T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
2
,
A
p
r
il
2
0
2
1
,
p
p
.
690
~
704
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i2
.
1
8
1
5
5
690
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Ro
o
f
ma
terials
id
entif
ica
tion ba
sed
on pleia
des spe
ct
ra
l
respo
nses
using
s
uperv
ised cla
ss
ifi
ca
tion
Ay
o
m
Widi
pa
m
into
1
,
Yo
ha
n
es F
rido
lin
H
estr
io
2
,
Yuv
it
a
Dia
n Sa
f
it
ri
3
,
Do
nn
a
M
o
nica
4
,
Dedi
I
ra
wa
di
5
,
Ro
k
hm
a
t
ulo
h
6
,
Dj
o
ko
T
riy
o
no
7
,
E
rna
Sri A
di
nin
g
s
i
h
8
1
,
2,
3,
4
,
5,
8
Re
m
o
te
S
e
n
sin
g
Tec
h
n
o
l
o
g
y
a
n
d
Da
ta Cen
ter,
Na
ti
o
n
a
l
I
n
st
it
u
te o
f
Ae
ro
n
a
u
ti
c
s a
n
d
S
p
a
c
e
,
In
d
o
n
e
sia
1,
6,
7
De
p
a
rtme
n
t
o
f
P
h
y
sic
s,
F
a
c
u
lt
y
o
f
M
a
th
e
m
a
ti
c
s a
n
d
S
c
ien
c
e
s,
Un
iv
e
rsity
o
f
I
n
d
o
n
e
sia
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
7
,
2
0
2
0
R
ev
is
ed
No
v
1
2
,
2
0
2
0
Acc
ep
ted
No
v
2
5
,
2
0
2
0
Th
e
c
u
rre
n
t
u
rb
a
n
e
n
v
ir
o
n
m
e
n
t
is
v
e
ry
d
y
n
a
m
ic
a
n
d
a
lwa
y
s
c
h
a
n
g
e
s
b
o
th
p
h
y
sic
a
ll
y
a
n
d
so
c
io
-
e
c
o
n
o
m
ica
l
ly
v
e
ry
q
u
ick
l
y
.
M
o
n
it
o
ri
n
g
u
r
b
a
n
a
re
a
s
is
o
n
e
o
f
th
e
m
o
st
re
lev
a
n
t
iss
u
e
s
re
late
d
to
e
v
a
l
u
a
ti
n
g
h
u
m
a
n
i
m
p
a
c
ts
o
n
e
n
v
iro
n
m
e
n
tal
c
h
a
n
g
e
.
No
wa
d
a
y
s
re
m
o
te
se
n
sin
g
tec
h
n
o
lo
g
y
is
i
n
c
re
a
sin
g
ly
b
e
in
g
u
se
d
in
a
v
a
riety
o
f
a
p
p
li
c
a
ti
o
n
s
i
n
c
lu
d
in
g
m
a
p
p
in
g
a
n
d
m
o
d
e
li
n
g
o
f
u
rb
a
n
a
re
a
s.
Th
e
p
u
rp
o
se
o
f
th
is
p
a
p
e
r
is
to
c
las
sify
t
h
e
P
leia
d
e
s
d
a
ta
fo
r
t
h
e
id
e
n
ti
fica
ti
o
n
o
f
ro
o
f
m
a
teria
ls.
Th
is
c
las
sifica
ti
o
n
is
b
a
se
d
o
n
d
a
ta
fro
m
sa
telli
te
ima
g
e
sp
e
c
tro
sc
o
p
y
re
su
lt
s
with
v
e
ry
h
i
g
h
re
so
lu
t
io
n
.
S
p
e
c
tro
sc
o
p
y
is
a
tec
h
n
iq
u
e
fo
r
o
b
tain
i
n
g
s
p
e
c
tru
m
o
r
wa
v
e
len
g
t
h
s
a
t
e
a
c
h
p
o
s
it
io
n
fro
m
va
rio
u
s
sp
a
ti
a
l
d
a
ta
so
th
a
t
i
m
a
g
e
s
c
a
n
b
e
re
c
o
g
n
ize
d
b
a
se
d
o
n
th
e
ir
re
sp
e
c
ti
v
e
sp
e
c
tral
wa
v
e
len
g
th
s.
Th
e
o
u
tco
m
e
o
f
t
h
is
stu
d
y
is
th
a
t
h
ig
h
-
re
so
lu
ti
o
n
re
m
o
te
se
n
sin
g
d
a
ta
c
a
n
b
e
u
se
d
to
i
d
e
n
t
ify
r
o
o
f
m
a
teri
a
l
a
n
d
c
a
n
m
a
p
fu
rth
e
r
in
th
e
c
o
n
tex
t
o
f
m
o
n
it
o
rin
g
u
r
b
a
n
a
re
a
s.
Th
e
o
v
e
ra
ll
v
a
lu
e
o
f
a
c
c
u
ra
c
y
a
n
d
Ka
p
p
a
C
o
e
fficie
n
t
o
n
t
h
e
m
e
th
o
d
th
a
t
we
u
se
is
e
q
u
a
l
to
9
2
.
9
2
%
a
n
d
0
.
9
0
6
9
.
K
ey
w
o
r
d
s
:
C
las
s
if
icatio
n
Ma
ch
in
e
lear
n
in
g
R
ef
lecta
n
ce
r
esp
o
n
s
e
R
o
o
f
m
ater
ials
Satellite im
ag
er
y
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
:
Ay
o
m
W
id
ip
am
in
to
R
em
o
te
Sen
s
in
g
T
ec
h
n
o
lo
g
y
a
n
d
Data
C
en
ter
Natio
n
al
I
n
s
titu
te
o
f
Aer
o
n
au
t
ics an
d
Sp
ac
e
J
ak
ar
ta,
I
n
d
o
n
esia
E
m
ail:
ay
o
m
.
wid
ip
am
i
n
to
@
lap
an
.
g
o
.
id
1.
I
NT
RO
D
UCT
I
O
N
Ur
b
an
ar
ea
s
a
r
e
n
o
w
a
t
y
p
e
o
f
lan
d
c
o
v
er
th
at
is
ch
a
n
g
in
g
v
e
r
y
r
a
p
id
ly
,
ev
en
th
o
u
g
h
th
ese
ar
ea
s
o
n
ly
m
ee
t
a
lo
w
p
er
ce
n
tag
e
o
f
th
e
en
tire
lan
d
s
u
r
f
ac
e
g
lo
b
ally
[1
,
2]
.
T
h
e
u
r
b
an
en
v
ir
o
n
m
e
n
t
i
s
also
d
y
n
a
m
ic
an
d
ex
p
er
ien
ce
s
r
ap
i
d
p
h
y
s
ical
an
d
s
o
cio
-
ec
o
n
o
m
ic
ch
a
n
g
es
[
3
]
.
Ur
b
an
ar
ea
p
lan
n
in
g
r
e
q
u
ir
e
s
in
f
o
r
m
atio
n
f
r
o
m
v
ar
io
u
s
p
h
en
o
m
en
a
an
d
s
p
atial
ch
ar
ac
ter
is
tics
s
u
ch
as r
o
o
f
m
ater
ials
,
u
r
b
a
n
s
tr
u
ctu
r
es,
an
d
b
u
ild
in
g
r
esil
ien
ce
[
4
]
.
B
u
ilt
-
u
p
lan
d
m
o
n
ito
r
in
g
i
n
m
o
r
e
d
etail
ab
o
u
t
lan
d
u
s
e
will
d
escr
ib
e
th
e
d
ev
elo
p
m
en
t
o
f
th
e
city
ar
ea
[
5
]
.
B
u
ilt
-
u
p
lan
d
m
o
n
ito
r
in
g
t
o
d
etail
th
e
ty
p
e
o
f
r
o
o
f
m
ate
r
ial
ca
n
also
d
escr
ib
e
in
m
o
r
e
d
etail
th
e
ec
o
n
o
m
ic
lev
el
o
f
t
h
e
co
m
m
u
n
ity
.
Kn
o
wled
g
e
o
f
r
o
o
f
m
ater
ials
ty
p
e
s
also
m
ak
es
it
p
o
s
s
ib
le
to
estab
lis
h
q
u
an
tific
atio
n
m
o
n
ito
r
in
g
o
f
p
o
llu
tan
t
em
is
s
io
n
s
f
r
o
m
h
az
a
r
d
o
u
s
m
ater
ials
[
6
]
.
