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
.
2
,
A
p
r
il
201
7
,
p
p
.
8
1
0
~
81
7
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v7
i
2
.
p
p
8
1
0
-
8
1
7
810
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
Para
m
e
tric Co
m
pa
riso
n of K
-
m
ea
ns a
nd Adap
tive
K
-
m
ea
ns
C
lustering
P
er
form
a
nce
on Di
ff
ere
nt
I
m
a
g
es
M
a
dh
us
m
it
a
Sa
hu
1
,
K.
P
a
rv
a
t
hi
2
,
M
.
Va
m
s
i K
rish
na
3
1
De
p
t.
o
f
ECE
,
CUT
M
,
P
a
ra
lak
h
m
u
n
d
i,
Od
is
h
a
,
In
d
ia
2
S
c
h
o
o
l
o
f
El
e
c
tro
n
ics
,
KIIT
Un
i
v
e
rsit
y
,
Bh
u
b
a
n
e
sw
a
r,
In
d
ia
3
De
p
t.
o
f
CS
E,
CUT
M
,
P
a
ra
lak
h
m
u
n
d
i,
Od
is
h
a
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
1
8
,
2
0
1
6
R
ev
i
s
ed
Feb
2
3
,
2
0
1
7
A
cc
ep
ted
Ma
r
7
,
2
0
1
7
Im
a
g
e
se
g
m
e
n
tatio
n
tak
e
s
a
m
a
j
o
r
ro
le
f
o
r
a
n
a
l
y
z
in
g
th
e
a
re
a
o
f
in
tere
st
in
im
a
g
e
p
ro
c
e
ss
in
g
.
M
a
n
y
r
e
se
a
r
c
h
e
rs
h
a
v
e
u
se
d
d
i
f
fe
re
n
t
t
y
p
e
s
o
f
t
e
c
h
n
iq
u
e
s
f
o
r
a
n
a
l
y
z
in
g
th
e
i
m
a
g
e
.
On
e
o
f
th
e
w
id
e
l
y
u
se
d
tec
h
n
iq
u
e
s
is
K
-
m
e
a
n
s
c
lu
ste
rin
g
.
In
th
is
p
a
p
e
r
w
e
u
se
tw
o
a
l
g
o
rit
h
m
s
K
-
mean
s
a
n
d
th
e
a
d
v
a
n
c
e
o
f
K
-
m
e
a
n
s
is
c
a
l
led
a
s
a
d
a
p
ti
v
e
K
-
m
e
a
n
s
c
lu
ste
rin
g
.
Bo
th
t
h
e
a
lg
o
rit
h
m
s
a
re
u
sin
g
in
d
if
f
e
re
n
t
t
y
p
e
s
o
f
i
m
a
g
e
a
n
d
g
o
t
a
su
c
c
e
ss
f
u
l
re
su
lt
.
By
c
o
m
p
a
rin
g
th
e
T
i
m
e
p
e
rio
d
,
P
S
NR
a
n
d
RM
S
E
v
a
lu
e
f
ro
m
th
e
re
su
lt
o
f
b
o
th
a
lg
o
rit
h
m
s
w
e
p
ro
v
e
th
a
t
th
e
A
d
a
p
ti
v
e
K
-
m
e
a
n
s
c
lu
ste
rin
g
a
lg
o
rit
h
m
g
i
v
e
s
a
b
e
st
re
su
lt
a
s co
m
p
a
rd
to
K
-
m
e
a
n
s clu
ste
rin
g
in
im
a
g
e
se
g
m
e
n
tatio
n
.
K
ey
w
o
r
d
:
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
I
m
ag
e
s
e
g
m
e
n
tatio
n
K
-
m
ea
n
s
clu
s
ter
i
n
g
MRI
co
lo
r
an
d
g
r
a
y
co
lo
r
im
a
g
e
P
SNR
&
R
MSE
Satellite i
m
a
g
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
m
t.
Ma
d
h
u
s
m
ita
Sa
h
u
,
Dep
t.
o
f
E
C
E
,
P
ar
alak
h
m
u
n
d
i,
Od
is
h
a
,
I
n
d
ia
.
E
m
ail:
m
ad
h
u
s
m
ita
s
7
9
@
g
m
a
i
l.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
No
w
a
d
a
y
s
ca
n
ce
r
i
s
a
co
m
m
o
n
d
i
s
ea
s
es
w
h
ich
s
p
r
ea
d
ed
all
o
v
er
w
o
r
ld
.
I
t
is
v
er
y
m
u
ch
es
s
en
tial
f
o
r
i
m
p
r
o
v
in
g
t
h
e
s
u
r
v
i
v
al
r
at
es
b
ec
au
s
e
it
is
o
n
e
o
f
t
h
e
m
o
s
t
d
em
a
n
d
i
n
g
a
n
d
d
elib
er
ated
to
p
ics
w
h
ic
h
ar
e
b
ein
g
d
i
s
cu
s
s
ed
a
m
o
n
g
t
h
e
r
esear
ch
s
c
ien
ti
s
t.
M
u
ltip
licat
io
n
o
f
a
s
et
o
f
ce
ll
s
in
a
p
ar
ticu
l
ar
ar
ea
o
f
o
u
r
b
o
d
y
an
d
th
e
g
r
o
w
t
h
o
f
th
e
ce
lls
ar
e
u
n
co
n
tr
o
llab
le,
th
i
s
is
o
n
e
ty
p
e
s
o
f
s
tr
u
ct
u
r
al
d
is
tr
o
r
tio
n
o
f
ce
lls
th
at
ar
e
m
en
tito
n
ed
as
t
u
m
o
r
.
I
t
is
v
er
y
i
m
p
o
r
tan
t
to
k
n
o
w
t
h
e
p
er
f
ec
t
s
p
r
ea
d
in
g
ar
ea
o
f
tu
m
o
r
ce
ll
f
o
r
cu
r
e
th
e
d
is
ea
s
e
s
.
[
1
]
I
n
i
m
ag
e
p
r
o
ce
s
s
i
n
g
s
o
m
e
tech
n
iq
u
es
ar
e
u
s
ed
to
d
ete
ct
th
e
t
u
m
o
r
ce
ll
a
m
o
n
g
t
h
at
i
m
a
g
e
s
eg
m
e
n
tatio
n
tak
e
s
a
m
aj
o
r
r
o
le
to
an
al
y
z
in
g
t
h
e
i
m
a
g
e
.
I
n
s
e
g
m
en
tatio
n
b
y
u
s
i
n
g
d
if
f
er
en
t
tec
h
n
iq
u
es
we
d
etec
t
th
e
tu
m
o
r
ce
ll
a
n
d
e
x
tr
ac
ted
th
e
ar
ea
f
o
r
a
n
al
y
zi
n
g
.
Ou
t
o
f
d
i
f
f
er
e
n
t
tec
h
n
iq
u
e
s
cl
u
s
ter
i
n
g
is
t
h
e
m
o
s
t
ef
f
icien
t
m
et
h
o
d
s
.
T
h
er
e
ar
e
d
if
f
er
e
n
t
t
y
p
es
o
f
cl
u
s
ter
in
g
K
-
m
ea
n
s
cl
u
s
ter
in
g
,
F
u
zz
y
C
-
m
ea
n
s
cl
u
s
ter
i
n
g
,
h
ier
ar
ch
al
cl
u
s
ter
i
n
g
an
d
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
in
g
.
Am
o
n
g
all
t
h
e
clu
s
ter
i
n
g
K
-
m
ea
n
s
an
d
A
d
ap
tiv
e
K
-
m
ea
n
s
clu
s
ter
in
g
m
eth
o
d
is
w
id
el
y
u
s
ed
in
i
m
a
g
e
s
e
g
m
en
ta
tio
n
.
T
h
ese
al
g
o
r
ith
m
s
ar
e
s
i
m
p
le
a
n
d
f
a
s
ter
th
an
o
th
er
cl
u
s
ter
i
n
g
.
B
o
th
ar
e
u
s
ed
f
o
r
w
o
r
k
i
n
a
lar
g
e
n
u
m
b
er
o
f
v
ar
iab
les.
I
n
ca
s
e
o
f
K
-
m
ea
n
s
,
i
n
i
tial
ize
th
e
n
u
m
b
er
o
f
K
tak
e
s
an
i
m
p
o
r
tan
t
tas
k
.
W
h
er
ea
s
in
A
d
ap
tiv
e
K
-
m
ea
n
s
it
i
s
n
o
t
r
eq
u
ir
ed
to
in
itialize
th
e
n
u
m
b
er
o
f
K,
b
ased
u
p
o
n
s
o
m
e
f
ea
t
u
r
es
a
n
d
ch
ar
ac
ter
is
tic
s
th
e
i
m
a
g
e
w
ill
au
to
m
at
ical
l
y
f
i
n
alize
t
h
e
n
u
m
b
er
o
f
K.
