I
n
t
ern
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
ces
i
n
A
p
p
l
i
ed
S
ci
en
ces
(
I
J
A
A
S
)
V
o
l
.
7
,
N
o
.
1
,
M
ar
ch
20
1
8
,
pp
.
78
~
85
I
S
S
N
:
225
2
-
88
14
,
D
OI
:
10.
115
91
/ij
a
a
s
.
v7
.
i
1
.
p
p
78
-
85
78
Jo
u
r
n
al
h
om
e
pa
ge
:
h
t
t
p
:
/
/
i
a
e
s
co
r
e
.
c
o
m/
o
n
l
i
n
e
/
i
n
d
e
x
.
p
h
p
/
I
J
A
A
S
SL
I
C
Supe
r
pi
xe
l
B
a
s
e
d Sel
f
O
r
ga
n
i
z
i
ng
M
aps
A
l
gori
t
hm f
or
Se
g
m
e
nt
a
t
i
o
n o
f
M
i
cr
oarr
ay
I
mag
es
D
ur
g
a
P
r
a
s
a
d
K
o
ndi
s
e
t
t
y
1
,
M
o
ha
m
m
e
d
A
l
i
H
us
s
a
i
n
2
,
1
D
e
pt
.
o
f
C
om
put
e
r
S
c
i
e
nc
e
,
B
ha
r
a
t
hi
ar
U
n
i
v
e
r
s
i
t
y
,
T
a
m
i
l
n
a
d
u
,
I
n
d
i
a
2
D
e
pt
.
o
f
C
om
put
e
r
S
c
i
e
nc
e
&
E
ngi
ne
e
r
i
ng
,
K
L
U
n
i
v
e
rs
i
t
y
,
Vi
j
a
y
a
w
a
d
a
,
A
P
,
I
n
d
i
a
A
rt
i
cl
e
I
n
f
o
A
BS
TR
A
C
T
Ar
t
i
c
l
e
h
i
s
t
o
r
y
:
R
ecei
v
ed
Ju
n
9
,
201
7
Re
v
i
s
e
d
J
a
n
13
,
201
8
A
ccep
t
e
d
F
e
b
16
,
2
01
8
W
e
c
a
n f
i
n
d t
he
s
im
ul
t
a
ne
o
us
m
oni
t
or
i
n
g of
t
h
ou
s
a
n
ds
o
f
ge
ne
s
i
n
p
ar
al
l
el
M
i
cr
o
ar
r
ay
t
ech
n
o
l
o
g
y
.
A
s
p
e
r
t
h
es
e m
ea
s
u
r
em
en
t
s
,
m
i
c
r
oa
r
r
a
y
t
e
c
hn
ol
o
gy
ha
ve
p
r
o
ve
n po
we
r
f
ul
i
n ge
ne
e
xp
r
e
s
s
i
o
n
pr
of
i
l
i
n
g f
or
di
s
c
o
ve
r
i
n
g
ne
w
t
y
pe
s
of
di
s
e
a
s
e
s
a
n
d f
or
p
r
e
d
i
c
t
i
ng t
he
t
y
pe
of
a
di
s
e
a
s
e
.
G
r
i
d
di
n
g,
I
nt
e
ns
i
t
y
e
xt
r
a
c
t
i
on,
E
n
ha
nc
e
m
e
nt
a
n
d
S
eg
m
en
t
at
i
o
n
ar
e i
m
p
o
r
t
an
t
s
t
ep
s
i
n
m
i
cr
o
ar
r
ay
i
m
ag
e an
al
y
s
i
s
.
T
h
i
s
p
a
p
e
r
g
i
v
e
s
s
i
m
p
l
e
l
i
n
e
a
r
i
t
e
r
a
t
i
v
e
c
l
u
s
t
e
r
i
n
g
(
S
L
I
C
)
b
a
s
e
d
s
e
l
f
or
ga
ni
z
i
ng m
a
ps
(
S
O
M
)
a
l
g
or
i
t
hm
f
or
s
e
g
m
e
nt
a
t
i
on o
f
m
i
c
ro
a
rra
y
i
m
ag
e.
T
h
e cl
u
s
t
er
s
o
f
p
i
x
el
s
w
h
i
c
h
s
h
ar
e s
i
m
i
l
ar
f
eat
u
r
es
ar
e cal
l
ed
S
u
pe
r
pi
xe
l
s
,
t
h
us
t
he
y
c
a
n
be
us
e
d a
s
m
i
d
-
l
ev
el
u
n
i
t
s
t
o
d
ecr
eas
e
t
h
e
c
om
put
a
t
i
ona
l
c
os
t
i
n m
a
ny
vi
s
i
on
a
p
pl
i
c
a
t
i
ons
.
T
h
e
p
r
o
p
os
e
d
a
l
g
o
r
i
t
h
m
u
t
i
l
i
z
e
s
s
u
p
e
r
p
i
x
e
l
s
a
s
c
l
u
s
t
e
r
i
n
g
o
b
j
e
c
t
s
i
n
s
t
e
a
d
o
f
p
i
x
e
l
s
.
T
he
qua
l
i
t
a
t
i
v
e
a
n
d q
ua
nt
i
t
a
t
i
ve
a
na
l
y
s
i
s
s
ho
w
s
t
ha
t
t
he
pr
o
pos
e
d
m
e
t
hod p
r
o
d
uc
e
s
be
t
t
e
r
s
e
gm
e
nt
a
t
i
on q
ua
l
i
t
y
t
ha
n k
-
m
ean
s
,
f
u
zzy
c
-
m
e
a
ns
a
n
d
s
e
l
f
o
r
ga
ni
z
i
n
g
m
a
ps
c
l
us
t
e
r
i
n
g
m
e
t
ho
ds
.
Ke
y
wo
r
d
:
E
m
pi
r
i
c
a
l
m
ode
de
c
om
pos
i
t
i
on
I
m
ag
e
S
e
g
m
e
n
t
a
t
i
o
n
M
i
cr
o
ar
r
ay
I
m
ag
es
S
e
l
f
or
ga
ni
z
i
ng
m
a
ps
S
i
m
p
l
e
l
i
n
e
a
r
i
t
e
r
a
t
i
v
e
c
l
u
s
t
e
r
i
n
g
Copy
r
i
ght
©
201
8
Ins
t
i
t
ut
e
o
f
A
d
v
anc
e
d
E
ngi
n
e
e
r
i
ng and S
c
i
e
nc
e
.
A
l
l
ri
g
h
t
s re
se
rv
e
d
.
C
or
r
e
s
po
n
di
n
g
A
u
t
h
or
:
D
ur
ga
P
r
a
s
a
d
K
ond
is
e
tty
,
R
es
ear
ch
S
c
h
o
l
ar
,
D
ep
t
.
o
f
C
o
m
p
u
t
er
S
ci
en
ce,
B
h
ar
at
h
i
ar
U
n
i
v
er
s
i
t
y
,
T
am
i
l
n
ad
u
,
I
n
d
i
a.
E
m
a
i
l
:
d
p
k
o
n
d
i
s
e
t
t
y
@
g
m
a
i
l
.
c
o
m
.
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
h
e m
o
s
t
p
o
w
er
f
u
l
t
o
o
l
i
n
m
o
l
ecu
l
ar
g
e
n
et
i
cs
f
o
r
b
i
o
m
ed
i
cal
r
es
ear
c
h
i
s
M
i
cr
o
ar
r
ay
,
w
h
i
ch
al
l
o
w
s
pa
r
a
l
l
e
l
a
na
l
y
s
i
s
of
t
he
e
x
p
r
e
s
s
i
on l
e
ve
l
o
f
t
ho
us
a
nds
o
f
ge
ne
s
.
T
he
m
os
t
i
m
por
t
a
nt
a
s
pe
c
t
i
n m
i
c
r
oa
r
r
a
y
e
xpe
r
i
m
e
nt
i
s
im
a
ge
a
na
l
y
s
i
s
.
