I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
7
,
No
.
3
,
J
u
n
e
201
7
,
p
p
.
1180
~
1
1
8
7
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v7
i
3
.
p
p
1
1
8
0
-
1187
1180
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
JE
C
E
Co
m
pa
riso
n of
A
ccuracy
Mea
sure
s for RS
I
m
a
g
e
Cl
a
ss
ificatio
n
using
SVM a
nd
A
NN Clas
sifie
rs
S.V.
S.P
ra
s
a
d
1
,
T
.
Sa
t
y
a
S
a
v
i
t
hri
2
,
I
y
y
a
nk
i V
.
M
ura
li K
rish
na
3
1
De
p
t.
o
f
ECE
,
M
L
RIT
,
H
y
d
e
ra
b
a
d
,
In
d
ia
2
De
p
t.
o
f
ECE
,
JN
T
U,
Hy
d
e
ra
b
a
d
,
In
d
ia
3
CS
IT
a
n
d
Dire
c
to
r
(R&
D),
JN
T
U,
Hy
d
e
ra
b
a
d
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
9
,
2
0
1
6
R
ev
i
s
ed
Ma
r
2
7
,
2
0
1
7
A
cc
ep
ted
A
p
r
1
1
,
2
0
1
7
T
h
e
a
c
c
u
ra
te
lan
d
u
se
lan
d
c
o
v
e
r
(
L
U
L
C)
c
las
si
f
ica
ti
o
n
s
f
ro
m
sa
telli
te
im
a
g
e
r
y
a
re
p
ro
m
in
e
n
t
f
o
r
lan
d
u
se
p
lan
n
i
n
g
,
c
li
m
a
ti
c
c
h
a
n
g
e
d
e
tec
ti
o
n
a
n
d
eco
-
e
n
v
iro
n
m
e
n
t
m
o
n
it
o
ri
n
g
.
T
h
is
p
a
p
e
r
in
v
e
stig
a
tes
th
e
a
c
c
u
ra
c
y
a
n
d
re
li
a
b
il
it
y
o
f
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
(
S
V
M
)
c
las
sif
ier
f
o
r
c
las
si
fy
in
g
m
u
lt
i
-
sp
e
c
tral
im
a
g
e
o
f
H
y
d
e
ra
b
a
d
a
n
d
it
s
su
rr
o
u
n
d
in
g
s
a
re
a
a
n
d
a
lso
c
o
m
p
a
re
it
s
p
e
rf
o
r
m
a
n
c
e
w
it
h
A
rti
f
icia
l
Ne
u
ra
l
Ne
t
w
o
rk
(
AN
N)
c
las
si
f
ier.
In
th
is
p
a
p
e
r,
a
h
y
b
rid
tec
h
n
iq
u
e
w
h
ich
w
e
re
f
e
r
to
a
s
F
u
z
z
y
In
c
o
rp
o
ra
ted
Hi
e
ra
rc
h
ic
a
l
c
lu
ste
rin
g
h
a
s
b
e
e
n
p
r
o
p
o
se
d
f
o
r
c
lu
ste
rin
g
th
e
m
u
lt
isp
e
c
tral
sa
telli
te
im
a
g
e
s
in
to
L
UL
C
se
c
to
rs.
T
h
e
e
x
p
e
ri
m
e
n
tal
re
su
lt
s
sh
o
w
th
a
t
o
v
e
ra
ll
a
c
c
u
ra
c
ies
o
f
L
U
L
C
c
la
ss
i
f
ica
ti
o
n
o
f
th
e
H
y
d
e
r
a
b
a
d
a
n
d
it
s
su
rro
u
n
d
in
g
s
a
re
a
a
re
a
p
p
ro
x
im
a
tel
y
9
3
.
1
5
9
%
f
o
r
S
V
M
a
n
d
8
9
.
9
2
5
%
f
o
r
AN
N.
T
h
e
c
o
rr
e
sp
o
n
d
in
g
k
a
p
p
a
c
o
e
ff
icie
n
t
v
a
lu
e
s
a
re
0
.
8
9
3
a
n
d
0
.
8
4
3
.
T
h
e
c
las
sif
ied
re
su
lt
s
sh
o
w
th
a
t
th
e
S
V
M
y
ield
s
a
v
e
r
y
p
ro
m
isi
n
g
p
e
rf
o
r
m
a
n
c
e
th
a
n
th
e
AN
N
in
L
U
L
C
c
las
si
f
ica
ti
o
n
o
f
h
ig
h
re
so
l
u
ti
o
n
L
a
n
d
sa
t
-
8
sa
telli
te i
m
a
g
e
s.
K
ey
w
o
r
d
:
C
las
s
i
f
icatio
n
C
lu
s
ter
i
n
g
M
u
lti
s
p
ec
tr
al
Satellite I
m
a
g
e
S
u
p
p
o
r
t V
ec
to
r
Ma
ch
in
e
Co
p
y
rig
h
t
©
2
0
1
7
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
S.V.
S.P
r
asad
,
Dep
ar
te
m
en
t
o
f
E
lectr
ical
an
d
C
o
m
p
u
ter
E
n
g
in
ee
r
i
n
g
,
ML
R
I
n
s
t
itu
te
o
f
T
ec
h
n
o
lo
g
y
,
H
y
d
er
ab
ad
,
T
elan
g
an
a,
I
n
d
ia
-
5
0
0
0
4
3
.
E
m
ail:
P
r
asad
.
s
is
ta
@
g
m
a
il.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
r
ec
en
t
ti
m
e
s
,
s
atell
ite
i
m
ag
es
ar
e
d
eliv
er
in
g
a
n
e
n
o
r
m
o
u
s
s
o
u
r
ce
o
f
in
f
o
r
m
a
tio
n
f
o
r
s
t
u
d
y
i
n
g
th
e
s
p
atial
an
d
te
m
p
o
r
al
v
ar
iab
ilit
y
o
f
en
v
ir
o
n
m
e
n
tal
co
n
d
itio
n
s
.
I
t
ca
n
b
e
u
tili
ze
d
in
a
n
u
m
b
er
o
f
ap
p
licatio
n
s
w
h
ic
h
co
n
s
i
s
ts
o
f
m
a
k
in
g
o
f
m
ap
p
i
n
g
p
r
o
d
u
cts
f
o
r
m
ilit
ar
y
a
n
d
civ
il
p
u
r
p
o
s
es,
ex
p
lo
r
ati
o
n
,
n
u
r
s
i
n
g
o
f
lan
d
u
s
e
lan
d
co
v
er
[
1
]
,
ass
es
s
m
e
n
t
o
f
e
n
v
ir
o
n
m
e
n
tal
d
a
m
a
g
e,
r
ad
iatio
n
lev
el
ch
ec
k
,
s
o
il
te
s
t,
g
r
o
w
th
d
ir
ec
tiv
e,
cr
o
p
o
u
tco
m
e
i
n
cr
e
m
en
t
a
n
d
u
r
b
an
p
lan
n
i
n
g
.
T
h
ese
m
u
l
tis
p
e
ctr
al
i
m
a
g
es
ca
n
b
e
u
s
ed
m
ai
n
l
y
i
n
t
h
e
co
u
r
s
e
o
f
class
i
f
icatio
n
an
d
al
s
o
in
m
ap
p
in
g
o
f
v
e
g
etatio
n
o
v
er
ex
ten
s
iv
e
s
p
atial
s
ca
le
s
.
T
h
is
is
b
e
ca
u
s
e
m
u
ltis
p
ec
tr
al
i
m
a
g
e
cla
s
s
i
f
ie
s
la
n
d
co
v
er
an
d
lan
d
u
s
a
g
e
f
ea
t
u
r
es
s
u
ch
as
v
eg
eta
tio
n
,
w
ater
,
o
il,
f
o
r
ests
an
d
u
r
b
a
n
izatio
n
,
it
also
d
eliv
er
s
v
er
y
g
o
o
d
s
co
p
e
an
d
m
ap
p
in
g
.
T
h
is
k
i
n
d
o
f
class
i
f
icat
io
n
tec
h
n
iq
u
e
r
ep
l
ac
es
t
h
e
tr
ad
itio
n
al
class
i
f
icatio
n
tech
n
iq
u
e
s
[
2
]
.
A
s
m
an
y
en
v
ir
o
n
m
e
n
tal
an
d
s
o
cio
-
ec
o
n
o
m
ic
p
r
o
p
o
s
als
ar
e
b
ased
o
n
th
ese
class
i
f
icatio
n
r
es
u
lt
s
,
r
esear
ch
es
a
n
d
s
t
u
d
ies
o
n
t
h
ese
s
at
ellite
i
m
ag
e
clas
s
i
f
icatio
n
s
h
av
e
ce
n
tr
alize
d
th
e
co
n
ce
n
tr
atio
n
o
f
t
h
e
s
cie
n
ti
f
i
c
c
o
m
m
u
n
it
y
.
A
c
lass
if
icatio
n
tech
n
iq
u
e
clas
s
if
ies
r
ele
v
a
n
t
co
m
p
o
n
e
n
t
s
o
r
class
es
o
f
la
n
d
co
v
er
s
ec
tio
n
s
o
v
er
a
p
ar
ticu
lar
ar
ea
[
3
]
.