I
m
ag
e
s
e
g
m
en
tatio
n
ta
k
es
an
i
m
p
o
r
ta
n
t
r
o
le
to
an
aly
ze
t
h
e
ex
tr
ac
t
r
eg
io
n
i
n
m
ed
ical
i
m
a
g
es
o
r
MRI
(
Ma
g
n
etic
r
eso
n
a
n
ce
i
m
a
g
e)
i
n
m
ed
ical
ar
ea
.
So
m
ed
ical
i
m
ag
e
s
ar
e
s
eg
m
e
n
ted
an
d
f
o
r
f
u
r
th
er
a
n
al
y
ze
b
y
u
s
i
n
g
d
i
f
f
er
e
n
t
tec
h
n
iq
u
e
s
i
n
i
m
ag
e
s
e
g
m
en
tatio
n
.
Fo
r
an
al
y
zin
g
t
h
e
m
ed
ical
i
m
a
g
es
o
n
e
o
f
t
h
e
f
u
n
d
a
m
en
ta
l
p
r
o
b
lem
s
is
to
id
e
n
ti
f
y
t
h
e
b
o
u
n
d
ar
ies
o
f
a
n
o
b
j
ec
t
s
u
c
h
as
o
r
g
an
s
,
o
r
ab
n
o
r
m
a
l
r
eg
io
n
s
i
n
i
m
a
g
es.
Fro
m
t
h
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
P
a
r
a
metric Co
mp
a
r
is
o
n
o
f K
-
mea
n
s
a
n
d
A
d
a
p
tive
K
-
mea
n
s
C
lu
s
ter
in
g
P
erfo
r
ma
n
ce
…
(
M
a
d
h
u
s
mita
S
a
h
u
)
811
r
esu
lt
o
f
s
eg
m
e
n
tatio
n
it
is
p
o
s
s
ib
le
f
o
r
s
h
ap
e
an
a
l
y
s
is
,
d
ete
ctin
g
v
o
lu
m
e
c
h
a
n
g
e
a
n
d
m
a
k
i
n
g
ac
cu
r
ate
r
ad
i
atio
n
th
er
ap
y
tr
ea
t
m
en
t p
l
an
.
[
2
]
I
n
t
h
is
p
ap
er
,
b
y
u
s
i
n
g
b
o
th
t
h
e
al
g
o
r
ith
m
w
e
p
o
in
t
o
u
t
t
h
e
tu
m
o
r
s
p
r
ea
d
in
g
ar
ea
in
b
r
ain
M
R
g
r
a
y
an
d
co
lo
r
im
a
g
es
w
h
er
ea
s
i
n
t
h
e
Satelli
te
i
m
a
g
e
w
e
f
in
d
o
u
t
th
e
s
p
r
ea
d
in
g
ar
ea
o
f
la
k
e.
Af
ter
th
e
d
etec
tio
n
o
f
th
e
ch
a
n
g
es
f
r
o
m
th
e
s
e
g
m
e
n
ted
i
m
ag
e
s
w
e
h
a
v
e
co
m
p
ar
ed
s
o
m
e
p
ar
a
m
eter
s
b
ased
o
n
P
SNR
,
R
MSE
an
d
T
im
e
P
er
io
d
.
Af
ter
th
e
co
m
p
ar
i
s
o
n
w
e
co
n
clu
d
ed
th
at
A
d
ap
ti
v
e
K
-
m
e
an
s
cl
u
s
ter
i
n
g
g
i
v
es
t
h
e
b
est
r
esu
lt
as
co
m
p
ar
ed
to
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
f
o
r
th
e
ac
cu
r
ate
d
etec
tio
n
o
f
tu
m
o
r
ce
ll
in
MR
g
r
a
y
an
d
co
lo
r
im
a
g
es
an
d
v
o
lu
m
e
d
etec
tio
n
i
n
s
ate
llit
e
i
m
ag
e
s
.
So
b
y
th
is
w
a
y
w
e
ca
n
d
etec
t
t
h
e
ac
c
u
r
ate
d
etec
tio
n
o
f
tu
m
o
r
ce
ll
a
n
d
cu
r
e
th
e
d
is
ea
s
es t
h
r
o
u
g
h
r
ad
iatio
n
th
er
ap
y
tr
ea
t
m
e
n
t.
2.
K
-
M
E
ANS
CL
U
ST
E
RI
NG
K
-
m
ea
n
s
alg
o
r
it
h
m
is
u
s
ed
f
o
r
in
itializa
tio
n
o
f
p
ar
a
m
eter
s
s
i
n
ce
it
is
s
i
m
p
le
a
n
d
w
o
r
k
s
w
el
l
f
o
r
lar
g
e
d
ata
s
ets
w
h
e
n
co
m
p
ar
ed
w
ith
h
ier
ar
ch
ical
cl
u
s
ter
i
n
g
.
[
3
]
A
cc
o
r
d
in
g
to
s
o
m
e
s
i
m
ilar
it
y
m
ea
s
u
r
e
(
e.
g
.
E
u
c
lid
ea
n
)
t
h
e
d
ata
ar
e
d
iv
id
es
i
n
to
a
s
p
ec
if
ic
n
u
m
b
er
o
f
g
r
o
u
p
s
t
h
at
g
r
o
u
p
n
u
m
b
er
i
s
n
o
ted
as
„
K‟
v
ar
iab
le
w
h
er
e
K=
1
,
2
,
3
………….
n
.
it
i
s
a
u
s
er
d
ef
in
ed
n
u
m
b
er
.
T
h
e
g
r
o
u
p
o
f
s
a
m
e
d
ata
is
ca
ll
ed
as c
lu
s
ter
i
n
g
a
n
d
th
e
m
et
h
o
d
is
ca
lled
as K
-
m
ea
n
s
clu
s
ter
i
n
g
.
T
h
is
alg
o
r
it
h
m
i
s
co
n
s
is
t
in
g
o
f
t
w
o
s
ep
ar
ate
p
h
a
s
es.
At
f
ir
s
t
i
t c
alcu
la
tes
th
e
K
ce
n
tr
o
id
an
d
s
ec
o
n
d
l
y
it
tak
es
ea
ch
p
o
in
t
to
th
e
cl
u
s
ter
w
h
ic
h
h
as
n
ea
r
est
ce
n
tr
o
id
f
r
o
m
t
h
e
ad
j
ac
en
t
d
ata
p
o
in
t.
T
o
f
in
d
o
u
t
th
e
d
is
tan
ce
o
f
n
ea
r
est
ce
n
tr
o
id
it
w
il
l
u
s
ed
t
h
e
E
u
clid
ea
n
d
is
ta
n
ce
m
et
h
o
d
.
A
f
ter
th
e
g
r
o
u
p
i
n
g
o
f
t
h
e
d
ata
i
t
f
in
d
s
o
u
t
ce
n
tr
o
id
o
f
ea
ch
clu
s
ter
.
T
h
e
E
u
clid
ea
n
d
is
tan
ce
is
ca
l
cu
lated
a
m
o
n
g
ea
ch
ce
n
ter
an
d
ea
c
h
d
ata
p
o
in
t
ass
i
g
n
s
i
n
th
e
c
lu
s
ter
w
h
ich
h
av
e
m
i
n
i
m
u
m
E
u
clid
ea
n
d
is
tan
ce
,
af
ter
r
ea
s
s
i
g
n
i
n
g
t
h
e
p
o
in
ts
ea
c
h
ti
m
e
t
h
e
ce
n
tr
o
id
o
f
th
e
c
lu
s
ter
s
ar
e
r
ec
alcu
lated
.
T
h
is
p
r
o
ce
s
s
is
co
n
t
in
u
ed
t
ill
t
h
e
en
d
o
f
th
e
r
est
p
o
in
ts
t
h
e
n
f
in
al
l
y
i
t
ca
lcu
lated
th
e
f
i
n
al
ce
n
tr
o
id
o
r
m
ea
n
o
f
th
e
cl
u
s
ter
s
.
So
K
-
m
ea
n
s
i
s
an
i
ter
ativ
e
a
l
g
o
r
ith
m
i
n
w
h
ic
h
it
m
ea
n
s
t
h
e
s
u
m
o
f
d
is
ta
n
ce
f
r
o
m
ea
ch
o
b
j
ec
t
to
its
clu
s
ter
ce
n
tr
o
id
,
o
v
er
all
clu
s
t
er
s
.