T
he
a
na
l
y
s
i
s
of
o
ut
p
ut
o
f
i
m
a
g
e
i
s
a
m
a
t
r
i
x
c
o
n
s
i
s
t
i
n
g
o
f
i
n
t
e
n
s
i
t
y
m
e
a
s
u
r
e
o
f
e
a
c
h s
p
ot
i
n
t
he
i
m
a
ge
.
T
hi
s
i
s
de
n
ot
e
s
g
e
ne
e
x
pr
e
s
s
i
on
r
a
t
i
o
(
t
r
a
ns
c
r
i
pt
i
on
a
bu
n
da
n
c
e
)
be
t
w
e
e
n c
ont
r
o
l
s
a
m
pl
e
s
f
or
t
h
e
c
or
r
e
s
po
n
di
n
g ge
ne
a
nd t
he
ge
ne
t
e
s
t
.
T
h
e
ne
ga
t
i
ve
e
x
p
r
e
s
s
i
o
n i
n
di
c
a
t
e
s
u
nde
r
-
ex
p
r
e
s
s
ion
w
hi
l
e
pos
i
t
i
ve
e
xp
r
e
s
s
i
on i
n
d
i
c
a
t
e
s
t
he
o
ve
r
-
e
x
pr
e
s
s
i
o
n
be
t
w
e
e
n t
he
c
ont
r
ol
a
nd
t
r
e
a
t
m
e
nt
ge
ne
s
.
T
he
m
a
i
n
c
om
pone
nt
s
i
n
m
i
c
r
oa
r
r
a
y
i
m
a
ge
a
na
l
y
s
i
s
a
r
e
l
oc
a
l
i
z
a
t
i
on
,
s
e
gm
e
nt
a
t
ion a
n
d s
pot
q
ua
nt
i
f
i
c
a
t
i
o
n [
1]
.
T
h
e
m
ai
n
ap
p
l
i
cat
i
o
n
s
o
f
m
i
cr
o
ar
r
ay
t
ech
n
ol
o
gy
a
r
e
Ge
ne
di
s
c
o
ve
r
y
,
D
r
u
g di
s
c
o
ve
r
y
,
Di
s
e
a
s
e
di
a
g
nos
i
s
,
T
o
x
i
co
l
o
g
i
cal
r
es
ear
ch
et
c
[
2
]
.
T
h
e
m
i
cr
o
ar
r
a
y
i
m
ag
e
an
al
y
s
i
s
i
s
s
h
o
w
n
i
n
f
i
g
u
r
e
1
.
R
a
w
16 b
i
t
D
N
A
mi
cr
o
a
r
r
a
y
i
ma
g
e
I
m
a
g
e
E
nha
nc
e
m
e
n
t
Se
g
m
e
n
t
a
t
i
o
n
O
ut
pu
t
:
S
p
ot
M
et
r
i
cs
G
r
i
ddi
ng
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
A
S
I
S
S
N
:
225
2
-
88
14
SL
I
C
S
u
pe
r
pi
x
e
l
B
as
e
d
Se
l
f
O
r
ga
ni
z
i
n
g
M
a
p
s
A
l
g
or
i
t
h
m
f
or
S
e
g
m
e
n
ta
tion…
(
D
u
r
ga
P
r
as
a
d
K
ond
is
e
tty
)
79
F
i
g
u
r
e
1
:
A
n
a
l
y
s
i
s
o
f
M
i
c
ro
a
r
ra
y
Im
a
g
e
T
h
e
m
i
cr
o
ar
r
a
y
i
m
ag
es
i
s
a
d
i
f
f
i
cu
l
t
t
as
k
as
t
h
e
f
l
u
o
r
es
ce
n
c
e
o
f
t
he
gl
a
s
s
s
l
i
de
a
dds
n
oi
s
e
f
l
oo
r
t
o
t
he
m
i
cr
o
ar
r
ay
i
m
ag
e
[
3]
[
18]
.
T
he
pr
oc
e
s
s
i
n
g of
t
he
m
i
c
r
oa
r
r
a
y
im
a
ge
r
e
q
ui
r
e
s
n
oi
s
e
s
u
pp
r
e
s
s
i
on
w
i
t
h m
ini
m
a
l
r
e
d
uc
t
i
o
n
of
s
pot
e
d
ge
i
nf
or
m
a
t
i
on t
ha
t
de
r
i
ve
s
t
he
s
e
gm
e
nt
a
t
i
on
p
r
oc
e
s
s
.
T
hi
s
pa
p
e
r
d
es
c
r
i
b
es
r
ed
u
c
e
t
h
e
noi
s
e
i
n
m
i
cr
o
ar
r
ay
i
m
ag
es
u
s
i
n
g
E
m
p
i
r
i
c
a
l
M
ode
De
c
om
pos
i
t
i
o
n
[E
M
D
] m
e
t
h
o
d
.
T
he
B
E
M
D
m
e
t
ho
d
[5
]
de
c
om
pos
e
s
t
h
e
im
a
ge
i
nt
o s
e
ve
r
a
l
I
nt
r
i
ns
i
c
M
ode
F
unc
t
i
o
ns
[
I
M
F
]
,
i
n
whi
c
h t
he
f
i
r
s
t
f
u
nc
t
i
o
n i
s
t
h
e
hi
g
h
f
r
e
que
nc
y
c
om
po
ne
nt
,
s
e
c
o
nd
f
unc
t
i
o
n
ne
xt
hi
g
h
f
r
e
que
nc
y
c
o
m
pone
nt
a
n
d s
o
o
n;
t
he
l
a
s
t
f
u
nc
t
i
o
n
de
n
ot
e
s
t
he
l
ow
f
r
e
q
ue
nc
y
c
om
pone
n
t
.
T
he
m
e
a
n
f
i
l
t
e
r
i
s
a
ppl
i
e
d
onl
y
t
o t
h
e
f
e
w f
i
r
s
t
hi
g
h f
r
e
que
nc
y
c
om
p
one
nt
s
l
e
a
vi
ng t
he
l
o
w
f
r
e
q
ue
nc
y
c
om
pone
nt
s
,
a
s
t
he
hi
g
h f
r
e
q
ue
nc
y
c
om
pon
e
nt
s
c
o
nt
a
i
n n
oi
s
e
.
T
he
i
m
a
ge
i
s
r
eco
n
s
t
r
uc
t
e
d
by
c
om
bi
ni
ng
t
he
f
i
l
t
e
r
e
d
hi
g
h f
r
e
q
ue
nc
y
c
om
pone
nt
s
a
n
d l
o
w
f
r
e
q
ue
n
c
y
c
om
pone
nt
s
.
A
ft
e
r
noi
s
e
r
e
m
ova
l
,
s
e
gm
e
nt
a
t
i
on,
Ex
p
r
e
ssi
o
n
r
a
t
i
o
an
d
gr
idd
ing
c
a
l
c
u
l
a
t
i
o
n
s
a
r
e
t
he
i
m
por
t
a
nt
t
a
s
ks
i
n
a
n
a
l
y
s
i
s
o
f
m
i
cr
o
ar
r
a
y
i
m
a
g
e
.
A
ny
n
oi
s
e
i
n
t
he
m
i
c
r
oa
r
r
a
y
im
a
ge
w
i
l
l
a
f
f
e
c
t
t
h
e
s
u
b
s
e
q
u
e
n
t
a
n
a
l
y
s
i
s
[6
].
I
n t
he
pr
o
pos
e
d l
i
t
e
r
a
t
ur
e
o
f
m
a
ny
m
i
c
r
oa
r
r
a
y
i
m
a
ge
s
e
gm
e
nt
a
t
i
on a
p
pr
oa
c
he
s
ha
ve
F
i
x
ed
ci
r
cl
e
s
e
gm
e
nt
a
t
i
on [
7
]
,
A
da
pt
i
ve
c
i
r
cl
e S
eg
m
en
t
at
i
o
n
T
ec
h
n
i
q
u
e [
8
],
S
e
e
de
d r
e
gi
o
n
gr
ow
i
ng m
e
t
hods
[
9]
a
n
d
c
l
u
s
t
e
r
i
n
g
a
l
g
o
r
i
t
h
m
s
[
1
0
]
a
r
e
t
he
m
e
t
hods
t
ha
t
de
a
l
wi
t
h
m
i
c
r
oa
r
r
a
y
i
m
a
ge
s
e
gm
e
nt
a
t
i
on
p
r
o
bl
e
m
.
