I
nte
rna
t
io
n
a
l J
o
urna
l o
f
Adv
a
nces in Applie
d Science
s
(
I
J
AAS)
Vo
l.
6
,
No
.
3
,
Sep
tem
b
er
2
0
1
7
,
p
p
.
2
5
8
~2
6
7
I
SS
N:
2252
-
8814
258
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
J
AAS
Bra
in I
m
a
g
e S
eg
m
en
tatio
n
using
Lev
el Set
:
A
n
H
y
brid
Appro
a
ch
P
Su
dh
a
rs
ha
n Dut
h
,
Vi
m
a
l
Vis
w
a
na
t
h,
P
a
nk
a
j
Sree
k
u
ma
r
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
,
Am
rit
a
V
ish
w
a
V
id
y
a
p
e
e
th
a
m
Un
iv
e
rsit
y
,
M
y
su
ru
Ca
m
p
u
s,
Ka
rn
a
tak
a
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
hi
s
to
r
y:
R
ec
eiv
ed
J
u
n
1
1
,
2
0
1
7
R
ev
i
s
ed
A
u
g
1
9
,
2
0
1
7
A
cc
ep
ted
A
u
g
2
5
,
2
0
1
7
A
d
v
a
n
c
e
s
in
m
e
d
ica
l
i
m
a
g
in
g
tec
h
n
o
l
o
g
ies
h
a
v
e
g
iv
e
n
rise
f
o
r
e
ff
e
c
ti
v
e
d
iag
n
o
stic
p
ro
c
e
d
u
re
s.
T
h
e
a
c
q
u
isit
io
n
p
r
o
m
p
tn
e
ss
a
n
d
re
so
lu
ti
o
n
e
n
h
a
n
c
e
m
e
n
ts
o
f
i
m
a
g
in
g
m
o
d
a
li
ti
e
s
h
a
v
e
g
iv
e
n
p
h
y
si
c
ian
s
m
o
re
in
f
o
rm
a
ti
o
n
,
les
s
in
v
a
siv
e
l
y
a
b
o
u
t
th
e
ir
p
a
ti
e
n
ts.
A
c
ti
v
e
c
o
n
to
u
rs
a
re
u
se
d
t
o
se
g
m
e
n
t,
m
a
tch
a
n
d
trac
k
i
m
a
g
e
s
o
f
a
n
a
to
m
ic
stru
c
tu
re
b
y
m
a
n
ip
u
latin
g
c
o
n
stra
in
ts
d
e
riv
e
d
f
ro
m
th
e
im
a
g
e
d
a
ta
to
g
e
th
e
r
w
it
h
p
r
io
r
k
n
o
w
led
g
e
a
b
o
u
t
th
e
lo
c
a
ti
o
n
,
siz
e
,
a
n
d
sh
a
p
e
o
f
th
e
se
stru
c
tu
re
s.
T
h
e
lev
e
l
se
t
m
e
th
o
d
is
re
fe
rre
d
a
s
a
p
a
rt
o
f
a
c
ti
v
e
c
o
n
to
u
r
f
a
m
il
y
.
T
h
e
m
a
jo
r
d
isa
d
v
a
n
tag
e
s
o
f
le
v
e
l
se
t
m
e
th
o
d
a
re
in
it
ializa
ti
o
n
o
f
c
o
n
tr
o
ll
i
n
g
p
a
ra
m
e
ters
a
n
d
ti
m
e
c
o
m
p
lex
it
y
.
T
h
e
p
ro
p
o
se
d
m
e
th
o
d
a
d
o
p
ts
Ro
b
u
st
S
p
a
ti
a
l
Ke
rn
e
l
F
u
z
z
y
C
-
M
e
a
n
s
(RS
KFCM
)
a
n
d
L
a
tt
ice
Bo
lt
z
m
a
n
n
M
e
th
o
d
(L
BM
)
to
o
v
e
rc
o
m
e
th
e
se
d
ra
w
b
a
c
k
s.
RS
KFCM
is
b
a
se
d
o
n
sta
n
d
a
r
d
F
u
z
z
y
C
-
M
e
a
n
s
a
l
g
o
rit
h
m
w
h
ich
u
se
s
Ga
u
ss
ian
RBF
k
e
rn
e
l
f
u
n
c
ti
o
n
a
s
d
istan
c
e
m
e
tri
c
a
n
d
in
c
o
rp
o
ra
tes
sp
a
ti
a
l
in
f
o
rm
a
ti
o
n
.
T
h
e
L
BM
u
se
s
th
e
e
n
e
rg
y
f
u
n
c
ti
o
n
to
d
e
ter
m
in
e
a
n
d
re
d
u
c
e
th
e
a
c
tu
a
l
p
r
o
c
e
ss
in
g
ti
m
e
w
h
ich
a
d
d
re
ss
e
s
th
e
ti
m
e
c
o
m
p
l
e
x
it
y
.
T
h
e
p
ro
p
o
se
d
sy
ste
m
c
o
m
b
in
e
s
b
o
t
h
RS
KFCM
a
n
d
L
BM
to
f
o
r
m
a
h
y
b
rid
a
p
p
ro
a
c
h
,
a
n
d
t
h
e
sy
ste
m
is
tes
te
d
o
n
a
larg
e
se
t
o
f
M
RI
b
ra
in
im
a
g
e
s
a
n
d
th
e
ex
p
e
ri
m
e
n
tal
re
su
lt
s
a
r
e
f
o
u
n
d
to
b
e
i
m
p
ro
v
e
d
w
it
h
re
sp
e
c
t
to
ti
m
e
c
o
m
p
lex
it
y
.
K
ey
w
o
r
d
:
C
-
m
ea
n
s
Fu
zz
y
C
-
m
ea
n
s
Ker
n
el
f
u
zz
y
C
-
m
ea
n
s
L
attice
b
o
ltz
m
an
n
m
eth
o
d
L
e
v
el
s
et
m
et
h
o
d
R
o
b
u
s
t sp
atia
l k
er
n
e
l f
u
zz
y
Sp
atial
f
u
zz
y
C
-
m
ea
n
s
Sp
atial
k
er
n
e
l f
u
zz
y
C
-
m
ea
n
s
Co
p
y
rig
h
t
©
201
7
In
s
t
it
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
:
P
Su
d
h
ar
s
h
a
n
D
u
t
h
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
Am
r
ita
Vi
s
h
w
a
Vid
y
ap
ee
th
a
m
Un
i
v
er
s
it
y
,
M
y
s
u
r
u
C
a
m
p
u
s
,
Kar
n
atak
a,
I
n
d
ia.
E
m
ail: s
u
d
h
is
p
@
g
m
ail.
c
om
1.
I
NT
RO
D
UCT
I
O
N
Me
d
ical
i
m
ag
i
n
g
h
as
u
n
d
er
g
o
n
e
ad
v
an
ce
m
e
n
t
in
t
h
e
p
r
ev
io
u
s
d
ec
ad
e
[
1
]
.
T
h
e
p
r
o
ce
s
s
o
f
m
ed
ical
i
m
a
g
in
g
i
n
v
o
lv
e
s
r
ep
r
esen
tati
o
n
o
f
t
h
e
b
o
d
y
’
s
in
ter
io
r
o
r
g
a
n
s
,
w
h
ic
h
h
elp
s
in
m
ed
ical
e
x
a
m
i
n
atio
n
a
n
d
o
th
er
m
ed
ical
p
u
r
p
o
s
es.
T
h
er
e
ar
e
m
an
y
tec
h
n
iq
u
e
s
f
o
r
m
ed
ical
i
m
a
g
in
g
at
p
r
ese
n
t
li
k
e
X
-
r
a
y
,
C
T
’
s
a
n
d
M
R
I
[
2
]
.
T
h
e
d
ata
p
r
o
d
u
ce
d
th
r
o
u
g
h
t
h
ese
m
eth
o
d
s
ar
e
ap
p
lied
in
d
if
f
er
en
t
ar
ea
s
o
f
m
ed
ical
a
n
al
y
s
i
s
li
k
e
m
o
lec
u
lar
i
m
a
g
in
g
,
n
a
n
o
tec
h
n
o
lo
g
y
etc.
Seg
m
en
tatio
n
is
th
e
co
r
e
p
ar
t
in
m
ed
ical
i
m
ag
e
p
r
o
ce
s
s
in
g
.
I
m
ag
e
s
e
g
m
en
tatio
n
is
th
e
p
r
o
ce
s
s
in
w
h
ic
h
an
i
m
ag
e
i
s
s
u
b
d
iv
id
ed
in
to
its
co
n
s
tit
u
e
n
t
p
ar
ts
.
Seg
m
e
n
tat
io
n
p
r
o
ce
s
s
ca
n
b
e
p
er
f
o
r
m
ed
b
ased
o
n
o
b
s
er
v
in
g
s
ets
o
f
p
i
x
els
wh
er
eb
y
p
ix
els
r
ep
r
esen
tin
g
a
r
eg
io
n
ca
n
b
e
id
en
tif
ied
ac
co
r
d
in
g
to
s
o
m
e
h
o
m
o
g
en
eit
y
cr
i
ter
ia
li
k
e
co
l
o
r
,
tex
tu
r
e
o
r
i
n
te
n
s
it
y
s
o
as
to
lo
ca
te
an
d
id
en
ti
f
y
b
o
u
n
d
ar
ies
in
an
i
m
a
g
e.
Seg
m
en
tatio
n
o
f
m
ed
ical
i
m
ag
es
is
co
m
p
lex
d
u
e
to
s
e
v
er
al
r
ea
s
o
n
s
s
u
ch
as
ch
ar
ac
t
er
is
tics
o
f
i
m
a
g
i
n
g
m
o
d
ali
t
y
,
g
eo
m
etr
y
o
f
a
n
ato
m
y
,
p
ar
tial
v
o
lu
m
e
e
f
f
ec
t,
p
r
ese
n
ce
o
f
ar
ti
f
ac
ts
an
d
n
o
is
e.
A
p
p
licatio
n
s
o
f
s
eg
m
e
n
tatio
n
i
n
m
ed
ical
i
m
a
g
e
p
r
o
ce
s
s
in
g
ar
e
to
lo
ca
te
tu
m
o
r
s
an
d
o
th
er
p
ath
o
lo
g
ies.
T
h
r
o
u
g
h
s
e
g
m
e
n
tatio
n
,
o
th
er
p
r
o
ce
s
s
es
li
k
e
b
o
u
n
d
ar
y
a
n
al
y
s
is
,
in
te
n
s
it
y
d
etec
tio
n
a
n
d
ab
n
o
r
m
alit
y
id
e
n
ti
f
icat
io
n
ar
e
ca
r
r
ied
o
u
t.
