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.
6
,
No
.
6
,
Dec
em
b
er
201
6
,
p
p
.
3
08
7
~
3
09
3
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
6i
6
.
1
2
1
1
6
3087
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
ISAR I
m
a
g
e Clas
sifica
tion w
ith
Wa
v
elet
a
nd Wate
r
shed
Tra
nsfo
r
m
s
B
.
M
a
m
a
t
ha
1
,
V.
Va
lli K
u
ma
r
2
1
NST
L
,
V
isa
k
h
a
p
a
tn
a
m
,
In
d
ia
2
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
&
S
y
ste
m
En
g
in
e
e
rin
g
,
A
n
d
h
ra
Un
iv
e
rsit
y
,
V
isa
k
h
a
p
a
tn
a
m
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
8
,
2
0
1
6
R
ev
i
s
ed
Sep
9
,
2
0
1
6
A
cc
ep
ted
Sep
2
3
,
2
0
1
6
In
v
e
rse
S
y
n
th
e
ti
c
A
p
e
rtu
re
Ra
d
a
r
im
a
g
e
s
a
r
e
p
la
y
in
g
a
sig
n
if
ic
a
n
t
ro
le
in
c
las
si
f
ica
ti
o
n
o
f
se
a
a
n
d
a
ir
tar
g
e
ts.
F
irst
w
e
a
c
q
u
ire
th
e
IS
A
R
im
a
g
e
s
o
f
targ
e
ts
u
sin
g
a
se
n
so
r
li
k
e
ra
d
a
r
a
n
d
e
x
trac
t
th
e
c
h
a
ra
c
teristics
o
f
targ
e
ts
f
ro
m
th
e
IS
A
R
im
a
g
e
s
in
th
e
f
o
rm
o
f
f
e
a
tu
re
v
e
c
to
rs.
T
h
e
c
o
m
p
u
ted
f
e
a
tu
re
v
e
c
to
rs
a
r
e
u
se
d
f
o
r
c
las
si
f
ic
a
ti
o
n
o
f
targ
e
ts.
In
th
is
w
o
rk
,
w
id
e
ly
u
se
d
a
n
d
e
ff
ici
e
n
t
se
g
m
e
n
tatio
n
to
o
l
W
a
ters
h
e
d
tran
sf
o
rm
a
n
d
th
e
m
u
lt
i
re
so
lu
ti
o
n
tec
h
n
iq
u
e
w
a
v
e
l
e
t
tran
sf
o
r
m
a
re
e
x
p
lo
re
d
to
d
e
riv
e
th
e
targ
e
t
f
e
a
tu
re
s.
A
n
a
rti
f
icia
l
n
e
u
ra
l
n
e
tw
o
rk
b
a
se
d
c
las
si
f
ier
is
u
se
d
f
o
r
c
las
si
f
ic
a
ti
o
n
.
T
h
e
W
a
v
e
let
a
n
a
l
y
sis
d
iv
id
e
s
th
e
in
f
o
rm
a
ti
o
n
o
f
a
n
im
a
g
e
in
to
a
p
p
ro
x
i
m
a
ti
o
n
a
n
d
d
e
tail
su
b
sig
n
a
ls.
T
h
e
a
p
p
ro
x
i
m
a
te
a
n
d
th
re
e
d
e
tail
su
b
sig
n
a
l
v
a
lu
e
s
a
re
tak
e
n
a
s
fe
a
tu
re
v
e
c
to
rs
a
n
d
g
iv
e
n
a
s
in
p
u
t
to
t
h
e
c
las
sif
ier
f
o
r
sh
ip
I
S
A
R
im
a
g
e
c
las
si
f
ica
ti
o
n
.
T
h
e
w
id
e
l
y
u
se
d
se
g
m
e
n
tatio
n
tec
h
n
i
q
u
e
,
W
a
ters
h
e
d
tran
sf
o
r
m
is
a
p
p
li
e
d
to
th
e
IS
A
R
i
m
a
g
e
s.
T
h
e
w
a
v
e
let
c
o
e
fficie
n
ts
a
re
c
o
m
p
u
ted
f
o
r
th
e
se
g
m
e
n
ted
IS
A
R
i
m
a
g
e
s
a
n
d
u
se
d
a
s
f
e
a
tu
re
v
e
c
to
rs
f
o
r
c
las
si
f
ica
ti
o
n
o
f
th
e
IS
A
R
ima
g
e
s.
A
lso
,
th
e
sta
ti
stica
l
m
o
m
e
n
ts
m
e
a
n
a
n
d
sta
n
d
a
rd
d
e
v
iatio
n
a
re
c
o
m
p
u
ted
f
o
r
th
e
c
o
l
o
r
IS
A
R
im
a
g
e
s
it
se
l
f
,
tak
e
n
in
RG
B
f
o
rm
a
t.
T
h
e
se
st
a
ti
stica
l
c
o
l
o
r
m
o
m
e
n
ts
a
re
u
se
d
a
s
f
e
a
tu
re
v
e
c
to
r
.
T
h
e
c
las
si
f
ica
ti
o
n
a
c
c
u
ra
c
y
is
c
o
m
p
a
re
d
f
o
r
th
e
f
e
a
tu
re
v
e
c
to
rs.
K
ey
w
o
r
d
:
C
las
s
i
f
icatio
n
a
cc
u
r
ac
y
C
las
s
i
f
ier
C
o
lo
r
m
o
m
e
n
ts
I
s
ar
i
m
ag
e
W
ater
s
h
ed
t
r
an
s
f
o
r
m
W
av
elet
t
r
an
s
f
o
r
m
Co
p
y
rig
h
t
©
2
0
1
6
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
:
B.
Ma
m
at
h
a,
Nav
al
Scie
n
ce
&
T
ec
h
n
o
lo
g
ic
al
L
ab
o
r
ato
r
y
,
Vig
y
an
n
a
g
ar
,
Vis
a
k
h
ap
at
n
a
m
-
5
3
0
0
2
7
,
I
n
d
ia.
E
m
ail:
m
a
m
ath
ab
a
s
av
a6
7
@
g
m
ail.
c
m
1.
I
NT
RO
D
UCT
I
O
N
C
las
s
i
f
icatio
n
o
f
tar
g
et
s
h
as
i
m
m
en
s
e
i
m
p
o
r
tan
ce
in
d
ef
e
n
s
e
an
d
civ
i
lian
ap
p
licatio
n
s
li
k
e
co
astal
s
u
r
v
eilla
n
ce
,
p
r
ev
en
tio
n
o
f
u
n
au
th
o
r
ized
f
i
s
h
i
n
g
etc.
C
las
s
if
icatio
n
is
to
ass
i
g
n
a
n
u
n
k
n
o
w
n
tar
g
et
to
a
clas
s
f
r
o
m
a
g
i
v
e
n
s
et
o
f
cla
s
s
e
s
.
T
h
e
class
i
f
icatio
n
is
b
a
s
ed
o
n
attr
ib
u
te
v
a
lu
e
s
o
f
th
e
tar
g
ets
o
b
tain
ed
f
r
o
m
d
if
f
er
e
n
t r
ep
r
esen
tatio
n
s
o
f
tar
g
ets.
I
n
t
h
is
w
o
r
k
I
n
v
er
s
e
S
y
n
t
h
etic
A
p
er
t
u
r
e
R
ad
ar
i
m
a
g
es
g
en
er
all
y
m
ea
s
u
r
ed
w
it
h
a
h
ig
h
r
e
s
o
lu
tio
n
R
ad
ar
ar
e
u
s
ed
f
o
r
tar
g
et
f
ea
t
u
r
e
ex
tr
ac
tio
n
.
T
h
e
I
S
AR
i
m
ag
e
s
r
ep
r
esen
t
t
h
e
t
w
o
d
i
m
en
s
io
n
al
d
i
s
tr
ib
u
tio
n
o
f
tar
g
et
s
ca
tter
er
s
i
n
d
o
w
n
r
an
g
e
(
li
n
e
o
f
s
ite
o
f
r
ad
ar
)
an
d
cr
o
s
s
r
an
g
e
(
p
er
p
en
d
icu
lar
to
d
o
w
n
r
an
g
e
)
d
ir
ec
tio
n
s
.
T
h
e
m
ea
s
u
r
e
m
e
n
t
o
f
I
S
AR
i
m
ag
e
s
is
a
v
er
y
co
m
p
le
x
tas
k
t
h
at
r
eq
u
ir
es
u
n
i
f
o
r
m
m
o
tio
n
o
f
th
e
tar
g
et
w
it
h
r
esp
ec
t
to
r
ad
ar
.
T
h
e
f
o
r
m
atio
n
o
f
an
I
S
A
R
i
m
ag
e
i
n
v
o
l
v
es
h
i
g
h
l
y
co
m
p
u
tatio
n
al
s
i
g
n
al
p
r
o
ce
s
s
i
n
g
.
T
h
e
r
o
le,
p
itch
a
n
d
y
a
w
i
n
f
o
r
m
atio
n
o
f
t
h
e
tar
g
et
s
i
n
c
ase
o
f
s
h
ip
s
is
al
s
o
co
n
s
id
er
ed
alo
n
g
w
it
h
th
e
tar
g
et
ec
h
o
d
ata
to
s
y
n
t
h
esize
an
I
S
AR
i
m
ag
e
Neu
r
al
n
et
w
o
r
k
s
ar
e
w
id
el
y
b
ein
g
u
s
ed
in
v
ar
io
u
s
i
n
telli
g
e
n
t
s
y
s
t
e
m
s
.
On
e
s
u
c
h
co
m
p
u
ter
v
i
s
io
n
s
y
s
te
m
f
o
r
r
ec
o
g
n
it
io
n
a
n
d
class
i
f
icatio
n
o
f
co
m
m
er
cial
co
r
k
tile
s
i
s
d
escr
ib
ed
in
[
1
]
.
Var
io
u
s
f
ea
t
u
r
e
v
ec
to
r
ex
tr
ac
tio
n
m
et
h
o
d
s
b
ase
d
o
n
co
lo
r
,
tex
tu
r
e
an
d
s
h
ap
e
ca
n
b
e
s
ee
n
i
n
[
2
]
.