R
o
o
f
m
ater
ials
ty
p
e
in
f
o
r
m
atio
n
also
n
ee
d
ed
f
o
r
th
e
d
ev
el
o
p
m
en
t
o
f
p
h
o
to
v
o
ltaic
p
o
ten
tial,
u
s
e
d
to
b
u
il
d
a
m
icr
o
wea
th
er
m
o
d
el
s
im
u
lato
r
,
an
d
is
n
ee
d
ed
f
o
r
th
e
m
ilit
ar
y
to
d
eter
m
in
e
t
h
e
ty
p
e
o
f
air
attac
k
,
u
r
b
a
n
p
la
n
n
in
g
,
an
d
d
is
aster
ass
es
s
m
en
t
[7
-
9]
Mo
n
ito
r
in
g
u
r
b
a
n
ar
ea
s
is
o
n
e
o
f
th
e
m
o
s
t
r
elev
an
t
is
s
u
es
r
elate
d
to
ev
alu
atin
g
th
e
im
p
a
ct
h
u
m
an
s
h
av
e
o
n
th
e
e
n
v
ir
o
n
m
en
t.
I
n
cu
r
r
en
t
r
em
o
te
s
en
s
in
g
,
im
ag
er
y
h
as
m
an
y
b
e
n
ef
its
,
o
n
e
o
f
wh
ich
ca
n
p
r
o
d
u
ce
th
e
latest
g
eo
g
r
ap
h
ic
d
ata
f
o
r
n
ee
d
s
s
u
ch
as
lan
d
u
s
e
m
ap
p
in
g
,
em
er
g
e
n
cy
r
esp
o
n
s
e
m
an
ag
em
e
n
t,
r
o
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
R
o
o
f m
a
teri
a
ls
id
en
tifi
ca
tio
n
b
a
s
ed
o
n
p
leia
d
es sp
ec
tr
a
l res
p
o
n
s
es u
s
in
g
s
u
p
ervis
ed
…
(
A
yo
m
Wid
ip
a
min
to
)
691
lo
ca
tio
n
s
with
h
ig
h
a
n
d
l
o
w
em
is
s
iv
ity
,
a
b
etter
u
n
d
e
r
s
tan
d
in
g
o
f
g
r
ee
n
in
g
in
itiativ
es,
an
d
s
u
s
tain
ab
le
city
d
ev
elo
p
m
e
n
t
[
1
0
]
.
R
em
o
te
s
en
s
in
g
tech
n
o
lo
g
y
is
in
cr
ea
s
in
g
ly
b
ein
g
u
s
ed
in
a
v
ar
ie
t
y
o
f
ap
p
licatio
n
s
in
clu
d
in
g
m
a
p
p
in
g
a
n
d
m
o
d
e
lin
g
o
f
u
r
b
an
ar
ea
s
.
Dr
iv
en
b
y
tech
n
o
l
o
g
ical
ad
v
an
ce
m
en
ts
an
d
co
m
m
u
n
ity
n
ee
d
s
,
r
em
o
te
s
en
s
in
g
in
u
r
b
a
n
ar
ea
s
is
in
cr
ea
s
in
g
ly
b
ec
o
m
in
g
a
n
ew
ar
en
a
o
f
g
e
o
s
p
atial
tech
n
o
lo
g
y
an
d
h
as
th
e
ap
p
licatio
n
o
f
b
en
ef
its
in
al
l
s
o
cio
-
ec
o
n
o
m
ic
s
ec
to
r
s
[
1
1
]
.
T
h
e
s
p
atial
r
eso
lu
tio
n
u
s
e
d
f
o
r
id
en
tific
atio
n
g
r
ea
tly
in
f
lu
en
ce
s
th
e
o
u
t
p
u
t
p
r
o
d
u
ce
d
.
So
f
ar
,
r
e
m
o
te
s
en
s
in
g
d
ata
th
at
ar
e
wid
ely
u
s
ed
in
lan
d
an
d
u
r
b
an
o
b
s
er
v
atio
n
s
s
till
u
s
e
m
ed
iu
m
r
eso
lu
tio
n
im
ag
e
d
ata
s
u
ch
as
Sen
tin
el
-
2
[
1
2
]
an
d
L
a
n
d
s
at
-
8
[
1
3
]
,
wh
er
e
Sen
tin
el
-
2
an
d
L
a
n
d
s
at
-
8
h
av
e
a
r
eso
lu
tio
n
o
f
a
r
o
u
n
d
1
0
m
an
d
3
0
m
.
T
h
is
r
esu
lts
in
in
ac
c
u
r
ate
id
en
tific
atio
n
b
ec
au
s
e
u
r
b
an
ar
ea
s
ar
e
s
o
v
a
r
ied
an
d
d
y
n
am
ic
th
at
h
ig
h
er
im
ag
e
r
eso
lu
tio
n
is
n
ee
d
ed
.
T
h
is
is
r
e
in
f
o
r
ce
d
b
y
th
e
ex
is
ten
ce
o
f
n
ew
f
ac
ts
o
r
p
ar
a
d
ig
m
s
th
at
v
er
y
h
ig
h
-
r
eso
lu
tio
n
d
ata
n
ee
d
s
to
b
e
a
d
o
p
ted
to
p
r
o
d
u
ce
s
atis
f
ac
to
r
y
im
ag
e
an
aly
s
is
[
1
4
]
.
T
h
e
id
en
tific
atio
n
m
eth
o
d
u
s
e
d
in
th
is
p
ap
er
is
s
p
ec
tr
o
s
co
p
y
-
th
e
s
tu
d
y
o
f
elec
tr
o
m
a
g
n
etic
r
ad
iatio
n
.
I
m
ag
in
g
s
p
ec
tr
o
s
co
p
y
h
as
m
an
y
n
am
es
in
th
e
r
e
m
o
te
s
en
s
in
g
co
m
m
u
n
ity
,
in
cl
u
d
in
g
i
m
a
g
i
n
g
s
p
ec
tr
o
m
etr
y
,
h
y
p
er
s
p
ec
tr
al,
an
d
u
ltra
-
s
p
ec
tr
al
im
ag
in
g
.
Sp
ec
tr
o
s
co
p
y
is
a
tech
n
iq
u
e
f
o
r
o
b
tain
in
g
s
p
ec
tr
u
m
o
r
wav
ele
n
g
th
s
at
ea
ch
p
o
s
itio
n
f
r
o
m
v
ar
io
u
s
s
p
atial
d
ata
s
o
th
at
im
ag
es
ca
n
b
e
r
ec
o
g
n
ized
b
ased
o
n
th
ei
r
r
esp
ec
tiv
e
s
p
ec
t
r
al
wav
elen
g
th
s
.
T
h
e
p
i
ctu
r
e
ca
n
b
e
a
r
o
c
k
in
a
lab
o
r
at
o
r
y
,
a
f
ie
ld
s
tu
d
y
o
n
an
air
p
lan
e,
o
r
a
n
en
tire
p
lan
et
f
r
o
m
a
s
p
ac
ec
r
af
t
o
r
an
E
ar
th
-
b
ased
telesco
p
e.
B
y
an
aly
zin
g
s
p
ec
tr
al
f
ea
tu
r
es,
an
d
th
u
s
s
p
ec
if
ic
ch
em
ical
b
o
n
d
s
in
a
m
ater
ial,
o
n
e
ca
n
m
ap
wh
er
e
th
o
s
e
b
o
n
d
s
o
cc
u
r
,
an
d
t
h
u
s
m
ap
th
e
m
ater
ial
[
1
5
]
.
T
h
e
p
u
r
p
o
s
e
o
f
th
is
p
ap
er
is
to
class
if
y
th
e
ty
p
es
o
f
m
ater
ials
b
ased
o
n
th
eir
s
p
ec
tr
al
r
e
s
p
o
n
s
es,
as
a
b
asis
f
o
r
th
e
id
en
tific
atio
n
o
f
r
o
o
f
m
ater
ials
.
Ou
r
co
n
tr
ib
u
tio
n
is
th
e
u
s
e
o
f
Pleiad
es
d
ata,
v
er
y
h
ig
h
-
r
eso
lu
tio
n
s
atellite
d
ata
with
a
m
u
ltis
p
ec
tr
al
r
eso
lu
tio
n
th
at
h
as n
o
t b
ee
n
u
s
ed
b
y
c
u
r
r
e
n
t c
o
n
d
itio
n
s
.
2.
RE
S
E
ARCH
M
AT
E
R
I
AL
S
AND
M
E
T
H
O
D
2
.
1
.