[
2
,
4
]
L
et
u
s
co
n
s
id
er
an
i
m
ag
e
m
×
n
an
d
t
h
e
i
m
a
g
e
h
as
to
b
e
clu
s
ter
in
to
K
n
u
m
b
er
o
f
cl
u
s
ter
.
Su
p
p
o
s
e
a
p
o
in
t
f
(
x
,
y
)
w
h
ic
h
i
s
an
in
p
u
t
p
ix
els
to
b
e
clu
s
ter
an
d
C
k
b
e
th
e
cl
u
s
ter
ce
n
ter
.
T
h
e
alg
o
r
ith
m
o
f
K
-
m
ea
n
s
clu
s
ter
i
n
g
is
s
h
o
w
n
a
s
f
o
llo
w
s
:
a.
First,
in
i
tialize
t
h
e
n
u
m
b
er
o
f
clu
s
ter
K
a
n
d
ca
lcu
late
t
h
eir
c
en
ter
C
k
b
y
u
s
i
n
g
th
e
f
o
r
m
u
la
b.
C
alcu
late
th
e
E
u
c
lid
ea
n
d
is
ta
n
ce
E
d
b
et
w
ee
n
t
h
e
ce
n
ter
a
n
d
ea
ch
p
ix
el
o
f
an
i
m
ag
e
u
s
i
n
g
t
h
e
r
elatio
n
g
iv
e
n
b
elo
w
.
E
d
=
|
|
f
(
x
,
y
)
-
C
k
||
(
1
)
c.
B
ased
u
p
o
n
th
e
E
u
cl
id
ea
n
d
is
t
an
ce
E
d
ass
i
g
n
s
th
e
r
es
t p
ix
el
s
to
th
e
n
ea
r
est cl
u
s
ter
.
d.
Af
ter
as
s
i
g
n
in
g
all
th
e
p
ix
el
s
i
n
t
h
e
ap
p
r
o
p
r
iate
clu
s
ter
t
h
e
v
alu
e
o
f
th
e
clu
s
ter
ce
n
ter
s
h
o
u
ld
b
e
ch
an
g
ed
an
d
it is
ca
lc
u
lated
b
y
u
s
in
g
t
h
e
r
elatio
n
g
i
v
en
b
elo
w
:
∑
∑
(
2
)
e.
R
eiter
ate
th
e
p
r
o
ce
s
s
til
ll it sat
is
f
ie
s
t
h
e
to
ler
an
ce
o
r
er
r
o
r
v
al
u
e.
f.
Up
d
ate
th
e
clu
s
ter
m
ea
n
i.e
.
ca
lcu
late
t
h
e
m
ea
n
o
f
t
h
e
o
b
j
ec
t
f
o
r
ea
ch
clu
s
ter
.
3.
ADAP
T
I
VE
K
-
M
E
ANS
CL
UST
E
RI
NG
I
n
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
in
g
a
lg
o
r
it
h
m
th
e
n
u
m
b
er
o
f
K
is
s
elec
ted
f
r
o
m
d
ata
s
et
an
d
it
i
s
r
an
d
o
m
l
y
s
elec
ted
b
ased
u
p
o
n
s
o
m
e
ch
ar
ac
ter
i
s
tics
a
n
d
f
ea
t
u
r
es.
T
h
e
p
r
o
p
er
ties
o
f
ea
ch
ele
m
en
t a
ls
o
f
o
r
m
th
e
p
r
o
p
er
ties
o
f
th
e
clu
s
ter
t
h
at
is
co
n
s
is
ti
n
g
b
y
t
h
e
el
e
m
e
n
t.
L
i
k
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
t
h
is
alg
o
r
ith
m
i
s
al
s
o
co
m
p
u
te
th
e
d
is
tan
ce
b
et
w
ee
n
a
g
i
v
e
n
ele
m
en
t
an
d
a
clu
s
ter
.
So
t
h
e
f
u
n
ct
io
n
o
f
t
h
e
alg
o
r
it
h
m
i
s
b
ased
u
p
o
n
to
c
o
m
p
u
te
th
e
d
is
ta
n
ce
b
et
w
ee
n
t
w
o
e
le
m
e
n
t
s
.
T
h
is
f
u
n
ctio
n
h
as a
n
i
m
p
o
r
tan
t c
o
n
s
id
er
atio
n
i.e
.
it
s
h
o
u
ld
b
e
ab
le
to
co
u
n
t
f
o
r
th
e
d
is
ta
n
ce
b
ase
d
o
n
p
r
o
p
er
ties
th
at
h
av
e
b
ee
n
n
o
r
m
alize
d
s
o
th
a
t
d
is
tan
ce
is
n
o
t
d
o
m
in
a
ted
b
y
o
n
e
p
r
o
p
er
ty
o
r
s
o
m
e
o
f
th
e
p
r
o
p
er
ty
is
n
o
t
ig
n
o
r
ed
in
t
h
e
co
m
p
u
tatio
n
o
f
d
is
tan
ce
.
I
t u
s
es E
u
c
lid
ea
n
d
is
tan
ce
f
o
r
m
u
la
f
o
r
c
o
m
p
u
te
t
h
e
d
is
tan
ce
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
2
,
A
p
r
il 2
0
1
7
:
8
1
0
–
81
7
812
Fo
r
ex
a
m
p
le:
w
h
en
s
p
ec
tr
a
l
d
ata
s
p
ec
if
ied
b
y
n
-
d
i
m
e
n
s
io
n
s
,
t
h
e
d
is
tan
ce
b
et
w
ee
n
t
w
o
d
ata
ele
m
e
n
ts
.
D
1
={D
11
,D
12
,
……
………D
1n
}a
n
d
D
2
={D
21
,D
22
,
…………
D
2n
}
T
h
e
E
u
clid
ea
n
d
is
ta
n
ce
is
g
iv
e
n
b
y
E
.
D=
√
(
3
)
B
ased
u
p
o
n
th
e
d
is
ta
n
ce
th
e
a
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
al
g
o
r
ith
m
is
a
s
f
o
llo
w
s
:
A
t
f
ir
s
t
f
in
d
o
u
t
th
e
d
is
ta
n
ce
o
f
ea
ch
clu
s
ter
f
r
o
m
ev
er
y
o
t
h
e
r
clu
s
ter
an
d
it
is
s
to
r
ed
in
a
2
D
ar
r
ay
as
a
tr
ian
g
u
lar
m
atr
ix
.
T
h
e
m
in
i
m
u
m
d
i
s
tan
ce
D
min
b
et
w
ee
n
a
n
y
t
w
o
cl
u
s
t
er
s
C
1
a
n
d
C
2
was
n
o
ted
.
C
o
m
p
u
te
th
e
E
.
D
d
is
ta
n
ce
f
r
o
m
ea
c
h
cl
u
s
ter
to
ea
c
h
u
n
cl
u
s
ter
ed
ele
m
en
t.
I
n
t
h
is
ca
s
e
t
h
er
e
ar
e
th
r
ee
p
o
s
s
ib
ilit
ies
f
o
r
ass
i
g
n
m
e
n
t o
f
t
h
e
r
est ele
m
e
n
t
to
th
e
s
u
i
tab
le
clu
s
ter
th
at
ar
e
lik
e
b
elo
w
f
o
u
r
f
o
r
m
s
.
a.
W
h
en
th
e
d
is
ta
n
ce
i
s
ze
r
o
f
r
o
m
a
ele
m
en
t
to
a
cl
u
s
ter
t
h
en
th
at
ele
m
e
n
t
is
a
s
s
i
g
n
ed
i
n
t
h
at
clu
s
ter
,
t
h
en
p
r
o
ce
ed
f
o
r
n
ex
t e
le
m
e
n
t.
b.
T
h
e
n
ex
t
ele
m
en
t
is
a
llo
ca
ted
to
its
clo
s
e
s
t
cl
u
s
ter
if
t
h
e
d
is
t
an
ce
o
f
a
ele
m
en
t
f
r
o
m
a
cl
u
s
ter
is
le
s
s
t
h
a
n
D
m
in
.
Af
ter
r
ea
s
s
i
g
n
ed
t
h
e
ele
m
en
t
i
n
t
h
e
clu
s
ter
t
h
e
ce
n
tr
o
id
o
f
th
e
c
lu
s
ter
is
c
h
a
n
g
ed
.
S
o
ac
co
r
d
in
g
to
th
e
p
r
o
p
er
ties
o
f
th
e
ele
m
e
n
t
s
th
e
ce
n
tr
o
id
is
r
ec
o
m
p
u
ted
.