T
h
ese
p
r
o
ce
s
s
es a
llo
w
a
p
h
y
s
i
cian
’
s
tas
k
o
f
d
ia
g
n
o
s
in
g
t
h
e
s
tag
e
o
f
t
h
e
tu
m
o
r
to
b
e
m
u
c
h
e
asier
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
B
r
a
in
I
ma
g
e
S
eg
men
ta
tio
n
Usi
n
g
Leve
l S
et:
A
n
Hyb
r
id
A
p
p
r
o
a
ch
(
S
u
d
h
a
r
s
h
a
n
Du
t
h
)
259
I
n
th
e
L
iter
at
u
r
e
s
u
r
v
e
y
th
a
t
w
e
h
av
e
d
o
n
e,
w
e
co
u
ld
f
in
d
t
h
at
m
an
y
r
esear
c
h
er
s
p
r
o
p
o
s
ed
s
eg
m
e
n
tatio
n
m
et
h
o
d
s
.
T
r
ad
it
io
n
al
s
e
g
m
en
ta
tio
n
m
et
h
o
d
s
ca
n
b
e
class
i
f
ied
i
n
to
th
r
ee
ca
t
eg
o
r
ies:
t
h
r
esh
o
l
d
b
ased
[
3
]
r
eg
io
n
b
ased
an
d
ed
g
e
b
ased
.
Ho
w
e
v
er
p
o
o
r
c
o
n
tr
ast,
lack
o
f
s
p
atia
l
in
f
o
r
m
atio
n
,
an
d
n
o
is
e
i
n
t
h
e
i
m
a
g
es
ca
u
s
ed
th
e
s
e
g
m
e
n
tat
io
n
to
b
e
a
h
ec
tic
p
r
o
ce
s
s
.
C
lu
s
ter
i
n
g
tec
h
n
iq
u
es
o
n
s
eg
m
en
tatio
n
p
r
o
ce
s
s
i
s
b
ein
g
u
s
ed
f
r
o
m
ea
r
l
y
d
ec
ad
es; C
an
n
y
[
4
]
u
s
ed
an
e
x
te
n
d
ed
FC
M
al
g
o
r
ith
m
f
o
r
ed
g
e
d
etec
tio
n
to
s
e
g
m
e
n
t t
h
e
i
m
a
g
es.
P
h
a
m
a
n
d
P
r
in
ce
[
5
]
p
r
o
p
o
s
ed
an
alg
o
r
ith
m
u
s
i
n
g
Fu
zz
y
C
m
ea
n
s
w
it
h
p
r
esen
ce
o
f
i
n
te
n
s
i
t
y
h
o
m
o
g
en
eit
ies
o
n
MRI
i
m
ag
es.
Sali
m
a
et
a
l.
[
6
]
p
r
o
p
o
s
ed
a
n
e
w
s
eg
m
e
n
tatio
n
m
et
h
o
d
b
ased
o
n
FC
M
clu
s
ter
i
n
g
in
w
h
ic
h
it
d
ea
ls
w
it
h
t
w
o
p
r
o
b
le
m
s
o
f
t
h
e
co
n
v
e
n
tio
n
al
F
C
M
clu
s
ter
in
g
a
lg
o
r
ith
m
.
Firs
t,
th
e
s
p
atial
co
n
s
tr
ain
s
d
u
r
in
g
th
e
clu
s
ter
in
g
p
r
o
ce
s
s
b
y
i
n
co
r
p
o
r
atin
g
lo
ca
l
n
ei
g
h
b
o
u
r
h
o
o
d
in
f
o
r
m
at
io
n
to
th
e
m
e
m
b
er
s
h
i
p
f
u
n
ctio
n
h
a
s
b
e
en
co
n
s
id
er
ed
.
T
h
e
lo
ca
l
i
n
f
o
r
m
atio
n
r
e
f
lects
t
h
e
s
p
atial
in
f
l
u
e
n
ce
o
f
t
h
e
n
eig
h
b
o
u
r
i
n
g
p
ix
el
s
o
n
th
e
c
en
tr
al
p
ix
e
l.
Seco
n
d
l
y
,
I
n
o
r
d
er
to
av
o
id
lo
ca
l
m
i
n
i
m
a
o
f
t
h
e
s
p
atia
l
F
C
M
alg
o
r
ith
m
,
A
B
C
al
g
o
r
ith
m
i
s
u
s
ed
.
C
h
u
a
n
g
[
1
]
alter
ed
th
e
FC
M
m
et
h
o
d
,
an
d
p
r
o
p
o
s
ed
an
alg
o
r
it
h
m
ca
lled
MS
-
FC
M
f
o
r
MRI
b
r
ain
i
m
ag
e
s
e
g
m
e
n
tat
io
n
,
w
h
ich
o
v
e
r
co
m
e
s
th
e
p
r
o
b
le
m
o
f
n
o
is
e
s
en
s
iti
v
it
y
in
t
h
e
tr
ad
itio
n
al
F
C
M
b
y
al
lo
w
in
g
s
p
atial
in
f
o
r
m
atio
n
to
b
e
u
t
iliz
ed
.
Z
h
o
u
et
al.
,
[
7
]
p
r
o
p
o
s
ed
f
u
zz
y
c
m
ea
n
b
ased
o
n
m
ea
n
s
h
i
f
t
w
h
ic
h
r
elate
s
s
i
m
ilar
to
t
h
e
lev
e
l
s
e
t
al
g
o
r
i
th
m
.
Z
a
n
at
y
et
a
l.,
[
8
]
p
r
o
p
o
s
ed
alter
n
at
iv
e
f
o
r
E
u
clid
ea
n
d
i
s
ta
n
ce
i
n
s
ta
n
d
ar
d
FC
M
b
y
i
n
tr
o
d
u
cin
g
n
e
w
k
er
n
elize
d
d
is
ta
n
ce
m
etr
ic
(
KF
C
M)
w
h
ich
co
n
s
id
er
s
p
atial
co
n
s
tr
ain
s
as
w
ell.
T
h
e
co
m
b
i
n
atio
n
o
f
Sp
atial
F
C
M
(
SF
C
M)
a
n
d
L
ev
e
l
Set
m
et
h
o
d
f
o
r
i
m
a
g
e
s
eg
m
e
n
tatio
n
r
esu
lts
i
n
b
etter
o
u
tco
m
e
s
.
L
i
et
al.
,
[
9
]
p
r
o
p
o
s
ed
th
e
v
er
y
s
a
m
e
m
et
h
o
d
f
o
r
m
ed
ical
i
m
a
g
e
s
eg
m
e
n
tatio
n
.
As
th
e
tr
ad
itio
n
al
FC
M
u
s
e
s
E
u
clid
ea
n
d
is
t
an
ce
,
it
lack
s
in
p
r
o
d
u
cin
g
ac
cu
r
ate
r
esu
lt
s
o
n
s
eg
m
e
n
tatio
n
.
A
b
is
h
a
a
n
d
S
h
i
j
i
[
1
0
]
p
r
o
p
o
s
ed
k
er
n
el
i
n
d
u
c
ed
d
is
tan
ce
m
etr
ic
f
o
r
t
h
e
b
et
ter
s
eg
m
e
n
tatio
n
i
n
w
h
ic
h
it
u
s
es
b
o
th
s
p
atial
a
n
d
k
er
n
el
in
f
o
r
m
a
tio
n
.
I
t
r
ed
u
ce
s
th
e
p
r
o
b
le
m
r
ai
s
ed
d
u
e
to
t
h
e
n
o
is
e
a
n
d
o
u
tlier
s
.
T
r
a
d
itio
n
al
FC
M
alo
n
e
lack
s
in
p
r
o
d
u
cin
g
b
etter
r
esu
lt
s
f
o
r
m
ed
ical
i
m
a
g
e
s
e
g
m
e
n
tat
io
n
s
in
ce
it
d
o
esn
’
t
u
s
e
s
s
p
atial
in
f
o
r
m
atio
n
as
s
aid
in
ab
o
v
e
p
ap
er
s
.
C
o
m
b
in
i
n
g
SF
C
M
an
d
KFC
M
to
g
eth
er
w
ill
p
r
o
v
id
e
o
p
tim
i
s
tic
r
esu
lt
s
s
i
n
ce
i
t
u
s
es
b
o
th
t
h
e
s
p
atial
a
n
d
k
er
n
el
i
n
f
o
r
m
ati
o
n
.
A
r
u
n
ak
u
m
ar
an
d
Har
is
h
[
1
1
]
im
p
le
m
en
ted
a
co
m
b
i
n
atio
n
o
f
s
p
atial
i
n
f
o
r
m
atio
n
w
it
h
d
is
tan
ce
a
s
k
er
n
el
m
etr
ic
ca
lli
n
g
th
is
m
et
h
o
d
as
SK
FC
M,
w
h
ic
h
p
r
o
d
u
ce
d
a
r
o
b
u
s
t
r
esu
lts
a
n
d
th
is
cl
u
s
ter
in
g
alg
o
r
it
h
m
w
a
s
n
a
m
ed
as
R
SKF
C
M.
S
h
ar
m
a
an
d
Mu
k
h
e
r
j
ee
[
1
2
]
p
r
o
p
o
s
ed
a
m
et
h
o
d
in
w
h
ich
i
t
u
s
e
s
Gr
e
y
lev
el
C
o
-
o
cc
u
r
r
en
ce
Ma
tr
ix
(
GL
C
M)
f
o
r
tex
t
u
r
e
f
ea
t
u
r
e
e
x
tr
ac
tio
n
,
A
N
FIS
(
A
d
ap
tiv
e
Net
w
o
r
k
F
u
zz
y
I
n
f
er
en
ce
S
y
s
te
m
)
p
lu
s
Gen
etic
Alg
o
r
it
h
m
f
o
r
f
ea
t
u
r
e
s
elec
tio
n
a
n
d
F
C
M
(
Fu
zz
y
C
Me
a
n
s
)
f
o
r
s
eg
m
e
n
ta
tio
n
o
f
A
s
tr
o
c
y
to
m
a
(
B
r
ain
T
u
m
o
r
)
w
i
th
all
f
o
u
r
Gr
ad
es.
Th
e
L
e
v
el
Set
m
et
h
o
d
h
elp
s
i
n
f
i
n
d
in
g
th
e
s
h
ap
es
a
n
d
b
o
u
n
d
ar
ies
e
f
f
icien
tl
y
i
n
an
i
m
a
g
e.
S
u
r
i
[
1
3
]
u
s
ed
L
e
v
el
S
et
Me
th
o
d
f
o
r
MRI
s
ca
n
n
ed
b
r
ain
i
m
a
g
e
s
e
g
m
e
n
t
atio
n
w
h
ic
h
g
av
e
f
aster
a
n
d
b
etter
r
esu
l
ts
.
