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.
37
,
No
.
1
,
J
an
u
ar
y
20
25
,
p
p
.
47
5
~
48
7
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
37.
i
1
.
p
p
47
5
-
4
8
7
475
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Field
-
l
ev
el sug
a
rc
a
ne yield estima
t
i
o
n ut
ilizing
S
enti
nel
-
2
t
im
e
-
series a
nd ma
chin
e learning
Rek
ha
B
.
U
.
1
,
Vee
na
V
.
Desa
i
1
,
Su
re
s
h K
ur
i
1
,
P
ra
t
ij
ny
a
S
Aj
a
wa
n
1
,
Su
nil
K
um
a
r
J
ha
2
, V
.
C
.
P
a
t
il
3
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
K
LS
G
o
g
t
e
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
A
f
f
i
l
i
a
t
e
d
t
o
V
i
s
v
e
s
v
a
r
a
y
a
T
e
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
B
e
l
a
g
a
v
i
,
I
n
d
i
a
2
A
p
p
l
i
e
d
A
g
r
i
c
u
l
t
u
r
a
l
R
e
m
o
t
e
S
e
n
s
i
n
g
C
e
n
t
r
e
,
U
n
i
v
e
r
si
t
y
o
f
N
e
w
E
n
g
l
a
n
d
,
A
r
mi
d
a
l
e
,
A
u
st
r
a
l
i
a
3
K
J
S
o
m
a
i
y
a
I
n
st
i
t
u
t
e
o
f
A
p
p
l
i
e
d
A
g
r
i
c
u
l
t
u
r
a
l
R
e
s
e
a
r
c
h
,
B
a
g
a
l
k
o
t
,
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
Ap
r
16
,
2
0
2
4
R
ev
is
ed
Sep
10
,
2
0
2
4
Acc
ep
ted
Sep
29
,
2
0
2
4
Th
is
wo
r
k
fo
c
u
se
d
o
n
d
e
v
e
l
o
p
i
n
g
a
m
e
th
o
d
o
lo
g
y
f
o
r
u
si
n
g
m
a
c
h
i
n
e
lea
rn
in
g
(M
L)
a
p
p
r
o
a
c
h
e
s
to
e
sta
b
li
sh
a
p
re
-
h
a
rv
e
st
y
ield
p
re
d
icti
o
n
m
o
d
e
l
fo
r
su
g
a
rc
a
n
e
a
t
field
lev
e
l
b
y
i
n
teg
ra
ti
n
g
ti
m
e
-
se
ries
re
m
o
te
se
n
sin
g
ima
g
e
ry
d
a
ta
with
M
L
tec
h
n
iq
u
e
s.
G
ro
u
n
d
tr
u
th
a
g
ro
d
a
ta
a
n
d
th
irt
y
-
o
n
e
sp
e
c
tral
v
e
g
e
tatio
n
in
d
ice
s
we
re
e
x
trac
ted
fro
m
S
e
n
ti
n
e
l
-
2
i
m
a
g
e
ry
a
n
d
we
re
c
o
n
sid
e
re
d
f
o
r
y
ield
m
o
d
e
li
n
g
.
A
two
-
lev
e
l
fe
a
tu
re
se
lec
ti
o
n
tec
h
n
iq
u
e
wa
s
u
se
d
to
d
e
term
in
e
th
e
m
o
st
sig
n
ifi
c
a
n
t
v
a
riab
les
th
a
t
b
e
st
c
o
rre
l
a
ted
with
su
g
a
rc
a
n
e
y
ield
to
p
re
d
ict
y
ield
i
n
a
d
v
a
n
c
e
.
S
e
v
e
n
M
L
a
lg
o
rit
h
m
s,
in
c
lu
d
in
g
t
h
o
se
b
a
se
d
o
n
re
g
u
lariz
a
ti
o
n
,
d
e
c
isio
n
tree
s,
a
n
d
e
n
se
m
b
le
m
e
t
h
o
d
s
li
k
e
b
o
o
st
in
g
,
we
re
u
se
d
t
o
p
re
d
ict
y
i
e
ld
.
T
h
e
a
p
p
ro
a
c
h
a
c
h
iev
e
d
t
h
e
h
ig
h
e
st
R
2
sc
o
re
o
f
0
.
7
3
a
n
d
th
e
l
o
we
st
ro
o
t
m
e
a
n
sq
u
a
re
d
e
rro
r
(RM
S
E
)
o
f
1
3
.
4
5
t/
h
a
with
ra
n
d
o
m
fo
re
st
(RF
)
a
m
o
n
g
th
e
s
e
v
e
n
M
L
m
o
d
e
ls
tes
ted
.
F
u
r
th
e
rm
o
re
,
a
ll
fe
a
tu
re
se
lec
ti
o
n
p
ro
c
e
d
u
re
s
id
e
n
ti
fie
d
n
o
rm
a
li
z
e
d
d
iffere
n
c
e
re
d
e
d
g
e
(ND
RE),
re
d
e
d
g
e
c
h
l
o
ro
p
h
y
ll
i
n
d
e
x
(RECI),
a
n
d
ra
ti
o
v
e
g
e
tati
o
n
in
d
e
x
(RVI)
a
s
m
a
jo
r
y
ield
-
d
r
iv
i
n
g
v
a
riab
les
.
Th
e
e
x
p
e
rime
n
ts
d
u
r
in
g
fe
a
tu
re
se
le
c
ti
o
n
d
e
m
o
n
stra
ted
th
e
p
o
ten
ti
a
l
o
f
re
d
e
d
g
e
sp
e
c
tral
b
a
n
d
s
in
d
e
v
e
lo
p
m
e
n
t
o
f
a
re
li
a
b
le
su
g
a
rc
a
n
e
-
y
ield
p
re
d
ictio
n
a
p
p
r
o
a
c
h
.
Th
e
RF
m
o
d
e
l
o
b
tain
e
d
u
sin
g
t
h
e
p
ro
p
o
se
d
m
e
th
o
d
o
lo
g
y
o
u
t
p
e
rfo
rm
s
th
e
two
b
a
se
li
n
e
m
o
d
e
ls
d
e
v
e
lo
p
e
d
u
si
n
g
ND
VI
a
n
d
G
ND
VI
in
d
ice
s,
w
it
h
a
n
imp
ro
v
e
d
RM
S
E
o
f
1
6
-
1
8
%
.
K
ey
w
o
r
d
s
:
Ma
ch
in
e
lear
n
in
g
R
ed
ed
g
e
s
p
ec
tr
al
b
an
d
s
R
em
o
te
s
en
s
in
g
Su
g
ar
ca
n
e
y
ield
Veg
etatio
n
in
d
ices tim
e
s
er
ies
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
:
R
ek
h
a
B
.
U
.
Dep
ar
tm
en
t o
f
E
lectr
o
n
ics an
d
C
o
m
m
u
n
icatio
n
E
n
g
in
ee
r
i
n
g
,
KL
S Go
g
te
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
Af
f
iliated
to
Vis
v
esv
ar
ay
a
T
e
ch
n
o
lo
g
ical
Un
iv
er
s
ity
B
elag
av
i,
5
9
0
0
1
8
,
I
n
d
ia
E
m
ail: d
h
d
r
e
k
h
a@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Su
g
ar
ca
n
e
is
a
m
u
ltifu
n
ctio
n
al
cr
o
p
m
ain
ly
u
tili
ze
d
f
o
r
e
x
tr
ac
tin
g
s
u
g
ar
a
n
d
with
in
a
d
iv
er
s
if
ied
p
o
r
tf
o
lio
.
Pre
cise
ea
r
ly
-
s
ea
s
o
n
also
k
n
o
wn
as
“with
in
-
th
e
-
s
ea
s
o
n
”
y
ield
p
r
ed
ictio
n
o
f
s
u
g
ar
ca
n
e
h
as
an
im
p
o
r
tan
t
r
o
le
in
th
e
m
a
n
ag
em
en
t
o
f
ag
r
o
ec
o
s
y
s
tem
s
[
1
]
.
T
h
e
c
o
n
v
e
n
tio
n
al
m
et
h
o
d
s
o
f
s
u
g
ar
c
a
n
e
y
iel
d
esti
m
atio
n
b
ef
o
r
e
h
ar
v
est
in
v
o
lv
e
m
an
u
al
s
co
u
tin
g
th
r
o
u
g
h
estab
lis
h
ed
s
u
g
ar
ca
n
e
f
ield
s
an
d
cr
o
p
cu
ttin
g
ex
p
er
im
en
ts
(
C
C
E
)
.
W
alk
in
g
th
r
o
u
g
h
th
e
d
en
s
e
s
u
g
ar
ca
n
e
c
r
o
p
s
th
at
ar
e
as
tall
as
6
m
p
r
o
b
e
ch
allen
g
e
t
o
th
e
em
p
lo
y
ee
s
.
As
a
r
esu
lt
th
e
co
n
v
en
t
io
n
al
m
eth
o
d
s
ar
e
p
r
o
n
e
to
i
n
ac
cu
r
ac
ies
an
d
o
f
ten
b
ia
s
ed
d
u
e
to
h
u
m
a
n
er
r
o
r
s
.
Sin
ce
th
e
estab
lis
h
m
en
t
o
f
co
m
m
e
r
cial
cu
ltiv
atio
n
o
f
s
u
g
ar
ca
n
e,
i
n
th
e
r
ec
e
n
t
p
ast
a
f
ew
s
u
g
ar
m
ills
ar
e
ex
p
lo
r
in
g
t
h
e
p
o
s
s
ib
ilit
i
es
o
f
esti
m
atin
g
s
u
g
ar
ca
n
e
y
ield
u
s
in
g
g
lo
b
al
p
o
s
itio
n
in
g
s
y
s
tem
(
GPS)
tech
n
o
lo
g
y
th
r
o
u
g
h
m
a
n
u
al
s
u
r
v
ey
o
f
f
ield
p
er
im
eter
s
in
th
eir
m
ill
ca
tch
m
en
t
ar
ea
p
r
o
m
o
t
in
g
in
ter
v
en
tio
n
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
1
,
J
an
u
ar
y
20
25
:
47
5
-
48
7
476
en
g
in
ee
r
in
g
tech
n
o
l
o
g
ies
in
s
u
g
ar
ca
n
e
cr
o
p
y
ield
p
r
ed
icti
o
n
v
ia
f
ield
-
b
ased
s
tu
d
ies
[
2
]
.
Hig
h
ly
d
etailed
r
em
o
te
-
s
en
s
in
g
d
ata
with
p
r
ec
is
e
tem
p
o
r
al
an
d
s
p
atial
r
eso
lu
tio
n
ca
n
o
f
f
e
r
v
alu
a
b
le
in
s
ig
h
ts
in
to
th
e
d
ev
elo
p
m
e
n
t
a
n
d
y
ield
o
f
s
u
g
ar
ca
n
e
cr
o
p
s
[
3
]
-
[
6
]
.
T
h
e
a
p
p
licatio
n
o
f
r
em
o
te
-
s
en
s
in
g
tech
n
o
l
o
g
ies
also
f
ac
ilit
ates
n
o
n
-
in
v
asiv
e
an
d
u
n
b
iased
esti
m
atio
n
s
o
f
cr
o
p
p
r
o
d
u
ctio
n
.
Fu
r
th
er
m
o
r
e,
v
e
g
etatio
n
in
d
ices
a
r
e
n
u
m
er
ical
i
n
d
icato
r
s
t
h
at
in
d
i
ca
te
th
e
lev
el
o
f
v
e
g
etatio
n
v
itality
.
Sev
er
al
v
eg
etatio
n
in
d
icato
r
s
,
in
clu
d
in
g
NDVI
[
7
]
,
[
8
]
,
R
VI
[
9
]
,
GNDV
I
[
1
0
]
,
E
VI
,
NDWI,
an
d
SAVI
[
1
1
]
,
ar
e
cu
r
r
en
tly
u
tili
ze
d
in
th
e
d
ev
elo
p
m
en
t
o
f
s
tati
s
tical
m
eth
o
d
s
f
o
r
s
u
g
a
r
ca
n
e
y
ield
p
r
ed
ictio
n
.
W
h
en
it
co
m
es
to
p
r
e
d
ictin
g
s
u
g
ar
ca
n
e
p
r
o
d
u
ctiv
ity
o
n
in
d
iv
id
u
al
f
ar
m
s
in
I
r
an
,
th
e
g
r
ee
n
v
eg
etatio
n
in
d
ex
(
GVI
)
is
m
o
r
e
ac
cu
r
ate
th
an
b
o
th
GNDV
I
an
d
NDVI
[
1
2
]
.