As
w
ell
as
w
e
r
ec
o
m
p
u
te
t
h
e
d
is
tan
ce
o
f
t
h
e
af
f
ec
ted
clu
s
ter
f
r
o
m
ev
er
y
o
t
h
er
clu
s
ter
,
also
th
e
m
i
n
i
m
u
m
d
is
ta
n
ce
b
et
w
ee
n
an
y
t
w
o
clu
s
ter
s
an
d
th
e
t
w
o
cl
u
s
ter
s
t
h
e
y
ar
e
clo
s
est to
ea
ch
o
th
er
.
c.
Fin
all
y
,
it
is
co
m
p
u
ted
w
h
e
n
th
e
d
is
ta
n
ce
D
m
in
i
s
less
t
h
an
th
e
d
is
ta
n
ce
o
f
th
e
ele
m
en
t
f
r
o
m
t
h
e
n
ea
r
es
t
clu
s
ter
.
d.
W
e
tak
e
t
w
o
clo
s
e
s
t
cl
u
s
ter
s
C
1
an
d
C
2
,
a
n
d
t
h
en
m
er
g
e
C
2
i
n
to
C
1
.
No
w
w
e
g
e
t
o
n
l
y
o
n
e
cl
u
s
ter
C
1
an
d
w
e
d
estro
y
t
h
e
ele
m
e
n
ts
a
s
w
ell
a
s
th
e
r
ep
r
esen
tatio
n
o
f
C
2
,
s
o
C
2
ca
lled
as
e
m
p
t
y
cl
u
s
ter
.
W
e
g
et
a
n
e
w
clu
s
ter
s
u
cc
e
s
s
f
u
ll
y
.
Ag
ain
t
h
e
d
is
ta
n
ce
s
b
et
w
ee
n
al
l
clu
s
ter
s
ar
e
r
ec
o
m
p
u
ted
an
d
t
w
o
clo
s
est
clu
s
ter
s
r
ec
o
g
n
ized
ag
a
in
.
T
ill
all
th
e
ele
m
e
n
ts
h
as
b
ee
n
clu
s
ter
ed
th
e
ab
o
v
e
th
r
ee
s
t
ep
s
ar
e
to
b
e
r
ep
ea
ted
.
B
y
th
is
w
a
y
w
e
s
h
o
u
ld
clu
s
ter
th
e
ele
m
en
ts
w
i
th
o
u
t d
e
f
in
in
g
t
h
e
n
u
m
b
er
o
f
k
in
ad
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
in
g
.
[
5
]
4.
P
RO
P
O
SE
D
AL
G
O
R
I
T
H
M
I
n
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
w
e
a
p
p
lied
b
o
th
th
e
tech
n
iq
u
es
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
a
n
d
A
d
ap
ti
v
e
K
-
m
ea
n
s
clu
s
ter
i
n
g
to
tr
ac
k
th
e
tu
m
o
r
o
b
j
ec
ts
in
MRI
b
r
ain
g
r
a
y
co
l
o
r
im
a
g
e,
co
lo
r
im
a
g
e
an
d
als
o
th
ese
tech
n
iq
u
e
s
ar
e
ap
p
lied
to
f
in
d
o
u
t t
h
e
lak
e
r
eg
io
n
in
s
atellite
i
m
a
g
e.
T
h
e
b
lo
ck
d
iag
r
a
m
o
f
t
h
e
p
r
o
p
o
s
ed
m
e
th
o
d
f
o
r
K
-
m
ea
n
s
a
n
d
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
in
g
r
a
y
co
lo
r
an
d
co
lo
r
im
a
g
e
is
:
Fig
u
r
e
1
.
B
lo
ck
Diag
r
a
m
o
f
p
r
o
p
o
s
ed
m
eth
o
d
K
-
m
e
a
n
s a
l
g
o
r
i
t
h
m
A
d
a
p
t
i
v
e
K
-
m
e
a
n
s a
l
g
o
r
i
t
h
m
S
e
l
e
c
t
i
o
n
o
f
c
l
u
st
e
r
S
e
l
e
c
t
i
o
n
o
f
c
l
u
st
e
r
P
a
r
a
me
t
e
r
s c
o
mp
a
r
i
n
g
F
i
n
d
o
u
t
t
h
e
b
e
st
r
e
su
l
t
(
o
/
p
)
i
mag
e
L
u
mi
n
o
si
t
y
(
L
)
C
h
r
o
ma
t
i
c
i
t
y
(
a
*
b
)
C
o
n
v
e
r
t
L
*
a
*
b
i
m
a
g
e
I
n
p
u
t
i
m
a
g
e
R
G
B
c
o
l
o
r
i
mag
e
(
i
f
i
/
p
g
r
a
y
c
o
l
o
r
i
mag
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
P
a
r
a
metric Co
mp
a
r
is
o
n
o
f K
-
mea
n
s
a
n
d
A
d
a
p
tive
K
-
mea
n
s
C
lu
s
ter
in
g
P
erfo
r
ma
n
ce
…
(
M
a
d
h
u
s
mita
S
a
h
u
)
813
T
h
e
alg
o
r
ith
m
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
is
g
i
v
en
b
elo
w
f
r
o
m
s
tep
1
-
s
tep
1
4
:
s
t
ep
1
R
ea
d
th
e
i
m
a
g
es
w
h
ich
ar
e
i
n
J
P
E
G
f
o
r
m
at,
t
h
at
ar
e
co
lle
cted
f
r
o
m
t
h
e
m
ed
ical
b
r
ain
i
m
ag
e
s
a
n
d
s
atellite
i
m
a
g
es i
n
th
e
g
o
g
g
le
s
ite.
s
t
ep
2
R
esize
t
h
e
i
m
ag
e
s
t
ep
3
C
o
n
v
er
t th
e
i
m
a
g
e
in
to
R
GB
co
lo
r
s
p
ac
e
im
a
g
e,
th
i
s
s
tep
is
o
n
l
y
ap
p
licab
le
f
o
r
g
r
a
y
co
lo
r
im
ag
e
s
.
s
t
ep
4
T
h
en
af
ter
co
n
v
er
t
t
h
e
i
m
a
g
e
i
n
to
L
*
a
*
b
co
lo
r
s
p
ac
e
i
m
a
g
e.
W
h
en
w
e
ig
n
o
r
in
g
t
h
e
b
r
ig
h
t
n
ess
i
n
t
h
e
i
m
a
g
es
.
W
e
ca
n
ea
s
il
y
v
is
u
all
y
d
is
tin
g
u
is
h
th
e
co
lo
r
s
f
r
o
m
t
h
e
i
m
a
g
e.
T
h
e
L
*
a
*
b
*
co
lo
r
s
p
ac
e
(
also
w
ell
-
k
n
o
w
n
a
s
C
I
E
L
A
B
o
r
C
I
E
L
*
a
*
b
*
)
e
n
ab
les
u
s
to
q
u
a
n
t
if
y
t
h
ese
v
is
u
al
d
i
f
f
er
en
ce
s
.
T
h
e
L
*
a
*
b
*
co
lo
r
s
p
ac
e
b
ased
o
n
lu
m
i
n
o
cit
y
la
y
er
a
n
d
C
h
r
o
m
atici
t
y
l
a
y
er
,
w
h
er
e
t
h
e
l
u
m
i
n
o
s
it
y
la
y
er
is
'L
*
'
w
h
er
e
as
ch
r
o
m
aticit
y
-
la
y
er
'
a
*
'
r
ep
r
esen
t
in
g
co
lo
r
f
alls
alo
n
g
t
h
e
r
ed
-
g
r
ee
n
a
x
i
s
a
n
d
ch
r
o
m
a
ticit
y
-
la
y
er
'
b
*
'
r
ep
r
esen
ti
n
g
t
h
e
co
lo
r
f
alls
alo
n
g
t
h
e
b
lu
e
-
y
ello
w
ax
is
.
A
l
l
o
f
t
h
e
co
lo
r
r
elate
d
in
f
o
r
m
atio
n
is
in
t
h
e
'
a
*
'
an
d
'
b
*
'
la
y
er
s
.
Usi
n
g
t
h
e
E
u
c
lid
ea
n
d
is
ta
n
ce
m
etr
ic
w
e
ca
n
m
ea
s
u
r
e
t
h
e
d
if
f
er
e
n
ce
b
et
w
ee
n
t
w
o
co
lo
r
s
.
C
o
n
v
er
t t
h
e
p
r
ev
io
u
s
i
m
a
g
e
i
n
to
L
*
a
*
b
*
co
lo
r
s
p
ac
e.
[
6
]
s
t
ep
5
B
y
U
s
i
n
g
K
-
Me
a
n
s
C
lu
s
ter
i
n
g
class
i
f
y
t
h
e
C
o
lo
u
r
in
'
a
*
b
*
'
S
p
ac
e.