Mu
lti
s
p
ec
tr
al
im
a
g
e
s
p
r
im
ar
i
l
y
co
n
s
i
s
ts
t
h
e
d
ata
th
at
is
co
llected
o
v
er
a
w
id
e
r
an
g
e
o
f
f
r
eq
u
en
cie
s
an
d
it
ch
an
g
es
f
o
r
d
if
f
er
en
t
lo
ca
tio
n
s
[
4
]
.
T
h
is
o
v
er
all
p
ec
u
liar
n
atu
r
e
o
f
s
atelli
te
i
m
a
g
e
d
ata
ca
n
b
e
r
elate
d
to
th
e
s
p
ec
tr
al
f
ea
t
u
r
es
th
a
t
co
r
r
elate
w
it
h
th
e
s
p
atial
f
ea
t
u
r
es
b
elo
n
g
i
n
g
to
th
e
s
a
m
e
b
an
d
a
n
d
th
i
s
i
s
ca
lled
as
s
p
atial
co
r
r
elatio
n
.
C
la
s
s
i
f
icatio
n
o
f
la
n
d
u
s
e
a
n
d
la
n
d
co
v
er
u
s
i
n
g
r
e
m
o
te
s
en
s
i
n
g
i
m
a
g
er
y
is
to
u
g
h
b
ec
au
s
e
o
f
co
m
p
o
u
n
d
la
n
d
s
ca
p
es
a
n
d
also
t
h
e
s
p
ec
tr
al
an
d
s
p
atial
r
eso
lu
tio
n
o
f
th
e
s
a
tellite i
m
a
g
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
is
o
n
o
f A
cc
u
r
a
cy
Mea
s
u
r
es fo
r
R
S
I
ma
g
e
C
la
s
s
ifica
t
i
o
n
u
s
in
g
S
V
M a
n
d
…
.
(
S
.
V
.
S
.
P
r
a
s
a
d
)
1181
Satellite
i
m
ag
e
cla
s
s
i
f
icatio
n
ca
n
b
e
p
r
im
ar
il
y
co
n
s
id
er
ed
as
a
co
m
b
i
n
atio
n
o
f
b
o
th
i
m
ag
e
clu
s
ter
i
n
g
an
d
cla
s
s
i
f
icatio
n
m
et
h
o
d
s
[
5
]
.
A
t
p
r
esen
t
t
h
er
e
ar
e
m
a
n
y
i
m
a
g
e
cla
s
s
i
f
icatio
n
tec
h
n
iq
u
es
b
ein
g
in
tr
o
d
u
ce
d
u
n
d
er
d
i
f
f
er
en
t
ca
teg
o
r
ies
li
k
e
p
atter
n
r
ec
o
g
n
it
io
n
a
n
d
i
m
a
g
e
a
n
al
y
s
i
s
[
6
]
.
Un
d
er
s
o
m
e
cir
cu
m
s
ta
n
ce
s
,
th
e
p
r
o
ce
s
s
o
f
c
lass
if
ica
tio
n
m
ig
h
t
b
e
co
n
s
id
er
ed
as
a
n
e
n
ti
t
y
o
f
an
al
y
s
is
,
b
u
t
i
n
o
th
er
ca
s
es
it
is
co
n
s
id
er
ed
as
a
ce
n
tr
e
s
ta
g
e
i
n
co
m
p
licated
ca
s
es
li
k
e
d
eg
r
ad
atio
n
o
f
la
n
d
r
u
les,
la
n
d
s
ca
p
e
m
o
d
eli
n
g
,
p
r
o
ce
s
s
s
t
u
d
ies,
en
v
ir
o
n
m
e
n
t
m
o
n
i
to
r
in
g
a
n
d
m
an
a
g
e
m
e
n
t o
f
co
astal z
o
n
e
s
.
T
o
o
v
er
co
m
e
v
ar
io
u
s
p
r
o
b
lem
s
o
f
i
m
a
g
e
clas
s
i
f
icatio
n
g
r
ea
t
liter
atu
r
es
o
f
s
u
p
er
v
i
s
ed
tech
n
iq
u
e
s
ar
e
cr
ea
ted
[
7
]
.
Am
o
n
g
th
e
m
„
Ma
x
i
m
u
m
li
k
eli
h
o
o
d
‟
clas
s
i
f
ier
is
o
n
e
s
tat
is
tical
tech
n
iq
u
e
th
a
t
w
as
u
s
ed
p
r
ev
io
u
s
l
y
f
o
r
th
e
c
lass
if
ica
tio
n
p
r
o
ce
s
s
.
B
u
t
a
t
p
r
esen
t,
ar
t
if
icial
in
telli
g
en
ce
tech
n
iq
u
e
is
ap
p
lied
to
s
atellite
i
m
a
g
e
cla
s
s
i
f
icatio
n
s
[
8
]
.
T
o
in
cr
ea
s
e
th
e
p
r
ec
is
io
n
o
f
class
if
ica
tio
n
d
i
f
f
er
en
t
tech
n
iq
u
e
s
h
av
e
b
ee
n
p
r
esen
ted
[
9
]
,
b
u
t scien
tis
ts
h
a
v
e
p
u
t
g
r
ea
t e
f
f
o
r
ts
i
n
f
i
n
d
in
g
o
u
t a
n
e
f
f
icien
t s
y
s
te
m
an
d
m
e
th
o
d
f
o
r
in
cr
ea
s
i
n
g
th
e
ac
cu
r
ac
y
o
f
clas
s
i
f
icatio
n
.
P
r
im
ar
y
f
o
c
u
s
o
f
t
h
is
r
esear
c
h
is
to
class
if
y
m
u
lti
s
p
ec
tr
al
i
m
a
g
e
i
n
to
la
n
d
u
s
e
an
d
la
n
d
co
v
er
.
C
h
ar
ac
ter
is
tic
f
ea
t
u
r
es
o
f
la
n
d
s
u
r
f
ac
e
ca
n
b
e
r
elate
d
to
lan
d
co
v
er
[
1
0
]
.
T
h
in
g
s
t
h
at
w
e
ca
n
co
n
s
id
er
as
lan
d
co
v
er
ca
n
b
e
n
atu
r
al,
s
e
m
i
n
at
u
r
al
o
r
co
m
p
lete
m
an
m
ad
e
s
tr
u
ctu
r
e
s
th
at
ar
e
o
b
s
er
v
ab
le.
W
h
er
ea
s
lan
d
u
s
e
ca
n
b
e
co
n
s
id
er
ed
as
u
s
ab
ilit
y
o
f
lan
d
,
s
o
cio
-
ec
o
n
o
m
ic
ac
ti
v
itie
s
li
k
e
a
g
r
ic
u
lt
u
r
e
[
1
1
]
.
T
h
ese
ar
e
th
e
co
m
m
o
n
l
y
u
s
ed
cla
s
s
e
s
o
f
u
s
ag
e.
C
la
s
s
i
f
i
ca
tio
n
o
f
la
n
d
co
v
er
i
s
o
n
e
o
f
th
e
m
aj
o
r
in
p
u
t
s
w
h
en
co
n
s
id
er
ed
in
p
lan
n
i
n
g
a
t
lo
ca
l,
r
eg
io
n
al
an
d
n
atio
n
al
le
v
els.
Her
e,
an
ef
f
icie
n
t
clas
s
i
f
icatio
n
tech
n
iq
u
e
i
s
p
r
o
p
o
s
ed
to
class
if
y
r
e
m
o
te
s
en
s
ed
i
m
a
g
e
s
i
n
to
la
n
d
co
v
er
an
d
la
n
d
u
s
e
b
y
u
s
i
n
g
SVM
class
i
f
ier
[
1
2
]
.
T
h
is
tech
n
iq
u
e
co
m
p
r
is
es
o
f
f
o
u
r
p
h
ases
n
a
m
el
y
:
p
r
e
-
p
r
o
ce
s
s
i
n
g
,
cl
u
s
ter
in
g
b
ased
s
e
g
m
e
n
tatio
n
,
tr
ain
i
n
g
d
ata
s
elec
ti
o
n
f
o
r
SVM
a
n
d
class
i
f
icatio
n
u
s
i
n
g
tr
ain
ed
S
VM
.
As,
t
h
e
s
atel
lite
i
m
a
g
e
c
an
n
o
t
b
e
f
ed
d
ir
ec
tl
y
i
n
to
S
V
M
f
o
r
tr
ain
i
n
g
a
n
d
test
i
n
g
,
i
n
itiall
y
p
r
e
-
p
r
o
ce
s
s
i
n
g
is
to
b
e
d
o
n
e
s
o
th
at
i
m
ag
e
ca
n
b
e
co
n
v
er
ted
p
r
ef
er
ab
ly
f
o
r
th
e
p
r
o
ce
s
s
o
f
s
eg
m
e
n
tatio
n
.
L
ater
,
f
u
zz
y
in
co
r
p
o
r
ated
h
ier
ar
ch
ical
b
ased
clu
s
ter
i
n
g
al
g
o
r
ith
m
i
s
u
s
ed
f
o
r
s
eg
m
en
t
in
g
th
e
i
m
a
g
e
i
n
to
cl
u
s
ter
s
.