P
ar
ag
io
s
[
1
4
]
in
tr
o
d
u
ce
d
k
n
o
w
led
g
e
-
b
ased
co
n
s
tr
ai
n
ts
b
y
d
e
ali
n
g
w
it
h
th
e
lo
ca
l
d
e
f
o
r
m
atio
n
s
in
L
e
v
el
Set.
M
itch
e
ll
[
1
5
]
in
tr
o
d
u
ce
d
f
le
x
ib
le
a
n
d
ex
ten
s
ib
le
L
e
v
el
Set
Me
th
o
d
s
.
L
i
et
al.
,
[
1
6
]
p
r
o
p
o
s
e
d
r
e
-
in
it
ializatio
n
m
e
th
o
d
wh
ich
is
ap
p
lied
to
p
er
io
d
ically
r
ep
lace
th
e
d
eg
r
a
d
ed
lev
el
s
et
f
u
n
ctio
n
w
it
h
a
s
ig
n
ed
d
is
tan
ce
f
u
n
ctio
n
.
T
h
e
lev
el
s
et
m
e
th
o
d
in
tr
o
d
u
ce
d
in
t
h
e
ea
r
lier
d
ec
ad
e
is
d
ep
en
d
ed
o
n
th
e
p
o
s
itio
n
o
f
th
e
i
n
itial
co
n
to
u
r
,
an
d
th
e
ev
o
lv
i
n
g
cu
r
v
e
ca
n
b
e
tr
ap
p
ed
in
to
lo
ca
l
m
i
n
i
m
a.
T
h
e
m
et
h
o
d
p
r
o
p
o
s
ed
b
y
C
h
an
a
n
d
Vese
[
1
7
]
is
al
s
o
n
o
t
s
u
itab
le
f
o
r
p
ar
allel
p
r
o
g
r
am
m
i
n
g
b
ec
au
s
e
t
h
e
a
v
er
ag
e
in
te
n
s
ities
i
n
s
id
e
an
d
o
u
ts
id
e
t
h
e
co
n
to
u
r
is
co
m
p
u
te
d
at
ea
ch
iter
atio
n
.
T
h
ese
r
esu
lts
i
n
to
m
o
r
e
C
P
U
ti
m
e
co
n
s
u
m
p
tio
n
b
y
i
n
cr
ea
s
i
n
g
i
n
ter
p
r
o
ce
s
s
co
m
m
u
n
icati
o
n
s
.
I
n
s
p
ir
ed
b
y
t
h
e
b
etter
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
R
S
KFC
M
m
et
h
o
d
,
Su
d
h
a
r
s
h
a
n
Du
t
h
[
1
8
]
p
r
o
p
o
s
ed
a
m
et
h
o
d
to
in
itialize
th
e
le
v
el
s
et
p
ar
a
m
eter
b
ased
o
n
R
o
b
u
s
t
Sp
atial
Ker
n
el
F
u
zz
y
C
-
Me
an
s
(
R
SKF
C
M)
.
B
alla
et
al.
,
[
1
9
]
p
r
o
p
o
s
ed
a
n
e
w
m
et
h
o
d
w
h
ic
h
ai
m
s
at
t
h
e
ev
o
lv
i
n
g
cu
r
v
e
to
b
e
s
to
p
p
ed
ac
co
r
d
in
g
to
th
e
m
e
m
b
er
s
h
ip
d
eg
r
ee
o
f
th
e
cu
r
r
en
t
p
i
x
el
to
b
e
in
s
id
e
o
r
o
u
ts
id
e
o
f
th
e
ac
tiv
e
co
n
to
u
r
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
d
e
v
elo
p
ed
w
i
th
th
e
h
elp
o
f
t
h
e
m
o
d
if
ied
f
u
zz
y
C
-
m
ea
n
s
(
F
C
M)
a
n
d
L
a
ttice
B
o
ltz
m
an
n
Me
th
o
d
)
.
L
B
M
ca
n
h
an
d
le
p
r
o
b
le
m
o
f
ti
m
e
co
n
s
u
m
p
tio
n
a
s
th
e
c
u
r
v
atu
r
e
i
m
p
licitl
y
co
m
p
u
ted
a
n
d
th
e
al
g
o
r
ith
m
i
s
s
i
m
p
le
f
o
r
p
ar
allel
p
r
o
g
r
am
m
i
n
g
.
L
B
M
ac
ts
as a
lter
n
ati
v
e
s
o
lv
er
f
o
r
L
e
v
el
Set E
q
u
atio
n
.
I
n
t
h
is
p
ap
er
w
e
m
a
k
e
u
s
e
o
f
R
SK
FC
M
an
d
L
a
ttice
B
o
ltzm
a
n
n
Me
t
h
o
d
f
o
r
m
ed
ical
i
m
ag
e
s
eg
m
e
n
tatio
n
.
R
SKF
C
M
is
b
a
s
ed
o
n
s
tan
d
ar
d
FC
M
alg
o
r
it
h
m
,
w
h
ic
h
co
n
s
id
er
s
s
p
atial
in
f
o
r
m
at
io
n
an
d
u
s
es
Gau
s
s
ia
n
R
B
F
k
er
n
el
f
u
n
ctio
n
as
d
i
s
tan
ce
m
etr
ic.
C
las
s
ica
l
m
et
h
o
d
s
ta
k
e
m
o
r
e
C
P
U
ti
m
e
f
o
r
s
o
lv
i
n
g
L
SE.
An
alter
n
ati
v
e
ap
p
r
o
ac
h
to
s
o
lv
e
L
e
v
el
Set
E
q
u
atio
n
(
L
SE)
is
L
attice
B
o
ltz
m
a
n
n
Me
th
o
d
(
L
B
M)
.
L
B
M
i
s
a
n
u
m
er
ical
f
r
a
m
e
w
o
r
k
f
o
r
m
o
d
ellin
g
B
o
ltz
m
an
n
p
ar
ticle
d
y
n
a
m
ics
o
n
a
2
-
D
o
r
3
-
D
lattic
e
w
h
ic
h
o
v
er
co
m
es
th
e
p
r
o
b
le
m
o
f
clas
s
ical
m
et
h
o
d
s
an
d
p
r
o
d
u
ce
s
ac
cu
r
ate
r
e
s
u
lt
s
.
I
t
w
a
s
f
ir
s
t
d
es
ig
n
ed
to
s
o
lv
e
m
ac
r
o
s
co
p
ic
f
l
u
id
d
y
n
a
m
ic
s
ch
alle
n
g
es.
T
h
e
m
et
h
o
d
is
ac
cu
r
ate
b
o
th
i
n
ti
m
e
an
d
i
n
s
p
ac
e
u
p
to
s
ec
o
n
d
o
r
d
er
.
L
B
M
h
as
m
o
r
e
ad
v
a
n
ta
g
es li
k
e
p
ar
alleli
za
b
ilit
y
an
d
s
i
m
p
lici
t
y
a
s
it is
l
o
ca
l a
n
d
ex
p
licit in
n
at
u
r
e.
T
h
e
r
est
o
f
th
e
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
w
s
:
i
n
s
ec
tio
n
2
,
w
e
g
iv
e
a
b
r
ief
b
ac
k
g
r
o
u
n
d
ab
o
u
t
R
SKF
C
M
a
n
d
its
al
g
o
r
ith
m
,
lev
el
s
et
m
et
h
o
d
an
d
L
att
ic
e
B
o
ltzm
a
n
n
m
et
h
o
d
.
Sectio
n
3
,
in
tr
o
d
u
ce
s
o
u
r
alg
o
r
ith
m
a
n
d
g
i
v
es
th
e
m
a
th
e
m
atica
l
an
al
y
s
is
o
f
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
.
Sectio
n
4
p
r
esen
ts
t
h
e
e
x
p
er
i
m
e
n
ta
l
r
esu
lt
s
.
Fin
a
ll
y
,
w
e
p
r
o
v
id
e
co
n
clu
d
i
n
g
r
e
m
ar
k
s
i
n
Sectio
n
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8814
IJ
AA
S
Vo
l.
6
,
No
.
3
,
Sep
tem
b
er
2
0
1
7
:
2
58
–
2
67
260
2.
B
ACK
G
RO
UN
D
2.1
R
obu
st
Sp
a
t
i
a
l
K
er
ne
l
F
C
M (
R
S
K
F
C
M)
C
lu
s
ter
i
n
g
a
n
d
s
e
g
m
e
n
tatio
n
ar
e
v
er
y
clo
s
e
ter
m
s
i
n
i
m
ag
e
p
r
o
ce
s
s
in
g
.
C
l
u
s
ter
in
g
tech
n
iq
u
es
s
ep
ar
ate
p
ix
els
h
a
v
in
g
th
e
s
a
m
e
ch
ar
ac
ter
i
s
tics
.
T
h
e
s
ep
ar
atio
n
is
b
ased
o
n
p
ar
a
m
eter
s
li
k
e
s
p
atial
in
f
o
r
m
atio
n
a
n
d
d
is
ta
n
ce
m
e
tr
ics.
T
h
e
r
esu
lt
o
f
cl
u
s
ter
in
g
is
m
ap
p
ed
to
s
p
atial
d
o
m
ai
n
as
s
ep
ar
ated
r
eg
io
n
s
i
n
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
.
I
n
clu
s
ter
in
g
t
h
e
d
ata
is
d
iv
id
ed
in
to
clu
s
ter
s
b
ased
o
n
s
o
m
e
s
i
m
i
l
ar
it
y
m
ea
s
u
r
es
li
k
e
d
is
tan
ce
,
co
n
n
ec
ti
v
it
y
,
in
ten
s
it
y
.
F
u
zz
y
c
-
m
ea
n
s
i
s
w
ell
k
n
o
w
n
cl
u
s
ter
in
g
m
eth
o
d
w
h
e
r
e
d
ata
p
o
in
ts
ar
e
clu
s
ter
ed
b
ased
o
n
m
e
m
b
er
s
h
ip
f
u
n
ct
io
n
a
s
s
i
g
n
ed
to
ea
c
h
o
f
t
h
e
m
.
T
h
e
tr
ad
itio
n
al
F
C
M
is
n
o
t
s
u
itab
le
f
o
r
r
ev
ea
lin
g
n
o
n
E
u
clid
ea
n
s
tr
u
c
tu
r
e,
as
t
h
e
alg
o
r
it
h
m
d
o
es
n
o
t
u
tili
ze
th
e
s
p
atial
in
f
o
r
m
a
tio
n
.