T
h
e
co
m
p
ar
is
o
n
o
f
cla
s
s
i
f
i
ca
tio
n
p
er
f
o
r
m
an
ce
o
f
d
i
f
f
er
en
t
f
ea
t
u
r
e
v
ec
to
r
s
av
er
ag
e
R
GB
,
co
lo
r
m
o
m
e
n
t
s
,
co
lo
r
h
i
s
to
g
r
a
m
s
etc.
i
s
g
i
v
en
i
n
[
3
]
.
Ho
w
t
h
e
f
ea
t
u
r
e
r
ep
r
esen
tatio
n
an
d
f
ea
t
u
r
e
ex
tr
ac
tio
n
ar
e
r
q
u
all
y
cr
itical
in
m
u
lti
m
ed
ia
p
r
o
ce
s
s
i
n
g
i
s
ex
p
lai
n
ed
in
[
4
]
.
I
n
th
is
w
o
r
k
a
p
r
o
b
ab
ilis
tic
n
eu
r
al
n
et
w
o
r
k
i
s
u
s
ed
f
o
r
I
S
AR
i
m
a
g
e
cla
s
s
i
f
icat
io
n
.
T
h
e
class
i
f
ier
p
er
f
o
r
m
an
ce
ac
cu
r
a
c
y
p
r
ese
n
ted
i
n
t
h
is
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.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
3
08
7
–
3
09
3
3088
p
ap
er
is
d
o
n
e
w
it
h
I
S
AR
i
m
a
g
es
o
f
t
w
o
s
h
ip
s
th
at
ar
e
m
ea
s
u
r
ed
w
it
h
s
tep
p
ed
f
r
eq
u
e
n
c
y
wav
ef
o
r
m
r
ad
ar
.
Fo
r
f
ea
t
u
r
e
ex
tr
ac
tio
n
,
an
I
S
AR
im
ag
e
i
s
co
n
s
id
er
ed
as
a
d
ig
ital
i
m
ag
e
r
ep
r
esen
ted
w
it
h
a
t
w
o
d
im
e
n
s
io
n
al
m
atr
i
x
o
f
p
ix
el
in
te
n
s
i
t
y
v
alu
e
s
.
I
n
th
is
s
t
u
d
y
m
at
h
e
m
a
tical
tr
an
s
f
o
r
m
s
W
ater
s
h
ed
tr
an
s
f
o
r
m
an
d
w
a
v
elet
tr
a
n
s
f
o
r
m
ar
e
ap
p
lied
to
th
e
I
SAR
i
m
ag
e
s
to
ex
tr
ac
t ta
r
g
et
attr
ib
u
tes.
T
h
e
w
ater
s
h
ed
tr
an
s
f
o
r
m
i
s
a
m
at
h
e
m
atica
l
tr
a
n
s
f
o
r
m
th
a
t
ca
n
b
e
ca
te
g
o
r
ized
as
a
r
e
g
io
n
b
ased
i
m
a
g
e
s
e
g
m
en
tat
io
n
tec
h
n
iq
u
e
.
T
h
e
s
eg
m
e
n
tatio
n
r
ef
er
s
to
s
u
b
d
iv
id
in
g
a
n
i
m
ag
e
i
n
to
it
s
c
o
n
s
tit
u
e
n
t
r
eg
io
n
s
.
I
n
s
eg
m
e
n
tatio
n
t
h
e
i
m
ag
e
ele
m
en
ts
t
h
at
ex
h
ib
it
s
i
m
ilar
ch
ar
ac
ter
is
tics
ar
e
g
r
o
u
p
ed
.
T
h
e
s
eg
m
e
n
tat
io
n
r
esu
lts
o
b
tain
ed
f
r
o
m
w
ater
s
h
ed
tr
an
s
f
o
r
m
ca
n
b
e
tak
e
n
as
in
p
u
t f
o
r
f
u
r
t
h
er
i
m
ag
e
a
n
al
y
s
i
s
.
Se
g
m
e
n
tatio
n
is
a
cr
itical
i
m
a
g
e
p
r
o
ce
s
s
in
g
tas
k
f
o
r
ac
cu
r
ate
in
ter
p
r
etatio
n
o
f
ce
r
tain
i
m
a
g
e
s
lik
e
m
ed
ical
i
m
ag
e
s
u
s
ed
f
o
r
m
ed
ical
d
iag
n
o
s
i
s
.
A
p
p
licatio
n
o
f
W
at
er
s
h
ed
s
e
g
m
en
tatio
n
to
m
ed
ic
al
i
m
ag
e
s
g
i
v
es
e
s
s
e
n
tial
i
n
f
o
r
m
atio
n
u
s
ef
u
l
f
o
r
b
etter
m
ed
ical
d
iag
n
o
s
is
[
5
]
.
An
o
b
j
ec
t
o
r
ien
ted
ap
p
r
o
ac
h
f
o
r
u
n
s
u
p
er
v
is
ed
cla
s
s
i
f
icatio
n
o
f
h
i
g
h
r
eso
l
u
tio
n
r
e
m
o
tesen
s
i
n
g
i
m
a
g
es
is
p
r
o
p
o
s
ed
.
T
h
e
f
r
ac
tal
n
et
ev
al
u
ati
o
n
tech
n
iq
u
e
is
u
s
ed
as
an
i
m
ag
e
s
eg
m
e
n
tatio
n
tech
n
iq
u
e
f
o
r
ex
tr
ac
tio
n
o
f
o
b
jects
f
r
o
m
t
h
e
i
m
ag
e
s
[
6
]
.
T
h
e
u
s
e
f
u
ln
e
s
s
a
n
d
ef
f
ec
ti
v
e
n
es
s
o
f
u
s
i
n
g
W
ater
s
h
e
d
s
eg
m
e
n
tatio
n
i
n
d
etec
tio
n
o
f
o
r
al
ca
n
ce
r
an
d
t
u
m
o
r
s
i
n
m
a
m
m
o
g
r
a
m
s
i
s
e
x
p
lain
ed
in
[
7
-
8
]
.
I
n
o
u
r
w
o
r
k
,
W
ater
s
h
ed
s
eg
m
e
n
tatio
n
is
ap
p
lied
to
th
e
s
h
ip
I
S
A
R
i
m
a
g
es
b
ef
o
r
e
f
ea
tu
r
e
ex
tr
ac
tio
n
.
W
av
elet
T
r
an
s
f
o
r
m
i
s
a
m
u
lti
r
eso
lu
tio
n
tech
n
iq
u
e
o
f
ten
u
s
e
d
to
r
eso
lv
e
s
o
m
e
p
ar
ts
o
f
s
i
g
n
al
i
n
ti
m
e
an
d
s
o
m
e
o
t
h
er
p
ar
ts
in
f
r
eq
u
e
n
c
y
.
T
h
e
w
av
e
let
co
ef
f
icie
n
ts
,
th
e
ap
p
r
o
x
i
m
atio
n
,
h
o
r
izo
n
tal,
v
er
tical
an
d
d
etail
co
ef
f
icie
n
t
s
ar
e
w
id
el
y
b
ein
g
u
s
ed
as
f
ea
t
u
r
e
v
ec
to
r
s
in
v
ar
io
u
s
i
m
a
g
e
cla
s
s
i
f
icat
io
n
a
n
d
i
m
ag
e
r
etr
iev
a
l
ap
p
licatio
n
s
.
T
h
e
s
tatis
t
ical
m
o
m
e
n
t
s
o
f
w
a
v
elet
d
etail
co
ef
f
icien
t
s
ar
e
u
s
ed
f
o
r
E
C
G
i
m
a
g
e
class
i
f
ica
tio
n
[
9
]
.
T
h
e
s
tu
d
y
an
d
an
al
y
s
i
s
o
f
elec
tr
o
m
y
o
g
r
ap
h
y
s
ig
n
al
s
h
av
e
ap
p
licatio
n
in
s
p
o
r
ts
s
cien
ce
.
T
h
ese
elec
tr
o
-
p
h
y
s
io
lo
g
ica
l
s
ig
n
al
s
ar
e
u
s
ed
to
d
eter
m
i
n
e
m
u
s
cl
e
s
tr
en
g
t
h
,
m
u
s
cle
f
o
r
ce
etc.
th
at
d
eter
m
i
n
e
th
e
h
u
m
a
n
ai
l
m
e
n
t
s
li
k
e
lo
w
b
a
ck
p
ai
n
etc.
T
h
e
w
av
ele
t
tr
an
s
f
o
r
m
is
u
s
ed
f
o
r
th
e
f
ea
t
u
r
e
e
x
tr
ac
tio
n
a
n
d
class
i
f
icatio
n
o
f
E
MG
s
i
g
n
al
s
[
1
0
]
.
Hea
r
t
R
ate
Var
iab
ilit
y
(
HR
V)
s
i
g
n
als
r
ep
r
esen
t
a
h
u
m
an
‟
s
Au
to
n
o
m
o
u
s
Ner
v
o
u
s
S
y
s
te
m
.
W
av
ele
t
co
ef
f
icien
ts
co
m
p
u
ted
f
o
r
th
e
H
R
V
‟
s
ar
e
u
s
ed
as
f
ea
tu
r
e
v
ec
t
o
r
s
to
r
e
p
r
es
en
t
th
e
s
ig
n
al
s
an
d
t
h
eir
f
u
r
th
er
a
n
al
y
s
is
[
1
1
]
.
I
n
g
e
n
d
er
clas
s
i
f
icati
o
n
th
r
o
u
g
h
f
ac
e
i
m
a
g
es,
t
h
e
l
o
g
ar
ith
m
ic
w
a
v
elet
o
f
Gab
o
r
is
u
s
ed
to
s
ep
ar
ate
d
ata
r
elate
d
to
tex
t
u
r
e
f
r
o
m
i
n
ter
n
m
ed
iate
f
r
eq
u
e
n
c
y
b
an
d
s
an
d
u
s
e
th
e
d
ata
i
n
i
m
a
g
e
s
e
g
m
e
n
tat
io
n
.
T
h
e
lo
ca
l
b
in
ar
y
p
atter
n
s
o
f
t
h
e
s
e
g
m
e
n
ted
i
m
a
g
es
ar
e
u
s
ed
f
o
r
g
e
n
d
er
class
i
f
icatio
n
[
1
2
]
.
T
h
e
w
a
v
ele
t tr
an
s
f
o
r
m
i
s
also
u
s
ed
f
o
r
i
m
ag
e
co
m
p
r
ess
io
n
.