T
hi
s
pa
pe
r
m
ai
n
l
y
f
o
cu
s
es
o
n
cl
u
s
t
er
i
n
g
al
g
o
r
i
t
h
m
s
.
T
h
es
e al
g
o
r
i
t
h
m
s
h
av
e t
h
e a
d
v
an
t
ag
es
t
h
at
t
h
ey
ar
e n
o
t
r
es
t
r
i
ct
ed
t
o
a p
ar
t
i
cu
l
a
r
s
p
o
t
s
i
ze an
d
s
h
ap
e,
d
o
e
s
n
o
t
r
eq
u
i
r
e
a
n i
ni
t
i
a
l
s
t
a
t
e
of
pi
x
e
l
s
a
nd
n
o ne
e
d o
f
pos
t
pr
oc
e
s
s
i
ng
.
T
he
s
e
a
l
go
r
i
t
h
m
s
ha
ve
be
e
n
de
ve
l
ope
d
ba
s
e
d
on
t
he
i
nf
or
m
a
t
i
on a
b
out
t
he
i
nt
e
ns
i
t
i
e
s
o
f
t
he
pi
xe
l
s
o
nl
y
(
o
n
e
fe
a
t
u
re
).
In
t
h
i
s
p
a
p
e
r,
S
L
I
C
s
up
e
r
pi
xe
l
ba
s
e
d s
e
l
f
o
r
ga
ni
z
i
ng m
a
ps
c
l
us
t
e
r
i
ng a
l
go
r
i
t
hm
i
s
pr
o
p
os
e
d
.
T
h
e
q
u
a
l
i
t
a
t
i
v
e
a
n
d
q
u
a
n
t
i
t
a
t
i
v
e
r
e
s
u
l
t
s
s
h
o
w
t
h
a
t
p
r
op
os
e
d
m
e
t
h
od
h
as
s
eg
m
e
n
t
ed
t
h
e
i
m
ag
e
b
et
t
er
t
h
an
k
-
m
ean
s
,
f
u
zzy
c
-
m
e
a
ns
a
nd
s
e
l
f
o
r
ga
ni
z
i
ng
m
a
ps
c
l
us
t
e
r
i
ng
a
l
g
or
i
t
h
m
s
.
2.
BI
-
D
I
M
E
NS
I
O
NA
L
E
MP
I
R
I
C
A
L
M
O
DE
D
E
C
O
MP
O
S
I
T
I
O
N
-
D
W
T
TH
R
ES
H
O
LD
I
N
G
ME
T
H
O
D
E
m
pi
r
i
c
a
l
m
ode
de
c
om
pos
i
t
i
on
[1
1
]
i
s
a
s
i
gna
l
pr
oc
e
s
s
i
n
g m
e
t
hod t
ha
t
no
n
de
s
t
r
uc
t
i
ve
l
y
f
r
a
gm
e
nt
s
a
n
y no
n
-
l
i
ne
a
r
a
nd n
on
-
s
t
a
t
i
o
n
a
r
y
s
i
g
n
a
l
i
n
t
o
o
s
c
i
l
l
a
t
o
r
y
f
u
n
c
t
i
o
n
s
b
y
m
e
a
n
s
o
f
a
m
e
c
h
a
n
i
s
m
c
a
l
l
e
d
s
h
i
f
t
i
n
g
p
r
o
ces
s
.
T
h
es
e
f
u
n
ct
i
o
n
s
ar
e
cal
l
ed
I
n
t
r
i
n
s
i
c M
ode
F
u
nc
t
i
o
ns
(
I
M
F
)
,
a
nd
i
t
s
a
t
i
s
f
i
e
s
t
w
o
p
r
o
p
e
r
t
i
e
s
,
(
i
) t
h
e
num
be
r
o
f
z
e
r
o
c
r
os
s
i
ngs
a
nd
e
xt
r
e
m
a
poi
nt
s
s
ho
u
l
d
be
e
q
u
a
l
or
di
f
f
e
r
by
one
.
(
i
i
)
S
y
m
m
et
r
i
c
en
v
el
o
p
es
(
zer
o
m
e
a
n)
i
nt
e
r
pr
e
t
by
l
oc
a
l
m
a
xi
m
a
a
nd m
i
nim
a
[
1
2
]
.
T
he
s
i
gna
l
a
f
t
e
r
d
e
c
om
pos
i
t
i
on
us
i
n
g
E
MD
i
s
n
o
n
-
de
s
t
r
uc
t
i
ve
m
e
a
ns
t
ha
t
t
he
or
i
gi
na
l
s
i
g
na
l
c
a
n be
obt
a
i
ne
d
by
a
d
di
n
g t
he
I
M
F
s
a
n
d r
e
s
i
d
ue
.
T
he
f
i
r
s
t
I
M
F
i
s
a
hi
g
h
f
r
e
que
nc
y
c
om
pone
nt
a
n
d t
he
s
u
bs
e
q
ue
nt
I
M
F
s
c
o
nt
a
i
n f
r
om
ne
xt
hi
gh
f
r
e
q
ue
nc
y
t
o t
he
l
o
w
f
r
e
q
ue
nc
y
c
om
pone
nt
s
.
T
he
s
hi
f
t
i
n
g
pr
o
c
e
s
s
u
s
e
d
to
obta
in
I
MF
s
on
a
2
-
D
s
i
g
n
al
(
i
m
ag
e)
i
s
s
u
m
m
ar
i
zed
as
f
o
l
l
o
w
s
:
a)
L
et
I
(
x
,
y
)
b
e a M
i
cr
o
ar
r
ay
i
m
ag
e u
s
e
d
f
o
r
E
M
D
d
ec
o
m
p
o
s
i
t
i
o
n
.
F
i
n
d
al
l
l
o
cal
m
ax
i
m
a a
n
d
l
o
cal
m
i
n
i
m
a
p
o
i
n
ts
in
I
(
x,
y)
.
b)
U
p
pe
r
e
nve
l
op
e
U
p(
x,
y
)
i
s
c
r
e
a
t
e
d by
i
nt
e
r
pol
a
t
i
n
g t
h
e
m
a
xi
m
a
poi
n
ts
a
n
d low
e
r
e
nv
e
lo
p
e
L
w(
x,
y)
i
s
cr
eat
ed
b
y
i
n
t
e
r
p
o
l
a
t
i
n
g
m
i
n
i
m
a
p
o
i
n
t
s
.
T
h
e
c
u
b
i
c
s
p
l
i
n
e
i
n
t
e
r
p
o
l
a
t
i
o
n
m
e
t
h
o
d
fo
r
i
n
t
e
r
p
o
l
a
t
i
o
n
i
s
car
r
i
ed
out
a
s
:
c)
C
om
put
e
t
he
m
e
a
n
o
f
l
o
w
e
r
a
nd
u
p
pe
r
e
n
ve
l
ope
s
de
n
ot
e
d
by
M
e
a
n
(
x
,
y
)
.
(
(,
)
(,
)
)
(,
)
2
U
p
xy
L
w
xy
M
e
a
n
x
y
+
=
(
1)
d)
T
h
i
s
m
e
a
n
s
i
g
n
a
l
i
s
s
u
b
t
r
a
c
t
e
d
f
r
o
m
t
h
e
i
n
p
u
t
s
i
g
n
a
l
.
(,
)
(,
)
(,
)
Sub
x
y
I
x
y
M
e
an
x
y
=
−
(
2)
e)
I
f
S
u
b
(
x
,
y
)
s
a
t
i
s
f
i
e
s
t
h
e
I
M
F
p
r
o
p
e
r
t
i
e
s
,
t
h
e
n
a
n
I
M
F
i
s
o
b
t
a
i
n
e
d
.