A
g
r
o
u
p
o
f
o
b
j
ec
ts
ar
e
s
ep
ar
ated
b
y
u
s
i
n
g
cl
u
s
ter
i
n
g
m
eth
o
d
.
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
tr
ea
ts
ea
ch
ele
m
en
t
as
h
av
i
n
g
a
lo
ca
tio
n
in
s
p
ac
e.
T
h
e
p
ar
titi
o
n
s
w
h
er
e
th
e
o
b
j
ec
ts
in
s
id
e
ea
ch
clu
s
ter
w
er
e
clo
s
er
to
ea
ch
o
t
h
er
as
p
o
s
s
ib
le,
a
n
d
as
d
is
tan
ce
f
r
o
m
t
h
e
o
b
j
ec
ts
i
n
o
t
h
er
cl
u
s
ter
s
as
p
o
s
s
ib
le
w
er
e
s
e
lecte
d
b
y
it.
I
n
K
-
m
ea
n
s
clu
s
ter
in
g
w
e
h
av
e
to
d
ef
i
n
e
th
e
n
u
m
b
er
o
f
clu
s
ter
s
to
b
e
p
a
r
titi
o
n
ed
an
d
a
d
is
tan
ce
m
etr
ic
to
co
m
p
u
te
h
o
w
clo
s
er
t
w
o
o
b
j
ec
ts
ea
ch
o
th
er
.
A
s
th
e
co
lo
r
in
f
o
r
m
atio
n
b
ased
in
th
e
'
a
*
b
*
'
s
p
ac
e,
y
o
u
r
o
b
j
ec
ts
ar
e
p
ix
els
b
y
m
ea
n
s
o
f
'
a
*
'
a
n
d
'
b
*
'
v
al
u
es.
B
y
u
s
i
n
g
K
-
m
ea
n
s
cl
u
s
ter
th
e
o
b
j
ec
ts
in
to
K
clu
s
ter
s
u
s
i
n
g
th
e
E
u
cli
d
ea
n
d
is
tan
ce
m
etr
ic.
[
7
,
8]
s
t
ep
6
L
ab
el
E
v
er
y
P
i
x
el
in
t
h
e
I
m
a
g
e
Usi
n
g
t
h
e
R
es
u
lt
s
f
r
o
m
K
-
M
ea
n
s
.
W
ith
o
u
t
i
n
d
ex
in
p
u
t
K
-
m
ea
n
s
r
etu
r
n
s
to
a
co
r
r
esp
o
n
d
in
g
c
l
u
s
ter
f
o
r
ev
er
y
o
b
j
ec
t.
L
ab
el
ev
er
y
p
i
x
el
in
th
e
i
m
a
g
e
w
it
h
it
s
clu
s
ter
in
d
ex
.
s
t
ep
7
A
cc
o
r
d
in
g
to
th
e
n
u
m
b
er
o
f
K
w
h
ic
h
d
ef
i
n
ed
b
y
t
h
e
u
s
er
b
ased
o
n
t
h
at
i
m
ag
e
we
w
il
l
g
et
s
eg
m
e
n
ted
i
m
a
g
es o
r
clu
s
ter
e
d
im
a
g
e
s
.
s
t
ep
8
Select
o
n
e
clu
s
ter
ed
i
m
a
g
e.
s
t
ep
9
I
n
th
e
s
a
m
e
w
a
y
,
C
la
s
s
i
f
y
th
e
C
o
lo
r
in
'
a
*
b
*
'
Sp
ac
e
Usi
n
g
Ad
ap
tiv
e
K
-
Me
a
n
s
C
l
u
s
ter
i
n
g
.
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
d
o
es
n
‟
t n
e
ed
s
p
ec
if
y
i
n
g
t
h
e
n
u
m
b
er
o
f
c
l
u
s
ter
s
to
b
e
p
ar
titi
o
n
ed
.
I
t a
u
to
m
atica
ll
y
f
i
n
d
s
o
u
t
th
e
n
u
m
b
er
o
f
clu
s
te
r
b
ased
u
p
o
n
s
o
m
e
p
r
o
p
er
ties
o
r
f
ea
tu
r
es.
Sin
ce
t
h
e
co
lo
r
in
f
o
r
m
a
tio
n
ex
is
t
s
i
n
t
h
e
'
a
*
b
*
'
s
p
ac
e
a
n
d
th
e
o
b
j
ec
ts
ar
e
p
ix
els
w
it
h
'
a
*
'
an
d
'
b
*
'
v
a
lu
e
s
.
Use
A
d
ap
ti
v
e
K
-
m
ea
n
s
to
clu
s
ter
th
e
o
b
j
ec
ts
in
to
K
cl
u
s
ter
s
u
s
in
g
t
h
e
E
u
clid
ea
n
d
is
t
an
ce
m
etr
ic.
s
t
ep
1
0
I
n
th
e
I
m
a
g
e
l
ab
el
E
v
er
y
P
ix
el
b
y
u
s
i
n
g
th
e
R
es
u
lt
s
f
r
o
m
A
d
ap
tiv
e
K
-
Me
an
s
clu
s
ter
in
g
.
W
ith
o
u
r
i
n
d
ex
i
n
p
u
t,
A
d
ap
ti
v
e
K
-
m
ea
n
s
cl
u
s
ter
ed
ev
er
y
o
b
ject.
L
ab
el
ev
er
y
p
ix
el
in
t
h
e
i
m
ag
e
w
it
h
its
clu
s
ter
in
d
e
x
.
s
t
ep
1
1
As
p
er
th
e
n
u
m
b
er
o
f
o
b
j
ec
ts
in
th
at
i
m
a
g
e
w
e
w
ill
g
e
t
th
e
s
eg
m
e
n
ted
i
m
ag
e
s
(
As
p
er
th
e
n
u
m
b
er
o
f
ce
n
tr
o
id
)
.
s
t
ep
1
2
Select
o
n
e
clu
s
ter
ed
i
m
a
g
e
o
r
s
eg
m
e
n
ted
i
m
a
g
e.
s
t
ep
1
3
C
o
m
p
ar
e
th
e
p
ar
a
m
eter
s
(
T
I
ME
,
P
SNR
,
an
d
R
MSE
)
o
f
th
e
o
r
ig
i
n
al
i
m
ag
e
w
i
th
t
h
e
s
eg
m
e
n
ted
i
m
a
g
e,
f
i
n
d
o
u
t t
h
e
b
est r
esu
l
t.
s
t
ep
1
4
Fin
all
y
,
f
r
o
m
t
h
e
s
e
g
m
e
n
ted
i
m
ag
e
w
e
ca
n
d
etec
t
th
e
ap
p
r
o
p
r
iate
tu
m
o
u
r
s
p
r
ea
d
in
g
ar
e
a
b
y
u
s
in
g
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
in
ca
s
e
o
f
MRI
i
m
a
g
es
an
d
f
r
o
m
t
h
e
s
a
tellite
i
m
a
g
e
w
e
f
in
d
o
u
t
t
h
e
ar
ea
co
v
er
ed
b
y
th
e
la
k
e.
5.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
I
n
th
e
p
r
o
p
o
s
ed
m
e
th
o
d
,
w
e
co
n
v
er
t
a
g
r
a
y
-
lev
e
l
MR
b
r
ai
n
i
m
ag
e
i
n
to
an
R
GB
co
lo
r
im
ag
e
b
u
t
w
h
e
n
w
e
u
s
e
co
lo
r
MRI
an
d
Satellite
i
m
a
g
e
w
e
h
a
v
e
n
o
t
co
n
v
er
ted
th
e
i
n
p
u
t
i
m
a
g
e
i
n
to
R
GB
co
lo
r
s
p
ac
e,
th
en
t
h
e
i
m
ag
e
co
n
v
er
ted
in
to
C
I
E
L
ab
co
lo
r
m
o
d
el.
C
o
lo
r
s
in
b
o
th
t
h
e
a
*
an
d
b
*
s
p
ac
es
ar
e
f
ea
tu
r
e
v
ec
to
r
s
f
o
r
K
-
m
ea
n
s
clu
s
ter
i
n
g
a
n
d
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
.
T
o
ex
p
r
ess
th
e
d
etec
tio
n
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
,
w
e
ta
k
e
a
MR
b
r
ain
i
m
ag
e
o
f
g
r
a
y
co
lo
r
an
d
co
lo
r
im
a
g
e
co
n
tai
n
i
n
g
t
h
e
p
ath
o
lo
g
ical
ch
an
g
e
ar
ea
s
h
o
w
n
in
Fi
g
u
r
e
2
(
a)
&
Fi
g
u
r
e
3
(
a)
s
er
v
ed
a
s
a
o
r
ig
in
a
l
i
m
a
g
e,
h
er
e
also
w
e
t
ak
e
a
s
atelli
te
co
lo
r
i
m
a
g
e
f
o
r
d
etec
tio
n
o
f
g
eo
g
r
ap
h
ical
ar
ea
co
v
er
ed
b
y
th
e
w
a
ter
s
h
o
w
n
in
Fi
g
u
r
e
4
(
a)
.