L
it
er
a
t
ure
s
t
ud
y
Pre
v
io
u
s
r
esear
ch
o
n
th
e
c
o
llectio
n
o
f
s
p
ec
tr
al
lib
r
ar
ies
f
o
r
t
h
e
d
etec
tio
n
o
f
u
r
b
an
m
ater
ial
h
as
b
ee
n
d
is
cu
s
s
ed
in
s
ev
er
al
p
ap
er
s
s
u
ch
as,
in
[
3
]
th
ey
wer
e
c
o
m
b
in
in
g
s
p
ec
tr
al
an
d
s
p
atial
attr
ib
u
tes,
class
if
icat
io
n
is
d
o
n
e
in
two
way
s
n
am
ely
p
er
-
p
ix
el
(
g
u
id
e
d
class
if
icatio
n
)
an
d
o
b
ject
-
b
ased
(
o
r
ien
ted
o
b
ject)
,
in
[
2
]
wh
ich
co
m
p
ar
es
th
e
ab
ilit
y
o
f
d
if
f
er
e
n
t
s
en
s
o
r
s
to
d
etec
t
m
ater
ial
c
o
v
er
i
n
u
r
b
an
ar
ea
s
b
y
ev
al
u
atin
g
an
d
also
a
d
d
in
g
h
y
p
er
s
p
ec
tr
al
s
en
s
o
r
v
alu
es
to
m
ap
u
r
b
an
ar
e
as.
A
n
ew
ap
p
r
o
ac
h
h
as
also
b
ee
n
tak
e
n
to
d
eter
m
in
e
a
n
d
ev
alu
ate
s
tr
o
n
g
s
p
ec
tr
al
f
ea
tu
r
es
ag
ain
s
t
s
p
ec
tr
al
o
v
er
lap
s
b
etwe
en
m
ater
ial
class
es
an
d
cl
ass
v
ar
iab
ilit
y
.
T
h
e
r
o
b
u
s
tn
ess
o
f
in
ter
ac
tiv
ely
d
et
er
m
in
ed
s
p
ec
tr
al
f
ea
tu
r
es
was
ev
alu
ated
b
y
s
ep
ar
a
tio
n
an
aly
s
is
.
T
h
is
m
eth
o
d
is
d
o
n
e
b
ased
o
n
th
e
c
o
n
f
u
s
io
n
m
atr
ix
f
o
r
ea
c
h
m
ater
ial
th
at
is
ca
lcu
lated
f
r
o
m
ea
c
h
class
if
icatio
n
r
esu
lt
[
1
6
]
.
DAI
S
(
d
ig
ital
ae
r
ial
im
ag
in
g
s
p
ec
tr
o
m
eter
s
)
Hy
p
er
s
p
ec
t
r
al
d
ata
an
d
R
OSI
S
(
r
ef
lectiv
e
o
p
tical
s
y
s
tem
im
ag
in
g
s
p
ec
tr
o
m
eter
s
)
h
av
e
b
ee
n
u
s
ed
f
o
r
d
ata
ex
tr
ac
tio
n
with
s
ev
er
al
ap
p
r
o
ac
h
es
f
o
r
u
r
b
an
a
r
ea
s
(
s
p
atial
in
f
o
r
m
atio
n
)
an
d
n
o
n
-
u
r
b
an
ar
ea
s
(
s
p
ec
tr
al
in
f
o
r
m
atio
n
)
[
1
7
]
.
R
esear
ch
in
[
1
8
]
h
as
d
e
v
elo
p
ed
a
u
to
m
ated
m
eth
o
d
s
f
o
r
h
y
p
er
s
p
ec
tr
al
i
m
a
g
e
an
aly
s
is
th
at
ca
n
ex
p
lo
it
s
p
ec
tr
al
an
d
s
p
atial
in
f
o
r
m
atio
n
co
n
ten
t
th
o
r
o
u
g
h
ly
f
r
o
m
d
ata
to
d
is
tin
g
u
is
h
ty
p
e
s
o
f
u
r
b
an
s
u
r
f
ac
e
co
v
er
i
n
g
s
u
s
in
g
m
ath
em
atica
l
m
o
d
els
th
at
ca
n
r
ed
u
ce
th
e
v
alu
e
s
o
th
at
it
will
b
e
p
o
s
s
ib
le
en
d
m
em
b
er
c
o
m
b
in
atio
n
s
f
o
r
ea
ch
p
ix
el
b
y
in
tr
o
d
u
cin
g
p
u
r
e
s
p
ec
tr
al
s
ee
d
lin
g
s
an
d
r
eg
is
ter
in
g
p
o
s
s
ib
le
en
d
m
em
b
er
co
m
b
in
atio
n
s
f
o
r
ad
jace
n
t p
ix
els in
th
e
u
n
m
i
x
in
g
p
r
o
ce
d
u
r
e.
R
esear
ch
co
n
d
u
cted
in
[
6
]
cr
ea
tes
a
s
u
p
er
s
p
ec
tr
al
s
en
s
o
r
d
esig
n
f
o
r
th
e
class
if
icatio
n
o
f
u
r
b
an
m
ater
ial,
b
an
d
s
elec
tio
n
is
ca
r
r
ied
o
u
t
to
id
en
tif
y
th
e
m
o
s
t
o
p
tim
al
s
u
b
s
et
o
f
b
an
d
s
to
class
if
y
u
r
b
an
m
a
p
s
in
th
e
co
n
tex
t
o
f
s
u
p
er
s
p
ec
tr
a
s
en
s
o
r
d
esig
n
b
y
co
llectin
g
s
p
ec
tr
al
ca
teg
o
r
ies
f
o
r
u
r
b
an
m
ater
ial
f
r
o
m
s
ev
e
n
s
p
ec
tr
al
lib
r
ar
ies
(
u
s
in
g
h
y
p
e
r
s
p
ec
tr
al
d
ata
an
d
g
r
o
u
n
d
tr
u
t
h
)
.
T
h
e
in
v
ar
ian
t
f
o
r
m
u
la
f
o
r
s
p
ec
tr
al
im
ag
es
o
f
n
atu
r
al
o
b
jects
r
etain
s
s
p
ec
tr
al
in
f
o
r
m
atio
n
an
d
d
o
es
n
o
t
d
if
f
er
f
r
o
m
h
i
g
h
lig
h
ts
,
s
h
ad
o
ws
,
s
u
r
f
ac
e
g
eo
m
etr
y
,
an
d
lig
h
tin
g
i
n
ten
s
ity
an
d
also
p
r
o
v
es
th
at
co
n
v
en
tio
n
al
s
p
ec
tr
al
in
v
ar
ian
t
tech
n
i
q
u
es
ca
n
b
e
ap
p
lied
to
m
etals
o
th
er
t
h
an
d
ielec
tr
ic
o
b
jects
[
1
9
]
.
T
h
e
co
m
b
in
ed
p
er
f
o
r
m
a
n
ce
o
f
d
ata
m
in
in
g
(
DM
)
an
d
o
b
ject
-
b
ased
im
ag
e
an
aly
s
i
s
(
OB
I
A)
d
ev
elo
p
s
p
r
o
ce
d
u
r
es
th
at
co
n
tr
i
b
u
te
to
a
u
to
m
atic
k
n
o
wled
g
e
d
is
co
v
er
y
an
d
m
ap
p
in
g
o
f
u
r
b
an
s
u
r
f
ac
e
m
ater
ial
f
r
o
m
im
ag
es
o
f
lar
g
e
h
y
p
er
s
p
ec
tr
a
l
s
p
ac
e
f
ea
tu
r
es
with
an
ac
c
u
r
ac
y
o
f
8
8
%
[
2
0
]
.
R
esear
ch
in
[
2
1
]
h
as
b
u
ilt
a
s
p
ec
tr
al
lib
r
ar
y
o
f
u
r
b
an
m
ater
i
als
an
d
u
s
es
th
em
to
an
aly
ze
u
r
b
an
en
v
ir
o
n
m
en
ts
with
a
co
m
p
r
eh
en
s
iv
e
co
llectio
n
o
f
s
p
ec
tr
al
r
ef
lecta
n
ce
o
f
v
ar
io
u
s
u
r
b
an
m
ater
ials
,
r
esu
ltin
g
in
an
ac
cu
r
ac
y
o
f
7
3
%
u
s
in
g
m
a
x
im
u
m
lik
elih
o
o
d
.
T
h
en
,
f
r
o
m
s
tu
d
ies
co
n
d
u
c
ted
b
y
[
2
2
]
,
s
p
ec
tr
o
m
eter
s
an
d
VNI
R
(
v
is
ib
le
an
d
alm
o
s
t
in
f
r
ar
ed
)
ca
m
er
as
p
r
o
d
u
ce
d
if
f
e
r
en
t
s
p
ec
tr
al
r
e
f
lecta
n
ce
v
alu
es
ev
e
n
th
o
u
g
h
th
e
ta
r
g
ets
ar
e
th
e
s
am
e
.