I
n
an
ef
f
o
r
t
t
o
esti
m
ate
t
h
e
s
u
g
ar
ca
n
e
y
ield
at
th
e
p
lo
t
lev
el
u
s
in
g
m
ac
h
in
e
lea
r
n
in
g
(
ML
)
co
n
s
id
er
in
g
u
n
m
an
n
ed
ae
r
ial
v
eh
i
cle
(
U
AV)
m
u
lti
-
s
p
ec
tr
al
im
ag
es,
an
in
v
esti
g
atio
n
was
co
n
d
u
c
ted
in
B
u
n
d
ab
er
g
,
Au
s
tr
alia.
Am
o
n
g
th
e
twen
ty
-
th
r
ee
id
en
tifie
d
in
d
icato
r
s
,
f
iv
e
s
p
ec
if
ic
in
d
ices,
n
o
tab
l
y
NDRE,
GR
NDVI
,
P
NDVI
,
C
C
C
I
,
an
d
W
DR
VI
,
d
em
o
n
s
tr
ated
a
n
o
ta
b
le
d
eg
r
ee
o
f
p
r
ec
is
io
n
in
th
e
p
r
ed
ictio
n
o
f
s
u
g
ar
ca
n
e
y
ield
.
Acc
o
r
d
in
g
to
a
n
in
v
esti
g
atio
n
in
[
1
3
]
,
m
etr
ics
o
b
tain
ed
f
r
o
m
s
tan
d
ar
d
ized
NDVI
tim
e
-
s
er
ies
co
llected
b
y
th
e
MO
DI
S
s
en
s
o
r
alo
n
g
with
an
en
s
em
b
le
ap
p
r
o
ac
h
o
f
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
AN
Ns)
wer
e
u
s
ed
f
o
r
p
r
ed
ictin
g
th
e
y
ield
o
f
s
u
g
ar
ca
n
e
in
São
Pau
lo
State,
B
r
az
il,
t
h
r
ee
m
o
n
th
s
p
r
io
r
to
h
a
r
v
esti
n
g
.
Pre
d
ictin
g
s
u
g
ar
ca
n
e
y
ield
at
th
e
r
eg
io
n
al
s
ca
le
(
d
is
tr
ict
-
wis
e)
in
Uttar
Pra
d
esh
,
I
n
d
ia
was
d
o
n
e
u
s
in
g
th
e
r
ea
d
ily
av
ailab
le
M
ODI
S
s
atel
lite
d
ata.
A
p
ilo
t
-
s
c
ale
r
esear
ch
was
co
n
d
u
cted
in
[
1
4
]
to
ev
alu
ate
th
e
cu
ltiv
atio
n
o
f
s
u
g
ar
ca
n
e
ac
r
o
s
s
f
o
u
r
s
u
g
a
r
p
r
o
ce
s
s
in
g
p
lan
ts
lo
ca
ted
in
th
e
two
I
n
d
ian
s
tates.
T
h
e
in
v
esti
g
atio
n
was
c
o
n
d
u
cted
t
h
r
o
u
g
h
o
u
t
th
e
p
er
i
o
d
b
e
twee
n
2
0
1
7
an
d
2
0
1
9
an
d
u
tili
ze
d
NDVI
,
wate
r
s
ca
lar
(
W
S),
an
d
lan
d
s
u
r
f
ac
e
wate
r
in
d
ex
(
L
SW
I
)
m
u
lti
-
d
ate
m
u
lti
-
s
p
ec
tr
al
in
f
o
r
m
atio
n
g
at
h
er
ed
f
r
o
m
L
I
SS
I
I
I
,
L
I
SS
I
V
o
f
R
eso
u
r
ce
s
at
-
2
an
d
2
A,
in
ad
d
itio
n
to
g
r
o
u
n
d
d
ata
an
d
GPS
in
f
o
r
m
atio
n
.
Ad
d
itio
n
ally
,
C
C
E
w
a
s
u
tili
ze
d
as
p
ar
t
o
f
th
e
ass
ess
m
en
t
p
r
o
ce
s
s
.
A
co
m
p
ar
ab
l
e
in
v
esti
g
atio
n
co
n
d
u
cted
s
im
ilar
to
[
1
4
]
,
was
p
r
esen
ted
in
[
1
5
]
w
h
er
e
s
im
il
ar
s
u
g
ar
p
r
o
ce
s
s
in
g
p
lan
t
s
in
t
wo
I
n
d
ian
s
tates
wer
e
co
n
s
id
e
r
ed
an
d
u
tili
ze
d
an
en
s
em
b
le
ap
p
r
o
ac
h
o
f
o
p
tic
al
in
f
o
r
m
a
tio
n
f
r
o
m
b
o
th
s
y
n
th
etic
a
p
er
tu
r
e
r
ad
a
r
(
SAR
)
an
d
Sen
tin
el
-
2
in
f
o
r
m
atio
n
co
llected
th
r
o
u
g
h
th
e
s
u
g
ar
ca
n
e
h
ar
v
est
cy
cle
’
s
m
ajo
r
s
tag
e
o
f
d
e
v
elo
p
m
e
n
t
f
o
r
y
ield
p
r
ed
ictio
n
.
A
n
u
m
b
er
o
f
f
ac
to
r
s
,
in
clu
d
in
g
a
co
m
b
in
atio
n
o
f
NDVI
,
NDRE,
E
VI
,
MT
VI
,
an
d
W
S,
ac
co
r
d
in
g
to
in
f
o
r
m
atio
n
g
ath
e
r
ed
b
y
Sen
tin
el
-
2
,
wer
e
u
s
ed
to
d
eter
m
i
n
e
th
e
cr
o
p
y
ield
.
On
th
e
c
o
n
t
r
ar
y
,
VV
a
n
d
VH
p
o
lar
iza
tio
n
s
,
in
ad
d
itio
n
to
t
h
e
cr
o
s
s
pol
r
atio
(
VV/VH)
,
co
n
s
is
ted
o
f
th
e
p
ar
am
eter
s
d
er
iv
ed
f
r
o
m
SAR
in
f
o
r
m
atio
n
th
at
wer
e
u
s
ed
t
o
d
r
iv
e
y
ield
m
o
d
elin
g
.
T
h
e
r
esear
ch
er
s
u
ltima
tely
em
p
l
o
y
ed
th
r
ee
f
ac
to
r
s
,
n
am
ely
E
VI
,
W
S,
an
d
VH,
at
d
if
f
er
en
t
s
tag
es
o
f
d
ev
el
o
p
m
en
t
in
o
r
d
er
to
o
b
tain
a
p
r
ec
is
e
es
tim
atio
n
o
f
s
u
g
ar
ca
n
e
y
ield
.
Facto
r
s
lik
e
less
cu
ltiv
atin
g
r
eg
io
n
,
a
h
y
b
r
i
d
cr
o
p
p
i
n
g
s
tr
u
ctu
r
e,
cr
o
p
v
ar
i
etal
s
h
if
ts
,
v
ar
iatio
n
in
th
e
d
ate
o
f
p
lan
tin
g
,
s
o
il
v
ar
ian
ts
,
an
d
ir
r
ig
atio
n
m
eth
o
d
s
co
n
tr
i
b
u
ted
to
th
e
d
if
f
icu
lty
o
f
ac
c
u
r
atel
y
p
r
ed
ictin
g
y
ield
s
at
a
s
p
atia
l
s
ca
le
.
ML
h
as
b
ee
n
wid
ely
r
ec
o
g
n
ize
d
f
o
r
its
p
o
ten
tial
to
ac
h
iev
e
h
ig
h
er
p
r
ed
ictin
g
ac
cu
r
ac
y
lev
els
i
n
ag
r
icu
ltu
r
al
y
ield
at
th
e
f
ield
lev
el
.
T
h
is
is
m
ai
n
ly
d
u
e
to
ML
’
s
ca
p
ab
ilit
y
o
f
ca
p
tu
r
in
g
th
e
c
o
m
p
licated
co
n
n
ec
tio
n
am
o
n
g
cr
o
p
y
ield
alo
n
g
s
id
e
th
e
v
a
r
io
u
s
f
ea
t
u
r
es
t
h
at
ar
e
u
tili
ze
d
f
o
r
y
ield
p
r
e
d
ictio
n
s
.
T
h
e
r
ef
o
r
e
,
t
h
e
u
tili
za
tio
n
o
f
s
atellite
im
ag
er
y
wh
en
co
m
b
in
ed
with
ML
ap
p
r
o
ac
h
en
h
an
ce
s
th
e
p
r
ec
is
io
n
o
f
s
u
g
ar
ca
n
e
y
ield
p
r
ed
ictio
n
s
.
Sev
er
al
ML
alg
o
r
ith
m
s
,
s
u
ch
as
m
u
ltip
le
lin
ea
r
r
eg
r
ess
io
n
(
ML
R
)
,
gr
ad
ien
t
b
o
o
s
tin
g
tr
ee
(
GB
R
)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(S
VM
)
,
r
an
d
o
m
f
o
r
est
(
R
F),
ANNs,
ex
tr
em
e
g
r
ad
ien
t
bo
o
s
tin
g
t
r
ee
(
XGB),
an
d
B
ay
esian
ap
p
r
o
ac
h
,
h
a
v
e
b
ee
n
wid
ely
em
p
lo
y
ed
f
o
r
p
r
ed
ictin
g
s
u
g
ar
ca
n
e
cr
o
p
y
ield
at
th
e
f
ield
t
o
r
eg
i
o
n
al
o
r
m
ill
-
lev
el
em
p
lo
y
in
g
ag
r
o
m
eteo
r
o
lo
g
ical
an
d
s
atellite
im
ag
er
y
[
1
5
]
-
[
1
8
]
.
On
e
o
f
th
e
m
o
s
t
im
p
o
r
ta
n
t
s
tep
s
in
d
ev
elo
p
in
g
an
y
k
in
d
o
f
c
r
o
p
y
ield
p
r
e
d
ictio
n
a
p
p
r
o
ac
h
is
ch
o
o
s
in
g
th
e
r
i
g
h
t
ML
m
o
d
el
an
d
in
co
r
p
o
r
atin
g
th
e
r
ig
h
t
f
e
atu
r
e
p
ar
am
eter
s
.
I
n
s
p
ite
o
f
a
k
n
o
wn
co
r
r
elatio
n
b
etwe
en
s
u
g
ar
ca
n
e
y
ield
a
n
d
d
if
f
e
r
en
t
v
eg
etatio
n
in
d
ices,
it
is
cr
u
cial
to
weig
h
t
h
eir
s
u
itab
ilit
y
f
o
r
th
e
in
ten
d
ed
a
p
p
licatio
n
at
a
g
iv
en
lo
ca
tio
n
an
d
g
r
o
wth
s
ta
g
es
o
f
s
u
g
ar
ca
n
e.
A
th
o
r
o
u
g
h
ex
am
i
n
atio
n
o
f
v
eg
etatio
n
i
n
d
ices
in
[
1
9
]
in
d
icate
d
t
h
at
v
ar
io
u
s
f
ac
to
r
s
ca
n
s
ig
n
if
ican
tly
im
p
ac
t
th
e
ac
c
u
r
ac
y
an
d
r
eliab
ilit
y
o
f
v
eg
etatio
n
in
d
ices.
T
h
ese
f
ac
to
r
s
in
clu
d
e
en
v
ir
o
n
m
en
tal
v
ar
iab
les,
s
en
s
o
r
m
ea
s
u
r
em
e
n
t
s
,
s
en
s
o
r
o
b
s
er
v
in
g
co
n
d
itio
n
s
,
s
u
n
’
s
lig
h
tin
g
g
eo
m
etr
y
,
s
o
il
m
o
is
tu
r
e
le
v
els,
b
r
ig
h
tn
ess
,
an
d
co
l
o
r
.
Xu
e
an
d
Su
[
2
0
]
,
ex
am
in
e
d
o
v
er
o
n
e
h
u
n
d
r
ed
v
e
g
etatio
n
in
d
ices.
T
h
ey
h
av
e
an
al
y
ze
d
th
eir
u
s
ef
u
ln
ess
in
r
elatio
n
to
th
e
s
p
ec
if
ic
v
eg
etatio
n
o
f
i
n
ter
est
an
d
th
e
s
u
r
r
o
u
n
d
in
g
e
n
v
ir
o
n
m
en
t.