Af
ter
th
a
t
ce
n
tr
o
id
o
f
ea
c
h
cl
u
s
ter
is
s
u
b
j
ec
ted
to
tr
ain
ed
SVM
a
n
d
A
N
N
C
las
s
i
f
ier
s
r
esp
ec
tiv
el
y
[
1
3
]
.
T
h
is
p
ap
er
is
d
is
cu
s
s
ed
u
n
d
er
f
iv
e
d
i
f
f
er
en
t
h
ea
d
in
g
s
.
I
n
s
ec
tio
n
2
Su
p
p
o
r
t
Vec
to
r
m
ac
h
in
e
tech
n
iq
u
e
is
elab
o
r
ated
.
Sectio
n
3
p
r
esen
ts
th
e
p
r
o
p
o
s
ed
t
ec
h
n
iq
u
e
an
d
s
t
u
d
y
ar
ea
.
Sec
tio
n
4
p
r
esen
ts
t
h
e
ex
p
er
i
m
e
n
tal
r
es
u
lts
a
n
d
d
is
cu
s
s
io
n
.
Sec
tio
n
5
g
i
v
es t
h
e
co
n
c
lu
s
io
n
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
Su
pp
o
rt
Vec
t
o
r
M
a
chine (
SVM
)
SVM
i
s
o
n
e
o
f
t
h
e
s
tati
s
tical
b
ased
clas
s
i
f
ier
.
SV
M
s
ec
tio
n
s
m
a
in
l
y
r
ela
te
to
th
e
d
ec
is
i
o
n
s
u
r
f
ac
e
th
at
m
a
g
n
i
f
ies
t
h
e
b
o
u
n
d
ar
y
a
m
o
n
g
th
e
clas
s
es.
T
h
ese
s
u
r
f
ac
es
ar
e
ter
m
ed
as
o
p
tim
a
l
h
y
p
er
p
lan
e
an
d
th
e
d
ata
p
o
in
ts
t
h
at
ar
e
clo
s
e
to
t
h
e
h
y
p
er
p
la
n
e
ar
e
ter
m
ed
as
s
u
p
p
o
r
t
v
ec
to
r
s
.
W
h
e
n
w
e
a
r
e
co
n
s
id
er
in
g
t
h
e
tr
ain
i
n
g
s
e
t,
s
u
p
p
o
r
t v
ec
to
r
s
ar
e
v
er
y
i
m
p
o
r
ta
n
t.
Dev
ia
tio
n
s
o
f
SVM
ar
e
s
tated
b
elo
w
.
1)
B
y
u
tili
z
in
g
n
o
n
li
n
ea
r
k
er
n
el
s
,
SVM
ca
n
b
e
alter
ed
to
a
n
o
n
lin
ea
r
class
if
ier
.
2)
B
y
g
r
o
u
p
in
g
lar
g
e
n
u
m
b
er
o
f
b
in
ar
y
SVM
cla
s
s
i
f
ier
s
,
m
u
lti
class
i
f
ier
ca
n
b
e
ac
q
u
ir
ed
.
P
air
w
i
s
e
class
if
icatio
n
s
tr
ate
g
y
is
g
e
n
er
all
y
u
ti
lized
f
o
r
th
e
p
u
r
p
o
s
e
o
f
m
u
l
ticlas
s
class
if
ic
atio
n
.
T
h
e
o
u
tp
u
t
o
f
S
VM
clas
s
if
icatio
n
is
th
e
d
ec
is
io
n
v
al
u
es
f
o
r
ea
ch
p
ix
el
h
a
v
in
g
a
p
lace
w
it
h
ev
e
r
y
i
n
d
i
v
id
u
al
cla
s
s
an
d
it is
m
o
s
tl
y
u
ti
lized
f
o
r
p
r
o
b
ab
ilit
y
e
s
ti
m
ates.
W
h
en
a
b
i
n
ar
y
cla
s
s
i
s
co
n
s
id
er
ed
,
th
is
clas
s
i
f
ier
w
ill
p
o
s
s
e
s
s
a
h
y
p
er
p
lan
e
th
at
g
r
ad
u
al
l
y
d
ec
r
ea
s
es
th
e
d
is
ta
n
ce
f
r
o
m
its
n
ea
r
b
y
p
ix
els
b
elo
n
g
i
n
g
to
ea
c
h
cla
s
s
with
r
esp
ec
t
to
th
e
h
y
p
er
p
lan
e.
T
h
is
class
if
ica
tio
n
ca
n
b
e
d
ef
in
ed
as
f
o
llo
w
s
:
C
o
n
s
id
er
th
er
e
ar
e
N
tr
ain
in
g
s
a
m
p
les
an
d
t
h
e
y
ca
n
b
e
r
ep
r
ese
n
ted
b
y
s
et
o
f
p
air
s
)}
......,
3
,
2
,
1
),
,
{(
N
i
y
x
i
i
w
h
er
e
i
x
th
e
class
lab
el
o
f
v
al
u
e
is
1
an
d
i
y
n
w
h
er
e
it
s
h
o
ws
th
e
v
ec
to
r
co
m
p
r
is
i
n
g
o
f
n
ele
m
en
ts
.
B
asical
l
y
c
lass
if
ier
f
u
n
c
tio
n
is
g
i
v
e
n
b
y
x
y
f
)
;
(
w
h
er
e
α
,
is
th
e
m
ea
s
u
r
e
o
f
class
i
f
ier
ele
m
e
n
t.
Fig
u
r
e
1
s
h
o
w
s
R
ep
r
ese
n
tati
o
n
o
f
m
ax
i
m
u
m
-
m
ar
g
i
n
h
y
p
er
p
lan
e
an
d
m
ar
g
in
s
f
o
r
SVM
tr
ain
ed
w
it
h
s
a
m
p
les
b
elo
n
g
in
g
to
t
w
o
class
e
s
.
T
h
er
e
is
an
o
p
ti
m
u
m
s
ep
ar
atin
g
h
y
p
er
p
lan
e
t
h
at
w
a
s
f
o
u
n
d
o
u
t
b
y
SVM
al
g
o
r
ith
m
i
n
s
u
ch
a
w
a
y
th
at:
1
)
Sa
m
p
les
co
n
s
is
ti
n
g
o
f
lab
els
±
1
ar
e
s
itu
ated
o
n
eith
er
s
id
e
o
f
th
e
h
y
p
er
p
lan
e.
2
)
T
h
e
d
is
tan
ce
o
f
n
ea
r
est
v
ec
to
r
w
h
ic
h
i
s
ter
m
e
d
as
o
p
ti
m
al
m
ar
g
i
n
w
h
ic
h
ar
e
clo
s
er
to
th
e
h
y
p
er
p
lan
e
is
ca
lled
as
a
s
u
p
p
o
r
t
v
ec
to
r
.
Usu
al
l
y
th
e
eq
u
at
io
n
f
o
r
a
h
y
p
er
p
lan
e
is
g
iv
e
n
b
y
0
.
b
y
w
w
h
er
e
)
,
(
b
w
ar
e
th
e
p
ar
a
m
eter
f
ac
to
r
s
o
f
h
y
p
er
p
lan
e.
T
h
e
in
eq
u
alit
y
0
.
b
y
w
b
asicall
y
d
en
o
tes
v
e
cto
r
s
th
at
ar
e
n
o
t
o
n
t
h
e
h
y
p
er
p
lan
e,
s
o
th
e
class
i
f
ier
is
u
s
u
a
ll
y
r
e
p
r
esen
ted
as
).
.
(
)
;
(
b
y
w
s
g
m
y
f
Su
p
p
o
r
t
v
ec
to
r
s
b
asicall
y
lie
o
n
th
e
t
w
o
h
y
p
er
p
lan
es,
w
h
ich
ar
e
p
ar
allel
to
o
p
ti
m
al
h
y
p
er
p
lan
e,
o
f
eq
u
atio
n
1
.
b
y
w
.
Ma
x
i
m
izatio
n
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
3
,
J
u
n
e
2
0
1
7
:
1
1
8
0
–
1
1
8
7
1182
m
ar
g
i
n
w
it
h
th
e
s
e
eq
u
atio
n
s
o
f
t
w
o
s
u
p
p
o
r
t
v
ec
to
r
h
y
p
er
p
lan
es
m
ai
n
l
y
co
n
tr
ib
u
tes
to
th
e
f
o
l
lo
w
i
n
g
co
n
s
tr
ain
ed
o
p
ti
m
izatio
n
p
r
o
b
le
m
2
||
||
2
1
m
i
n
w
w
it
h
M
i
b
y
w
x
i
,.......,
2
,
1
,
1
)
.
(
.
Fo
r
th
e
p
u
r
p
o
s
e
o
f
clas
s
if
y
i
n
g
t
h
e
s
ate
llit
e
i
m
a
g
e
i
n
to
la
n
d
u
s
e
a
n
d
lan
d
co
v
er
ef
f
ec
ti
v
el
y
,
w
e
u
s
e
co
n
ce
p
ts
lik
e
H
ier
ar
ch
ical
cl
u
s
ter
in
g
an
d
F
u
zz
y
C
-
Me
an
s
al
g
o
r
ith
m
s
i
n
o
u
r
H
y
b
r
id
tech
n
i
q
u
e.