Fig
u
r
e
1
s
h
o
ws
th
e
s
p
ec
tr
al
p
r
o
f
ile
o
f
th
e
r
o
o
f
an
d
s
u
r
f
ac
e
m
ater
ial,
an
d
Fig
u
r
e
2
s
h
o
ws
th
e
s
u
r
f
ac
e
r
ef
lectio
n
o
f
a
s
elec
tiv
e
u
r
b
an
s
u
r
f
ac
e
m
ater
ial
r
e
co
r
d
e
d
b
y
th
e
H
y
p
er
s
p
ec
tr
al
H
y
Ma
p
s
en
s
o
r
as
a
r
ef
e
r
en
ce
f
o
r
th
is
s
tu
d
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
2
,
Ap
r
il 2
0
2
1
:
6
9
0
-
704
692
Fig
u
r
e
1
.
Sp
ec
tr
al
p
r
o
f
iles
o
f
r
o
o
f
an
d
s
u
r
f
ac
e
m
ater
ials
[
2
3
]
Fig
u
r
e
2
.
At
s
u
r
f
ac
e
r
eflec
tan
c
e
o
f
s
elec
tiv
e
u
r
b
a
n
s
u
r
f
ac
e
m
ater
ials
r
ec
o
r
d
ed
b
y
th
e
h
y
p
e
r
s
p
ec
tr
al
Hy
Ma
p
s
en
s
o
r
[
2
4
]
2
.
2
.
L
o
ca
t
io
n a
nd
da
t
a
T
h
e
r
esear
ch
ar
ea
is
lo
ca
ted
a
t
th
e
R
em
o
te
Sen
s
in
g
T
ec
h
n
o
l
o
g
y
an
d
Data
C
en
ter
,
Natio
n
a
l
I
n
s
titu
te
o
f
Aer
o
n
au
tics
an
d
Sp
ac
e,
E
a
s
t
J
ak
ar
ta.
R
o
o
f
s
am
p
lin
g
is
c
ar
r
ied
o
u
t
in
th
e
R
em
o
te
Sen
s
in
g
T
ec
h
n
o
lo
g
y
an
d
Data
C
en
ter
b
u
ild
in
g
wh
ich
h
as
a
v
ar
iety
o
f
m
ater
ials
in
clu
d
in
g
alu
m
i
n
u
m
r
o
o
f
,
asb
e
s
to
s
,
ce
r
am
ic
tile,
co
n
cr
ete,
a
n
d
s
an
d
m
etal
tile.
Fig
u
r
e
3
s
h
o
ws
th
e
im
a
g
e
w
e
u
s
ed
in
t
h
is
s
tu
d
y
.
T
h
e
n
o
is
e
co
u
ld
af
f
ec
t
d
ata
q
u
ality
[
2
5
,
26
,
27
,
28
,
29]
,
esp
ec
ially
im
ag
e
q
u
ality
in
th
is
r
esear
ch
.
I
n
s
tead
o
f
d
is
cu
s
s
in
g
lo
w
-
q
u
ality
im
ag
es,
we
ch
o
o
s
e
h
ig
h
-
r
eso
l
u
tio
n
im
ag
es.
T
h
er
e
f
o
r
e,
t
h
e
d
ata
u
s
ed
f
o
r
t
h
is
s
tu
d
y
is
clo
u
d
-
f
r
ee
im
ag
e
d
ata
f
r
o
m
t
h
e
Pleiad
es
s
atellite
o
b
tain
ed
o
n
1
Ma
y
2
0
1
8
.
R
ep
r
es
en
tativ
e
s
p
atial
an
d
tem
p
o
r
al
r
eso
lu
tio
n
s
en
s
o
r
s
f
o
r
th
e
s
atellite
ar
e
d
escr
ib
e
d
in
T
ab
le
1
.
T
ab
le
1
.
R
ep
r
esen
tativ
e
s
p
atia
l a
n
d
tem
p
o
r
al
r
eso
lu
tio
n
s
en
s
o
r
s
[
3
0
]
S
e
n
s
o
r
B
a
n
d
T
y
p
e
S
p
a
t
i
a
l
R
e
s
o
l
u
t
i
o
n
G
l
o
b
a
l
R
e
v
i
si
t
C
y
c
l
e
O
p
e
r
a
t
i
o
n
a
l
P
e
r
i
o
d
A
c
c
e
ss
P
l
e
i
a
d
e
s
P
a
n
c
h
r
o
ma
t
i
c
M
u
l
t
i
s
p
e
c
t
r
a
l
5
0
c
m
2
m
D
a
i
l
y
2
0
1
1
-
p
r
e
s
e
n
t
(
P
l
e
i
a
d
e
s
-
1
A
)
2
0
1
2
-
p
r
e
s
e
n
t
(
P
l
e
i
a
d
e
s
-
1
B
)
C
o
mm
e
r
c
i
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
R
o
o
f m
a
teri
a
ls
id
en
tifi
ca
tio
n
b
a
s
ed
o
n
p
leia
d
es sp
ec
tr
a
l res
p
o
n
s
es u
s
in
g
s
u
p
ervis
ed
…
(
A
yo
m
Wid
ip
a
min
to
)
693
Fig
u
r
e
3
.
Stu
d
y
ar
ea
2
.
3
.
M
et
ho
d
T
h
e
m
eth
o
d
t
h
at
we
p
r
o
p
o
s
e
was
ca
r
r
ied
o
u
t
as
f
o
llo
ws.
First,
we
o
b
tain
ed
th
e
d
ata
u
s
ed
in
th
is
s
tu
d
y
,
Pleiad
es
s
atell
ite
im
ag
e
r
y
.
W
e
f
ilter
d
ata
to
0
%
clo
u
d
co
v
er
with
an
an
g
le
o
f
v
iew
th
at
less
th
an
2
0
°.
T
h
e
d
ata
th
en
p
r
o
ce
s
s
ed
in
to
a
p
an
-
s
h
ar
p
en
ed
m
u
ltis
p
ec
tr
al
(
PMS)
u
s
in
g
th
e
E
h
ler
s
m
et
h
o
d
[
3
1
]
.
T
h
e
n
e
x
t
s
tep
o
f
o
u
r
m
et
h
o
d
is
to
g
et
tr
a
in
in
g
d
ata
b
y
s
elec
tin
g
p
ix
els f
r
o
m
th
e
d
a
ta.
T
r
ain
in
g
d
ata
is
s
elec
ted
b
y
clu
s
ter
s
am
p
lin
g
m
eth
o
d
an
d
s
am
p
lin
g
with
a
h
ar
m
o
n
io
u
s
s
y
s
te
m
wh
er
e
ea
ch
s
am
p
le
h
as
p
ix
el
s
p
ac
es.
C
lu
s
ter
s
am
p
lin
g
is
u
s
ed
to
ca
teg
o
r
ize
m
ater
ial
ty
p
es
wh
ile
s
y
s
tem
a
tic
s
am
p
lin
g
is
u
s
ed
to
tak
e
t
h
e
r
ep
r
esen
tatio
n
o
f
ad
jace
n
t
p
ix
els.
W
e
s
elec
ted
t
r
ain
in
g
d
ata
f
o
r
ea
ch
ty
p
e
o
f
r
o
o
f
m
ater
ial,
n
am
el
y
alu
m
in
u
m
r
o
o
f
,
asb
esto
s
,
ce
r
am
ic
tile,
co
n
cr
ete,
an
d
s
an
d
m
etal
tile.
C
lu
s
ter
in
g
was
d
o
n
e
u
s
in
g
a
m
u
lti
-
co
r
e
clu
s
ter
,
th
e
u
s
e
o
f
th
e
clu
s
ter
is
au
to
m
atic
in
th
ese
f
u
n
ctio
n
s
.
T
h
e
lib
r
a
r
y
th
a
t
we
u
s
ed
is
p
ar
ticu
lar
ly
u
s
ef
u
l
to
s
p
ee
d
u
p
co
m
p
u
tatio
n
s
in
f
u
n
ctio
n
s
lik
e
p
r
ed
ict,
in
ter
p
o
late,
an
d
p
e
r
h
a
p
s
ca
lc
[
3
2
]
.
T
h
e
n
ex
t
s
tag
e
is
th
e
s
elec
tio
n
o
f
f
ea
tu
r
es
b
y
ex
tr
ac
tin
g
th
e
r
ef
lecta
n
ce
v
alu
e
o
f
ea
ch
p
i
x
el
o
f
th
e
im
ag
e.
At
th
is
s
tag
e,
th
e
v
al
u
e
o
f
ea
ch
p
ix
el
(
r
an
g
e
1
6
p
ix
els
0
-
6
5
5
3
5
)
o
f
th
e
d
ata
o
b
tain
ed
is
r
ea
d
to
ca
lcu
late
th
e
em
is
s
io
n
an
d
r
e
f
lectio
n
v
alu
es
u
s
in
g
f
o
r
m
u
la
(
3
)
an
d
(
4
)
.