(,
)
(,
)
i
IMF
x
y
S
u
b
x
y
=
(
3)
f)
S
u
b
t
r
a
c
t
t
h
e
e
x
t
r
a
c
t
e
d
I
M
F
f
r
o
m
t
h
e
i
n
p
u
t
s
i
g
n
a
l
.
N
o
w
t
h
e
v
a
l
u
e
o
f
I
(
x
,
y
)
i
s
(,
)
(,
)
(,
)
i
I
xy
I
xy
I
M
F
xy
=
−
(
4)
R
e
pe
a
t
t
he
a
b
o
ve
s
t
e
ps
(
b
)
t
o
(
f
)
f
o
r
t
he
ge
ne
r
a
t
i
on
o
f
ne
xt
I
M
F
s
.
g)
T
hi
s
pr
oc
e
s
s
i
s
r
e
pe
a
t
e
d
u
nt
i
l
I
(
x,
y
)
d
oe
s
n
ot
ha
ve
m
a
xim
a
or
m
i
nim
a
poi
nt
s
t
o
c
r
e
a
t
e
e
nv
e
l
ope
s
.
O
r
i
gi
na
l
I
m
a
ge
c
a
n
be
r
e
c
ons
t
r
uc
t
e
d
by
i
n
ve
r
s
e
E
M
D
gi
ve
n
by
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-
88
14
IJ
A
A
S
V
o
l
.
7
,
N
o
.
1
,
M
a
r
ch
2
018
:
7
8
–
85
80
1
(,
)
(,
)
(,
)
n
i
i
I
xy
I
M
F
xy
r
e
s
xy
=
=
+
∑
(
5)
T
he
m
e
c
ha
ni
s
m
of
de
-
n
o
i
s
i
n
g
u
s
i
n
g
BE
M
D
-
D
W
T
i
s
s
u
m
m
ar
i
zed
as
f
o
l
l
o
w
s
a.
A
pp
ly
2
-
D
E
M
D
f
o
r
n
o
i
s
y
m
i
c
ro
a
rra
y
t
o
o
b
t
a
i
n
IM
F
i
(i
=
1
,
2
,
…
k
).
T
h
e
k
t
h
IM
F
i
s
c
a
l
l
e
d
re
s
idu
e
.
b.
T
he
f
i
r
s
t
i
nt
r
i
ns
i
c
m
ode
f
u
n
c
t
i
on
(
I
M
F
1)
c
ont
a
i
n
s
hi
g
h
f
r
e
que
nc
y
c
o
m
pone
nt
s
a
n
d
i
t
i
s
s
ui
t
a
bl
e
f
o
r
de
n
oi
s
i
n
g.
T
hi
s
I
M
F
1
i
s
de
no
i
s
e
d
w
i
t
h
m
e
a
n
f
i
l
t
e
r
.
T
hi
s
de
-
noi
s
e
d
I
M
F
1
i
s
r
e
pr
e
s
e
nt
e
d
w
i
t
h
D
N
I
M
F
1
.
c.
T
he
de
noi
s
e
d
i
m
a
ge
i
s
r
e
c
on
s
t
r
uc
t
e
d
by
t
he
s
um
m
a
t
i
on
o
f
D
N
I
MF
1
a
n
d
r
e
m
a
i
ni
ng
I
M
F
s
gi
ve
n
by
2
1
k
i
i
R
I
D
N
IMF
IMF
=
=
+
∑
(
6)
w
he
r
e
R
I
i
s
t
he
r
e
c
ons
t
r
uc
t
e
d
ba
n
d
a
n
d
t
he
f
l
ow
di
a
gr
a
m
o
f
B
E
M
D
-
D
W
T
f
i
l
t
e
r
i
n
g
i
s
s
h
o
w
n
i
n
f
i
g
u
r
e
F
i
gure
2:
F
l
ow
D
i
a
gra
m
of B
E
M
D
-
m
e
a
n fi
l
t
e
ri
ng m
e
t
hod
3.
M
I
C
R
O
AR
R
A
Y
I
MA
G
E
G
R
I
D
D
I
N
G
T
he
p
r
oc
e
s
s
o
f
di
vi
di
ng t
he
m
i
c
r
oa
r
r
a
y
i
m
a
ge
i
nt
o bl
o
c
ks
(
s
ub
-
gr
i
d
d
i
ng
)
a
nd e
a
c
h
bl
oc
k a
ga
i
n
di
vi
de
d i
nt
o s
ub
-
bl
oc
ks
(
s
p
o
t
-
d
e
t
e
c
t
i
o
n
)
i
s
c
a
l
l
e
d
G
r
i
d
d
i
n
g
.
T
h
e
f
i
n
a
l
s
u
b
-
bl
oc
k c
ont
a
i
ns
a
s
i
ng
le
s
po
t a
nd
ha
vi
ng
o
nl
y
t
w
o
r
e
gi
ons
s
pot
a
nd
ba
c
k
gr
o
un
d.
E
xi
s
t
i
n
g a
l
g
or
i
t
hm
s
f
o
r
gr
i
ddi
ng
a
r
e
s
e
m
i
-
au
t
o
m
at
i
c i
n
n
at
u
r
e
r
e
q
ui
r
i
n
g
s
e
ve
r
a
l
pa
r
a
m
e
t
e
r
s
s
uc
h
a
s
s
i
z
e
o
f
s
pot
,
n
um
be
r
of
r
ows
o
f
s
p
ot
s
,
n
um
be
r
o
f
c
ol
um
ns
of
s
pot
e
t
c
.
I
n
t
h
i
s
p
a
p
e
r
,
a
f
u
l
l
y
a
u
t
o
m
a
t
i
c
gr
i
d
di
n
g
a
l
g
or
i
t
hm
de
s
i
gne
d
i
n
[
13]
i
s
us
e
d
f
o
r
s
ub
-
g
r
i
d
di
n
g
a
nd
s
pot
-
d
et
ect
i
o
n
.
.
4.
S
LI
C
S
U
P
E
R
P
I
X
ELS
S
i
m
p
l
e
l
i
n
e
a
r
i
t
e
r
a
t
i
v
e
c
l
u
s
t
e
r
i
n
g
(
S
L
I
C)
i
s
a
n
a
d
a
p
t
i
o
n
o
f
k
-
m
e
a
ns
f
o
r
S
u
p
e
r
pi
xe
l
ge
ne
r
a
t
i
on,
wi
t
h
t
wo
i
m
por
t
a
nt
di
s
t
i
nc
t
i
ons
:
i
)
t
he
n
um
be
r
of
d
i
s
t
a
n
c
e
c
a
l
c
u
l
a
t
i
o
n
s
i
n
t
h
e
o
p
t
i
m
i
z
a
t
i
o
n
i
s
d
r
a
m
a
t
i
c
a
l
l
y
r
e
d
u
c
e
d
by
l
im
i
t
i
ng
t
he
s
e
a
r
c
h
s
pa
c
e
t
o
a
r
e
gi
o
n
pr
o
po
r
t
i
ona
l
t
o
t
he
S
upe
r
pi
xe
l
s
i
z
e
.
T
hi
s
r
e
d
uc
e
s
t
h
e
c
om
pl
e
xi
t
y
t
o
be
l
i
ne
a
r
i
n
t
he
n
u
m
be
r
of
pi
xe
l
s
N
a
nd
i
n
de
pe
n
de
nt
o
f
t
he
n
u
m
be
r
of
s
u
pe
r
p
i
xe
l
s
k
.
i
i
)
A
w
e
i
g
h
t
e
d
d
i
s
t
a
n
c
e
m
e
a
s
ur
e
c
om
bi
ne
s
c
ol
o
r
a
nd
s
pa
t
i
a
l
pr
o
xi
m
i
ty
,
w
hi
l
e
s
i
m
ul
ta
ne
o
us
l
y
pr
ovi
di
n
g
c
ont
r
ol
o
v
e
r
t
he
s
i
z
e
a
nd
c
om
pa
c
t
ne
s
s
o
f
t
he
s
u
pe
r
pi
xe
l
s
.