I
m
a
g
es c
o
n
t
ain
lar
g
e
a
m
o
u
n
t o
f
in
f
o
r
m
atio
n
t
h
at
r
eq
u
ir
es
m
u
c
h
s
to
r
ag
e
s
p
ac
e.
Ma
n
y
a
ti
m
e
it
is
v
er
y
u
s
e
f
u
l
to
co
m
p
r
e
s
s
th
e
i
m
a
g
e.
I
m
ag
e
s
co
n
tain
r
ed
u
n
d
a
n
t
in
f
o
r
m
atio
n
.
T
h
is
r
ed
u
n
d
a
n
c
y
f
ea
tu
r
e
i
s
ex
p
lo
ited
to
co
m
p
r
es
s
i
m
a
g
es
.
On
l
y
t
h
e
ess
e
n
tia
l
in
f
o
r
m
atio
n
n
ee
d
ed
to
r
ec
o
n
s
tr
u
ct
an
i
m
ag
e
is
o
n
l
y
s
to
r
ed
.
W
av
elets
ar
e
also
u
s
ed
f
o
r
s
eg
m
en
tatio
n
o
f
i
m
a
g
es.
T
h
e
co
m
p
u
tatio
n
al
o
v
er
h
ea
d
o
f
co
n
v
er
s
io
n
o
f
co
lo
r
i
m
ag
e
to
g
r
a
y
i
m
a
g
e
b
ef
o
r
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
ca
n
b
e
av
o
id
ed
b
y
e
x
tr
ac
tio
n
o
f
f
ea
t
u
r
e
v
ec
to
r
s
f
r
o
m
t
h
e
co
lo
r
I
SA
R
i
m
a
g
es
a
n
d
u
s
ed
f
o
r
class
i
f
icatio
n
.
T
h
e
co
lo
r
im
a
g
es
ar
e
r
ep
r
esen
ted
in
d
if
f
er
en
t
co
lo
r
m
o
d
els
l
i
k
e
R
GB
,
I
n
d
e
x
ed
i
m
a
g
e,
NT
SC
,
YC
b
C
r
,
C
MY
,
C
MY
K
an
d
HSV.
T
h
e
i
m
a
g
e
in
o
n
e
m
o
d
el
ca
n
b
e
co
n
v
er
te
d
to
an
y
o
th
er
m
o
d
el
b
y
s
u
ita
b
le
tr
an
s
f
o
r
m
a
tio
n
.
E
ac
h
o
f
t
h
ese
m
o
d
el
s
is
s
u
i
ta
b
le
f
o
r
o
n
e
p
ar
ticu
lar
ap
p
licatio
n
[
1
3
]
.
I
n
th
is
w
o
r
k
R
GB
co
lo
r
m
o
d
el
is
ta
k
en
an
d
s
u
cc
e
s
s
f
u
ll
y
u
s
ed
f
o
r
f
ea
t
u
r
e
ex
tr
ac
tio
n
an
d
s
h
ip
I
S
A
R
i
m
a
g
e
clas
s
if
icatio
n
.
C
o
lo
r
f
ea
tu
r
e
is
a
n
ef
f
icie
n
t
v
is
u
al
ch
ar
ac
ter
is
tic
o
f
i
m
a
g
e
s
a
n
d
s
o
t
h
e
co
lo
r
i
m
a
g
es
ar
e
b
ein
g
u
s
ed
e
x
te
n
s
i
v
el
y
i
n
co
n
ten
t
b
ased
i
m
a
g
e
r
etr
iev
al
f
r
o
m
h
u
g
e
d
atab
ases
.
T
h
e
s
in
g
le
co
lo
r
m
o
m
e
n
t
s
a
n
d
t
w
o
co
lo
r
m
o
m
e
n
t
s
ar
e
b
ein
g
u
s
ed
f
o
r
I
S
AR
i
m
a
g
e
clas
s
i
f
icatio
n
[
1
4
]
.
R
GB
co
lo
r
h
is
to
g
r
a
m
an
d
C
an
n
y
ed
g
e
d
etec
tio
n
i
s
u
s
ed
as
a
f
ea
t
u
r
e
v
ec
to
r
f
o
r
i
m
a
g
e
r
etr
iev
al
f
r
o
m
a
d
atab
ase
[
1
5
]
.
A
n
i
m
p
o
r
tan
t
c
h
ar
ac
t
er
is
tic
o
f
co
lo
r
f
ea
t
u
r
e
i
s
in
v
ar
ian
ce
to
r
o
tatio
n
w
h
ic
h
is
v
er
y
es
s
en
tial f
o
r
I
S
AR
i
m
ag
e
cla
s
s
i
f
icatio
n
s
i
n
ce
I
SAR
i
m
ag
e
s
ar
e
asp
ec
t v
ar
ia
n
t.
2.
F
E
AT
U
RE
E
XT
RAC
T
I
O
N
2
.
1
.
Wa
v
elet
T
ra
ns
f
o
r
m
T
h
e
co
n
tin
u
o
u
s
w
a
v
elet
tr
a
n
s
f
o
r
m
i
s
th
e
s
u
m
o
v
er
ti
m
e
o
f
s
c
aled
an
d
s
h
i
f
ted
w
av
ele
t Ψ a
n
d
is
g
iv
e
n
(
)
∫
(
)
(
)
(
1
)
W
h
er
e„
s
‟
i
s
s
ca
le
o
f
a
w
a
v
elet
w
h
ic
h
is
p
r
o
p
o
r
tio
n
al
to
th
e
i
n
v
er
s
e
o
f
t
h
e
f
r
eq
u
e
n
c
y
i
n
f
o
r
m
atio
n
.
T
h
e
tr
an
s
latio
n
is
p
r
o
p
o
r
tio
n
al
to
ti
m
e
in
f
o
r
m
a
tio
n
.
W
av
elet
s
ar
e
m
at
h
e
m
atica
l
f
u
n
ct
io
n
s
t
h
at
d
iv
id
e
d
ata
in
to
d
if
f
er
e
n
t
f
r
eq
u
e
n
c
y
co
m
p
o
n
e
n
ts
.
T
h
e
i
m
ag
e
ca
n
b
e
d
ec
o
m
p
o
s
ed
u
s
in
g
lo
w
p
a
s
s
an
d
h
i
g
h
p
ass
f
ilter
.
T
h
is
lo
w
p
ass
an
d
h
i
g
h
p
ass
f
ilter
p
air
i
s
ca
lled
an
a
l
y
s
is
f
il
ter
p
air
.
Fi
r
s
t
lo
w
p
ass
f
il
ter
is
ap
p
lied
f
o
r
ea
ch
r
o
w
o
f
d
ata,
to
o
b
tain
lo
w
f
r
eq
u
en
c
y
co
m
p
o
n
en
t
s
o
f
t
h
e
r
o
w
s
.
I
t
b
ein
g
a
h
al
f
b
an
d
f
il
ter
,
th
e
o
u
tp
u
t
o
f
L
P
F
co
n
tai
n
s
o
n
l
y
f
r
eq
u
en
c
ies
in
t
h
e
f
ir
s
t
h
al
f
o
f
th
e
o
r
ig
in
al
f
r
eq
u
en
c
y
r
an
g
e.
Si
m
i
lar
l
y
h
ig
h
p
ass
f
ilter
is
ap
p
lied
f
o
r
th
e
s
a
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
I
S
A
R
I
ma
g
e
C
la
s
s
ifica
tio
n
w
it
h
W
a
ve
let
a
n
d
W
a
ter
s
h
ed
Tr
a
n
s
fo
r
ms
(
B
.
Ma
ma
th
a
)
3089
r
o
w
s
o
f
d
ata
an
d
n
o
w
t
h
e
h
ig
h
p
as
s
co
m
p
o
n
e
n
ts
ar
e
s
ep
ar
ated
an
d
p
lace
d
b
y
t
h
e
s
id
e
o
f
th
e
lo
w
p
a
s
s
co
m
p
o
n
e
n
t
s
.
T
h
e
ap
p
licatio
n
o
f
w
av
ele
t
tr
a
n
s
f
o
r
m
to
an
i
m
ag
e,
t
h
e
o
b
tai
n
ed
ap
p
r
o
x
i
m
atio
n
s
u
b
s
i
g
n
al
s
h
o
w
s
t
h
e
g
en
er
al
tr
en
d
o
f
p
ix
e
l
v
alu
e
s
a
n
d
th
r
ee
d
etail
s
u
b
s
ig
n
al
s
s
h
o
w
th
e
v
er
tica
l,
h
o
r
izo
n
ta
l
a
n
d
d
iag
o
n
al
d
etail
s
o
r
ch
an
g
es
i
n
th
e
i
m
ag
e.
T
h
is
ca
n
b
e
r
ep
ea
ted
till
r
eq
u
ir
ed
lev
el
o
f
d
ec
o
m
p
o
s
itio
n
i
s
r
ea
ch
e
d
.
T
h
e
f
ilter
s
h
av
e
th
e
ef
f
ec
t
o
f
s
ep
ar
atin
g
o
u
t
f
i
n
er
an
d
f
i
n
er
d
etails.
T
h
is
f
il
ter
in
g
p
r
o
ce
s
s
u
s
i
n
g
w
a
v
elet
ca
n
b
e
u
n
d
er
s
to
o
d
f
r
o
m
t
h
e
Fi
g
u
r
e
1
to
Fig
u
r
e
3
.
LL
HL
LH
HH
LL
HL
HL
LH
HH
LH
HH
HL
HL
HL
LH
HH
LH
HH
LH
HH
Fig
u
r
e
1
.
1
st
L
ev
el
Dec
o
m
p
o
s
itio
n
Fig
u
r
e
2
.
2
nd
L
ev
el
d
ec
o
m
p
o
s
itio
n
Fi
g
u
r
e
3
.
3
rd
L
ev
el
Dec
o
m
p
o
s
itio
n
I
n
2
D,
th
e
i
m
a
g
es
ar
e
co
n
s
id
er
ed
to
b
e
m
atr
ice
s
w
it
h
M
r
o
w
s
a
n
d
N
co
l
u
m
n
s
.
A
t
e
v
er
y
le
v
el
f
o
u
r
s
u
b
i
m
a
g
es
ar
e
o
b
tai
n
ed
,
th
e
ap
p
r
o
x
im
a
tio
n
,
t
h
e
v
er
tical
d
e
tail,
th
e
h
o
r
izo
n
ta
l
d
etail
a
n
d
th
e
d
ia
g
o
n
al
d
etail.
