T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
1
9
,
No
.
2
,
A
p
r
il
2
0
2
1
,
p
p
.
46
3
~
47
0
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
De
cr
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i
2
.
1
6
4
8
8
463
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
An ef
fec
tive fe
a
t
u
re e
x
traction
met
ho
d f
o
r ric
e leaf
d
isea
se
cla
ss
ificatio
n
M
uh
a
m
m
a
d Anwa
rul A
zim
,
M
o
ha
m
m
a
d K
ha
irul Is
la
m
,
M
d.
M
a
ruf
ur
Ra
hm
a
n
,
F
a
ra
h J
a
ha
n
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
Un
iv
e
rsity
o
f
C
h
it
tag
o
n
g
,
C
h
it
tag
o
n
g
,
Ba
n
g
lad
e
s
h
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ap
r
2
8
,
2
0
2
0
R
ev
is
ed
Sep
2
5
,
2
0
2
0
Acc
ep
ted
Oct
7
,
2
0
2
0
Ou
r
so
c
iety
is
g
e
tt
i
n
g
m
o
re
a
n
d
m
o
re
tec
h
n
o
lo
g
y
d
e
p
e
n
d
e
n
t
d
a
y
b
y
d
a
y
.
Ne
v
e
rth
e
les
s,
a
g
ricu
lt
u
re
is
imp
e
ra
ti
v
e
fo
r
o
u
r
su
r
v
i
v
a
l.
Rice
is
o
n
e
o
f
th
e
p
rima
ry
fo
o
d
g
ra
i
n
s.
It
p
ro
v
i
d
e
s
su
ste
n
a
n
c
e
to
a
lmo
st
fift
y
p
e
rc
e
n
t
o
f
th
e
wo
rl
d
p
o
p
u
lati
o
n
a
n
d
p
r
o
m
o
tes
h
u
g
e
a
m
o
u
n
t
o
f
e
m
p
l
o
y
m
e
n
ts.
He
n
c
e
,
p
ro
p
e
r
m
it
ig
a
ti
o
n
o
f
r
ice
p
la
n
t
d
ise
a
se
s
is
o
f
p
a
ra
m
o
u
n
t
imp
o
rtan
c
e
.
A
m
o
d
e
l
to
d
e
tec
t
th
re
e
rice
lea
f
d
ise
a
s
e
s,
n
a
m
e
ly
b
a
c
teria
l
lea
f
b
li
g
h
t,
b
ro
wn
s
p
o
t
,
a
n
d
lea
f
sm
u
t
is
p
ro
p
o
se
d
in
t
h
is
p
a
p
e
r.
Ba
c
k
g
r
o
u
n
d
s
o
f
t
h
e
ima
g
e
s
a
re
re
m
o
v
e
d
b
y
sa
tu
ra
ti
o
n
th
re
sh
o
ld
wh
il
e
d
ise
a
se
a
ffe
c
ted
a
re
a
s
a
r
e
se
g
m
e
n
ted
u
sin
g
h
u
e
th
re
sh
o
l
d
.
Distin
c
t
iv
e
fe
a
tu
re
s
fro
m
c
o
lo
r,
sh
a
p
e
,
a
n
d
tex
t
u
re
d
o
m
a
in
a
re
e
x
trac
ted
fro
m
a
ffe
c
ted
a
re
a
s.
Th
e
se
fe
a
tu
re
s
c
a
n
ro
b
u
stl
y
d
e
sc
rib
e
lo
c
a
l
a
n
d
g
lo
b
a
l
sta
ti
stics
o
f
su
c
h
ima
g
e
s.
Try
in
g
a
c
o
u
p
le
o
f
c
las
sifica
ti
o
n
a
lg
o
rit
h
m
s,
e
x
trem
e
g
ra
d
ien
t
b
o
o
stin
g
d
e
c
isio
n
tree
e
n
se
m
b
le
is
i
n
c
o
rp
o
ra
ted
i
n
th
is
m
o
d
e
l
fo
r
it
s
su
p
e
rio
r
p
e
rfo
rm
a
n
c
e
.
O
u
r
m
o
d
e
l
a
c
h
iev
e
s
8
6
.
5
8
%
a
c
c
u
ra
c
y
o
n
rice
lea
f
d
ise
a
se
s
d
a
tas
e
t
fro
m
UCI,
wh
ich
is
h
i
g
h
e
r
th
a
n
p
re
v
io
u
s
w
o
rk
s
o
n
th
e
sa
m
e
d
a
tas
e
t.
Clas
s
-
wise
a
c
c
u
ra
c
y
o
f
th
e
m
o
d
e
l
is
a
lso
c
o
n
siste
n
t
a
m
o
n
g
t
h
e
c
las
se
s.
K
ey
w
o
r
d
s
:
C
las
s
if
icatio
n
E
x
tr
em
e
g
r
a
d
ien
t b
o
o
s
tin
g
I
m
ag
e
p
r
o
ce
s
s
in
g
Ma
ch
in
e
lear
n
in
g
Plan
t d
is
ea
s
e
d
etec
tio
n
R
ice
leaf
d
is
ea
s
es
XGBo
o
s
t
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
:
Mu
h
am
m
ad
A
n
war
u
l A
zim
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
Un
iv
er
s
ity
o
f
C
h
ittag
o
n
g
C
h
ittag
o
n
g
-
4
3
3
1
,
B
an
g
lad
esh
E
m
ail:
az
im
@
cu
.
ac
.
b
d
1.
I
NT
RO
D
UCT
I
O
N
B
ein
g
th
e
p
r
im
ar
y
f
o
o
d
o
f
m
o
r
e
th
an
th
r
ee
b
illi
o
n
p
eo
p
le
o
f
th
e
wo
r
ld
,
r
ice
is
o
n
e
o
f
th
e
m
o
s
t
im
p
o
r
tan
t
cr
o
p
s
.
I
t
is
th
e
s
tap
le
f
o
o
d
o
f
B
an
g
lad
esh
,
C
h
in
a,
I
n
d
ia,
I
n
d
o
n
esia,
Vietn
am
,
a
n
d
Nig
e
r
ia
.
An
y
t
h
in
g
th
at
in
f
lu
en
ce
s
th
e
q
u
ality
an
d
q
u
a
n
tity
o
f
r
ice
p
r
o
d
u
ctio
n
af
f
ec
ts
th
e
m
ass
p
o
p
u
latio
n
.
Hen
ce
,
c
o
n
tin
u
al
m
o
n
ito
r
in
g
an
d
d
etec
tio
n
o
f
d
is
ea
s
es is
es
s
en
tial f
o
r
p
r
o
p
er
m
itig
atio
n
.
W
ith
o
u
t tim
ely
in
ter
v
en
tio
n
,
d
is
ea
s
es
ca
n
s
ev
er
ely
af
f
ec
t
th
e
p
r
o
d
u
ctio
n
.
T
h
u
s
,
au
to
m
atic
p
lan
t
d
is
ea
s
e
d
etec
ti
o
n
is
a
to
p
ic
o
f
g
r
ea
t
in
ter
est
in
ag
r
o
-
in
f
o
r
m
atics.
b
ac
ter
ial
leaf
b
lig
h
t
(
B
L
B
)
,
b
r
o
wn
s
p
o
t
(
B
S)
,
leaf
b
last
,
s
h
ea
th
b
lig
h
t,
tu
n
g
r
o
,
an
d
leaf
s
m
u
t
(
L
S)
ar
e
s
o
m
e
o
f
th
e
m
o
s
t p
r
ev
alen
t
r
ice
p
lan
t d
is
ea
s
es.
Af
f
ec
ted
p
lan
ts
p
r
o
d
u
ce
r
ice
o
f
less
am
o
u
n
t
an
d
lo
w
er
q
u
ality
th
an
t
h
at
o
f
h
ea
lth
y
p
lan
ts
.
T
h
is
lo
s
s
o
f
y
ield
h
as
d
ir
ec
t
an
d
n
eg
ativ
e
e
f
f
ec
t
o
n
m
ass
p
eo
p
le,
f
o
o
d
s
ec
u
r
ity
,
an
d
n
atio
n
al
ec
o
n
o
m
y
.
Acc
o
r
d
in
g
to
I
n
ter
n
atio
n
al
R
ice
R
esear
ch
I
n
s
titu
te,
b
ac
ter
ial
leaf
b
lig
h
t
ca
n
r
ed
u
ce
th
e
y
ield
u
p
to
7
0
%
[
1
]
.
B
r
o
wn
s
p
o
t
d
is
ea
s
e
c
an
d
am
ag
e
r
ice
p
r
o
d
u
ctio
n
u
p
to
4
5
%
[
2
]
.
I
t
is
o
n
e
o
f
th
e
p
r
im
ar
y
r
ea
s
o
n
s
b
eh
in
d
th
e
“Gr
ea
t
B
en
g
al
Fam
in
e”
o
f
1
9
4
3
.
So
m
e
o
f
th
ese
d
is
ea
s
es
h
av
e
s
im
ilar
s
y
m
p
to
m
s
wh
ich
co
n
f
u
s
e
th
e
u
n
tr
ain
ed
f
ar
m
er
s
a
n
d
ca
u
s
e
lo
s
s
o
f
h
ar
v
est.
Had
t
h
er
e
b
e
en
a
s
im
p
le,
au
to
m
at
ed
,
an
d
r
eliab
l
e
d
is
ea
s
e
d
etec
tio
n
f
ac
ilit
y
,
th
is
p
r
o
d
u
ctio
n
lo
s
s
co
u
ld
h
av
e
b
ee
n
av
o
id
e
d
.
Su
ch
a
s
y
s
tem
ca
n
s
u
r
ely
b
r
in
g
tan
g
ib
le
r
esu
lt
s
in
s
h
o
r
t
tim
e.
I
n
th
is
p
ap
er
,
we
in
ten
d
to
d
ev
is
e
a
s
et
o
f
ef
f
ec
tiv
e
f
ea
tu
r
es
to
class
if
y
r
ice
leaf
d
is
ea
s
es
with
h
ig
h
er
ac
cu
r
ac
y
.
T
h
e
ap
p
r
o
ac
h
p
r
esen
ted
in
th
is
p
ap
er
is
b
r
ief
ly
lai
d
o
u
t
in
f
iv
e
s
tep
s
.
Firstl
y
,
r
ice
leaf
d
is
ea
s
e
im
ag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
1
9
,
No
.
2
,
Ap
r
il
202
1
:
46
3
-
47
0
464
d
ataset
is
co
llected
f
r
o
m
[
3
]
wh
ich
is
p
r
ep
ar
ed
an
d
p
u
b
lis
h
ed
b
y
Pra
jap
ati
et
a
l.
[
4
]
.
Af
te
r
im
ag
e
ac
q
u
is
itio
n
,
b
ac
k
g
r
o
u
n
d
s
o
f
th
e
im
ag
es
a
r
e
r
em
o
v
ed
a
n
d
th
e
n
d
is
ea
s
e
af
f
ec
ted
p
o
r
tio
n
s
ar
e
s
eg
m
en
ted
.
C
o
lo
r
,
s
h
a
p
e,
an
d
tex
tu
r
e
f
ea
t
u
r
es
ar
e
ex
tr
ac
ted
f
r
o
m
th
e
im
ag
es.