B
y
u
s
i
n
g
s
p
atial
in
f
o
r
m
atio
n
i
n
to
th
e
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
to
i
m
p
r
o
v
e
th
e
s
eg
m
en
ta
tio
n
r
es
u
lts
Sp
atial
FC
M
(
SF
C
M)
w
a
s
in
tr
o
d
u
ce
d
w
h
ich
i
s
m
o
r
e
r
o
b
u
s
t
to
n
o
is
es
a
n
d
o
u
tl
ier
s
b
u
t
o
n
l
y
r
e
m
o
v
es
t
h
e
n
o
is
e
p
ar
tiall
y
.
A
k
er
n
el
p
ar
am
eter
,
u
s
i
n
g
all
d
ata
p
o
in
ts
in
t
h
e
co
llectio
n
is
d
eter
m
i
n
ed
in
Ker
n
el
F
C
M
(
KFC
M)
.
Or
ig
in
al
E
u
clid
ea
n
d
is
tan
ce
i
n
t
h
e
F
C
M
i
s
r
ep
lace
d
b
y
a
k
er
n
el
-
i
n
d
u
ce
d
d
is
tan
ce
in
it.K
FC
M
w
it
h
s
p
atial
co
n
s
tr
ai
n
ts
(
SKF
C
M)
,
it
p
r
o
v
id
es
b
etter
an
d
m
o
r
eo
v
er
f
aster
s
eg
m
e
n
tatio
n
r
esu
lt
t
h
an
FC
M
an
d
S
FC
M
alo
n
e.
R
SKFC
M
is
b
a
s
e
d
o
n
s
tan
d
ar
d
FC
M
al
g
o
r
ith
m
,
w
h
i
ch
in
co
r
p
o
r
ates
s
p
atial
in
f
o
r
m
atio
n
an
d
u
s
es
Ga
u
s
s
ian
R
B
F
k
er
n
el
f
u
n
ctio
n
.
I
t
p
r
o
d
u
ce
s
g
o
o
d
r
esu
lts
as it
u
s
e
s
m
u
ltip
le
ad
v
a
n
ta
g
es o
f
d
i
f
f
er
en
t
m
e
th
o
d
s
[
1
1
]
.
2
.
1
.
1
Alg
o
rit
h
m
T
h
e
w
o
r
k
i
n
g
o
f
R
SK
FC
M
al
g
o
r
ith
m
i
s
b
ased
o
n
th
e
f
o
llo
w
i
n
g
s
tep
s
:
a.
Dis
tr
ib
u
te
s
th
e
p
i
x
el
o
f
t
h
e
in
p
u
t i
m
ag
e
i
n
to
d
ata
s
et
X
an
d
i
n
itialize
th
e
v
al
u
e
o
f
ce
n
ter
s
Ɛ
,
m
b.
C
o
m
p
u
te
all
m
e
m
b
er
s
h
ip
v
al
u
es
ij
u
o
f
ea
ch
p
ix
el
a
g
ai
n
s
t c
e
n
ter
s
s
u
c
h
as
1
(
1
)
1
(
1
)
1
(
1
(
,
)
)
(
1
(
,
)
)
m
ji
ij
c
m
jk
k
K
x
v
u
K
x
v
c.
C
o
m
p
u
te
th
e
n
e
w
m
e
m
b
er
s
h
ip
v
alu
e
ij
w
1
pq
i
j
i
j
ij
c
pq
k
j
k
j
k
us
w
us
()
j
i
j
i
k
k
N
K
x
su
W
h
er
e
()
j
N
K
x
r
ep
r
esen
ts
a
s
q
u
ar
e
w
i
n
d
o
w
ca
n
ter
ed
o
n
p
ix
el
i
n
th
e
j
x
s
p
atial
d
o
m
a
in
d.
C
alcu
la
te
th
e
o
b
j
ec
tiv
e
f
u
n
c
ti
o
n
J
as f
o
llo
w
s
:
11
2
(
1
(
,
)
)
cN
m
i
j
j
i
ij
J
w
K
x
v
e.
C
alcu
late
n
e
w
ce
n
tr
e
v
al
u
es
i
v
1
1
n
m
i
j
j
j
i
n
m
ij
j
wx
v
w
f.
E
v
alu
a
te
th
e
t
h
r
es
h
o
ld
o
f
ter
m
in
atio
n
co
n
d
itio
n
{
(
)
(
1
)
}
J
i
J
i
Ɛ
,
w
h
er
e
Ɛ
is
a
ter
m
i
n
atio
n
cr
iter
io
n
.
Sto
p
if
it is
s
a
tis
f
ied
o
th
er
w
i
s
e
g
o
to
s
tep
2
.
2
.
2
L
ev
el
Set
M
et
ho
d (
L
S
M
)
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
B
r
a
in
I
ma
g
e
S
eg
men
ta
tio
n
Usi
n
g
Leve
l S
et:
A
n
Hyb
r
id
A
p
p
r
o
a
ch
(
S
u
d
h
a
r
s
h
a
n
Du
t
h
)
261
T
h
e
L
SM
is
a
n
u
m
er
ical
tec
h
n
iq
u
e
f
o
r
tr
ac
k
i
n
g
in
ter
f
ac
e
s
an
d
s
h
ap
es.
I
t
h
as
t
h
e
ad
v
an
tag
e
o
f
h
an
d
li
n
g
au
to
m
atica
l
l
y
to
p
o
lo
g
ical
c
h
an
g
es
o
f
t
h
e
tr
ac
k
ed
s
h
ap
e
as
it
u
s
es
a
n
i
m
p
licit
r
ep
r
esen
tat
io
n
o
f
ac
t
iv
e
co
n
to
u
r
s
.
I
n
2
-
D
i
m
a
g
e
s
eg
m
en
tatio
n
,
t
h
e
L
SM
r
ep
r
esen
ts
a
clo
s
ed
cu
r
v
e
as
th
e
ze
r
o
lev
el
s
et
o
f
,
ca
lled
th
e
lev
el
s
et
f
u
n
c
tio
n
.
T
h
e
ev
o
lu
tio
n
o
f
t
h
e
cu
r
v
e
s
tar
t
s
f
r
o
m
an
ar
b
itra
r
y
s
tar
ti
n
g
co
n
to
u
r
an
d
ev
o
lv
e
s
its
e
l
f
d
r
iv
en
b
y
t
h
e
L
SE
w
h
ich
ca
n
b
e
s
ee
n
as a
co
n
v
ec
tio
n
–
d
if
f
u
s
io
n
eq
u
atio
n
.
Vb
t
(
1
)
W
h
er
e
an
d
ar
e
th
e
g
r
ad
ie
n
t
an
d
th
e
L
ap
lacia
n
o
f
r
esp
ec
tiv
el
y
.
T
h
e
ter
m
b
is
ca
lled
ar
ti
f
ici
al
v
is
co
s
it
y
(
Set
h
ian
s
u
g
g
e
s
ted
r
ep
lacin
g
it
w
it
h
||
bk
w
h
ich
is
b
ette
r
f
o
r
h
a
n
d
li
n
g
t
h
e
e
v
o
lu
tio
n
o
f
lo
w
er
d
i
m
en
s
io
n
al
i
n
ter
f
ac
es
[
1
9
]
)
,
an
d
k
is
th
e
c
u
r
v
at
u
r
e
o
f
th
e
d
is
tan
ce
f
u
n
ctio
n
.
T
h
e
L
SE
ca
n
th
er
ef
o
r
e
b
e
w
r
itte
n
as
.
||
V
b
k
t
(
2)
2
.
3
L
a
t
t
ice
B
o
lt
zm
a
n
n M
et
ho
d (
L
B
M
)
T
h
e
L
B
M
i
s
a
n
u
m
er
ical
f
r
a
m
e
w
o
r
k
f
o
r
m
o
d
ellin
g
B
o
ltz
m
an
n
p
ar
ticle
d
y
n
a
m
ics
o
n
a
2
-
D
o
r
3
-
D
lattice,
w
h
ich
w
as
f
ir
s
t
d
e
s
ig
n
ed
to
s
o
lv
e
m
ac
r
o
s
co
p
ic
f
lu
id
d
y
n
a
m
ics
c
h
alle
n
g
es.
T
h
e
m
et
h
o
d
is
ac
cu
r
at
e
b
o
th
in
ti
m
e
a
n
d
in
s
p
ac
e
u
p
t
o
s
ec
o
n
d
o
r
d
er
.
T
h
e
A
d
v
a
n
tag
es o
f
L
B
M
ar
e
p
ar
alleliza
b
ilit
y
a
n
d
s
i
m
p
licit
y
d
u
e
to
lo
ca
l
an
d
ex
p
licit
n
at
u
r
e.
L
B
M
ca
n
h
an
d
le
p
r
o
b
le
m
o
f
ti
m
e
co
n
s
u
m
p
tio
n
a
s
th
e
c
u
r
v
atu
r
e
i
s
i
m
p
l
icitl
y
co
m
p
u
ted
an
d
t
h
e
al
g
o
r
ith
m
is
s
i
m
p
le
f
o
r
p
ar
allel
p
r
o
g
r
am
m
in
g
.
D2
Q9
m
eth
o
d
,
i
n
w
h
i
ch
t
h
e
2
D
i
m
ag
e
is
p
ar
titi
o
n
ed
in
to
eig
h
t
li
n
k
s
w
i
th
its
n
ei
g
h
b
o
u
r
s
an
d
o
n
e
li
n
k
f
o
r
th
e
ce
ll
its
e
lf
.
T
h
e
L
B
M
ev
o
lu
tio
n
eq
u
atio
n
ca
n
b
e
w
r
itte
n
as
f
o
llo
w
s
u
s
i
n
g
th
e
B
h
a
tn
a
g
ar
,
Gr
o
s
s
an
d
Kr
o
o
k
m
o
d
el
[
2
0
]
.
1
(
,
1
)
(
,
)
[
(
,
)
(
,
)
]
eq
i
i
i
i
i
f
r
e
t
f
r
t
f
r
t
f
r
t
T
(
3
)
W
h
er
e
r
ep
r
esen
ts
t
h
e
r
elax
atio
n
ti
m
e
d
eter
m
i
n
in
g
t
h
e
k
i
n
e
m
a
tic
v
is
co
s
it
y
o
f
t
h
e
f
l
u
id
b
y
11
32
(
4
)
an
d
eq
i
f
is
t
h
e
eq
u
ilib
r
iu
m
p
ar
ticle
d
is
tr
ib
u
tio
n
d
e
f
in
ed
as
22
,
(
)
(
)
(
)
eq
i
i
i
i
i
i
i
f
u
A
B
e
u
C
e
u
D
u
(
5
)
w
h
er
e
i
A
to
i
D
ar
e
co
n
s
tan
t
co
e
f
f
ici
en
ts
d
ep
en
d
in
g
o
n
t
h
e
g
eo
m
etr
y
o
f
th
e
latt
ice
lin
k
s
,
an
d
u
ar
e
th
e
m
ac
r
o
s
co
p
ic
f
lu
id
d
en
s
it
y
a
n
d
v
elo
cit
y
,
r
esp
ec
ti
v
el
y
,
co
m
p
u
ted
f
r
o
m
t
h
e
p
ar
ticle
d
is
tr
ib
u
tio
n
a
s
1
i
i
ii
i
f
u
f
e
(
6
)
Fo
r
m
o
d
eli
n
g
t
y
p
ical
d
if
f
u
s
io
n
co
m
p
u
tatio
n
s
,
th
e
eq
u
ilib
r
iu
m
f
u
n
ctio
n
ca
n
b
e
s
i
m
p
li
f
ied
as f
o
llo
w
s
[
2
2
]
.
