
0
2
0
,
p
p
.
6
5
9
–
6
6
9
.
[
1
8
]
K
.
S
a
h
u
a
n
d
S
.
M
i
n
z
,
“
A
d
a
p
t
i
v
e
s
e
g
men
t
a
t
i
o
n
w
i
t
h
i
n
t
e
l
l
i
g
e
n
t
R
e
sN
e
t
a
n
d
LST
M
–
D
N
N
f
o
r
p
l
a
n
t
l
e
a
f
mu
l
t
i
-
d
i
s
e
a
se
c
l
a
ss
i
f
i
c
a
t
i
o
n
mo
d
e
l
,
”
S
e
n
s
i
n
g
a
n
d
I
m
a
g
i
n
g
,
v
o
l
.
2
4
,
n
o
.
1
,
J
u
l
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
2
2
0
-
0
2
3
-
0
0
4
2
8
-
3.
[
1
9
]
B
.
S
a
i
R
e
d
d
y
a
n
d
S
.
N
e
e
r
a
j
a
,
“
P
l
a
n
t
l
e
a
f
d
i
s
e
a
s
e
c
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
d
a
mag
e
d
e
t
e
c
t
i
o
n
s
y
s
t
e
m
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s,
”
Mu
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
1
,
n
o
.
1
7
,
p
p
.
2
4
0
2
1
–
2
4
0
4
0
,
M
a
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
2
-
1
2
1
4
7
-
0.
[
2
0
]
M
.
A
.
P
a
t
i
l
a
n
d
M
.
M
,
“
E
n
h
a
n
c
e
d
r
a
d
i
a
l
b
a
si
s
f
u
n
c
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
t
o
ma
t
o
p
l
a
n
t
d
i
se
a
se
l
e
a
f
i
ma
g
e
se
g
me
n
t
a
t
i
o
n
,
”
Ec
o
l
o
g
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
7
0
,
p
.
1
0
1
7
5
2
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
c
o
i
n
f
.
2
0
2
2
.
1
0
1
7
5
2
.
[
2
1
]
X
.
G
u
o
a
n
d
M
.
Zh
a
n
g
,
“
I
ma
g
e
o
f
p
l
a
n
t
d
i
se
a
se
se
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
b
a
se
d
o
n
i
m
p
r
o
v
e
d
p
u
l
s
e
-
c
o
u
p
l
e
d
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
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
a
t
i
o
n
a
l
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
2
3
,
n
o
.
1
,
p
p
.
1
–
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
5
0
4
/
i
j
c
s
e
.
2
0
2
0
.
1
1
0
1
9
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.