Fin
ally
,
e
x
tr
em
e
g
r
ad
ien
t
b
o
o
s
tin
g
d
ec
is
io
n
tr
ee
en
s
em
b
le
(
XGBo
o
s
t)
class
if
ier
is
u
s
ed
f
o
r
class
if
icatio
n
.
2.
B
ACK
G
RO
UND
AN
D
RE
L
AT
E
D
WO
RK
2
.
1
.
B
a
c
k
g
r
o
un
d
T
h
e
d
ataset
u
s
ed
in
th
is
wo
r
k
is
p
r
ep
ar
e
d
b
y
Pra
ja
p
ati
et
a
l
.
[
4
]
w
h
ich
c
o
n
tain
s
1
2
0
R
GB
im
ag
es
o
f
th
r
ee
class
es
e.
g
.
,
b
ac
ter
ial
lea
f
b
lig
h
t,
b
r
o
wn
s
p
o
t,
a
n
d
lea
f
s
m
u
t
ea
ch
co
n
tain
i
n
g
4
0
im
ag
e
s
.
I
m
ag
es a
r
e
tak
e
n
o
n
a
wh
ite
b
ac
k
g
r
o
u
n
d
u
n
d
er
n
atu
r
al
lig
h
tin
g
.
Am
o
n
g
th
e
m
B
L
B
d
i
s
ea
s
e
is
ca
u
s
ed
b
y
a
b
ac
ter
iu
m
n
am
ed
Xan
th
o
m
o
n
as o
r
y
za
e
.
H
o
t a
n
d
h
u
m
id
a
r
ea
is
co
n
g
e
n
ial
f
o
r
th
e
s
p
r
ea
d
o
f
t
h
e
d
is
ea
s
e.
T
h
at
is
th
e
r
ea
s
o
n
b
e
h
in
d
its
p
r
ev
alen
ce
in
r
ice
-
p
r
o
d
u
cin
g
ar
ea
s
o
f
Asi
a,
Af
r
ica,
an
d
th
e
C
ar
ib
b
ea
n
.
Du
r
in
g
a
B
L
B
ep
id
em
ic,
cr
o
p
l
o
s
s
ca
n
b
e
as
h
ig
h
as
7
5
%.
I
t
ca
n
af
f
ec
t
p
ad
d
y
leav
es
an
d
s
ee
d
li
n
g
s
;
o
o
zin
g
m
ilk
y
ju
ice
at
th
e
p
r
im
ar
y
s
tag
e
an
d
f
in
ally
d
r
y
in
g
o
u
t
t
h
e
leav
es
to
wilt
an
d
d
ie.
B
r
o
wn
s
p
o
t
(
B
S)
d
is
ea
s
e
is
ca
u
s
ed
b
y
C
o
ch
li
o
b
o
lu
s
m
iy
ab
ea
n
u
s
,
a
f
u
n
g
u
s
.
I
t
af
f
ec
ts
th
e
r
ice
p
lan
t
in
v
ar
io
u
s
p
ar
ts
lik
e
leav
es,
s
ee
d
lin
g
s
,
s
h
ea
th
,
an
d
s
tem
s
.
T
h
e
lo
s
s
o
f
y
ield
ca
n
b
e
u
p
to
4
5
%
o
r
m
o
r
e.
On
p
a
d
d
y
leav
es,
th
e
d
is
ea
s
e
af
f
ec
ted
s
p
o
ts
ar
e
cir
cu
lar
o
r
o
v
al
in
s
h
ap
e.
I
t
ca
u
s
es
th
e
af
f
ec
ted
p
ar
t
o
f
t
h
e
p
lan
t
to
d
i
e,
ef
f
ec
tiv
ely
d
im
i
n
is
h
in
g
t
h
e
q
u
ality
an
d
q
u
an
tity
o
f
h
a
r
v
est.
L
ea
f
s
m
u
t
(
L
S)
is
also
ca
u
s
ed
b
y
a
f
u
n
g
u
s
n
am
e
d
E
n
ty
lo
m
a
o
r
y
za
e
.
T
h
o
u
g
h
it
is
n
o
t
a
m
ajo
r
d
is
ea
s
e,
it
ca
n
m
ak
e
way
f
o
r
o
th
er
d
is
ea
s
es
b
y
cr
ea
tin
g
a
f
av
o
r
ab
le
en
v
i
r
o
n
m
en
t
f
o
r
o
th
er
f
u
n
g
i
.
T
h
e
lesi
o
n
s
ar
e
cir
cu
lar
to
o
v
al
o
r
n
o
n
-
u
n
if
o
r
m
in
s
h
ap
e
with
a
r
o
u
g
h
tex
tu
r
e
.
Fig
u
r
e
1
s
h
o
ws
a
f
ew
s
am
p
le
im
ag
es
o
f
th
e
se
d
is
ea
s
es
f
r
o
m
th
e
d
ataset.
T
o
d
ev
elo
p
a
r
eliab
le
class
if
icatio
n
m
o
d
el,
ex
tr
ac
tin
g
im
p
o
r
ta
n
t
f
ea
tu
r
es
is
ess
e
n
tial.
Fo
r
th
at
p
u
r
p
o
s
e,
t
h
e
m
o
s
t
v
ital
ch
ar
ac
ter
is
tics
o
f
t
h
e
d
is
ea
s
es
ar
e
b
r
ief
ly
p
r
esen
te
d
in
T
ab
le
1
[
1
,
2
]
.
T
h
ese
co
lo
r
an
d
s
h
ap
e
in
f
o
r
m
atio
n
ca
n
b
e
an
al
y
ze
d
to
g
en
er
ate
d
i
s
tin
ctiv
e
f
ea
tu
r
es.
(
a)
(
c)
(
e)
(
b
)
(
d
)
(f)
Fig
u
r
e
1
.
R
ice
leav
es a
f
f
ec
ted
with
v
ar
io
u
s
d
is
ea
s
es
;
(a
)
-
(
b
)
B
L
B
,
(
c)
-
(
d
)
B
S,
(
e)
-
(
f
)
L
S
T
ab
le
1
.
R
ice
leaf
d
is
ea
s
e
lesi
o
n
co
m
p
ar
is
o
n
N
a
me
o
f
d
i
s
e
a
se
S
y
mp
t
o
m c
o
l
o
r
Le
si
o
n
s
h
a
p
e
B
LB
Y
e
l
l
o
w
t
o
G
r
a
y
-
W
h
i
t
e
Lo
n
g
sp
o
t
s,
n
o
n
-
u
n
i
f
o
r
m
sh
a
p
e
BS
Li
g
h
t
B
r
o
w
n
t
o
G
r
a
y
a
t
t
h
e
c
e
n
t
e
r
,
R
e
d
d
i
s
h
B
r
o
w
n
su
r
r
o
u
n
d
i
n
g
C
i
r
c
u
l
a
r
t
o
o
v
a
l
LS
D
a
r
k
R
e
d
d
i
s
h
B
r
o
w
n
t
o
D
a
r
k
G
r
a
y
C
i
r
c
u
l
a
r
t
o
o
v
a
l
o
r
n
o
n
-
u
n
i
f
o
r
m
2
.
2
.
Rela
t
ed
wo
rk
Ap
p
licatio
n
o
f
c
o
m
p
u
ter
an
d
in
f
o
r
m
atio
n
tech
n
o
lo
g
y
in
ag
r
icu
ltu
r
al
d
o
m
ain
is
a
p
r
o
m
in
e
n
t
f
ield
o
f
s
tu
d
y
n
o
wad
a
y
s
.
Var
io
u
s
r
esear
ch
er
s
ar
e
wo
r
k
i
n
g
in
d
i
f
f
er
e
n
t
s
ec
tio
n
s
o
f
th
is
f
ield
.
Gar
cia
et
a
l.
[
5
]
ca
r
r
ied
o
u
t
a
s
u
r
v
ey
o
n
v
a
r
io
u
s
im
ag
e
p
r
o
ce
s
s
in
g
an
d
cl
ass
if
icatio
n
tech
n
iq
u
es
o
f
th
is
d
o
m
ai
n
,
wh
ile
Pra
jap
ati
et
a
l.
[
6
]
s
u
r
v
ey
ed
v
ar
i
o
u
s
tech
n
iq
u
es
u
s
ed
in
r
ice
p
lan
t
d
is
ea
s
e
d
etec
tio
n
.
Sh
im
am
u
r
a
et
a
l.
[
7
]
wo
r
k
ed
o
n
au
t
o
m
atic
d
etec
tio
n
o
f
tip
b
u
r
n
in
ar
tific
ially
illu
m
in
ated
p
lan
t
f
ac
to
r
ie
s
u
s
in
g
C
N
Ns.
Su
g
im
o
to
et
a
l
.
[
8
]
ad
d
r
ess
ed
th
e
y
ield
lo
s
s
d
u
e
to
th
e
p
r
esen
ce
o
f
wee
d
s
in
p
ad
d
y
f
ield
s
.
T
h
e
y
d
e
v
elo
p
e
d
a
tech
n
i
q
u
e
o
f
p
a
d
d
y
wee
d
r
em
o
v
al
u
s
in
g
r
o
b
o
tics
.
A
lo
t
o
f
r
esear
ch
er
s
u
n
d
er
to
o
k
th
e
p
r
o
b
lem
o
f
a
u
to
m
atic
d
etec
tio
n
o
f
p
lan
t
d
is
ea
s
es
f
r
o
m
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
effec
tive
fea
tu
r
e
ex
tr
a
ctio
n
meth
o
d
fo
r
r
ice
lea
f d
is
ea
s
e
c
la
s
s
ifica
tio
n
(
Mu
h
a
mma
d
A
n
w
a
r
u
l A
z
im
)
465
leaf
im
ag
es.
K
h
ir
ad
e
et
a
l.
[
9
]
u
s
ed
h
u
e
-
s
atu
r
atio
n
th
r
esh
o
ld
in
g
f
o
r
im
a
g
e
s
eg
m
e
n
tat
io
n
an
d
SVM
f
o
r
class
if
icatio
n
in
th
eir
m
eth
o
d
.
W
an
g
et
a
l.
[
1
0
]
ex
tr
ac
ted
5
0
f
ea
tu
r
es
f
r
o
m
a
d
ataset
c
o
n
tai
n
in
g
two
g
r
ap
e
leaf
class
es
an
d
two
wh
ea
t
leaf
clas
s
es.
T
h
ey
u
s
ed
K
-
m
ea
n
s
f
o
r
s
e
g
m
en
tatio
n
a
n
d
PC
A
f
o
r
d
im
e
n
s
io
n
ality
r
ed
u
ctio
n
with
b
ac
k
p
r
o
p
ag
atio
n
n
e
u
r
al
n
etwo
r
k
f
o
r
d
is
ea
s
e
class
if
icat
io
n
.
Or
illo
et
a
l.
[
1
1
]
also
u
s
e
d
b
ac
k
p
r
o
p
a
g
atio
n
n
eu
r
al
n
etwo
r
k
with
s
tatis
tical
f
ea
tu
r
es
f
r
o
m
co
lo
r
im
ag
e
d
at
aset
co
n
tain
in
g
B
last
,
B
r
o
wn
Sp
o
t,
an
d
b
ac
ter
ial
leaf
b
lig
h
t
d
is
ea
s
es.
Su
m
an
et
a
l.