W
h
ile
th
er
e
ar
e
n
u
m
er
o
u
s
in
d
ice
s
m
en
tio
n
ed
in
th
e
liter
atu
r
e,
o
n
l
y
a
f
ew
o
f
th
e
m
h
av
e
b
ee
n
ess
en
tially
u
tili
ze
d
o
r
ass
ess
ed
f
o
r
p
r
ed
ictin
g
s
u
g
ar
ca
n
e
y
ield
.
Acc
o
r
d
in
g
to
p
r
ev
i
o
u
s
r
esear
ch
,
th
e
NDVI
h
as
b
ee
n
id
en
tifie
d
as
th
e
m
o
s
t
co
m
m
o
n
ly
u
tili
ze
d
in
d
ex
f
o
r
p
r
ed
ictin
g
c
r
o
p
y
ield
[
2
1
]
-
[
2
6
]
.
T
h
e
u
s
e
o
f
R
S
an
d
ML
f
o
r
s
u
g
ar
c
an
e
y
ield
p
r
ed
icti
o
n
is
p
ick
in
g
th
e
m
o
m
en
tu
m
in
r
ec
en
t
y
ea
r
s
af
t
er
s
atellite
d
ata
p
r
o
v
i
d
er
s
lik
e
USGS
an
d
E
SA
ch
an
g
ed
th
e
ir
p
o
licies
to
m
ak
e
th
e
d
ata
f
r
ee
an
d
o
p
en
t
o
u
s
er
s
.
Hen
ce
it
r
em
ain
s
a
s
ite
-
s
p
ec
i
f
ic
o
p
en
r
esear
ch
p
r
o
b
lem
th
e
s
o
lu
tio
n
s
o
f
wh
ich
ca
n
lead
to
a
u
n
iv
er
s
al
m
o
d
el
g
r
a
d
u
ally
with
m
in
im
al
ch
an
g
es
to
its
o
p
er
atio
n
.
Fro
m
o
u
r
u
n
d
er
s
tan
d
in
g
o
f
r
ec
en
t
s
tu
d
ies
to
t
h
e
b
est
o
f
o
u
r
k
n
o
wled
g
e
,
a
s
et
o
f
v
eg
etat
io
n
in
d
ices
t
o
b
e
in
clu
d
e
d
f
o
r
y
ield
m
o
d
elin
g
a
n
d
a
d
etailed
ML
b
ased
d
ata
d
r
iv
en
ap
p
r
o
ac
h
to
m
o
d
el
s
u
g
ar
ca
n
e
cr
o
p
y
ield
at
f
ield
lev
el
was
u
n
k
n
o
wn
in
o
u
r
s
tu
d
y
ar
ea
.
T
h
er
e
is
n
o
t
a
s
p
ec
if
ic
v
eg
etatio
n
in
d
e
x
o
r
a
p
r
e
s
et
s
et
o
f
in
d
ices
th
at
is
u
n
iv
er
s
ally
ap
p
licab
le
to
s
u
g
ar
ca
n
e
cr
o
p
m
o
d
elin
g
.
T
h
u
s
,
th
e
ch
o
ice
o
f
o
n
e
v
eg
etatio
n
in
d
ex
o
v
er
th
e
o
th
er
,
f
o
r
an
a
p
p
licatio
n
,
is
q
u
ite
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:
2
5
0
2
-
4
7
52
F
ield
-
leve
l su
g
a
r
ca
n
e
yield
es
tima
tio
n
u
tili
z
in
g
S
en
tin
el
-
2
time
-
s
eries
a
n
d
ma
ch
in
e
lea
r
n
i
n
g
(
R
ek
h
a
B
.
U)
477
d
elica
te
to
m
ak
e.
T
h
er
ef
o
r
e,
th
e
p
r
esen
t
wo
r
k
aim
s
to
p
r
o
p
o
s
e
a
m
eth
o
d
o
lo
g
y
to
ev
alu
at
e
th
e
r
elatio
n
s
h
ip
o
f
g
r
o
u
n
d
d
ata
v
a
r
iab
les
an
d
s
p
ec
tr
al
v
eg
etatio
n
in
d
ices
d
er
i
v
ed
f
r
o
m
h
y
p
er
s
p
ec
tr
al
im
a
g
e
r
y
with
y
ield
at
th
e
f
ield
lev
el
an
d
d
ev
el
o
p
a
ML
b
ased
s
u
g
ar
ca
n
e
y
ield
p
r
e
d
ictio
n
m
o
d
el
f
o
r
a
tr
o
p
ical
r
eg
io
n
lik
e
I
n
d
ia
p
ar
ticu
lar
ly
in
o
u
r
s
tu
d
y
ar
ea
wh
ile
tar
g
etin
g
th
e
f
o
llo
win
g
o
b
jectiv
es:
−
I
d
en
tify
an
d
ev
alu
ate
th
e
s
en
s
itiv
ity
o
f
in
d
ices
ca
lcu
lated
f
r
o
m
n
ar
r
o
wb
an
d
an
d
r
ed
e
d
g
e
b
an
d
s
with
s
u
g
ar
ca
n
e
y
ield
.
−
Dete
r
m
in
e
th
e
o
p
tim
al
tim
e
f
o
r
s
u
g
ar
ca
n
e
y
ield
p
r
ed
ictio
n
.
−
I
d
en
tify
t
h
e
s
ig
n
if
ican
t v
e
g
etatio
n
in
d
ices th
at
b
est ex
p
lain
th
e
y
ield
at
th
e
f
ield
lev
el.
−
C
r
ea
te
an
d
ev
alu
ate
v
ar
i
o
u
s
ML
tech
n
iq
u
es
f
o
r
p
r
e
d
ict
in
g
s
u
g
ar
ca
n
e
y
ield
u
s
in
g
Sen
ti
n
el
-
2
d
ata
in
a
n
en
v
ir
o
n
m
en
t w
ith
s
p
ar
s
e
g
r
o
u
n
d
d
ata.
2.
M
AT
E
R
I
AL
S AN
D
M
E
T
H
O
DS
2
.
1
.
Study
a
re
a
T
h
e
in
v
esti
g
atio
n
was c
ar
r
ied
o
u
t in
th
e
o
p
er
atio
n
al
r
eg
i
o
n
o
f
Go
d
av
ar
i Bi
o
-
R
ef
in
er
ies L
td
.
(
GB
L
)
,
a
s
u
g
ar
m
ill
lo
ca
te
d
in
Sam
e
er
wad
i,
wh
ich
is
p
o
s
itio
n
ed
with
in
th
e
B
ag
alk
o
t
a
n
d
B
elag
av
i
d
is
tr
icts
o
f
Kar
n
atak
a,
I
n
d
ia.
T
h
e
g
e
o
g
r
ap
h
ical
co
o
r
d
in
ates
o
f
th
e
s
t
u
d
y
a
r
ea
ar
e
1
6
.
3
8
9
8
°N
an
d
7
5
.
0
3
7
1
°E.
T
h
e
r
esear
ch
r
e
g
io
n
is
d
ep
icted
in
Fig
u
r
e
1
.
T
h
e
r
eg
io
n
was
ch
ar
ac
ter
ized
b
y
a
h
u
m
id
s
u
b
tr
o
p
ical
en
v
ir
o
n
m
e
n
t,
with
th
e
m
o
n
s
o
o
n
p
e
r
io
d
f
r
o
m
J
u
n
e
t
o
Sep
tem
b
e
r
.
T
h
e
r
eg
io
n
was
d
r
y
a
n
d
s
p
ar
s
e
with
t
em
p
er
atu
r
es
r
an
g
in
g
f
r
o
m
1
6
.
2
0
°C
to
3
9
.
0
0
°C
.
T
h
e
m
ea
n
p
r
ec
i
p
itatio
n
in
B
elag
av
i
a
nd
B
ag
alk
o
t
was
5
4
5
m
illi
m
eter
s
(mm)
an
d
8
0
8
m
m
r
esp
ec
tiv
ely
.
Su
g
ar
ca
n
e
cu
ltiv
atio
n
in
th
e
s
tu
d
y
ar
ea
o
b
s
er
v
es
th
r
ee
p
lan
tin
g
s
ea
s
o
n
s
:
s
ea
s
o
n
al
(
b
etwe
en
J
an
-
Ma
r
an
d
J
an
-
Fe
b
)
,
p
r
e
-
s
ea
s
o
n
al
(
b
etwe
en
Oc
t
-
No
v
an
d
Dec
-
Feb
)
,
a
n
d
a
d
s
ali
(
b
etwe
en
J
u
ly
-
Au
g
an
d
Oct
-
Dec
)
.
T
h
e
c
o
m
m
o
n
s
u
g
a
r
ca
n
e
v
ar
ieties
g
r
o
wn
h
er
e
in
clu
d
e
C
O8
6
0
3
2
a
n
d
C
O9
1
0
1
0
.
Ap
ar
t
f
r
o
m
s
u
g
a
r
ca
n
e,
o
th
er
c
r
o
p
s
g
r
o
wn
in
clu
d
e
m
aize
,
tu
r
m
e
r
ic,
g
r
ee
n
leav
es,
s
ea
s
o
n
al
f
r
u
its
,
an
d
v
eg
etab
les.
Fig
u
r
e
1
.
L
o
ca
tio
n
m
a
p
o
f
th
e
s
tu
d
y
ar
ea
2
.
2
.
F
ield
d
a
t
a
A
GPS
s
u
r
v
ey
o
f
s
u
g
ar
ca
n
e
f
ield
s
r
eg
is
ter
ed
with
GB
L
f
o
r
cr
o
p
y
ield
m
o
d
elin
g
was
p
er
f
o
r
m
ed
b
etwe
en
J
an
u
ar
y
a
n
d
A
p
r
il
2
0
1
9
.
T
h
e
f
ield
b
o
u
n
d
a
r
y
g
eo
-
co
o
r
d
in
ates
wer
e
co
llected
u
s
in
g
a
h
an
d
h
eld
Gar
m
in
Mo
n
tan
a
6
8
0
GPS d
ev
ice.
T
h
e
GPS tr
ac
k
s
wer
e
d
ig
itized
in
to
KM
L
f
iles
u
s
in
g
G
o
o
g
le
E
ar
th
Pro
a
n
d
co
m
b
in
ed
in
QGI
S
to
cr
ea
te
a
s
in
g
le
-
s
h
ap
e
f
ile
r
ep
r
esen
tin
g
th
e
s
u
r
v
e
y
ed
f
a
r
m
s
in
2
D.
Du
r
in
g
d
i
g
itizatio
n
,
we
en
s
u
r
ed
to
ca
r
ef
u
lly
m
ap
t
h
e
f
ield
b
o
u
n
d
ar
y
co
o
r
d
in
ates
an
d
av
o
id
a
n
y
in
ter
s
ec
tio
n
o
f
r
esu
ltin
g
p
o
ly
g
o
n
s
.
A
r
an
d
o
m
s
u
b
s
et
o
f
f
ield
s
f
r
o
m
d
if
f
er
e
n
t
v
illag
es
in
th
e
s
tu
d
y
ar
ea
wer
e
c
h
o
s
en
f
o
r
th
e
s
u
r
v
ey
.
T
h
is
was
d
o
n
e
to
en
s
u
r
e
th
e
o
v
er
all
s
am
p
le
r
ef
lecte
d
th
e
ty
p
ical
g
r
o
win
g
c
o
n
d
itio
n
s
an
d
h
en
ce
,
d
iv
e
r
s
e
cr
o
p
v
i
g
o
r
ca
te
g
o
r
ies
in
o
u
r
s
am
p
lin
g
s
tr
ata.
W
h
ile
we
ac
k
n
o
wled
g
e
th
e
p
o
s
s
ib
ilit
y
o
f
ce
r
tain
s
o
u
r
ce
s
o
f
b
ias,
s
u
ch
as
v
a
r
iatio
n
in
f
ar
m
m
an
a
g
em
en
t
p
r
ac
tices,
s
o
il
f
er
tili
ty
,
o
r
wea
th
er
co
n
d
itio
n
s
,
th
o
u
g
h
we
m
ad
e
ef
f
o
r
ts
to
m
itig
ate
th
ese
f
ac
to
r
s
.
Ag
r
o
n
o
m
ic
d
ata
f
o
r
1
9
6
s
u
r
v
e
y
ed
s
u
g
a
r
ca
n
e
f
ield
s
,
in
clu
d
in
g
cr
o
p
v
ar
iety
,
ca
n
e
ty
p
e,
an
d
y
ield
f
o
r
th
e
s
u
r
v
ey
ed
f
iel
d
s
wer
e
o
b
tain
ed
f
r
o
m
th
e
GB
L
ag
r
o
d
atab
ase
f
o
r
th
r
ee
co
n
s
ec
u
tiv
e
g
r
o
win
g
s
ea
s
o
n
s
2
0
1
7
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
1
,
J
an
u
ar
y
20
25
:
47
5
-
48
7
478
2
0
1
8
,
2
0
1
8
-
2
0
1
9
,
an
d
2
0
1
9
-
2
0
2
0
.
