T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
24
,
No
.
4
,
A
u
g
u
s
t
20
26
,
p
p
.
1
307
~1
3
1
9
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
24
i
4
.
27747
1307
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//telko
mn
ika
.
u
a
d
.
a
c.
i
d
Decisio
n
-
t
r
ee
-
b
a
s
ed
m
a
chine
l
ea
rni
ng
f
o
r
de
tect
ing
c
o
ff
ee
a
g
ro
forestry
u
sing
SP
O
T
-
7
I
M
a
de
K
hris
na
Yo
g
a
Dev
a
n
dra
,
I
Neng
a
h Sura
t
i J
a
y
a
,
T
a
t
a
ng
T
iry
a
na
D
e
p
a
r
t
me
n
t
o
f
F
o
r
e
st
M
a
n
a
g
e
me
n
t
,
F
a
c
u
l
t
y
o
f
F
o
r
e
st
r
y
a
n
d
En
v
i
r
o
n
me
n
t
,
I
P
B
U
n
i
v
e
r
si
t
y
,
B
o
g
o
r
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
20
,
2025
R
ev
i
s
ed
A
p
r
7
,
2
0
2
6
A
cc
ep
ted
Ma
y
25
,
2026
T
h
is
stu
d
y
d
e
v
e
lo
p
s
a
d
e
c
isio
n
-
tr
e
e
-
b
a
se
d
m
a
c
h
in
e
-
lea
rn
in
g
(M
L
)
a
p
p
ro
a
c
h
to
id
e
n
ti
fy
c
o
ffe
e
a
g
ro
f
o
re
str
y
p
lan
ts
u
sin
g
S
P
OT
-
7
sa
telli
te
im
a
g
e
r
y
.
T
h
e
a
lg
o
rit
h
m
wa
s
d
e
v
e
lo
p
e
d
b
y
e
x
a
m
in
in
g
th
e
c
o
m
b
in
a
ti
o
n
o
f
im
a
g
e
in
d
ice
s
d
e
riv
e
d
f
ro
m
S
P
OT
-
7
a
n
d
b
i
o
p
h
y
sic
a
l
v
a
riab
les
.
De
tec
ti
o
n
u
sin
g
sp
e
c
tral
v
a
riab
les
is
o
f
ten
h
a
m
p
e
re
d
b
y
s
p
e
c
tral
sim
il
a
rit
y
b
e
t
w
e
e
n
v
e
g
e
ta
ti
o
n
c
o
v
e
r
c
las
se
s.
T
h
is
stu
d
y
f
o
u
n
d
th
a
t
a
ML
m
e
th
o
d
t
h
a
t
c
o
m
b
in
e
s
sp
e
c
tral
a
n
d
b
io
p
h
y
sic
a
l
v
a
riab
les
c
a
n
sig
n
if
ica
n
tl
y
i
m
p
ro
v
e
o
v
e
ra
ll
a
c
c
u
ra
c
y
,
f
ro
m
6
0
.
4
%
(u
si
n
g
c
o
n
v
e
n
ti
o
n
a
l
s
p
e
c
tral
v
a
riab
les
a
lo
n
e
)
t
o
9
4
%
(
u
sin
g
in
teg
ra
ted
sp
e
c
tral
-
b
io
p
h
y
sic
a
l
v
a
riab
les
).
F
o
r
d
e
tec
ti
n
g
a
n
d
i
d
e
n
ti
f
y
in
g
a
g
ro
f
o
re
str
y
c
o
ff
e
e
c
las
s
e
s
t
y
p
ica
ll
y
f
o
u
n
d
u
n
d
e
r
tree
c
a
n
o
p
ies
,
th
e
a
d
d
i
ti
o
n
o
f
th
e
“
lan
d
c
o
v
e
r”
v
a
riab
le
p
u
b
li
s
h
e
d
b
y
t
h
e
M
in
istry
o
f
En
v
iro
n
m
e
n
t
a
n
d
F
o
re
stry
c
o
n
tri
b
u
tes
sig
n
if
ica
n
tl
y
to
th
e
c
las
si
f
ica
ti
o
n
o
f
a
g
ro
f
o
re
str
y
c
o
ffe
e
.
Im
p
o
rtan
t
v
a
riab
les
id
e
n
ti
f
ied
in
th
is
m
o
d
e
l
a
re
n
o
rm
a
li
z
e
d
d
if
fe
re
n
c
e
v
e
g
e
tatio
n
in
d
e
x
(
ND
V
I
)
,
v
isi
b
le
d
if
f
e
re
n
c
e
v
e
g
e
tatio
n
in
d
e
x
(
V
D
V
I
)
,
n
o
rm
a
li
z
e
d
re
d
-
g
re
e
n
v
e
g
e
tatio
n
i
n
d
e
x
(
NRG
I
)
,
e
lev
a
ti
o
n
,
a
n
d
la
n
d
c
o
v
e
r.
K
ey
w
o
r
d
s
:
B
io
p
h
y
s
ical
v
ar
iab
les
C
o
f
f
ee
ag
r
o
f
o
r
estr
y
Dec
is
io
n
tr
ee
Ma
ch
i
n
e
l
ea
r
n
i
n
g
Sp
ec
tr
al
v
ar
iab
les
SP
OT
-
7
im
a
g
e
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
:
I
Nen
g
a
h
S
u
r
ati
J
a
y
a
Dep
ar
t
m
en
t o
f
Fo
r
est M
an
a
g
e
m
en
t,
Facu
l
t
y
o
f
Fo
r
estr
y
a
n
d
E
n
v
ir
o
n
m
e
n
t,
I
P
B
Un
iv
er
s
it
y
B
o
g
o
r
(
1
6
6
8
0
)
,
I
n
d
o
n
esia
E
m
ail: i
n
s
-
j
a
y
a
@
ap
p
s
.
ip
b
.
ac
.
i
d
1.
I
NT
RO
D
UCT
I
O
N
Ov
er
th
e
las
t
t
w
o
d
ec
ad
es,
ag
r
o
f
o
r
estr
y
h
as
e
m
er
g
ed
as
a
s
tr
ateg
ic
ap
p
r
o
ac
h
to
s
u
s
tai
n
ab
le
f
o
r
est
r
eso
u
r
ce
m
an
a
g
e
m
en
t
w
it
h
i
n
th
e
f
r
a
m
e
w
o
r
k
o
f
m
u
lti
-
b
u
s
i
n
ess
f
o
r
estr
y
,
also
k
n
o
w
n
a
s
“
m
u
lti
-
u
s
a
h
a
k
eh
u
ta
n
an
(
MU
K)
”
in
I
n
d
o
n
esia.
T
h
is
m
u
lti
-
b
u
s
in
e
s
s
s
y
s
te
m
h
as
i
n
te
g
r
ated
f
o
r
estr
y
an
d
ag
r
ic
u
lt
u
r
al
co
m
m
o
d
itie
s
u
n
d
er
a
s
in
g
le
m
an
a
g
e
m
e
n
t
u
n
it
,
en
ab
li
n
g
s
i
m
u
ltan
eo
u
s
ec
o
lo
g
ical
a
n
d
ec
o
n
o
m
ic
b
en
e
f
its
.
I
t
h
as
b
ee
n
s
h
o
w
n
th
a
t
ag
r
o
f
o
r
es
tr
y
p
r
ac
tices
en
h
a
n
ce
lan
d
p
r
o
d
u
ctiv
i
t
y
,
b
o
th
in
ter
m
s
o
f
f
o
r
est
co
v
er
u
tili
za
tio
n
an
d
ag
r
icu
ltu
r
al
y
ield
[
1
]
.
T
an
g
g
a
m
u
s
R
e
g
e
n
c
y
i
n
L
a
m
p
u
n
g
P
r
o
v
in
ce
is
a
g
o
o
d
ex
a
m
p
le
o
f
ag
r
o
f
o
r
estr
y
co
f
f
ee
p
r
ac
t
ices,
w
ith
an
ex
te
n
s
i
v
e
f
o
r
est
ar
ea
o
f
ap
p
r
o
x
i
m
atel
y
1
,
4
7
6
.
6
9
k
m
²,
i
n
clu
d
in
g
p
r
o
t
ec
ted
f
o
r
ests
an
d
co
n
s
er
v
atio
n
ar
ea
s
.
A
t
t
h
i
s
s
t
u
d
y
s
ite,
R
o
b
u
s
ta
c
o
f
f
ee
i
s
th
e
d
o
m
in
a
n
t
s
p
ec
ie
s
,
p
la
y
in
g
a
cr
u
cia
l
r
o
le
in
e
co
n
o
m
ic
v
alu
e
a
n
d
ec
o
lo
g
ical
ad
ap
tab
ilit
y
.
T
h
is
s
p
ec
ies
d
em
o
n
s
tr
ates
a
b
r
o
ad
ec
o
l
o
g
ical
to
ler
an
ce
,
g
r
o
w
i
n
g
ac
r
o
s
s
a
w
id
e
altitu
d
i
n
a
l r
an
g
e
u
n
d
er
s
h
ad
ed
co
n
d
itio
n
s
[
2
]
,
[
3
]
.
T
h
e
p
r
esen
ce
o
f
s
h
ad
e
tr
ee
s
is
es
s
en
t
ial
i
n
r
eg
u
lati
n
g
m
icr
o
cli
m
atic
co
n
d
it
io
n
s
,
p
ar
ti
cu
lar
l
y
b
y
r
ed
u
cin
g
ex
ce
s
s
i
v
e
r
ad
iatio
n
ex
p
o
s
u
r
e
a
n
d
i
m
p
r
o
v
i
n
g
p
lan
t
p
r
o
d
u
ctiv
it
y
.
C
o
n
s
eq
u
en
tl
y
,
co
f
f
ee
-
b
a
s
ed
ag
r
o
f
o
r
estr
y
s
y
s
te
m
s
h
a
v
e
b
ec
o
m
e
an
i
m
p
o
r
tan
t
co
m
p
o
n
en
t
o
f
b
o
th
lo
ca
l
li
v
eli
h
o
o
d
s
an
d
g
lo
b
al
ag
r
ic
u
lt
u
r
al
m
ar
k
et
s
[
1
]
,
[
3
]
.
C
o
f
f
ee
g
r
o
w
n
in
a
g
r
o
f
o
r
estr
y
s
y
s
te
m
s
m
a
k
es
a
s
i
g
n
i
f
ican
t
co
n
tr
ib
u
tio
n
to
co
m
m
u
n
i
t
y
w
el
l
-
b
ei
n
g
w
h
ile
a
ls
o
g
e
n
er
ati
n
g
p
o
s
itiv
e
ec
o
lo
g
ical
i
m
p
ac
t
s
.
Ho
w
e
v
er
,
ac
cu
r
ate
d
ata
o
n
t
h
e
ex
ten
t
o
f
t
h
ese
ag
r
o
f
o
r
estr
y
co
f
f
ee
p
la
n
tatio
n
s
is
e
x
tr
e
m
e
l
y
d
i
f
f
icu
lt
to
o
b
tain
.
T
h
e
p
r
o
b
lem
i
s
th
a
t
id
en
t
if
y
in
g
a
n
d
m
ap
p
in
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
16
93
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
4
,
A
u
g
u
s
t
20
26
:
1
3
0
7
-
1
3
1
9
1308
th
e
s
p
atial
d
is
tr
ib
u
tio
n
o
f
co
f
f
ee
ag
r
o
f
o
r
estr
y
s
y
s
te
m
s
,
esp
e
ciall
y
th
o
s
e
lo
ca
ted
b
en
ea
th
th
e
f
o
r
est
ca
n
o
p
y
,
is
ex
tr
e
m
e
l
y
d
if
f
ic
u
lt.
T
er
r
estrial
m
ap
p
in
g
o
r
in
ter
v
ie
w
s
ar
e
in
ef
f
icie
n
t
i
n
v
e
n
to
r
y
m
eth
o
d
s
f
o
r
lar
g
e
-
s
ca
le
ap
p
licatio
n
s
.
F
u
r
th
er
m
o
r
e,
th
e
av
ailab
ilit
y
o
f
o
p
tical
r
e
m
o
te
s
en
s
in
g
d
ata
i
s
li
m
i
ted
to
o
n
l
y
o
b
j
ec
ts
ex
p
o
s
ed
at
th
e
to
p
o
f
th
e
tr
ee
ca
n
o
p
y
o
r
th
o
s
e
ca
p
tu
r
ed
b
y
ca
m
er
as.
Mi
s
class
if
ica
tio
n
o
f
ten
o
cc
u
r
s
a
m
o
n
g
ca
n
o
p
y
co
v
e
r
class
es
w
h
e
n
s
i
m
ilar
class
e
s
ar
e
s
p
ec
tr
ally
m
i
x
ed
,
d
u
e
to
s
p
ec
tr
al
m
ix
in
g
b
et
w
ee
n
co
f
f
ee
p
lan
ts
an
d
o
th
er
ca
n
o
p
y
v
e
g
etati
o
n
.
T
h
er
ef
o
r
e,
r
eliab
le
s
p
atial
in
f
o
r
m
ati
o
n
o
n
t
h
e
ex
te
n
t
a
n
d
d
is
tr
ib
u
tio
n
o
f
co
f
f
ee
ag
r
o
f
o
r
estr
y
s
y
s
te
m
s
is
s
ev
er
e
l
y
li
m
ited
,
h
a
m
p
er
in
g
ef
f
ec
ti
v
e
p
lan
n
i
n
g
,
m
o
n
i
to
r
in
g
,
an
d
p
o
lic
y
m
a
k
i
n
g
.