T
o
g
et
th
e
n
e
x
t
le
v
el
o
f
d
ec
o
m
p
o
s
i
tio
n
th
e
ap
p
r
o
x
i
m
atio
n
d
etail
is
u
s
ed
.
T
h
e
s
h
ip
I
S
A
R
i
m
a
g
e
an
d
i
ts
d
ec
o
m
p
o
s
itio
n
u
p
to
3
rd
lev
el
u
s
i
n
g
w
av
ele
t tr
an
s
f
o
r
m
ar
e
s
h
o
w
n
in
F
ig
u
r
e
4
an
d
Fig
u
r
e
5
.
Fig
u
r
e
4
.
I
SA
R
I
m
a
g
e
o
f
a
s
h
ip
Fig
u
r
e
5
.
I
SA
R
I
m
a
g
e
o
f
a
s
h
ip
T
h
e
w
a
v
elet
tr
a
n
s
f
o
r
m
is
a
p
p
lied
to
th
e
s
et
o
f
I
S
AR
i
m
ag
e
s
co
n
s
id
er
ed
.
T
h
e
ap
p
r
o
x
i
m
at
io
n
,
h
o
r
izo
n
tal,
v
er
tical
a
n
d
d
iag
o
n
al
d
etails
ar
e
co
m
p
u
ted
.
T
h
e
w
a
v
elet
tr
an
s
f
o
r
m
is
ap
p
lied
t
o
th
e
ap
p
r
o
x
i
m
at
io
n
d
etail
u
p
to
3
rd
le
v
el
to
co
m
p
u
te
th
e
d
etai
ls
at
ea
ch
lev
el.
T
h
e
ap
p
r
o
x
i
m
atio
n
,
h
o
r
izo
n
tal,
v
er
tical
an
d
d
iag
o
n
al
d
etails
an
d
th
eir
co
m
b
i
n
atio
n
at
ea
ch
lev
el
ar
e
tak
e
n
in
d
ep
en
d
en
tl
y
o
n
e
s
et
at
a
ti
m
e.
P
r
in
cip
al
co
m
p
o
n
e
n
t
an
al
y
s
is
is
ap
p
lied
to
ea
ch
s
e
t
o
f
v
al
u
e
s
.
I
n
P
C
A
,
th
e
ab
o
v
e
m
en
tio
n
ed
f
ea
t
u
r
e
v
al
u
es
o
f
e
ac
h
s
et
ar
e
ar
r
an
g
ed
in
th
e
f
o
r
m
o
f
a
m
atr
ix
.
T
h
en
co
v
ar
ian
ce
m
a
tr
ix
i
s
co
m
p
u
te
d
.
T
h
e
E
ig
en
v
a
lu
e
s
o
f
th
e
co
v
ar
ian
ce
m
atr
i
x
ar
e
co
m
p
u
ted
a
n
d
u
s
ed
as
a
f
ea
t
u
r
e
v
ec
to
r
.
T
h
e
class
i
f
icat
io
n
p
er
f
o
r
m
an
ce
o
f
all
th
e
co
m
b
in
at
io
n
s
o
f
w
a
v
elet
co
ef
f
icie
n
t
s
tak
e
n
as
f
ea
t
u
r
e
v
ec
to
r
s
is
co
m
p
u
ted
f
o
r
t
h
e
co
n
s
id
er
ed
d
ata
s
et
o
f
m
ea
s
u
r
ed
s
h
ip
I
S
A
R
i
m
a
g
es.
2
.
2
.
Wa
t
er
s
hed T
ra
n
s
f
o
r
m
Dir
ec
t
ap
p
licatio
n
o
f
W
ater
s
h
ed
tr
an
s
f
o
r
m
to
a
g
r
ad
ien
t
i
m
a
g
e
ca
u
s
e
s
o
v
er
s
eg
m
e
n
t
atio
n
.
T
h
e
co
n
ce
p
t
o
f
m
ar
k
er
s
i
s
u
s
ed
to
co
n
tr
o
l
o
v
er
s
eg
m
e
n
tat
io
n
.
A
m
ar
k
er
is
a
co
n
n
ec
ted
co
m
p
o
n
en
t
b
elo
n
g
i
n
g
to
an
i
m
a
g
e.
First a
s
e
t o
f
i
n
ter
n
al
m
ar
k
er
s
th
at
ar
e
i
n
s
id
e
ea
c
h
o
f
t
h
e
o
b
j
ec
t o
f
in
ter
est an
d
a
s
et
o
f
ex
ter
n
al
m
ar
k
er
s
th
at
ar
e
co
n
tain
ed
in
th
e
b
ac
k
g
r
o
u
n
d
ar
e
to
B
E
id
en
t
if
ied
.
T
h
ese
m
ar
k
er
s
ar
e
u
s
ed
to
m
o
d
i
f
y
t
h
e
g
r
ad
ien
t
i
m
a
g
e.
T
h
e
to
o
lb
o
x
f
u
n
ctio
n
i
m
ex
ten
d
ed
m
in
(
f
,
h
)
i
s
u
s
ed
t
o
co
m
p
u
te
t
h
e
s
et
o
f
lo
w
s
p
o
t
s
in
th
e
i
m
a
g
e
t
h
at
ar
e
d
ee
p
er
th
an
th
e
ir
s
u
r
r
o
u
n
d
in
g
s
b
y
a
ce
r
tain
h
ei
g
h
t
t
h
r
es
h
o
ld
h
to
o
b
tain
th
e
in
ter
n
a
l
m
a
r
k
er
i
m
a
g
e
i
m
.
T
h
e
n
ex
t
s
tep
i
s
to
f
in
d
t
h
e
e
x
ter
n
al
m
ar
k
er
s
o
r
p
ix
els
th
at
b
elo
n
g
to
th
e
b
ac
k
g
r
o
u
n
d
b
y
a
p
p
ly
i
n
g
W
ater
s
h
ed
tr
an
s
f
o
r
m
o
f
t
h
e
d
is
ta
n
ce
tr
a
n
s
f
o
r
m
o
f
t
h
e
i
n
ter
n
a
l
m
ar
k
e
r
i
m
a
g
e
i
m
.
T
h
u
s
o
b
tain
ed
i
n
ter
n
al
an
d
e
x
ter
n
a
l
m
ar
k
er
s
ar
e
u
s
ed
to
m
o
d
if
y
t
h
e
g
r
ad
ien
t
i
m
a
g
e
b
y
i
m
p
o
s
i
n
g
r
eg
io
n
al
m
i
n
i
m
a
at
t
h
e
lo
ca
ti
o
n
s
o
f
in
ter
n
al
a
n
d
ex
ter
n
al
m
ar
k
er
s
u
s
in
g
t
h
e
f
o
ll
o
w
i
n
g
t
w
o
M
A
T
L
A
B
p
r
o
g
r
am
s
tate
m
e
n
ts
.
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.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
3
08
7
–
3
09
3
3090
g
2
=
i
m
i
m
p
o
s
e
m
i
n
(
g
,
i
m
|
e
m
)
L
2
=
w
ater
s
h
ed
(
g
2
)
T
h
e
g
2
is
th
e
m
ar
k
er
m
o
d
if
i
ed
g
r
ad
ien
t
i
m
ag
e.
F
in
al
l
y
t
h
e
W
ater
s
h
ed
tr
a
n
s
f
o
r
m
o
f
t
h
e
m
ar
k
e
r
m
o
d
i
f
ied
g
r
ad
ien
t
i
m
ag
e
is
co
m
p
u
ted
to
o
b
tain
L
2
.
W
it
h
t
h
e
b
elo
w
p
r
o
g
r
a
m
s
tate
m
en
t,
t
h
e
W
ater
s
h
ed
r
id
g
e
lin
es a
r
e
s
u
p
er
i
m
p
o
s
ed
o
n
t
h
e
o
r
ig
in
al
i
m
a
g
e
to
ac
h
ie
v
e
m
u
c
h
i
m
p
r
o
v
ed
s
e
g
m
en
ta
tio
n
.
F (
L
2
==
0
)
=2
5
5
T
h
e
I
SA
R
i
m
a
g
e
o
f
a
s
h
ip
a
n
d
th
e
i
m
a
g
e
w
it
h
i
m
p
r
o
v
ed
s
eg
m
e
n
tat
io
n
ar
e
s
h
o
w
n
i
n
Fi
g
u
r
e
6
an
d
Fig
u
r
e
7
.
F
ig
u
r
e
6
.
Sh
ip
I
S
A
R
I
m
ag
e
i
n
B
in
ar
y
Fo
r
m
at
Fig
u
r
e
7
.
Seg
m
e
n
ted
S
h
ip
I
S
AR
I
m
a
g
e
in
B
i
n
ar
y
Fo
r
m
at
No
w
w
a
v
elet
tr
an
s
f
o
r
m
is
a
p
p
lied
to
th
e
im
a
g
e
w
it
h
i
m
p
r
o
v
ed
s
eg
m
e
n
tatio
n
ac
h
ie
v
e
d
th
r
o
u
g
h
W
ater
s
h
ed
s
e
g
m
e
n
tatio
n
u
p
t
o
th
ir
d
le
v
el.
T
h
e
o
b
tain
ed
d
et
ail
m
a
tr
ix
at
ea
c
h
le
v
el
i
s
u
s
ed
as
a
f
ea
t
u
r
e
v
ec
to
r
an
d
g
i
v
e
n
as
i
n
p
u
t
to
th
e
c
l
ass
i
f
ier
.
T
h
e
ap
p
r
o
x
i
m
atio
n
,
h
o
r
izo
n
tal,
v
er
tical
a
n
d
d
iag
o
n
al
d
etail
m
atr
ice
s
co
m
p
u
ted
u
p
to
t
h
ir
d
lev
el
d
ec
o
m
p
o
s
itio
n
ar
e
tak
e
n
a
n
d
p
r
in
cip
al
co
m
p
o
n
e
n
t
a
n
al
y
s
is
i
s
ap
p
lied
an
d
co
m
p
u
ted
E
i
g
en
v
a
lu
e
s
ar
e
u
s
ed
as
f
ea
tu
r
e
v
ec
to
r
s
to
b
e
g
iv
en
as
i
n
p
u
t
to
t
h
e
cla
s
s
i
f
ier
.