[
12
]
wo
r
k
ed
with
b
ac
te
r
ial
leaf
b
lig
h
t,
b
r
o
wn
s
p
o
t
,
n
ar
r
o
w
b
r
o
wn
s
p
o
t,
an
d
r
ice
b
last
d
is
ea
s
e
s
.
T
h
ey
u
s
ed
8
-
co
n
n
ec
ted
co
m
p
o
n
e
n
t
an
aly
s
is
f
o
r
d
is
ea
s
e
s
eg
m
en
tatio
n
an
d
SVM
f
o
r
class
if
icatio
n
.
T
h
ey
ex
tr
ac
ted
v
ar
io
u
s
co
lo
r
an
d
s
h
a
p
e
f
ea
tu
r
es
f
o
r
th
eir
m
o
d
el
wh
ich
ac
h
ie
v
ed
7
0
%
ac
cu
r
ac
y
.
K
-
m
ea
n
s
clu
s
ter
in
g
was
u
s
ed
b
y
Sin
g
h
et
a
l.
[
13
]
f
o
r
a
f
f
ec
ted
s
p
o
t
s
eg
m
en
tatio
n
.
T
h
ey
u
s
ed
SVM
f
o
r
class
if
icatio
n
an
d
g
ain
ed
8
2
%
clas
s
if
icatio
n
ac
cu
r
ac
y
.
K
-
m
ea
n
s
was
also
u
s
ed
b
y
Seth
y
et
a
l.
[
1
4
]
.
Yao
et
a
l.
[
15
]
u
s
ed
s
h
ap
e
an
d
tex
tu
r
e
f
ea
tu
r
es
with
SVM
as
th
e
clas
s
if
ier
.
Ph
ad
ik
ar
et
a
l
.
[
16
]
wo
r
k
e
d
with
r
ice
b
last
an
d
b
r
o
wn
s
p
o
t
d
is
ea
s
es.
T
h
ey
ex
tr
ac
ted
co
lo
r
a
n
d
s
h
ap
e
f
ea
tu
r
es
lik
e
r
ad
ial
d
is
tr
ib
u
tio
n
o
f
h
u
e.
Seg
m
en
tatio
n
was
ca
r
r
ied
o
u
t
u
s
in
g
Ost
u
'
s
s
eg
m
en
tatio
n
tech
n
iq
u
e
with
h
u
e
th
r
esh
o
l
d
.
Fo
r
1
0
-
f
o
l
d
cr
o
s
s
v
alid
atio
n
,
th
ey
ac
h
iev
ed
7
9
.
5
%
ac
cu
r
ac
y
with
B
ay
esian
class
if
ier
an
d
6
8
.
1
%
ac
c
u
r
ac
y
u
s
in
g
SVM.
Ph
ad
ik
ar
et
a
l.
[
1
7
]
wo
r
k
e
d
with
f
o
u
r
d
is
ea
s
e
class
e
s
an
d
ac
h
iev
ed
8
2
%
ac
cu
r
ac
y
with
p
atter
n
r
ec
o
g
n
itio
n
tech
n
iq
u
e.
C
h
ar
liep
a
u
l
[
18
]
s
eg
m
en
ted
th
e
d
is
ea
s
e
af
f
ec
ted
ar
ea
s
u
s
in
g
Fer
m
i
en
er
g
y
,
ex
tr
a
cted
co
lo
r
an
d
s
h
ap
e
f
ea
tu
r
es
an
d
ac
q
u
i
r
ed
7
5
%
ac
c
u
r
ac
y
.
Fer
m
i
en
er
g
y
was
also
u
s
ed
b
y
Ph
ad
ik
a
r
et
a
l.
[
1
9
]
.
K
u
r
n
iawa
ti
et
a
l.
[
2
0
]
u
s
ed
Ost
u
’
s
m
eth
o
d
f
o
r
s
eg
m
e
n
tatio
n
an
d
ex
tr
ac
ted
s
tatis
tical
f
ea
tu
r
es f
r
o
m
im
ag
es o
f
th
r
e
e
r
ice
leaf
d
is
ea
s
es
.
T
h
ey
u
s
ed
p
r
o
d
u
ctio
n
r
u
le
-
b
ased
class
if
ier
f
o
r
ac
h
iev
in
g
s
u
p
er
io
r
ac
cu
r
ac
y
.
An
th
o
n
y
s
et
a
l.
[
21
]
p
r
o
ce
s
s
ed
r
ice
b
last
,
b
r
o
wn
s
p
o
t
,
an
d
s
h
ea
th
b
lig
h
t
d
is
ea
s
es.
T
h
ey
u
s
ed
m
ath
em
atica
l
m
o
r
p
h
o
l
o
g
y
f
o
r
s
eg
m
en
tatio
n
an
d
m
em
b
er
s
h
ip
f
u
n
ctio
n
as
class
if
ier
.
W
ith
co
l
o
r
,
s
h
ap
e,
an
d
te
x
tu
r
e
f
ea
tu
r
es,
t
h
ey
ac
q
u
ir
e
d
o
v
er
7
0
%
ac
cu
r
ac
y
.
Var
io
u
s
co
lo
r
an
d
s
h
ap
e
f
ea
tu
r
es
wer
e
also
e
x
tr
ac
ted
as
f
ea
tu
r
es
in
th
e
wo
r
k
o
f
C
h
awa
th
e
[
2
2
]
.
Ma
jid
et
a
l.
[
2
3
]
u
s
ed
f
u
zz
y
e
n
tr
o
p
y
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
a
n
d
p
r
o
b
a
b
ilis
tic
n
eu
r
al
n
etwo
r
k
as
class
if
ier
o
n
a
d
ataset
o
f
f
o
u
r
d
is
ea
s
es.
Pra
jap
ati
et
a
l.
[
4
]
wo
r
k
ed
with
b
ac
te
r
ial
leaf
b
lig
h
t,
b
r
o
wn
s
p
o
t,
an
d
leaf
s
m
u
t
di
s
ea
s
es.
T
h
ey
u
s
ed
K
-
m
ea
n
s
clu
s
ter
in
g
alg
o
r
ith
m
f
o
r
d
is
ea
s
e
s
eg
m
en
tatio
n
.
T
h
ey
u
s
ed
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
f
o
r
r
eg
io
n
f
illi
n
g
,
r
eg
io
n
m
ask
i
n
g
an
d
n
o
is
e
r
em
o
v
al.
T
h
ey
e
x
tr
a
cted
a
to
tal
o
f
4
0
f
ea
tu
r
es
f
r
o
m
co
lo
r
,
s
h
ap
e,
a
n
d
tex
tu
r
e.
T
h
eir
d
ataset
co
n
tain
e
d
4
0
im
ag
es
o
f
ea
ch
d
is
ea
s
e,
o
u
t
o
f
wh
ich
3
5
wer
e
u
s
ed
f
o
r
tr
ain
in
g
an
d
5
f
o
r
test
in
g
.
T
h
ey
u
s
ed
an
SVM
class
if
ier
with
r
ad
ial
b
asis
f
u
n
ctio
n
(
R
B
F)
k
er
n
el
a
n
d
o
b
tain
ed
7
3
.
3
3
%
class
if
icatio
n
ac
cu
r
ac
y
.
I
t is o
b
s
er
v
ed
th
at
th
er
e
is
n
o
s
tan
d
ar
d
d
ataset
f
o
r
r
ice
leaf
d
is
ea
s
e.
All th
e
r
esear
ch
er
s
m
en
tio
n
ed
ab
o
v
e
p
r
ep
a
r
ed
an
d
u
s
ed
th
eir
o
wn
d
ataset.
An
o
p
en
an
d
s
tan
d
ar
d
r
ice
leaf
d
i
s
ea
s
e
im
ag
e
d
ataset
wo
u
ld
im
m
en
s
ely
h
elp
t
h
e
r
esear
ch
er
s
o
f
th
is
f
ield
.
I
t
is
also
n
o
ted
th
at
K
-
m
ea
n
s
clu
s
ter
in
g
alg
o
r
ith
m
is
th
e
p
r
im
e
alg
o
r
ith
m
f
o
r
d
is
ea
s
e
af
f
ec
ted
s
p
o
t seg
m
en
tatio
n
,
as u
tili
ze
d
in
[
4
,
1
0
,
1
3
]
.
3.
P
RO
P
O
SE
D
M
E
T
H
O
D
Fig
u
r
e
2
s
h
o
ws
th
e
b
lo
ck
d
ia
g
r
am
o
f
o
u
r
p
r
o
p
o
s
ed
ap
p
r
o
a
ch
.
I
t
co
n
s
is
ts
o
f
b
ac
k
g
r
o
u
n
d
r
em
o
v
al,
af
f
ec
ted
s
p
o
t
s
eg
m
en
tatio
n
,
f
ea
tu
r
e
ex
tr
ac
tio
n
,
an
d
class
if
icatio
n
.
F
o
r
th
e
s
ak
e
o
f
co
m
p
ar
is
o
n
with
o
th
er
r
esear
ch
wo
r
k
s
,
we
u
s
e
th
e
im
ag
es
o
f
th
e
d
ataset
u
s
ed
in
Pra
jap
ati
et
a
l.
[
4
]
as
in
p
u
t
.
T
h
e
i
m
ag
es
o
f
th
e
u
s
ed
d
at
aset
ar
e
ca
p
t
u
r
ed
with
a
w
h
ite
b
ac
k
g
r
o
u
n
d
,
wh
ich
ar
e
u
s
u
ally
s
atu
r
atio
n
d
o
m
in
ated
.
H
en
ce
,
we
co
n
v
er
t
a
n
R
GB
im
ag
e
to
HSV
an
d
r
em
o
v
e
its
b
ac
k
g
r
o
u
n
d
.
T
h
is
en
s
u
r
e
s
th
at
s
h
ad
o
ws
an
d
o
th
er
o
u
tlier
s
d
o
n
o
t
af
f
ec
t
th
e
f
ea
tu
r
e
ex
tr
ac
ti
o
n
p
r
o
ce
s
s
.
Ne
x
t
s
tep
is
ex
tr
ac
tin
g
th
o
s
e
p
o
r
ti
o
n
s
o
f
lea
v
es
wh
er
e
d
is
ea
s
e
s
y
m
p
to
m
s
ar
e
l
o
ca
ted
.
I
t is d
o
n
e
by
u
s
in
g
a
th
r
esh
o
l
d
m
ask
o
n
th
e
h
u
e
p
lan
e
o
f
th
e
HSV
im
ag
es.
T
o
g
en
er
ate
a
s
ig
n
atu
r
e
o
f
a
d
is
ea
s
e
we
ex
tr
ac
t
v
ar
io
u
s
f
ea
tu
r
es
f
r
o
m
co
lo
r
,
s
h
ap
e,
a
n
d
tex
tu
r
e
o
f
th
e
in
f
ec
ted
ar
e
as
.
Fin
ally
,
ex
tr
em
e
g
r
ad
ien
t
b
o
o
s
tin
g
d
ec
is
io
n
tr
ee
en
s
em
b
le
is
u
s
ed
f
o
r
clas
s
if
y
in
g
th
e
d
is
ea
s
es
.
T
h
e
f
o
llo
win
g
s
ec
tio
n
s
illu
s
tr
ate
th
ese
s
tep
s
in
d
etail.
Fig
u
r
e
2
.
Ov
e
r
v
iew
o
f
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
3
.
1
.