R
ec
en
t
ad
v
a
n
ce
s
i
n
r
e
m
o
te
s
e
n
s
i
n
g
an
d
m
ac
h
in
e
lear
n
in
g
(
ML
)
h
a
v
e
p
r
o
v
id
ed
n
e
w
o
p
p
o
r
tu
n
i
ties
f
o
r
lan
d
co
v
er
class
if
icatio
n
.
Am
o
n
g
t
h
e
s
e,
d
ec
is
io
n
tr
ee
alg
o
r
ith
m
s
h
a
v
e
g
ai
n
ed
atten
tio
n
f
o
r
th
eir
s
i
m
p
licit
y
,
in
ter
p
r
etab
ilit
y
,
a
n
d
r
ep
r
o
d
u
ci
b
ilit
y
ac
r
o
s
s
d
iv
er
s
e
d
atasets
[
4
]
–
[
6
]
.
P
r
ev
io
u
s
s
tu
d
ie
s
h
a
v
e
d
em
o
n
s
tr
ated
th
a
t
in
te
g
r
atin
g
s
p
ec
tr
al
v
ar
iab
les
d
er
iv
ed
f
r
o
m
s
atelli
te
i
m
a
g
er
y
w
it
h
n
o
n
-
s
p
ec
tr
al
v
ar
iab
le
s
,
s
u
ch
as
b
io
p
h
y
s
ica
l
an
d
g
eo
g
r
ap
h
ical
f
ac
to
r
s
,
c
an
s
i
g
n
if
ica
n
tl
y
i
m
p
r
o
v
e
cla
s
s
i
f
icatio
n
ac
cu
r
ac
y
.
Ho
w
e
v
er
,
ap
p
ly
in
g
t
h
ese
ap
p
r
o
ac
h
es
to
co
m
p
le
x
ag
r
o
f
o
r
estry
s
y
s
te
m
s
,
p
ar
tic
u
lar
l
y
f
o
r
d
etec
tin
g
u
n
d
er
s
to
r
y
cr
o
p
s
s
u
c
h
as
co
f
f
ee
,
r
e
m
ain
s
ch
al
len
g
i
n
g
.
P
r
ev
io
u
s
r
esear
ch
h
a
s
id
e
n
ti
f
i
ed
s
ev
er
al
cr
itical
is
s
u
e
s
,
p
ar
ticu
lar
l
y
i
n
t
h
e
u
s
e
o
f
v
ar
iab
les
d
er
iv
ed
s
o
lel
y
f
r
o
m
s
p
ec
tr
al
d
ata
[
4
]
,
[
7
]
–
[
9
]
.
First,
th
e
ef
f
ec
ti
v
e
n
ess
o
f
s
p
ec
tr
al
im
a
g
e
-
b
ased
d
ec
is
io
n
-
tr
ee
al
g
o
r
ith
m
s
f
o
r
d
etec
t
in
g
u
n
d
er
s
to
r
y
co
f
f
ee
ag
r
o
f
o
r
ests
r
e
m
a
in
s
lo
w
b
ec
au
s
e
o
p
tical
i
m
a
g
er
y
p
r
o
v
id
es
o
n
l
y
s
p
ec
tr
al
in
f
o
r
m
atio
n
ab
o
u
t
t
h
e
v
e
g
et
atio
n
ca
n
o
p
y
[
1
0
]
.
Seco
n
d
,
u
n
d
er
s
ta
n
d
in
g
t
h
e
m
o
s
t
i
n
f
l
u
en
t
ial
v
ar
iab
les
i
n
d
eter
m
in
i
n
g
co
f
f
ee
a
g
r
o
f
o
r
est
r
y
lo
ca
tio
n
s
is
s
tr
o
n
g
l
y
in
f
l
u
e
n
ce
d
b
y
s
i
te
s
u
i
tab
ilit
y
(
b
io
p
h
y
s
ical
f
ac
to
r
s
)
an
d
s
o
cio
ec
o
n
o
m
ic
f
ac
to
r
s
(
r
o
ad
ac
ce
s
s
,
d
is
tan
ce
f
r
o
m
s
ett
lem
en
t
ce
n
ter
s
)
[
1
1
]
.
T
h
ir
d
,
c
h
alle
n
g
e
s
r
elate
d
to
m
o
d
el
g
e
n
er
aliza
tio
n
,
in
c
lu
d
i
n
g
o
v
er
f
itti
n
g
an
d
tr
a
n
s
f
er
ab
i
lit
y
,
r
e
m
ai
n
i
n
s
u
f
f
ic
ien
t
l
y
ad
d
r
ess
ed
[
1
2
]
.
T
h
ese
co
n
d
itio
n
s
cr
ea
te
u
n
ce
r
tai
n
t
y
i
n
m
o
d
el
d
ev
elo
p
m
e
n
t
an
d
li
m
it
th
e
o
p
er
atio
n
al
ap
p
licatio
n
o
f
ML
ap
p
r
o
ac
h
es
f
o
r
ag
r
o
f
o
r
estr
y
m
ap
p
in
g
in
tr
o
p
ical
r
eg
io
n
s
.
T
o
ad
d
r
ess
th
ese
ch
alle
n
g
e
s
,
th
e
o
b
j
ec
tiv
e
o
f
th
is
s
t
u
d
y
is
to
d
ev
elo
p
a
d
ec
is
io
n
tr
ee
-
b
ased
ML
m
o
d
el
th
at
in
te
g
r
ates
s
p
ec
tr
al
v
ar
iab
les
d
er
iv
ed
f
r
o
m
SP
O
T
-
7
im
a
g
er
y
w
it
h
n
o
n
-
s
p
ec
tr
al
v
ar
iab
les
s
u
c
h
as
b
io
p
h
y
s
ical
a
n
d
s
o
cio
-
ec
o
n
o
m
ic
v
ar
ia
b
les
i
n
T
an
g
g
a
m
u
s
R
eg
e
n
c
y
.
Sp
ec
i
f
icall
y
,
th
is
s
t
u
d
y
s
ee
k
s
to
id
en
ti
f
y
k
e
y
v
ar
iab
les
an
d
th
eir
r
elativ
e
im
p
o
r
tan
ce
,
ev
a
lu
ate
t
h
e
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
an
d
r
eliab
ilit
y
,
an
d
ass
e
s
s
it
s
in
ter
p
r
etab
ilit
y
.
T
h
is
s
t
u
d
y
is
ex
p
ec
ted
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
an
d
ap
p
lica
b
ilit
y
o
f
s
p
atial
m
ap
p
in
g
f
o
r
co
f
f
ee
a
g
r
o
f
o
r
estr
y
s
y
s
te
m
s
.
T
h
e
d
ev
elo
p
ed
m
o
d
el
i
s
e
x
p
ec
ted
to
p
r
o
v
id
e
a
r
o
b
u
s
t
a
n
d
ea
s
il
y
u
n
d
er
s
to
o
d
f
r
a
m
e
w
o
r
k
to
s
u
p
p
o
r
t su
s
tain
a
b
le
lan
d
m
an
a
g
e
m
e
n
t a
n
d
ag
r
o
f
o
r
estr
y
m
o
n
ito
r
i
n
g
i
n
tr
o
p
ical
en
v
ir
o
n
m
e
n
ts
.
2.
M
E
T
H
O
D
T
an
g
g
a
m
u
s
R
eg
e
n
c
y
i
s
lo
ca
ted
at
1
0
4
°
1
8
'
–
1
0
5
°1
2
'
E
an
d
5
°
0
5
'
–
5
°5
6
'
S,
en
co
m
p
ass
i
n
g
4
6
5
4
.
9
5
k
m
²
w
it
h
ele
v
atio
n
s
r
ea
ch
in
g
2
1
1
5
m
(
s
ee
Fig
u
r
e
1
)
.
A
p
p
r
o
x
i
m
atel
y
4
0
%
o
f
th
e
r
eg
io
n
co
m
p
r
is
es
p
latea
u
lan
d
s
ca
p
es,
in
cl
u
d
in
g
h
i
ll
y
t
o
m
o
u
n
tai
n
o
u
s
ter
r
ain
.
T
an
g
g
a
m
u
s
R
eg
e
n
c
y
h
as
a
co
f
f
ee
p
lan
tatio
n
ar
ea
o
f
4
1
,
5
1
2
h
ec
tar
es
an
d
a
p
r
o
d
u
ctio
n
o
f
3
3
,
4
8
2
to
n
s
.
Fo
r
o
n
-
th
e
-
g
r
o
u
n
d
o
b
s
er
v
atio
n
a
n
d
d
ata
p
r
o
ce
s
s
in
g
a
n
d
an
al
y
s
is
,
w
e
u
s
ed
a
g
lo
b
al
p
o
s
itio
n
i
n
g
s
y
s
te
m
(
GP
S)
d
ev
ic
e,
s
p
ec
if
ic
s
o
f
t
w
ar
e
s
u
ch
a
s
Ar
cGI
S
an
d
E
R
D
A
S
I
m
ag
i
n
e
f
o
r
s
p
atial
an
a
l
y
s
is
an
d
s
tatis
tical
d
at
a
ex
tr
ac
ti
o
n
,
an
d
R
ap
id
Min
er
Stu
d
io
f
o
r
d
ev
elo
p
in
g
t
h
e
d
ec
is
io
n
tr
ee
alg
o
r
it
h
m
.
T
h
e
al
g
o
r
ith
m
w
as t
h
e
n
i
m
p
le
m
e
n
te
d
in
P
y
t
h
o
n
.
T
h
e
m
ain
d
ata
u
s
ed
ar
e
SP
OT
-
7
i
m
ag
er
y
,
w
h
ich
in
cl
u
d
es
o
n
e
p
an
ch
r
o
m
a
tic
(
P
A
N)
b
a
n
d
w
it
h
a
h
i
g
h
s
p
atial
r
es
o
lu
tio
n
o
f
1
.
5
m
a
n
d
f
o
u
r
m
u
ltis
p
ec
tr
al
b
an
d
s
s
p
an
n
in
g
0
.
4
5
0
to
0
.
7
4
5
μ
m
,
en
ab
l
in
g
d
etailed
s
p
atial
o
b
s
er
v
atio
n
.
T
h
e
m
u
lti
s
p
ec
tr
al
b
an
d
s
ar
e
p
r
o
v
id
ed
at
a
s
p
atial
r
eso
lu
tio
n
o
f
6
m
,
in
cl
u
d
in
g
t
h
r
ee
v
i
s
ib
le
b
an
d
s
(
b
lu
e,
g
r
ee
n
,
a
n
d
r
ed
)
an
d
a
n
ea
r
-
in
f
r
ar
ed
b
an
d
s
p
an
n
i
n
g
0
.
7
6
0
–
0
.
8
9
0
μ
m
.
T
h
ese
s
p
ec
tr
al
b
an
d
s
co
ll
ec
ti
v
el
y
s
u
p
p
o
r
t
a
w
id
e
r
an
g
e
o
f
a
p
p
licatio
n
s
,
i
n
clu
d
i
n
g
v
e
g
et
atio
n
an
a
l
y
s
is
,
la
n
d
co
v
er
class
i
f
icatio
n
,
an
d
en
v
ir
o
n
m
e
n
tal
m
o
n
ito
r
i
n
g
[
1
3
]
.
T
h
ese
b
an
d
s
w
er
e
th
e
n
co
n
v
er
ted
in
to
s
ev
er
al
in
d
ices,
n
a
m
el
y
th
e
n
o
r
m
a
lized
d
if
f
er
e
n
ce
v
eg
eta
tio
n
in
d
e
x
(
NDVI
)
,
th
e
n
o
r
m
al
ized
r
ed
-
g
r
ee
n
v
e
g
etat
io
n
i
n
d
ex
(
NR
GI
)
,
th
e
v
i
s
ib
le
d
if
f
er
e
n
ce
v
eg
e
tatio
n
in
d
e
x
(
VDVI
)
,
an
d
th
e
v
is
ib
le
at
m
o
s
p
h
er
e
r
esis
tan
ce
(
V
A
R
I
)
.
T
h
e
s
e
in
d
ices
lev
er
a
g
e
s
p
ec
tr
al
d
ata
to
p
r
o
v
id
e
in
s
i
g
h
ts
in
to
v
e
g
etatio
n
d
y
n
a
m
ic
s
,
o
f
f
er
i
n
g
a
n
o
n
-
d
estru
ct
iv
e
m
ea
n
s
o
f
m
o
n
ito
r
in
g
an
d
m
a
n
a
g
i
n
g
f
o
r
es
t
r
e
s
o
u
r
c
es,
p
ar
ticu
lar
l
y
f
o
r
ass
es
s
i
n
g
v
eg
eta
tio
n
cla
s
s
e
s
i
n
d
en
s
e
-
c
an
o
p
y
a
g
r
o
f
o
r
estr
y
p
lan
tatio
n
s
.