T
h
e
co
m
b
i
n
atio
n
o
f
ap
p
r
o
x
im
a
tio
n
an
d
o
t
h
er
t
h
r
ee
d
etails
(
th
e
co
m
p
lete
s
et
o
f
wav
elet
co
e
f
f
icie
n
ts
)
co
m
p
u
ted
u
p
to
t
h
ir
d
le
v
el
is
also
g
i
v
en
as i
n
p
u
t to
th
e
clas
s
if
ier
an
d
t
h
e
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
is
co
m
p
u
ted
.
2
.
3
.
RG
B
I
m
a
g
es
T
h
e
co
lo
r
im
ag
e
s
in
R
GB
f
o
r
m
at
ar
e
tak
e
n
.
An
i
m
ag
e
i
n
R
GB
f
o
r
m
at
ca
n
b
e
co
n
s
id
er
ed
as a
s
tack
o
f
th
r
ee
t
w
o
d
i
m
en
s
io
n
al
m
atr
ic
es.
T
h
e
th
r
ee
i
m
ag
e
s
f
o
r
m
in
g
an
R
GB
co
lo
r
im
a
g
e
ar
e
th
e
r
ed
,
g
r
ee
n
an
d
b
lu
e
co
m
p
o
n
e
n
t i
m
a
g
es.
I
f
t
h
e
R
G
B
im
a
g
e
b
elo
n
g
s
to
a
clas
s
d
o
u
b
le,
th
e
r
an
g
e
o
f
p
i
x
el
in
ten
s
i
t
y
v
al
u
e
s
is
[
0
,
1
]
.
I
f
th
e
R
GB
i
m
a
g
e
is
o
f
clas
s
u
n
i
t8
,
th
e
r
an
g
e
o
f
p
i
x
el
v
al
u
e
s
li
e
b
et
w
ee
n
0
an
d
2
5
5
.
Fo
r
u
n
it
1
6
class
i
m
a
g
e,
t
h
e
r
an
g
e
o
f
p
i
x
el
v
al
u
es
w
ill
b
e
[
0
,
6
6
5
5
5
]
.
T
h
e
n
u
m
b
er
o
f
b
it
s
th
at
r
ep
r
esen
t
th
e
p
ix
e
l
v
alu
e
s
o
f
t
h
e
co
m
p
o
n
e
n
t
i
m
a
g
e
is
ca
lled
th
e
b
it
d
ep
th
o
f
th
e
R
GB
i
m
a
g
e.
W
h
en
th
e
co
m
p
o
n
en
t
i
m
a
g
e
is
a
t
w
o
b
it
i
m
a
g
e,
t
h
e
co
r
r
esp
o
n
d
in
g
d
ep
th
o
f
t
h
e
R
GB
i
m
a
g
e
i
s
2
4
b
its
d
ee
p
.
I
f
b
is
t
h
e
n
u
m
b
er
o
f
b
its
i
n
ea
c
h
co
m
p
o
n
e
n
t
i
m
ag
e
th
en
t
h
e
n
u
m
b
er
o
f
co
lo
r
s
in
a
n
i
m
a
g
e
i
s
(2
b
)
3
.
T
h
e
R
,
G,
B
co
m
p
o
n
en
t
i
m
a
g
e
m
atr
ice
s
ar
e
e
x
tr
ac
ted
f
r
o
m
th
e
R
GB
i
m
a
g
e
m
atr
i
x
.
Fo
r
t
h
e
o
b
tain
ed
ea
ch
co
m
p
o
n
e
n
t
i
m
ag
e
m
atr
i
x
,
th
e
m
ea
n
an
d
s
ta
n
d
ar
d
d
ev
iatio
n
ar
e
co
m
p
u
ted
.
T
h
e
v
alu
e
s
o
b
tain
ed
f
r
o
m
th
e
th
r
ee
co
m
p
o
n
en
t
i
m
a
g
es
ar
e
co
m
b
i
n
ed
to
f
o
r
m
a
f
ea
tu
r
e
v
ec
to
r
th
at
h
a
s
g
i
v
en
g
o
o
d
cl
ass
i
f
icatio
n
r
esu
lt
s
.
Si
m
i
lar
l
y
,
t
h
e
co
m
p
o
n
en
t
m
a
tr
ices
ar
e
co
n
ca
te
n
ated
in
co
m
b
in
at
io
n
s
R
&
G,
G
&
B
a
n
d
R
&
B
.
T
h
e
m
ea
n
,
m
ed
ian
a
n
d
s
tan
d
ar
d
d
ev
iatio
n
v
al
u
es
ar
e
co
m
p
u
ted
f
o
r
th
e
co
m
b
i
n
ed
co
m
p
o
n
en
t
m
atr
i
ce
s
(
R
G,
GB
,
an
d
R
B
)
.
T
h
e
o
b
tain
ed
s
tatis
t
ical
v
alu
e
s
o
f
t
h
e
co
m
p
o
n
e
n
t
i
m
a
g
es
R
,
G,
B
an
d
co
m
b
in
at
io
n
s
R
G,
GB
an
d
R
B
ar
e
g
iv
e
n
a
s
a
f
ea
t
u
r
e
v
ec
to
r
s
to
th
e
n
eu
r
al
n
et
w
o
r
k
f
o
r
class
i
f
ica
tio
n
.
T
h
e
t
w
o
f
ea
t
u
r
e
v
ec
to
r
s
d
is
cu
s
s
ed
ar
e
g
i
v
e
n
b
elo
w
:
(
2
)
(
3
)
T
h
e
co
lu
m
n
s
o
f
th
e
m
a
tr
ix
ar
e
co
n
ca
ten
a
ted
to
f
o
r
m
a
s
i
n
g
le
co
lu
m
n
f
ea
tu
r
e
v
ec
to
r
s
i
n
ce
t
h
e
i
n
p
u
t
to
th
e
n
eu
r
al
n
et
w
o
r
k
b
ased
cla
s
s
if
ier
is
r
eq
u
ir
ed
to
b
e
s
i
n
g
le
d
i
m
e
n
s
io
n
al
v
ec
to
r
.
I
n
b
o
th
th
e
ca
s
es
th
e
len
g
t
h
o
f
th
e
f
ea
tu
r
e
v
ec
to
r
is
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
I
S
A
R
I
ma
g
e
C
la
s
s
ifica
tio
n
w
it
h
W
a
ve
let
a
n
d
W
a
ter
s
h
ed
Tr
a
n
s
fo
r
ms
(
B
.
Ma
ma
th
a
)
3091
3.
DATA S
E
T
G
E
N
E
RAT
I
O
N
I
n
th
i
s
w
o
r
k
,
t
h
e
d
ata
s
et
is
g
e
n
er
ated
b
y
r
o
tatin
g
t
h
e
s
h
ip
I
S
AR
i
m
ag
e
s
o
b
tain
ed
f
r
o
m
liter
atu
r
e.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
clas
s
i
f
icatio
n
o
f
ea
ch
o
f
th
e
f
ea
t
u
r
e
v
ec
to
r
s
d
is
cu
s
s
ed
i
n
p
r
ev
io
u
s
s
ec
tio
n
s
is
co
m
p
u
ted
f
o
r
t
w
o
s
h
ip
s
I
S
AR
i
m
ag
e
s
m
ea
s
u
r
ed
u
s
i
n
g
I
n
s
tr
u
m
e
n
tatio
n
R
a
d
ar
.
T
h
er
e
ar
e
t
w
el
v
e
i
m
a
g
es
f
o
r
ea
ch
o
f
t
h
e
t
w
o
s
h
ip
s
.
So
th
er
e
ar
e
t
w
e
n
t
y
f
o
u
r
i
m
ag
e
s
in
t
h
e
tr
ain
i
n
g
s
et
an
d
test
s
et
in
d
ep
en
d
e
n
tl
y
.
T
h
e
class
i
f
icatio
n
p
er
f
o
r
m
a
n
ce
f
o
r
ea
ch
o
f
th
e
f
e
atu
r
e
v
ec
to
r
u
s
in
g
t
h
e
ar
tif
ic
ial
n
eu
r
al
n
et
w
o
r
k
is
g
i
v
en
i
n
T
ab
le
1
to
T
a
b
le
9
.
4.
F
E
AT
U
RE
V
E
C
T
O
R
DE
SC
RIPT
I
O
N
T
h
e
s
ize
o
f
th
e
i
m
a
g
es
i
n
tr
a
in
i
n
g
a
n
d
test
s
et
i
s
3
0
x
3
0
.
E
x
ce
p
t
f
o
r
R
GB
i
m
a
g
es,
ea
c
h
i
m
a
g
e
is
r
esized
to
3
.
I
n
ca
s
e
o
f
W
av
elet
co
ef
f
ic
ien
t
s
,
th
e
s
ize
o
f
t
h
e
co
m
b
i
n
ed
m
atr
i
x
o
f
ap
p
r
o
x
i
m
atio
n
,
h
o
r
izo
n
tal,
v
er
tical
an
d
d
iag
o
n
al
co
ef
f
ici
en
ts
is
1
5
x
6
0
,
8
x
3
2
an
d
4
x
1
6
f
o
r
th
e
th
r
ee
lev
els
o
f
d
ec
o
m
p
o
s
itio
n
r
esp
.
So
th
e
s
ize
o
f
th
e
co
m
p
u
ted
f
ea
t
u
r
e
v
ec
to
r
af
ter
ap
p
licatio
n
o
f
P
C
A
is
9
0
0
,
2
5
6
an
d
1
2
8
f
o
r
th
e
f
ir
s
t,
s
ec
o
n
d
an
d
th
ir
d
lev
el
d
ec
o
m
p
o
s
itio
n
.
T
h
e
s
ize
o
f
th
e
in
d
i
v
id
u
a
l
ap
p
r
o
x
i
m
atio
n
,
h
o
r
izo
n
tal,
v
er
tical
o
r
d
iag
o
n
al
d
etails
is
1
5
x
1
5
,
8
x
8
an
d
4
x
4
f
o
r
th
e
t
h
r
ee
lev
el
s
r
asp
.
.
.
So
t
h
e
s
ize
o
f
t
h
e
f
i
n
al
f
ea
t
u
r
e
v
ec
to
r
w
h
e
n
ap
p
r
o
x
i
m
a
tio
n
o
r
h
o
r
izo
n
tal
o
r
v
er
tical
o
r
d
iag
o
n
al
d
etail
alo
n
e
u
s
ed
is
2
5
6
,
6
4
an
d
1
6
r
esp
ec
tiv
el
y
f
o
r
th
e
o
n
e,
t
w
o
an
d
th
r
ee
d
ec
o
m
p
o
s
itio
n
s
r
esp
ec
ti
v
el
y
.