,
eq
ii
f
u
A
(
7)
I
n
th
e
ca
s
e
o
f
D2
Q9
m
o
d
el,
i
A
=4
/9
f
o
r
th
e
ze
r
o
lin
k
,
i
A
=1
/9
f
o
r
th
e
ax
ial
li
n
k
s
,
an
d
i
A
=1
/3
6
f
o
r
th
e
d
iag
o
n
a
l lin
k
s
.
No
w
,
t
h
e
r
elax
atio
n
t
i
m
e
is
d
eter
m
i
n
ed
b
y
th
e
d
i
f
f
u
s
io
n
co
ef
f
icie
n
t
d
ef
in
ed
as
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8814
IJ
AA
S
Vo
l.
6
,
No
.
3
,
Sep
tem
b
er
2
0
1
7
:
2
58
–
2
67
262
2
(2
1
)
9
(
8
)
As
s
h
o
w
n
in
[
2
1
]
,
L
B
M
ca
n
b
e
u
s
ed
to
s
o
l
v
e
t
h
e
p
ar
ab
o
lic
d
if
f
u
s
io
n
w
h
ic
h
ca
n
b
e
r
ec
o
v
er
ed
b
y
t
h
e
C
h
ap
m
an
-
E
n
s
k
o
g
e
x
p
an
s
io
n
.
t
(
9
)
I
n
th
i
s
ca
s
e,
th
e
e
x
ter
n
a
l f
o
r
ce
ca
n
b
e
in
cl
u
d
ed
as f
o
llo
w
s
21
(
.
)
2
i
i
i
i
f
f
B
F
e
(
1
0
)
Mo
r
eo
v
er
,
th
u
s
,
(
9
)
b
ec
o
m
es
F
t
(
1
1
)
R
ep
lacin
g
b
y
t
h
e
s
i
g
n
ed
d
is
ta
n
ce
f
u
n
ctio
n
,
th
e
L
SE
ca
n
b
e
r
ec
o
v
er
ed
3.
P
R
OPOS
E
D
M
E
T
HO
D
L
e
v
el
s
et
m
et
h
o
d
is
a
p
ar
t
o
f
th
e
ac
ti
v
e
co
n
to
u
r
f
a
m
il
y
.
L
e
v
el
s
et
m
et
h
o
d
s
h
a
v
e
s
h
o
w
n
t
r
e
m
en
d
o
u
s
r
esu
lt
s
f
o
r
m
ed
ical
i
m
a
g
e
s
e
g
m
en
tatio
n
.
T
h
e
m
ai
n
d
r
a
w
b
ac
k
s
o
f
L
e
v
el
s
et
m
et
h
o
d
s
ar
e
th
e
r
esu
lts
d
ep
en
d
s
o
n
th
e
i
n
itializa
tio
n
o
f
co
n
tr
o
lli
n
g
p
ar
a
m
eter
s
a
n
d
ti
m
e
co
m
p
le
x
it
y
.
I
n
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
w
e
o
v
er
co
m
e
t
h
ese
t
w
o
d
r
a
w
b
ac
k
s
.
T
o
in
itia
lize
th
e
co
n
tr
o
llin
g
p
ar
a
m
eter
we
u
s
ed
R
o
b
u
s
t
Sp
atial
Ker
n
e
l
Fu
zz
y
C
-
Me
a
n
s
(
R
SKF
C
M)
.
R
SKF
C
M
is
b
a
s
e
d
o
n
s
tan
d
ar
d
FC
M
alg
o
r
it
h
m
,
w
h
ic
h
i
n
co
r
p
o
r
ates
s
p
ati
al
in
f
o
r
m
at
io
n
a
n
d
u
s
e
s
Gau
s
s
ia
n
R
B
F
k
er
n
el
f
u
n
cti
o
n
.
T
o
o
v
er
co
m
e
t
h
e
s
ec
o
n
d
d
r
aw
b
ac
k
w
e
m
ak
e
u
s
e
o
f
L
attice
B
o
ltz
m
an
n
Me
th
o
d
(
L
B
M)
.
A
s
d
o
n
e
i
n
[
2
0
]
,
th
e
b
ias
f
ield
is
i
n
co
r
p
o
r
ate
d
in
to
t
h
e
R
S
KFC
M
f
r
a
m
e
w
o
r
k
b
y
m
o
d
elli
n
g
t
h
e
o
b
s
er
v
ed
i
m
ag
e
a
s
f
o
llo
w
s
:
{
1
,
2
,
.
.
.
,
}
i
i
i
Y
X
G
i
N
(
1
2
)
W
h
er
e
Y
i
,
X
i
,
an
d
G
i
ar
e
th
e
o
b
s
er
v
ed
in
te
n
s
it
y
,
tr
u
e
i
n
te
n
s
it
y
,
an
d
g
a
in
f
ield
at
th
e
i
th
p
ix
e
l,
r
esp
ec
tiv
el
y
.
N
is
th
e
to
tal
n
u
m
b
er
o
f
p
i
x
els
i
n
t
h
e
m
ag
n
etic
r
eso
n
a
n
ce
i
m
a
g
e.
T
h
e
ar
tef
ac
t
ca
n
b
e
m
o
d
elled
as
a
n
ad
d
iti
v
e
b
ia
s
f
ield
b
y
ap
p
l
y
i
n
g
a
lo
g
ar
ith
m
i
c
tr
an
s
f
o
r
m
atio
n
to
b
o
th
s
id
es o
f
eq
u
atio
n
(
1
2
)
[
2
0
]
,
{
1
,
2
,
.
.
.
,
}
i
i
i
y
x
i
N
(
1
3
)
W
h
er
e
y
i
an
d
x
i
ar
e
th
e
o
b
s
er
v
ed
an
d
tr
u
e
lo
g
-
tr
an
s
f
o
r
m
ed
in
ten
s
itie
s
at
t
h
e
i
th
v
o
x
el,
r
esp
ec
tiv
el
y
,
an
d
i
is
th
e
b
ias
f
ield
at
th
e
i
th
v
o
x
el.
B
y
in
co
r
p
o
r
atin
g
th
e
b
ia
s
f
iel
d
m
o
d
el
in
to
a
n
R
SKF
C
M
f
r
a
m
e
w
o
r
k
,
w
e
w
ill
b
e
ab
le
to
iter
ativ
el
y
esti
m
ate
b
o
th
th
e
tr
u
e
in
te
n
s
it
y
a
n
d
th
e
b
ias
f
ield
f
r
o
m
t
h
e
o
b
s
er
v
ed
in
ten
s
it
y
.
B
y
s
u
b
s
t
itu
t
in
g
e
q
u
atio
n
1
3
in
to
R
SKF
C
M
o
b
j
ec
tiv
e
f
u
n
c
tio
n
eq
u
atio
n
t
h
e
clu
s
ter
in
g
cr
iter
i
o
n
to
m
in
i
m
ize
in
t
h
e
p
r
esen
ce
o
f
b
ias f
ield
b
ec
o
m
e
s
a
co
n
s
tr
ain
ed
o
p
ti
m
izat
io
n
p
r
o
b
lem
2
11
(
,
,
,
)
|
|
|
|
cN
p
k
i
i
k
ki
J
W
V
B
Y
w
1
1
c
ki
k
wi
0
1
,
ki
w
k
i
(
1
4
)
w
h
er
e
1
{}
N
ii
Yy
is
t
h
e
o
b
s
er
v
ed
i
m
a
g
e
an
d
1
{}
N
ii
B
is
th
e
b
ias
f
ield
i
m
ag
e.
I
n
a
co
n
tin
u
o
u
s
f
o
r
m
,
th
e
a
f
o
r
e
m
en
tio
n
ed
cr
iter
io
n
ca
n
b
e
wr
itten
as
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
B
r
a
in
I
ma
g
e
S
eg
men
ta
tio
n
Usi
n
g
Leve
l S
et:
A
n
Hyb
r
id
A
p
p
r
o
a
ch
(
S
u
d
h
a
r
s
h
a
n
Du
t
h
)
263
2
1
(
,
,
,
)
(
,
)
|
|
(
,
)
|
|
k
c
p
kk
k
J
W
V
B
Y
W
x
y
Y
x
y
d
x
d
y
1
(
,
)
1
,
c
k
W
x
y
x
y
0
(
,
)
1
,
,
k
W
x
y
k
x
y
(
1
5
)
C
o
n
s
id
er
th
e
t
w
o
-
p
h
ase
le
v
el
s
et
alth
o
u
g
h
t
h
e
m
et
h
o
d
ca
n
b
e
ea
s
il
y
ex
te
n
d
ed
to
m
o
r
e
th
an
t
w
o
p
h
ases
.
T
h
e
i
m
a
g
e
d
o
m
ai
n
Ω
is
s
eg
m
e
n
ted
i
n
to
t
w
o
d
is
j
o
in
t
r
eg
io
n
s
Ω
1
a
n
d
Ω
2
,
i.e
.
,
c
=
2
.
I
n
th
is
ca
s
e,
w
e
ca
n
in
tr
o
d
u
ce
a
le
v
el
s
et
f
u
n
cti
o
n
as f
o
llo
w
s
:
2
11
2
22
(
,
,
,
,
)
(
,
)
|
|
(
,
)
(
,
)
|
|
(
)
(
,
)
|
|
(
,
)
(
,
)
|
|
(
1
(
)
)
p
p
J
W
V
B
Y
W
x
y
Y
x
y
B
x
y
H
d
x
d
y
W
x
y
Y
x
y
B
x
y
v
H
d
x
d
y
12
(
,
)
(
,
)
1
,
W
x
y
W
x
y
x
y
0
(
,
)
1
,
,
k
W
x
y
k
x
y
(
1
6
)
W
h
er
e
is
a
s
ig
n
ed
d
is
tan
t
f
u
n
c
tio
n
.