E
ac
h
i
n
d
ex
h
a
s
u
n
iq
u
e
s
tr
en
g
t
h
s
: N
DVI
is
u
s
ed
t
o
ass
ess
v
eg
e
tatio
n
h
ea
lt
h
d
u
e
to
its
s
en
s
iti
v
it
y
to
ch
lo
r
o
p
h
y
ll
co
n
ten
t
an
d
ca
n
o
p
y
s
tr
u
c
tu
r
e,
a
n
d
NR
GI
,
w
h
i
c
h
is
s
i
m
ilar
to
t
h
e
R
ed
-
E
d
g
e
NDVI
,
h
as
p
o
te
n
tial
f
o
r
b
io
m
as
s
esti
m
at
io
n
,
p
ar
tic
u
lar
l
y
w
h
e
n
co
m
b
in
ed
w
i
th
o
th
er
i
n
d
ices
[
1
4
]
.
T
h
e
VDVI
,
ak
i
n
to
V
AR
I
,
is
d
esig
n
ed
to
b
e
less
s
en
s
i
tiv
e
t
o
atm
o
s
p
h
er
ic
co
n
d
itio
n
s
,
m
a
k
in
g
it
u
s
ef
u
l
f
o
r
m
o
n
ito
r
in
g
v
eg
eta
tio
n
h
ea
l
th
i
n
v
ar
iab
le
w
ea
th
er
co
n
d
itio
n
s
[
1
0
]
.
VA
R
I
is
e
f
f
ec
ti
v
e
in
d
is
tin
g
u
i
s
h
in
g
v
eg
eta
tio
n
f
r
o
m
s
o
il
an
d
o
th
er
n
o
n
-
v
eg
eta
tiv
e
s
u
r
f
ac
es,
w
h
ic
h
i
s
cr
u
cial
in
d
en
s
e
ca
n
o
p
y
e
n
v
ir
o
n
m
e
n
t
s
w
h
er
e
g
r
o
u
n
d
co
v
er
ca
n
o
b
s
cu
r
e
v
eg
eta
tio
n
s
ig
n
als
[
1
5
]
.
T
o
im
p
r
o
v
e
m
o
d
el
p
er
f
o
r
m
a
n
ce
,
th
e
s
tu
d
y
also
ev
al
u
ated
th
e
co
n
tr
ib
u
tio
n
o
f
b
io
p
h
y
s
ica
l
v
ar
iab
les
s
u
c
h
as e
le
v
atio
n
,
s
lo
p
e,
v
i
s
u
a
l c
o
m
m
o
n
la
n
d
co
v
er
,
an
d
p
r
o
x
i
m
it
y
to
r
iv
er
s
,
r
o
ad
s
,
an
d
s
et
tle
m
e
n
ts
.
E
le
v
atio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Dec
is
io
n
-
tr
ee
-
b
a
s
ed
ma
ch
in
e
lea
r
n
in
g
fo
r
d
etec
tin
g
c
o
ffee
a
g
r
o
fo
r
estr
y
…
(
I
Ma
d
e
K
h
r
is
n
a
Yo
g
a
Dev
a
n
d
r
a
)
1309
an
d
s
lo
p
e
w
er
e
d
er
iv
ed
f
r
o
m
N
atio
n
al
A
er
o
n
a
u
tics
a
n
d
Sp
ac
e
A
d
m
in
is
tr
atio
n
d
ig
it
al
elev
atio
n
m
o
d
el
(
NA
S
ADE
M
)
,
w
h
ile
v
i
s
u
al
la
n
d
co
v
er
w
a
s
in
ter
p
r
eted
f
r
o
m
SP
O
T
-
7
im
a
g
er
y
.
Fu
r
t
h
er
m
o
r
e,
p
r
o
x
i
m
it
y
-
b
a
s
ed
v
ar
iab
les
w
er
e
i
n
co
r
p
o
r
ated
,
in
cl
u
d
in
g
d
is
ta
n
ce
s
to
r
iv
er
s
,
r
o
ad
s
,
an
d
s
ettle
m
en
ts
,
o
b
tai
n
ed
f
r
o
m
th
e
R
u
p
a
B
u
mi
I
n
d
o
n
esia
(
R
B
I
)
m
ap
.
F
ield
-
b
ased
s
a
m
p
le
p
lo
ts
co
llec
ted
th
r
o
u
g
h
g
r
o
u
n
d
s
u
r
v
e
y
s
i
n
2
0
2
3
w
er
e
u
s
ed
as
r
ef
er
en
ce
d
ata,
w
h
ile
ad
m
i
n
i
s
tr
ativ
e
b
o
u
n
d
ar
ies
(
d
is
tr
ict
an
d
s
u
b
d
is
tr
ict
lev
el
s
)
w
er
e
also
s
o
u
r
ce
d
f
r
o
m
R
B
I
2
0
2
0
d
ata
to
s
u
p
p
o
r
t
s
p
atial
an
al
y
s
i
s
.
Fig
u
r
e
1
.
R
esear
ch
lo
ca
tio
n
i
n
T
an
g
g
a
m
u
s
Dis
tr
ict
2
.
1
.
I
m
a
g
e
pre
pro
ce
s
s
ing
P
r
io
r
to
an
al
y
s
i
s
,
th
e
s
p
at
ial
d
ata
w
er
e
g
eo
-
co
r
r
ec
te
d
an
d
p
r
ep
r
o
ce
s
s
ed
,
in
clu
d
in
g
i
n
d
ex
d
ev
elo
p
m
en
t,
s
p
atial
f
il
ter
in
g
,
an
d
i
m
ag
e
[
1
6
]
.
R
ad
io
m
etr
ic
co
r
r
ec
tio
n
w
as
p
er
f
o
r
m
ed
to
i
m
p
r
o
v
e
i
m
a
g
e
q
u
alit
y
b
y
m
iti
g
ati
n
g
s
u
r
f
ac
e
r
ef
lecta
n
ce
er
r
o
r
s
an
d
at
m
o
s
p
h
er
ic
i
n
f
l
u
en
ce
s
,
o
b
tain
in
g
m
o
r
e
ac
cu
r
ate
in
f
o
r
m
atio
n
[
1
7
]
.
2
.
2
.
Cla
s
s
s
che
m
e
a
nd
t
ra
ini
ng
a
re
a
dev
elo
p
m
ent
T
h
e
class
s
c
h
e
m
e
w
a
s
d
ev
el
o
p
ed
th
r
o
u
g
h
v
is
u
al
i
n
ter
p
r
etatio
n
o
f
i
m
a
g
er
y
b
ased
o
n
ele
m
e
n
ts
s
u
ch
a
s
s
h
ap
e,
co
lo
r
/h
u
e,
p
atter
n
,
tex
t
u
r
e,
s
h
ad
o
w
s
ize,
lo
ca
tio
n
,
an
d
a
s
s
o
ciatio
n
[
1
7
]
.
T
h
e
class
s
ch
e
m
e
w
a
s
d
iv
id
ed
in
to
1
0
class
es:
w
ater
b
o
d
ies
(
W
T
B
)
,
p
lan
tatio
n
s
/e
s
tate
cr
o
p
(
P
L
T
)
,
sc
r
u
b
(
SC
B
)
,
n
atu
r
al
f
o
r
ests
(
FR
S),
b
ar
e
lan
d
(
B
R
L
)
,
s
ett
le
m
e
n
t
s
(
ST
L
)
,
m
i
x
ed
d
r
y
la
n
d
ag
r
icu
ltu
r
e
(
A
G
C
)
,
p
ad
d
y
f
ield
s
(
P
DF)
,
co
f
f
ee
a
g
r
o
f
o
r
estr
y
(
C
A
F),
an
d
n
o
n
-
c
lo
u
d
lan
d
co
v
er
(
C
L
D)
.
A
tr
ain
i
n
g
ar
ea
o
f
6
,
6
6
6
p
o
in
ts
w
as
cr
ea
ted
b
ased
o
n
th
e
class
s
ch
e
m
e
an
d
d
is
tr
ib
u
ted
ev
e
n
l
y
ac
r
o
s
s
lan
d
co
v
er
.
T
h
e
l
o
c
a
t
i
o
n
o
f
th
e
t
r
ai
n
in
g
a
r
ea
w
a
s
d
e
t
e
r
m
in
e
d
b
a
s
e
d
o
n
im
ag
e
a
p
p
e
a
r
an
c
e
an
d
f
i
e
l
d
s
u
r
v
ey
s
o
f
th
e
c
o
f
f
e
e
c
l
as
s
,
b
o
t
h
a
g
r
o
f
o
r
e
s
t
r
y
a
n
d
m
o
n
o
cu
l
tu
r
e.
T
h
e
s
tu
d
y
u
s
e
d
a
t
r
a
in
in
g
a
r
e
a
w
i
th
a
2
.
2
5
m
b
u
f
f
e
r
,
o
r
4
.
5
m
×
4
.
5
m
(
2
0
.
2
5
m
2
)
.
W
e
u
s
e
d
a
s
am
p
l
e
s
iz
e
o
f
4
.
5
×
4
.
5
(
o
r
a
2
.
2
5
m
p
o
i
n
t
b
u
f
f
e
r
)
b
e
c
au
s
e
th
e
im
a
g
e
p
ix
el
s
i
z
e
is
1
.
5
m
.
T
h
is
s
i
z
e
a
ll
o
w
s
f
o
r
r
e
p
r
e
s
en
t
at
iv
e
d
a
ta
,
w
it
h
o
n
e
o
b
s
e
r
v
at
i
o
n
p
o
i
n
t
c
o
v
e
r
in
g
9
p
ix
e
ls
(
3
×
3
p
ix
e
ls
)
.
2
.
3
.
Dec
is
io
n t
re
e
a
l
g
o
rit
h
m
dev
elo
p
m
ent
T
h
e
d
ec
is
io
n
tr
ee
alg
o
r
ith
m
i
s
a
n
o
n
-
p
ar
a
m
etr
ic
m
eth
o
d
ca
p
ab
le
o
f
h
an
d
lin
g
lar
g
e
a
n
d
co
m
p
le
x
d
atasets
w
it
h
o
u
t
r
eq
u
ir
i
n
g
a
c
o
m
p
le
x
p
ar
a
m
etr
ic
s
tr
u
ctu
r
e.
Fig
u
r
e
2
s
h
o
w
s
th
e
p
r
o
ce
d
u
r
e
f
o
r
d
ev
elo
p
in
g
th
e
d
ec
is
io
n
tr
ee
alg
o
r
ith
m
o
f
t
h
i
s
s
tu
d
y
[
4
]
,
[
1
8
]
.
T
h
e
d
ata
w
er
e
d
iv
id
ed
in
to
a
tr
ain
in
g
d
ataset
(
7
0
%)
an
d
a
v
alid
atio
n
d
ataset
(
3
0
%).
T
h
e
tr
ain
in
g
d
ataset
w
as
u
s
ed
to
b
u
ild
a
d
ec
is
io
n
tr
e
e
m
o
d
el,
an
d
th
e
v
alid
atio
n
d
ataset
w
a
s
u
s
ed
to
d
eter
m
i
n
e
th
e
o
p
ti
m
al
tr
ee
s
ize
to
ac
h
iev
e
t
h
e
f
in
al
m
o
d
el
[
1
9
]
.
T
h
e
im
p
o
r
tan
ce
o
f
v
ar
iab
les
w
as
d
eter
m
i
n
ed
u
s
i
n
g
th
e
B
r
u
t
e
Fo
r
ce
m
et
h
o
d
(
1
)
,
in
f
o
r
m
atio
n
g
a
in
(
2
)
,
g
ai
n
r
atio
(
3
)
,
an
d
g
in
i
in
d
ex
(
4
)
u
s
i
n
g
th
e
f
o
r
m
u
las a
s
d
ep
icted
f
r
o
m
(
1
)
to
(
6
)
[
1
8
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
16
93
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
4
,
A
u
g
u
s
t
20
26
:
1
3
0
7
-
1
3
1
9
1310
(
)
=
−
∑
=
1
2
(
)
(
1
)
(
,
)
=
(
)
−
∑
|
|
|
|
Î
(
)
(
)
(
2
)
(
,
)
=
(
,
)
(
)
(
3
)
(
)
=
−
∑
|
|
|
|
Î
(
)
2
(
|
|
|
|
)
(
4
)
(
)
=
1
−
∑
(
)
2
=
1
(
5
)
(
,
)
=
∑
|
|
|
|
Î
(
)
(
)
(
6
)
Fu
r
t
h
er
m
o
r
e,
th
e
m
o
d
el
p
er
f
o
r
m
an
ce
e
v
al
u
atio
n
w
a
s
co
n
d
u
cted
b
y
m
ea
s
u
r
in
g
o
v
er
all
ac
c
u
r
ac
y
a
n
d
k
ap
p
a
ac
cu
r
ac
y
u
s
in
g
(
7
)
,
(
8
)
an
d
(
9
)
[
1
2
]
,
[
1
4
]
,
[
1
5
]
,
[
2
0
]
,
[
2
1
]
.
T
h
e
r
elev
an
ce
o
f
th
is
e
v
al
u
atio
n
ap
p
r
o
ac
h
is
s
u
p
p
o
r
ted
b
y
r
ec
en
t
s
t
u
d
ies
s
u
ch
as
N
u
t
h
a
m
m
ac
h
o
t
a
n
d
Stra
to
u
lias
[
2
0
]
,
w
h
o
s
u
cc
ess
f
u
ll
y
ap
p
lied
v
eg
etatio
n
in
d
ices
d
er
iv
ed
f
r
o
m
Se
n
ti
n
e
l‑
2
s
atellite
i
m
ag
er
y
to
class
i
f
y
h
ea
lt
h
y
an
d
d
is
ea
s
ed
o
il
p
al
m
tr
ee
s
,
t
h
er
eb
y
d
em
o
n
s
tr
ati
n
g
th
e
ap
p
licab
ilit
y
o
f
d
ec
is
io
n
tr
ee
‑
b
ased
clas
s
i
f
icatio
n
in
a
g
r
icu
l
tu
r
al
co
n
te
x
t
s
.