T
h
e
f
ield
lev
el
s
u
g
a
r
ca
n
e
y
ield
m
o
d
elin
g
was
d
o
n
e
f
o
r
th
e
g
r
o
win
g
s
ea
s
o
n
2017
-
2
0
1
8
.
T
h
e
2
0
1
8
-
2
0
1
9
an
d
2
0
1
9
-
2
0
2
0
s
ea
s
o
n
d
ata
wer
e
u
s
ed
to
v
alid
ate
th
e
f
in
al
m
o
d
el.
2
.
3
.
Sa
t
ellit
e
im
a
g
er
y
da
t
a
T
h
e
Sen
tin
el
-
2
to
p
-
of
-
atm
o
s
p
h
er
e
l
ev
el
-
1
C
(
L
1
C
)
c
o
llectio
n
in
Go
o
g
le
ea
r
th
en
g
in
e
(
GE
E
)
is
u
s
ed
f
o
r
th
e
s
tu
d
y
.
T
h
e
Sen
tin
el
-
2
s
u
r
f
ac
e
r
ef
lecta
n
ce
L
ev
el
-
2
A
co
llectio
n
in
GE
E
d
o
es
n
o
t
e
x
ten
d
b
ac
k
t
o
2
0
1
7
.
So
we
u
s
ed
th
e
L
1
C
co
llecti
o
n
.
L
1
C
d
ata
is
also
r
e
p
o
r
te
d
to
p
r
o
v
id
e
g
o
o
d
r
esu
lts
with
o
u
t
th
e
n
ee
d
f
o
r
atm
o
s
p
h
er
ic
co
r
r
ec
tio
n
[
2
7
]
-
[
2
9
]
.
T
h
e
e
f
f
ec
t
o
f
clo
u
d
s
was r
ed
u
ce
d
in
two
lev
els.
Fir
s
t,
Sen
tin
el
-
2
tiles
with
a
clo
u
d
co
v
er
o
f
less
th
an
2
0
%
wer
e
f
ilter
ed
b
y
th
e
“c
lo
u
d
p
i
x
el
p
er
ce
n
tag
e”
(
C
PP
)
m
etad
ata
f
ield
.
Seco
n
d
,
at
th
e
p
ix
el
lev
el
clo
u
d
s
wer
e
m
ask
ed
u
s
in
g
th
e
“QA
6
0
”
q
u
ali
ty
b
an
d
[
2
7
]
.
T
h
e
2
0
m
s
p
atia
l
r
eso
lu
tio
n
s
o
f
th
e
r
ed
e
d
g
e
an
d
SW
I
R
b
an
d
s
wer
e
d
o
w
n
s
ca
led
to
1
0
m
u
s
in
g
b
ilin
ea
r
in
ter
p
o
latio
n
[
2
8
]
with
th
e
o
r
ig
in
a
l
p
r
o
jectio
n
m
ai
n
tain
ed
,
t
o
im
p
r
o
v
e
in
ter
p
r
etatio
n
an
d
ac
h
iev
e
co
m
m
o
n
d
if
f
e
r
en
tiab
ilit
y
.
2
.
4
.
Sp
ec
t
ra
l
v
eg
et
a
t
io
n ind
ices
3
1
v
e
g
etatio
n
i
n
d
ices
lis
ted
in
T
ab
le
1
,
s
p
ec
if
ic
to
cr
o
p
y
iel
d
wer
e
s
elec
ted
f
o
r
t
h
e
s
tu
d
y
.
I
n
T
ab
le
1
,
th
e
n
o
tatio
n
s
ar
e
as
f
o
llo
ws
N
-
NI
R
,
R
-
R
ed
,
G
-
Gr
ee
n
,
B
-
B
lu
e,
R
E
1
-
R
ed
E
d
g
e
1
,
R
E
2
-
R
ed
E
d
g
e
2
,
R
E
3
-
R
ed
E
d
g
e
3
b
an
d
,
N
2
-
R
ed
E
d
g
e
4
b
a
n
d
.
T
h
e
s
elec
tio
n
o
f
t
h
ese
in
d
ices
was
b
ased
o
n
th
e
k
n
o
wled
g
e
th
at
th
e
NI
R
,
Gr
ee
n
,
an
d
R
ed
s
p
ec
tr
al
b
an
d
s
p
lay
a
cr
u
cial
r
o
le
in
r
ep
r
esen
tin
g
th
e
d
e
n
s
e
ca
n
o
p
y
s
u
r
f
ac
e
s
u
ch
as
th
at
o
f
a
s
u
g
ar
ca
n
e
c
r
o
p
.
I
n
ad
d
itio
n
,
r
ec
en
t
s
tu
d
ies
f
o
r
r
ice
y
ield
esti
m
atio
n
f
o
u
n
d
n
ar
r
o
w
b
an
d
s
to
b
e
u
s
ef
u
l
in
im
p
r
o
v
in
g
t
h
e
y
ield
m
o
d
el
ac
cu
r
ac
y
[
2
9
]
.
T
o
ex
a
m
in
e
th
e
p
o
ten
tial
o
f
n
ar
r
o
w
b
an
d
s
in
s
u
g
ar
ca
n
e
y
ield
p
r
ed
ictio
n
,
a
s
et
o
f
i
n
d
ic
es b
ased
o
n
Nar
r
o
w
NI
R
an
d
r
ed
ed
g
e
b
an
d
s
was a
ls
o
in
clu
d
ed
.
T
ab
le
1
.
T
h
e
v
eg
etatio
n
in
d
ice
s
u
s
ed
in
th
is
s
tu
d
y
I
n
d
e
x
B
a
n
d
c
o
mb
i
n
a
t
i
o
n
R
e
f
e
r
e
n
c
e
R
e
d
e
d
g
e
b
a
n
d
g
r
e
e
n
n
e
ss
i
n
d
i
c
e
s
C
h
l
o
r
o
p
h
y
l
l
c
a
r
o
t
e
n
o
i
d
i
n
d
e
x
(
C
C
C
I
)
(N
-
R
E
1
)
*
(
N
+
R
)
/
(
N
+
R
E
1
)
*
(
N
-
R)
[
2
0
]
M
o
d
i
f
i
e
d
C
h
l
o
r
o
p
h
y
l
l
A
b
s
o
r
p
t
i
o
n
i
n
R
e
f
l
e
c
t
a
n
c
e
I
n
d
e
x
(
M
C
A
R
I
)
(
(
R
E1
-
R
)
-
0
.
2
*
(
R
E
1
-
G
)
)
*
(
R
E
1
/
R
)
[
2
0
]
M
o
d
i
f
i
e
d
r
e
d
e
d
g
e
N
D
V
I
(
M
R
EN
D
V
I
)
(
R
E1
-
N
)
/
(
R
E
1
+
N
-
(
2
*
B
)
)
[
3
0
]
N
o
r
mal
i
z
e
d
d
i
f
f
e
r
e
n
c
e
r
e
d
e
d
g
e
i
n
d
e
x
(
N
D
R
E)
(N
-
R
E
1
)
/
(
N
+
R
E
1
)
[
3
0
]
R
e
d
e
d
g
e
c
h
l
o
r
o
p
h
y
l
l
i
n
d
e
x
(
R
E
C
I
)
(
N
/
R
E
1
)
-
1
[
3
0
]
S
e
n
t
i
n
e
l
-
2
r
e
d
e
d
g
e
p
o
s
i
t
i
o
n
(
S
2
R
EP)
7
0
5
.
0
+
3
5
.
0
*
(
(
(
(
R
E3
+
R
)
/
2
.
0
)
-
R
E
1
)
/
(
R
E2
-
R
E1
)
)
[
3
1
]
A
t
mo
s
p
h
e
r
i
c
a
l
l
y
r
e
s
i
st
a
n
t
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
A
R
V
I
)
(
N
2
-
(
R
-
g
a
mm
a
*
(
R
-
B
)
)
)
/
(
N
2
+
(
R
-
g
a
mm
a
*
(
R
-
B
)
)
)
[
3
2
]
B
r
o
a
d
b
a
n
d
a
n
d
n
a
r
r
o
w
b
a
n
d
g
r
e
e
n
n
e
ss i
n
d
i
c
e
s
B
l
u
e
n
o
r
ma
l
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
B
N
D
V
I
)
(N
-
B
)
/
(
N
+
B
)
[
3
0
]
C
h
l
o
r
o
p
h
y
l
l
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
C
V
I
)
(
N
*
R
)
/
(
G
*
*
2
.
0
)
[
3
0
]
D
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
D
V
I
)
N
–
R
[
2
0
]
En
h
a
n
c
e
d
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
EV
I
)
g
*
(
N
-
R
)
/
(
N
+
C
1
*
R
-
C
2
*
B
+
L)
[
2
0
]
G
r
e
e
n
-
b
l
u
e
n
o
r
ma
l
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
G
B
N
D
V
I
)
(N
-
(
G
+
B
)
)
/
(
N
+
(
G
+
B
)
)
[
3
0
]
G
r
e
e
n
c
h
l
o
r
o
p
h
y
l
l
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
G
C
I
)
(
N
/
G
)
–
1
[
3
0
]
G
l
o
b
a
l
e
n
v
i
r
o
n
m
e
n
t
m
o
n
i
t
o
r
i
n
g
i
n
d
e
x
(
G
EM
I
)
(
(
2
.
0
*
(
(
N
*
*
2
.
0
)
-
(
R
*
*
2
.
0
)
)
+
1
.
5
*
N
+
0
.
5
*
R
)
/
(
N
+
R
+
0
.
5
)
)
*
(
1
.
0
-
0
.
2
5
*
(
(
2
.
0
*
(
(
N
*
*
2
.
0
)
-
(
R
*
*
2
)
)
+
1
.
5
*
N
+
0
.
5
*
R
)
/
(
N
+
R
+
0
.
5
)
)
)
-
(
(
R
-
0
.
1
2
5
)
/
(
1
-
R
)
)
[
2
0
]
G
r
e
e
n
l
e
a
f
i
n
d
e
x
(
G
LI
)
(
2
.
0
*
G
-
R
-
B
)
/
(
2
.
0
*
G
+
R
+
B
)
[
2
0
]
G
r
e
e
n
n
o
r
ma
l
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
G
N
D
V
I
)
(N
-
G
)
/
(
N
+
G
)
[
2
0
]
G
r
e
e
n
r
e
d
n
o
r
ma
l
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
G
R
N
D
V
I
)
(N
-
(
G
+
R
)
)
/
(
N
+
(
G
+
R
)
)
[
3
0
]
G
r
e
e
n
r
a
t
i
o
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
G
R
V
I
)
N
/
G
[
2
0
]
M
o
d
i
f
i
e
d
s
o
i
l
-
a
d
j
u
st
e
d
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
M
S
A
V
I
)
0
.
5
*
(
2
.
0
*
N
+
1
-
(
(
(
2
*
N
+
1
)
*
*
2
)
-
8
*
(
N
-
R
)
)
*
*
0
.
5
)
[
2
0
]
M
o
d
i
f
i
e
d
s
i
m
p
l
e
r
a
t
i
o
(
M
S
R
)
(
N
/
R
-
1
)
/
(
(
N
/
R
+
1
)
*
*
0
.
5
)
[
2
0
]
M
o
d
i
f
i
e
d
t
r
i
a
n
g
u
l
a
r
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
M
TV
I
)
1
.
2
*
(
1
.
2
*
(
N
-
G)
-
2
.
5
*
(
R
-
G))
[
2
0
]
N
o
r
mal
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
N
D
V
I
)
(N
-
R
)
/
(
N
+
R
)
[
2
0
]
N
o
r
mal
i
z
e
d
d
i
f
f
e
r
e
n
c
e
w
a
t
e
r
i
n
d
e
x
(
N
D
W
I
)
(G
-
N
)
/
(
G
+
N
)
[
2
0
]
O
p
t
i
mi
z
e
d
s
o
i
l
-
a
d
j
u
st
e
d
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
O
S
A
V
I
)
(N
-
R
)
/
(
N
+
R
+
0
.