Af
te
r
d
ata
co
ll
ec
tio
n
an
d
f
ea
tu
r
e
s
elec
tio
n
,
n
ex
t
is
th
e
clas
s
if
i
ca
tio
n
u
s
in
g
th
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
class
if
ic
atio
n
m
eth
o
d
-
a
m
ac
h
in
e
lear
n
in
g
(
ML
)
-
b
ased
class
i
fier
wh
ich
is
estab
lis
h
e
d
b
ased
o
n
th
e
s
tatis
tical
lear
n
in
g
th
eo
r
y
[
3
3
]
.
I
n
th
is
s
tu
d
y
,
we
u
s
ed
SVM
s
in
ce
SVM
is
r
ar
ely
u
s
ed
f
o
r
r
em
o
te
s
en
s
in
g
im
ag
e
class
if
icatio
n
.
Fo
r
th
e
SVM
ML
m
o
d
els,
we
ad
a
p
t
th
e
m
o
d
els
f
r
o
m
r
a
n
d
o
m
f
o
r
est
ML
f
r
o
m
[
3
4
]
,
we
tak
e
th
e
s
p
ec
ies
d
is
tr
ib
u
tio
n
m
o
d
elin
g
t
o
a
p
p
ly
it
u
s
in
g
SVM
f
o
r
ML
.
W
e
u
s
e
C
-
class
if
icat
io
n
(
m
u
lti
-
class
cla
s
s
if
icatio
n
)
f
o
r
th
e
SVM
-
ty
p
e,
th
e
SVM
k
er
n
el
is
r
ad
ial
b
asis
f
u
n
ctio
n
[
3
5
,
36
,
37]
.
T
h
e
C
-
class
if
icatio
n
an
d
SVM
-
k
er
n
el
(
r
ad
ial
b
asis
f
u
n
ctio
n
)
f
o
r
m
u
las ar
e
e
x
p
r
ess
ed
in
th
e
(
1
)
an
d
(
2
)
.
C
-
class
if
ica
tio
n
:
min
1
2
⊤
−
⊤
.
.
0
≤
≤
,
I
=
1
,
…
,
,
(
1
)
⊤
=
0
,
wh
er
e
e
is
th
e
u
n
it
v
ec
to
r
,
C
is
th
e
u
p
p
er
b
o
u
n
d
,
Q
is
an
l
by
l
p
o
s
itiv
e
s
em
id
ef
in
ite
m
atr
ix
,
Q
ij
≡
y
i
y
j
K
(
x
i
,
x
j
)
,
an
d
K
(
x
i
, x
j
)
≡
ϕ(
x
i
)
⊤
ϕ(
x
j
)
is
th
e
k
er
n
el
[
3
8
]
.
SVM
-
Ker
n
e
l,
r
ad
ial
b
asis
f
u
n
ctio
n
[
3
9
]
:
(
1
,
2
)
=
e
xp
(
−
‖
1
−
2
‖
2
)
(
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
2
,
Ap
r
il 2
0
2
1
:
6
9
0
-
704
694
W
e
clas
s
if
y
o
u
r
d
ata
in
to
f
iv
e
class
e
s
o
f
ty
p
es
o
f
r
o
o
f
m
at
er
ials
,
n
am
ely
a
lu
m
in
u
m
r
o
o
f
,
asb
esto
s
,
ce
r
am
ic
tile,
co
n
cr
ete,
a
n
d
s
an
d
m
etal
tile.
Fin
ally
,
th
e
e
v
alu
atio
n
wa
s
d
o
n
e
b
y
v
is
u
ally
c
o
m
p
a
r
in
g
th
e
r
esu
lts
o
f
th
e
class
if
icatio
n
with
s
atell
ite
im
ag
es.
T
h
e
wh
o
le
s
tu
d
y
m
eth
o
d
is
illu
s
tr
ated
in
th
e
d
iag
r
am
i
n
Fig
u
r
e
4
.
Fig
u
r
e
4
.
R
esear
ch
d
ia
g
r
am
2
.
3
.
1
.
P
M
S P
ro
ce
s
s
PMS
wa
s
d
o
n
e
with
th
e
f
o
llo
win
g
s
tep
s
:
−
I
n
p
u
t
d
ata
th
at
co
n
tai
n
p
an
ch
r
o
m
atic
d
ata
lev
el
s
en
s
o
r
p
r
im
ar
y
,
m
u
ltis
p
ec
tr
al
d
ata
lev
el
s
en
s
o
r
p
r
im
ar
y
,
an
d
d
ig
ital
elev
atio
n
m
o
d
el
(
DE
M)
d
ata
o
f
th
e
ar
ea
o
f
in
te
r
est
(
AOI
)
r
eg
io
n
with
th
e
s
am
e
r
eso
lu
tio
n
as
p
an
ch
r
o
m
atic
r
eso
lu
tio
n
o
r
h
ig
h
er
.
−
T
h
e
o
r
th
o
r
e
ctif
icatio
n
p
r
o
ce
s
s
u
s
es DE
M
d
ata
.
−
Pan
s
h
ar
p
en
in
g
p
r
o
ce
s
s
b
etwe
e
n
p
an
c
h
r
o
m
atic
an
d
m
u
ltis
p
ec
tr
al.
T
h
e
p
an
s
h
ar
p
e
n
in
g
m
eth
o
d
u
s
ed
was
th
e
Un
iv
er
s
ity
o
f
New
B
r
u
n
s
wick
’
s
wh
ich
was
d
e
v
elo
p
ed
b
y
Dr
.
Yu
n
Z
h
a
n
g
.
T
h
is
m
et
h
o
d
ca
n
p
r
o
d
u
ce
im
a
g
es
with
g
o
o
d
co
n
s
is
ten
t
q
u
ality
f
o
r
all
s
en
s
o
r
s
.
T
h
is
h
as
b
ee
n
ev
alu
ated
an
d
an
aly
ze
d
with
n
in
e
co
m
p
ar
is
o
n
alg
o
r
ith
m
s
[
4
0
]
.
−
C
o
n
v
er
t
Dig
ital Co
u
n
t to
T
o
p
o
f
Atm
o
s
p
h
er
e
(
T
OA)
R
ad
ian
ce
.
Fo
r
a
r
esp
ec
tiv
e
b
an
d
(
b
)
,
th
e
co
n
v
er
s
io
n
o
f
th
e
Dig
ital
C
o
u
n
t
o
f
a
p
ix
el
DC
(
p
)
to
T
OA
r
ad
ian
ce
L
b
(
p
)
(
in
W
·
s
r
−1
·
m
−2
·
μ
m
−1
)
is
d
o
n
e
b
y
th
e
ab
s
o
lu
te
r
ad
io
m
etr
ic
ca
lib
r
atio
n
co
e
f
f
icien
ts
GAI
N
an
d
B
I
AS
[
4
1
]
:
L
b
(
p
)
=
(
)
(
)
+ B
I
A
S
(
b
)
(
3
)
−
T
OA
R
ad
ian
ce
to
T
OA
R
ef
lecta
n
ce
p
r
o
ce
s
s
T
h
e
T
o
p
Of
Atm
o
s
p
h
er
e
(
T
OA)
s
p
ec
tr
al
r
ef
lecta
n
ce
is
th
e
r
atio
o
f
th
e
T
OA
r
a
d
ian
ce
n
o
r
m
alize
d
b
y
th
e
in
co
m
in
g
s
o
lar
ir
r
a
d
ian
ce
[
4
1
]
.
T
h
e
o
u
t
p
u
t d
ata
is
a
p
a
n
-
s
h
ar
p
en
e
d
m
u
ltis
p
ec
tr
al
d
ata
le
v
el
o
r
th
o
s
tan
d
ar
d
.
ρ
b
(
p
)
=
.
(
)
(
)
.
co
s
(
)
(
4
)
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
First,
we
o
b
tain
ed
7
3
4
tr
ain
i
n
g
d
ata
f
r
o
m
Pleiad
es
im
ag
es,
as
s
h
o
wn
in
Fig
u
r
e
5
.
T
h
e
n
u
m
b
er
o
f
d
etailed
tr
ain
in
g
d
ata
p
er
m
at
er
ial
is
s
h
o
wn
in
T
ab
le
2
.
T
h
e
tr
ain
in
g
d
ata
th
at
we
u
s
e
w
as
th
e
ar
ea
th
at
wa
s
d
ir
ec
tly
ex
p
o
s
ed
to
t
h
e
s
u
n
ca
u
g
h
t
i
n
th
e
s
atellite
d
ata.
Fo
r
t
h
e
f
ea
tu
r
es
u
s
ed
in
t
h
e
class
if
icatio
n
,
s
p
ec
tr
al
d
ata
wer
e
o
b
tain
ed
f
o
r
5
(
f
iv
e)
ty
p
es
o
f
r
o
o
f
m
ater
ial
f
r
o
m
h
ig
h
-
r
eso
lu
tio
n
s
atellite
im
ag
er
y
,
i.e
.
alu
m
in
u
m
r
o
o
f
,
asb
esto
s
,
ce
r
am
ic
tile,
co
n
cr
et
e,
an
d
s
an
d
m
etal
tile.