Fig
u
r
e
1
.
SVM
class
i
f
icat
io
n
f
o
r
t
w
o
d
if
f
er
e
n
t c
las
s
es
w
i
th
m
ar
g
i
n
le
v
els
2
.
2
.
H
y
brid T
ec
hn
iqu
e
(
F
uzzy
I
n
co
rpo
ra
t
ed
H
iera
rc
hica
l C
lus
t
er
ing
)
Her
e
H
y
b
r
id
tech
n
iq
u
e
w
ill
i
m
p
r
o
v
e
th
e
d
r
a
w
b
ac
k
s
o
f
b
o
th
h
ier
ar
ch
ical
a
n
d
FC
M
clu
s
ter
i
n
g
m
et
h
o
d
s
.
T
h
is
m
et
h
o
d
is
ca
lled
as
Fu
zz
y
i
n
co
r
p
o
r
ated
h
ier
a
r
ch
ical
clu
s
ter
in
g
(
H
y
b
r
id
T
e
ch
n
iq
u
e)
.
B
y
u
s
i
n
g
th
is
tec
h
n
iq
u
e
ac
c
u
r
ac
y
le
v
el
i
n
cr
ea
s
es
w
h
e
n
w
e
co
m
p
ar
e
w
i
th
th
e
n
o
r
m
al
FC
M.
Fig
u
r
e
2
.
Me
th
o
d
o
lo
g
y
f
lo
w
c
h
ar
t
La
n
ds
a
t
-
8
i
m
a
ge
Pr
e
-
p
r
o
c
e
s
si
n
g
st
a
ge
T
r
a
i
n
i
n
g
sa
m
p
le
se
le
c
t
i
o
n
C
las
si
f
i
c
a
t
i
o
n
u
si
n
g SV
M
c
las
si
f
i
e
r
A
c
c
ur
a
c
y
a
ss
e
ss
m
e
n
t
o
f
c
las
si
f
i
e
d
o
ut
p
ut
Fi
n
a
l
c
las
si
f
i
e
d
o
u
t
p
ut
C
o
m
p
a
r
a
t
i
v
e
a
n
a
lys
i
s
Hy
b
r
i
d
c
lus
t
e
r
i
n
g
m
e
t
h
o
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
is
o
n
o
f A
cc
u
r
a
cy
Mea
s
u
r
es fo
r
R
S
I
ma
g
e
C
la
s
s
ifica
t
i
o
n
u
s
in
g
S
V
M a
n
d
…
.
(
S
.
V
.
S
.
P
r
a
s
a
d
)
1183
As
th
e
s
ate
llit
e
i
m
ag
e
s
ca
n
n
o
t
b
e
f
ed
d
ir
ec
tly
in
to
t
h
e
SVM
class
i
f
ier
f
o
r
th
e
p
u
r
p
o
s
e
o
f
tr
ain
i
n
g
a
n
d
test
i
n
g
.
T
h
e
in
p
u
t
L
a
n
d
s
at
-
8
im
ag
e
s
h
o
u
ld
u
n
d
er
g
o
p
r
e
p
r
o
ce
s
s
i
n
g
p
r
o
ce
d
u
r
e
s
o
th
at
i
m
a
g
e
ca
n
b
e
ch
an
g
ed
f
o
r
f
u
r
t
h
er
p
r
o
ce
s
s
in
g
.
I
n
t
h
e
s
ec
o
n
d
s
tep
o
f
p
r
e
p
r
o
ce
s
s
i
n
g
th
e
g
i
v
e
n
i
m
a
g
e
i
s
co
n
s
id
er
ed
f
r
o
m
R
GB
to
L
a
b
co
lo
u
r
s
p
ac
e
w
h
ich
ca
n
b
e
f
u
r
th
er
s
e
g
m
en
ted
b
y
u
s
in
g
h
y
b
r
i
d
clu
s
ter
i
n
g
tec
h
n
iq
u
e.
F
ig
u
r
e
2
s
h
o
w
s
t
h
e
b
lo
ck
d
iag
r
a
m
f
o
r
th
e
p
r
o
p
o
s
ed
tech
n
iq
u
e
i
s
s
h
o
w
n
b
elo
w
.
T
h
e
r
esu
ltin
g
i
m
a
g
e
th
a
t
is
o
b
tain
ed
af
t
er
p
r
e
p
r
o
ce
s
s
in
g
co
n
s
is
t
s
o
f
th
o
u
s
an
d
s
o
f
p
i
x
els
an
d
to
class
i
f
y
b
ased
o
n
s
i
n
g
e
p
ix
el
is
a
g
r
ea
t
task
an
d
also
it
is
tim
e
co
n
s
u
m
in
g
.
.
Hen
ce
,
w
e
g
r
o
u
p
th
e
p
ix
els
in
to
clu
s
ter
s
an
d
f
o
r
ea
ch
cl
u
s
ter
ce
n
tr
o
id
is
s
elec
ted
.
T
h
is
ce
n
tr
o
id
w
ill
r
ep
r
esen
t
all
p
i
x
e
ls
in
a
g
i
v
en
cl
u
s
ter
.
E
v
er
y
p
i
x
el
i
n
a
cl
u
s
ter
w
ill
h
av
e
n
ea
r
l
y
s
a
m
e
p
i
x
el
v
al
u
e
an
d
it
d
i
f
f
er
s
o
n
l
y
b
y
t
h
e
ce
n
tr
o
id
v
al
u
e
j
u
s
t
b
y
s
m
al
l
a
m
o
u
n
t.
B
y
clas
s
i
f
y
i
n
g
th
e
ce
n
tr
o
id
w
il
l
i
n
d
ir
ec
tl
y
c
lass
i
f
y
all
th
e
p
ix
el
s
i
n
a
clu
s
ter
.
T
h
is
i
n
r
et
u
r
n
d
ec
r
ea
s
es th
e
i
n
p
u
ts
to
t
h
e
cla
s
s
i
f
ier
s
y
s
te
m
w
h
ic
h
m
ai
n
l
y
r
ed
u
ce
s
th
e
co
m
p
lex
i
t
y
a
n
d
also
ti
m
e.
No
w
,
w
e
d
is
c
u
s
s
ab
o
u
t
tr
ai
n
in
g
d
ata
s
elec
tio
n
t
h
at
is
g
i
v
e
n
t
o
th
e
SVM
c
lass
if
ier
f
o
r
th
e
p
u
r
p
o
s
e
o
f
class
i
f
icatio
n
.
C
la
s
s
i
f
ica
tio
n
i
s
d
o
n
e
ef
f
ec
ti
v
el
y
b
y
m
a
k
i
n
g
u
s
e
o
f
f
e
at
u
r
ed
co
lo
u
r
s
i
n
t
h
e
m
u
lti
s
p
ec
tr
al
i
m
a
g
e.
E
v
er
y
d
ata
ele
m
e
n
t
p
r
ese
n
t
o
n
ea
r
th
h
a
s
a
s
p
ec
if
ic
co
lo
u
r
b
y
w
h
ich
it
i
s
d
escr
ib
ed
.
I
n
-
o
r
d
er
to
class
if
y
t
h
e
g
iv
e
n
m
u
lti
s
p
ec
tr
al
i
m
a
g
e
u
s
i
n
g
th
e
SVM
cla
s
s
i
f
ier
,
w
e
u
s
e
th
e
co
lo
u
r
s
o
f
t
h
ese
ea
r
t
h
l
y
ele
m
e
n
ts
.
Sp
ec
if
i
c
co
lo
u
r
s
in
th
e
i
m
ag
e
r
ep
r
esen
t
s
f
o
r
„
lan
d
u
s
e
‟
an
d
f
e
w
f
o
r
„
l
an
d
co
v
er
‟
.
Af
ter
id
en
ti
f
ica
tio
n
co
lo
u
r
d
etails
ar
e
g
iv
e
n
to
t
h
e
SVM
cla
s
s
i
f
ier
f
o
r
f
u
r
th
er
cla
s
s
i
f
icatio
n
p
u
r
p
o
s
e.
Fig
u
r
e
3
s
h
o
w
s
th
e
C
h
ar
t
d
escr
ib
in
g
co
lo
u
r
s
w
it
h
w
h
ic
h
ele
m
e
n
t
s
o
f
t
h
e
ea
r
th
ar
e
r
ep
r
esen
ted
.
T
h
e
Fig
u
r
e
3
m
ai
n
l
y
d
escr
ib
es
d
if
f
er
e
n
t
co
lo
u
r
s
an
d
it‟s
r
ep
r
esen
tatio
n
in
s
ate
llit
e
i
m
a
g
e.
I
t
al
s
o
s
h
o
w
s
t
h
e
class
i
f
icat
io
n
o
f
l
an
d
u
s
e
an
d
lan
d
co
v
er
.