1
6
)
[
2
0
]
P
a
n
N
D
V
I
(
P
N
D
V
I
)
(N
-
(
G
+
R
+
B
)
)
/
(
N
+
(
G
+
R
+
B
)
)
[
3
0
]
R
e
d
b
l
u
e
N
D
V
I
(
R
B
N
D
V
I
)
N
-
(
R
+
B
)
/
N
+
(
R
+
B
)
[
3
0
]
R
e
n
o
r
m
a
l
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
R
D
V
I
)
((N
–
R
)
/
(
N
+
R
)
)
1/
2
[
2
0
]
R
a
t
i
o
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
R
V
I
)
N
/
R
[
2
0
]
S
o
i
l
-
a
d
j
u
st
e
d
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
S
A
V
I
)
(
1
.
0
+
L)
*
(
N
-
R
)
/
(
N
+
R
+
L)
[
2
0
]
Tr
a
n
sf
o
r
me
d
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
TV
I
)
(((N
-
R
)
/
(
N
+
R
)
)
+
0
.
5
)
*
*
0
.
5
[
2
0
]
W
i
d
e
d
y
n
a
m
i
c
r
a
n
g
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
(
W
D
R
V
I
)
(
a
l
p
h
a
*
N
-
R
)
/
(
a
l
p
h
a
*
N
+
R
)
[
2
0
]
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:
2
5
0
2
-
4
7
52
F
ield
-
leve
l su
g
a
r
ca
n
e
yield
es
tima
tio
n
u
tili
z
in
g
S
en
tin
el
-
2
time
-
s
eries
a
n
d
ma
ch
in
e
lea
r
n
i
n
g
(
R
ek
h
a
B
.
U)
479
2
.
5
.
F
e
a
t
ure
e
ng
ineering
W
e
s
tar
ted
co
n
s
tr
u
ctin
g
a
tim
e
s
er
ies
f
o
r
all
th
e
s
e
lecte
d
3
1
v
eg
etatio
n
in
d
ices
u
s
in
g
th
e
eq
u
atio
n
s
d
escr
ib
ed
in
T
ab
le
1
f
r
o
m
J
an
u
ar
y
2
0
1
7
to
Ma
r
c
h
2
0
1
8
b
y
f
ilter
in
g
th
e
Sen
tin
el
-
2
L
1
C
co
llectio
n
a
n
d
ap
p
ly
in
g
a
clo
u
d
-
m
ask
in
g
f
u
n
ctio
n
in
GE
E
[
3
3
]
.
T
h
e
m
ask
ed
clo
u
d
y
p
i
x
els
wer
e
r
ep
lace
d
b
y
th
e
tem
p
o
r
al
g
ap
-
f
illi
n
g
tec
h
n
iq
u
e
with
lin
ea
r
in
ter
p
o
latio
n
[
3
4
]
.
Fo
ll
o
win
g
th
at,
with
th
e
z
o
n
al
s
tatis
t
ics
to
o
l,
we
ca
lcu
lated
th
e
m
ea
n
v
alu
e
o
f
all
p
ix
els
with
in
ea
ch
p
o
ly
g
o
n
f
o
r
e
v
er
y
s
p
ec
tr
al
in
d
ex
b
an
d
,
to
d
er
iv
e
th
e
r
esp
ec
tiv
e
v
eg
etatio
n
in
d
ex
f
o
r
a
f
ield
o
n
ea
ch
d
ay
an
im
ag
e
was
av
ailab
le
f
o
r
it.
S
in
ce
t
h
e
f
ield
s
ar
e
s
p
r
ea
d
g
eo
g
r
a
p
h
ically
,
th
ey
wer
e
ca
p
tu
r
ed
b
y
m
u
ltip
le
tiles
.
Hen
ce
,
all
th
e
ca
lcu
lated
v
eg
etati
o
n
in
d
ices
wer
e
n
o
t
av
ailab
le
at
t
h
e
s
am
e
tem
p
o
r
a
l
r
eso
lu
tio
n
f
o
r
ea
c
h
f
ield
.
T
o
h
ar
m
o
n
ize
th
is
d
ata
we
m
o
d
eled
th
e
tim
e
s
er
ies
in
d
ices
to
b
e
av
ailab
le
at
a
r
e
g
u
lar
in
ter
v
al
i.e
.
,
at
e
v
er
y
5
-
d
ay
tim
e
-
s
tep
.
W
e
u
s
ed
a
tim
e
win
d
o
w
s
ize
o
f
21
d
ay
s
to
lo
o
k
f
o
r
an
u
n
m
ask
ed
p
ix
el
in
th
e
tim
e
s
er
ies.
T
h
i
s
h
elp
s
in
v
is
u
alizin
g
m
o
r
e
g
r
a
d
u
al
ch
an
g
es
in
th
e
tem
p
o
r
al
p
r
o
f
iles
o
f
in
d
ices
c
o
m
p
ar
ed
to
th
e
o
r
ig
i
n
al
o
n
es.
T
h
e
ef
f
ec
t
o
f
in
te
r
p
o
latio
n
an
d
s
m
o
o
t
h
in
g
o
n
th
e
tem
p
o
r
al
p
r
o
f
ile
o
f
NDVI
ca
n
b
e
s
e
en
in
Fig
u
r
e
2
.
Fig
u
r
e
2
.
T
h
e
av
er
a
g
ed
5
-
d
a
y
NDVI
tem
p
o
r
al
p
r
o
f
ile
ac
r
o
s
s
all
f
ield
s
.
*
T
h
e
r
is
e
in
NDVI
b
etwe
en
Ma
y
-
Sep
co
u
ld
b
e
d
u
e
t
o
in
ter
cr
o
p
p
in
g
f
ar
m
in
g
p
r
ac
tices in
th
e
s
tu
d
y
ar
ea
2
.
6
.
O
pti
m
um
t
i
m
e
f
o
r
y
ield pre
dict
io
n
T
h
e
m
o
n
th
ly
av
er
a
g
e
v
al
u
es
o
f
esti
m
ated
tim
e
-
s
er
ies
v
eg
etat
io
n
in
d
ices
wer
e
u
tili
ze
d
.
Ho
wev
er
,
d
u
e
to
th
e
clo
u
d
p
ix
el
p
er
ce
n
ta
g
e
th
r
esh
o
ld
u
s
ed
t
o
f
ilter
clo
u
d
y
im
ag
es,
n
o
i
n
d
ices
wer
e
av
ai
lab
le
b
etwe
en
J
u
n
e
an
d
Sep
tem
b
er
2
0
1
7
,
ev
en
w
ith
in
ter
p
o
latio
n
.
T
h
e
co
r
r
elat
io
n
b
etwe
en
ea
ch
f
ea
tu
r
e
v
ar
i
ab
le
an
d
o
b
s
er
v
ed
s
u
g
ar
ca
n
e
y
iel
d
was
ev
alu
ate
d
u
s
in
g
Pear
s
o
n
’
s
co
r
r
elatio
n
co
ef
f
icien
t
[
1
]
.
T
h
e
co
r
r
elatio
n
co
ef
f
icie
n
ts
wer
e
co
m
p
u
ted
f
o
r
g
r
a
n
d
g
r
o
wth
(
1
2
0
-
2
7
0
d
ay
s
af
ter
s
o
win
g
)
an
d
m
at
u
r
ity
p
h
ase
(
2
7
0
-
3
6
0
d
ay
s
af
ter
s
o
win
g
)
ex
ce
p
t d
u
r
in
g
th
e
clo
u
d
y
p
e
r
io
d
.
2
.
7
.
F
e
a
t
ure
s
elec
t
io
n
A
to
tal
o
f
3
7
f
ea
tu
r
es
f
o
r
th
e
y
ield
m
o
d
el
wer
e
en
g
in
ee
r
e
d
f
r
o
m
th
e
c
o
m
b
in
atio
n
o
f
g
r
o
u
n
d
tr
u
th
(
6
ca
teg
o
r
ical
v
ar
iab
les
th
at
in
clu
d
e
ca
n
e
v
ar
iety
,
cr
o
p
ty
p
e
-
p
lan
tatio
n
,
r
ato
o
n
1
,
r
ato
o
n
2
,
r
ato
o
n
3
,
m
ix
o
f
r
ato
o
n
an
d
p
lan
tatio
n
)
an
d
tim
e
-
s
er
ies
r
em
o
te
s
en
s
in
g
d
ata
(
3
1
v
ar
iab
les)
f
o
r
ev
er
y
f
ield
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
an
y
ML
m
o
d
el
is
lar
g
ely
b
ased
o
n
th
e
f
ea
t
u
r
es
it
is
tr
ain
ed
o
n
.
T
h
er
e
f
o
r
e,
t
o
s
elec
t
o
n
ly
th
e
m
o
s
t
s
u
itab
le
o
r
s
ig
n
if
ican
t
f
ea
tu
r
es
f
o
r
p
r
ed
i
ct
in
g
s
u
g
ar
ca
n
e
y
ield
we
p
e
r
f
o
r
m
e
d
f
ea
tu
r
e
s
elec
tio
n
in
two
p
h
ases
u
s
in
g
Pear
s
o
n
’
s
co
r
r
elatio
n
co
e
f
f
ici
en
t
(
r
)
[
1
5
]
an
d
r
ec
u
r
s
iv
e
f
ea
tu
r
e
elim
in
atio
n
with
cr
o
s
s
-
v
alid
atio
n
(
R
FECV)
[
3
5
]
,
[
3
6
]
.
2
.
8
.
M
o
del e
v
a
lua
t
i
o
n a
nd
v
a
lid
a
t
io
n
T
h
e
s
u
g
ar
ca
n
e
y
ield
p
r
ed
ictio
n
m
o
d
els
wer
e
d
ev
elo
p
ed
u
s
in
g
th
e
y
ield
d
ata
f
o
r
1
6
0
f
iel
d
s
am
p
les.
As
p
ar
t
o
f
th
e
m
o
d
elin
g
s
tr
ateg
y
,
r
an
d
o
m
ly
7
0
%
d
ata
wer
e
s
elec
ted
f
o
r
m
o
d
el
tr
ain
in
g
an
d
r
est
3
0
%
d
ata
wer
e
h
eld
o
u
t
as
b
lin
d
d
ata
f
o
r
m
o
d
el
test
in
g
.
Sev
en
ML
alg
o
r
ith
m
s
n
am
ely
,
ML
R
,
R
id
g
e,
E
last
icn
et,
s
u
p
p
o
r
t
v
ec
to
r
r
eg
r
ess
io
n
(
SVR
)
,
R
F,
GB
R
,
an
d
XG
B
wer
e
f
itted
o
v
er
th
e
tr
ain
in
g
d
ata.
T
h
e
r
e
liab
ilit
y
o
f
th
e
f
in
al
m
o
d
el
was
v
alid
ated
u
s
in
g
th
e
2
0
1
8
-
2
0
1
9
an
d
2
0
1
9
-
2
0
2
0
g
r
o
win
g
s
ea
s
o
n
s
u
g
a
r
ca
n
e
y
ield
d
ata.
T
h
e
f
r
am
ewo
r
k
f
o
r
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
is
p
r
esen
ted
i
n
Fig
u
r
e
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
1
,
J
an
u
ar
y
20
25
:
47
5
-
48
7
480
Fig
u
r
e
3
.
Fra
m
ewo
r
k
f
o
r
th
e
s
tu
d
y
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
v
e
g
etatio
n
tim
e
s
er
ies
o
f
all
in
d
ices
v
ar
y
o
v
er
tim
e
as
a
f
u
n
ctio
n
o
f
th
e
s
tan
d
i
n
g
ca
n
e
’
s
d
ev
elo
p
m
e
n
tal
p
h
ases
.
T
h
is
te
m
p
o
r
al
v
ar
iatio
n
is
an
al
y
ze
d
t
o
id
en
tify
th
e
o
p
tim
u
m
tim
e
f
o
r
y
ield
p
r
e
d
ictio
n
u
s
in
g
R
S
d
ata
in
th
e
s
tu
d
y
.
T
h
e
co
r
r
elatio
n
b
etwe
en
ea
ch
f
ea
tu
r
e
v
ar
iab
le
an
d
o
b
s
er
v
e
d
s
u
g
ar
ca
n
e
y
ield
was
ev
alu
ated
u
s
in
g
Pear
s
o
n
’
s
co
r
r
elatio
n
co
ef
f
icien
t.
T
h
e
‘
r
’
v
a
lu
es
f
o
r
g
r
o
u
n
d
t
r
u
th
v
ar
ia
b
le
s
r
an
g
ed
b
etwe
en
-
0
.
3
0
t
o
0
.
18
,
i
n
d
icatin
g
a
n
eg
li
g
ib
le
co
r
r
elatio
n
with
y
ield
.