T
h
e
m
ea
s
u
r
em
en
t
o
f
th
e
r
o
o
f
s
h
o
ws
a
u
n
iq
u
e
s
p
ec
tr
al
s
ig
n
wh
ich
ca
n
b
e
s
ee
n
in
T
a
b
le
3
.
Fo
r
b
etter
u
n
d
e
r
s
tan
d
in
g
,
we
a
n
aly
ze
t
h
e
s
p
ec
tr
al
r
esp
o
n
s
e
lin
e
g
r
a
p
h
s
as
s
h
o
wn
in
Fig
u
r
e
6
(
a)
.
R
ef
lect
an
ce
is
u
s
ed
in
th
e
an
aly
s
is
s
in
ce
r
e
f
lecta
n
ce
is
a
p
r
o
p
er
t
y
o
f
th
e
tar
g
et
m
ater
ial
its
elf
.
I
t
m
ea
s
u
r
es
h
o
w
m
u
ch
en
er
g
y
(
as
a
p
er
ce
n
t
)
a
s
u
r
f
ac
e
r
ef
lects
at
a
s
p
ec
if
ic
wav
elen
g
th
[
4
2
]
.
Ma
n
y
s
u
r
f
ac
es
r
ef
lect
a
d
if
f
er
e
n
t
a
m
o
u
n
t
o
f
e
n
er
g
y
in
d
if
f
er
e
n
t
p
o
r
tio
n
s
o
f
t
h
e
s
p
ec
tr
u
m
.
T
h
ese
d
if
f
er
e
n
ce
s
in
r
ef
lecta
n
ce
m
a
k
e
it
p
o
s
s
ib
le
to
id
en
tify
d
if
f
er
en
t
ea
r
th
s
u
r
f
ac
e
f
ea
tu
r
es
o
r
m
ater
ials
b
y
an
aly
zin
g
t
h
eir
s
p
ec
tr
a
l
r
ef
lecta
n
ce
s
ig
n
atu
r
es
[
4
3
]
.
Sp
ec
tr
al
r
ef
lecta
n
ce
v
alu
es
ar
e
also
ea
s
ier
to
r
ea
d
an
d
u
n
d
er
s
tan
d
as
a
class
if
icatio
n
b
ec
au
s
e
th
ey
h
a
v
e
a
r
an
g
e
v
alu
e
o
f
0
-
1.
Fro
m
th
e
s
ca
tter
p
lo
t
in
Fig
u
r
e
6
(
b
)
,
it
ca
n
b
e
s
ee
n
th
at
th
e
s
p
ec
tr
al
v
alu
es
we
ca
lcu
lated
ar
e
q
u
ite
n
ar
r
o
w
f
o
r
ea
c
h
b
an
d
ex
ce
p
t
f
o
r
th
e
s
p
ec
tr
al
r
esp
o
n
s
e
o
f
NI
R
b
an
d
s
th
at
h
av
e
a
lar
g
e
en
o
u
g
h
s
tan
d
ar
d
d
ev
iatio
n
.
I
t
s
h
o
ws
th
at
th
e
s
p
ec
tr
al
r
esp
o
n
s
e
in
t
h
e
r
ed
b
an
d
is
g
en
er
ally
lo
wer
th
a
n
t
h
e
o
th
er
b
an
d
s
a
n
d
in
cr
ea
s
es
s
ig
n
if
ican
tly
in
th
e
NI
R
b
an
d
.
T
h
e
s
ca
tter
p
lo
t
r
esu
lt
b
etwe
en
th
e
b
an
d
s
also
s
h
o
ws
th
at
th
e
o
v
er
all
b
an
d
s
h
o
ws
a
s
ig
n
if
ican
t
d
is
tr
ib
u
tio
n
to
th
e
NI
R
b
an
d
,
esp
ec
ially
th
e
r
ed
b
a
n
d
th
at
h
as
th
e
m
o
s
t
d
o
m
in
a
n
t
r
esu
lt
wh
en
f
ac
in
g
th
e
NI
R
b
an
d
.
T
h
is
b
eh
av
io
r
s
ee
m
s
to
b
e
u
n
iq
u
e
in
ev
e
r
y
m
ater
ial
a
n
d
th
er
ef
o
r
e
ca
n
b
e
u
s
ed
as a
f
ea
tu
r
e
to
d
is
tin
g
u
is
h
b
etwe
en
o
n
e
m
ater
ial
an
d
an
o
th
er
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
R
o
o
f m
a
teri
a
ls
id
en
tifi
ca
tio
n
b
a
s
ed
o
n
p
leia
d
es sp
ec
tr
a
l res
p
o
n
s
es u
s
in
g
s
u
p
ervis
ed
…
(
A
yo
m
Wid
ip
a
min
to
)
695
Af
ter
th
at,
we
p
r
o
ce
ed
with
t
h
e
class
if
icatio
n
u
s
in
g
SVM.
Usi
n
g
th
e
tr
ain
in
g
d
ata,
we
o
b
tain
ed
a
class
if
i
ca
tio
n
m
o
d
el
with
an
a
cc
u
r
ac
y
o
f
9
4
.
0
1
%
an
d
a
Kap
p
a
co
ef
f
icie
n
t
o
f
0
.
9
1
6
5
.
T
h
e
n
th
is
SVM
Mo
d
el
is
u
s
ed
to
class
if
y
Pleiad
es
d
ata
th
at
is
u
s
ed
co
m
p
r
eh
en
s
iv
ely
,
th
e
r
esu
lts
o
f
wh
ich
ar
e
s
h
o
wn
in
Fig
u
r
e
6
(
c)
.
Pleiad
es
d
ata
u
s
ed
wer
e
s
u
cc
ess
f
u
lly
class
if
ied
i
n
to
f
iv
e
class
es,
wh
er
e
class
es
1
,
2
,
3
,
4
,
an
d
5
ea
ch
r
ep
r
esen
ted
a
lu
m
in
u
m
r
o
o
f
,
as
b
esto
s
,
ce
r
am
ic
tile,
co
n
cr
ete,
an
d
s
an
d
m
etal
tile.
W
e
p
er
f
o
r
m
SVM
co
n
f
u
s
io
n
m
atr
ix
ca
lcu
latio
n
s
to
s
h
o
w
t
h
e
r
esu
lts
o
f
m
ac
h
in
e
lear
n
in
g
in
id
en
tify
i
n
g
r
o
o
f
m
ate
r
ial
(
T
ab
l
e
4
)
.
I
t
ca
n
b
e
s
ee
n
in
th
e
tab
le
th
at
SVM
d
etec
ts
d
if
f
er
en
ce
s
in
tr
ain
in
g
d
ata
with
ML
r
esu
lts
o
n
asb
esto
s
an
d
s
an
d
m
etal
tile
m
ater
ials
.
W
e
al
s
o
p
r
o
v
id
e
S
VM
s
tati
s
tics
p
er
clas
s
an
d
SVM
r
ec
eiv
er
o
p
er
atin
g
ch
ar
ac
ter
is
tic
(
R
O
C
)
clas
s
as
s
h
o
wn
in
T
a
b
le
5
an
d
Fig
u
r
e
7
.
I
n
g
en
e
r
al,
th
e
class
if
icatio
n
co
r
r
ec
tly
id
en
tifie
s
th
e
m
ater
ial
ar
o
u
n
d
th
e
s
tu
d
y
ar
ea
d
escr
ib
e
d
in
Fig
u
r
e
8
.
Fig
u
r
e
5
.
T
r
ain
in
g
d
ata
co
llect
ed
f
r
o
m
th
e
im
ag
e
T
ab
le
2
.
Deta
ils
o
f
tr
ain
in
g
d
at
a
u
s
ed
M
a
t
e
r
i
a
l
s
P
l
e
i
a
d
e
s
t
r
a
i
n
i
n
g
d
a
t
a
A
l
u
m
i
n
u
m
R
o
o
f
1
1
4
A
sb
e
s
t
o
s
1
1
9
C
e
r
a
m
i
c
Ti
l
e
1
4
3
C
o
n
c
r
e
t
e
42
S
a
n
d
M
e
t
a
l
T
i
l
e
3
1
6
T
ab
l
e
3
.
Un
iq
u
e
s
p
ec
tr
al
s
ig
n
s
o
f
ea
ch
m
ater
ial
M
a
t
e
r
i
a
l
s
B
a
n
d
R
e
f
l
e
c
t
a
n
c
e
V
a
l
u
e
B
l
u
e
G
r
e
e
n
R
e
d
N
I
R
A
l
u
m
i
n
u
m
R
o
o
f
0
.
3
2
2
0
.
2
9
9
0
.