T
he
a
l
go
r
i
t
hm
of
S
L
I
C
s
upe
r
pi
xe
l
s
ge
ne
r
a
t
i
on
i
s
gi
ve
n
be
l
ow
[
14]
.
1.
I
n
i
t
i
a
l
i
z
e
p
i
n
i
t
i
a
l
c
l
u
s
t
e
r
c
e
n
t
e
r
s
i
n
C =
[
k
,
x
,
y
,
r
,
s
]
T
b
y
s
a
m
p
l
i
n
g
p
i
x
e
l
s
a
t
r
e
g
u
l
a
r
g
r
i
d
s
t
e
p
s
S
.
2.
F
o
r
g
en
e
r
at
i
o
n
o
f
eq
u
al
s
i
zed
s
u
p
er
p
i
x
el
s
t
h
e
g
r
i
d
i
nt
e
r
v
a
l
S
i
s
gi
ve
n
by
S=
p
N
3.
S
e
t
l
a
b
e
l
k
(
j
)
=
-
1
f
o
r
e
a
c
h
pi
xe
l
j.
4.
4.
S
e
t
d
i
s
t
a
n
c
e
d
(
j
)
=
∞
fo
r
e
a
c
h
p
i
x
e
l
j
.
5.
F
o
r
eac
h
cl
u
s
t
e
r
ce
n
t
er
C
d
o
6.
F
or
e
a
c
h
pi
xe
l
j
i
n
a
2S
X
2S
r
e
gi
on
a
r
ou
n
d
C
d
o
7.
C
o
m
p
u
t
e
t
h
e
d
i
s
t
an
ce
D
b
et
w
een
C
a
n
d
j
.
8.
T
he
di
s
t
a
nc
e
D
de
pe
n
ds
on
pi
xe
l
’
s
c
ol
o
r
(
c
ol
o
r
pr
o
xi
m
ity
)
a
nd
pi
xe
l
p
os
i
t
i
on
(
s
pa
t
i
a
l
pr
oxi
m
i
ty
)
,
w
ho
s
e
va
l
ue
s
i
s
k
n
ow
n.
T
he
va
l
ue
of
D
i
s
gi
ve
n
by
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
A
S
I
S
S
N
:
225
2
-
88
14
SL
I
C
S
u
pe
r
pi
x
e
l
B
as
e
d
Se
l
f
O
r
ga
ni
z
i
n
g
M
a
p
s
A
l
g
or
i
t
h
m
f
or
S
e
g
m
e
n
ta
tion…
(
D
u
r
ga
P
r
as
a
d
K
ond
is
e
tty
)
81
(
7)
T
h
e
m
a
x
i
m
u
m
s
p
a
t
i
a
l
di
s
t
a
nc
e
e
xpe
c
t
e
d w
i
t
hi
n a
gi
ve
n c
l
us
t
e
r
s
h
o
ul
d c
or
r
e
s
po
n
d t
o t
he
s
a
m
pl
i
ng
i
n
t
e
r
v
a
l
,
N
S
=
S
.
D
e
t
e
r
m
i
n
i
n
g
t
h
e
m
a
x
i
m
u
m
c
o
l
o
r
d
i
s
t
a
n
c
e
N
c
i
s
n
o
t
s
o
s
t
r
a
i
g
h
t
f
o
r
w
a
r
d
,
a
s
c
o
l
o
r
d
i
s
t
a
n
c
e
s
c
a
n
va
r
y
s
i
gni
f
i
c
a
nt
l
y
f
r
om
c
l
us
t
e
r
t
o
c
l
u
s
t
e
r
a
nd
i
m
a
ge
t
o
i
m
a
ge
.
T
he
va
l
ue
of
N
c
i
n
t
he
r
a
nge
f
r
om
[
1,
4
0]
.
9.
i
f
D
<
d
(i
)
t
h
e
n
s
e
t
d
(i
)=
D
a
n
d
k
(i
)=
p
g
o
t
o
6
.
1
0.
G
o
t
o
5
,
t
h
e
s
a
m
e
p
r
o
ces
s
f
o
r
each
cl
u
s
t
er
1
1.
C
om
put
e
ne
w
c
l
us
t
e
r
c
e
nt
e
r
s
.
1
2.
T
h
e
cl
u
s
t
er
i
n
g
an
d
u
p
d
at
i
n
g
p
r
o
ces
s
es
a
r
e
r
e
p
eat
ed
u
n
t
i
l
a
p
r
e
d
ef
i
n
ed
n
u
m
b
er
o
f
i
t
er
at
i
o
n
i
s
ac
h
i
ev
e
d
.
T
h
e
S
L
I
C
a
l
g
or
i
t
h
m
c
a
n ge
ne
r
a
t
e
c
om
pa
c
t
a
nd
ne
a
r
l
y
u
ni
f
o
r
m
s
upe
r
pi
x
e
l
s
w
i
t
h a
l
ow
c
om
put
a
t
i
o
na
l
ove
r
he
a
d
.
5.
F
U
ZZY
C
-
M
E
AN
S
CL
U
S
T
E
R
I
NG
AL
G
O
R
I
T
H
M
T
h
e
F
CM
a
l
g
o
r
i
t
h
m
f
o
r
s
e
g
m
e
n
t
a
t
i
o
n
o
f
m
i
c
r
o
a
r
r
a
y
i
m
a
ge
i
s
de
s
c
r
i
be
d
be
l
ow
[
15
]:
1.
T
a
ke
r
a
n
d
om
l
y
K
i
ni
t
i
a
l
c
l
us
t
e
r
s
f
r
om
t
he
m
*n
i
m
a
ge
pi
xe
l
s
.
2.
I
n
i
t
i
a
l
i
z
e
m
e
m
b
e
r
s
h
i
p
m
a
t
r
i
x
u
ij
w
i
t
h
va
l
ue
i
n
r
a
n
ge
0
t
o
1
a
nd
va
l
ue
of
m
=
2.
A
s
s
i
g
n
e
a
c
h
p
i
x
e
l
t
o
t
h
e
c
l
u
s
t
e
r
Cj
{
j
=
1
,
2
,
…
.
.
K
}
i
f
i
t
s
a
t
i
s
f
i
e
s
t
h
e
f
o
l
l
o
w
i
n
g
c
o
n
d
i
t
i
o
n
[
D
(
.
,
.
)
]
i
s
t
h
e
E
u
cl
i
d
ea
n
d
i
s
t
an
ce
m
eas
u
r
e
b
et
w
een
t
w
o
v
al
u
es
.
(
,
)
(
,
)
,
1
,
2
,
...,
mm
ij
i
j
iq
i
q
u
DI
C
u
DI
C
q
K
jq
<=
≠
(
8)
T
h
e
n
e
w
m
e
m
b
er
s
h
i
p
an
d
cl
u
s
t
er
cen
t
r
o
i
d
v
al
u
es
as
cal
c
u
l
at
ed
as
1
1
1
1
,1
(,
)
()
(
,)
ik
K
ik
m
j
jk
u
f
or
i
K
DC
I
DC
I
−
=
=
≤≤
∑
1
^
1
n
m
ij
j
j
j
n
m
ij
j
uI
C
u
=
=
=
∑
∑
(
9)
3.
Co
n
t
i
n
u
e
2
-
3
u
nt
i
l
e
a
c
h
pi
xe
l
i
s
a
s
s
i
g
ne
d
t
o
t
he
m
a
xim
u
m
m
e
m
be
r
s
hi
p
c
l
us
t
e
r
[
16]
.
6.
S
LI
C
S
U
P
E
R
P
I
X
EL
BA
S
E
D
SO
M
C
L
U
ST
E
R
I
N
G
A
L
G
O
R
I
T
H
M
T
he
S
L
I
C
a
l
go
r
i
t
hm
ge
ne
r
a
t
e
s
s
upe
r
pi
xe
l
s
w
hi
c
h a
r
e
us
e
d i
n o
ur
c
l
us
t
e
r
i
n
g a
l
g
or
i
t
hm
.