T
h
e
af
o
r
em
e
n
tio
n
ed
ter
m
J
(
U,
V
,
B
,
Y
,
)
is
u
s
ed
as
th
e
d
ata
lin
k
in
o
u
r
e
n
er
g
y
f
u
n
ctio
n
a
l w
h
i
c
h
is
d
ef
i
n
ed
as
f
o
llo
w
s
:
(
,
,
,
,
)
(
,
,
,
)
|
|
E
W
V
B
Y
J
W
V
B
Y
C
(
1
7
)
W
h
er
e
ν
|
C
|
is
a
r
eg
u
lar
izatio
n
ter
m
w
it
h
ν
>
0
b
ein
g
a
f
i
x
ed
p
ar
am
eter
an
d
C
b
ein
g
a
g
iv
en
c
u
r
v
e
w
h
ic
h
is
r
ep
r
esen
ted
i
m
p
licitl
y
a
s
th
e
ze
r
o
lev
el
o
f
φ
an
d
|
C
|
is
th
e
len
g
t
h
o
f
C
an
d
ca
n
b
e
ex
p
r
ess
ed
b
y
th
e
f
o
llo
w
in
g
eq
u
at
io
n
[
2
3
]
.
|
|
|
(
)
|
C
H
d
x
d
y
(
1
8
)
As
d
o
n
e
i
n
[
2
3
]
,
to
o
b
tain
L
S
E
,
w
e
m
i
n
i
m
ize
(
,
,
,
,
)
E
W
V
B
Y
w
it
h
r
esp
ec
t
to
.
Fo
r
f
ix
ed
U,
V
an
d
B
,
w
e
u
s
e
t
h
e
g
r
ad
ien
t d
es
ce
n
t
m
e
th
o
d
E
t
(
1
9
)
W
e
th
u
s
o
b
tai
n
th
r
ee
n
ec
e
s
s
ar
y
co
n
d
itio
n
s
,
*
2
(
1
)
1
1
(
,
)
|
|
(
,
)
(
,
)
|
|
|
|
(
,
)
(
,
)
|
|
k
p
c
k
l
l
W
x
y
Y
x
y
B
x
y
v
Y
x
y
B
x
y
v
(
2
0
)
*
(
,
)
(
(
,
)
(
,
)
)
(
,
)
p
k
k
p
k
W
x
y
Y
x
y
B
x
y
d
x
d
y
W
x
y
d
x
d
y
(
2
1
)
*
1
1
(
,
)
(
,
)
(
,
)
(
,
)
c
p
kk
k
c
p
k
k
W
x
y
v
B
x
y
Y
x
y
W
x
y
(
2
2
)
B
y
u
s
in
g
t
h
e
g
r
ad
ie
n
t
p
r
o
j
ec
tio
n
m
e
th
o
d
o
f
R
o
s
en
[
2
4
]
,
w
e
ca
n
r
ep
lace
δ
(
)
b
y
|
∇
|
i
n
th
e
p
r
o
p
o
s
ed
L
SE,
an
d
as
is
a
d
i
s
tan
ce
f
u
n
ctio
n
,
w
e
h
a
v
e
|
∇
|
=
1
E
q
u
a
tio
n
(
2
3
)
an
d
w
ill
s
t
a
y
at
ea
ch
s
tep
s
in
ce
a
n
ad
ap
tiv
e
ap
p
r
o
ac
h
is
n
o
t
u
s
ed
an
d
th
e
d
is
ta
n
t
f
ield
i
s
v
alid
i
n
th
e
w
h
o
le
d
o
m
ai
n
.
T
h
u
s
,
th
e
p
r
o
p
o
s
ed
L
SE
b
ec
o
m
e
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8814
IJ
AA
S
Vo
l.
6
,
No
.
3
,
Sep
tem
b
er
2
0
1
7
:
2
58
–
2
67
264
2
11
2
22
(
,
)
|
|
(
,
)
(
,
)
|
|
(
,
)
|
|
(
,
)
(
,
)
|
|
(
)
p
p
W
x
y
Y
x
y
B
x
y
t
W
x
y
Y
x
y
B
x
y
d
i
v
12
(
,
)
(
,
)
1
,
W
x
y
W
x
y
x
y
0
(
,
)
1
,
,
k
W
x
y
k
x
y
(2
3)
R
ep
lacin
g
b
y
t
h
e
s
i
g
n
ed
d
is
ta
n
ce
f
u
n
ctio
n
,
(
1
1
)
b
ec
o
m
es
()
d
i
v
F
t
(
2
4
)
B
y
s
etti
n
g
t
h
e
ex
ter
n
al
f
o
r
ce
22
1
1
2
2
(
(
,
)
|
|
(
,
)
(
,
)
|
|
(
,
)
|
|
(
,
)
(
,
)
|
|
)
pp
F
W
x
y
Y
x
y
B
x
y
v
W
x
y
Y
x
y
B
x
y
v
(
2
5
)
W
h
er
e
is
a
p
o
s
it
iv
e
p
ar
a
m
ete
r
;
w
e
ca
n
s
ee
th
a
t
E
q
u
atio
n
(
2
3
)
is
o
n
ly
a
v
ar
iati
o
n
f
o
r
m
u
la
o
f
E
q
u
atio
n
(
2
4
)
an
d
th
u
s
,
ca
n
b
e
s
o
l
v
ed
b
y
th
e
L
B
M
w
it
h
t
h
e
ab
o
v
e
d
ef
i
n
ed
FEF.
T
h
e
c
h
o
ice
o
f
p
a
r
a
m
eter
p
is
a
g
r
ea
t
i
m
p
o
r
tan
ce
f
o
r
t
h
e
s
eg
m
e
n
tati
o
n
r
esu
lt.
3
.
1
.
Alg
o
rit
h
m
T
h
e
co
m
p
u
ta
tio
n
al
s
tep
s
ar
e
as f
o
llo
w
s
a.
I
n
itialize
t
h
e
d
is
ta
n
ce
f
u
n
ctio
n
an
d
class
ce
n
tr
o
id
v
alu
e
s
v
1
an
d
v
2
.
I
n
itialize
B
w
ith
ze
r
o
s
.
b.
C
o
m
p
u
te
L
e
v
el
s
e
t f
u
n
c
tio
n
s
1
(
,
)
p
W
x
y
an
d
2
(
,
)
p
W
x
y
u
s
in
g
(
2
0
)
c.
C
o
m
p
u
te
v
1
an
d
v
2
w
it
h
(
2
1
)
d.
C
o
m
p
u
te
B
w
it
h
t
h
e
eq
u
atio
n
(
2
2
)
e.
C
o
m
p
u
te
th
e
e
x
ter
n
al
f
o
r
ce
u
s
in
g
(
2
5
)
f.
I
n
clu
d
e
t
h
e
ex
ter
n
al
f
o
r
ce
b
ase
d
o
n
(
2
5
)
g.
R
eso
l
v
e
th
e
co
n
v
ec
tio
n
-
d
i
f
f
u
s
i
o
n
eq
u
atio
n
w
it
h
L
B
M
w
it
h
(
3
)
h.
A
cc
u
m
u
la
te
(
,
)
i
f
r
t
v
alu
e
s
at
ea
ch
g
r
i
d
p
o
in
t
b
y
eq
u
atio
n
(
6
)
,
w
h
ic
h
g
e
n
er
ates
an
u
p
d
ated
d
is
tan
ce
v
alu
e
at
ea
c
h
p
o
in
t.
i.
Fin
d
t
h
e
co
n
to
u
r
j.
I
f
th
e
s
e
g
m
e
n
tatio
n
i
s
n
o
t d
o
n
e,
in
cr
ea
s
e
th
e
v
al
u
e
o
f
an
d
g
o
b
ac
k
to
s
tep
5
4.
E
X
PE
R
IM
E
N
T
A
L
R
E
S
U
L
T
S
T
h
is
s
ec
tio
n
e
v
al
u
ates
t
h
e
p
e
r
f
o
r
m
an
ce
o
f
t
h
e
p
r
o
p
o
s
ed
R
SKFC
M.
T
h
e
MRI
i
m
a
g
e
o
f
th
e
b
r
ain
ch
o
s
en
f
o
r
th
e
e
x
p
er
i
m
e
n
t
i
s
av
ailab
le
i
n
th
r
ee
b
an
d
s
:
T
1
-
w
ei
g
h
ted
,
p
r
o
to
n
d
en
s
it
y
(
p
d
)
-
w
ei
g
h
ted
an
d
T
2
-
w
ei
g
h
ted
.
T
h
e
n
o
r
m
al
b
r
ai
n
i
m
ag
e
s
ar
e
o
b
tain
ed
f
r
o
m
B
r
ain
w
eb
d
atab
ase
[
2
6
]
.
I
n
th
is
p
a
p
er
,
w
e
u
s
e
th
e
tr
an
s
v
er
s
al
s
lice
m
ap
,
th
e
s
li
ce
th
ick
n
es
s
is
1
m
m
a
n
d
th
e
s
ize
is
2
1
7
x
1
8
1
p
ix
els.
W
e
u
s
ed
th
e
s
a
m
e
w
ei
g
h
ti
n
g
e
x
p
o
n
en
t
m
=2
.
0
,
an
d
a
3
x
3
s
q
u
ar
e
o
f
th
e
b
r
ain
im
ag
e
p
i
x
els.
W
e
i
m
p
le
m
en
te
d
an
d
s
i
m
u
la
ted
all
th
e
alg
o
r
it
h
m
s
w
it
h
Ma
tlab
R
R
2
0
1
3
a.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
B
r
a
in
I
ma
g
e
S
eg
men
ta
tio
n
Usi
n
g
Leve
l S
et:
A
n
Hyb
r
id
A
p
p
r
o
a
ch
(
S
u
d
h
a
r
s
h
a
n
Du
t
h
)
2
65
Fig
u
r
e
1
.
I
n
itial
Se
g
m
e
n
tat
io
n
Usi
n
g
R
SKF
C
M
Fig
u
r
e
2
.
Fin
al
Se
g
m
en
tatio
n
Usi
n
g
P
r
o
p
o
s
ed
Me
th
o
d
Fig
u
r
e
1
s
h
o
w
s
th
e
i
n
itial
s
e
g
m
e
n
tat
io
n
r
es
u
lt
s
u
s
i
n
g
R
S
KFC
M
m
e
th
o
d
.
Fi
g
u
r
e
2
s
h
o
w
s
t
h
e
f
i
n
al
s
eg
m
e
n
tatio
n
r
es
u
lt
o
f
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
.
I
n
s
u
m
m
ar
y
,
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
g
iv
e
s
a
g
o
o
d
s
eg
m
e
n
tat
io
n
r
esu
lt.
Si
n
ce
,
R
SK
FC
M
i
n
co
r
p
o
r
ate
s
p
atial
in
f
o
r
m
atio
n
a
n
d
u
s
es
k
er
n
el
m
etr
ic,
it
s
h
o
w
s
less
s
u
s
ce
p
tib
le
to
d
if
f
er
e
n
t
t
y
p
es
o
f
n
o
i
s
e.