B
a
c
k
g
ro
un
d r
em
o
v
a
l
T
h
e
im
ag
es in
th
e
d
ataset
b
elo
n
g
to
R
GB
co
lo
r
s
p
ac
e.
Als
o
,
th
e
b
ac
k
g
r
o
u
n
d
s
ar
e
wh
ite
in
all
im
ag
es.
Fo
r
b
ac
k
g
r
o
u
n
d
r
em
o
v
al,
i
m
ag
es
ar
e
f
ir
s
t
co
n
v
er
ted
f
r
o
m
R
GB
to
HSV
co
lo
r
s
p
ac
e.
T
h
e
s
atu
r
atio
n
o
f
a
b
ac
k
g
r
o
u
n
d
is
co
m
p
letely
d
if
f
er
en
t
f
r
o
m
th
at
o
f
a
leaf
.
B
in
ar
y
th
r
esh
o
ld
i
n
g
is
ca
r
r
ied
o
u
t
o
n
th
e
s
atu
r
atio
n
p
la
n
e
o
f
th
e
im
ag
es.
W
e
s
et
th
r
esh
o
l
d
to
71
f
o
r
b
ac
k
g
r
o
u
n
d
s
ep
ar
a
tio
n
.
T
h
e
o
r
ig
i
n
al
im
ag
e
is
th
en
m
ask
ed
with
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
1
9
,
No
.
2
,
Ap
r
il
202
1
:
46
3
-
47
0
466
th
r
esh
o
ld
ed
im
ag
e
t
o
g
et
th
e
b
ac
k
g
r
o
u
n
d
r
em
o
v
ed
im
a
g
e.
Fig
u
r
e
3
illu
s
tr
ates
th
e
b
ac
k
g
r
o
u
n
d
r
em
o
v
al
p
r
o
ce
d
u
r
e
an
d
Fig
u
r
e
4
s
h
o
ws
th
e
o
u
tp
u
t im
ag
e.
Fig
u
r
e
3
.
B
ac
k
g
r
o
u
n
d
r
em
o
v
al
p
r
o
ce
s
s
(
a)
(
b
)
Fig
u
r
e
4
.
B
ac
k
g
r
o
u
n
d
r
em
o
v
al
;
(
a)
o
r
ig
i
n
al
im
ag
e
a
n
d
(
b
)
b
a
ck
g
r
o
u
n
d
r
em
o
v
ed
im
ag
e
3
.
2
.
Dis
ea
s
e
a
f
f
ec
t
ed
s
po
t
s
eg
m
ent
a
t
i
o
n
T
h
e
af
f
ec
te
d
ar
ea
o
f
a
lea
f
tu
r
n
s
b
r
o
wn
,
g
r
ay
,
d
ar
k
g
r
ay
,
o
r
r
ed
d
is
h
b
r
o
wn
w
h
ich
ar
e
c
o
m
p
letely
d
is
s
im
ilar
to
th
e
u
n
af
f
ec
ted
g
r
ee
n
p
o
r
tio
n
o
f
th
e
lea
f
.
I
n
o
u
r
wo
r
k
,
h
u
e
b
ased
s
eg
m
en
tatio
n
is
f
o
u
n
d
ef
f
ec
tiv
e
an
d
tim
e
ef
f
icien
t c
o
m
p
a
r
ed
t
o
K
-
M
ea
n
s
.
B
in
ar
y
th
r
esh
o
l
d
in
g
is
ap
p
lied
o
n
th
e
h
u
e
p
la
n
e
o
f
th
e
im
a
g
es with
a
th
r
esh
o
ld
v
al
u
e
o
f
2
2
,
wh
ich
is
f
o
u
n
d
s
u
itab
le
f
r
o
m
h
is
to
g
r
a
m
an
aly
s
is
.
Fig
u
r
e
5
s
h
o
ws
th
e
b
lo
ck
d
iag
r
am
o
f
th
e
s
eg
m
en
tatio
n
s
tep
an
d
Fig
u
r
e
6
illu
s
tr
ates
an
o
u
tp
u
t
af
t
er
u
s
in
g
th
e
p
r
o
ce
s
s
.
Fig
u
r
e
5
.
Dis
ea
s
e
s
p
o
t seg
m
en
tatio
n
p
r
o
ce
s
s
(
a)
(
b
)
Fig
u
r
e
6
.
Dis
ea
s
e
s
p
o
t seg
m
en
tatio
n
;
(
a)
b
ac
k
g
r
o
u
n
d
r
em
o
v
e
d
im
ag
e
an
d
(
b
)
af
f
ec
te
d
s
p
o
ts
3
.
3
.
F
e
a
t
ure
s
elec
t
io
n a
nd
ex
t
ra
ct
io
n
T
h
e
r
esear
ch
es
,
u
s
in
g
th
e
s
am
e
d
ataset,
tr
ied
to
d
ec
r
ea
s
e
th
e
n
u
m
b
er
o
f
f
ea
tu
r
es
to
r
ed
u
ce
m
o
d
el
co
m
p
lex
ity
a
n
d
ac
h
ie
v
e
h
ig
h
er
class
if
icatio
n
ac
cu
r
ac
y
.
W
e
h
av
e
th
o
r
o
u
g
h
ly
i
n
v
esti
g
ated
th
e
f
ea
tu
r
es
o
f
d
is
ea
s
es
o
n
th
e
leav
es.
Fin
all
y
,
we
h
av
e
co
m
e
u
p
with
a
s
et
o
f
2
6
f
ea
tu
r
es
f
r
o
m
th
e
co
lo
r
,
s
h
ap
e,
an
d
te
x
tu
r
e
d
o
m
ain
th
at
ca
n
r
o
b
u
s
tly
d
escr
ib
e
d
is
ea
s
e
af
f
ec
ted
leaf
s
p
o
ts
.
T
h
ese
ar
e
m
en
tio
n
ed
b
el
o
w.
C
o
lo
r
f
ea
tu
r
es:
th
ese
f
ea
tu
r
es
d
escr
ib
e
co
lo
r
p
r
o
p
er
ties
o
f
th
e
s
eg
m
en
ted
lea
f
s
p
o
ts
as
well
as
g
lo
b
al
co
lo
r
p
r
o
p
er
ties
o
f
th
e
im
ag
es.
1
2
co
lo
r
f
ea
tu
r
es
ar
e
ex
tr
ac
te
d
f
r
o
m
th
e
s
eg
m
en
te
d
im
ag
es.
T
h
ese
ar
e:
i)
m
ea
n
o
f
r
ed
ch
an
n
el
,
ii)
m
ea
n
o
f
g
r
ee
n
ch
an
n
el,
iii)
m
ea
n
o
f
b
lu
e
ch
an
n
el,
iv
)
m
ea
n
o
f
h
u
e
c
h
an
n
el,
v
)
m
ea
n
o
f
s
atu
r
atio
n
ch
an
n
el,
v
i)
s
tan
d
ar
d
d
ev
iatio
n
o
f
r
e
d
ch
a
n
n
el,
v
ii)
s
tan
d
ar
d
d
e
v
iatio
n
o
f
g
r
ee
n
ch
an
n
el,
v
iii)
s
tan
d
ar
d
d
ev
iatio
n
o
f
b
lu
e
c
h
an
n
el,
ix
)
n
o
r
m
alize
d
h
u
e
-
s
atu
r
atio
n
h
is
to
g
r
am
,
x
)
s
k
ewn
ess
,
x
i)
k
u
r
to
s
is
,
an
d
x
ii)
s
tan
d
ar
d
d
ev
iatio
n
o
f
th
e
in
ten
s
ity
v
alu
es
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
effec
tive
fea
tu
r
e
ex
tr
a
ctio
n
meth
o
d
fo
r
r
ice
lea
f d
is
ea
s
e
c
la
s
s
ifica
tio
n
(
Mu
h
a
mma
d
A
n
w
a
r
u
l A
z
im
)
467
Sh
ap
e
f
ea
tu
r
es
:
s
h
ap
e
f
ea
tu
r
es
ar
e
v
er
y
h
elp
f
u
l
wh
er
e
c
o
lo
r
s
o
f
m
u
ltip
le
d
is
ea
s
e
s
y
m
p
to
m
s
ar
e
s
im
ilar
.
W
e
u
s
e
Hu
m
o
m
en
t
in
v
ar
ian
t
s
[
2
4
]
as
s
h
ap
e
f
ea
tu
r
es.
T
h
es
e
ar
e
s
ev
en
m
ath
em
atica
l
m
o
m
en
ts
o
f
th
e
im
a
g
es
w
h
ich
ar
e
i
n
v
ar
ian
t
to
r
o
tatio
n
,
tr
an
s
latio
n
,
s
ca
le,
an
d
b
r
i
g
h
tn
ess
.
W
e
ex
tr
ac
t
a
to
tal
o
f
9
s
h
ap
e
f
ea
tu
r
es
f
r
o
m
th
e
af
f
ec
ted
ar
e
as
o
f
leaf
im
ag
es.
T
h
ese
ar
e:
i)
av
e
r
ag
e
c
o
n
to
u
r
p
er
im
eter
,
ii)
av
er
a
g
e
co
n
to
u
r
ar
ea
,
a
n
d
iii
-
ix
)
7
Hu
m
o
m
en
t in
v
ar
ian
ts
.
T
ex
tu
r
e
f
ea
tu
r
es:
ro
b
u
s
t
tex
tu
r
e
f
ea
tu
r
es
ar
e
test
ed
o
n
th
e
d
at
aset.
W
e
s
elec
t
g
r
ay
-
lev
el
co
-
o
cc
u
r
r
en
c
e
m
atr
ix
(
GL
C
M)
an
d
lo
ca
l
b
in
a
r
y
p
atter
n
(
L
B
P)
as
tex
tu
r
al
f
e
atu
r
e
d
escr
ip
to
r
s
.
GL
C
M
s
h
o
w
s
th
e
co
-
o
cc
u
r
r
en
ce
o
f
g
r
a
y
-
lev
el
p
atter
n
s
in
t
h
e
s
e
g
m
en
ted
s
p
o
ts
.
W
e
c
alcu
late
th
e
GL
C
M
in
4
d
ir
ec
tio
n
s
at
a
3
-
p
ix
el
d
is
tan
ce
f
r
o
m
th
e
r
ef
er
e
n
ce
p
ix
el.
Fro
m
GL
C
M,
th
e
f
o
u
r
f
e
atu
r
es
u
s
ed
ar
e
co
n
tr
ast,
c
o
r
r
elatio
n
,
en
e
r
g
y
,
an
d
h
o
m
o
g
e
n
eity
.
W
e
tak
e
th
e
m
ea
n
o
f
t
h
ese
v
al
u
es
in
4
s
p
ec
if
ie
d
d
ir
ec
tio
n
s
e.
g
.
,
0
,
4
5
,
9
0
,
1
3
5
d
eg
r
ee
s
.
W
e
em
p
lo
y
L
B
P
f
ea
tu
r
e
in
its
b
asic
f
o
r
m
as
p
r
o
p
o
s
ed
in
[
2
5
]
to
ca
p
tu
r
e
th
e
in
ten
s
ity
p
atter
n
f
r
o
m
th
e
s
eg
m
en
t
ed
im
ag
es.
L
B
P
is
b
r
ig
h
tn
ess
an
d
r
o
tatio
n
in
v
ar
ia
n
t.
I
t
r
e
p
r
esen
ts
th
e
lo
ca
l
c
h
an
g
e
in
in
ten
s
ity
p
atter
n
.