T
he
s
up
e
r
pi
xe
l
s
a
r
e
ge
n
e
r
a
t
e
d
ba
s
e
d
on t
he
c
ol
or
s
i
m
i
l
a
r
i
ty
a
nd
p
r
o
xi
m
it
y
in t
he
i
m
a
ge
p
l
a
ne
.
T
he
a
l
g
o
r
i
t
h
m
de
pe
n
ds
o
n
t
w
o
va
l
ue
s
N
S
a
n
d N
c
,
t
he
hi
g
he
r
va
l
ue
of
N
S
c
o
r
r
e
s
p
on
ds
t
o m
or
e
r
e
gul
a
r
a
nd
gr
i
d
-
l
i
k
e
S
up
e
r
p
i
x
el
s
t
r
u
ct
u
r
e an
d
l
o
w
er
v
al
u
e o
f
N
c
cap
t
u
r
es
m
o
r
e i
m
ag
e d
et
ai
l
s
.
T
h
e S
L
I
C
S
u
p
er
p
i
x
el
b
as
e
d
S
O
M
c
l
u
s
t
e
r
i
n
g
a
l
g
o
r
i
t
h
m
i
s
g
i
v
e
n
b
e
l
o
w
:
1.
C
ol
l
e
c
t
ne
c
e
s
s
a
r
y
i
nf
o
r
m
a
ti
on
of
s
u
pe
r
pi
xe
l
s
by
ge
ne
r
a
t
e
t
he
s
u
pe
r
pi
xe
l
s
r
e
p
r
e
s
e
nt
a
t
i
on
of
o
r
i
gi
na
l
i
m
a
ge
.
2.
I
n
i
t
i
a
l
i
z
e
c
l
u
s
t
e
r
c
e
nt
r
oi
ds
v
i
,
i
=
1
,
.
.
.
,
C
.
3.
T
he
o
bje
c
t
i
ve
f
unc
t
i
o
n
F
i
s
gi
ve
n
by
(
10
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-
88
14
IJ
A
A
S
V
o
l
.
7
,
N
o
.
1
,
M
a
r
ch
2
018
:
7
8
–
85
82
4.
T
he
m
e
m
be
r
s
h
i
p
va
l
ue
s
ui
j
i
s
u
pd
a
t
e
d
gi
ve
n
by
(
11
)
Wh
e
re
γ
j
i
s
t
h
e
num
be
r
of
pi
xe
l
s
i
n
s
upe
r
pi
xe
l
s
j
,
u
ij
d
e
n
ot
e
s
t
he
m
e
m
be
r
s
hi
p
o
f
s
u
pe
r
pi
xe
l
s
j
t
o
t
he
i
t
h
c
l
u
s
t
e
r
.
Q
i
s
t
h
e
n
um
be
r
of
s
u
pe
r
pi
xe
l
s
i
n
i
m
a
ge
s
a
nd
ξ
j
i
s
t
h
e
av
er
a
g
e
co
l
o
r
va
l
ue
o
f
s
upe
r
pi
xe
l
s
j
,
N
j
s
t
a
n
d
s
f
o
r
t
h
e
s
e
t
o
f
ne
i
g
hb
or
i
ng
s
u
pe
r
pi
x
e
l
s
t
ha
t
a
r
e
a
d
j
a
c
e
nt
t
o
sj
a
nd
NR
i
s
t
h
e
c
a
r
d
i
n
a
l
i
t
y
o
f
N
j
.
|
|
·
|
|
is
a
n
or
m
m
e
tr
ic
,
de
n
ot
i
ng
E
uc
l
i
de
a
n
di
s
t
a
nc
e
b
e
t
w
e
e
n
p
i
x
e
l
s
a
n
d
c
l
u
s
t
e
r
i
n
g
c
e
nt
r
oi
ds
.
T
he
pa
r
a
m
e
t
e
r
m
i
s
a
we
i
ght
i
ng e
x
po
ne
nt
o
n e
a
c
h
S
O
M
m
e
m
be
r
s
hi
p a
nd
de
t
e
r
m
i
ne
s
t
he
a
m
ount
o
f
s
e
l
f
m
a
p
p
i
n
g
o
f
t
h
e
r
e
s
u
l
t
i
n
g
c
l
a
s
s
i
f
i
c
a
t
i
o
n
.
5.
T
he
c
l
us
t
e
r
c
e
nt
r
oi
ds
vi
i
s
u
p
da
t
e
d
gi
ve
n
by
(
12
)
6.
R
e
pe
a
t
s
S
t
e
ps
3
t
o
4
,
un
til
|
|
V
n
e
w
-
Vo
ld
|
|
<
ε
.
7.
E
X
P
ER
I
M
EN
TA
L
R
ES
U
L
TS
Q
u
a
n
t
i
t
a
t
i
v
e
A
n
a
l
y
s
i
s
:
Q
u
a
n
t
i
t
a
t
i
v
e
a
n
a
l
y
s
i
s
i
s
a
n
u
m
e
r
i
c
a
l
l
y
o
r
i
e
n
t
e
d
p
r
o
c
e
d
u
r
e
t
o
f
i
g
u
r
e
o
u
t
t
h
e
pe
r
f
o
r
m
a
nc
e
of
a
l
gor
i
t
hm
s
w
i
tho
ut
a
ny
h
um
a
n
e
r
r
or
.
T
he
M
e
a
n
S
q
ua
r
e
E
r
r
or
(
M
S
E
)
[
1
7]
i
s
s
i
g
n
i
f
i
c
a
n
t
m
e
t
r
i
c
t
o va
l
i
da
t
e
t
he
qua
l
i
t
y
o
f
i
m
a
ge
.
I
t
m
e
a
s
ur
e
s
t
he
s
qua
r
e
e
r
r
or
be
t
w
e
e
n
pi
x
e
l
s
of
t
he
or
i
gi
na
l
a
n
d t
he
r
e
s
ul
t
a
nt
i
m
ag
es
.
Q
u
a
l
i
t
a
t
i
v
e
A
n
a
l
y
s
i
s
:
T
he
pr
op
os
e
d c
l
us
t
e
r
i
ng a
l
g
or
i
t
hm
i
s
pe
r
f
or
m
e
d on t
w
o m
i
c
r
oa
r
r
a
y
im
a
ge
s
dr
a
w
n
f
r
om
t
h
e
s
t
a
nda
r
d
m
i
c
r
o
a
r
r
ay
d
at
a
b
a
s
e
co
r
r
es
p
o
n
d
s
t
o
b
r
eas
t
cat
e
g
o
ry
a
C
GH
t
u
m
o
r
t
i
s
s
u
e
[1
8
]
.
I
m
a
g
e
1 c
o
ns
i
s
t
s
of
a
t
ot
a
l
of
38
80
8 pi
xe
l
s
a
n
d I
m
a
ge
2 c
ons
i
s
t
s
of
64
8
80
pi
xe
l
s
.
Gr
i
ddi
n
g i
s
pe
r
f
o
r
m
e
d on t
he
i
np
ut
im
a
ge
s
by
t
he
m
e
t
hod p
r
o
p
os
e
d i
n
[
1
3]
,
t
o s
e
g
m
e
nt
t
he
im
a
ge
i
n
t
o
co
m
p
ar
t
m
en
t
s
,
w
h
er
e each
c
om
pa
r
tm
e
nt
is
ha
vi
ng
onl
y
one
s
p
ot
r
e
gi
o
n
a
n
d
ba
c
k
gr
o
un
d.
T
he
g
r
i
d
d
i
ng
o
ut
put
i
s
s
ho
w
n
i
n
f
i
g
u
r
e
3.
A
f
t
e
r
gr
i
ddi
ng t
he
i
m
a
ge
i
nt
o c
o
m
pa
r
tm
e
nt
s
,
s
uc
h t
ha
t
e
a
c
h
c
om
pa
r
tm
e
nt
i
s
ha
vi
n
g s
i
n
gl
e
s
p
ot
a
n
d
ba
c
kg
r
ou
n
d,
c
om
pa
r
tm
e
nt
no
1 f
r
om
im
a
ge
1 a
n
d c
o
m
pa
r
tm
e
nt
no 12
f
r
om
im
a
ge
2 a
r
e
e
xt
r
a
c
t
e
d.