T
h
e
co
r
r
elatio
n
o
f
v
eg
etatio
n
in
d
ices
s
tar
ted
to
s
ee
an
in
cr
ea
s
in
g
tr
en
d
af
ter
Ma
y
,
attain
ed
a
p
ea
k
in
th
e
f
ir
s
t
f
o
r
tn
ig
h
t
o
f
No
v
em
b
er
,
an
d
th
en
d
ec
r
ea
s
ed
in
Decem
b
er
.
No
v
em
b
er
wh
er
e
t
h
e
h
ig
h
est co
r
r
elatio
n
was o
b
s
er
v
ed
f
o
r
th
e
m
ajo
r
ity
o
f
th
e
i
n
d
ices,
co
r
r
esp
o
n
d
s
to
th
e
m
o
n
th
wh
en
m
o
s
t
o
f
th
e
s
u
g
ar
ca
n
e
is
to
war
d
s
th
e
en
d
o
f
th
e
g
r
a
n
d
g
r
o
wth
p
h
ase
an
d
th
e
s
tar
t
o
f
th
e
m
atu
r
ity
p
h
ase,
i.e
.
,
o
n
l
y
two
m
o
n
th
s
b
ef
o
r
e
h
a
r
v
esti
n
g
[
3
7
]
.
T
h
is
tr
en
d
in
co
r
r
elatio
n
l
ea
d
u
s
to
id
en
tif
y
No
v
em
b
er
m
o
n
th
as
th
e
r
ig
h
t
tim
e
f
o
r
y
ield
p
r
ed
ictio
n
.
T
h
e
v
ar
iatio
n
in
‘
r
’
f
o
r
v
eg
etatio
n
in
d
ices
with
y
ield
is
s
h
o
wn
in
Fig
u
r
e
4
.
Am
o
n
g
all
o
p
tical
v
ar
iab
les,
MREND
VI
(
r
min
=
-
0
.
1
3
,
r
max
=0
.
1
7
,
r
nov
=
-
0
.
1
3
)
an
d
S2
R
E
P
(r
min
=
-
0
.
1
1
,
r
max
=0
.
2
1
,
r
nov
=0
.
2
1
)
in
d
icate
d
lo
w
c
o
r
r
elatio
n
with
y
ield
th
r
o
u
g
h
o
u
t
th
e
s
u
g
ar
ca
n
e
cr
o
p
cy
cle.
Hen
ce
,
th
ese
m
ay
n
o
t b
e
th
e
b
est in
d
icato
r
s
o
f
s
u
g
ar
ca
n
e
y
ie
ld
.
Fig
u
r
e
4
.
T
h
e
v
ar
iatio
n
in
co
r
r
elatio
n
co
ef
f
icien
t
f
o
r
ea
c
h
VI
with
tar
g
et
y
ield
b
etwe
en
J
an
u
ar
y
to
Dec
em
b
er
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:
2
5
0
2
-
4
7
52
F
ield
-
leve
l su
g
a
r
ca
n
e
yield
es
tima
tio
n
u
tili
z
in
g
S
en
tin
el
-
2
time
-
s
eries
a
n
d
ma
ch
in
e
lea
r
n
i
n
g
(
R
ek
h
a
B
.
U)
481
Sin
ce
th
e
co
r
r
elatio
n
co
ef
f
icie
n
t
q
u
an
tifie
d
th
e
r
elatio
n
s
h
i
p
b
etwe
en
th
e
f
ea
tu
r
es
an
d
y
ield
,
f
ea
tu
r
es
with
r
≥
±
0
.
5
wer
e
s
elec
ted
f
o
r
f
u
r
th
er
a
n
aly
s
is
.
T
h
is
r
esu
lted
in
2
9
o
u
t
o
f
3
7
f
ea
tu
r
es
we
s
tar
ted
with
.
I
n
ter
esti
n
g
ly
,
all
2
9
f
ea
tu
r
es
wer
e
th
e
s
p
ec
tr
al
v
eg
etatio
n
in
d
ices
h
av
in
g
p
-
v
alu
e<
0
.
0
5
,
in
d
icatin
g
th
eir
s
tatis
t
ical
s
ig
n
i
f
ican
ce
with
y
i
eld
.
Desp
ite
h
av
i
n
g
a
s
tr
o
n
g
ass
o
ciatio
n
with
s
u
g
ar
ca
n
e
y
i
eld
,
th
e
2
9
s
p
ec
tr
al
v
eg
etatio
n
in
d
ices
th
at
wer
e
ch
o
s
en
h
a
d
m
o
d
er
ate
t
o
h
ig
h
co
r
r
elatio
n
s
with
o
n
e
an
o
t
h
er
,
ca
u
s
in
g
f
ea
tu
r
e
r
ed
u
n
d
an
cy
.
T
o
m
in
im
ize
f
e
atu
r
e
r
ed
u
n
d
an
c
y
in
th
e
m
o
d
el,
w
e
d
ec
id
ed
to
r
etain
o
n
ly
th
o
s
e
v
eg
etatio
n
in
d
ices
th
at
wer
e
o
f
th
e
h
ig
h
es
t
im
p
o
r
tan
ce
in
p
r
ed
ictin
g
s
u
g
ar
ca
n
e
y
ield
u
s
in
g
R
FECV.
T
h
e
ex
ac
t
n
u
m
b
er
o
f
f
ea
tu
r
es
th
at
ar
e
im
p
o
r
ta
n
t
f
o
r
y
ield
m
o
d
elin
g
is
u
n
k
n
o
wn
i
n
ad
v
a
n
ce
.
Fo
r
th
is
r
ea
s
o
n
,
w
e
u
s
ed
R
FECV
o
v
er
t
r
ad
itio
n
al
R
FE
m
eth
o
d
.
R
FECV
au
to
m
atica
lly
d
ete
r
m
in
es
th
e
o
p
tim
al
n
u
m
b
e
r
o
f
f
e
atu
r
es
u
s
in
g
cr
o
s
s
-
v
alid
atio
n
with
R
FE.
T
h
is
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
’
s
s
en
s
itiv
ity
to
esti
m
ato
r
c
h
o
ice
is
its
lim
itatio
n
.
I
n
o
th
e
r
wo
r
d
s
,
a
s
u
b
s
et
o
f
f
ea
tu
r
es
s
elec
ted
b
y
o
n
e
es
tim
ato
r
u
s
ed
with
R
FECV
alg
o
r
ith
m
m
a
y
o
r
m
ig
h
t
n
o
t
b
e
s
i
g
n
i
f
i
c
a
n
t
t
o
a
n
o
t
h
e
r
e
s
t
i
m
ato
r
.
T
o
m
i
t
i
g
a
t
e
t
h
is
li
m
i
t
at
i
o
n
,
w
e
e
v
a
l
u
a
t
e
d
t
h
e
R
FE
C
V
w
i
t
h
4
d
i
f
f
e
r
e
n
t
e
s
t
i
m
a
t
o
r
s
.
T
h
e
R
F
E
C
V
r
e
s
u
l
t
s
f
o
r
v
a
r
i
o
u
s
e
s
ti
m
a
t
o
r
s
a
r
e
p
r
e
s
e
n
t
e
d
i
n
F
i
g
u
r
e
5
,
i
n
cl
u
d
i
n
g
F
ig
u
r
e
s
5
(a
)
t
o
5
(
d
)
.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
5
.
Ap
p
ly
in
g
:
(
a)
R
id
g
e
-
b
ased
R
FECV
,
(
b
)
SVR
-
b
ased
R
FECV
,
(
c)
R
F
R
-
b
ased
R
FE
C
V
,
an
d
(
d
)
XGBo
o
s
t
r
eg
r
ess
o
r
-
b
ased
R
FEC
V
to
th
e
in
itial f
ea
tu
r
e
s
p
ac
e
As
ex
p
ec
ted
,
th
e
o
p
tim
u
m
f
e
atu
r
es
o
b
tain
ed
d
u
r
in
g
th
e
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
v
ar
ied
d
ep
en
d
i
n
g
o
n
th
e
esti
m
ato
r
u
s
ed
,
with
R
FECV.
T
h
e
f
ea
tu
r
e
s
u
b
s
et
ch
o
s
en
b
y
ea
ch
R
FECV
-
esti
m
ato
r
(
R
id
g
e,
SVR
,
R
F,
an
d
XGB)
co
m
b
in
atio
n
h
ad
n
o
tab
le
d
if
f
er
e
n
ce
s
.
R
FECV
-
R
id
g
e
an
d
R
FECV
-
R
F
ch
o
s
e
f
ewe
r
f
ea
tu
r
es
with
co
m
p
ar
ativ
ely
b
etter
R
MSE
d
u
e
to
th
eir
a
b
ilit
y
to
r
ed
u
ce
r
ed
u
n
d
an
c
y
an
d
m
o
d
el
c
o
m
p
lex
ity
.
Du
r
i
n
g
o
u
r
ex
p
er
im
en
tatio
n
with
d
i
f
f
er
en
t
esti
m
ato
r
s
,
we
ev
alu
ated
th
e
f
ea
tu
r
e
im
p
o
r
ta
n
ce
co
m
p
u
t
ed
b
y
th
e
R
FECV
m
eth
o
d
s
f
o
r
th
e
s
elec
ted
f
ea
t
u
r
es.
I
t
ca
n
b
e
n
o
ticed
,
th
at
N
DR
E
an
d
R
VI
ar
e
d
ee
m
ed
as
s
ig
n
if
ican
t
b
y
all
4
R
FEC
V
esti
m
ato
r
s
w
ith
co
n
s
is
ten
t
h
ig
h
f
ea
tu
r
e
im
p
o
r
tan
c
e
v
alu
e.
An
d
R
E
C
I
was
ch
o
s
en
b
y
3
o
u
t
o
f
4
esti
m
ato
r
s
test
ed
.
T
h
e
o
th
er
in
d
ices
wer
e
s
h
o
r
tlis
ted
eith
er
o
n
ce
o
r
twice
o
n
ly
.
T
h
u
s
,
th
ese
3
in
d
ices
p
lay
a
p
r
o
m
in
e
n
t
r
o
le
i
n
p
r
e
d
ictin
g
s
u
g
ar
ca
n
e
y
ield
.
T
h
e
o
p
tim
u
m
f
ea
tu
r
es
s
elec
ted
b
y
ea
c
h
R
FECV
esti
m
ato
r
alo
n
g
with
th
e
im
p
o
r
tan
ce
v
alu
es o
f
th
e
f
ea
tu
r
es a
r
e
p
r
esen
ted
in
F
ig
u
r
e
6
,
in
clu
d
in
g
Fig
u
r
e
6
(
a)
to
6
(
d
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
1
,
J
an
u
ar
y
20
25
:
47
5
-
48
7
482
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
6
.
Featu
r
e
im
p
o
r
ta
n
ce
o
f
p
r
e
d
icto
r
s
esti
m
ated
b
y
th
e
alg
o
r
ith
m
s
;
(
a)
R
FECV
-
R
id
g
e
,
(
b
)
R
FECV
-
SVR
,
(
c)
R
FECV
-
RF
,
an
d
(
d
)
R
FEC
V
-
XGB
T
h
o
u
g
h
,
th
e
c
h
o
ice
o
f
3
v
ar
ia
b
les
b
y
R
FECV
-
R
id
g
e,
s
ee
m
ed
an
id
ea
l
f
ea
tu
r
e
s
u
b
s
et,
c
h
o
o
s
in
g
t
h
e
b
est
esti
m
ato
r
f
o
r
R
FECV
ca
n
d
ep
en
d
o
n
v
ar
io
u
s
f
ac
to
r
s
,
i
n
clu
d
in
g
th
e
p
er
f
o
r
m
an
ce
m
etr
ics,
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
,
an
d
th
e
s
p
ec
if
ic
c
h
a
r
ac
ter
is
tics
o
f
o
u
r
d
ata.
T
o
eli
m
in
ate
th
e
p
o
s
s
ib
le
tr
ad
e
-
o
f
f
b
etwe
en
th
e
n
u
m
b
er
o
f
o
p
tim
u
m
f
ea
tu
r
es
an
d
m
o
d
el
p
er
f
o
r
m
an
ce
,
all
s
ev
en
ML
m
o
d
els
ch
o
s
en
f
o
r
th
is
s
tu
d
y
wer
e
tr
ain
ed
u
s
in
g
th
e
o
p
tim
al
f
ea
tu
r
es
f
r
o
m
b
o
t
h
R
FECV
-
R
id
g
e
an
d
R
FECV
-
R
F
m
eth
o
d
s
.