Data
elem
e
n
t
s
th
at
co
m
e
u
n
d
er
lan
d
u
s
e
ar
e
r
o
o
f
s
,
co
n
cr
ete
b
u
ild
in
g
s
a
n
d
th
o
s
e
t
h
at
c
o
m
e
u
n
d
er
lan
d
co
v
er
m
a
in
l
y
in
cl
u
d
e
v
e
g
etatio
n
,
m
u
d
,
s
o
il a
n
d
cr
o
p
s
.
Fig
u
r
e
3
.
C
h
ar
t d
escr
ib
in
g
co
l
o
u
r
s
w
it
h
w
h
ic
h
ele
m
en
ts
o
f
t
h
e
ea
r
th
ar
e
r
ep
r
esen
ted
.
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
3
.
1
.
Study
Are
a
Her
e
in
t
h
is
s
ec
tio
n
w
e
d
is
c
u
s
s
ab
o
u
t
t
h
e
m
u
ltis
p
ec
tr
al
s
at
el
lite
i
m
ag
e
s
.
T
h
ese
i
m
ag
e
s
a
r
e
ca
p
tu
r
ed
f
r
o
m
t
h
e
L
a
n
d
s
at
-
8
s
atell
ite.
T
h
is
o
b
s
er
v
atio
n
s
atelli
te
w
h
ich
is
a
n
Am
er
ican
E
ar
th
s
atellite
m
ai
n
l
y
co
n
s
is
t
s
o
f
n
in
e
s
p
ec
tr
al
b
an
d
s
.
Am
o
n
g
t
h
ese
b
a
n
d
s
,
b
an
d
1
to
b
an
d
7
h
a
v
e
a
s
p
atia
l
r
eso
lu
tio
n
o
f
3
0
m
e
ter
s
a
n
d
th
e
r
eso
lu
tio
n
f
o
r
b
an
d
8
is
1
5
m
eter
s
an
d
th
i
s
b
an
d
is
p
an
ch
r
o
m
atic.
Ultr
a
-
b
lu
e
w
h
ic
h
is
b
an
d
1
is
m
ain
l
y
u
s
ed
f
o
r
ae
r
o
s
o
l a
n
d
co
astal st
u
d
ies
.
T
a
b
le
1
Sh
o
w
s
all
b
an
d
Sp
ec
if
icatio
n
s
o
f
L
a
n
d
Sat
-
8
Satelli
te.
F
ig
u
r
e
4
s
h
o
w
s
all
b
an
d
s
o
f
t
h
e
in
p
u
t
m
u
ltis
p
ec
tr
al
s
a
tellite
i
m
a
g
e
o
f
H
y
d
er
ab
ad
d
is
tr
ict
i
n
th
e
y
ea
r
2
0
1
4
w
h
ich
w
a
s
tak
e
n
f
r
o
m
t
h
e
L
a
n
d
s
at
-
8
s
atell
ite.
C
o
n
s
id
er
in
g
all
8
b
an
d
s
o
f
t
h
e
i
n
p
u
t
i
m
ag
e,
H
y
d
er
ab
ad
cit
y
a
n
d
it
s
s
u
r
r
o
u
n
d
i
n
g
s
r
e
g
io
n
is
cr
o
p
p
ed
b
ased
o
n
it
s
latit
u
d
e
an
d
lo
n
g
it
u
d
e
v
al
u
es
b
y
t
h
e
h
elp
o
f
H
y
d
er
ab
ad
cit
y
an
d
it
s
s
u
r
r
o
u
n
d
i
n
g
s
to
p
o
-
s
h
ee
t.
A
l
l
t
h
es
e
8
b
an
d
s
ar
e
s
tack
ed
to
ac
q
u
ir
e
th
e
co
m
p
lete
in
f
o
r
m
atio
n
p
r
o
v
id
ed
b
y
th
e
m
.
T
h
e
m
et
h
o
d
o
lo
g
y
i
s
test
ed
o
n
a
L
a
n
d
s
at
-
8
R
S
i
m
a
g
e
w
h
i
ch
is
s
h
o
w
n
in
F
ig
u
r
e
5
w
h
i
ch
is
h
ig
h
r
eso
lu
tio
n
Gr
a
y
-
s
ca
le
a
n
d
f
a
ls
e
co
lo
u
r
ed
R
S i
m
ag
e
s
o
f
H
y
d
e
r
ab
ad
city
a
n
d
its
s
u
r
r
o
u
n
d
ig
s
w
h
ic
h
ar
e
o
b
tain
ed
b
y
co
m
b
i
n
i
n
g
all
8
b
an
d
s
o
f
L
an
d
s
at
-
8.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
3
,
J
u
n
e
2
0
1
7
:
1
1
8
0
–
1
1
8
7
1184
T
ab
le
1
.
L
an
d
s
at
-
8
B
an
d
Sp
ec
if
icatio
n
s
S
a
t
e
l
l
i
t
e
n
a
me
B
a
n
d
s
W
a
v
e
l
e
n
g
t
h
(
mi
c
r
o
me
t
e
r
s)
R
e
so
l
u
t
i
o
n
(
me
t
e
r
s)
L
a
n
d
sat
-
8
B
a
n
d
1
C
o
a
st
a
l
a
e
r
o
so
l
0
.
4
3
-
0
.
4
5
30
B
a
n
d
2
B
l
u
e
0
.
4
5
-
0
.
5
1
30
B
a
n
d
3
G
r
e
e
n
0
.
5
3
-
0
.
5
9
30
B
a
n
d
4
R
e
d
0
.
6
4
-
0
.
6
7
30
B
a
n
d
5
N
e
a
r
I
n
f
r
a
r
e
d
(
N
I
R
)
0
.
8
5
-
0
.
8
8
30
B
a
n
d
6
S
W
I
R
1
1
.
5
7
-
1
.
6
5
30
B
a
n
d
7
S
W
I
R
2
2
.
1
1
-
2
.
2
9
30
B
a
n
d
8
P
a
n
c
h
r
o
ma
t
i
c
0
.
5
0
-
0
.
6
8
15
B
an
d
-
1
B
an
d
-
2
B
an
d
-
3
B
an
d
-
4
B
an
d
-
5
B
an
d
-
6
B
an
d
-
7
B
an
d
-
8
Fig
u
r
e
4
.
I
n
p
u
t satell
ite
i
m
a
g
e
(
H
y
d
er
ab
ad
d
is
tr
ict: y
ea
r
-
2014)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
is
o
n
o
f A
cc
u
r
a
cy
Mea
s
u
r
es fo
r
R
S
I
ma
g
e
C
la
s
s
ifica
t
i
o
n
u
s
in
g
S
V
M a
n
d
…
.
(
S
.
V
.
S
.
P
r
a
s
a
d
)
1185
(
a)
Gr
ay
-
s
ca
le
(
b
)
R
GB
Fig
u
r
e
5
Hig
h
r
eso
lu
tio
n
Gr
a
y
-
s
ca
le/
f
al
s
e
co
lo
r
ed
R
S i
m
a
g
e
o
f
H
y
d
er
ab
ad
city
a
n
d
its
s
u
r
r
o
u
n
d
i
g
s
ar
ea
3
.
2
.
Cla
s
s
if
ica
t
io
n Re
s
ult
s
:
T
h
e
s
tu
d
y
ar
ea
w
as
ca
te
g
o
r
iz
ed
in
to
f
o
u
r
la
n
d
t
y
p
es,
i.e
.
v
eg
etatio
n
,
Ur
b
an
,
w
ater
,
a
n
d
o
th
er
s
(
d
r
y
lan
d
,
r
o
ck
etc.
.
.
)
u
s
i
n
g
t
h
e
s
a
m
e
r
e
g
io
n
o
f
I
n
ter
est
(
R
OI
)
tr
ain
i
n
g
s
a
m
p
les
an
d
co
v
er
in
g
a
m
aj
o
r
it
y
o
f
t
h
e
la
n
d
co
v
er
f
ea
t
u
r
es.
L
ater
,
t
h
e
tr
ai
n
in
g
s
a
m
p
le
s
ar
e
co
n
s
id
er
ed
f
o
r
clu
s
ter
i
n
g
H
y
d
er
ab
ad
cit
y
an
d
its
s
u
r
r
o
u
n
d
i
g
s
ar
ea
in
to
class
es
u
s
in
g
F
u
zz
y
I
n
co
r
p
o
r
ated
Hier
ar
ch
ical
clu
s
t
e
r
in
g
tec
h
n
iq
u
e.
T
h
e
class
i
f
ied
i
m
ag
e
s
o
f
G
HM
C
u
s
i
n
g
A
NN
an
d
SVM
class
i
f
ier
s
ar
e
s
h
o
w
n
in
Fi
g
u
r
e
6
.
I
n
Fig
u
r
e
6
,
w
e
ca
n
s
ee
d
i
f
f
er
en
t
co
lo
r
s
ea
ch
co
lo
r
r
ep
r
esen
ts
d
if
f
er
en
t
c
lass
.
I
n
cla
s
s
i
f
icatio
n
o
f
H
y
d
er
ab
ad
cit
y
an
d
its
s
u
r
r
o
u
n
d
i
g
s
ar
ea
w
e
c
o
n
s
id
er
ed
f
o
u
r
class
e
s
s
u
c
h
as
u
r
b
an
,
v
e
g
etatio
n
,
w
ater
a
n
d
o
th
er
s
.