(
)
=
∑
=
1
100%
(
7
)
(
)
=
−
1
−
(
8
)
=
∑
(
+
.
+
)
=
1
2
(
9
)
W
h
er
e
is
t
h
e
d
ata
s
et
(
s
a
m
p
l
e
d
ata)
b
eiin
g
te
s
ted
,
is
th
e
p
r
o
p
o
r
ti
o
n
o
f
t
h
e
-
th
c
lass
,
is
t
h
e
n
u
m
b
er
o
f
class
es
.
is
t
h
e
n
u
m
b
er
o
f
d
ata
p
o
in
ts
at
v
al
u
e
f
o
r
ea
ch
v
ar
ia
b
le
,
w
h
er
e
(
s
u
b
-
attr
ib
u
te)
i
s
a
m
e
m
b
er
o
f
attr
ib
u
te
.
is
th
e
at
tr
ib
u
te
o
r
v
ar
iab
le
b
ein
g
test
ed
,
is
th
e
p
r
o
p
o
r
tio
n
o
f
th
e
-
t
h
clas
s
in
t
h
e
to
tal
d
ata
s
et
,
(
,
)
is
i
n
f
o
r
m
atio
n
g
a
in
f
o
r
v
ar
ia
b
le
f
r
o
m
th
e
e
n
tire
d
ata
s
et
(
)
.
A
h
i
g
h
er
v
al
u
e
f
o
r
a
v
ar
iab
le
in
d
icate
s
b
etter
clas
s
s
ep
ar
atio
n
in
t
h
e
d
ec
is
io
n
tr
ee
.
Fig
u
r
e
2
.
ML
-
b
ased
m
o
d
els a
n
d
ex
p
er
i
m
e
n
tal
m
et
h
o
d
s
ap
p
lied
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Dec
is
io
n
-
tr
ee
-
b
a
s
ed
ma
ch
in
e
lea
r
n
in
g
fo
r
d
etec
tin
g
c
o
ffee
a
g
r
o
fo
r
estr
y
…
(
I
Ma
d
e
K
h
r
is
n
a
Yo
g
a
Dev
a
n
d
r
a
)
1311
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
F
e
a
t
ure
s
elec
t
io
n
T
h
e
r
esu
lts
o
f
a
s
tu
d
y
u
s
in
g
a
f
u
ll
f
ea
t
u
r
e
s
et
o
f
1
0
v
ar
iab
les
s
h
o
w
ed
s
i
g
n
i
f
ica
n
t
o
v
er
f
itti
n
g
,
r
esu
lti
n
g
in
a
s
o
m
e
w
h
at
o
d
d
an
d
u
n
n
atu
r
al
s
p
atial
m
ap
.
A
cc
u
r
ac
y
w
as
h
i
g
h
,
b
u
t
th
e
v
is
u
al
cl
ass
i
f
icatio
n
r
esu
l
t
s
r
ese
m
b
led
t
h
e
s
h
ap
es
o
f
t
h
e
i
n
p
u
t
v
ar
iab
les.
T
h
e
r
ef
o
r
e,
t
h
i
s
s
t
u
d
y
s
elec
ted
v
ar
iab
les
to
r
ed
u
ce
o
v
er
f
i
ttin
g
,
th
er
eb
y
en
ab
li
n
g
o
p
ti
m
al
m
o
d
el
p
er
f
o
r
m
a
n
ce
[
9
]
,
[
2
1
]
.
T
h
e
lar
g
e
lan
d
co
v
er
v
ar
iab
le
v
al
u
es
i
n
b
o
th
t
h
e
o
v
er
f
itti
n
g
a
n
d
b
est
al
g
o
r
ith
m
r
es
u
lt
s
m
a
y
b
e
in
f
lu
e
n
ce
d
b
y
la
n
d
co
v
er
v
ar
i
ab
le
d
ata
g
en
er
ated
f
r
o
m
v
i
s
u
al
in
ter
p
r
etatio
n
o
f
s
atell
ite
i
m
a
g
er
y
a
n
d
b
y
th
e
cr
ea
tio
n
o
f
tr
ain
i
n
g
ar
ea
d
ata
tailo
r
ed
to
th
e
s
a
m
e
lan
d
co
v
er
class
.
T
h
ese
f
ac
to
r
s
ca
n
s
i
g
n
i
f
ica
n
tl
y
e
n
h
an
ce
class
i
f
icatio
n
ac
cu
r
ac
y
.
I
n
2
0
2
2
,
as
s
h
o
w
n
i
n
T
ab
le
1
,
r
ai
n
f
a
ll
in
T
an
g
g
a
m
u
s
R
e
g
e
n
c
y
w
as
1
5
4
m
m
p
er
m
o
n
t
h
,
a
m
o
d
er
ate
r
ain
f
all
ca
t
eg
o
r
y
.
A
w
ei
g
h
t
o
f
0
in
d
icate
s
th
at
co
f
f
ee
f
ar
m
er
s
in
T
an
g
g
a
m
u
s
R
e
g
en
c
y
te
n
d
n
o
t
to
r
ely
o
n
r
iv
er
w
ater
b
u
t
in
s
tead
o
n
r
ain
f
all
[
9
]
.
A
v
ar
iab
le
w
ith
a
v
al
u
e
o
f
1
in
d
icate
s
th
at
th
e
v
ar
iab
le
p
lay
s
a
d
o
m
i
n
a
n
t
r
o
le
in
th
e
a
lg
o
r
ith
m
.
C
o
n
v
er
s
el
y
,
a
v
al
u
e
clo
s
e
to
0
in
d
icate
s
th
at
th
e
a
ttrib
u
te
is
n
o
t
v
er
y
s
ig
n
i
f
ica
n
t
i
n
r
ed
u
ci
n
g
e
n
tr
o
p
y
.
A
v
a
lu
e
o
f
ze
r
o
in
d
icate
s
it
s
lo
w
e
s
t
i
m
p
o
r
tan
ce
a
n
d
th
u
s
is
s
u
b
o
r
d
in
ate
.
T
h
is
s
tu
d
y
f
o
u
n
d
i
m
p
o
r
tan
t
v
ar
ia
b
les
th
at
ca
n
s
ig
n
i
f
ica
n
tl
y
i
m
p
r
o
v
e
class
i
f
icatio
n
ac
cu
r
ac
y
.
I
n
2
0
2
2
,
it
w
a
s
r
ep
o
r
ted
th
at
r
ain
f
all
i
n
T
an
g
g
a
m
u
s
R
eg
e
n
c
y
w
as 1
5
4
m
m
p
er
m
o
n
t
h
,
w
h
ich
f
all
s
w
it
h
i
n
t
h
e
m
o
d
er
ate
r
ain
f
all
ca
teg
o
r
y
.
A
w
ei
g
h
t
o
f
0
in
d
ic
ates
th
at
co
f
f
ee
f
ar
m
er
s
i
n
T
a
n
g
g
a
m
u
s
R
e
g
en
c
y
te
n
d
n
o
t
to
r
ely
o
n
r
iv
er
w
ater
b
u
t o
n
r
ain
f
all.
T
ab
le
1
.
W
eig
h
t r
esu
lt
s
f
o
r
ea
ch
v
ar
iab
le
(
o
v
er
f
i
tti
n
g
)
B
r
u
t
e
f
o
r
c
e
I
n
f
o
r
mat
i
o
n
g
a
i
n
G
a
i
n
r
a
t
i
o
G
i
n
i
i
n
d
ex
V
a
r
i
a
b
l
e
W
e
i
g
h
t
V
a
r
i
a
b
l
e
W
e
i
g
h
t
V
a
r
i
a
b
l
e
W
e
i
g
h
t
V
a
r
i
a
b
l
e
W
e
i
g
h
t
L
a
n
d
c
o
v
e
r
1
L
a
n
d
c
o
v
e
r
1
S
e
t
t
l
e
me
n
t
1
L
a
n
d
c
o
v
e
r
1
El
e
v
a
t
i
o
n
1
N
D
V
I
0
.
3
7
N
D
V
I
0
.
9
0
S
e
t
t
l
e
me
n
t
0
.
3
2
S
l
o
p
e
1
N
R
G
I
0
.
3
4
L
a
n
d
c
o
v
e
r
0
.
8
2
N
D
V
I
0
.
2
2
R
i
v
e
r
1
V
A
R
I
0
.
3
3
N
R
G
I
0
.
7
2
N
R
G
I
0
.
2
0
R
o
a
d
1
S
e
t
t
l
e
me
n
t
0
.
2
9
V
A
R
I
0
.
7
0
V
A
R
I
0
.
2
0
S
e
t
t
l
e
me
n
t
1
S
l
o
p
e
0
.
2
0
S
l
o
p
e
0
.
3
3
R
o
a
d
0
.
1
6
N
D
V
I
1
V
D
V
I
0
.
1
9
R
o
a
d
0
.
3
2
S
l
o
p
e
0
.
1
3
N
R
G
I
1
El
e
v
a
t
i
o
n
0
.
1
7
V
D
V
I
0
.
3
1
V
D
V
I
0
.
1
3
V
D
V
I
1
R
o
a
d
0
.
1
3
El
e
v
a
t
i
o
n
0
.
3
0
El
e
v
a
t
i
o
n
0
.
1
0
V
A
R
I
1
R
i
v
e
r
0
R
i
v
e
r
0
R
i
v
e
r
0
B
as
ed
o
n
th
is
i
n
f
o
r
m
atio
n
,
v
ar
iab
les
s
u
c
h
as
s
et
tle
m
en
t
s
ex
h
ib
it
i
n
co
n
s
is
ten
t
w
e
ig
h
t
s
eq
u
en
ce
s
,
in
d
icati
n
g
o
v
er
f
it
tin
g
,
as
th
e
y
y
ield
m
o
r
e
b
r
an
ch
n
o
d
es
th
a
n
s
p
ec
tr
al
v
ar
iab
les,
l
ea
d
in
g
to
il
lo
g
ica
l
class
i
f
icatio
n
r
esu
l
ts
.
Ov
er
f
itt
i
n
g
i
s
a
co
m
m
o
n
p
r
o
b
lem
i
n
ma
ch
in
e
lear
n
in
g
,
esp
ec
iall
y
w
it
h
d
ec
is
io
n
tr
ee
s
,
an
d
ca
n
n
o
t
b
e
a
v
o
id
ed
en
tire
l
y
[
9
]
.
T
h
is
ca
n
b
e
d
u
e
to
li
m
i
tatio
n
s
in
tr
ain
in
g
d
ata,
w
h
ic
h
ca
n
b
e
l
i
m
i
ted
in
s
ize
o
r
en
co
m
p
a
s
s
m
a
n
y
v
ar
ia
b
les,
o
r
to
li
m
itatio
n
s
in
al
g
o
r
ith
m
s
th
at
ar
e
to
o
co
m
p
le
x
a
n
d
r
eq
u
ir
e
to
o
m
a
n
y
p
ar
am
eter
s
.
Var
iab
les
s
u
ch
a
s
r
o
ad
p
r
o
x
im
it
y
,
s
et
tle
m
en
t
s
,
r
iv
er
s
,
an
d
s
lo
p
es
w
er
e
r
e
m
o
v
ed
u
s
in
g
t
h
e
d
r
o
p
o
u
t
m
et
h
o
d
as
in
T
ab
le
2
.
T
h
is
m
et
h
o
d
s
ig
n
i
f
ican
t
l
y
r
ed
u
ce
s
co
m
p
u
tatio
n
al
ef
f
o
r
t
an
d
is
s
u
itab
le
f
o
r
lar
g
e
an
d
co
m
p
le
x
d
ata
s
et
s
[
9
]
.
L
an
d
co
v
er
r
ec
eiv
ed
th
e
h
ig
h
es
t
weig
h
tin
g
f
o
r
al
m
o
s
t
e
v
er
y
p
ar
a
m
eter
.
L
a
n
d
co
v
er
ca
r
r
ies
a
h
ig
h
w
e
ig
h
ti
n
g
b
ec
au
s
e
it
r
e
f
lects
th
e
s
o
cial
a
n
d
n
atu
r
al
co
n
d
itio
n
s
o
f
a
r
e
g
io
n
an
d
p
la
y
s
a
cr
u
cia
l
r
o
le
in
u
n
d
er
s
tan
d
i
n
g
th
e
co
m
p
lex
r
elat
io
n
s
h
ip
s
b
et
w
ee
n
ac
tiv
itie
s
a
n
d
ch
a
n
g
e
s
o
cc
u
r
r
in
g
o
n
t
h
e
E
ar
th
’
s
s
u
r
f
ac
e.
T
h
e
r
ef
o
r
e,
it
h
as
h
i
g
h
s
e
n
s
i
tiv
it
y
a
n
d
is
th
e
m
o
s
t
in
f
lu
e
n
tia
l
in
clas
s
if
y
i
n
g
lan
d
co
v
er
[
1
1
]
.