So
,
th
e
tex
tu
r
e
f
ea
tu
r
es
ar
e:
i)
m
ea
n
co
n
tr
ast
,
ii)
m
ea
n
co
r
r
elatio
n
,
iii)
m
ea
n
e
n
er
g
y
,
iv
)
m
ea
n
h
o
m
o
g
e
n
eity
,
an
d
v
)
n
o
r
m
alize
d
lo
ca
l
b
in
a
r
y
p
atter
n
h
is
to
g
r
am
.
W
e
ex
tr
ac
t
a
s
et
o
f
d
if
f
er
en
t
f
ea
tu
r
es
co
m
p
ar
ed
to
Pra
jap
ati
et
a
l.
[
4
]
f
o
r
class
if
icat
io
n
.
W
e
u
tili
ze
hue
-
s
atu
r
atio
n
h
is
to
g
r
a
m
wh
ich
ca
p
tu
r
es
th
e
r
elatio
n
o
f
p
i
x
el
co
l
o
r
s
.
A
co
m
p
letely
d
if
f
er
en
t
s
et
o
f
s
h
ap
e
f
ea
tu
r
e
d
escr
ip
t
o
r
is
u
s
ed
in
th
i
s
wo
r
k
.
W
e
f
in
d
th
ese
to
b
e
h
ig
h
ly
r
o
b
u
s
t
an
d
ef
f
ec
ti
v
e
in
d
is
tin
g
u
is
h
in
g
b
etwe
en
d
is
ea
s
e
class
e
s
with
s
im
ilar
co
lo
r
s
y
m
p
to
m
s
.
L
B
P h
is
to
g
r
am
is
also
u
s
ed
f
o
r
ef
f
ec
tiv
e
tex
tu
r
e
a
n
aly
s
is
.
3
.
4
.
Cla
s
s
if
ica
t
io
n
E
x
tr
em
e
g
r
a
d
ien
t
b
o
o
s
tin
g
o
r
XGBo
o
s
t
i
s
p
r
o
p
o
s
ed
b
y
C
h
en
et
a
l.
[
26
]
in
2
0
1
6
w
h
ich
is
an
ex
ten
s
io
n
of
d
ec
is
io
n
tr
ee
al
g
o
r
ith
m
.
T
h
is
b
r
in
g
s
s
ig
n
if
ican
t
p
er
f
o
r
m
a
n
ce
im
p
r
o
v
em
en
t
o
v
er
o
th
er
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
ith
m
s
in
clu
d
i
n
g
th
e
o
r
ig
i
n
al
g
r
ad
ien
t
b
o
o
s
ted
tr
ee
alg
o
r
ith
m
.
T
h
e
o
b
jectiv
e
f
u
n
ctio
n
i
s
g
iv
en
b
y
(
1
)
.
(
)
=
(
)
+
(
)
(
1
)
wh
er
e,
(
)
an
d
(
)
r
ep
r
esen
ts
tr
ain
in
g
lo
s
s
an
d
r
e
g
u
lar
izatio
n
,
i.
e.
m
o
d
el
c
o
m
p
lex
ity
o
f
th
e
e
n
s
em
b
le
r
esp
ec
tiv
ely
.
T
h
e
class
if
ier
b
u
i
ld
s
a
n
u
m
b
er
o
f
wea
k
lear
n
e
r
s
an
d
en
s
em
b
les
th
em
.
T
h
e
al
g
o
r
ith
m
f
ir
s
t
b
u
ild
s
a
tr
ee
b
ased
o
n
a
f
ea
tu
r
e.
Usi
n
g
th
e
o
b
jectiv
e
f
u
n
ctio
n
,
it g
en
e
r
ates a
n
o
th
er
tr
ee
wh
ich
im
p
r
o
v
es u
p
o
n
th
e
e
r
r
o
r
s
o
r
r
esid
u
als
o
f
t
h
e
p
r
ev
io
u
s
tr
e
e.
Du
r
in
g
n
ew
tr
ee
b
u
ild
in
g
,
e
r
r
o
r
o
r
r
esid
u
al
is
ca
lcu
lated
an
d
m
in
im
ized
u
s
in
g
g
r
ad
ien
t
d
escen
t.
T
r
ee
p
r
u
n
in
g
is
d
o
n
e
g
r
ee
d
ily
b
ased
o
n
ac
cu
r
ac
y
g
ain
in
ea
ch
s
p
lit.
Dep
th
-
f
ir
s
t
tr
ee
p
r
u
n
in
g
an
d
g
r
ad
ien
t
lo
s
s
m
in
im
izatio
n
b
o
o
s
t
th
e
d
ec
is
io
n
tr
ee
b
u
ild
in
g
p
r
o
ce
s
s
wh
ich
r
esu
lts
in
f
aster
ex
ec
u
tio
n
an
d
im
p
r
o
v
e
d
ac
cu
r
ac
y
.
4.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S
4
.
1
.
E
x
perim
ent
a
l
s
et
up
Fo
r
our
im
p
lem
en
tatio
n
,
Py
th
o
n
2
.
7
.
1
5
is
u
s
ed
alo
n
g
with
Op
e
n
C
V
2
.
4
.
1
3
in
a
d
esk
to
p
en
v
ir
o
n
m
e
n
t.
Scik
it
-
lear
n
an
d
XGBo
o
s
t
p
y
th
o
n
p
ac
k
ag
es
ar
e
u
s
ed
f
o
r
m
ac
h
in
e
lear
n
in
g
task
s
.
W
e
u
s
e
s
tan
d
ar
d
m
o
d
el
ev
alu
atio
n
m
etr
ics
lik
e
ac
c
u
r
a
cy
s
co
r
e,
an
d
f
1
s
co
r
e.
W
e
tu
n
e
th
r
ee
p
ar
a
m
eter
s
e.
g
.
lear
n
in
g
r
ate
,
m
ax
im
u
m
d
ep
th
,
a
n
d
m
in
im
u
m
ch
ild
wei
g
h
t
o
f
XGBo
o
s
t
class
if
ier
to
o
b
tain
th
e
b
est
p
e
r
f
o
r
m
an
ce
.
T
h
e
n
u
m
b
er
o
f
d
ec
is
io
n
tr
ee
s
an
d
tr
ee
lev
els
ar
e
two
v
e
r
y
im
p
o
r
tan
t
p
a
r
am
eter
s
th
at
c
o
n
tr
o
l
d
ata
f
itti
n
g
.
Hu
g
e
n
u
m
b
er
o
f
d
ec
is
io
n
tr
ee
s
ca
n
m
ak
e
th
e
en
s
em
b
le
m
o
d
el
co
m
p
le
x
a
n
d
ex
tr
em
ely
tim
e
co
n
s
u
m
in
g
.
Hig
h
v
alu
e
o
f
tr
e
e
lev
els
i.e
.
d
ee
p
e
r
tr
ee
s
ca
n
o
v
er
f
it
th
e
m
o
d
el.
Min
im
u
m
ch
ild
weig
h
t
in
d
icate
s
th
e
m
in
im
u
m
n
u
m
b
er
o
f
d
ata
s
am
p
les
th
at
n
ee
d
to
ex
is
t
in
a
d
ec
is
io
n
s
p
lit.
T
h
e
lo
s
s
f
u
n
ctio
n
ca
lcu
lates
th
e
er
r
o
r
d
u
r
in
g
n
ew
iter
atio
n
s
o
f
tr
ee
b
u
ild
in
g
.
L
o
g
is
tic
lo
s
s
f
u
n
ctio
n
is
u
s
ed
h
er
e
f
o
r
t
h
is
p
u
r
p
o
s
e.
2
0
%
d
ata
is
u
s
ed
f
o
r
test
in
g
th
e
m
o
d
el,
wh
ile
r
e
s
t
o
f
th
e
im
ag
es
is
allo
ca
ted
f
o
r
tr
ain
in
g
.
Fo
r
o
p
ti
m
al
p
er
f
o
r
m
an
ce
,
th
e
m
o
d
el
n
ee
d
s
to
b
e
ac
c
u
r
ate,
r
o
b
u
s
t,
s
im
p
le,
an
d
ef
f
ec
tiv
e
.
Var
io
u
s
co
m
b
in
atio
n
s
o
f
th
e
a
f
o
r
em
en
tio
n
ed
p
ar
am
eter
v
alu
es
ar
e
test
ed
f
o
r
ac
cu
r
ac
y
im
p
r
o
v
em
en
t.
T
ab
le
2
s
h
o
ws th
e
v
alu
es o
f
th
e
t
u
n
ed
p
ar
am
eter
s
f
o
r
th
e
class
if
ier
.
T
ab
le
2
.
T
u
n
ed
p
ar
am
eter
v
al
u
es o
f
XGBo
o
s
t
Le
a
r
n
i
n
g
r
a
t
e
M
a
x
i
m
u
m
d
e
p
t
h
M
i
n
i
m
u
m
c
h
i
l
d
w
e
i
g
h
t
0
.
1
3
5
4
.
2
.
Resul
t
s
W
e
also
class
if
y
with
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
u
s
in
g
r
ad
ial
b
asis
f
u
n
ctio
n
(
Gau
s
s
ian
)
k
er
n
el.
I
n
o
u
r
ap
p
r
o
ac
h
,
SVM
p
er
f
o
r
m
s
b
et
ter
th
an
th
e
f
in
d
in
g
s
o
f
Pra
jap
ati
et
a
l.
[
4
]
b
y
m
o
r
e
th
an
8
%
ac
cu
r
ac
y
.
T
h
is
im
p
r
o
v
em
e
n
t
clea
r
l
y
d
em
o
n
s
t
r
ates
th
e
ef
f
ec
tiv
en
ess
o
f
o
u
r
f
ea
tu
r
e
ex
tr
ac
tio
n
m
eth
o
d
.
Mo
r
eo
v
er
,
XGBo
o
s
t
o
u
tp
er
f
o
r
m
s
SVM
b
y
an
ac
cu
r
ac
y
o
f
ab
o
u
t
5
%,
ac
h
iev
i
n
g
o
v
er
all
ac
cu
r
ac
y
im
p
r
o
v
e
m
en
t
o
f
ab
o
u
t
1
3
%.
T
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
1
9
,
No
.
2
,
Ap
r
il
202
1
:
46
3
-
47
0
468
r
esu
lt
o
f
class
if
icat
io
n
p
r
o
ce
s
s
is
g
iv
en
in
T
ab
le
3
.
Fig
u
r
e
7
illu
s
tr
ates
th
e
c
lass
-
wi
s
e
an
d
o
v
er
all
ac
cu
r
ac
y
co
m
p
ar
is
o
n
b
etwe
en
o
u
r
ap
p
r
o
ac
h
an
d
th
e
p
r
ev
io
u
s
wo
r
k
o
f
Pra
jap
ati
e
t a
l
.
[
4
]
o
n
t
h
e
s
am
e
d
ataset.
T
ab
le
3
.
Per
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
b
etwe
en
XGBo
o
s
t a
n
d
SVM
with
R
B
F k
er
n
el
C
l
a
s
si
f
i
e
r
A
c
c
u
r
a
c
y
(
%)
W
e
i
g
h
t
e
d
a
v
e
r
a
g
e
F
-
1
sc
o
r
e
X
G
B
o
o
st
8
6
.
5
8
0
.