Gen
er
all
y
,
a
n
M
R
b
r
ai
n
i
m
a
g
e
co
n
s
is
t
s
o
f
r
eg
io
n
s
th
at
r
ep
r
esen
t
th
e
b
o
n
e
,
s
o
f
t
ti
s
s
u
e,
af
f
ec
ted
ca
n
ce
r
ce
ll
ar
ea
,
f
at
an
d
b
ac
k
g
r
o
u
n
d
[
6
]
.
B
ased
u
p
o
n
s
o
m
e
v
i
s
u
a
l
j
u
d
g
e
m
en
t
a
n
d
s
o
m
e
f
ea
tu
r
e
s
th
e
K
-
m
ea
n
s
a
n
d
A
d
ap
tiv
e
K
-
m
ea
n
s
clu
s
ter
i
n
g
al
g
o
r
ith
m
cl
u
s
ter
ed
th
e
g
r
a
y
co
lo
r
an
d
co
lo
r
b
r
ain
i
m
a
g
e
an
d
t
h
e
s
e
lecte
d
clu
s
ter
i
m
a
g
es
ar
e
s
h
o
wn
in
Fi
g
u
r
e
2
(
b
)
,
2
(
c
)
&
Fig
u
r
e
3
(
b
)
,
3
(
c
)
.
Af
ter
s
elec
t
t
h
e
c
lu
s
ter
w
e
ca
lcu
late
th
e
ti
m
e
p
er
io
d
,
R
MS
E
v
al
u
e
a
n
d
P
SNR
v
al
u
e
a
n
d
th
e
n
we
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
2
,
A
p
r
il 2
0
1
7
:
8
1
0
–
81
7
814
co
m
p
ar
i
n
g
t
h
e
r
es
u
lts
o
f
th
e
e
n
tire
i
m
ag
e.
W
h
ic
h
r
es
u
lts
ar
e
s
h
o
w
n
in
p
ar
a
m
eter
s
co
m
p
ar
i
n
g
tab
le.
Fig
u
r
e
2
. (
a
)
Or
ig
in
al
i
m
ag
e
(
b
)
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
(
c
)
A
d
a
p
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
in
g
Her
e
w
e
d
i
s
cu
s
s
th
e
M
R
I
b
r
ain
co
lo
r
i
m
a
g
e
a
n
d
Satelli
te
i
m
a
g
e.
W
h
e
n
w
e
ap
p
lied
b
o
th
th
e
alg
o
r
ith
m
i
n
th
o
s
e
i
m
a
g
e
s
i
n
c
o
lo
r
b
r
ain
MRI
i
m
ag
e
w
e
tr
ac
k
t
h
e
t
u
m
o
r
ce
l
l a
r
ea
clea
r
l
y
b
y
u
s
i
n
g
A
d
ap
tiv
e
k
-
m
ea
n
s
cl
u
s
ter
i
n
g
is
co
m
p
ar
ed
to
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
.
I
n
Ot
h
er
ca
s
e
w
h
e
n
b
o
th
al
g
o
r
it
h
m
s
ar
e
i
m
p
le
m
e
n
ted
i
n
Satellite
i
m
a
g
e
th
e
s
e
g
m
en
tat
io
n
o
f
lak
e
f
r
o
m
t
h
e
s
atelli
te
i
m
ag
e
i
s
clea
r
l
y
i
s
o
lated
in
A
d
ap
ti
v
e
K
-
m
ea
n
s
clu
s
ter
i
n
g
alg
o
r
it
h
m
w
h
ic
h
w
e
ca
n
v
i
s
u
al
ize
in
t
h
e
clu
s
ter
i
m
ag
e
4
(
c
)
.
(
a)
(
b
)
(
c
)
Fig
u
r
e
3.
(
a
)
Or
ig
in
al
i
m
ag
e
(
b
)
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
(
c
)
A
d
a
p
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
in
g
(
a)
(
b
)
(
c
)
Fig
u
r
e
4
.
(
a)
Or
ig
in
al
i
m
ag
e
(
b
)
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
(
c
)
A
d
a
p
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
in
g
Fro
m
t
h
e
Fi
g
u
r
e
2
,
Fig
u
r
e
3
an
d
Fi
g
u
r
e
4
w
e
o
b
s
er
v
ed
th
at
th
e
o
u
tp
u
t
r
e
s
u
l
t
u
s
i
n
g
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
i
n
g
al
g
o
r
ith
m
h
a
s
b
etter
s
eg
m
en
tatio
n
r
es
u
lt a
s
co
m
p
ar
ed
to
th
e
k
-
m
ea
n
s
al
g
o
r
ith
m
.
T
h
e
q
u
alit
y
o
f
t
h
e
s
e
g
m
e
n
ted
i
m
ag
e
i
s
an
al
y
ze
d
u
s
in
g
t
h
e
m
ea
s
u
r
e
m
e
n
t
a
m
o
u
n
t
o
f
P
ea
k
to
Sig
n
al
No
is
e
R
atio
n
(
P
S
N
R
)
an
d
R
o
o
t M
ea
n
Sq
u
ar
e
E
r
r
o
r
(
R
MSE
)
[
2
]
.
5
.
1
RM
SE
(
Ro
o
t
M
ea
n Squ
a
re
E
rr
o
r)
I
t
h
as
b
ee
n
u
s
ed
f
o
r
th
e
s
tan
d
ar
d
p
er
f
o
r
m
an
ce
m
ea
s
u
r
e
m
e
n
t
o
f
th
e
o
u
tp
u
t
i
m
a
g
e.
I
t
g
iv
e
s
h
o
w
m
u
c
h
o
u
tp
u
t i
m
ag
e
i
s
d
ev
iated
f
r
o
m
th
e
in
p
u
t i
m
ag
e.
o
r
i
g
i
n
a
l
i
m
a
g
e
o
b
j
e
c
t
s
i
n
c
l
u
s
t
e
r
2
o
b
j
e
c
t
s
i
n
c
l
u
s
t
e
r
2
o
r
i
g
i
n
a
l
i
m
a
g
e
o
b
j
e
c
t
s
i
n
c
l
u
s
t
e
r
6
o
b
j
e
c
t
s
i
n
c
l
u
s
t
e
r
4
o
r
i
g
i
n
a
l
i
m
a
g
e
o
b
j
e
c
t
s
i
n
c
l
u
s
t
e
r
1
o
b
j
e
c
t
s
i
n
c
l
u
s
t
e
r
1
(
a)
(
b
)
(
c)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
P
a
r
a
metric Co
mp
a
r
is
o
n
o
f K
-
mea
n
s
a
n
d
A
d
a
p
tive
K
-
mea
n
s
C
lu
s
ter
in
g
P
erfo
r
ma
n
ce
…
(
M
a
d
h
u
s
mita
S
a
h
u
)
815
R
MSE
=
√
∑
∑
[
(
)
]
∑
∑
[
]
(
4
)
5
.
2
P
SNR(
P
ea
k
t
o
Sig
na
l N
o
is
e
Ra
t
io
)
T
h
e
p
ea
k
to
s
ig
n
al
n
o
is
e
r
atio
is
th
e
p
r
o
p
o
r
tio
n
b
etw
ee
n
m
ax
i
m
u
m
attai
n
ab
le
p
o
w
er
s
an
d
th
e
co
r
r
u
p
tin
g
n
o
is
e
t
h
at
ef
f
ec
t li
k
en
es
s
o
f
i
m
ag
e.
I
t is
u
s
ed
to
m
ea
s
u
r
e
th
e
q
u
alit
y
o
f
t
h
e
o
u
tp
u
t i
m
a
g
e.
P
SNR
=
[
(
)
√
∑
∑
[
(
)
]
∑
∑
[
]
]
(
5
)
5
.
3
Co
m
pa
riso
n
Ana
ly
s
i
s
o
f
t
he
P
a
ra
m
et
er
s
I
n
th
e
T
ab
le
1
w
e
h
av
e
co
m
p
a
r
ed
th
e
p
ar
am
eter
s
r
es
u
lt
s
o
f
t
h
e
alg
o
r
it
h
m
u
s
ed
i
n
d
if
f
er
en
t i
m
a
g
es.
T
ab
le
1
.