He
n
c
e
it
is
s
u
itab
le
to
i
n
itiate
lev
el
s
et
ev
o
l
u
tio
n
f
o
r
i
m
a
g
e
s
e
g
m
e
n
tatio
n
.
T
h
e
b
elo
w
T
ab
le
1
s
h
o
w
s
t
h
e
ac
h
ie
v
ed
C
P
U
tim
e
i
n
s
ec
o
n
d
s
f
o
r
th
e
p
r
o
p
o
s
ed
m
eth
o
d
w
ith
o
t
h
er
ex
i
s
tin
g
m
e
th
o
d
s
w
h
ich
p
r
o
v
es r
ed
u
ctio
n
in
t
i
m
e
co
m
p
lex
it
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8814
IJ
AA
S
Vo
l.
6
,
No
.
3
,
Sep
tem
b
er
2
0
1
7
:
2
58
–
2
67
266
T
ab
le
1
(
a)
.
Seg
m
e
n
tatio
n
r
esu
lts
o
f
t
h
e
C
V
m
et
h
o
d
(
b
)
Seg
m
en
tatio
n
r
es
u
lt
s
o
f
B
alla
et
al
.
,
(
c)
Seg
m
en
ta
tio
n
r
esu
lt
s
o
f
p
r
o
p
o
s
ed
m
et
h
o
d
.
M
e
t
h
o
d
s
I
mag
e
1
I
mag
e
2
I
mag
e
3
C
P
U
t
i
me
(
se
c
)
A
C
h
a
n
-
V
e
se
1
6
4
.
3
6
9
3
.
2
8
1
0
2
.
7
4
1
0
B
B
a
l
l
a
e
t
a
l
.
,
0
.
7
9
0
.
7
8
0
.
7
7
C
P
r
o
p
o
se
d
M
e
t
h
o
d
0
.
6
0
0
.
6
9
0
.
7
2
I
mag
e
d
i
me
n
s
i
o
n
s
4
8
3
X
3
2
1
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
w
e
p
r
o
p
o
s
ed
a
h
y
b
r
id
m
et
h
o
d
f
o
r
i
m
ag
e
s
eg
m
en
tatio
n
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
b
ased
o
n
lev
e
l
s
e
t
m
et
h
o
d
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
u
tili
ze
s
R
SKFC
M
al
g
o
r
ith
m
to
in
i
tial
ize
th
e
co
n
tr
o
lli
n
g
p
ar
am
eter
s
o
f
lev
e
l
s
et
a
n
d
L
B
M
m
e
th
o
d
to
r
ed
u
ce
th
e
ti
m
e
co
m
p
le
x
it
y
.
P
er
f
o
r
m
a
n
ce
ev
alu
a
tio
n
h
as
b
ee
n
ca
r
r
ied
o
u
t o
n
MRI
b
r
ain
i
m
a
g
es.
T
h
e
r
esu
lts
w
er
e
co
n
f
ir
m
e
d
p
r
o
m
is
i
n
g
ef
f
icie
n
t.
RE
F
E
R
E
NC
E
S
[1
]
K.
S.
C
h
u
an
g
,
H.
L
.
T
ze
n
g
,
S
.
C
h
e
n
,
J
.
W
u
an
d
T
.
J
.
C
h
en
.
,
“
F
u
zz
y
C
-
Me
an
s
C
l
u
s
ter
i
n
g
W
ith
Sp
atial
I
n
f
o
r
m
a
tio
n
f
o
r
I
m
a
g
e
Se
g
m
e
n
tatio
n
,
”
C
o
mp
u
teri
z
ed
Med
ica
l
I
ma
g
in
g
a
n
d
Gra
p
h
ics
,
Vo
l.
3
0
,
Pp
.
9
1
5
.
E
ls
ev
ier
.
2
0
0
6
.
[2
]
A
.
Hald
er
A
n
d
D.
K.
Ko
le.
,
“Au
to
m
a
tic
B
r
ain
T
u
m
o
r
Dete
ctio
n
a
n
d
I
s
o
latio
n
o
f
T
u
m
o
r
C
ells
Fro
m
Mr
i
I
m
ag
e
s
,
”
I
n
tern
a
tio
n
a
l J
o
u
r
n
a
l
o
f Co
mp
u
ter A
p
p
lica
tio
n
s
,
V
o
l.
3
9
,
N
o
.
2
,
2
0
1
2
.
[3
]
Ots
u
N,
“A
T
h
r
esh
o
ld
Select
i
o
n
Me
th
o
d
Fro
m
Gr
a
y
-
L
ev
el
His
to
g
r
a
m
s
,
”
I
ee
e
T
r
an
s
.
S
y
s
t
.
Ma
n
C
y
b
er
Net,
Vo
l.
9
,
P
p
.
6
2
-
6
6
,
1
9
7
9
.
[4
]
C
an
n
y
J
.
R
,
“A
C
o
m
p
u
tatio
n
al
A
p
p
r
o
ac
h
T
o
E
d
g
e
Dete
ctio
n
,
”
I
ee
e
Tr
a
n
s
a
ctio
n
s
o
n
P
a
tter
n
A
n
a
lysi
s
a
n
d
Ma
ch
in
ein
tellig
en
ce
,
Vo
l.
8
,
Pp
.
6
7
9
-
6
9
8
,
1
9
8
6
.
[5
]
P
h
a
m
D.
L
a
n
d
P
r
in
ce
J
.
L
,
“A
d
ap
tiv
e
F
u
zz
y
Seg
m
e
n
tati
o
n
o
f
Ma
g
n
et
ic
R
e
s
o
n
a
n
ce
I
m
ag
e
s
,
”
I
ee
e
Tr
a
n
s
a
ctio
n
On
med
ica
l I
ma
g
in
g
,
Vo
l.1
8
,
P
p
.
7
3
7
-
7
5
2
,
1
9
9
9
.
[6
]
Sali
m
a,
O.
,
T
aleb
-
Ah
m
ed
,
A
.
,
&
Mo
h
a
m
ed
,
B
,
“
Sp
atial
I
n
f
o
r
m
atio
n
B
ased
I
m
a
g
e
C
lu
s
t
er
in
g
W
it
h
A
S
w
ar
m
A
p
p
r
o
ac
h
.
”
I
a
es I
n
t.
J.
A
r
tif.
I
n
tell.
(
I
j
-
A
i)
,
1
(
3
)
,
1
4
9
-
1
6
0
,
2
0
1
2
.
[7
]
Z
h
o
u
H,
Sc
h
ae
f
er
G
An
d
Sh
i
C
,
“
A
Mea
n
S
h
ift
B
a
s
ed
F
u
z
z
y
C
-
Mea
n
s
A
lg
o
r
ith
m
F
o
r
I
ma
g
e
S
eg
men
ta
tio
n
,
”3
0
th
an
n
u
al
I
n
te
r
n
atio
n
al
C
o
n
f
er
en
ce
o
f
T
h
e
I
ee
e
E
n
g
i
n
ee
r
i
n
g
in
Me
d
icin
e
an
d
B
io
lo
g
y
So
ciet
y
,
Vo
l.1
,
P
p
.
3
0
9
1
-
3
0
9
4
,
2
0
0
8
.
[8
]
Z
a
n
at
y
E
.
A
,
A
lj
ah
d
ali
S,
an
d
Deb
n
ath
N.
,
“A
Ker
n
elize
d
Fu
zz
y
C
-
Me
an
s
A
l
g
o
r
ith
m
Fo
r
A
u
to
m
atic
Ma
g
n
e
tic
R
eso
n
an
ce
i
m
a
g
e
Seg
m
en
tatio
n
,
”
Jo
u
r
n
a
l
o
f
C
o
mp
u
ta
tio
n
a
l
Meth
o
d
s
in
S
cien
ce
a
n
d
E
n
g
in
ee
r
in
g
,
Vo
l
.
9
,
P
p
.
1
2
3
-
1
3
6
,
2
0
0
9
.
[9
]
L
i,
B
.
N.
,
C
h
u
i,
C
.
K.
,
C
h
a
n
g
,
S
an
d
On
g
,
S.
H.
,
“
I
n
te
g
r
ati
n
g
Sp
atia
l
Fu
zz
y
C
lu
s
ter
in
g
W
ith
L
e
v
el
Set
Me
th
o
d
s
Fo
r
au
to
m
ated
Me
d
ical
I
m
ag
e
Se
g
m
e
n
tat
io
n
,
”
C
o
mp
u
ters
in
B
io
lo
g
y
a
n
d
Med
icin
e
,
4
1
(
1
)
,
Pp
.
1
-
1
0
,
2
0
1
1
.
[1
0
]
G.
J
.
A
b
is
h
a
a
n
d
S
h
ij
i,
“
F
u
zz
y
C
l
u
s
ter
in
g
W
it
h
Sp
atial
I
n
f
o
r
m
atio
n
Fo
r
I
m
ag
e
Seg
m
e
n
tatio
n
Usi
n
g
Ker
n
el
Me
tr
ic,
”
I
n
tern
a
tio
n
a
l
Jo
u
r
n
a
l
Of
E
n
g
i
n
ee
r
in
g
R
ese
a
r
ch
A
n
d
A
p
p
lica
tio
n
s
(
I
jera
)
I
s
s
n
:
2
2
4
8
-
9
6
2
2
,
2
0
1
4
.
[1
1
]
A
r
u
n
a
Ku
m
ar
,
S.
V
a
n
d
Har
is
h
,
B
.
S.,
“
S
eg
men
tin
g
Mr
i
B
r
a
in
I
ma
g
es
Usi
n
g
N
o
ve
l
R
o
b
u
s
t
S
p
a
tia
l
K
er
n
el
F
cm
(
R
s
kfcm
)
,
”
I
n
I
n
t
er
n
atio
n
al
C
o
n
f
er
en
ce
o
n
E
i
g
h
th
I
n
ter
n
a
tio
n
al
C
o
n
f
er
en
ce
o
n
I
m
ag
e
a
n
d
Sig
n
al
P
r
o
ce
s
s
in
g
(
P
p
.
3
8
-
4
4
)
.
E
ls
ev
ier
,
2
0
1
4
.
[1
2
]
Sh
ar
m
a,
M.
,
&
M
u
k
h
er
j
ee
,
S,
“
F
u
zz
y
C
-
Me
a
n
s
,
An
f
is
An
d
Gen
et
ic
A
l
g
o
r
ith
m
F
o
r
Seg
m
e
n
ti
n
g
Ast
r
o
c
y
to
m
a
-
A
T
y
p
e
Of
B
r
ain
T
u
m
o
r
.