8
7
S
V
M
w
i
t
h
R
B
F
k
e
r
n
e
l
8
1
.
6
7
0
.
8
2
Fig
u
r
e
7
.
C
lass
-
wis
e
an
d
o
v
er
all
ac
cu
r
ac
y
4
.
3
.
Dis
cus
s
io
n
Am
o
n
g
th
e
s
elec
ted
f
ea
tu
r
es,
s
o
m
e
h
ad
s
ig
n
if
ican
t
im
p
ac
t
o
n
th
e
m
o
d
el
p
e
r
f
o
r
m
an
ce
.
N
o
r
m
alize
d
Hu
e
-
Satu
r
atio
n
h
is
to
g
r
am
is
wo
r
th
m
e
n
tio
n
in
g
.
I
t
is
a
s
tr
o
n
g
f
ea
tu
r
e
d
escr
ip
to
r
o
f
co
l
o
r
d
o
m
ain
.
I
n
o
u
r
o
b
s
er
v
atio
n
,
H
ue
-
S
atu
r
atio
n
2
D
h
is
to
g
r
am
h
ig
h
ly
c
o
n
tr
ib
u
t
ed
to
th
e
ac
c
u
r
ac
y
g
ain
.
GL
C
M
f
ea
tu
r
es
ar
e
also
p
r
o
m
in
e
n
t
with
r
esp
ec
t
to
ac
c
u
r
ac
y
g
ain
.
Me
an
c
o
r
r
elatio
n
a
n
d
co
n
tr
ast
o
f
th
e
s
eg
m
en
ted
im
ag
e
im
p
r
o
v
e
d
th
e
s
u
cc
ess
f
u
l
clas
s
if
icatio
n
r
ate
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el.
B
r
o
wn
sp
o
t
an
d
leaf
s
m
u
t
d
is
ea
s
es
h
av
e
alm
o
s
t
s
im
ilar
ef
f
ec
t
o
n
th
e
r
ice
lea
v
es.
B
o
th
o
f
th
e
d
is
ea
s
es
cr
ea
te
s
y
m
p
to
m
s
o
f
cir
c
u
lar
t
o
o
v
al
s
h
ap
e.
T
h
e
lesi
o
n
c
o
lo
r
s
ar
e
also
s
im
ilar
.
Pra
jap
ati
et
a
l.
[
4
]
f
ailed
t
o
ex
tr
ac
t
p
r
o
p
er
f
ea
tu
r
es
wh
ich
co
u
ld
d
is
tin
g
u
is
h
b
etw
ee
n
th
e
B
S
an
d
L
S
class
es
ac
co
r
d
in
g
to
th
eir
s
u
b
tle
s
h
ap
e
d
is
tin
ctio
n
.
Hen
c
e,
class
-
wis
e
ac
cu
r
ac
y
o
f
leaf
s
m
u
t
was
as
lo
w
as
4
0
%
in
th
eir
wo
r
k
.
I
n
o
u
r
wo
r
k
,
ca
r
ef
u
lly
s
elec
ted
s
h
ap
e
f
ea
tu
r
es
ar
e
f
o
u
n
d
h
i
g
h
ly
ef
f
ec
tiv
e
in
d
if
f
er
e
n
tiatin
g
b
etwe
en
th
ese
two
c
o
n
f
u
s
in
g
class
es.
W
ith
th
e
ch
o
s
en
f
ea
t
u
r
es,
b
o
t
h
SVM
an
d
XGBo
o
s
t
ac
h
iev
e
c
o
n
s
is
ten
t
class
-
wi
s
e
ac
cu
r
ac
y
.
T
h
e
d
et
ec
tio
n
r
ate
o
f
B
L
B
is
m
u
ch
h
ig
h
er
th
a
n
th
e
o
th
er
s
as
it
h
as
d
is
tin
ct
d
is
ea
s
e
s
ig
n
atu
r
e.
On
th
e
o
th
e
r
h
an
d
,
B
S
an
d
L
S
h
a
v
e
s
im
ilar
d
is
ea
s
e
s
p
o
ts
.
Fu
r
th
er
m
o
r
e,
r
eso
lu
tio
n
o
f
m
a
n
y
L
S
im
ag
es
in
th
e
d
ataset
is
s
ig
n
if
ican
tly
lo
wer
th
an
th
e
a
v
er
ag
e
r
eso
lu
tio
n
o
f
im
ag
es o
f
o
th
er
class
es.
Fo
r
th
is
r
ea
s
o
n
,
th
e
ac
cu
r
ac
y
r
ate
o
f
L
S d
etec
tio
n
,
alth
o
u
g
h
co
m
p
a
r
ab
le,
y
et
is
m
ar
g
in
ally
lo
wer
t
h
an
th
at
o
f
B
S.
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
we
p
r
o
p
o
s
e
an
a
p
p
r
o
ac
h
f
o
r
r
ice
leaf
d
is
ea
s
e
c
lass
if
icatio
n
.
T
h
e
ap
p
r
o
ac
h
d
e
tects
th
r
ee
r
ice
leaf
d
is
ea
s
es
e.
g
.
B
ac
ter
i
al
L
ea
f
B
lig
h
t,
B
r
o
wn
Sp
o
t,
a
n
d
L
ea
f
Sm
u
t.
I
n
th
is
wo
r
k
,
b
ac
k
g
r
o
u
n
d
s
o
f
t
h
e
im
ag
es
ar
e
r
em
o
v
ed
u
s
in
g
s
atu
r
atio
n
th
r
esh
o
ld
,
w
h
ile
af
f
ec
te
d
p
o
r
tio
n
s
ar
e
s
eg
m
en
ted
u
s
in
g
h
u
e
th
r
esh
o
ld
.
T
h
e
d
is
ea
s
es
m
an
if
est
d
if
f
er
en
t
s
y
m
p
to
m
s
o
n
in
f
ec
ted
leav
es.
S
ev
er
al
s
tatis
tical
f
ea
tu
r
es
p
er
tain
in
g
co
lo
r
,
s
h
ap
e,
an
d
tex
tu
r
e
d
o
m
ain
ar
e
ex
tr
ac
ted
f
r
o
m
th
e
im
a
g
es.
XGBo
o
s
t
is
u
s
ed
f
o
r
c
lass
if
icatio
n
p
r
o
ce
s
s
.
Mo
d
el
p
ar
am
eter
s
ar
e
tu
n
ed
f
o
r
o
p
ti
m
al
p
er
f
o
r
m
a
n
ce
.
T
h
e
n
o
v
el
co
n
tr
ib
u
tio
n
o
f
o
u
r
ap
p
r
o
ac
h
is
f
ea
tu
r
e
s
elec
tio
n
an
d
co
m
b
in
atio
n
.
T
h
is
r
esu
lted
i
n
h
ig
h
er
ac
c
u
r
ac
y
o
f
class
if
icati
o
n
f
r
o
m
p
r
ev
io
u
s
r
esear
ch
es
o
n
th
e
d
ataset
alo
n
g
with
co
n
s
is
ten
t
class
-
wis
e
ac
c
u
r
ac
y
.
8
6
.
5
8
%
ac
cu
r
ac
y
o
n
th
e
test
d
ataset
is
ac
h
iev
ed
with
a
s
m
all
n
u
m
b
er
o
f
f
ea
tu
r
es
(
2
6
f
ea
tu
r
es).
Pr
o
p
er
f
ea
tu
r
e
s
elec
tio
n
r
esu
lted
in
a
s
i
g
n
if
ican
t
ac
c
u
r
ac
y
im
p
r
o
v
em
e
n
t
o
v
e
r
th
e
p
r
ev
io
u
s
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
effec
tive
fea
tu
r
e
ex
tr
a
ctio
n
meth
o
d
fo
r
r
ice
lea
f d
is
ea
s
e
c
la
s
s
ifica
tio
n
(
Mu
h
a
mma
d
A
n
w
a
r
u
l A
z
im
)
469
wo
r
k
s
o
n
th
e
s
am
e
d
ataset.
W
e
aim
to
co
n
t
in
u
e
o
u
r
in
v
esti
g
at
io
n
f
o
r
ex
tr
ac
tin
g
a
n
ew
s
et
o
f
d
is
tin
ctiv
e
f
ea
tu
r
es
f
r
o
m
th
e
in
f
ec
te
d
ar
ea
s
wh
ich
will
f
u
r
th
er
in
cr
ea
s
e
th
e
ac
cu
r
ac
y
o
f
d
is
ea
s
e
d
etec
tio
n
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
will b
e
im
p
lem
en
ted
in
m
o
b
ile
ap
p
f
o
r
m
at
s
o
th
at
th
e
f
ar
m
e
r
s
ca
n
u
s
e
it o
n
-
fi
eld
f
o
r
d
etec
tin
g
d
is
ea
s
es.
RE
F
E
R
E
NC
E
S
[
1
]
In
tern
a
ti
o
n
a
l
R
ice
Re
se
a
rc
h
In
s
ti
tu
te
,
“
Ba
c
teria
l
b
li
g
h
t
-
IRRI
Ri
c
e
Kn
o
wle
d
g
e
Ba
n
k
,
”
Ri
c
e
Kn
o
wled
g
e
Ba
n
k
.
[On
li
n
e
].
A
v
a
il
a
b
le
:
h
tt
p
:/
/
ww
w.k
n
o
wle
d
g
e
b
a
n
k
.
irri
.
o
r
g
/t
ra
i
n
in
g
/fac
t
-
sh
e
e
ts/p
e
st
-
m
a
n
a
g
e
m
e
n
t/
d
ise
a
se
s/it
e
m
/
b
a
c
teria
l
-
b
li
g
h
t
(
Ac
c
e
ss
e
d
No
v
.
2
8
,
2
0
1
9
).
[
2
]
In
tern
a
ti
o
n
a
l
Rice
Re
se
a
rc
h
In
stit
u
te
,
“
Bro
wn
sp
o
t
-
IRRI
Rice
Kn
o
wle
d
g
e
Ba
n
k
,
”
Ri
c
e
Kn
o
wled
g
e
Ba
n
k
.
[On
li
n
e
]
.
Av
a
il
a
b
le
:
h
t
tp
:/
/www
.
k
n
o
w
led
g
e
b
a
n
k
.
irri
.
o
r
g
/t
ra
in
in
g
/fac
t
-
sh
e
e
ts/p
e
st
-
m
a
n
a
g
e
m
e
n
t/
d
ise
a
se
s/it
e
m
/b
ro
wn
-
sp
o
t
(
Ac
c
e
ss
e
d
No
v
.
2
8
,
2
0
1
9
).
[
3
]
H.
B.
P
ra
jap
a
ti
,
J.
P
.
S
h
a
h
,
a
n
d
V.
K.
Da
b
h
i
,
“
Rice
lea
f
d
ise
a
se
s
d
a
ta
se
t,
”
Distrib
u
ted
b
y
UCI
M
a
c
h
in
e
L
e
a
rn
in
g
Rep
o
sit
o
ry
,
2
0
1
9
.
[On
li
n
e
].
A
v
a
il
a
b
le
a
t:
h
tt
p
s://
a
rc
h
i
v
e
.
ics
.
u
c
i.
e
d
u
/ml/d
a
tas
e
ts/Ri
c
e
+
Lea
f
+
Dise
a
se
s
[
4
]
H.
B.
P
ra
jap
a
ti
,
J.
P
.