T
h
e
s
p
ec
tr
al
v
ar
iab
le
w
it
h
th
e
h
i
g
h
est
w
ei
g
h
ti
n
g
a
m
o
n
g
th
e
b
est
f
ea
t
u
r
es
w
a
s
NDVI
,
w
it
h
a
g
ai
n
r
atio
o
f
1
,
a
Gin
i
in
d
ex
o
f
0
.
1
1
,
an
d
an
in
f
o
r
m
a
t
io
n
g
ain
o
f
0
.
2
5
(
s
ee
T
a
b
le
2
)
.
T
h
is
in
d
icate
s
t
h
at
th
e
NDVI
v
eg
eta
tio
n
in
d
e
x
is
h
ig
h
l
y
r
eli
ab
le
i
n
lan
d
co
v
e
r
class
i
f
icatio
n
.
N
DVI
r
ef
lec
ts
ec
o
lo
g
ical
i
n
f
o
r
m
a
tio
n
,
i
n
clu
d
in
g
v
eg
eta
tio
n
g
r
o
w
t
h
,
v
e
g
etat
io
n
co
v
er
lev
e
l
,
an
d
b
io
m
a
s
s
[
2
2
]
.
T
ab
le
2
.
W
eig
h
t r
esu
lt
f
o
r
ea
ch
v
ar
iab
le
(
b
est)
B
r
u
t
e
f
o
r
c
e
I
n
f
o
r
mat
i
o
n
g
a
i
n
G
a
i
n
r
a
t
i
o
G
i
n
i
i
n
d
e
x
V
a
r
i
a
b
l
e
W
e
i
g
h
t
V
a
r
i
a
b
l
e
W
e
i
g
h
t
V
a
r
i
a
b
l
e
W
e
i
g
h
t
V
a
r
i
a
b
l
e
W
e
i
g
h
t
L
a
n
d
c
o
v
e
r
1
L
a
n
d
c
o
v
e
r
1
N
D
V
I
1
L
a
n
d
c
o
v
e
r
1
El
e
v
a
t
i
o
n
1
N
D
V
I
0
.
2
5
L
a
n
d
c
o
v
e
r
0
.
9
7
N
D
V
I
0
.
1
1
N
D
V
I
1
V
A
R
I
0
.
1
4
V
A
R
I
0
.
5
9
V
A
R
I
0
.
0
7
N
R
G
I
1
N
R
G
I
0
.
1
3
N
R
G
I
0
.
5
7
N
R
G
I
0
.
0
6
V
D
V
I
1
El
e
v
a
t
i
o
n
0
.
0
0
El
e
v
a
t
i
o
n
0
.
0
5
V
D
V
I
0
.
0
2
V
A
R
I
1
V
D
V
I
0
V
D
V
I
0
El
e
v
a
t
i
o
n
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
16
93
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
4
,
A
u
g
u
s
t
20
26
:
1
3
0
7
-
1
3
1
9
1312
3
.
2
.
M
o
del
o
pti
m
iza
t
io
n
P
ar
am
eter
tu
n
i
n
g
in
d
ec
is
io
n
tr
ee
s
u
s
es
g
r
id
s
ea
r
ch
,
a
s
y
s
te
m
atic
p
r
o
ce
s
s
th
at
tes
ts
v
ar
io
u
s
co
m
b
i
n
atio
n
s
o
f
p
ar
a
m
eter
v
a
lu
es
to
au
to
m
atica
ll
y
f
in
d
t
h
e
o
p
tim
a
l
o
n
e
[
2
3
]
.
M
o
d
e
l
(
g
r
i
d
)
o
p
t
im
iz
a
t
i
o
n
is
an
o
p
e
r
a
t
i
o
n
th
at
f
in
d
s
th
e
o
p
tim
al
v
al
u
e
f
o
r
e
ac
h
s
e
le
c
t
e
d
p
a
r
am
e
t
e
r
,
th
e
r
e
b
y
d
et
e
r
m
in
in
g
th
e
v
a
l
u
e
th
a
t
y
i
el
d
s
th
e
m
ax
im
u
m
r
e
s
u
l
ts
w
i
th
o
u
t
h
av
in
g
t
o
r
e
-
en
t
e
r
d
a
t
a
.
T
a
b
l
e
3
s
h
o
w
s
th
a
t
th
e
in
f
o
r
m
a
t
i
o
n
g
a
in
p
a
r
am
et
e
r
a
c
h
i
ev
e
d
t
h
e
h
ig
h
es
t
a
c
cu
r
a
cy
o
f
9
4
.
0
%
w
ith
t
h
e
t
r
e
a
tm
en
t
w
i
th
o
u
t
p
r
u
n
in
g
a
n
d
w
ith
o
u
t
p
r
e
-
p
r
u
n
in
g
,
a
m
ax
im
u
m
d
e
p
th
o
f
3
5
,
a
m
in
im
u
m
l
ea
f
s
i
ze
o
f
6
0
,
a
m
in
im
u
m
s
p
l
it
s
i
z
e
o
f
4
8
,
a
n
d
a
p
r
e
-
p
r
u
n
i
n
g
al
t
e
r
n
a
t
iv
e
o
f
6
0
.
T
h
e
r
esu
lts
o
f
m
o
d
el
o
p
tim
iz
atio
n
u
s
i
n
g
th
e
in
f
o
r
m
atio
n
g
ain
p
ar
am
e
ter
d
o
m
i
n
ated
th
e
p
ar
am
eter
s
p
ac
e,
w
it
h
t
h
e
1
0
b
est
m
o
d
el
p
ar
am
eter
co
m
b
i
n
atio
n
s
o
u
t
o
f
1
0
,
3
6
8
.
B
ased
o
n
th
e
m
o
d
el
o
p
ti
m
izatio
n
,
t
h
e
h
ig
h
e
s
t
ac
cu
r
ac
y
w
as
9
4
.
0
%,
d
em
o
n
s
tr
ati
n
g
th
e
d
ec
i
s
io
n
tr
ee
m
o
d
el
’
s
ab
ilit
y
to
clas
s
i
f
y
ag
r
o
f
o
r
estr
y
co
f
f
ee
w
it
h
h
i
g
h
ac
cu
r
ac
y
.
In
f
o
r
m
at
io
n
g
a
in
p
r
o
v
e
d
t
o
b
e
t
h
e
m
o
s
t
e
f
f
e
ct
iv
e
s
e
p
a
r
a
ti
o
n
p
a
r
am
e
te
r
i
n
th
is
c
as
e
,
y
ie
l
d
in
g
c
o
n
s
i
s
t
en
t
ly
h
ig
h
a
c
cu
r
a
cy
.
I
n
f
o
r
m
a
ti
o
n
g
ai
n
i
s
a
s
u
it
a
b
l
e
m
e
tr
i
c
f
o
r
s
e
l
e
ct
in
g
d
a
t
a
s
p
l
it
s
[
5
]
,
[
1
6
]
,
[
2
3
]
.
T
ab
le
3
.
1
0
h
ig
h
e
s
t a
cc
u
r
ac
y
p
ar
a
m
eter
co
m
b
in
at
io
n
P
a
r
a
me
t
e
r
P
r
u
n
n
i
ng
M
a
x
i
m
a
l
d
e
p
t
h
Pre
-
p
r
u
n
n
i
n
g
M
i
n
i
m
a
l
l
e
a
f
si
z
e
M
i
n
i
m
a
l
si
z
e
f
o
r
sp
l
i
t
Pre
-
p
r
u
n
n
i
n
g
a
l
t
e
r
n
a
t
i
v
e
A
c
c
u
r
a
c
y
(
%)
IG
F
35
F
60
48
60
9
4
.
0
IG
F
24
F
80
80
0
9
3
.
9
IG
T
35
F
1
64
1
0
0
9
3
.
7
IG
T
24
F
21
17
40
9
3
.
7
IG
F
35
F
41
1
40
9
3
.
6
GI
F
35
F
21
80
40
9
3
.
6
IG
F
24
F
21
17
80
9
3
.
6
IG
T
30
F
80
48
1
0
0
9
3
.
6
IG
F
35
F
41
1
20
9
3
.
5
IG
F
19
F
80
48
20
9
3
.
5
*
N
o
t
e
:
G
I
=
g
i
n
i
i
n
d
e
x
,
I
G
=
i
n
f
o
r
mat
i
o
n
g
a
i
n
,
F
=
f
a
l
se
,
T
=
t
r
u
e
T
h
e
ac
cu
r
ac
y
-
clas
s
f
r
eq
u
en
c
y
r
esu
lt
s
f
o
r
th
e
d
ec
is
io
n
tr
ee
p
ar
a
m
eter
s
i
n
d
icate
t
h
at
i
n
f
o
r
m
atio
n
g
ai
n
w
a
s
s
elec
ted
a
s
t
h
e
b
est
p
ar
a
m
eter
f
o
r
b
u
ild
in
g
a
d
ec
is
io
n
tr
ee
m
o
d
el,
w
it
h
1
0
3
6
8
co
m
b
in
atio
n
s
ac
h
iev
i
n
g
90%
-
9
5
%
ac
cu
r
ac
y
,
a
s
s
h
o
w
n
in
T
ab
le
4
.
Gain
r
atio
h
as
m
o
r
e
co
m
b
in
a
tio
n
s
w
it
h
ac
c
u
r
ac
y
b
elo
w
7
0
%
th
a
n
o
th
er
p
ar
am
eter
s
,
n
a
m
el
y
1
3
7
9
.
T
h
i
s
s
h
o
w
s
t
h
at
th
e
g
ai
n
r
atio
d
o
es
n
o
t
p
er
f
o
r
m
well
in
d
ec
is
io
n
-
tr
ee
class
i
f
icatio
n
.
Si
m
i
lar
r
esu
lts
w
er
e
r
ep
o
r
ted
b
y
[
4
]
,
w
h
o
id
en
ti
f
ied
in
f
o
r
m
atio
n
g
ain
as
o
n
e
o
f
th
e
1
0
b
est
p
ar
am
eter
s
in
t
h
e
o
p
ti
m
iza
tio
n
m
o
d
el,
ac
h
ie
v
i
n
g
a
n
ac
cu
r
a
c
y
o
f
8
4
.
6
%.
T
h
e
g
in
i
in
d
e
x
w
a
s
also
s
elec
ted
,
ac
h
iev
in
g
th
e
h
i
g
h
est
ac
cu
r
ac
y
o
f
8
4
.
2
%.
T
h
e
r
o
les
o
f
th
e
g
in
i
in
d
ex
an
d
in
f
o
r
m
atio
n
g
a
in
w
er
e
f
r
eq
u
e
n
tl
y
r
ec
o
g
n
ized
in
t
h
e
u
s
e
o
f
t
h
e
d
ec
is
io
n
tr
ee
alg
o
r
it
h
m
[
1
8
]
.
T
ab
le
4
.
C
lass
ac
cu
r
ac
y
f
r
eq
u
en
c
y
f
o
r
ea
ch
d
ec
is
io
n
tr
ee
p
ar
a
m
eter
C
l
a
ss
a
c
c
u
r
a
c
y
P
a
r
a
me
t
e
r
T
o
t
a
l
I
n
f
o
r
mat
i
o
n
g
a
i
n
G
a
i
n
r
a
t
i
o
G
i
n
i
i
n
d
e
x
90
−
95%
1
7
2
7
1
6
9
7
1
7
0
9
5
1
3
3
85
−
90%
1
3
2
6
19
3
4
6
80
−
85%
0
26
0
26
75
−
80%
1
7
2
2
19
1
7
2
8
3
4
6
9
70
−
75%
6
9
0
15
<
7
0
%
0
1
3
7
9
0
1
3
7
9
T
o
t
a
l
3
4
5
6
3
4
5
6
3
4
5
6
1
0
3
6
8
3
.
3
.
Acc
ura
cy
a
s
s
ess
m
e
nt
T
h
e
ac
cu
r
ac
y
o
f
tr
ai
n
in
g
-
ar
e
a
d
ata
in
th
e
d
ec
i
s
io
n
-
tr
ee
cl
ass
i
f
icatio
n
a
l
g
o
r
ith
m
i
s
as
s
e
s
s
ed
u
s
i
n
g
p
r
o
d
u
ce
r
,
u
s
er
,
to
tal,
an
d
k
ap
p
a
ac
cu
r
ac
ies.
P
r
o
d
u
ce
r
ac
c
u
r
ac
y
a
n
d
u
s
er
ac
cu
r
ac
y
ar
e
co
m
p
o
n
e
n
t
s
o
f
t
h
e
o
v
er
all
ac
cu
r
ac
y
e
s
ti
m
ate
t
h
a
t
d
escr
ib
e
th
e
to
tal
ac
cu
r
ac
y
o
f
th
e
cla
s
s
i
f
icat
io
n
r
es
u
lts
;
k
ap
p
a
ac
cu
r
ac
y
,
i
n
co
n
tr
ast,
is
d
ete
r
m
in
ed
n
o
t
o
n
l
y
b
y
co
r
r
ec
tl
y
clas
s
i
f
ied
o
b
j
ec
ts
b
u
t
also
b
y
clas
s
i
f
icatio
n
er
r
o
r
s
[
1
9
]
.
T
h
r
ee
d
if
f
er
e
n
t
co
m
b
in
a
tio
n
s
u
s
ed
in
th
e
test
ar
e
:
(
i
)
s
p
ec
tr
a
l,
b
io
-
g
eo
p
h
y
s
ical,
(
ii)
s
p
ec
tr
al,
an
d
(
iii)
b
io
-
g
eo
p
h
y
s
ica
l.