”
I
a
es
I
n
tern
a
tio
n
a
l
Jo
u
r
n
a
l
o
f
A
r
tifi
cia
l
I
n
tellig
en
ce
,
3
(
1
)
,
1
6
,
2014.
[1
3
]
Su
r
i,
J
.
S,
“
T
w
o
-
Di
m
e
n
s
io
n
al
Fas
t
Ma
g
n
etic
R
eso
n
a
n
ce
B
r
ain
Se
g
m
e
n
tat
io
n
,
”
E
n
g
in
ee
r
in
g
in
Med
icin
e
a
n
d
B
io
l
o
g
yma
g
a
z
in
e
,
I
ee
e,
2
0
(
4
)
,
8
4
-
9
5
,
2
0
0
1
.
[1
4
]
P
ar
ag
io
s
,
N,
“A
L
e
v
el
Set
A
p
p
r
o
ac
h
Fo
r
Sh
ap
e
-
Dr
iv
e
n
Seg
m
e
n
tatio
n
a
n
d
T
r
ac
k
in
g
o
f
T
h
e
L
e
f
t
Ven
tr
icle,
”M
ed
icali
m
a
g
in
g
,
I
ee
e
T
r
an
s
ac
tio
n
s
[1
5
]
Flex
ib
le,
E
x
te
n
s
ib
le
an
d
E
f
f
i
cien
t
T
o
o
lb
o
x
o
f
L
ev
e
l
Set
Me
th
o
d
s
,
”
Jo
u
r
n
a
l
o
f
S
cie
n
tifi
c
C
o
mp
u
tin
g
,
3
5
(
2
-
3
)
,
3
0
0
-
3
2
9
,
2
0
0
8
.
[1
6
]
C
.
L
i,
C
.
Xu
,
C
.
G
u
i,
O
n
,
2
2
(
6
)
,
7
7
3
-
7
7
6
,
2
0
0
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
B
r
a
in
I
ma
g
e
S
eg
men
ta
tio
n
Usi
n
g
Leve
l S
et:
A
n
Hyb
r
id
A
p
p
r
o
a
ch
(
S
u
d
h
a
r
s
h
a
n
Du
t
h
)
267
[1
7
]
Mitch
ell,
I
.
M.
,
“
T
h
e
An
d
m
.
Fo
x
,
“
D
is
ta
n
ce
R
e
g
u
lar
ized
L
e
v
el
Set
E
v
o
l
u
tio
n
a
n
d
its
A
p
p
licatio
n
T
o
I
m
ag
e
Se
g
m
e
n
tatio
n
,
”
I
ee
e
Tr
a
n
s
.
I
ma
g
e
P
r
o
ce
s
s
.
,
Vo
l.
1
9
,
No
.
1
2
,
P
p
.
3
2
4
3
3
2
5
4
,
Dec
.
2
0
1
0
.
[1
8
]
T
.
C
h
an
An
d
L
.
Ve
s
e,
“A
cti
v
e
C
o
n
to
u
r
s
W
ith
o
u
t E
d
g
e
s
,
”I
ee
e
T
r
an
s
.
I
m
a
g
e
P
r
o
ce
s
s
.
,
Vo
l.
1
0
,
No
.
2
,
P
p
.
2
6
6
2
7
7
,
Feb
.
2
0
0
1
.
[1
9
]
P
Su
d
h
ar
s
h
an
D
u
t
h
,
“A
Fas
t
a
n
d
R
o
b
u
s
t
L
ev
el
Set
Me
t
h
o
d
Fo
r
Me
d
ical
I
m
ag
e
Seg
m
en
tatio
n
,
”
I
n
tern
a
tio
n
a
l
Jo
u
r
n
a
lo
f
A
p
p
lied
E
n
g
in
ee
r
in
g
R
esea
r
ch
,
I
s
s
n
0
9
7
3
-
4
5
6
2
Vo
lu
m
e
1
0
,
Nu
m
b
er
1
1
,
2
0
1
5
.
[2
0
]
B
all
a
-
A
r
ab
,
S.,
Gao
,
X
an
d
W
an
g
,
B
,
“A
Fas
t
a
n
d
R
o
b
u
s
t
L
e
v
el
Set
Me
t
h
o
d
Fo
r
I
m
a
g
e
Seg
m
e
n
tatio
n
Usi
n
g
F
u
zz
y
cl
u
s
ter
in
g
a
n
d
L
a
ttice
B
o
ltz
m
a
n
n
Me
t
h
o
d
,
”
C
y
b
er
n
etics,
I
ee
e
T
r
an
s
ac
tio
n
s
O
n
,
4
3
(
3
)
,
9
1
0
-
9
2
0
,
2
0
1
3
.
[2
1
]
Y.
C
h
e
n
,
Z
.
Ya
n
,
an
d
Y
.
C
h
u
,
“
C
el
lu
lar
Au
to
m
ata
B
ased
L
ev
e
l
Set
Me
t
h
o
d
Fo
r
I
m
a
g
e
Seg
m
en
tatio
n
,
”I
n
p
r
o
c.
I
ee
e/I
c
m
e,
B
eij
in
g
,
C
h
i
n
a,
Ma
y
2
0
0
7
,
P
p
.
2
3
2
7
.
[2
2
]
W
.
C
h
en
An
d
M.
L
.
Gi
g
er
,
“
A
F
u
z
z
y
C
-
Mea
n
s
(
F
cm)
B
a
s
ed
A
lg
o
r
ith
m
F
o
r
I
n
ten
s
ity
I
n
h
o
mo
g
e
n
eity
C
o
r
r
ec
tio
n
a
n
d
S
eg
men
t
a
tio
n
o
f
Mr
I
ma
g
es
,
”I
n
P
r
o
c
.
I
ee
e
I
n
t.
S
y
m
p
.
B
io
m
e
d
.
I
m
a
g
i
n
g
:
Nan
o
Ma
cr
o
,
2
0
0
4
,
Vo
l.
2
,
Pp
.
1
3
0
7
1
3
1
0
.
[2
3
]
Y.
Z
h
ao
,
“L
attice
B
o
ltz
m
an
n
B
ased
P
d
e
So
lv
er
o
n
T
h
e
Gp
u
,
Vis
u
alco
m
p
u
t”
Vo
l.
2
4
,
No
.
5
,
P
p
.
3
2
3
3
3
3
,
Ma
r
.
2
0
0
7
.
[2
4
]
L
.
C
.
E
v
a
n
s
An
d
R
.
F.
Gar
i
ep
y
,
“
Me
a
s
u
r
e
T
h
eo
r
y
a
n
d
F
in
e
P
r
o
p
er
ties
Of
F
u
n
ctio
n
s
,
”B
o
ca
R
ato
n
,
Fl:C
r
c
P
r
ess
,
1
9
9
2
.
[2
5
]
J
.
G.
R
o
s
en
,
“
T
h
e
Gr
ad
ien
t
P
r
o
j
ec
tio
n
Me
th
o
d
Fo
r
No
n
li
n
ea
r
P
r
o
g
r
am
m
i
n
g
,
”
I
i,
No
n
li
n
ea
r
C
o
n
s
tr
ain
t
s
,
J
.
Siam
,
Vo
l.
9
,
No
.
4
,
P
p
.
5
1
4
5
3
2
,
Dec
.
1
9
6
1
.
[2
6
]
J
.
A
u
j
o
l
an
d
G.
Au
b
er
t,
“Sig
n
ed
Dis
ta
n
ce
F
u
n
c
tio
n
s
a
n
d
Vis
co
s
it
y
So
l
u
tio
n
s
o
f
Dis
co
n
ti
n
u
o
u
s
Ha
m
ilto
n
j
ac
o
b
ieq
u
atio
n
s
,
”I
n
r
i
a,
L
e
C
h
es
n
a
y
ce
d
e
x
,
Fra
n
ce
,
2
0
0
2
,
I
n
r
ia
-
0
0
0
7
2
0
8
1
,
Ver
s
io
n
1
,
R
ef
.
R
r
-
4507.
[2
7
]
Http
://B
r
ain
w
eb
.
B
ic.
Mn
i.M
c
g
ill.C
a/B
r
ain
w
eb
/
.
B
I
B
L
I
O
G
R
AP
H
Y
O
F
AUT
H
O
RS
P
Su
d
h
ar
s
h
a
n
D
u
t
h
:
r
ec
eiv
ed
h
is
B
.
E
d
eg
r
ee
in
I
n
f
o
r
m
at
io
n
Scien
ce
a
n
d
M.
T
ec
h
d
eg
r
ee
in
So
f
t
w
ar
e
E
n
g
in
ee
r
i
n
g
f
r
o
m
Vis
v
e
s
v
ar
a
y
a
T
ec
h
n
o
lo
g
ical
Un
i
v
er
s
it
y
.
He
is
c
u
r
r
en
tl
y
s
er
v
i
n
g
a
s
A
s
s
is
ta
n
t
P
r
o
f
es
s
o
r
in
C
o
m
p
u
ter
Sc
ie
n
ce
Dep
ar
t
m
e
n
t,
Am
r
ita
Vis
h
w
aVid
y
ap
ee
t
h
a
m
Un
i
v
er
s
it
y
,
M
y
s
u
r
u
C
a
m
p
u
s
,
M
y
s
u
r
u
.
His
ar
ea
o
f
r
esear
c
h
in
cl
u
d
es M
ac
h
i
n
e
L
ea
r
n
i
n
g
,
I
m
ag
e
P
r
o
ce
s
s
i
n
g
a
n
d
Data
Mi
n
in
g
.
Vi
m
al
Vis
w
an
at
h
:
i
s
cu
r
r
en
tl
y
p
u
r
s
u
i
n
g
h
i
s
M
C
A
d
eg
r
ee
in
Am
r
ita
Vi
s
h
w
a
Vid
y
ap
ee
th
a
m
U
n
i
v
er
s
it
y
,
M
y
s
u
r
u
C
a
m
p
u
s
,
M
y
s
u
r
u
.
His
ar
ea
o
f
r
esear
ch
in
clu
d
es
I
m
ag
e
P
r
o
ce
s
s
i
n
g
a
n
d
Data
Min
i
n
g
.
P
an
k
aj
Sre
ek
u
m
ar
:
is
cu
r
r
en
tl
y
p
u
r
s
u
in
g
h
is
M
C
A
d
eg
r
ee
in
Am
r
i
ta
Vis
h
w
a
Vid
y
ap
ee
th
a
m
U
n
i
v
er
s
it
y
,
M
y
s
u
r
u
C
a
m
p
u
s
,
M
y
s
u
r
u
.
His
ar
ea
o
f
r
esear
ch
in
clu
d
es
I
m
ag
e
P
r
o
ce
s
s
i
n
g
a
n
d
Data
Min
i
n
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.