S
h
a
h
,
a
n
d
V.
K.
Da
b
h
i,
“
De
tec
ti
o
n
a
n
d
c
las
sifica
ti
o
n
o
f
r
ice
p
lan
t
d
ise
a
se
s,”
In
telli
g
e
n
t
De
c
isio
n
T
e
c
h
n
o
l
o
g
ies
,
IOS
Pre
ss
,
v
o
l.
1
1
,
n
o
.
3
,
p
p
.
3
5
7
-
3
7
3
,
2
0
1
7
.
[
5
]
J.
G
a
r
c
ia
a
n
d
J.
A.
Ba
rb
e
d
o
,
"
Di
g
it
a
l
ima
g
e
p
ro
c
e
ss
in
g
tec
h
n
i
q
u
e
s
fo
r
d
e
tec
ti
n
g
,
q
u
a
n
ti
f
y
in
g
a
n
d
c
las
sify
in
g
p
lan
t
d
ise
a
se
s,"
S
p
i
n
g
e
r P
l
u
s
,
v
o
l
.
2
,
p
p
.
1
-
1
2
,
2
0
1
3
.
[
6
]
H.
B.
P
ra
jap
a
ti
,
J.
P
.
S
h
a
h
,
a
n
d
V
.
K.
Da
b
h
i,
“
A
s
u
rv
e
y
o
n
d
e
tec
ti
o
n
a
n
d
c
las
sifica
ti
o
n
o
f
r
ice
p
lan
t
d
ise
a
se
s,"
2
0
1
6
IEE
E
In
t.
Co
n
f.
Cu
rr
e
n
t
T
re
n
d
s i
n
Ad
v
a
n
c
e
d
C
o
mp
u
ti
n
g
(ICCT
AC
)
,
2
0
1
6
,
p
p
.
1
-
8
.
[
7
]
S
.
S
h
ima
m
u
ra
,
K.
Ue
h
a
ra
,
a
n
d
S
.
Ko
a
k
u
ts
u
,
“
Au
to
m
a
ti
c
i
d
e
n
ti
f
ica
ti
o
n
o
f
p
lan
t
p
h
y
sio
l
o
g
ica
l
d
i
so
rd
e
rs
i
n
p
lan
t
fa
c
to
ries
with
a
rti
f
icia
l
li
g
h
t
u
si
n
g
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
two
r
k
s,
”
In
t.
J
o
u
rn
a
l
o
f
Ne
w
Co
mp
u
ter
A
rc
h
it
e
c
tu
re
s
a
n
d
th
e
ir A
p
p
li
c
a
t
io
n
s
,
v
o
l.
9
,
n
o
.
1
,
p
p
.
2
5
-
3
0
,
2
0
1
9
.
[
8
]
M.
S
u
g
imo
t
o
,
Y.
I
n
o
k
i,
T.
S
h
ira
k
a
wa
,
K.
Tak
e
u
c
h
i,
D.
Y
o
sh
i
o
k
a
,
H.
F
u
k
it
a
,
e
t
a
l
.
,
“
A
stu
d
y
fo
r
d
e
v
e
lo
p
m
e
n
t
o
f
a
u
to
n
o
m
o
u
s
p
a
d
d
y
-
we
e
d
i
n
g
r
o
b
o
t
sy
ste
m
–
An
e
x
p
e
rime
n
tatio
n
fo
r
a
u
to
n
o
m
o
u
sly
stra
ig
h
t
-
ru
n
n
i
n
g
b
a
s
e
d
o
n
c
o
m
p
a
ss
-
c
o
m
p
e
n
sa
ti
o
n
,
”
In
t.
J
o
u
rn
a
l
o
f
Ne
w
Co
mp
u
ter
Arc
h
it
e
c
tu
re
s
a
n
d
t
h
e
ir
Ap
p
l
ica
ti
o
n
s
,
v
o
l.
8
,
n
o
.
4
,
p
p
.
1
8
6
-
1
9
7
,
2
0
1
8
.
[
9
]
S
.
D.
Kh
ira
d
e
a
n
d
A.
B.
P
a
ti
l,
“
P
lan
t
d
ise
a
se
d
e
tec
ti
o
n
u
sin
g
ima
g
e
p
r
o
c
e
ss
in
g
,
”
In
t
.
C
o
n
f
.
o
n
Co
mp
u
ti
n
g
Co
mm
u
n
ica
ti
o
n
C
o
n
tro
l
a
n
d
Au
t
o
ma
ti
o
n
(ICCUBEA
)
,
I
EE
E
,
2
0
1
5
,
p
p
.
7
6
8
-
771
.
[
1
0
]
H.
Wan
g
,
G
.
Li
,
Z.
M
a
,
a
n
d
X.
L
i,
"
Im
a
g
e
re
c
o
g
n
it
i
o
n
o
f
p
lan
t
d
is
e
a
se
s
b
a
se
d
o
n
Ba
c
k
p
ro
p
a
g
a
ti
o
n
Ne
two
rk
s,"
5
th
In
t.
C
o
n
g
re
ss
o
n
Im
a
g
e
a
n
d
S
i
g
n
a
l
Pro
c
e
ss
in
g
(CIS
P),
IEE
E,
2
0
1
2
,
p
p
.
8
9
4
-
9
0
0
.
[
1
1
]
J.
W.
Orill
o
,
J.
D.
Cru
z
,
L.
A
g
a
p
it
o
,
P
.
J.
S
a
ti
m
b
re
,
a
n
d
I
.
Va
len
z
u
e
l
a
,
"
Id
e
n
ti
f
ica
ti
o
n
o
f
d
ise
a
se
s
in
rice
p
lan
t
(Ory
z
a
sa
ti
v
a
)
u
sin
g
Ba
c
k
p
r
o
p
a
g
a
ti
o
n
A
rti
ficia
l
Ne
u
ra
l
Ne
two
rk
,
"
In
t.
C
o
n
f.
o
n
Hu
ma
n
o
id
,
Na
n
o
tec
h
n
o
l
o
g
y
,
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
Co
mm
u
n
ica
ti
o
n
a
n
d
Co
n
tro
l,
E
n
v
iro
n
me
n
t
a
n
d
M
a
n
a
g
e
me
n
t
(HNICEM
),
IE
EE
,
2
0
1
4
,
p
p
.
1
-
6
.
[
1
2
]
T.
S
u
m
a
n
a
n
d
T.
Dh
r
u
v
a
k
u
m
a
r,
“
Clas
sifica
ti
o
n
o
f
p
a
d
d
y
lea
f
d
ise
a
se
s
u
sin
g
sh
a
p
e
a
n
d
c
o
lo
r
fe
a
tu
re
s,”
In
t.
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
E
lec
tro
n
ics
En
g
in
e
e
rs
,
v
o
l
.
7
,
n
o
.
1
,
p
p
.
2
3
9
-
2
5
0
,
2
0
1
5
.
[
1
3
]
A.
K.
S
i
n
g
h
,
A.
Ru
b
a
iy
a
,
a
n
d
B.
S
.
Ra
ja,
“
Clas
sifica
ti
o
n
o
f
rice
d
ise
a
se
u
sin
g
d
ig
i
tal
ima
g
e
p
ro
c
e
ss
in
g
a
n
d
S
VM
c
las
sifier,”
In
t.
J
o
u
r
n
a
l
o
f
El
e
c
trica
l
a
n
d
El
e
c
tro
n
ics
E
n
g
i
n
e
e
rs
,
v
o
l.
7
,
n
o
.
1
,
p
p
.
2
9
4
-
2
9
9
,
2
0
1
5
.
[
1
4
]
P
.
S
e
t
h
y
,
B.
Ne
g
i
,
a
n
d
N.
B
h
o
i
,
“
D
e
te
c
ti
o
n
o
f
h
e
a
lt
h
y
a
n
d
d
e
fe
c
ted
d
i
se
a
se
d
lea
f
o
f
rice
c
ro
p
u
si
n
g
K
-
M
e
a
n
s
c
lu
ste
rin
g
tec
h
n
iq
u
e
,
”
In
t
.
J
o
u
rn
a
l
o
f
Co
mp
u
ter
Ap
p
li
c
a
ti
o
n
s
,
v
o
l.
1
5
7
,
n
o
.
1
,
2
0
1
7
.
[
1
5
]
Q.
Ya
o
,
Z.
G
u
a
n
,
Y.
Zh
o
u
,
J.
Tan
g
,
Y.
Hu
,
a
n
d
B.
Ya
n
g
,
“
Ap
p
li
c
a
ti
o
n
o
f
S
u
p
p
o
rt
Ve
c
to
r
M
a
c
h
i
n
e
fo
r
d
e
tec
ti
n
g
rice
d
ise
a
se
s u
sin
g
sh
a
p
e
a
n
d
c
o
lo
r
te
x
tu
re
fe
a
tu
re
s,”
I
n
t.
C
o
n
f
.
o
n
E
n
g
i
n
e
e
rin
g
C
o
mp
u
ta
ti
o
n
,
IEE
E
,
2
0
0
9
,
p
p
.
7
9
-
83
.
[
1
6
]
S
.
P
h
a
d
ik
a
r,
J.
S
il
,
a
n
d
A.
Da
s,
“
C
las
sifica
ti
o
n
o
f
rice
lea
f
d
ise
a
se
s
b
a
se
d
o
n
m
o
r
p
h
o
lo
g
ica
l
c
h
a
n
g
e
s,”
In
t.
J
o
u
rn
a
l
o
f
In
fo
rm
a
t
io
n
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
,
v
o
l.
2
,
n
o
.
3
,
p
p
.
4
6
0
-
4
6
3
,
2
0
1
2
.
[
1
7
]
S
.
P
h
a
d
ik
a
r
a
n
d
J.
S
il
,
"
Rice
d
ise
a
se
id
e
n
ti
fica
ti
o
n
u
sin
g
p
a
tt
e
r
n
re
c
o
g
n
it
io
n
tec
h
n
iq
u
e
s,"
1
1
th
I
n
t.
C
o
n
f.
o
n
Co
m
p
u
ter
a
n
d
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
IE
E
E,
2
0
0
8
,
p
p
.
4
2
0
-
4
2
3
.
[
1
8
]
C.
K.
Ch
a
rli
e
p
a
u
l,
“
Clas
sifica
ti
o
n
o
f
rice
p
lan
t
lea
f
u
sin
g
fe
a
tu
re
m
a
tch
in
g
,
”
In
t.
J
o
u
rn
a
l
o
n
E
n
g
in
e
e
rin
g
T
e
c
h
n
o
l
o
g
y
a
n
d
S
c
ien
c
e
s
,
v
o
l.
1
,
n
o
.
7
,
p
p
.
2
9
0
-
2
9
5
,
2
0
1
4
.
[
1
9
]
S
.
P
h
a
d
i
k
a
r,
J.
S
i
l,
a
n
d
A.
Da
s,
“
Rice
d
ise
a
se
s
c
las
sifica
ti
o
n
u
si
n
g
fe
a
tu
re
se
lec
ti
o
n
a
n
d
ru
le
g
e
n
e
ra
ti
o
n
te
c
h
n
i
q
u
e
s,"
Co
mp
u
ter
s
a
n
d
e
lec
tro
n
ics
i
n
a
g
r
icu
lt
u
re
,
v
o
l
.
9
0
,
p
p
.
7
6
-
8
5
,
2
0
1
3
.