Hy
p
er
p
ar
am
ete
r
o
p
tim
izatio
n
an
d
10
-
f
o
l
d
c
r
o
s
s
-
v
alid
atio
n
wer
e
also
em
p
lo
y
e
d
t
o
m
itig
ate
th
e
ch
an
ce
s
o
f
u
n
d
er
f
itti
n
g
o
r
o
v
er
f
itti
n
g
th
e
m
o
d
el.
T
h
is
m
eth
o
d
o
lo
g
y
r
esu
lted
in
f
o
u
r
teen
s
u
g
a
r
ca
n
e
y
ield
m
o
d
els.
B
ased
o
n
th
e
R
M
S
E
test
s
co
r
e,
th
e
to
p
3
m
o
d
els
ar
e
R
F
tr
ain
ed
with
R
FECV
-
R
id
g
e
f
ea
tu
r
e
s
u
b
s
et
(
R
2
=0
.
7
3
1
)
,
R
F
tr
ain
e
d
with
R
FECV
-
R
F
f
ea
tu
r
e
s
u
b
s
et
(
R
2
=0
.
7
1
3
)
,
a
n
d
R
id
g
e
tr
ain
ed
with
R
FE
C
V
-
R
F
f
ea
tu
r
e
s
u
b
s
et
(
R
2
=0
.
6
9
8
)
.
T
h
e
ac
cu
r
ac
y
m
etr
ics
o
f
all
s
ev
en
ML
m
o
d
els
co
n
s
id
er
ed
f
o
r
y
ield
m
o
d
elin
g
ar
e
tab
u
lated
b
elo
w
i
n
T
ab
le
2
.
A
K
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
p
r
o
ce
d
u
r
e
w
ith
k
=1
0
was
u
s
ed
to
s
elec
t
o
n
e
b
est
m
o
d
el
f
r
o
m
th
e
to
p
th
r
ee
p
e
r
f
o
r
m
in
g
m
o
d
els s
tated
ab
o
v
e.
T
ab
le
2
.
T
h
e
p
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
ML
m
o
d
els o
n
tr
ain
in
g
an
d
test
in
g
d
atasets
M
o
d
e
l
R
F
EC
V
e
st
i
ma
t
o
r
:
R
i
d
g
e
R
F
EC
V
e
st
i
ma
t
o
r
:
RF
M
A
E
(
t
/
h
a
)
R
M
S
E
(
t
/
h
a
)
M
A
E
(
t
/
h
a
)
R
M
S
E
(
t
/
h
a
)
Tr
a
i
n
Te
st
Tr
a
i
n
Te
st
Tr
a
i
n
Te
st
Tr
a
i
n
Te
st
LR
1
1
.
2
5
0
1
0
.
8
5
5
1
3
.
7
8
9
1
4
.
4
3
3
1
1
.
3
4
8
1
1
.
7
0
4
1
4
.
5
2
5
1
5
.
1
6
9
R
i
d
g
e
1
0
.
8
2
9
1
0
.
9
3
4
1
3
.
5
6
0
1
4
.
5
0
3
1
1
.
0
1
9
1
0
.
7
5
1
1
3
.
8
5
1
1
4
.
2
8
3
El
a
s
t
i
c
n
e
t
1
5
.
4
5
9
1
8
.
3
8
5
1
8
.
5
2
8
2
1
.
5
4
7
1
4
.
0
6
2
1
5
.
8
1
8
1
7
.
0
2
4
1
9
.
1
3
4
S
V
R
1
0
.
8
1
1
1
0
.
8
8
9
1
3
.
4
9
7
1
4
.
3
2
0
1
1
.
1
8
6
1
0
.
9
8
1
1
3
.
9
3
1
1
4
.
4
6
3
RF
1
1
.
3
0
6
9
.
7
3
4
1
4
.
0
8
8
1
3
.
4
4
8
1
1
.
1
5
9
1
0
.
0
4
6
1
4
.
0
6
1
1
3
.
6
4
4
G
B
R
1
1
.
6
1
9
1
0
.
4
0
9
1
4
.
4
0
1
1
4
.
3
2
8
1
1
.
6
9
6
1
0
.
4
8
1
1
4
.
2
7
5
1
3
.
9
9
9
XGB
1
1
.
7
0
3
1
1
.
7
3
5
1
4
.
2
9
7
1
5
.
3
5
3
1
1
.
8
5
5
1
1
.
7
4
3
1
4
.
5
9
1
1
5
.
2
6
6
T
h
e
R
F
m
o
d
el
tr
ain
ed
with
f
ea
tu
r
es
s
elec
ted
b
y
R
FECV
-
R
id
g
e
em
er
g
ed
as
th
e
b
est
-
f
itted
m
o
d
el
with
an
R
MSE
=1
3
.
7
9
t/h
a
d
u
r
in
g
th
e
m
o
d
el
s
elec
tio
n
p
r
o
c
ed
u
r
e.
I
n
a
d
d
itio
n
,
an
R
2
=0
.
7
3
1
r
ef
lecte
d
th
at
th
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:
2
5
0
2
-
4
7
52
F
ield
-
leve
l su
g
a
r
ca
n
e
yield
es
tima
tio
n
u
tili
z
in
g
S
en
tin
el
-
2
time
-
s
eries
a
n
d
ma
ch
in
e
lea
r
n
i
n
g
(
R
ek
h
a
B
.
U)
483
m
o
d
el
was
ab
le
to
ex
p
lain
th
e
v
ar
ian
ce
in
y
ield
r
ea
s
o
n
ab
ly
well.
T
h
e
s
u
g
ar
ca
n
e
y
ield
p
r
e
d
ictio
n
b
ased
o
n
th
e
RF
m
o
d
el
h
as
h
ig
h
er
ac
cu
r
ac
y
th
an
o
th
er
s
.
T
h
is
co
u
l
d
b
e
a
ttrib
u
ted
to
th
e
R
F
’
s
ab
ilit
y
t
o
h
an
d
le
n
o
n
-
lin
ea
r
an
d
d
y
n
am
ic
in
p
u
t
d
ata,
s
u
ch
as
en
v
ir
o
n
m
e
n
tal
im
p
ac
ts
,
an
d
cr
o
p
p
in
g
s
y
s
tem
s
.
T
h
e
R
F
m
o
d
el
tr
ain
ed
with
R
FEC
V
-
R
F
f
ea
tu
r
es
s
u
b
s
et
was
th
e
s
ec
o
n
d
-
b
est
m
o
d
el
with
a
clo
s
er
R
MSE
s
co
r
e.
T
h
is
co
m
b
in
atio
n
is
co
m
p
u
tatio
n
ally
ex
p
e
n
s
iv
e,
p
ar
ticu
lar
ly
f
o
r
lar
g
e
s
am
p
les
at
th
e
p
r
o
d
u
ctio
n
lev
el.
I
n
o
u
r
an
aly
s
is
[
3
8
]
we
ac
h
iev
ed
an
R
2
=0
.
7
1
with
GN
DVI
alo
n
e
as
th
e
y
ield
p
r
ed
ic
to
r
u
s
in
g
Po
l
y
n
o
m
ial
r
eg
r
ess
io
n
.
Ho
wev
e
r
,
f
r
o
m
o
u
r
cu
r
r
e
n
t
m
eth
o
d
o
lo
g
y
,
we
s
ee
th
at
a
co
m
b
in
atio
n
o
f
in
d
ices
is
g
iv
in
g
u
s
an
in
cr
ea
s
ed
R
2
=0
.
7
3
1
,
r
ef
lectin
g
th
e
f
ac
t
t
h
at
a
co
m
b
in
atio
n
o
f
s
elec
ted
f
ea
tu
r
es
ar
e
ab
le
to
i
m
p
r
o
v
e
th
e
y
ield
ac
cu
r
ac
y
.
I
n
ad
d
itio
n
,
th
e
c
u
r
r
e
n
t
s
tu
d
y
,
p
er
f
o
r
m
e
d
m
o
d
el
v
alid
atio
n
o
n
2
0
1
8
-
2
0
1
9
an
d
2
0
1
9
-
2
0
2
0
g
r
o
win
g
s
ea
s
o
n
d
ata.
T
h
e
b
est
m
o
d
el
i.e
.
,
RF
,
u
s
ed
ND
R
E
,
R
E
C
I
,
an
d
R
VI
as th
e
p
r
ed
icto
r
v
ar
iab
les
as sh
o
wn
in
Fig
u
r
e
7
.
T
h
e
m
o
d
el
ex
h
ib
its
a
d
e
ce
n
t
f
it
o
n
tr
ain
in
g
d
ata
as
s
h
o
wn
in
Fig
u
r
e
7
(
a)
.
Fro
m
th
e
r
esid
u
al
p
lo
t
in
Fig
u
r
e
7
(
b
)
,
we
s
ee
a
f
air
ly
r
an
d
o
m
an
d
u
n
if
o
r
m
d
is
tr
ib
u
tio
n
o
f
th
e
r
e
s
id
u
als
ag
ain
s
t
th
e
tar
g
et
y
ield
.
Fro
m
th
e
h
is
to
g
r
am
,
we
ca
n
also
s
ee
th
at
th
e
er
r
o
r
is
n
o
r
m
ally
d
is
tr
ib
u
te
d
ar
o
u
n
d
z
er
o
,
w
h
ich
g
e
n
er
al
ly
in
d
icate
s
a
well
-
f
itted
m
o
d
el.
(
a)
(
b
)
Fig
u
r
e
7
.
Plo
t o
f
(
a)
a
p
r
e
d
ictio
n
er
r
o
r
with
ac
tu
al
y
ield
p
lo
tt
ed
ag
ain
s
t p
r
e
d
icted
y
ield
v
alu
es b
y
m
o
d
el
with
b
est
-
f
it lin
e
an
d
(
b
)
r
esid
u
al
s
h
o
win
g
th
e
d
is
tr
ib
u
tio
n
o
f
th
e
d
if
f
er
en
ce
b
etwe
en
th
e
ac
tu
al
y
ield
an
d
th
e
p
r
ed
icted
y
ield
I
n
o
u
r
s
tu
d
y
,
th
e
ch
o
ice
o
f
in
d
ices
-
NDR
E
,
R
E
C
I
,
an
d
R
VI
f
o
r
s
u
g
ar
ca
n
e
y
ield
p
r
ed
ic
tio
n
in
th
e
s
tu
d
y
ar
ea
is
co
m
p
letely
d
ata
-
d
r
iv
en
.
Fro
m
all
th
e
e
x
p
er
im
e
n
tatio
n
s
,
we
co
u
ld
n
o
te
th
at
in
d
ices
d
er
iv
ed
u
s
in
g
r
ed
e
d
g
e
b
an
d
s
ar
e
f
av
o
r
ed
c
o
m
p
ar
ed
to
o
th
e
r
in
d
ices.
He
n
ce
th
e
in
tr
o
d
u
ctio
n
o
f
in
d
ices
in
th
e
r
ed
ed
g
e
b
an
d
s
ca
n
p
la
y
a
s
ig
n
if
ican
t
r
o
le
in
im
p
r
o
v
in
g
th
e
s
u
g
ar
ca
n
e
y
ield
p
r
ed
ictio
n
m
o
d
el.
W
h
ile
NDRE
an
d
R
VI
ar
e
a
c
o
m
b
in
atio
n
o
f
t
h
e
NI
R
b
an
d
an
d
r
ed
ed
g
e
r
a
n
g
e
b
etw
ee
n
v
is
ib
le
R
ed
an
d
NI
R
,
R
VI
u
s
es
th
e
NI
R
a
n
d
R
ed
b
an
d
o
f
Sen
tin
el
-
2
.
Giv
e
n
ag
r
icu
ltu
r
al
m
o
n
ito
r
in
g
,
th
e
r
o
les
o
f
NDVI
an
d
ND
R
E
a
r
e
co
m
p
lem
en
tar
y
.
NDRE
o
f
f
er
s
ad
v
an
tag
es
in
c
er
tain
s
itu
atio
n
s
,
p
ar
ticu
lar
ly
f
o
r
d
ee
p
e
r
ca
n
o
p
y
m
ea
s
u
r
em
e
n
ts
an
d
ar
ea
s
wh
er
e
NDVI
m
ay
r
ea
ch
s
atu
r
atio
n
[
3
9
]
,
[
4
0
]
.
Su
g
ar
ca
n
e
is
also
a
p
er
en
n
ial
cr
o
p
.