T
h
e
ac
cu
r
ac
y
co
m
p
ar
i
s
o
n
f
o
r
ea
ch
co
m
b
in
at
i
o
n
is
s
h
o
w
n
i
n
Fig
u
r
e
3
.
Data
in
th
e
f
o
r
m
o
f
a
co
n
f
u
s
io
n
m
atr
ix
ca
n
n
o
t
b
e
as
s
ess
ed
s
o
lel
y
b
ased
o
n
n
u
m
b
e
r
s
an
d
e
x
is
ti
n
g
ac
c
u
r
ac
y
.
Ho
wev
er
,
o
v
er
f
i
tti
n
g
is
ev
id
en
t i
n
t
h
e
m
o
d
el
’
s
v
is
u
aliz
atio
n
s
[
8
]
,
[
9
]
,
[
2
4
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Dec
is
io
n
-
tr
ee
-
b
a
s
ed
ma
ch
in
e
lea
r
n
in
g
fo
r
d
etec
tin
g
c
o
ffee
a
g
r
o
fo
r
estr
y
…
(
I
Ma
d
e
K
h
r
is
n
a
Yo
g
a
Dev
a
n
d
r
a
)
1313
Fig
u
r
e
3
.
T
h
e
ac
cu
r
ac
y
d
i
f
f
er
e
n
ce
b
et
w
ee
n
ea
c
h
v
ar
iab
le
co
m
b
in
at
io
n
T
h
e
b
est
ac
cu
r
ac
y
w
a
s
ac
h
ie
v
ed
u
s
in
g
a
co
m
b
in
a
tio
n
o
f
s
p
e
ctr
al
an
d
b
io
p
h
y
s
ical
v
ar
iab
les,
w
i
th
a
n
o
v
er
all
ac
cu
r
ac
y
o
f
9
4
.
0
%
an
d
a
k
ap
p
a
o
f
9
3
.
3
%.
T
h
e
lo
w
est
ac
cu
r
ac
y
w
a
s
ac
h
iev
ed
w
ith
o
n
l
y
s
p
ec
tr
al
v
ar
iab
les,
w
it
h
a
n
o
v
er
all
ac
c
u
r
ac
y
o
f
6
0
.
4
%
an
d
a
k
ap
p
a
o
f
5
5
.
4
%.
Sp
ec
tr
al
v
ar
iab
les
al
o
n
e
ar
e
in
s
u
f
f
icie
n
t
to
class
if
y
ag
r
o
f
o
r
estr
y
co
f
f
e
e
u
n
d
er
a
tr
ee
ca
n
o
p
y
.
S
h
ad
i
n
g
ca
n
in
f
l
u
en
ce
t
h
e
v
e
g
etati
o
n
in
d
ex
v
alu
e
i
n
co
f
f
ee
f
ield
s
,
s
o
th
e
v
al
u
e
o
b
tain
ed
d
o
es
n
o
t
r
ep
r
esen
t
th
e
e
n
tire
co
f
f
ee
p
la
n
t
v
e
g
etatio
n
in
d
ex
b
u
t
r
ath
er
a
m
i
x
tu
r
e
o
f
co
f
f
ee
p
la
n
t a
n
d
s
h
ad
e
tr
ee
v
eg
e
tatio
n
i
n
d
ices
[
1
4
]
,
[
1
5
]
,
[
2
0
]
,
[
2
5
]
.
T
h
e
o
v
er
all
ac
cu
r
ac
y
v
al
u
e
o
f
ea
ch
lan
d
co
v
er
class
i
s
a
g
o
o
d
v
alu
e
(
s
ee
T
ab
le
5
)
,
w
h
ic
h
is
in
li
n
e
w
it
h
[
2
4
]
,
w
h
ich
s
tate
s
th
at
a
c
cu
r
ac
y
v
al
u
es
i
n
th
e
r
an
g
e
o
f
7
8
.
8
%
–
1
0
0
%
ar
e
in
clu
d
ed
in
th
e
m
ed
i
u
m
-
h
i
g
h
ca
teg
o
r
y
,
m
ea
n
in
g
t
h
e
ac
cu
r
a
c
y
o
f
th
e
class
i
f
icatio
n
r
esu
lt
s
f
o
r
all
lan
d
c
o
v
er
class
es
p
r
o
d
u
ce
d
ca
n
b
e
f
u
ll
y
tr
u
s
ted
.
C
o
f
f
ee
ag
r
o
f
o
r
estr
y
h
as
t
h
e
lo
w
e
s
t
u
s
er
ac
c
u
r
ac
y
,
at
8
3
.
6
%,
in
d
i
ca
tin
g
a
n
8
3
.
6
%
ch
a
n
ce
t
h
at
t
h
e
tr
ain
i
n
g
p
i
x
el
s
clas
s
i
f
ied
r
ep
r
esen
t
a
g
r
o
f
o
r
estr
y
co
f
f
ee
.
Ho
wev
er
,
d
r
y
lan
d
a
g
r
ic
u
ltu
r
e
h
a
s
t
h
e
lo
w
est
p
r
o
d
u
ce
r
ac
cu
r
ac
y
o
f
8
9
.
2
%.
T
h
is
i
n
d
i
ca
tes
t
h
at
o
n
l
y
8
9
.
2
%
o
f
t
h
e
d
r
y
la
n
d
a
g
r
icu
lt
u
r
al
cla
s
s
tr
ai
n
in
g
d
ata
m
a
tch
e
s
f
ield
c
o
n
d
itio
n
s
.
T
h
e
s
tu
d
y
al
s
o
s
h
o
w
ed
t
h
at
t
h
e
F1
-
s
co
r
e
f
o
r
ea
ch
class
e
x
ce
ed
ed
8
5
%.
T
ab
le
5
.
T
h
e
ac
cu
r
ac
y
r
es
u
lt
s
f
r
o
m
ea
ch
v
ar
iab
le
L
a
n
d
c
o
v
e
r
c
l
a
ss
U
se
r
a
c
c
u
r
a
c
y
(
r
e
c
a
l
l
)
P
r
o
d
u
c
e
r
a
c
c
u
r
a
c
y
(
p
r
e
c
i
si
o
n
)
F
1
s
c
o
r
e
O
v
e
r
a
l
l
a
c
c
u
r
a
c
y
K
a
p
p
a
a
c
c
u
r
a
c
y
C
l
o
u
d
9
6
.
8
%
9
7
.
7
%
9
7
.
3
%
9
4
.
0
%
9
3
.
3
%
C
o
f
f
e
e
a
g
r
o
f
o
r
e
st
r
y
8
3
.
6
%
8
9
.
5
%
8
6
.
5
%
F
o
r
e
st
9
5
.
1
%
9
6
.
1
%
9
5
.
6
%
P
l
a
n
t
a
t
i
o
n
(
e
st
a
t
e
c
r
o
p
)
9
9
.
0
%
9
8
.
1
%
9
8
.
6
%
A
g
r
i
c
u
l
t
u
r
e
(
d
r
y
l
a
n
d
a
g
r
i
c
u
l
t
u
r
e
)
9
2
.
3
%
8
9
.
2
%
9
0
.
7
%
S
e
t
t
l
e
me
n
t
9
7
.
6
%
9
5
.
3
%
9
6
.
4
%
S
c
r
u
b
9
2
.
7
%
8
9
.
8
%
9
1
.
2
%
P
a
d
d
y
f
i
e
l
d
9
4
.
4
%
9
9
.
2
%
9
6
.
7
%
W
a
t
e
r
b
o
d
y
9
9
.
2
%
9
6
.
7
%
9
7
.
9
%
B
a
r
e
l
a
n
d
9
7
.
8
%
9
7
.
8
%
9
7
.
8
%
A
v
e
r
a
g
e
9
4
.
8
%
9
4
.
9
%
9
4
.
9
%
Si
m
i
lar
r
esu
lt
s
w
er
e
also
f
o
u
n
d
in
[
2
6
]
,
w
h
o
o
b
tain
ed
p
r
o
d
u
ce
r
ac
cu
r
ac
y
f
o
r
o
n
e
class
th
at
d
id
n
o
t
r
ea
ch
8
5
%,
b
u
t
ac
h
iev
ed
an
o
v
er
all
ac
cu
r
ac
y
o
f
9
4
.
7
%
an
d
a
k
ap
p
a
c
o
ef
f
icie
n
t
o
f
0
.
9
,
in
d
i
ca
tin
g
th
at
th
e
lev
e
l
o
f
class
i
f
icat
io
n
s
u
itab
ili
t
y
i
s
v
er
y
h
ig
h
co
m
p
ar
ed
to
r
an
d
o
m
clas
s
i
f
icatio
n
.
T
h
e
m
o
s
t
co
m
m
o
n
clas
s
i
f
ic
atio
n
er
r
o
r
s
f
o
u
n
d
in
th
e
d
r
y
la
n
d
a
g
r
icu
l
tu
r
al
c
lass
ar
e
ca
u
s
ed
b
y
t
h
e
lo
ca
tio
n
o
f
d
r
y
la
n
d
ag
r
icu
lt
u
r
e,
esp
ec
iall
y
th
o
s
e
lo
ca
ted
o
r
ass
o
ciate
d
w
it
h
ag
r
o
f
o
r
estr
y
co
f
f
ee
a
n
d
s
h
r
u
b
class
es,
s
o
t
h
at
t
h
e
y
m
a
y
h
a
v
e
v
e
g
etatio
n
i
n
d
ex
v
alu
e
s
t
h
at
ar
e
n
o
t
m
u
c
h
d
if
f
e
r
en
t.
T
h
e
d
ec
is
io
n
tr
ee
alg
o
r
ith
m
t
en
d
s
to
ac
h
iev
e
h
i
g
h
er
ac
cu
r
ac
y
.
T
h
is
is
in
lin
e
w
it
h
th
e
r
esu
lt
s
o
f
r
esear
ch
b
y
[
6
]
,
w
h
ic
h
s
h
o
w
s
th
e
s
u
p
er
io
r
it
y
o
f
r
an
d
o
m
f
o
r
est
(
an
en
s
e
m
b
le
o
f
d
ec
is
io
n
-
tr
ee
alg
o
r
ith
m
s
)
o
v
er
SVM
in
cla
s
s
i
f
y
i
n
g
u
r
b
an
an
d
f
o
r
est
ar
ea
s
[
4
]
,
[
5
]
,
[
2
7
]
.
I
n
id
en
ti
f
y
i
n
g
co
n
i
f
er
o
u
s
f
o
r
es
t
class
es,
t
h
e
SV
M
alg
o
r
ith
m
ac
h
iev
e
s
9
5
.
8
%
ac
cu
r
ac
y
,
b
u
t
ac
c
u
r
ac
y
i
s
lo
w
e
r
f
o
r
o
th
er
f
o
r
est
t
y
p
e
s
,
r
ea
ch
in
g
o
n
l
y
8
8
.
3
%.
A
lt
h
o
u
g
h
M
L
a
lg
o
r
it
h
m
s
o
f
te
n
p
r
o
m
is
e
b
etter
ac
cu
r
ac
y
,
c
h
a
llen
g
es
r
e
m
ai
n
,
esp
ec
iall
y
in
d
iv
er
s
e
f
o
r
est
t
y
p
es,
p
ar
ticu
lar
l
y
i
n
m
ix
ed
v
e
g
etati
o
n
en
v
ir
o
n
m
e
n
ts
,
w
h
er
e
s
p
ec
tr
al
v
ar
iatio
n
is
v
er
y
lo
w
.
T
h
is
asp
ec
t
en
co
u
r
ag
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
16
93
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
4
,
A
u
g
u
s
t
20
26
:
1
3
0
7
-
1
3
1
9
1314
r
esear
ch
er
s
to
co
n
tin
u
e
s
t
u
d
y
i
n
g
to
s
elec
t
t
h
e
o
p
ti
m
al
al
g
o
r
ith
m
an
d
s
elec
t
t
h
e
m
o
s
t
co
n
s
is
ten
t
attr
ib
u
tes
to
ac
h
iev
e
h
i
g
h
ac
c
u
r
ac
y
.
T
h
e
r
esu
lti
n
g
d
ec
i
s
io
n
tr
ee
m
o
d
el
ca
n
class
i
f
y
a
g
r
o
f
o
r
estr
y
co
f
f
ee
f
r
o
m
o
t
h
er
la
n
d
co
v
er
t
y
p
e
s
,
w
it
h
a
r
o
o
t
n
o
d
e
r
ep
r
esen
tin
g
v
is
u
a
l
lan
d
co
v
er
in
Fi
g
u
r
e
4
,
E
L
E
V
=
elev
atio
n
,
P
L
T
=
p
lan
tatio
n
(
estate
cr
o
p
)
,
an
d
P
L
VI
S
=
v
is
u
al
lan
d
co
v
er
.
T
h
eo
r
etica
lly
,
it
in
d
icate
s
th
a
t
ag
r
o
f
o
r
estr
y
c
o
f
f
ee
p
lan
t
s
ar
e
f
o
u
n
d
in
ce
r
tain
lan
d
-
u
s
e
t
y
p
e
s
,
s
u
c
h
as
d
r
y
la
n
d
ag
r
icu
lt
u
r
e,
p
lan
tatio
n
f
o
r
ests
,
an
d
n
at
u
r
al
f
o
r
ests
.