P
ar
am
etr
ic
C
o
m
p
ar
i
s
o
n
T
ab
le
I
mag
e
R
M
S
E
P
S
N
R
El
a
p
se
d
t
i
me
K
-
m
e
a
n
s
A
d
a
p
t
i
v
e
K
-
me
a
n
s
K
-
m
e
a
n
s
A
d
a
p
t
i
v
e
K
-
me
a
n
s
K
-
m
e
a
n
s
A
d
a
p
t
i
v
e
K
-
me
a
n
s
G
r
a
y
c
o
l
o
r
B
r
a
i
n
(
M
R
I
)
1
4
.
5
9
1
4
1
3
.
7
5
6
4
9
.
6
8
8
7
1
0
.
3
5
4
0
0
.
2
3
5
1
6
3
0
.
1
5
2
0
3
4
C
o
l
o
r
B
r
a
i
n
(
M
R
I
)
2
2
.
7
3
1
0
2
0
.
0
2
4
1
8
.
2
6
1
3
1
2
.
4
9
0
5
0
.
3
1
4
1
6
2
0
.
1
5
8
1
8
0
S
a
t
e
l
l
i
t
e
2
5
.
0
6
4
7
2
4
.
1
4
2
9
8
.
5
7
3
8
8
.
7
0
6
0
0
.
2
8
3
9
9
9
0
.
0
6
7
3
5
8
5
.
4
G
ra
ph
ica
l r
epre
s
ent
a
t
io
n o
f
P
a
ra
m
et
er
s
Fig
u
r
e
5
g
r
ap
h
r
ep
r
esen
ts
t
h
e
R
MSE
in
K
-
m
ea
n
s
is
m
o
r
e
as
co
m
p
ar
ed
to
A
d
ap
tiv
e
K
-
m
ea
n
s
alg
o
r
ith
m
.
I
t
d
ef
i
n
es
t
h
at
t
h
e
A
d
ap
tiv
e
K
-
m
ea
n
s
cl
u
s
ter
in
g
i
m
a
g
e
is
le
s
s
d
ev
ia
ted
as
co
m
p
ar
ed
to
K
-
m
ea
n
s
clu
s
ter
i
n
g
i
m
a
g
e.
W
h
er
e
as
in
P
SNR
,
w
h
ic
h
d
ef
in
e
s
t
h
e
q
u
alit
y
o
f
th
e
i
m
ag
e,
is
m
o
r
e
in
A
d
ap
ti
v
e
K
-
m
ea
n
s
as
co
m
p
ar
ed
to
K
-
m
ea
n
s
cl
u
s
ter
in
g
al
g
o
r
ith
m
th
a
t
s
h
o
w
s
i
n
t
h
e
Fig
u
r
e6
g
r
ap
h
[
1
0
,
1
1
]
.
A
ls
o
in
ca
s
e
o
f
ti
m
e
p
er
io
d
A
d
ap
tiv
e
K
-
m
ea
n
s
ta
k
e
s
v
er
y
le
s
s
ti
m
e
as c
o
m
p
ar
ed
t
o
K
-
m
ea
n
s
t
h
at
w
il
l d
ef
i
n
e
f
r
o
m
th
e
f
i
g
u
r
e
7
.
Fig
u
r
e
5
.
Gr
ap
h
ical
r
ep
r
esen
ta
tio
n
o
f
R
MSE
Fig
u
r
e
6
.
Gr
ap
h
ical
r
ep
r
esen
ta
tio
n
o
f
P
SNR
Fig
u
r
e
7
.
Gr
ap
h
ical
r
ep
r
esen
ta
tio
n
o
f
E
lap
s
ed
ti
m
e
p
er
io
d
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
2
,
A
p
r
il 2
0
1
7
:
8
1
0
–
81
7
816
6.
CO
NCLU
SI
O
N
Seg
m
en
tatio
n
b
ei
n
g
an
i
m
p
o
r
t
an
t a
r
ea
o
f
r
e
s
ea
r
ch
,
d
eter
m
i
n
i
n
g
its
p
er
f
o
r
m
an
ce
is
al
s
o
i
m
p
o
r
tan
t [
9
]
.
B
y
u
s
in
g
K
-
m
ea
n
s
a
n
d
A
d
ap
t
iv
e
K
-
m
ea
n
s
al
g
o
r
ith
m
w
e
h
a
v
e
s
e
g
m
en
ted
th
e
b
r
ain
g
r
a
y
co
lo
r
&
co
lo
r
MRI
i
m
a
g
e
also
Satell
ite
i
m
ag
e
o
f
a
lak
e.
Du
r
i
n
g
s
u
ch
ti
m
e
th
ese
al
g
o
r
ith
m
u
s
es
t
h
e
c
o
lo
r
s
ep
ar
atio
n
an
d
co
n
v
er
s
io
n
tec
h
n
iq
u
es
i.e
.
R
GB
to
L
*
a
*
b
co
n
v
er
s
io
n
b
as
ed
u
p
o
n
th
e
a
*
b
c
h
r
o
m
aticit
y
th
e
al
g
o
r
ith
m
s
ar
e
u
s
ed
an
d
s
e
g
m
en
ted
t
h
e
i
m
ag
e.
W
e
ca
n
s
u
cc
es
s
f
u
ll
y
f
in
d
o
u
t
th
e
t
u
m
o
r
ce
ll
s
p
r
ea
d
in
g
ar
ea
f
r
o
m
th
e
g
r
a
y
a
n
d
co
lo
r
b
r
ain
MRI
clu
s
ter
ed
i
m
a
g
e
af
ter
th
e
s
eg
m
e
n
tatio
n
alo
n
g
w
it
h
id
en
ti
f
icatio
n
o
f
t
h
e
la
k
e
co
v
er
ar
ea
f
r
o
m
th
e
s
atel
lite i
m
ag
e.
T
h
e
R
MSE
an
d
P
SNR
v
alu
e
s
o
f
th
e
i
m
ag
e
s
b
o
th
in
K
-
m
ea
n
s
&
A
d
ap
tiv
e
K
-
m
ea
n
s
w
er
e
f
o
u
n
d
an
d
co
m
p
ar
ed
in
T
ab
le
1
.
Fro
m
wh
ich
it
is
o
b
s
er
v
ed
th
at
t
h
e
R
MSE
v
al
u
e
i
n
A
d
ap
tiv
e
K
-
m
e
an
s
i
s
les
s
t
h
en
t
h
at
o
f
K
-
m
ea
n
s
,
a
ls
o
th
e
P
SNR
v
alu
e
is
h
ig
h
er
in
A
d
ap
tiv
e
K
-
m
ea
n
s
th
e
n
t
h
at
o
f
K
-
m
ea
n
s
.
T
h
is
is
th
e
co
n
d
itio
n
o
f
g
o
o
d
im
a
g
e
s
eg
m
e
n
tatio
n
q
u
alit
y
.
Fro
m
th
e
co
m
p
ar
is
o
n
o
f
R
MSE
,
P
SNR
,
ti
m
e
p
er
io
d
&
clu
s
ter
in
g
p
ar
am
eter
s
A
d
ap
tiv
e
K
-
m
ea
n
s
clu
s
ter
i
n
g
g
i
v
es b
est a
n
d
b
etter
r
esu
lt
s
as c
o
m
p
ar
ed
to
K
-
m
ea
n
s
al
g
o
r
ith
m
.
RE
F
E
R
E
NC
E
S
[1
]
S
u
sh
m
a
S
J,
S
C
P
ra
sa
n
n
a
Ku
m
a
r,
e
t
a
l.
,
“
A
d
v
a
n
c
e
m
e
n
t
in
Re
se
a
rc
h
T
e
c
h
n
iq
u
e
s
o
n
M
e
d
ica
l
Im
a
g
in
g
P
ro
c
e
ss
in
g
f
o
r
Bre
a
st
Ca
n
c
e
r
De
tec
ti
o
n
”
,
p
u
b
li
sh
e
d
i
n
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
Co
m
p
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
V
o
l
.
6
,
No
.
2
,
A
p
ril
2
0
1
6
,
p
p
.
7
1
7
~
7
2
4
.
[2
]
Na
m
e
ira
k
p
a
m
Dh
a
n
a
c
h
a
n
d
ra
,
e
t
a
l.
,
“
Im
a
g
e
S
e
g
m
e
n
tatio
n
u
si
n
g
K
-
m
e
a
n
s
Clu
ste
rin
g
A
lg
o
rit
h
m
a
n
d
S
u
b
trac
ti
v
e
Clu
ste
rin
g
A
lg
o
rit
h
m
”
,
P
u
b
li
sh
e
d
in
El
se
v
ier
a
t
(
IM
CIP
-
2
0
1
5
)
Pro
c
e
d
ia
Co
m
p
u
ter
S
c
ien
c
e
,
p
p
.