I
n
Fi
g
u
r
e
6
,
r
ed
co
l
o
r
in
d
icate
s
u
r
b
an
ar
ea
,
Gr
ee
n
co
lo
r
in
d
icate
s
v
eg
e
tat
io
n
,
B
lu
e
co
lo
r
in
d
icate
s
w
a
te
r
an
d
Yello
w
co
lo
r
in
d
icate
s
o
t
h
er
s
.
I
n
s
ec
tio
n
2
,
Fig
u
r
e
3
s
h
o
ws
th
e
ch
ar
t
d
escr
ib
in
g
co
lo
r
s
w
it
h
w
h
ich
ele
m
e
n
ts
o
f
t
h
e
ea
r
th
ar
e
r
ep
r
esen
ted
.
Fig
u
r
e
6
(
a)
C
lass
i
f
ied
i
m
a
g
e
u
s
i
n
g
A
NN
clas
s
i
f
ier
(
b
)
C
lass
if
ied
i
m
ag
e
u
s
i
n
g
SVM
clas
s
i
f
ier
3
.
3
.
Acc
ura
cy
Ass
es
s
m
ent
I
n
th
is
s
ec
tio
n
,
w
e
p
r
esen
t
t
h
e
p
er
f
o
r
m
an
ce
e
v
al
u
atio
n
r
es
u
lts
o
f
t
h
e
p
r
o
p
o
s
ed
tech
n
iq
u
e.
Her
e
w
e
h
av
e
e
v
al
u
ated
an
d
co
m
p
ar
ed
th
e
r
es
u
lts
w
it
h
v
ar
io
u
s
cl
u
s
te
r
in
g
al
g
o
r
ith
m
s
an
d
al
s
o
w
it
h
v
ar
io
u
s
c
lass
if
ier
s
.
T
h
e
f
ir
s
t
p
ar
t
i
n
th
i
s
s
ec
tio
n
,
d
ea
ls
w
it
h
t
h
e
c
o
m
p
ar
i
s
o
n
w
i
t
h
v
ar
io
u
s
clu
s
ter
i
n
g
alg
o
r
it
h
m
s
.
I
n
t
h
e
later
p
ar
t,
ev
alu
a
tio
n
a
n
d
co
m
p
ar
is
o
n
is
m
ad
e
b
y
co
m
p
ar
i
n
g
v
ar
io
u
s
class
i
f
ier
s
.
A
d
etailed
an
al
y
s
i
s
,
f
o
llo
w
ed
b
y
t
h
e
ev
alu
a
tio
n
g
r
ap
h
is
m
ad
e
in
e
ac
h
p
ar
t.
A
n
al
y
s
is
m
ad
e
an
d
r
esu
lt
s
o
b
tain
ed
clea
r
l
y
d
e
m
o
n
s
tr
ate
th
e
ef
f
icie
n
c
y
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
in
class
i
f
y
in
g
t
h
e
m
u
lti
s
p
ec
tr
al
i
m
ag
e
in
to
la
n
d
u
s
e
a
n
d
lan
d
co
v
er
r
eg
io
n
s
.
T
h
e
o
v
er
all
ac
cu
r
ac
y
i
s
t
h
e
p
er
ce
n
tag
e
o
f
co
r
r
ec
tl
y
clas
s
i
f
i
ed
p
ix
els.
Kap
p
a
co
ef
f
ic
ien
t
i
s
an
o
t
h
er
cr
iter
io
n
class
ica
ll
y
u
s
ed
i
n
r
e
m
o
te
s
e
n
s
i
n
g
clas
s
i
f
icatio
n
to
m
ea
s
u
r
e
th
e
d
e
g
r
ee
o
f
a
g
r
ee
m
en
t
an
d
ta
k
es
i
n
to
ac
co
u
n
t
th
e
co
r
r
ec
t
class
if
ic
atio
n
t
h
at
m
a
y
h
a
v
e
b
ee
n
o
b
tain
ed
b
y
c
h
a
n
ce
b
y
w
ei
g
h
tin
g
t
h
e
m
ea
s
u
r
ed
ac
cu
r
ac
ies.
3
.
3
.
1
.
E
v
a
lua
t
io
n
us
ing
diff
er
ent
C
lus
t
er
ing
Alg
o
rit
h
m
Fo
r
p
er
f
o
r
m
a
n
ce
ev
al
u
atio
n
,
th
e
p
r
o
p
o
s
ed
tech
n
iq
u
e
(
p
r
o
p
o
s
ed
clu
s
ter
i
n
g
+
SVM)
is
ev
al
u
ated
w
it
h
th
e
tr
ad
itio
n
al
clu
s
ter
in
g
alg
o
r
ith
m
(
p
r
o
p
o
s
ed
clu
s
ter
in
g
+
A
NN)
.
T
h
e
ac
cu
r
ac
y
v
alu
e
is
co
m
p
u
ted
b
y
d
iv
id
in
g
t
h
e
to
tal
n
u
m
b
er
o
f
s
i
m
ilar
p
ix
els
id
en
ti
f
ied
as
la
n
d
u
s
e
to
t
h
e
n
u
m
b
er
o
f
p
ix
els
in
t
h
e
la
n
d
u
s
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
3
,
J
u
n
e
2
0
1
7
:
1
1
8
0
–
1
1
8
7
1186
r
eg
io
n
.
A
s
u
m
m
ar
y
o
f
ac
cu
r
a
c
y
an
d
k
ap
p
a
co
ef
f
icien
t
o
f
A
N
N
an
d
SVM
clas
s
i
f
ier
s
ca
n
b
e
f
o
u
n
d
in
T
ab
le
2
&
3
r
esp
ec
tiv
el
y
.
T
ab
le
2
.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
A
N
N
class
if
ica
tio
n
o
n
H
y
d
er
ab
ad
city
a
n
d
its
s
u
r
r
o
u
n
d
i
g
s
ar
ea
Ov
er
all
A
cc
u
r
ac
y
:
8
9
.
9
2
5
%
Kap
p
a:
0
.
8
4
3
T
ab
le
3
.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
SVM
class
if
icatio
n
o
n
H
y
d
er
ab
ad
city
a
n
d
its
s
u
r
r
o
u
n
d
i
g
s
ar
ea
Ov
er
all
A
cc
u
r
ac
y
:
9
3
.
1
5
9
%
Kap
p
a:
0
.
8
9
3
Her
e,
T
a
b
le
2
an
d
3
s
h
o
w
th
e
co
n
f
u
s
io
n
m
atr
ice
s
o
f
t
h
e
d
if
f
er
en
t
m
et
h
o
d
s
i.e
.
ANN
an
d
SVM
i
n
lan
d
u
s
e
cla
s
s
i
f
icatio
n
.
Am
o
n
g
t
h
e
t
w
o
cla
s
s
i
f
ied
o
u
tp
u
t
s
t
h
e
i
n
ter
esti
n
g
o
b
s
er
v
atio
n
w
a
s
t
h
at
t
h
e
S
VM
g
o
t
g
o
o
d
class
if
icatio
n
r
es
u
lt
s
th
a
n
A
NN
w
it
h
an
o
v
er
al
l
ac
cu
r
a
c
y
o
f
9
3
.
1
5
9
%
f
o
r
SVM
an
d
8
9
.
9
2
5
%
f
o
r
A
NN.
T
h
e
k
ap
p
a
co
ef
f
icien
t
v
al
u
es
ar
e
0
.
8
9
3
an
d
0
.
8
4
3
f
o
r
SVM
an
d
A
N
N
r
esp
ec
t
iv
el
y
.
T
h
e
o
b
tain
ed
r
esu
lts
ar
e
co
m
p
ar
ed
b
ased
o
n
th
eir
ac
cu
r
ac
ies o
f
in
d
i
v
id
u
a
l c
lass
e
s
an
d
it is
in
d
icate
d
i
n
T
ab
le
4
.
T
ab
le
4
: Co
m
p
ar
is
o
n
o
f
C
la
s
s
i
f
icatio
n
A
cc
u
r
ac
ie
s
O
v
e
r
a
l
l
I
d
e
n
t
i
f
i
e
d
P
i
x
e
l
s (%)
A
N
N
S
V
M
V
e
g
e
t
a
t
i
o
n
9
3
.
0
0
%
9
5
.
4
5
8
%
U
r
b
a
n
9
0
.
7
2
4
%
9
3
.
8
8
7
%
W
a
t
e
r
9
3
.
8
8
%
9
9
.
1
0
%
O
t
h
e
r
s
7
7
.
5
7
5
%
8
3
.
6
8
%
O
v
e
r
a
l
l
A
c
c
u
r
a
c
y
8
9
.
9
2
5
%
9
3
.
1
5
9
%
K
a
p
p
a
0
.
8
4
3
0
.
8
9
3
Fro
m
ab
o
v
e
tab
le
it
is
o
b
s
er
v
ed
th
at
th
e
SV
M
class
if
ier
h
as
d
etec
ted
all
class
es
s
u
ch
a
s
Veg
etatio
n
,
Ur
b
an
,
w
ater
a
n
d
o
th
er
s
co
m
p
ar
ativ
el
y
b
etter
t
h
an
ANN
clas
s
if
ier
.