2
7
6
0
.
3
3
4
A
sb
e
s
t
o
s
0
.
2
3
4
0
.
2
0
4
0
.
1
8
3
0
.
2
3
7
C
e
r
a
m
i
c
Ti
l
e
0
.
2
1
5
0
.
1
8
4
0
.
1
9
2
0
.
2
6
0
C
o
n
c
r
e
t
e
0
.
3
1
7
0
.
2
7
5
0
.
2
5
3
0
.
3
9
0
S
a
n
d
M
e
t
a
l
T
i
l
e
0
.
2
2
0
0
.
1
8
7
0
.
1
6
7
0
.
2
2
2
T
ab
l
e
4
.
SVM
co
n
f
u
s
io
n
m
atr
i
x
P
r
e
d
i
c
t
i
o
n
A
c
t
u
a
l
A
l
u
m
i
n
u
m
A
sb
e
s
t
o
s
C
e
r
a
m
i
c
t
i
l
e
C
o
n
c
r
e
t
e
S
a
n
d
m
e
t
a
l
t
i
l
e
A
l
u
m
i
n
u
m
1
1
4
0
0
0
0
A
sb
e
s
t
o
s
0
90
0
0
15
C
e
r
a
m
i
c
Ti
l
e
0
0
1
4
3
0
0
C
o
n
c
r
e
t
e
0
0
0
42
0
S
a
n
d
m
e
t
a
l
t
i
l
e
0
29
0
0
3
0
1
T
ab
l
e
5
.
SVM
s
tatis
tics
b
y
cla
s
s
C
l
a
s
s:
A
l
u
m
i
n
u
m
A
sb
e
s
t
o
s
C
e
r
a
m
i
c
t
i
l
e
C
o
n
c
r
e
t
e
S
a
n
d
m
e
t
a
l
t
i
l
e
S
e
n
s
i
t
i
v
i
t
y
1
.
0
0
0
0
0
.
7
5
6
3
1
.
0
0
0
0
1
.
0
0
0
0
0
0
.
9
5
2
5
S
p
e
c
i
f
i
c
i
t
y
1
.
0
0
0
0
0
.
9
7
5
6
1
.
0
0
0
0
1
.
0
0
0
0
0
0
.
9
3
0
6
P
o
s Pr
e
d
V
a
l
u
e
1
.
0
0
0
0
0
.
8
5
7
1
1
.
0
0
0
0
1
.
0
0
0
0
0
0
.
9
1
2
1
N
e
g
P
r
e
d
V
a
l
u
e
1
.
0
0
0
0
0
.
9
5
3
9
1
.
0
0
0
0
1
.
0
0
0
0
0
0
.
9
6
2
9
P
r
e
v
a
l
e
n
c
e
0
.
1
5
5
3
0
.
1
6
2
1
0
.
1
9
4
8
0
.
0
5
7
2
2
0
.
4
3
0
5
D
e
t
e
c
t
i
o
n
R
a
t
e
0
.
1
5
5
3
0
.
1
2
2
6
0
.
1
9
4
8
0
.
0
5
7
2
2
0
.
4
1
0
1
D
e
t
e
c
t
i
o
n
P
r
e
v
a
l
e
n
c
e
0
.
1
5
5
3
0
.
1
4
3
1
0
.
1
9
4
8
0
.
0
5
7
2
2
0
.
4
4
9
6
B
a
l
a
n
c
e
d
A
c
c
u
r
a
c
y
1
.
0
0
0
0
0
.
8
6
6
0
1
.
0
0
0
0
1
.
0
0
0
0
0
0
.
9
4
1
6
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
2
,
Ap
r
il 2
0
2
1
:
6
9
0
-
704
696
(
a)
(
b
)
(
c)
Fig
u
r
e
6
.
(
a
)
Sp
ec
tr
al
r
e
f
lecta
n
ce
p
r
o
f
ile
f
r
o
m
t
h
e
Pleiad
es
,
(
b
)
Scatter
p
lo
t o
f
s
p
ec
tr
al
f
ea
t
u
r
es
,
(
c)
C
lass
if
icatio
n
r
esu
l
t
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
R
o
o
f m
a
teri
a
ls
id
en
tifi
ca
tio
n
b
a
s
ed
o
n
p
leia
d
es sp
ec
tr
a
l res
p
o
n
s
es u
s
in
g
s
u
p
ervis
ed
…
(
A
yo
m
Wid
ip
a
min
to
)
697
Fro
m
Fig
u
r
e
8
,
we
ca
n
s
ee
th
a
t th
e
r
esu
lts
ar
e
lar
g
ely
d
if
f
icu
lt to
d
is
tin
g
u
is
h
b
etwe
en
s
an
d
m
etal
tile
,
co
n
cr
ete,
an
d
v
eg
etatio
n
.
T
h
is
is
ca
u
s
ed
b
y
s
ev
er
al
f
ac
to
r
s
,
am
o
n
g
o
t
h
er
s
;
atm
o
s
p
h
er
ic
co
r
r
ec
tio
n
th
at
h
as
n
o
t
b
ee
n
ap
p
lied
,
Pleiad
es
im
ag
er
y
th
at
o
n
ly
u
s
es
4
(
f
o
u
r
)
,
s
p
ec
tr
al
b
an
d
s
,
also
s
p
ec
tr
al
r
esp
o
n
s
es
o
f
s
an
d
m
etal
tile
an
d
v
eg
etatio
n
th
at
h
a
v
e
th
e
s
am
e
r
an
g
e.
Fro
m
t
h
e
f
a
cto
r
s
ab
o
v
e,
we
tr
ied
to
e
x
tr
ac
t
th
e
No
r
m
alize
d
Dif
f
er
en
ce
Veg
etatio
n
I
n
d
ex
(
NDVI
)
v
alu
e
o
n
th
e
AOI
u
s
ed
.
W
e
co
m
p
ar
e
th
e
r
esu
lts
o
f
NDVI
(
F
ig
u
r
e
9
(
a)
)
with
o
u
r
class
if
icatio
n
r
esu
lts
,
f
r
o
m
th
er
e
it
is
f
o
u
n
d
th
at
t
h
e
v
eg
etatio
n
ar
ea
s
th
at
h
a
v
e
n
o
t
b
ee
n
s
ep
ar
ated
f
r
o
m
u
r
b
an
m
ater
ial
ar
e
m
o
s
tly
d
etec
ted
as c
o
n
cr
ete
o
r
s
an
d
m
etal
tiles
.
Fro
m
th
ese
r
esu
lts
,
we
d
o
h
av
e
a
p
lan
to
d
o
m
ask
in
g
v
eg
etatio
n
u
s
in
g
th
e
s
p
ec
tr
al
in
d
ex
f
r
o
m
NDVI
f
o
r
th
e
n
ex
t
wo
r
k
s
.
W
e
also
co
m
p
ar
ed
o
u
r
f
in
d
in
g
s
with
s
o
m
e
s
p
ec
tr
a
l
r
esp
o
n
s
e
r
ef
er
en
ce
s
.
Fo
r
s
an
d
m
etal
tile,
we
u
s
e
a
s
p
ec
tr
al
r
esp
o
n
s
e
r
ef
er
e
n
ce
f
r
o
m
s
an
d
b
ec
au
s
e
t
h
e
m
ate
r
ial
u
s
ed
o
n
th
e
r
o
o
f
is
s
an
d
c
o
ated
m
etal
(
F
ig
u
r
e
9
(
b
)
)
.
A
s
f
o
r
co
n
c
r
ete
an
d
v
eg
etatio
n
we
u
s
e
th
e
s
p
ec
tr
al
r
esp
o
n
s
e
r
ef
er
en
ce
a
v
ailab
le
o
n
th
e
E
C
O
s
y
s
tem
Sp
ac
eb
o
r
n
e
T
h
er
m
al
R
ad
io
m
eter
E
x
p
er
im
e
n
t
o
n
Sp
ac
e
Statio
n
(
E
C
OSTR
E
SS
)
s
p
ec
tr
al
lib
r
ar
y
web
s
ite
b
y
NASA/J
PL
(
F
ig
u
r
e
s
9
(
c)
,
(
d
)
,
(
e)
).
Fig
u
r
e
7
.
SVM
R
OC
cu
r
v
e
Fig
u
r
e
8
.
Vis
u
al
co
m
p
ar
is
o
n
o
f
class
if
icatio
n
r
esu
lts
with
s
at
ellite
im
ag
e
d
ata
(
r
ed
f
o
r
m
o
d
el
tr
ain
d
ata,
an
d
cy
an
f
o
r
o
th
e
r
o
b
jects th
at
h
a
v
e
b
ee
n
s
u
cc
ess
f
u
lly
d
etec
ted
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
2
,
Ap
r
il 2
0
2
1
:
6
9
0
-
704
698
(
a)
(
b
)
(
c
)
(
d
)
(
e
)
Fig
u
r
e
9
.