S
upe
r
pi
x
e
l
s
a
r
e
ge
ne
r
a
t
e
d
f
o
r
t
he
s
e
t
w
o c
om
pa
r
t
m
e
nt
s
us
i
n
g S
L
I
C
a
nd s
e
gm
e
nt
e
d us
i
n
g
S
L
I
C
ba
s
e
d S
O
M
a
l
go
r
i
t
hm
.
T
he
S
u
pe
r
pi
xe
l
ge
n
e
r
a
t
i
on
a
n
d
s
e
g
m
e
nt
a
t
i
on
i
s
s
h
ow
n
i
n
f
i
g
ur
e
3
.
Th
e
M
S
E
i
s
[
1
8
]
m
a
t
h
e
m
a
t
i
c
a
l
l
y
d
e
f
i
n
e
d
a
s
MS
E
=
|
|
v
i
-
c
j|
|
2
(
13
)
W
he
r
e
N
i
s
t
h
e
t
ot
a
l
num
be
r
of
pi
xe
l
s
i
n a
n i
m
a
ge
a
nd
x
i
i
s
t
he
pi
xe
l
w
hi
c
h be
l
on
gs
t
o t
he
jt
h
c
l
u
s
t
e
r
.
T
h
e
l
o
w
e
r
d
i
f
f
e
r
e
n
ce b
et
w
een
t
h
e r
es
u
l
t
an
t
an
d
t
h
e o
r
i
g
i
n
al
i
m
ag
e r
ef
l
ect
s
t
h
at
al
l
t
h
e d
at
a i
n
t
h
e r
eg
i
o
n
ar
e
l
oc
a
t
e
d ne
a
r
t
o i
t
s
c
e
nt
r
e
.
T
a
bl
e
1 s
ho
w
s
t
he
q
ua
nt
i
t
a
t
i
ve
e
va
l
ua
t
i
o
ns
o
f
c
l
us
t
e
r
i
ng a
l
go
r
i
t
hm
s
.
T
he
r
e
s
ul
t
s
c
o
n
f
i
r
m
t
h
a
t
S
L
I
C b
a
s
e
d
S
O
M
a
l
gor
i
t
hm
pr
o
d
uc
e
s
t
he
l
o
w
e
s
t
M
S
E
va
l
ue
f
or
s
e
gm
e
nt
i
ng t
he
m
i
c
r
oa
r
r
a
y
i
m
ag
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
A
S
I
S
S
N
:
225
2
-
88
14
SL
I
C
S
u
pe
r
pi
x
e
l
B
as
e
d
Se
l
f
O
r
ga
ni
z
i
n
g
M
a
p
s
A
l
g
or
i
t
h
m
f
or
S
e
g
m
e
n
ta
tion…
(
D
u
r
ga
P
r
as
a
d
K
ond
is
e
tty
)
83
I
m
ag
e
1
G
r
i
dde
d
I
m
a
ge
I
m
ag
e
2
G
r
i
dde
d
I
m
a
ge
C
om
pa
r
tm
e
nt
N
o
1
i
n
i
m
a
ge
1
S
u
pe
r
pi
xe
l
s
C
om
pa
r
tm
e
nt
N
o
1
2
i
n
i
m
ag
e
2
S
u
pe
r
pi
xe
l
s
S
eg
m
en
t
ed
I
m
ag
e
C
om
pa
r
tm
e
nt
N
o
1
S
eg
m
en
t
ed
I
m
ag
e
C
om
pa
r
tm
e
nt
N
o
1
2
F
i
gu
r
e
3:
S
u
pe
r
pi
xe
l
ba
s
e
d
S
O
M
s
e
gm
e
nt
a
ti
on
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-
88
14
IJ
A
A
S
V
o
l
.
7
,
N
o
.
1
,
M
a
r
ch
2
018
:
7
8
–
85
84
T
a
b
l
e
1
:
M
S
E
V
al
u
es
M
e
t
hod
C
o
m
pa
r
t
m
e
nt
No
1
C
o
m
pa
r
t
m
e
nt
N
o 12
K
-
me
a
n
s
96.
4
93.
6
Fu
zzy
c
-
me
a
n
s
93.
1
89.
4
Self
o
r
g
an
izin
g
m
a
p
s
84.
7
80.
9
S
LIC
S
O
M
82.
8
77.
4
8.
C
O
N
L
U
SI
O
N
S
M
i
c
r
oa
r
r
a
y
t
e
c
hn
ol
o
gy
i
s
us
e
d f
or
pa
r
a
l
l
e
l
a
na
l
y
s
i
s
of
ge
n
e
e
xp
r
e
s
s
i
o
n r
a
t
i
o of
di
f
f
e
r
e
nt
ge
ne
s
i
n
a
s
i
ngl
e
e
x
pe
r
i
m
e
nt
.
T
he
a
na
l
y
s
i
s
of
m
i
c
r
oa
r
r
a
y
im
a
ge
i
s
done
wi
t
h s
e
gm
e
nt
a
t
i
on,
i
nf
or
m
a
t
i
on e
xt
r
a
c
t
i
o
n a
n
d
gr
i
ddi
ng
.
T
he
t
r
a
ns
c
r
i
pt
i
o
n a
bu
n
da
nc
e
be
t
w
e
e
n t
w
o
ge
ne
s
un
de
r
e
x
pe
r
i
m
e
nt
i
s
t
he
e
x
pr
e
s
s
i
o
n r
a
t
i
o
of
e
a
c
h
a
nd
e
ve
r
y
ge
n
e
s
p
ot
.
C
l
us
t
e
r
i
n
g a
l
g
or
i
t
hm
s
ha
ve
be
e
n
u
s
e
d
f
o
r
m
i
c
r
oa
r
r
a
y
i
m
a
ge
s
e
gm
e
nt
a
t
i
on w
i
t
h a
n
ad
v
a
n
t
ag
e t
h
at
t
h
ey
ar
e n
o
t
r
es
t
r
i
ct
ed
t
o
a p
ar
t
i
cu
l
ar
s
i
ze an
d
s
h
a
p
e f
o
r
t
h
e s
p
o
t
s
.
T
h
i
s
p
ap
e
r
d
es
c
r
i
b
e
s S
LI
C
ba
s
e
d s
e
l
f
or
g
a
ni
z
i
ng
m
a
ps
c
l
us
t
e
r
i
n
g a
l
g
o
r
i
t
hm
f
or
s
e
g
m
e
nt
a
t
i
on o
f
m
i
c
r
oa
r
r
a
y
i
m
a
ge
.
S
p
ot
i
nf
o
r
m
a
ti
on
i
nc
l
ude
s
t
he
c
a
l
c
ul
a
t
i
on o
f
E
x
pr
e
s
s
i
o
n R
a
t
i
o i
n t
he
r
e
gi
o
n
of
e
ve
r
y
ge
ne
s
pot
on t
he
m
ic
r
oa
r
r
a
y
i
m
a
ge
.
T
h
e
ex
p
r
es
s
i
o
n
-
r
at
i
o
m
eas
u
r
es
t
h
e
t
r
an
s
c
r
i
pt
i
o
n a
bu
n
da
nc
e
be
t
w
e
e
n t
he
t
wo
s
a
m
pl
e
ge
ne
s
.
T
he
p
r
op
os
e
d m
e
t
ho
d
pe
r
f
o
r
m
s
be
t
t
e
r
n
oi
s
e
s
up
p
r
e
s
s
i
on
a
n
d
pr
o
duc
e
s
be
t
t
e
r
s
e
gm
e
nt
a
t
i
on
r
e
s
ul
t
s
.
R
EF
ER
E
N
C
ES
[1]
J
.
H
a
ri
ki
ra
n
,
A
.
R
a
ghu,
D
r.
P
.
V
.