4.
CO
NCLU
SI
O
N
I
n
t
h
is
p
ap
er
,
w
e
e
x
p
lo
r
ed
th
e
ac
c
u
r
ac
y
a
n
d
r
eliab
ilit
y
o
f
S
VM
cla
s
s
i
f
ier
f
o
r
clas
s
i
f
y
in
g
m
u
l
ti
-
s
p
ec
tr
al
i
m
a
g
e
o
f
H
y
d
er
ab
ad
cit
y
an
d
i
ts
s
u
r
r
o
u
n
d
ig
s
ar
ea
,
I
n
d
ia
an
d
al
s
o
co
m
p
ar
ed
it
s
p
er
f
o
r
m
an
ce
w
it
h
A
N
N
clas
s
if
ier
.
Her
e
in
S
VM
,
w
e
h
a
v
e
u
s
ed
an
e
f
f
i
cien
t
i
m
a
g
e
clu
s
ter
i
n
g
tec
h
n
iq
u
e
n
a
m
ed
F
u
zz
y
C
L
A
S
S
V
e
g
e
t
a
t
i
o
n
U
r
b
a
n
a
t
e
r
O
t
h
e
r
s
T
o
t
a
l
V
e
g
e
t
a
t
i
o
n
5
0
4
2
9
4
2
3
7
4
1
1
7
8
2
0
4
4
2
5
2
8
2
8
8
U
r
b
a
n
2
1
7
7
8
5
2
5
3
1
7
5
6
7
1
1
8
5
6
5
5
9
5
1
8
W
a
t
e
r
2
9
9
0
3
3
0
8
8
0
3
3
3
8
7
O
t
h
e
r
s
1
6
1
7
9
1
5
6
7
7
0
1
6
3
3
5
1
1
9
5
2
0
7
T
o
t
a
l
5
4
2
5
5
0
5
4
3
3
6
8
3
4
8
3
3
1
9
5
6
4
9
1
3
1
6
4
0
0
C
L
A
S
S
V
e
g
e
t
a
t
i
o
n
U
r
b
a
n
a
t
e
r
O
t
h
e
r
s
T
o
t
a
l
V
e
g
e
t
a
t
i
o
n
4
9
1
3
4
2
1
0
1
2
4
2
0
0
1
2
4
8
2
1
5
2
8
2
8
8
U
r
b
a
n
2
1
4
8
0
5
0
7
6
1
6
1
7
5
6
6
1
2
8
5
6
5
5
9
5
1
8
W
a
t
e
r
2
0
4
7
0
3
1
3
4
0
0
3
3
3
8
7
O
t
h
e
r
s
2
3
9
8
9
1
9
7
8
6
0
1
5
1
4
3
2
1
9
5
2
0
7
T
o
t
a
l
5
3
8
8
5
8
5
3
7
5
2
6
5
0
9
0
7
1
8
9
1
0
9
1
3
1
6
4
0
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
is
o
n
o
f A
cc
u
r
a
cy
Mea
s
u
r
es fo
r
R
S
I
ma
g
e
C
la
s
s
ifica
t
i
o
n
u
s
in
g
S
V
M a
n
d
…
.
(
S
.
V
.
S
.
P
r
a
s
a
d
)
1187
I
n
co
r
p
o
r
ated
Hier
a
r
ch
ical
clu
s
ter
in
g
to
clu
s
ter
th
e
m
u
ltis
p
ec
tr
al
r
em
o
te
s
en
s
ed
s
atellite
i
m
ag
e
s
.
A
n
d
t
h
eir
class
i
f
icatio
n
ac
cu
r
ac
ies
w
er
e
co
m
p
ar
ed
u
s
in
g
t
h
e
C
o
n
f
u
s
io
n
m
atr
ix
a
n
al
y
s
i
s
.
T
h
e
o
v
er
all
ac
cu
r
ac
ies
o
f
L
UL
C
clas
s
i
f
icatio
n
o
f
t
h
e
H
y
d
er
ab
ad
an
d
it
s
s
u
r
r
o
u
n
d
in
g
ar
ea
ar
e
ap
p
r
o
x
im
a
tel
y
9
3
.
1
5
9
%
an
d
8
9
.
9
2
5
%
an
d
th
e
k
ap
p
a
co
ef
f
ic
ien
ts
ar
e
0
.
8
9
3
an
d
0
.
8
4
3
f
o
r
t
h
e
ANN
a
n
d
S
VM
class
i
f
icatio
n
m
et
h
o
d
s
r
esp
ec
tiv
el
y
.
A
l
th
o
u
g
h
th
e
s
e
L
U
L
C
class
i
f
icatio
n
r
esu
lt
s
class
if
ie
d
b
y
th
e
A
NN
an
d
th
e
SVM
b
o
th
ac
h
ie
v
e
r
elati
v
el
y
h
i
g
h
ac
cu
r
ac
y
a
n
d
is
ad
eq
u
ate
en
o
u
g
h
f
o
r
th
e
s
t
u
d
y
o
n
u
r
b
an
lan
d
u
s
e/co
v
er
class
i
f
icatio
n
,
a
g
r
ea
t
n
u
m
b
er
o
f
u
n
ce
r
tai
n
tie
s
ex
is
t
i
n
th
e
p
r
ac
tice
d
u
e
to
m
a
n
y
in
f
l
u
en
c
in
g
f
ac
to
r
s
s
u
c
h
a
s
tr
ain
i
n
g
R
OI
s
elec
tio
n
s
,
at
m
o
s
p
h
er
ic
s
itu
a
t
io
n
s
,
u
n
d
er
l
y
in
g
s
u
r
f
ac
e
co
n
d
itio
n
s
an
d
m
i
x
in
g
p
ix
el
e
f
f
ec
ts
.
Ho
w
e
v
er
,
r
esu
lt
s
ar
e
in
d
icati
n
g
th
at
SVM
h
a
s
t
h
e
b
etter
f
in
a
l c
lass
i
f
icatio
n
ac
cu
r
ac
y
a
m
o
n
g
th
e
t
w
o
m
et
h
o
d
s
.
RE
F
E
R
E
NC
E
S
[1
]
Hu
a
-
M
e
i
Ch
e
n
,
V
a
rsh
n
e
y
,
P
.
K.
a
n
d
A
ro
ra
,
M
.
K,
“
P
e
rf
o
rm
a
n
c
e
o
f
M
u
tu
a
l
In
f
o
rm
a
ti
o
n
S
im
il
a
ri
ty
M
e
a
su
re
f
o
r
Re
g
istratio
n
o
f
M
u
lt
i
T
e
m
p
o
ra
l
Re
m
o
te
S
e
n
sin
g
Im
a
g
e
s”
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Ge
o
sc
ien
c
e
a
n
d
Rem
o
te
S
e
n
si
n
g
,
V
o
l
.
4
1
No
.
1
1
,
p
p
.
2
4
4
5
–
2
4
5
4
,
2
0
0
3
.
[2
]
M
in
g
-
Hs
e
n
g
T
se
n
g
,
S
h
e
n
g
-
Jh
e
Ch
e
n
,
Gwo
-
Ha
u
r
Hw
a
n
g
,
M
in
g
-
Yu
S
h
e
n
,
“
A
Ge
n
e
ti
c
A
lg
o
rit
h
m
Ru
le
-
b
a
se
d
A
p
p
ro
a
c
h
f
o
r
L
a
n
d
-
c
o
v
e
r
Clas
si
fica
ti
o
n
”
,
J
o
u
rn
a
l
o
f
Ph
o
to
g
ra
mm
e
try
a
n
d
Rem
o
te
S
e
n
si
n
g
,
Vo
l.
6
3
,
No
.
2
,
(3
)
,
p
p
.
202
-
2
1
2
,
2
0
0
8
.
[3
]
S
.
V.S
P
ra
sa
d
,
Dr.T
.
S
a
ty
a
S
a
v
it
ri,
Dr.I.
V.M
u
ra
li
Kris
h
n
a
,
“
Clas
sif
ica
ti
o
n
o
f
M
u
lt
isp
e
c
tral
S
a
telli
t
e
I
m
a
g
e
u
sin
g
Clu
ste
rin
g
w
it
h
S
VM
Clas
sif
ier
”
,
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
Co
mp
u
ter
Ap
p
li
c
a
t
io
n
s
,
V
o
l
.
3
5
,
N
o
.
5
,
p
p
.
4
3
9
9
-
6
1
0
7
,
2
0
1
1
.
[4
]
Hu
a
n
g
C.
,
L
.
S
.
D
a
v
is,
J.
R.
G
.
T
o
w
n
sh
e
n
d
2
0
0
2
,
“
A
n
A
s
se
ss
m
e
n
ts
o
f
S
u
p
p
o
r
t
V
e
c
to
r
M
a
c
h
i
n
e
s
f
o
r
L
a
n
d
c
o
v
e
r
C
las
sif
ic
a
ti
o
n
,
”
In
t
.
J
.
Rem
o
te
se
n
sin
g
,
v
o
l.