T
h
e
s
ize
o
f
t
h
e
f
ea
t
u
r
e
v
ec
to
r
s
is
s
a
m
e
w
h
e
n
i
m
a
g
e
s
w
it
h
W
ater
s
h
ed
s
eg
m
e
n
tatio
n
ar
e
u
s
ed
.
I
n
ca
s
e
o
f
R
GB
i
m
ag
e
s
,
t
h
e
s
i
ze
o
f
t
h
e
i
m
a
g
e
is
1
0
0
x
1
0
0
x
3
.
T
h
e
s
ize
o
f
th
e
ea
ch
co
m
p
o
n
en
t
i
m
a
g
e
is
1
0
0
x
1
0
0
.
T
h
e
f
ea
t
u
r
e
v
ec
to
r
s
o
b
tain
ed
f
r
o
m
in
d
i
v
id
u
al
R
,
G,
B
i
m
ag
e
s
o
r
co
m
b
i
n
ed
i
m
ag
es
R
G
,
GB
,
R
B
th
e
f
ea
t
u
r
e
v
ec
to
r
g
iv
e
n
b
y
(
2
)
an
d
(
3
)
.
T
h
e
m
ea
n
an
d
s
ta
n
d
ar
d
d
ev
iatio
n
v
al
u
es
ar
e
co
m
p
u
ted
f
o
r
t
h
e
s
h
ip
I
SAR
i
m
ag
e
s
s
h
o
w
n
i
n
Fi
g
u
r
e
8
to
Fig
u
r
e
11
o
f
t
w
o
tar
g
et
s
at
t
w
o
d
if
f
er
e
n
t a
s
p
ec
t a
n
g
le
s
.
Fig
u
r
e
8
.
T
a
r
g
et
1
I
SA
R
i
m
a
g
e
at
asp
ec
t 0
0
Fig
u
r
e
9
.
T
ag
et
2
I
SA
R
i
m
a
g
e
at
asp
ec
t 0
0
Fig
u
r
e
1
0
.
T
ar
g
et
1
I
SA
R
i
m
a
g
e
at
asp
ec
t 3
0
0
Fig
u
r
e
1
1
.
T
ar
g
et
2
I
SA
R
i
m
a
g
e
at
asp
ec
t 3
0
0
T
h
e
co
m
p
u
ted
s
tatis
tical
m
o
m
e
n
t
v
al
u
es
f
o
r
s
in
g
le
R
,
G,
B
co
m
p
o
n
e
n
t
i
m
ag
e
s
an
d
co
m
b
i
n
ed
R
G,
GB
,
R
B
o
f
th
e
I
S
A
R
i
m
ag
es a
t a
s
p
ec
t a
n
g
le
0
° a
n
d
3
0
° o
f
t
h
e
t
w
o
tar
g
et
s
s
h
o
w
n
ar
e
g
i
v
en
in
T
ab
les 1
an
d
T
ab
le
2
.
T
ab
le
1
.
Statis
tical
Mo
m
e
n
ts
o
f
Si
n
g
le
C
o
m
p
o
n
en
t I
m
a
g
e
s
(
R
,
G,
B
)
f
o
r
th
e
T
w
o
T
ar
g
et
s
A
t
a
sp
e
c
t
a
n
g
l
e
0
°
A
t
a
sp
e
c
t
a
n
g
l
e
3
0
°
T
a
r
g
e
t
1
T
a
r
g
e
t
2
T
a
r
g
e
t
1
T
a
r
g
e
t
2
1
5
.
0
4
4
7
1
4
.
1
1
6
2
1
4
.
9
7
2
8
1
4
.
1
2
8
8
3
0
.
0
2
8
4
2
5
.
1
1
0
7
2
7
.
6
7
7
5
2
2
.
2
3
2
4
6
.
0
4
9
3
5
.
0
3
2
2
5
.
7
1
5
8
4
.
7
7
7
6
1
2
.
5
1
2
0
1
1
.
6
6
6
1
1
0
.
6
7
8
6
9
.
4
1
7
6
1
5
.
3
8
8
3
7
.
7
9
3
2
1
5
.
3
4
1
5
7
.
6
4
1
3
3
0
.
9
8
2
5
1
9
.
8
5
2
3
2
8
.
3
5
3
9
1
6
.
6
2
9
6
T
ab
le
2
.
Statis
tical
Mo
m
e
n
ts
o
f
C
o
m
p
o
n
en
t I
m
a
g
es C
o
m
b
in
ed
(
R
G,
GB
,
an
d
R
B
)
A
t
a
sp
e
c
t
a
n
g
l
e
0
°
A
t
a
sp
e
c
t
a
n
g
l
e
3
0
°
T
a
r
g
e
t
1
T
a
r
g
e
t
2
T
a
r
g
e
t
1
T
a
r
g
e
t
2
1
0
.
5
4
7
0
9
.
5
7
4
2
1
0
.
3
4
4
3
9
.
4
5
2
9
1
2
.
3
8
6
2
9
.
5
0
6
8
1
2
.
0
2
0
0
9
.
0
6
1
4
1
0
.
7
1
8
8
6
.
4
1
2
7
1
0
.
5
2
8
6
6
.
2
0
9
5
1
3
.
0
6
0
6
5
.
7
8
8
5
1
2
.
4
9
8
4
5
.
0
9
9
7
1
5
.
2
1
6
5
1
0
.
9
5
4
7
1
5
.
1
5
7
2
1
0
.
8
8
4
8
0
.
6
7
4
7
3
.
7
1
8
3
0
.
4
7
8
3
3
.
9
6
1
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.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
3
08
7
–
3
09
3
3092
5.
CL
AS
SI
F
I
CAT
I
O
N
P
E
RF
O
RM
ANCE
C
las
s
i
f
icatio
n
is
d
o
n
e
u
s
i
n
g
P
r
o
b
ab
ilis
tic
n
e
u
r
al
n
et
w
o
r
k
.
M
A
T
L
A
B
n
eu
r
al
n
et
w
o
r
k
to
o
lb
o
x
is
u
s
ed
.
T
h
e
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
is
ca
lcu
lated
f
r
o
m
t
h
e
n
u
m
b
er
o
f
s
u
cc
ess
f
u
l
clas
s
i
f
ica
tio
n
s
a
n
d
th
e
to
tal
n
u
m
b
er
o
f
i
m
a
g
es
te
s
ted
.
Fo
r
ea
ch
o
f
t
h
e
f
ea
t
u
r
e
v
ec
t
o
r
s
t
u
d
ied
in
th
i
s
p
ap
er
,
th
e
clas
s
i
f
i
ca
tio
n
is
te
s
ted
f
o
r
th
e
I
S
AR
i
m
a
g
es
o
f
t
w
o
s
h
i
p
tar
g
ets
m
ea
s
u
r
ed
u
s
i
n
g
h
i
g
h
r
eso
l
u
tio
n
r
ad
ar
an
d
p
er
f
o
r
m
an
ce
is
g
i
v
e
n
i
n
T
ab
le
3
to
T
ab
le
8
.
T
h
e
class
i
f
icatio
n
p
er
f
o
r
m
an
ce
o
f
a
n
y
f
e
atu
r
e
v
ec
to
r
s
p
ec
i
f
ied
i
n
t
h
is
w
o
r
k
is
n
o
t
s
a
m
e
f
o
r
th
e
t
w
o
tar
g
et
s
co
n
s
id
er
ed
.
T
h
e
class
if
icatio
n
p
er
f
o
r
m
an
c
e
w
h
e
n
w
av
e
let
co
ef
f
icien
ts
,
ap
p
r
o
x
im
a
tio
n
an
d
th
e
o
th
er
th
r
ee
d
etails
co
m
b
i
n
ed
to
g
et
h
er
u
s
ed
as
f
e
atu
r
e
v
ec
to
r
f
o
r
I
S
A
R
i
m
a
g
e
s
w
it
h
a
n
d
w
it
h
o
u
t
s
eg
m
e
n
tat
io
n
f
o
r
t
h
r
ee
lev
e
l
d
ec
o
m
p
o
s
itio
n
s
is
g
i
v
e
n
i
n
T
ab
le
3
.
Si
m
ilar
l
y
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
ap
p
r
o
x
i
m
atio
n
,
h
o
r
iz
o
n
tal,
v
er
t
ical
a
n
d
d
iag
o
n
al
d
etail
s
ta
k
en
in
d
ep
e
n
d
en
tl
y
f
o
r
th
e
s
h
ip
I
S
A
R
i
m
ag
es
w
i
th
an
d
w
it
h
o
u
t
ap
p
licatio
n
o
f
W
ater
s
h
ed
s
eg
m
e
n
tatio
n
i
s
g
i
v
e
n
in
T
ab
le
4
to
T
ab
le
7
.
T
h
e
cl
ass
i
f
icatio
n
p
er
f
o
r
m
a
n
c
e
o
f
co
lo
r
m
o
m
e
n
ts
co
m
p
u
te
d
f
o
r
s
i
n
g
le
R
,
G,
B
co
m
p
o
n
en
t
i
m
a
g
e
s
an
d
co
m
b
in
ed
co
m
p
o
n
e
n
t i
m
a
g
es
R
G,
GB
an
d
R
B
is
g
i
v
e
n
i
n
T
ab
le
8
.
T
ab
le
3
.
C
lass
if
icatio
n
P
er
f
o
r
m
an
ce
P
er
ce
n
t
a
g
e
f
o
r
t
h
e
A
p
p
r
o
x
i
m
atio
n
,
Ho
r
izo
n
tal,
V
er
t
ic
al
an
d
Diag
o
n
a
l
Deta
ils
C
o
m
b
in
ed
W
i
t
h
o
u
t
S
e
g
me
n
t
a
t
i
o
n
W
i
t
h
se
g
me
n
t
a
t
i
o
n
T
a
r
g
e
t
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
1
5
8
.
3
25
25
3
3
.
3
50
8
3
.
3
2
9
1
.
6
9
1
.
6
8
3
.
3
9
1
.
6
5
8
.
3
5
8
.
3
T
ab
le
4
.