A
g
r
o
f
o
r
estr
y
p
lan
ts
ar
e
g
en
er
all
y
p
lan
ted
in
d
r
y
ar
ea
s
b
ec
au
s
e
co
f
f
ee
p
lan
ts
ar
e
in
to
ler
an
t
o
f
w
ater
lo
g
g
i
n
g
[
1
]
,
[
3
]
.
Fu
r
th
er
m
o
r
e,
co
f
f
ee
p
la
n
ts
ar
e
g
r
o
w
n
as
a
s
h
o
r
t
-
ter
m
s
a
v
in
g
s
cr
o
p
f
o
r
f
ar
m
er
s
,
w
i
th
p
r
o
d
u
ctiv
e
h
ar
v
e
s
ts
o
cc
u
r
r
in
g
ap
p
r
o
x
im
a
tel
y
ev
er
y
t
w
o
m
o
n
th
s
.
C
o
f
f
ee
p
lan
ts
r
eq
u
ir
e
g
o
o
d
d
r
ain
ag
e,
an
d
th
is
is
o
f
te
n
f
o
u
n
d
in
d
r
y
lan
d
ag
r
icu
l
tu
r
e.
Fu
r
t
h
er
m
o
r
e,
co
f
f
ee
p
lan
ts
ar
e
a
m
o
n
g
th
e
m
o
s
t
s
tu
d
ied
p
lan
ts
.
C
o
f
f
ee
p
la
n
ts
ty
p
ical
l
y
t
h
r
iv
e
o
n
m
o
u
n
tai
n
s
lo
p
es a
n
d
f
o
r
ested
ar
ea
s
.
B
y
n
atu
r
e,
co
f
f
ee
is
a
d
r
y
lan
d
p
lan
t.
Fig
u
r
e
4
.
Dec
is
io
n
tr
ee
m
o
d
el
u
s
ed
f
o
r
lan
d
co
v
er
clas
s
i
f
icati
o
n
Fig
u
r
e
4
s
h
o
w
s
t
h
at
t
h
e
d
ec
is
io
n
tr
ee
m
o
d
el
r
elies
o
n
a
c
o
m
b
i
n
atio
n
o
f
s
p
ec
tr
al
an
d
g
eo
p
h
y
s
ica
l
v
ar
iab
les,
w
h
ic
h
ar
e
cr
u
cia
l
f
o
r
ac
cu
r
atel
y
d
is
t
in
g
u
is
h
i
n
g
lan
d
co
v
er
cla
s
s
e
s
.
T
h
e
d
ec
is
io
n
tr
ee
d
ia
g
r
a
m
p
r
o
v
id
es
co
n
tex
t
f
o
r
u
n
d
er
s
t
an
d
in
g
h
o
w
v
ar
io
u
s
e
n
v
ir
o
n
m
en
tal
p
r
ed
icto
r
s
in
ter
ac
t
an
d
ar
e
s
eq
u
en
tiall
y
s
elec
ted
b
y
t
h
e
M
L
alg
o
r
it
h
m
to
d
is
tin
g
u
is
h
d
if
f
er
en
t
lan
d
c
o
v
er
t
y
p
es.
I
n
th
i
s
m
o
d
el,
a
g
r
o
f
o
r
estr
y
co
f
f
ee
w
as
s
elec
ted
b
ased
o
n
V
DVI
≤
0
.
1
0
6
,
NR
GI
≤
0
.
1
7
3
,
E
lev
atio
n
≤
2
1
0
,
an
d
NDVI
b
et
w
ee
n
0
.
7
1
9
an
d
0
.
7
3
7
[
3
]
s
tated
th
a
t
r
o
b
u
s
ta
co
f
f
ee
ca
n
b
e
p
lan
ted
in
lo
w
lan
d
ar
ea
s
w
i
th
a
n
ele
v
atio
n
o
f
le
s
s
th
a
n
8
0
0
m
eter
s
ab
o
v
e
s
ea
lev
el.
NDVI
v
alu
e
f
o
r
co
f
f
ee
a
g
r
o
f
o
r
estr
y
is
also
s
elec
ted
ar
o
u
n
d
0
.
7
,
in
d
icatin
g
th
e
c
o
f
f
ee
p
lan
tatio
n
is
co
v
er
ed
w
ith
tr
ee
s
h
ad
e,
ca
u
s
in
g
th
e
h
ig
h
NDVI
v
al
u
e
s
i
m
ilar
to
[
4
]
w
h
o
s
tate
s
th
at
t
h
e
NDVI
v
al
u
e
f
o
r
T
r
o
p
ical
R
ain
f
o
r
est is
ab
o
v
e
0
.
5
an
d
clo
s
e
to
1
,
an
d
th
a
t t
h
e
g
r
ee
n
er
y
/v
e
g
etatio
n
v
alu
e
is
a
r
o
u
n
d
0
.
1
–
0
.
4
.
T
h
is
is
in
li
n
e
w
it
h
p
r
ev
io
u
s
r
es
ea
r
c
h
r
eg
ar
d
in
g
th
e
s
i
g
n
i
f
ica
n
t r
o
l
e
o
f
NDVI
in
d
etec
ti
n
g
v
eg
e
ta
tio
n
v
ar
iatio
n
[
2
8
]
.
3
.
4
.
Cla
s
s
if
ica
t
io
n m
o
de
l
T
h
e
im
p
ac
t
o
f
o
v
er
f
itti
n
g
o
n
th
e
clas
s
i
f
icatio
n
r
e
s
u
lt
s
is
cl
ea
r
l
y
s
h
o
w
n
i
n
Fi
g
u
r
es
5
(
a)
an
d
(
b
)
to
Fig
u
r
es
8
(
a)
a
n
d
(
b
)
;
th
e
clas
s
if
icatio
n
r
es
u
lts
ex
h
ib
it
a
s
tr
a
n
g
e
p
atter
n
th
at
f
o
llo
w
s
t
h
e
i
n
p
u
t
v
ar
iab
le
’
s
s
h
ap
e
(
r
o
ad
s
,
r
iv
er
s
,
s
ettle
m
e
n
ts
,
s
l
o
p
es).
T
h
is
in
d
icate
s
th
at
th
e
m
o
d
el
is
o
v
er
l
y
s
e
n
s
iti
v
e
to
s
lig
h
t
v
ar
iatio
n
s
i
n
th
ese
v
ar
iab
les,
li
m
i
tin
g
it
s
ab
ilit
y
to
d
is
ti
n
g
u
is
h
a
m
o
n
g
la
n
d
co
v
er
c
lass
es.
O
v
er
f
itti
n
g
o
cc
u
r
s
w
h
en
a
m
o
d
el
ac
h
iev
e
s
h
i
g
h
ac
cu
r
ac
y
o
n
th
e
tr
ain
i
n
g
d
ata
b
u
t
f
a
ils
to
g
e
n
er
alize
w
ell
ac
r
o
s
s
m
u
ltip
le
d
atasets
[
8
]
,
[
1
9
]
.
If
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Dec
is
io
n
-
tr
ee
-
b
a
s
ed
ma
ch
in
e
lea
r
n
in
g
fo
r
d
etec
tin
g
c
o
ffee
a
g
r
o
fo
r
estr
y
…
(
I
Ma
d
e
K
h
r
is
n
a
Yo
g
a
Dev
a
n
d
r
a
)
1315
th
e
d
ata
is
o
v
er
f
it,
it
w
i
ll
h
a
v
e
an
u
n
n
at
u
r
al
s
h
ap
e.
Data
co
m
p
ar
is
o
n
s
ar
e
p
er
f
o
r
m
ed
to
v
is
u
alize
an
o
m
alie
s
.
I
n
s
o
m
e
ca
s
e
s
,
o
v
er
f
it
tin
g
o
cc
u
r
s
w
h
e
n
th
e
al
g
o
r
ith
m
o
v
er
lear
n
s
a
s
i
n
g
le
attr
ib
u
te.
T
h
is
i
s
ca
u
s
ed
b
y
a
li
m
ited
tr
ain
i
n
g
d
ataset,
a
n
i
m
b
ala
n
ce
in
th
e
tr
ai
n
i
n
g
d
ataset,
an
d
m
o
d
el
co
m
p
le
x
it
y
[
1
0
]
,
[
2
9
]
.
(
a)
(
b
)
(
a)
(
b
)
Fig
u
r
e
5
.
O
v
er
f
it
tin
g
i
n
estate
cr
o
p
class
if
icatio
n
:
(
a)
s
ettle
m
e
n
t
-
li
k
e
estate
cr
o
p
p
atter
n
an
d
(
b
)
s
ettle
m
en
t p
r
o
x
i
m
it
y
p
atter
n
Fig
u
r
e
6
.
Ov
er
f
it
tin
g
i
n
ag
r
ic
u
ltu
r
al
clas
s
i
f
icatio
n
:
(
a)
r
iv
er
-
p
r
o
x
i
m
it
y
-
li
k
e
ag
r
ic
u
ltu
r
al
p
atter
n
a
n
d
(
b
)
r
iv
er
p
r
o
x
i
m
it
y
p
atter
n
(
a)
(
b
)
(
a)
(
b
)
Fig
u
r
e
7
.
Ov
er
f
it
tin
g
i
n
s
cr
u
b
class
i
f
icatio
n
:
(
a)
r
o
a
d
-
p
r
o
x
i
m
it
y
-
li
k
e
s
cr
u
b
p
atter
n
an
d
(
b
)
r
o
a
d
p
r
o
x
i
m
it
y
p
atter
n
Fig
u
r
e
8
.
Ov
er
f
it
tin
g
i
n
s
ettle
m
en
t c
lass
if
icatio
n
:
(
a)
s
lo
p
e
-
lik
e
s
et
tle
m
e
n
t p
atter
n
an
d
(
b
)
s
lo
p
e
p
atter
n
Al
t
h
o
u
g
h
t
h
e
clas
s
if
icatio
n
r
e
s
u
lt
s
d
e
m
o
n
s
tr
ate
r
elati
v
el
y
h
ig
h
ac
c
u
r
ac
y
,
t
h
e
p
r
o
p
o
s
ed
m
eth
o
d
s
til
l
ex
h
ib
it
s
s
e
v
er
al
li
m
ita
tio
n
s
.
T
h
is
h
i
g
h
ac
c
u
r
ac
y
f
ai
ls
to
d
etec
t
w
ea
k
n
e
s
s
e
s
i
n
M
L
a
p
p
r
o
ac
h
es,
s
u
ch
as
o
v
er
f
itti
n
g
.
T
h
er
ef
o
r
e,
th
e
m
ain
li
m
itat
io
n
o
f
th
i
s
m
eth
o
d
is
th
e
lack
o
f
a
s
ta
n
d
ar
d
m
e
th
o
d
f
o
r
m
ea
s
u
r
i
n
g
m
o
d
el
r
el
iab
ilit
y
.
Ov
er
all
ac
cu
r
ac
y
,
t
h
e
k
ap
p
a
co
ef
f
icien
t,
r
ec
all,
an
d
p
r
ec
is
io
n
ar
e
r
eliab
le
s
tatis
tical
m
ea
s
u
r
es
f
o
r
ass
e
s
s
i
n
g
ac
cu
r
a
c
y
.
Ho
w
e
v
er
,
th
e
y
s
till
h
av
e
l
i
m
itat
io
n
s
i
n
m
ea
s
u
r
i
n
g
s
p
atia
l
er
r
o
r
s
,
in
clu
d
in
g
p
r
o
x
i
m
it
y
an
d
o
v
er
f
itti
n
g
.
I
n
th
is
s
t
u
d
y
,
b
ec
au
s
e
o
v
er
f
i
tt
in
g
p
atter
n
s
ca
n
n
o
t
b
e
id
en
tifie
d
s
o
lel
y
w
it
h
n
u
m
er
ical
m
e
tr
ics,
a
v
is
u
al
ap
p
r
o
ac
h
w
as
u
s
ed
to
d
iag
n
o
s
e
th
e
m
[
8
]
,
[
9
]
.
Ho
w
e
v
er
,
th
is
ap
p
r
o
ac
h
is
s
til
l
en
t
ir
el
y
q
u
alitati
v
e
a
n
d
lack
s
s
tan
d
ar
d
i
za
tio
n
as
a
m
et
h
o
d
f
o
r
d
iag
n
o
s
in
g
er
r
o
r
s
in
s
p
atial
m
ac
h
i
n
e
lear
n
i
n
g
.
T
h
er
ef
o
r
e,
th
e
ap
p
r
o
ac
h
r
elies
h
ea
v
i
l
y
o
n
th
e
an
a
l
y
s
t
’
s
k
n
o
w
led
g
e
an
d
s
k
ill
s
(
ex
p
er
t
j
u
d
g
m
e
n
t)
,
w
h
ic
h
ar
e
h
ig
h
l
y
s
u
b
j
ec
tiv
e.
T
h
er
ef
o
r
e,
f
u
r
t
h
er
r
esear
c
h
is
n
ee
d
ed
to
d
ev
elo
p
a
s
ta
n
d
ar
d
iz
ed
s
p
atial
d
iag
n
o
s
t
ic
m
eth
o
d
t
h
at
i
n
te
g
r
a
tes
s
p
atial
p
atter
n
s
an
d
ac
cu
r
ac
y
m
etr
ics.