2
3
,
n
o
.
4
,
p
p
.
7
2
5
–
7
4
9
,
2
0
0
2
.
[5
]
A
n
d
e
rso
n
,
J.R.
,
E.
E.
Ha
rd
y
,
J.T
.
Ro
a
c
h
,
R.
E
.
W
it
m
e
r,
1
9
7
6
,
“
A
L
a
n
d
Us
e
a
n
d
L
a
n
d
C
o
v
e
r
Clas
sif
i
c
a
ti
o
n
S
y
ste
m
f
o
r
Us
e
W
it
h
Re
m
o
te
S
e
n
so
r
Da
ta
”
,
USG
S
P
ro
f
e
ss
io
n
a
l
P
a
p
e
r
N
o
9
6
4
,
U.
S
.
G
e
o
lo
g
ica
l
S
u
rv
e
y
,
U.
S
.
G
o
v
e
rn
m
e
n
t
P
ri
n
ti
n
g
o
f
f
ice
,
W
a
sh
in
g
to
n
,
D.C.
[6
]
D.
L
u
,
Q.
W
e
n
g
,
“
A
S
u
rv
e
y
o
f
I
m
a
g
e
Clas
sif
ic
a
ti
o
n
M
e
th
o
d
s
a
n
d
T
e
c
h
n
iq
u
e
s
f
o
r
Im
p
ro
v
in
g
Clas
sif
ic
a
ti
o
n
P
e
rf
o
rm
a
n
c
e
”
,
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Rem
o
te S
e
n
si
n
g
,
V
o
l
.
2
8
,
N
o
.
5
,
p
p
.
8
2
3
-
8
7
0
,
Ja
n
u
a
ry
2
0
0
7
.
[7
]
Ja
n
Kn
o
r
n
,
A
n
d
re
a
s
Ra
b
e
,
Vo
lk
e
r
C.
Ra
d
e
lo
f
f
,
T
o
b
ias
Ku
e
m
m
e
rle,
Ja
c
e
k
Ko
z
a
k
,
P
a
tri
c
k
Ho
ste
rt,
“
L
a
n
d
c
o
v
e
r
M
a
p
p
i
n
g
o
f
L
a
rg
e
a
re
a
s
u
sin
g
Ch
a
in
Clas
sif
ica
ti
o
n
o
f
N
e
ig
h
b
o
ri
n
g
L
a
n
d
sa
t
S
a
telli
te
Im
a
g
e
s”
,
Re
mo
te
S
e
n
si
n
g
o
f
En
v
iro
n
me
n
t
,
V
o
l.
1
1
8
,
p
a
g
e
s 9
5
7
-
9
6
4
,
2
0
0
9
.
[8
]
S
.
V.S
P
ra
sa
d
,
D
r.
T
.
S
a
ty
a
S
a
v
it
ri,
Dr.I.
V
.
M
u
ra
li
Kris
h
n
a
.
,
“
T
e
c
h
n
iq
u
e
s
i
n
Im
a
g
e
Clas
si
f
ic
a
ti
o
n
;
A
S
u
rv
e
y
"
,
Glo
b
a
l
J
o
u
rn
a
l
o
f
Res
e
a
rc
h
e
s
i
n
E
n
g
i
n
e
e
rin
g
:
El
e
c
trica
l
a
n
d
El
e
c
tro
n
ics
En
g
in
e
e
rin
g
“
V
o
l.
1
5
,
No
.
6
,
p
p
.
2
2
4
9
-
4
5
9
6
,
2
0
1
5
.
[9
]
M
.
A
.
He
a
rst
;
Ca
li
f
o
rn
ia
Un
iv
.
,
Be
rk
e
le
y
,
C
A
;
S
.
T
.
Du
m
a
is
;
E.
Os
m
a
n
;
J.
P
latt,
“
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
s”
,
I
EE
E
In
telli
g
e
n
t
S
y
ste
ms
a
n
d
t
h
e
i
r A
p
p
l
ica
ti
o
n
s
(
Vo
lu
m
e
:1
3
,
Iss
u
e
:
4
),
I
S
S
N :
1
0
9
4
-
7
1
6
7
.
[1
0
]
B
S
o
w
m
y
a
a
n
d
B
S
h
e
e
lara
n
i
,
“
L
a
n
d
c
o
v
e
r
c
las
sif
ica
ti
o
n
u
si
n
g
re
fo
rm
e
d
f
u
z
z
y
C
-
m
e
a
n
s
”
,
S
a
d
h
a
n
a
,
Vo
l.
3
6
,
N
o
.
2
,
p
p
.
1
5
3
–
1
6
5
,
2
0
1
1
.
[1
1
]
S
h
iv
a
li
A
.
K
a
r
;
V
ish
a
k
h
a
V
.
Ke
l
k
a
r,
“C
la
ss
if
ica
ti
o
n
o
f
M
u
lt
isp
e
c
t
ra
l
S
a
telli
te
I
ma
g
e
s”
,
A
d
v
a
n
c
e
s
in
T
e
c
h
n
o
lo
g
y
a
n
d
E
n
g
in
e
e
rin
g
(ICA
T
E),
2
0
1
3
I
n
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
2
3
-
2
5
Ja
n
.
2
0
1
3
,
I
S
BN:
9
7
8
-
1
-
4
6
7
3
-
5
6
1
8
-
3.
[1
2
]
Ra
n
z
h
e
Jin
g
;
S
c
h
.
o
f
In
f
.
M
a
n
a
g
e
.
&
En
g
.
,
S
h
a
n
g
h
a
i
Un
iv
.
o
f
F
i
n
a
n
c
e
&
Eco
n
.
,
S
h
a
n
g
h
a
i,
C
h
in
a
;
Yo
n
g
Zh
a
n
g
,
“
A
Vi
e
w
o
f
S
u
p
p
o
rt
Vec
to
r
M
a
c
h
in
e
s
Al
g
o
rit
h
m
o
n
Cl
a
ss
if
ica
t
io
n
Pr
o
b
lem
s”
,
M
u
lt
im
e
d
ia
Co
m
m
u
n
ica
ti
o
n
s
(M
e
d
iac
o
m
),
2
0
1
0
I
n
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
7
-
8
A
u
g
.
2
0
1
0
,
P
ri
n
t
IS
BN:
9
7
8
-
0
-
7
6
9
5
-
4
1
3
6
-
5.
[1
3
]
Ca
m
p
b
e
ll
,
J.B.
,
1
9
8
1
,
“
S
p
a
ti
a
l
C
o
rre
latio
n
Ef
f
e
c
ts
u
p
o
n
A
c
c
u
ra
c
y
o
f
S
u
p
e
rv
ise
d
C
las
sif
ic
a
ti
o
n
o
f
lan
d
c
o
v
e
r”
,
Ph
o
t
o
g
r
a
mm
e
tric E
n
g
i
n
e
e
rin
g
a
n
d
Rem
o
te
S
e
n
sin
g
,
4
7
,
3
5
5
–
3
6
3
.
[1
4
]
M
.
A
.
He
a
rst
;
Ca
li
f
o
rn
ia
Un
iv
.
,
Be
rk
e
le
y
,
C
A
;
S
.
T
.
Du
m
a
is
;
E.
Os
m
a
n
;
J.
P
latt
,
“
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
s”
,
IEE
E
In
telli
g
e
n
t
S
y
ste
m
s a
n
d
t
h
e
i
r A
p
p
l
ica
ti
o
n
s
(
Vo
lu
m
e
:1
3
,
Iss
u
e
:
4
),
I
S
S
N :
1
0
9
4
-
7
1
6
7
.
[1
5
]
Hu
ss
a
in
A
b
u
Da
lb
o
u
h
;
F
a
c
.
o
f
S
c
i.
&
T
e
c
h
n
o
l,
Un
iv
.
S
a
in
s
Isla
m
M
a
la
y
sia
(USIM
),
Nilai,
M
a
la
y
s
ia
;
No
rit
a
M
d
.
No
rw
a
w
i
,
“
Imp
ro
v
e
me
n
t
o
n
A
g
g
l
o
me
ra
ti
v
e
Hie
ra
rc
h
ica
l
Clu
ste
rin
g
Al
g
o
rit
h
m
B
a
se
d
o
n
T
re
e
Da
t
a
S
tru
c
tu
re
wi
t
h
Bi
d
ire
c
ti
o
n
a
l
A
p
p
r
o
a
c
h
”
,
2
0
1
2
T
h
ird
I
n
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
In
telli
g
e
n
t
S
y
ste
m
s
M
o
d
e
ll
in
g
a
n
d
S
im
u
latio
n
8
-
1
0
F
e
b
.
2
0
1
2
,
IS
S
N
:2
1
6
6
-
0
6
6
2
.
[1
6
]
W
a
n
g
F
,
1
9
9
0
,
“
F
u
z
z
y
S
u
p
e
rv
ise
d
C
las
sif
i
c
a
ti
o
n
o
f
Re
m
o
te
S
e
n
sin
g
Im
a
g
e
s”
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Ge
o
-
sc
ien
c
e
a
n
d
Rem
o
te
S
e
n
sin
g
,
2
8
,
1
9
4
–
2
0
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.