C
lass
if
icatio
n
P
er
f
o
r
m
an
ce
P
er
ce
n
ta
g
e
f
o
r
t
h
e
A
p
p
r
o
x
i
m
atio
n
Deta
il
s
T
ab
le
5
.
C
lass
if
icatio
n
P
er
f
o
r
m
an
ce
P
er
ce
n
ta
g
e
f
o
r
t
h
e
Ho
r
izo
n
tal
Det
ail
s
W
i
t
h
o
u
t
S
e
g
me
n
t
a
t
i
o
n
W
i
t
h
S
e
g
me
n
t
a
t
i
o
n
T
a
r
g
e
t
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
1
9
1
.
6
6
6
.
6
6
6
.
6
6
6
.
6
6
6
.
6
4
1
.
6
2
75
6
6
.
6
75
3
3
.
3
3
3
.
3
6
6
.
6
T
ab
le
6
.
C
lass
if
icatio
n
P
er
f
o
r
m
an
ce
P
er
ce
n
ta
g
e
f
o
r
t
h
e
Ver
tical
De
tails
W
i
t
h
o
u
t
S
e
g
me
n
t
a
t
i
o
n
W
i
t
h
S
e
g
me
n
t
a
t
i
o
n
T
a
r
g
e
t
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
1
6
6
.
6
50
6
6
.
6
75
5
8
.
3
75
2
3
3
.
3
5
8
.
3
5
8
.
3
4
1
.
6
5
8
.
3
4
1
.
6
T
ab
le
7
.
C
lass
if
icatio
n
P
er
f
o
r
m
an
ce
P
er
ce
n
ta
g
e
f
o
r
t
h
e
Dia
g
o
n
al
De
tai
ls
W
i
t
h
o
u
t
S
e
g
me
n
t
a
t
i
o
n
W
i
t
h
S
e
g
me
n
t
a
t
i
o
n
T
a
r
g
e
t
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
1
5
8
.
3
5
8
.
3
8
3
.
3
3
3
.
3
75
50
2
5
8
.
3
50
6
6
.
6
50
3
3
.
3
5
8
.
3
T
ab
le
8
.
C
lass
if
icatio
n
P
er
f
o
r
m
an
ce
P
er
ce
n
t
a
g
e
o
f
t
h
e
C
o
lo
r
Mo
m
e
n
ts
C
o
mp
u
t
e
d
f
o
r
S
i
n
g
l
e
R
,
G,
B
C
o
m
p
o
n
e
n
t
s
C
o
mp
u
t
e
d
f
o
r
Tw
o
C
o
mp
o
n
e
n
t
s
R
G
,
G
B
,
RB
T
a
r
g
e
t
1
6
6
.
6
9
1
.
6
T
a
r
g
e
t
2
8
6
.
3
3
75
6.
CO
NCLU
SI
O
N
I
n
t
h
is
w
o
r
k
,
w
e
co
m
p
ar
ed
t
h
e
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
o
f
th
e
d
i
f
f
er
en
t
f
ea
tu
r
e
v
ec
to
r
s
i.
e.
w
a
v
elet
co
ef
f
icie
n
t
s
co
m
b
i
n
ed
an
d
in
d
iv
id
u
al
a
t
d
if
f
er
en
t
le
v
els
o
f
d
ec
o
m
p
o
s
it
io
n
a
n
d
s
ta
tis
tica
l
m
o
m
en
ts
f
o
r
th
e
m
ea
s
u
r
ed
s
h
ip
I
S
AR
i
m
a
g
e
d
ata
s
et.
T
h
e
p
er
f
o
r
m
a
n
ce
is
g
i
v
e
n
co
n
s
id
er
in
g
th
e
clas
s
i
f
icatio
n
o
f
b
o
th
th
e
W
i
t
h
o
u
t
S
e
g
me
n
t
a
t
i
o
n
W
i
t
h
S
e
g
me
n
t
a
t
i
o
n
T
a
r
g
e
t
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
L
e
v
e
l
1
L
e
v
e
l
2
L
e
v
e
l
3
1
3
3
.
3
4
1
.
6
3
3
.
3
50
50
6
6
.
6
2
4
1
.
6
9
1
.
6
9
1
.
6
5
8
.
3
5
8
.
3
3
3
.
3
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
I
S
A
R
I
ma
g
e
C
la
s
s
ifica
tio
n
w
it
h
W
a
ve
let
a
n
d
W
a
ter
s
h
ed
Tr
a
n
s
fo
r
ms
(
B
.
Ma
ma
th
a
)
3093
tar
g
ets
co
n
s
id
er
ed
.
Fro
m
t
h
e
T
ab
le
3
to
ab
le
8
,
it
is
o
b
s
e
r
v
ed
th
at
t
h
e
w
h
o
le
s
et
o
f
wav
elet
co
e
f
f
icien
ts
(
ap
p
r
o
x
im
atio
n
,
h
o
r
izo
n
tal,
v
er
tical
an
d
d
iag
o
n
al)
at
f
ir
s
t
lev
el
d
ec
o
m
p
o
s
itio
n
h
as
s
h
o
w
n
s
at
is
f
ac
to
r
y
class
i
f
icatio
n
p
er
f
o
r
m
a
n
ce
f
o
r
th
e
I
S
A
R
i
m
a
g
es
w
it
h
o
u
t
s
eg
m
e
n
tatio
n
(
T
ab
le
3
)
.
T
h
is
co
m
b
i
n
ed
s
et
o
f
w
a
v
elet
co
e
f
f
ic
ien
t
s
co
m
p
u
ted
at
th
ir
d
lev
el
d
ec
o
m
p
o
s
i
tio
n
h
as
s
h
o
w
n
s
a
tis
f
ac
to
r
y
cla
s
s
i
f
ic
atio
n
p
er
f
o
r
m
a
n
ce
f
o
r
I
S
A
R
i
m
ag
e
s
w
it
h
s
e
g
m
e
n
tatio
n
(
T
a
b
le
3
)
.
T
h
e
ap
p
r
o
x
i
m
atio
n
d
etails
ar
e
n
o
t
g
o
o
d
f
o
r
class
i
f
icatio
n
o
f
I
SAR
i
m
a
g
es
w
i
th
an
d
w
i
th
o
u
t
s
e
g
m
e
n
tatio
n
(
T
ab
le
4
)
.
W
h
er
ea
s
t
h
e
h
o
r
izo
n
tal
d
etail
co
ef
f
icien
ts
co
m
p
u
ted
at
all
th
e
t
h
r
ee
le
v
el
d
ec
o
m
p
o
s
itio
n
s
h
av
e
s
h
o
w
n
g
o
o
d
clas
s
if
ica
tio
n
p
er
f
o
r
m
a
n
ce
f
o
r
I
S
AR
i
m
a
g
es
w
it
h
o
u
t
s
eg
m
e
n
tatio
n
(
T
ab
le
5
)
.
T
h
e
v
er
tical
d
etail
co
ef
f
icie
n
t
s
co
m
p
u
ted
at
th
ir
d
lev
el
an
d
s
ec
o
n
d
lev
el
d
ec
o
m
p
o
s
itio
n
h
a
v
e
clas
s
i
f
ied
th
e
I
S
A
R
i
m
a
g
es
w
it
h
o
u
t
s
eg
m
en
tatio
n
a
n
d
w
ith
s
eg
m
e
n
tat
io
n
r
esp
ec
tiv
el
y
u
p
to
th
e
lev
e
l
o
f
ac
ce
p
tan
ce
(
T
a
b
le
6
)
.
T
h
e
d
iag
o
n
al
d
etai
ls
co
m
p
u
ted
at
f
ir
s
t
a
n
d
t
h
ir
d
le
v
el
d
ec
o
m
p
o
s
it
io
n
h
av
e
cla
s
s
i
f
ied
t
h
e
I
S
A
R
i
m
ag
e
s
w
it
h
o
u
t
s
eg
m
e
n
tatio
n
s
atis
f
ac
to
r
il
y
(
T
ab
le
7
)
.
Fro
m
t
h
i
s
s
t
u
d
y
it
i
s
u
n
d
er
s
to
o
d
th
at
th
e
w
a
v
elet
co
ef
f
icie
n
t
s
h
a
v
e
s
h
o
w
n
b
ett
er
class
if
ica
tio
n
p
er
f
o
r
m
a
n
ce
f
o
r
th
e
s
h
ip
I
S
A
R
i
m
a
g
e
d
ata
s
et
w
it
h
o
u
t
s
eg
m
en
tatio
n
t
h
a
n
w
it
h
s
e
g
m
en
ta
t
io
n
.
T
h
e
s
tatis
tical
m
o
m
en
t
s
h
av
e
s
h
o
w
n
g
o
o
d
class
i
f
icatio
n
p
er
f
o
r
m
a
n
ce
f
o
r
th
e
co
n
s
id
er
ed
I
S
AR
i
m
a
g
e
d
ata
s
et
(
T
ab
le
8
)
.
Am
o
n
g
t
h
e
d
if
f
er
en
t
w
av
e
let
co
ef
f
icie
n
t
s
co
n
s
id
er
ed
,
th
e
h
o
r
izo
n
tal
co
ef
f
icie
n
ts
co
m
p
u
t
ed
at
lev
e
l
1
d
ec
o
m
p
o
s
itio
n
h
av
e
s
h
o
w
n
t
h
e
b
est
class
i
f
icatio
n
p
er
f
o
r
m
a
n
ce
.
W
e
ca
n
s
ee
th
a
t,
o
u
t
o
f
al
l
th
e
f
ea
tu
r
e
v
ec
to
r
s
co
n
s
id
er
ed
in
th
i
s
w
o
r
k
,
t
h
e
s
tatis
t
ical
m
o
m
e
n
ts
co
m
p
u
ted
f
r
o
m
t
h
e
co
m
b
i
n
ed
co
m
p
o
n
e
n
t
i
m
a
g
es
R
G,
G
B
an
d
R
B
is
th
e
m
o
s
t
s
u
i
tab
le
f
ea
t
u
r
e
v
ec
to
r
f
o
r
s
h
ip
I
S
A
R
i
m
ag
e
cla
s
s
i
f
icatio
n
s
in
ce
t
h
e
a
d
d
itio
n
al
ad
v
a
n
ta
g
e
w
i
th
th
is
f
ea
tu
r
e
v
ec
to
r
is
t
h
e
s
m
al
l
le
n
g
th
o
f
t
h
e
f
ea
t
u
r
e
v
e
cto
r
ir
r
esp
ec
tiv
e
o
f
t
h
e
i
m
a
g
e
s
ize.