[
2
0
]
N.
N.
Ku
rn
iaw
a
ti
a
n
d
S
.
Ab
d
u
l
lah
,
"
Tex
tu
re
a
n
a
ly
sis
f
o
r
d
iag
n
o
sin
g
p
a
d
d
y
d
ise
a
se
,
"
In
t.
Co
n
f.
o
n
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
I
n
fo
rm
a
ti
c
s
,
IEE
E,
v
o
l.
1
,
2
0
0
9
,
p
p
.
2
3
-
27
.
[
2
1
]
G
.
An
th
o
n
y
s a
n
d
N.
Wi
c
k
ra
m
a
ra
c
h
c
h
i,
“
An
ima
g
e
re
c
o
g
n
it
i
o
n
sy
st
e
m
fo
r
c
ro
p
d
ise
a
se
id
e
n
ti
fica
ti
o
n
o
f
p
a
d
d
y
fiel
d
s
in
S
ri
Lan
k
a
,
”
In
t.
Co
n
f.
o
n
I
n
d
u
s
tria
l
a
n
d
In
f
o
rm
a
ti
o
n
S
y
ste
ms
,
IE
EE
,
2
0
0
9
,
p
p
.
4
0
3
-
4
0
7
.
[
2
2
]
S
.
S
.
C
h
a
wa
th
e
,
"
Rice
d
ise
a
se
d
e
tec
ti
o
n
b
y
ima
g
e
a
n
a
ly
sis,"
1
0
t
h
A
n
n
u
a
l
Co
m
p
u
ti
n
g
a
n
d
C
o
mm
u
n
ica
ti
o
n
W
o
rk
sh
o
p
a
n
d
Co
n
f.
(CCW
C),
Las
Ve
g
a
s,
NV
,
USA,
2
0
2
0
,
p
p
.
0
5
2
4
-
0
5
3
0
.
[
2
3
]
K.
M
a
ji
d
,
Y
.
He
rd
iy
e
n
i,
a
n
d
A.
M
.
Ra
u
f,
"
I
-
P
EDIA:
M
o
b
i
le
a
p
p
li
c
a
ti
o
n
f
o
r
p
a
d
d
y
d
ise
a
se
i
d
e
n
ti
fica
ti
o
n
u
sin
g
F
u
z
z
y
e
n
tro
p
y
a
n
d
P
ro
b
a
b
il
isti
c
Ne
u
ra
l
Ne
two
rk
,
"
I
n
t.
C
o
n
f.
o
n
Ad
v
a
n
c
e
d
Co
mp
u
ter
S
c
ien
c
e
a
n
d
In
f
o
r
ma
ti
o
n
S
y
ste
ms
(ICACS
IS
),
IEE
E
,
2
0
1
3
,
p
p
.
4
0
3
-
406
.
[
2
4
]
M
.
K.
Hu
,
“
Visu
a
l
p
a
t
tern
re
c
o
g
n
it
io
n
b
y
m
o
m
e
n
t
i
n
v
a
ria
n
ts,”
I
RE
T
r
a
n
sa
c
t
io
n
s
o
n
In
f
o
rm
a
ti
o
n
T
h
e
o
ry
,
v
o
l
.
8
,
no.
2
,
p
p
.
1
7
9
-
1
8
7
,
1
9
6
2
.
[
2
5
]
T.
Oja
la,
M
.
P
ietik
ä
i
n
e
n
,
a
n
d
D.
Ha
rwo
o
d
,
"
P
e
rfo
rm
a
n
c
e
e
v
a
lu
a
ti
o
n
o
f
te
x
tu
re
m
e
a
su
re
s
with
c
las
sifica
ti
o
n
b
a
se
d
o
n
K
u
ll
b
a
c
k
d
isc
rimi
n
a
ti
o
n
o
f
d
istri
b
u
ti
o
n
s,"
Pro
c
.
1
2
th
IAP
R
In
t
.
C
o
n
f
.
o
n
Pa
t
ter
n
Rec
o
g
n
i
ti
o
n,
v
o
l.
1
,
1
9
9
4
,
p
p
.
5
8
2
-
585
.
[
2
6
]
T.
C
h
e
n
a
n
d
C.
G
u
e
strin
,
“
Xg
b
o
o
st:
A
sc
a
lab
le
tree
b
o
o
st
in
g
s
y
ste
m
,
”
Pro
c
.
2
2
n
d
ACM
S
IGKD
D
In
t.
Co
n
f.
o
n
Kn
o
wled
g
e
Disc
o
v
e
ry
a
n
d
Da
t
a
M
in
i
n
g
,
2
0
1
6
,
p
p
.
7
8
5
-
7
9
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
1
9
,
No
.
2
,
Ap
r
il
202
1
:
46
3
-
47
0
470
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Mu
h
a
m
m
a
d
Anwa
r
u
l
Az
im
re
c
e
iv
e
d
B.
S
c
.
(E
n
g
g
.
)
d
e
g
re
e
i
n
El
e
c
tro
n
ics
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
fro
m
S
h
a
h
jala
l
Un
i
v
e
rsit
y
o
f
S
c
ien
c
e
&
Tec
h
n
o
l
o
g
y
,
S
y
l
h
e
t,
Ba
n
g
lad
e
sh
i
n
2
0
0
0
a
n
d
M
a
ste
r
o
f
En
g
i
n
e
e
rin
g
i
n
C
o
m
p
u
ter
E
n
g
i
n
e
e
rin
g
fr
o
m
Ko
re
a
Ae
ro
sp
a
c
e
Un
iv
e
rsity
,
S
o
u
t
h
Ko
re
a
in
2
0
0
8
.
He
is
w
o
rk
in
g
a
s
th
e
Ch
a
irma
n
a
n
d
P
ro
fe
s
so
r
in
th
e
d
e
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
ri
n
g
,
U
n
iv
e
rsit
y
o
f
Ch
it
tag
o
n
g
,
B
a
n
g
lad
e
sh
.
His
re
se
a
rc
h
in
tere
st
in
c
l
u
d
e
s
ima
g
e
p
ro
c
e
ss
in
g
,
c
o
m
p
u
ter v
isio
n
,
p
a
tt
e
rn
re
c
o
g
n
it
io
n
,
m
a
c
h
in
e
lea
rn
i
n
g
,
n
e
two
rk
i
n
g
,
a
n
d
se
c
u
r
it
y
.
Mo
h
a
m
m
a
d
K
h
a
irul
Is
la
m
re
c
e
i
v
e
d
th
e
B.
S
c
.
(
E
n
g
g
.
)
d
e
g
re
e
in
El
e
c
tro
n
ics
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
fro
m
S
h
a
h
jala
l
Un
i
v
e
rsit
y
o
f
S
c
ien
c
e
&
Tec
h
n
o
l
o
g
y
,
S
y
l
h
e
t,
Ba
n
g
la
d
e
sh
i
n
2
0
0
0
.
He
re
c
e
iv
e
d
th
e
M
a
ste
r
o
f
E
n
g
in
e
e
rin
g
a
n
d
Do
c
to
r
o
f
E
n
g
i
n
e
e
rin
g
d
e
g
re
e
s
i
n
In
f
o
rm
a
ti
o
n
a
n
d
Tele
c
o
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
fr
o
m
Ko
re
a
Ae
ro
sp
a
c
e
Un
iv
e
rsity
,
S
o
u
th
Ko
re
a
in
2
0
0
7
a
n
d
2
0
1
1
re
sp
e
c
ti
v
e
ly
.
Dr.
Isla
m
is
wo
rk
in
g
a
s
a
P
ro
fe
ss
o
r
in
t
h
e
d
e
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
in
Un
iv
e
rsit
y
o
f
Ch
it
tag
o
n
g
,
Ch
it
ta
g
o
n
g
-
43
3
1
,
Ba
n
g
lad
e
sh
.
His
re
se
a
rc
h
in
tere
st
in
c
lu
d
e
s
ima
g
e
u
n
d
e
rsta
n
d
in
g
,
a
c
t
iv
it
y
a
n
a
ly
sis,
c
o
m
p
u
ter
v
isio
n
,
p
a
tt
e
r
n
re
c
o
g
n
it
io
n
,
m
a
c
h
in
e
lea
rn
in
g
,
3
D
m
o
d
e
li
n
g
a
n
d
f
u
sio
n
o
f
re
m
o
te
se
n
sin
g
d
a
ta.
Md
.
Ma
r
u
fu
r
Ra
h
m
a
n
c
o
m
p
lete
d
h
is
B.
S
c
.
E
n
g
g
.
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
ri
n
g
fro
m
Un
iv
e
rsit
y
o
f
Ch
it
tag
o
n
g
,
Ch
it
tag
o
n
g
-
4
3
3
1
,
Ba
n
g
lad
e
sh
in
2
0
1
6
.
At
p
re
se
n
t
h
e
is
p
u
rsu
i
n
g
M
S
i
n
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
in
t
h
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
,
U
n
iv
e
rsit
y
o
f
Ch
it
tag
o
n
g
,
Ba
n
g
lad
e
s
h
.
His
re
s
e
a
rc
h
in
tere
st
in
c
lu
d
e
s
m
a
c
h
in
e
lea
rn
in
g
,
p
a
tt
e
r
n
re
c
o
g
n
i
ti
o
n
,
ima
g
e
p
r
o
c
e
ss
in
g
,
a
n
d
c
o
m
p
u
ter v
isi
o
n
.
Fa
r
a
h
J
a
h
a
n
re
c
e
iv
e
d
th
e
B.
S
c
.
(
Ho
n
s.)
d
e
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
fro
m
Un
iv
e
rsity
o
f
Ch
it
ta
g
o
n
g
,
C
h
it
tag
o
n
g
-
4
3
3
1
,
Ba
n
g
la
d
e
sh
in
2
0
0
8
.
S
h
e
re
c
e
iv
e
d
th
e
M
a
ste
r
o
f
En
g
i
n
e
e
rin
g
d
e
g
re
e
in
In
f
o
rm
a
ti
o
n
a
n
d
Tele
c
o
m
m
u
n
ica
ti
o
n
En
g
in
e
e
rin
g
fr
o
m
Ko
re
a
Ae
ro
sp
a
c
e
Un
iv
e
rsity
,
S
o
u
t
h
Ko
r
e
a
in
2
0
1
1
.
S
h
e
re
c
e
iv
e
d
th
e
P
h
.
D.
d
e
g
re
e
fro
m
G
riff
it
h
Un
iv
e
rsity
,
Au
stra
li
a
in
A
u
g
u
st
2
0
1
9
.
Dr.
Ja
h
a
n
is
a
n
As
so
c
iate
P
ro
fe
ss
o
r
in
th
e
d
e
p
a
rtme
n
t
o
f
C
o
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
in
Un
i
v
e
rsity
o
f
Ch
i
tt
a
g
o
n
g
,
Ch
i
tt
a
g
o
n
g
-
4
3
3
1
,
Ba
n
g
lad
e
sh
.
He
r
re
se
a
rc
h
in
tere
st
in
c
lu
d
e
s p
a
tt
e
rn
re
c
o
g
n
it
i
o
n
,
m
a
c
h
in
e
lea
rn
i
n
g
,
c
o
m
p
u
ter
v
isio
n
,
a
n
d
d
a
ta
fu
sio
n
i
n
t
h
e
field
o
f
re
m
o
te se
n
si
n
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.