(
a
)
NDVI
r
esu
lt
,
(
b
)
San
d
s
p
ec
tr
al
r
esp
o
n
s
e,
c
o
atin
g
m
ater
ial
f
o
r
s
an
d
m
etal
[
4
4
]
,
(
c)
C
o
n
cr
ete
s
p
ec
tr
al
r
esp
o
n
s
e
,
(
d
)
Sp
ec
tr
al
r
esp
o
n
s
e
o
f
g
r
as
s
,
(
e)
tr
ee
s
[
4
5
]
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
R
o
o
f m
a
teri
a
ls
id
en
tifi
ca
tio
n
b
a
s
ed
o
n
p
leia
d
es sp
ec
tr
a
l res
p
o
n
s
es u
s
in
g
s
u
p
ervis
ed
…
(
A
yo
m
Wid
ip
a
min
to
)
699
Fro
m
th
e
s
p
ec
tr
al
r
esp
o
n
s
e
r
e
f
er
en
ce
th
at
we
h
av
e,
it
ap
p
e
ar
s
th
at
th
e
r
esp
o
n
s
es
p
r
o
d
u
c
ed
b
y
s
an
d
m
etal
tile
,
co
n
cr
ete,
an
d
v
eg
et
atio
n
d
o
h
av
e
s
im
ilar
ities
,
esp
ec
ially
in
th
e
wav
elen
g
th
r
an
g
e
o
f
7
5
0
-
1
5
0
0
n
m
.
T
h
is
r
ein
f
o
r
ce
s
th
e
r
ea
s
o
n
s
wh
y
th
e
m
eth
o
d
we
u
s
e
is
d
if
f
icu
lt
to
d
is
tin
g
u
is
h
b
etwe
en
s
an
d
m
etal
tile,
co
n
cr
ete,
an
d
v
e
g
etatio
n
in
R
,
G,
B
,
NI
R
b
an
d
s
.
T
o
o
v
er
c
o
m
e
th
is
s
h
o
r
tco
m
in
g
,
we
p
l
a
n
f
u
r
th
e
r
r
esear
ch
u
s
in
g
th
e
f
o
r
m
u
latio
n
o
f
in
d
e
x
v
alu
es,
f
r
o
m
ex
is
tin
g
s
p
ec
tr
al
v
alu
es,
wh
ich
h
av
e
g
r
ea
ter
d
if
f
er
en
ce
s
b
etwe
en
ea
ch
o
th
er
,
d
ata
ass
im
ilatio
n
,
an
d
m
u
ltil
ev
el
class
if
icatio
n
m
eth
o
d
s
.
Fo
r
f
u
r
th
e
r
v
er
if
icati
o
n
,
we
h
a
v
e
tr
ied
t
o
u
s
e
th
e
m
eth
o
d
in
d
if
f
e
r
en
t
r
eg
io
n
s
as
a
r
ep
r
esen
tativ
e
f
o
r
m
u
ch
wid
er
a
r
ea
s
an
d
v
ar
iatio
n
s
,
n
am
ely
Pu
r
wo
k
er
to
an
d
Sem
ar
a
n
g
.
T
h
e
d
ata
u
s
ed
f
o
r
Pu
r
w
o
k
er
to
i
s
clo
u
d
-
f
r
ee
im
a
g
e
d
ata
f
r
o
m
t
h
e
Pleiad
es
s
atellite
o
b
tain
ed
o
n
3
0
J
an
u
a
r
y
2
0
1
9
,
th
e
d
etailed
l
o
ca
tio
n
is
at
T
er
m
in
a
l
B
u
lu
p
itu
.
Me
a
n
wh
ile
,
th
e
d
ata
u
s
ed
f
o
r
Sem
ar
an
g
is
o
b
tain
ed
o
n
3
M
ay
2
0
1
8
,
a
d
etailed
lo
ca
tio
n
is
at
State
Po
ly
tech
n
ic
o
f
Sem
ar
an
g
.
T
h
e
r
esu
lts
is
s
h
o
wn
in
T
ab
le
6
.
Fig
u
r
e
10
(
a)
an
d
10
(
b
)
s
h
o
w
SVM
d
ata
an
d
s
p
ec
tr
al
r
ef
lecta
n
ce
p
r
o
f
ile
f
r
o
m
th
e
P
u
r
wo
k
er
to
r
eg
io
n
.
Fig
u
r
es
11
(
a)
a
n
d
11
(
b
)
s
h
o
w
SVM
d
ata
an
d
s
p
ec
t
r
al
r
ef
lecta
n
ce
p
r
o
f
ile
f
r
o
m
t
h
e
Sem
ar
an
g
r
eg
i
o
n
.
Fro
m
th
e
v
is
u
al
r
esu
lts
an
d
v
alu
es
g
en
er
ated
b
y
th
e
SVM
p
r
o
g
r
a
m
,
it
ap
p
e
ar
s
th
at
th
e
m
eth
o
d
we
p
r
o
p
o
s
ed
ca
n
class
if
y
a
lu
m
i
n
u
m
r
o
o
f
,
asb
esto
s
,
ce
r
am
ic
tile,
co
n
cr
e
te,
an
d
s
an
d
m
etal
tile
m
ater
ials
.
Ho
wev
er
,
t
h
e
s
p
ec
tr
al
r
ef
lecta
n
ce
r
esu
lts
also
s
h
o
w
a
s
lig
h
tly
d
if
f
er
e
n
t
p
atter
n
f
r
o
m
th
e
r
esu
lts
f
r
o
m
t
h
e
J
ak
ar
ta
ar
ea
.
I
n
T
ab
le
7
it
ca
n
b
e
s
ee
n
th
at
th
e
r
e
ar
e
s
ig
n
if
ican
t
r
esu
lts
,
esp
e
cially
in
th
e
asb
esto
s
m
ater
ial
in
Sem
ar
an
g
.
B
ased
o
n
T
ab
le
8
,
t
h
e
o
v
er
all
ac
cu
r
a
cy
an
d
K
ap
p
a
c
o
ef
f
icien
t
v
alu
e
o
n
th
e
m
eth
o
d
th
at
we
u
s
e
a
r
e
eq
u
al
t
o
9
2
.
9
2
%
an
d
0
.
9
0
6
9
,
th
is
r
esu
lt
is
b
etter
if
co
m
p
ar
e
d
to
p
r
ev
io
u
s
s
tu
d
ies
th
at
o
b
tain
ed
an
o
v
er
all
a
cc
u
r
ac
y
v
alu
e
o
f
8
3
%
an
d
Kap
p
a
c
o
ef
f
icien
t
0
.
7
6
to
d
o
r
o
o
f
d
etec
tio
n
u
s
in
g
W
o
r
ld
View
-
2
s
atellite
th
at
h
as
8
(
eig
h
t
)
m
u
ltis
p
ec
tr
al
b
an
d
s
[
4
6
]
.
Ou
r
r
esu
lt
s
h
o
ws
th
at
f
ield
s
u
r
v
ey
s
ar
e
s
till
n
ee
d
ed
in
t
h
e
Pu
r
wo
k
er
to
an
d
Sem
ar
an
g
ar
ea
s
to
en
s
u
r
e
th
e
ty
p
e
o
f
r
o
o
f
m
ater
ial
s
av
ailab
le.
T
h
e
class
if
icatio
n
we
d
id
is
also
lim
ited
to
ju
s
t
5
(
f
iv
e)
m
ater
ials
,
s
o
o
th
er
u
n
k
n
o
wn
m
ater
ials
ar
e
id
en
tifie
d
as
th
em
in
s
tead
.
I
n
f
u
tu
r
e
wo
r
k
s
,
we
p
lan
to
d
o
t
h
e
ex
h
au
s
tiv
e
f
ield
s
u
r
v
e
y
s
,
T
OA
b
id
ir
ec
tio
n
al
r
ef
lecta
n
ce
d
is
tr
ib
u
tio
n
f
u
n
ctio
n
(
B
R
DF)
,
an
d
v
eg
etatio
n
m
ask
in
g
,
s
o
th
at
th
e
id
e
n
tific
atio
n
ca
n
b
e
u
p
g
r
a
d
ed
in
to
a
b
etter
r
esu
lt
with
o
n
ly
u
r
b
an
r
o
o
f
m
ater
ial
as
a
s
tu
d
y
ar
ea
.
(
a)
(
b
)
Fig
u
r
e
1
0
.
(
a)
SVM
d
ata
o
f
Pu
r
wo
k
er
to
r
eg
io
n
;
(
b
)
Sp
ec
t
r
al
r
ef
lecta
n
ce
p
r
o
f
ile
o
f
Pu
r
w
o
k
er
to
r
eg
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.