T
h
e
f
ea
tu
r
e
v
ec
to
r
s
s
t
u
d
ied
in
t
h
i
s
p
ap
er
ca
n
b
e
u
s
ed
f
o
r
cla
s
s
i
f
icatio
n
o
f
i
m
ag
e
s
o
f
d
if
f
er
en
t
ap
p
licatio
n
s
.
T
h
e
p
er
f
o
r
m
a
n
ce
w
il
l
v
ar
y
w
h
en
th
e
d
ata
s
et
ch
an
g
es.
RE
F
E
R
E
NC
E
S
[1
]
I
.
Jo
rd
a
n
o
v
a
n
d
A
.
G
.
Nu
ff
ield
,
“
F
e
e
d
F
o
rw
a
rd
Ne
u
ra
l
Ne
t
w
o
rk
s
fo
r
A
u
to
m
a
ted
Clas
si
f
ic
a
ti
o
n
,
”
in
Pro
c
.
9
th
IE
EE
In
t.
C
o
n
f
.
o
n
C
o
g
n
it
ive
In
f
o
rm
a
ti
c
s (
ICCI’1
0
)
,
2
0
1
0
.
[2
]
R
.
S
.
Ch
o
ra
sIm
a
g
e
,
“
F
e
a
tu
re
Ex
trac
ti
o
n
T
e
c
h
n
iq
u
e
s
a
n
d
T
h
e
ir
Ap
p
li
c
a
ti
o
n
s
f
o
r
CBIR
a
n
d
Bi
o
m
e
tri
c
s
S
y
ste
m
s,
”
In
ter
n
a
t
io
n
a
l
jo
u
rn
a
l
o
f
b
i
o
l
o
g
y
a
n
d
b
io
me
d
ica
l
e
n
g
i
n
e
e
rin
g
,
vol
/i
s
su
e
:
1
(1
)
,
2
0
0
7
.
[3
]
A
.
Ch
a
d
h
a
,
e
t
a
l.
,
“
Co
m
p
a
ra
ti
v
e
S
tu
d
y
a
n
d
Op
ti
m
iza
ti
o
n
o
f
F
e
a
tu
re
-
Ex
trac
ti
o
n
T
e
c
h
n
iq
u
e
s
f
o
r
Co
n
ten
t
b
a
se
d
Im
a
g
e
Re
tri
e
v
a
l,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
Ap
p
li
c
a
ti
o
n
s
,
v
o
l
/i
ss
u
e
:
5
2
(2
0
),
p
p
.
0
9
7
5
–
8
8
8
7
,
2
0
1
2
.
[4
]
D
.
P.
T
ian
,
e
t
a
l.
,
“
A
Re
v
ie
w
o
n
Im
a
g
e
F
e
a
tu
re
Ex
t
ra
c
ti
o
n
a
n
d
Re
p
re
se
n
tatio
n
T
e
c
h
n
iq
u
e
s,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
M
u
lt
ime
d
ia
a
n
d
Ub
iq
u
it
o
u
s E
n
g
in
e
e
rin
g
,
v
o
l
/
issu
e
:
8
(
4
),
2
0
1
3
.
[5
]
S.
S
iv
a
p
e
ru
m
a
l
a
n
d
M
.
S
u
n
d
h
a
ra
ra
jan
,
“
A
d
v
a
n
c
e
f
e
a
tu
re
e
x
trac
ti
o
n
o
f
M
RI
Bra
in
Im
a
g
e
a
n
d
De
tec
ti
o
n
u
si
n
g
L
o
c
a
l
S
e
g
m
e
n
tatio
n
M
e
th
o
d
w
it
h
W
a
te
rsh
e
d
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
Res
e
a
rc
h
(
IJ
EE
ER
)
,
v
o
l
/i
ss
u
e
:
3
(
4
),
p
p
.
87
-
9
4
,
2
0
1
3
.
[6
]
P
.
Zh
a
n
g
,
e
t
a
l.
,
“
A
Ne
w
F
ra
m
e
w
o
rk
o
f
th
e
Un
su
p
e
rv
ise
d
Clas
sif
ica
ti
o
n
f
o
r
Hi
g
h
-
Re
so
lu
ti
o
n
R
e
m
o
te
S
e
n
sin
g
Im
a
g
e
,
”
T
EL
KOM
NIKA
,
v
ol
/i
ss
u
e
:
1
0
(
7
),
p
p
.
1
7
4
6
~
1
7
5
5
,
2
0
1
2
.
[7
]
K.
A
n
u
ra
d
h
a
a
n
d
K.
S
a
n
k
a
ra
n
a
ra
y
a
n
a
n
,
“
Co
m
p
a
riso
n
o
f
F
e
a
tu
re
Ex
trac
ti
o
n
T
e
c
h
n
iq
u
e
s
to
c
las
sif
y
Ora
l
Ca
n
c
e
r
s
u
sin
g
Im
a
g
e
P
r
o
c
e
ss
in
g
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Ap
p
li
c
a
t
io
n
o
r
I
n
n
o
v
a
ti
o
n
in
En
g
in
e
e
rin
g
&
M
a
n
a
g
e
me
n
t
(
IJ
AIE
M
)
,
v
o
l
/i
ss
u
e
:
2
(6
),
2
0
1
3
.
[8
]
Zah
e
e
ru
d
d
i
n
,
e
t
a
l.
,
“
De
tec
ti
o
n
a
n
d
S
h
a
p
e
F
e
a
tu
re
Ex
trac
ti
o
n
o
f
Bre
a
st
T
u
m
o
r
in
M
a
m
m
o
g
ra
m
s,”
in
Pro
c
.
o
f
W
CE
,
L
o
n
d
o
n
,
U.K.
,
2
0
1
2
.
[9
]
M
.
B.
T
a
y
e
l
a
n
d
M
.
E.
E
.
Bo
u
rid
y
,
“
EC
G
I
m
a
g
e
s
Clas
si
f
ic
a
ti
o
n
u
s
in
g
A
rti
f
icia
l
Ne
u
ra
l
Ne
t
w
o
rk
Ba
se
d
o
n
S
e
v
e
ra
l
F
e
a
tu
re
Ex
trac
ti
o
n
M
e
t
h
o
d
s,”
IE
EE
,
2
0
0
8
.
[1
0
]
A
.
P
h
in
y
o
m
a
rk
,
e
t
a
l.
,
“
F
e
a
tu
re
Ex
trac
ti
o
n
a
n
d
Re
d
u
c
ti
o
n
o
f
W
a
v
e
let
T
ra
n
s
f
o
r
m
Co
e
ff
icie
n
ts
f
o
r
EM
G
P
a
tt
e
rn
Clas
sif
ic
a
ti
o
n
,
”
El
e
c
tro
n
ics
a
n
d
El
e
c
trica
l
En
g
in
e
e
rin
g
,
v
o
l
/i
ss
u
e
:
6
(1
2
2
)
,
2
0
1
2
.
[1
1
]
Kh
e
d
e
r
G
.
,
e
t
a
l.
,
“
F
e
a
tu
re
e
x
tr
a
c
ti
o
n
b
y
w
a
v
e
let
tran
s
f
o
rm
s
to
a
n
a
l
y
z
e
th
e
h
e
a
rt
ra
te
v
a
riab
il
it
y
d
u
rin
g
tw
o
m
e
d
it
a
ti
o
n
tec
h
n
iq
u
e
,
”
in
6
th
W
S
EA
S
I
n
t.
Co
n
f.
o
n
Circ
u
it
s,
S
y
ste
ms
,
El
e
c
tro
n
ics
,
Co
n
tro
l
&
S
ig
n
a
l
Pro
c
e
ss
in
g
,
Ca
iro
,
Eg
y
p
t.
,
pp
.
3
7
4
,
2
0
0
7
.
[1
2
]
M
.
Na
z
a
rlo
o
,
e
t
a
l.
,
“
G
e
n
d
e
r
Clas
sif
i
c
a
ti
o
n
Us
in
g
H
y
b
rid
o
f
G
a
b
o
r
F
il
ters
a
n
d
Bi
n
a
ry
F
e
a
tu
re
s
o
f
a
n
I
m
a
g
e
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
ol
/i
ss
u
e
:
4
(4
)
,
p
p
.
5
3
9
~
5
4
7
,
2
0
1
4
.
[1
3
]
R.
C.
G
o
n
z
a
lez
,
e
t
a
l.
,
“
Co
l
o
r
Im
a
g
e
P
ro
c
e
ss
in
g
,
W
a
v
e
lets
a
n
d
Im
a
g
e
S
e
g
m
e
n
tatio
n
,
”
in
Di
g
it
a
l
Im
a
g
e
Pro
c
e
ss
in
g
Us
in
g
M
AT
L
AB
,
2
nd
e
d
.
,
Ne
w
De
l
h
i
,
T
a
ta M
c
G
ra
w
-
Hill
,
p
p
.
2
7
2
-
3
3
0
,
3
3
1
-
3
7
3
,
4
8
9
-
5
5
0
,
2
0
1
0
.
[1
4
]
R.
V
.
R
.
Ch
a
ry
,
e
t
a
l
.
,
“
F
e
a
t
u
re
Ex
trac
ti
o
n
M
e
th
o
d
s
f
o
r
C
o
lo
r
Im
a
g
e
S
im
il
a
rit
y
,
”
Ad
v
a
n
c
e
d
Co
mp
u
ti
n
g
:
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l,
v
o
l
/i
ss
u
e
:
3
(2
),
2
0
1
2
.
[1
5
]
S
.
M
a
h
a
jan
a
n
d
D
.
P
a
ti
l
,
“
Co
m
p
a
riso
n
o
f
Co
lo
r
a
n
d
C
o
lo
r
w
it
h
Ed
g
e
F
e
a
tu
re
Ex
trac
ti
o
n
Us
in
g
Co
n
tri
b
u
t
io
n
-
b
a
se
d
Clu
ste
rin
g
A
lg
o
rit
h
m
,
”
in
4
th
In
t.
Co
n
f.
o
n
Co
mm
u
n
ica
ti
o
n
S
y
ste
ms
a
n
d
Ne
two
rk
T
e
c
h
n
o
l
o
g
ies
,
2
0
1
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.