T
h
is
m
a
k
es
NDRE
p
ar
ticu
lar
ly
u
s
ef
u
l f
o
r
d
ee
p
er
ca
n
o
p
y
ass
ess
m
en
t o
f
s
u
g
ar
ca
n
e
cr
o
p
s
at
m
atu
r
ity
.
Als
o
at
p
ix
el
lev
el,
wh
en
NDVI
s
atu
r
ates
an
d
d
etec
ts
a
co
n
s
is
ten
t
+1
,
NDRE
ca
n
o
cc
asio
n
ally
o
f
f
er
a
m
o
r
e
ac
cu
r
ate
ass
ess
m
en
t
o
f
v
ar
iab
ilit
y
in
th
o
s
e
r
eg
io
n
s
[
2
9
]
,
[
4
1
]
.
A
t
m
atu
r
it
y
,
s
u
g
ar
ca
n
e
is
ass
o
ciate
d
wit
h
h
ig
h
ca
n
o
p
y
d
en
s
ity
m
a
k
in
g
R
VI
a
v
iab
le
in
d
ex
f
o
r
th
is
s
tu
d
y
[
2
0
]
.
I
n
No
v
em
b
er
,
wh
ile
s
o
m
e
ca
n
e
h
as
r
ea
c
h
ed
a
s
tag
e
o
f
m
atu
r
ity
,
a
ce
r
tain
p
o
r
tio
n
o
f
it
is
s
till
in
th
e
m
id
-
late
m
atu
r
ity
s
t
ag
e
d
u
e
to
d
if
f
er
en
ce
s
in
th
e
d
ate
o
f
p
lan
tin
g
.
I
n
s
u
ch
a
s
ce
n
ar
io
,
R
E
C
I
h
elp
s
in
ass
es
s
in
g
th
e
cr
o
p
at
a
s
tag
e
o
f
ac
tiv
e
v
eg
etatio
n
d
ev
el
o
p
m
e
n
t
[
3
]
.
T
h
e
f
i
n
al
m
o
d
el
is
v
alid
ated
f
o
r
2
0
f
ield
s
an
d
1
6
f
ield
s
f
o
r
th
e
g
r
o
win
g
s
ea
s
o
n
2
0
1
8
-
2
0
1
9
a
n
d
2
0
1
9
-
2
0
2
0
r
esp
ec
tiv
ely
.
T
h
e
d
ata
o
f
th
ese
f
ield
s
was
n
o
t
ex
p
o
s
ed
to
t
h
e
m
o
d
el
d
u
r
in
g
t
h
e
tr
ain
in
g
p
r
o
ce
s
s
.
T
h
e
v
alid
atio
n
r
esu
lts
ar
e
p
r
esen
ted
i
n
T
ab
le
3
.
T
ab
le
3
.
T
h
e
p
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
f
in
al
ML
m
o
d
el
o
n
v
alid
atio
n
d
atasets
M
o
d
e
l
2
0
1
8
–
2
0
1
9
2
0
1
9
–
2
0
2
0
M
A
E
(
t
/
h
a
)
R
M
S
E
(
t
/
h
a
)
M
A
E
(
t
/
h
a
)
R
M
S
E
(
t
/
h
a
)
R
a
n
d
o
mF
o
r
e
st
R
e
g
r
e
ss
o
r
(
n
_
e
st
i
ma
t
o
r
s=1
0
0
,
m
a
x
_
d
e
p
t
h
=
2
,
mi
n
_
sa
mp
l
e
s
_
l
e
a
f
=
2
,
r
a
n
d
o
m
_
st
a
t
e
=
4
2
)
1
0
.
0
6
5
1
2
.
5
3
1
1
1
.
6
3
4
1
3
.
5
3
9
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
1
,
J
an
u
ar
y
20
25
:
47
5
-
48
7
484
W
e
also
p
er
f
o
r
m
ed
an
e
x
ten
s
iv
e
f
ea
tu
r
e
s
elec
tio
n
o
n
th
e
f
in
al
R
F
m
o
d
el.
Her
e,
all
p
o
s
s
ib
le
f
ea
tu
r
e
co
m
b
in
atio
n
s
we
r
e
test
ed
wit
h
o
u
t
elim
in
atin
g
a
n
y
f
ea
tu
r
e.
W
e
co
u
ld
s
ee
th
at
th
e
c
o
m
b
in
atio
n
o
f
NDRE
an
d
R
E
C
I
h
ad
R
MSE
=1
4
.
0
0
t/h
a
with
a
s
tan
d
ar
d
d
e
v
iatio
n
o
f
3
.
0
0
t/h
a
c
o
m
p
a
r
ed
to
R
MSE
=1
3
.
4
9
t/h
a
an
d
a
s
tan
d
ar
d
d
ev
iatio
n
o
f
3
.
1
5
t/h
a
f
o
r
NDRE
an
d
R
VI
.
T
h
is
s
h
o
ws
th
at
a
co
m
b
i
n
atio
n
o
f
two
o
r
m
o
r
e
i
n
d
ices
as
y
ield
p
r
ed
icto
r
s
is
ef
f
ec
tiv
e
in
s
u
g
ar
ca
n
e
y
ield
m
o
d
eli
n
g
co
m
p
ar
e
d
to
m
o
d
els
with
o
n
e
y
ield
p
r
e
d
icto
r
.
I
n
ad
d
itio
n
,
r
e
d
ed
g
e
b
a
n
d
s
h
a
v
e
th
e
p
o
ten
tial
to
lead
to
im
p
r
o
v
ed
ac
cu
r
ac
y
in
y
ield
p
r
e
d
ic
tio
n
.
B
ased
o
n
t
h
e
liter
atu
r
e,
it
was
f
o
u
n
d
NDVI
an
d
GNDV
I
ar
e
th
e
m
o
s
t
p
r
e
v
alen
t
v
eg
etatio
n
i
n
d
ices
u
s
ed
f
o
r
s
u
g
ar
ca
n
e
y
ield
p
r
ed
ictio
n
in
ad
v
a
n
ce
of
h
ar
v
est.
W
ith
an
o
b
jectiv
e
to
h
av
e
a
b
aselin
e
r
esu
lt
ag
ain
s
t
wh
ich
th
e
r
esu
lts
o
f
o
u
r
ap
p
r
o
ac
h
co
u
ld
b
e
co
m
p
ar
ed
,
we
im
p
lem
en
ted
two
R
F
m
o
d
els,
o
n
e
tr
ai
n
ed
with
NDVI
(
R
F1
)
an
d
a
n
o
th
e
r
tr
ain
ed
with
GNDV
I
(
R
F2
)
f
ea
tu
r
e
as
b
aselin
e
m
o
d
els.
T
h
e
r
esu
lts
o
f
th
ese
m
o
d
els
wh
en
co
m
p
ar
ed
t
o
th
e
R
F
r
esu
lts
o
b
tain
ed
th
r
o
u
g
h
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
(
R
F3
)
v
alid
ate
an
d
h
ig
h
lig
h
t
th
e
u
s
ef
u
ln
ess
o
f
th
e
ap
p
r
o
ac
h
p
r
o
p
o
s
ed
wh
ich
h
as
b
etter
ac
cu
r
ac
y
co
m
p
ar
ativ
el
y
.
R
F3
h
as
s
h
o
wn
im
p
r
o
v
ed
a
cc
u
r
ac
y
b
y
1
6
.
5
%
an
d
1
7
.
9
% c
o
m
p
ar
ed
t
o
R
F1
an
d
R
F2
r
esp
ec
tiv
ely
as sh
o
wn
in
Fig
u
r
e
8
.
Fig
u
r
e
8
.
C
o
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
with
b
aselin
e
m
o
d
els
4.
CO
NCLU
SI
O
N
T
h
e
s
tu
d
y
aim
ed
to
estab
lis
h
a
m
eth
o
d
to
d
ev
elo
p
a
s
u
g
a
r
ca
n
e
y
ield
p
r
ed
ictio
n
m
o
d
el
u
s
in
g
ML
alg
o
r
ith
m
s
an
d
r
e
m
o
te
s
en
s
in
g
d
ata.
Fo
r
t
h
is
p
u
r
p
o
s
e,
we
i
n
v
esti
g
ated
th
e
s
u
itab
ilit
y
o
f
3
1
v
e
g
etatio
n
in
d
ices
in
clu
d
in
g
b
o
th
r
ed
ed
g
e
an
d
n
ar
r
o
wb
an
d
g
r
ee
n
n
ess
in
d
ices
t
o
ex
p
lain
s
u
g
ar
ca
n
e
y
iel
d
in
th
e
s
tu
d
y
ar
ea
.
W
e
f
in
d
th
at
R
F
tr
ain
ed
with
N
DR
E
,
R
E
C
I
an
d
R
VI
in
d
ices
as
y
ield
p
r
ed
i
cto
r
s
ef
f
ici
en
tly
p
r
o
v
i
d
es
a
d
e
ce
n
t
esti
m
ate
o
f
p
r
ed
icted
y
ield
two
m
o
n
th
s
in
ad
v
a
n
ce
to
h
a
r
v
e
s
t
in
th
e
s
tu
d
y
ar
ea
.
An
o
th
er
p
r
o
m
in
en
t
f
i
n
d
in
g
is
th
at,
d
u
e
to
its
h
ig
h
ca
n
o
p
y
d
en
s
ity
at
m
atu
r
ity
,
s
u
g
ar
ca
n
e
y
ield
m
a
y
b
e
m
o
r
e
ac
c
u
r
ately
esti
m
ated
b
y
v
eg
etatio
n
i
n
d
ices
b
a
s
ed
o
n
wav
elen
g
th
s
lo
ca
ted
in
t
h
e
r
ed
ed
g
e
r
e
g
io
n
o
f
elec
tr
o
m
ag
n
etic
s
p
ec
tr
u
m
.
T
h
r
o
u
g
h
o
u
r
ex
p
e
r
im
en
tatio
n
s
b
ased
o
n
m
o
s
t
co
m
m
o
n
l
y
u
s
ed
in
d
ices
f
o
r
s
u
g
ar
ca
n
e
y
ield
,
it
is
f
o
u
n
d
th
at
a
co
m
b
in
atio
n
o
f
NDRE,
R
E
C
I
,
an
d
R
VI
in
d
ices
p
r
o
v
id
e
h
ig
h
er
ac
cu
r
ac
y
t
h
an
NDVI
a
n
d
GNDV
I
b
ased
ML
m
o
d
els.
Sin
ce
s
u
g
ar
ca
n
e
cr
o
p
g
r
o
wth
v
ar
ies
b
etwe
en
lo
ca
liti
es
wi
th
d
if
f
er
en
t
g
r
o
win
g
wea
th
er
co
n
d
itio
n
s
,
p
r
ec
ip
itatio
n
,
an
d
e
n
v
ir
o
n
m
en
tal
v
ar
iab
les th
at
in
tu
r
n
d
r
iv
e
s
u
g
ar
ca
n
e
y
ield
,
th
e
s
tu
d
y
r
e
f
l
ec
ts
o
n
th
e
f
ac
t th
at
in
d
ices
th
at
ex
p
lain
y
ield
h
a
v
e
to
b
e
c
h
o
s
en
with
ca
u
tio
n
a
f
ter
th
eir
s
ig
n
if
ican
ce
is
test
ed
to
ex
p
lain
y
ield
in
th
e
s
tu
d
y
ar
ea
.
T
h
e
m
o
d
el
ca
n
b
e
im
p
r
o
v
ed
b
y
ex
p
lo
r
in
g
m
o
r
e
r
e
d
ed
g
e
b
an
d
b
ased
v
eg
e
tatio
n
in
d
ices
f
r
o
m
h
y
p
er
s
p
ec
tr
al
im
ag
er
ies.
Als
o
,
v
er
if
icatio
n
o
f
th
e
m
eth
o
d
’
s
co
n
s
is
ten
cy
th
r
o
u
g
h
ad
d
itio
n
al
g
r
o
win
g
s
ea
s
o
n
s
is
s
u
g
g
ested
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
s
g
r
atef
u
lly
ac
k
n
o
wled
g
e
th
e
s
u
p
p
o
r
t
ex
ten
d
ed
f
o
r
th
is
wo
r
k
b
y
Sh
r
i
Sam
ir
So
m
aiy
a
(
C
h
air
m
an
a
n
d
MD
,
So
m
aiy
a
g
r
o
u
p
)
,
s
u
g
ar
ca
n
e
f
ar
m
er
s
o
f
th
e
r
e
g
io
n
an
d
G
o
d
av
a
r
i
B
io
-
r
ef
in
er
ies L
td
.
Evaluation Warning : The document was created with Spire.PDF for Python.