I
n
t
h
is
s
t
u
d
y
,
to
r
ed
u
c
e
o
v
er
f
it
tin
g
,
w
e
ap
p
lied
a
r
eg
u
lar
izatio
n
s
tr
ateg
y
[
3
0
]
th
a
t
in
cl
u
d
es
(
1
)
d
r
o
p
p
in
g
th
e
at
tr
ib
u
tes
t
h
at
ca
u
s
e
o
v
er
f
itti
n
g
b
y
eli
m
i
n
ati
n
g
th
e
d
ep
en
d
en
ce
o
f
th
e
m
o
d
el
o
n
t
h
at
s
p
ec
if
ic
f
ea
t
u
r
e,
(
2
)
ap
p
ly
i
n
g
p
r
e
-
p
r
u
n
i
n
g
,
p
o
s
t
-
p
r
u
n
i
n
g
,
s
ett
in
g
t
h
e
m
in
i
m
u
m
cla
s
s
m
e
m
b
er
s
b
ef
o
r
e
s
p
litt
i
n
g
,
an
d
ad
j
u
s
tin
g
t
h
e
tr
ee
d
ep
th
.
T
h
e
f
ir
s
t
s
tr
ate
g
y
i
s
i
m
p
le
m
en
te
d
b
ec
au
s
e
th
e
n
u
m
b
er
o
f
p
o
in
t
s
a
m
p
les
is
li
m
ited
[
8
]
.
I
t
is
v
er
y
d
if
f
ic
u
lt
to
i
n
cr
ea
s
e
th
e
n
u
m
b
er
o
f
lar
g
e
g
r
o
u
n
d
-
s
u
r
v
e
y
s
a
m
p
les,
s
o
o
u
r
m
ai
n
f
o
c
u
s
i
s
to
eli
m
i
n
ate
v
ar
iab
les t
h
a
t le
ad
to
s
ev
er
e
o
v
er
f
i
tti
n
g
.
As
s
h
o
w
n
in
Fig
u
r
es
5
to
8
,
th
e
s
ettle
m
en
ts
,
r
o
ad
s
,
r
iv
er
s
,
a
n
d
s
lo
p
es
v
ar
iab
les
s
tr
o
n
g
l
y
co
n
tr
ib
u
te
to
th
e
o
v
er
f
it
tin
g
.
T
h
er
ef
o
r
e,
th
ese
v
ar
iab
les
ar
e
n
o
t
u
s
ed
in
th
e
m
o
d
el.
T
h
e
v
ar
iab
les
r
etai
n
ed
ar
e
th
e
im
a
g
e
in
d
ices
(
ND
VI
,
NR
GI
,
AR
VI
,
an
d
VDVI
)
,
t
h
e
ele
v
atio
n
v
ar
iab
les,
an
d
t
h
e
v
is
u
al
v
e
g
etati
o
n
co
v
er
v
ar
iab
le
s
.
T
h
e
s
ec
o
n
d
s
tr
ate
g
y
is
i
m
p
le
m
en
ted
to
eli
m
i
n
ate
n
o
is
e
f
r
o
m
s
a
m
p
le
s
t
h
at
d
o
n
o
t
r
ep
r
esen
t
f
ield
co
n
d
it
io
n
s
.
T
h
e
p
r
e
-
p
r
u
n
in
g
(
ea
r
l
y
s
to
p
p
in
g
)
l
i
m
its
tr
ee
g
r
o
w
th
d
u
r
in
g
t
h
e
s
ea
r
ch
f
o
r
m
ax
i
m
u
m
i
n
f
o
r
m
atio
n
g
ai
n
b
y
en
f
o
r
ci
n
g
co
n
s
tr
ai
n
ts
s
u
c
h
as
m
a
x
i
m
u
m
tr
ee
d
ep
th
an
d
,
i
m
p
o
r
tan
tl
y
,
t
h
e
m
in
i
m
u
m
n
u
m
b
er
o
f
s
a
m
p
le
s
r
eq
u
ir
ed
f
o
r
n
o
d
e
s
p
litt
in
g
an
d
f
o
r
ter
m
i
n
al
leaf
n
o
d
es
[
4
]
.
B
y
i
n
cr
ea
s
i
n
g
th
e
m
i
n
i
m
u
m
n
u
m
b
er
o
f
s
a
m
p
les
f
o
r
s
p
lits
a
n
d
leav
es,
t
h
e
m
o
d
el
a
v
o
id
s
cr
ea
tin
g
o
v
er
l
y
s
p
ec
i
f
ic
b
r
an
ch
es
t
h
at
ar
e
s
e
n
s
it
iv
e
to
s
m
all
f
l
u
ctu
a
tio
n
s
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
16
93
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
4
,
A
u
g
u
s
t
20
26
:
1
3
0
7
-
1
3
1
9
1316
th
e
d
ata,
th
er
eb
y
r
ed
u
cin
g
v
ar
ian
ce
an
d
i
m
p
r
o
v
i
n
g
g
en
er
al
iz
atio
n
.
P
o
s
t
-
p
r
u
n
i
n
g
(
p
r
u
n
i
n
g
)
,
in
co
n
tr
ast,
allo
w
s
th
e
d
ec
is
io
n
tr
ee
to
f
u
ll
y
g
r
o
w
b
ef
o
r
e
s
y
s
te
m
atica
ll
y
r
e
m
o
v
in
g
b
r
an
c
h
es
t
h
at
co
n
tr
ib
u
t
e
litt
le
to
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
.
T
h
e
f
o
cu
s
o
n
p
e
r
f
o
r
m
an
ce
(
o
v
er
f
it
tin
g
)
a
n
d
in
ter
p
r
etab
ilit
y
(
d
ec
is
io
n
tr
ee
)
i
s
v
ital
f
o
r
f
o
s
ter
in
g
tr
u
s
t
an
d
f
ac
ilit
ati
n
g
in
f
o
r
m
e
d
d
ec
is
io
n
-
m
ak
i
n
g
in
cr
itical
s
ec
to
r
s
w
h
er
e
A
I
is
i
n
cr
ea
s
i
n
g
l
y
d
ep
lo
y
ed
[
3
1
]
.
Ulti
m
a
tel
y
,
ad
d
r
ess
in
g
o
v
er
f
i
ttin
g
w
h
ile
en
h
a
n
ci
n
g
m
o
d
el
in
ter
p
r
etab
ilit
y
ca
n
lead
to
m
o
r
e
r
eliab
le
an
d
tr
u
s
t
w
o
r
t
h
y
A
I
s
y
s
te
m
s
,
f
o
s
ter
in
g
g
r
ea
ter
ac
ce
p
tan
ce
an
d
e
f
f
ec
tiv
e
u
s
e
i
n
cr
itical
d
ec
is
io
n
-
m
ak
in
g
co
n
te
x
t
s
.
T
h
e
r
esu
lti
n
g
cla
s
s
i
f
icat
io
n
m
ap
f
o
r
th
e
1
0
lan
d
co
v
er
clas
s
es
is
m
o
r
e
lo
g
ical
a
n
d
n
at
u
r
al
,
an
d
d
o
es
n
o
t
r
ev
ea
l
an
y
tr
e
n
d
p
atter
n
s
f
o
r
an
y
p
ar
ticu
lar
v
ar
iab
le
as
in
Fi
g
u
r
e
9
.
T
h
e
ag
r
o
f
o
r
estry
co
f
f
ee
lan
d
co
v
er
r
esu
lti
n
g
f
r
o
m
t
h
e
al
g
o
r
ith
m
ic
m
o
d
el
clas
s
i
f
icatio
n
is
s
h
o
w
n
in
d
a
r
k
b
r
o
w
n
.
I
n
g
en
er
al,
T
an
g
g
a
m
u
s
R
eg
e
n
c
y
is
d
o
m
i
n
ated
b
y
d
r
y
la
n
d
a
g
r
i
cu
lt
u
r
al
l
a
n
d
co
v
er
.
I
n
ac
co
r
d
an
ce
w
it
h
t
h
e
o
b
j
ec
tiv
es
o
f
th
is
s
t
u
d
y
,
t
h
e
la
n
d
co
v
er
class
if
ica
tio
n
m
ap
ai
m
s
to
f
ac
ilit
ate
th
e
v
is
u
aliza
tio
n
o
f
lan
d
co
v
er
r
es
u
lts
.
T
h
i
s
v
i
s
u
al
izatio
n
h
elp
s
d
ep
ict
an
ar
ea
f
o
r
p
r
ed
ictin
g
l
an
d
co
v
er
a
n
d
d
etec
tin
g
a
g
r
o
f
o
r
estry
co
f
f
ee
i
n
h
ar
d
-
to
-
r
ea
ch
p
lace
s
.
I
t
is
a
ls
o
h
elp
f
u
l
f
o
r
f
o
r
est p
lan
n
in
g
,
p
ar
ticu
lar
l
y
t
h
r
o
u
g
h
ag
r
o
f
o
r
estr
y
s
ch
e
m
es.
Fig
u
r
e
9
.
Sp
atial
d
is
tr
ib
u
tio
n
c
lass
i
f
icatio
n
m
ap
4.
CO
NCLU
SI
O
N
T
h
is
r
esear
ch
s
u
cc
ess
f
u
ll
y
d
ev
elo
p
ed
a
co
f
f
ee
ag
r
o
f
o
r
estr
y
class
i
f
icatio
n
alg
o
r
it
h
m
b
ase
d
o
n
ML
d
ec
is
io
n
tr
ee
s
b
y
i
n
teg
r
ati
n
g
s
p
ec
tr
al
v
ar
iab
les
f
r
o
m
SP
OT
-
7
im
a
g
er
y
(
NDVI
,
VDV
I
,
an
d
NR
GI
)
an
d
b
io
p
h
y
s
ical
v
ar
iab
les
(
elev
ati
o
n
an
d
v
i
s
u
al
lan
d
co
v
e
r).
Th
e
s
tu
d
y
d
e
m
o
n
s
tr
ate
s
t
h
at
a
d
ec
is
io
n
-
tr
ee
-
b
ase
d
ap
p
r
o
ac
h
ef
f
ec
ti
v
el
y
cla
s
s
i
f
ie
s
co
f
f
ee
a
g
r
o
f
o
r
estr
y
.
T
h
e
m
o
d
el
w
i
th
t
h
ese
i
n
teg
r
ated
v
ar
iab
les
i
m
p
r
o
v
ed
o
v
er
all
ac
cu
r
ac
y
f
r
o
m
6
0
.
4
%
to
9
4
.
0
%
an
d
k
ap
p
a
ac
cu
r
ac
y
f
r
o
m
5
5
.
4
%
to
9
3
.
3
%
co
m
p
ar
ed
to
th
e
s
p
ec
tr
al
-
o
n
l
y
m
o
d
el.
T
h
is
m
o
d
el
al
s
o
ac
h
iev
ed
u
s
er
ac
cu
r
ac
y
(
p
r
ec
is
io
n
)
o
f
8
3
.
6
%
an
d
p
r
o
d
u
ce
r
ac
cu
r
ac
y
(
r
ec
all)
o
f
8
9
.
5
%.
T
h
e
in
teg
r
atio
n
o
f
th
e
s
e
v
ar
iab
les
p
r
o
v
ed
ef
f
ec
ti
v
e
in
i
m
p
r
o
v
i
n
g
m
o
d
el
ac
cu
r
ac
y
w
h
ile
m
ai
n
tai
n
i
n
g
g
o
o
d
in
ter
p
r
etab
ilit
y
an
d
a
v
o
id
in
g
o
v
er
f
i
t
ti
n
g
,
t
h
er
eb
y
allo
w
i
n
g
f
o
r
a
clea
r
er
u
n
d
er
s
tan
d
in
g
o
f
t
h
e
d
eter
m
in
a
n
ts
o
f
co
f
f
ee
a
g
r
o
f
o
r
estry
s
y
s
te
m
s
.
Slo
p
e
an
d
p
r
o
x
i
m
it
y
to
r
o
ad
s
,
s
ettle
m
en
t
s
,
an
d
r
iv
er
s
w
er
e
id
en
ti
f
ied
as
s
o
u
r
ce
s
o
f
o
v
er
f
it
tin
g
an
d
w
er
e
eli
m
i
n
ated
f
r
o
m
th
e
m
o
d
el
to
p
r
o
d
u
ce
a
class
if
ic
atio
n
o
u
tp
u
t
t
h
at
w
a
s
m
o
r
e
lo
g
ical,
s
tab
le,
an
d
r
ep
r
esen
tativ
e
o
f
r
ea
l
-
w
o
r
ld
co
n
d
itio
n
s
.
T
h
is
al
g
o
r
ith
m
was
d
ev
elo
p
ed
u
s
i
n
g
in
f
o
r
m
atio
n
g
ain
cr
iter
ia
w
it
h
o
u
t
p
r
u
n
i
n
g
o
r
p
r
e
-
p
r
u
n
in
g
,
with
a
p
ar
am
eter
co
n
f
i
g
u
r
atio
n
o
f
m
ax
i
m
u
m
d
ep
t
h
3
5
,
m
i
n
i
m
u
m
lea
f
s
ize
6
0
,
m
i
n
i
m
u
m
s
p
lit s
ize
4
8
,
an
d
a
p
r
e
-
p
r
u
n
i
n
g
alter
n
ati
v
e
o
f
6
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.