7
6
4
–
7
7
1
.
[3
]
T
.
J
y
o
th
irm
a
y
i1
,
K.
S
ri
n
iv
a
sa
Ra
o
,
P
.
S
rin
iv
a
sa
Ra
o
,
Ch
.
S
a
ty
a
n
a
ra
y
a
n
a
,
e
t
a
l.
,
“
Im
a
g
e
S
e
g
m
e
n
tatio
n
Ba
se
d
o
n
Do
u
b
ly
T
ru
n
c
a
ted
G
e
n
e
ra
li
z
e
d
L
a
p
lac
e
M
i
x
tu
re
M
o
d
e
l
a
n
d
K
M
e
a
n
s
Clu
ste
rin
g
”
,
p
u
b
li
sh
e
d
i
n
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
C
o
mp
u
t
e
r E
n
g
i
n
e
e
rin
g
(
IJ
ECE
)
,
V
o
l
.
6
,
No
.
5
,
Oc
to
b
e
r
2
0
1
6
,
p
p
.
2
1
8
8
~
2
1
9
6
[4
]
M
in
g
-
Ni
W
u
,
e
t
a
l.
,
“
Bra
in
T
u
m
o
r
De
tec
ti
o
n
Us
in
g
Co
lo
r
-
Ba
se
d
K
-
M
e
a
n
s
Clu
ste
rin
g
S
e
g
m
e
n
tatio
n
”
,
p
u
b
li
sh
e
d
in
IEE
E
X
p
l
o
re
2
0
0
9
.
[5
]
S
a
n
ji
v
K.
Bh
a
ti
a
,
e
t
a
l.
,
“
A
d
a
p
ti
v
e
K
-
M
e
a
n
s
Clu
ste
rin
g
”
,
p
u
b
l
ish
e
d
in
Ame
ric
a
n
Asso
c
ia
t
io
n
fo
r
Arti
fi
c
ia
l
In
telli
g
e
n
c
e
,
2
0
0
4
.
[6
]
A
n
il
Z
c
h
it
a
d
e
,
e
t
a
l
.
,
“
Co
l
o
r
b
a
se
d
im
a
g
e
se
g
m
e
n
tatio
n
u
sin
g
k
-
m
e
a
n
s clu
ste
rin
g
”
,
p
u
b
l
ish
e
d
i
n
IJ
E
S
T
,
v
o
l.
2
,
2
0
1
0
[7
]
M
.
S
a
h
u
&
Dr.
K.
P
a
rv
a
th
i,
e
t
a
l.
,
”
Clas
sif
ica
ti
o
n
a
n
d
De
tec
ti
o
n
o
f
T
u
m
o
r
c
e
ll
f
ro
m
c
o
lo
r
Bra
in
M
RI
im
a
g
e
b
y
Im
p
le
m
e
n
ti
n
g
K
-
m
e
a
n
s
c
lu
ste
rin
g
a
n
d
Histo
g
ra
m
P
ro
c
e
ss
in
g
”
,
p
u
b
li
s
h
e
d
in
PI
S
ER
j
o
u
rn
a
l
2
0
1
4
,
P
IS
ER
1
3
,
V
o
l
.
0
2
,
Iss
u
e
:
0
3
/0
6
M
a
y
-
Ju
n
e
,
P
a
g
e
(s) 0
1
9
-
0
2
3
.
[8
]
M
.
sa
h
u
&
Dr
.
K.
P
a
rv
a
th
i,
e
t
a
l.
,
“
S
e
g
m
e
n
tatio
n
o
f
Co
l
o
r
Da
ta
Ba
se
Im
a
g
e
b
y
I
m
p
le
m
e
n
ti
n
g
K
-
M
e
a
n
s
Clu
ste
ri
n
g
”
,
p
u
b
li
sh
e
d
in
IJ
IET
,
v
o
l
.
2
issu
e
4
A
u
g
2
0
1
3
.
[9
]
M
.
A
k
h
il
,
R.
A
ish
w
a
r
y
a
,
V
a
n
d
h
a
n
a
L
a
l
a
n
d
S
h
a
n
t
h
i
M
a
h
e
sh
,
e
t
a
l.
“
Co
m
p
a
riso
n
a
n
d
Ev
a
lu
a
ti
o
n
o
f
S
e
g
m
e
n
tatio
n
T
e
c
h
n
iq
u
e
s f
o
r
Bra
in
M
RI
u
sin
g
G
o
ld
S
tan
d
a
r
d
”
,
p
u
b
l
ish
e
d
in
IJ
S
T
jo
u
rn
a
l
,
V
o
l
-
9
,
De
c
e
m
b
e
r
2
0
1
6
.
[1
0
]
S
.
M
a
h
a
lak
sh
m
i
a
n
d
T
.
V
e
lm
u
ru
g
a
n
,
e
t
a
l.
“
De
tec
ti
o
n
o
f
Bra
in
T
u
m
o
r
b
y
P
a
rti
c
le S
w
a
r
m
Op
ti
m
iza
t
io
n
u
sin
g
Im
a
g
e
S
e
g
m
e
n
tatio
n
”
,
p
u
b
l
ish
e
d
i
n
IJ
S
T
jo
u
r
n
a
l
,
V
o
l
-
8
,
S
e
p
tm
b
e
r
2
0
1
5
.
[1
1
]
A
.
Na
v
e
e
n
a
n
d
T
.
V
e
lm
u
ru
g
a
n
,
e
t
a
l.
“
Id
e
n
t
if
ica
ti
o
n
o
f
Ca
l
c
if
ic
a
ti
o
n
i
n
M
RI
Bra
in
Im
a
g
e
s
b
y
k
-
M
e
a
n
s
A
l
g
o
rit
h
m
”
,
p
u
b
l
ish
e
d
i
n
IJ
S
T
j
o
u
rn
a
l
,
Vo
l
-
8
,
No
v
e
m
b
e
r
2
0
1
5
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
S
m
t.
M
a
d
h
u
s
m
ita
S
a
h
u
re
c
e
iv
e
d
th
e
M
.
S
c
.
(El
e
c
tro
n
ic
sc
ien
c
e
s)
f
ro
m
Be
rh
a
m
p
u
r
u
n
iv
e
rsity
,
G
a
n
ja
m
.
M
.
tec
h
(Co
m
m
u
n
ica
ti
o
n
sy
ste
m
o
f
e
n
g
g
in
e
rin
g
)
f
ro
m
V
S
S
u
n
iv
e
rsit
y
o
f
T
e
c
h
n
o
lo
g
y
,
Bu
rla,
S
a
m
b
a
lp
u
r,
O
d
ish
a
Dr
.
K
.
Pa
r
v
a
th
i
re
c
e
iv
e
d
th
e
B
E
(e
c
e
)
f
ro
m
Os
m
a
n
ia
u
n
iv
e
rsity
,
M
T
e
c
h
&
P
h
D
f
ro
m
A
n
d
h
ra
Un
iv
e
rsit
y
,
A
P
.
Re
se
a
rc
h
in
tere
sts a
re
sig
n
a
l
&
I
m
a
g
e
P
ro
c
e
ss
in
g
,
Co
m
p
u
tatio
n
a
l
I
n
telli
g
e
n
c
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
P
a
r
a
metric Co
mp
a
r
is
o
n
o
f K
-
mea
n
s
a
n
d
A
d
a
p
tive
K
-
mea
n
s
C
lu
s
ter
in
g
P
erfo
r
ma
n
ce
…
(
M
a
d
h
u
s
mita
S
a
h
u
)
817
Dr
.
M
.
V
a
m
si
K
r
ish
n
a
re
c
e
iv
e
d
th
e
M
T
e
c
h
(CS
)
in
A
ll
a
h
a
b
a
d
Un
iv
e
rsit
y
,
M
.
T
e
c
h
(
A
I
&
R
)
d
e
g
re
e
in
A
n
d
h
ra
Un
iv
e
rsit
y
,
a
n
d
P
h
.
D
f
ro
m
Ce
n
tu
rio
n
Un
iv
e
rsity
,
Od
ish
a
.
Cu
rre
n
tl
y
h
e
is
w
o
rk
in
g
a
s
P
ro
f
e
ss
o
r
&
HO
D
in
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
.
He
h
a
s
1
5
y
e
a
rs
o
f
e
x
p
e
rien
c
e
in
tea
c
h
in
g
.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
A
rti
f
icia
l
in
telli
g
e
n
c
e
,
c
o
m
p
u
ter
n
e
tw
o
rk
s,
i
m
a
g
e
p
ro
c
e
ss
in
g
a
n
d
Da
ta
A
n
a
l
y
ti
c
s.
Evaluation Warning : The document was created with Spire.PDF for Python.