L
a
ks
hm
i
,
D
r.
R.
K
i
r
a
n K
um
a
r,
“
E
dge
D
e
t
e
c
t
i
on us
i
n
g M
a
t
he
m
a
t
i
c
a
l
M
orphol
o
gy
f
or
G
ri
ddi
ng of M
i
c
roa
rra
y
Im
a
ge
”
,
Int
e
r
nat
i
onal
J
o
ur
nal
of
A
dv
anc
e
d R
e
s
e
ar
c
h i
n
Com
put
e
r
Sc
i
e
n
c
e
,
V
ol
um
e
3,
N
o
2,
pp
.
172
-
176,
A
pri
l
2012.
[2]
J
.H
ar
i
k
i
r
an
, D
r
.
P
.V
.L
ak
s
h
m
i
, D
r
.R
.K
i
r
an
K
um
a
r,
“
F
a
s
t
Cl
us
t
e
ri
ng A
l
gori
t
hm
s
f
or S
e
gm
e
nt
a
t
i
on
of M
i
c
roa
rr
a
y
I
m
ag
es
”,
In
t
e
r
na
t
i
onal
J
our
nal
o
f
Sc
i
e
nt
i
f
i
c
&
E
n
gi
ne
e
r
i
ng
R
e
s
e
a
r
c
h
,
V
ol
um
e
5
,
I
s
s
ue
10,
pp
569
-
574,
2014
.
[3]
Bogda
n S
m
ol
ka
,
e
t
.
a
l
.
“
U
l
t
ra
fa
s
t
T
e
c
hni
qu
e
of Im
pul
s
i
ve
N
oi
s
e
Re
m
ova
l
w
i
t
h A
ppl
i
c
a
t
i
on t
o M
i
c
roa
rra
y
Im
a
ge
De
-
noi
s
i
ng”
,
ICI
A
R
2005
,
L
N
CS
3656,
pp
.
990
–
9
97,
2005
.
S
pri
ng
e
r
-
V
e
rl
a
g
Be
r
l
i
n
H
e
i
de
l
be
rg
200
5.
[4]
H
ar
a S
t
ef
an
o
u
,
et
.
al
. “
M
i
cr
o
ar
r
a
y
I
m
ag
e D
en
o
i
s
i
n
g
U
s
i
n
g
a
T
w
o
-
St
a
g
e
M
ul
t
i
re
s
ol
ut
i
on T
e
c
hni
que
”
,
2007
I
EEE
Int
e
r
nat
i
onal
Co
nf
e
r
e
nc
e
on
B
i
o
i
nf
or
m
at
i
c
s
and
B
i
om
e
di
c
i
ne
.
[5]
L
a
ks
hm
a
na
P
ha
ne
e
ndra
M
a
gul
u
ri
,
e
t
.
a
l
,
“
BE
M
D
w
i
t
h Cl
us
t
e
ri
ng A
l
gori
t
hm
f
or S
e
gm
e
nt
a
t
i
o
n of M
i
c
ro
a
rr
a
y
I
m
ag
e”,
Int
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
oni
c
s
Com
m
uni
c
at
i
on and Com
put
e
r
E
ngi
ne
e
r
i
ng
V
ol
um
e
4,
Is
s
ue
2,
IS
S
N
:
2278
–
4209 2013
.
[6]
J
.
H
a
ri
ki
ra
n
,
D
.
R
a
m
a
kri
s
hna
,
B
.
A
vi
na
s
h,
D
r.
P
.
V
.
L
a
ks
hm
i
,
D
r.
R.
K
i
ra
n K
um
a
r,
“
A
N
e
w
M
e
t
hod of G
ri
ddi
ng
for
S
pot
D
e
t
e
c
t
i
on
i
n M
i
c
roa
rr
a
y
Im
a
ge
s
”
,
Com
put
e
r
E
ngi
n
e
e
r
i
ng
an
d Int
e
l
l
i
g
e
nt
Sy
s
t
e
m
s
,
V
ol
5
,
N
o 3
,
pp
.
25
-
33,
2014
[7]
M
.
E
i
s
e
n
,
S
c
a
nA
l
y
z
e
U
s
e
r’s
m
a
nua
l
,
1999,
[8]
J
.
Buhl
e
r,
T
.
Id
e
k
e
r a
nd D
.
H
a
y
no
r,
“
D
a
ppl
e
:
Im
pr
ove
d T
e
c
hni
que
s
for F
i
ndi
ng s
pot
s
on D
M
A
M
i
c
roa
rra
y
Im
a
ge
s
”
,
T
ech
. R
ep
.
U
W
T
R
2
0
0
0
-
08
-
05,
U
ni
ve
rs
i
t
y
of
W
a
s
hi
ngt
on,
2000.
[9]
F
ra
nk Y
.
S
hi
h a
nd S
ho
uxi
a
n Che
ng,
“
A
ut
om
a
t
i
c
s
e
e
de
d re
gi
on g
row
i
ng for c
ol
or i
m
a
ge
s
e
gm
e
nt
a
t
i
on”
,
0262
-
8856/
2005.
[10]
A
l
i
aa S
aad
E
l
-
G
a
w
a
d
y
,
e
t
.
a
l
.
“
S
e
gm
e
nt
a
t
i
on of Com
pl
e
m
e
nt
a
r
y
D
N
A
M
i
c
roa
rra
y
Im
a
ge
s
us
i
ng M
a
rke
r
-
Cont
ro
l
l
e
d
W
at
er
s
h
ed
T
ech
n
i
q
u
e”
,
Int
e
r
nat
i
onal
J
our
nal
of
Com
put
e
r
A
ppl
i
c
at
i
ons
(0975
–
8887) V
ol
um
e
110
–
N
o.
12
,
J
a
nua
r
y
2015
.
[11]
J
.H
ar
i
k
i
r
an
, e
t
.
al
. “
M
u
l
t
i
p
l
e F
e
at
u
r
e F
u
zz
y
C
-
m
e
a
ns
Cl
us
t
e
ri
ng A
l
gori
t
hm
for S
e
gm
e
nt
a
t
i
on of M
i
c
roa
rra
y
i
m
a
ge
”
,
IA
E
S In
t
e
r
nat
i
on
al
J
our
nal
of
E
l
e
c
t
r
i
c
al
and Com
put
e
r
E
ng
i
ne
e
r
i
n
g
,
V
ol
.
5
,
N
o.
5,
pp.
1045
-
1053
,
2015.
[12]
J
.H
ar
i
k
i
r
an
e
t
.
al
. “
F
u
zz
y
C
-
m
e
a
ns
w
i
t
h Bi
-
di
m
e
ns
i
ona
l
e
m
pi
ri
c
a
l
M
ode
de
c
om
pos
i
t
i
on for
s
e
gm
e
nt
a
t
i
on o
f
M
i
c
roa
rra
y
Im
a
ge
”
,
Int
e
r
nat
i
on
al
J
our
nal
of
Co
m
put
e
r
Sc
i
e
nc
e
Is
s
ue
s
,
vol
um
e
9,
Is
s
ue
5
,
N
um
be
r 3,
pp.
273
-
27
9,
2012.
[13]
J
.H
ar
i
k
i
r
an
, B
.A
vi
na
s
h,
D
r.
P
.
V
.
L
a
ks
hm
i
,
D
r.
R.
K
i
ra
nkum
a
r,
“
A
ut
om
a
t
i
c
G
ri
dd
i
ng M
e
t
hod for
m
i
c
roa
rra
y
i
m
a
g
e
s
”
,
J
our
nal
of
T
he
or
e
t
i
c
al
and
A
pp
l
i
e
d Inf
or
m
at
i
on
T
e
c
hnol
og
y
,
vol
u
m
e
65,
N
um
be
r
1,
pp
.
235
-
241,
2
014.
[14]
Ra
dha
kri
s
hna
A
c
ha
nt
a
e
t
.
a
l
.
”
S
L
I
C S
upe
rpi
xe
l
s
Com
pa
re
d t
o s
t
a
e
-
of
-
t
he
-
a
r
t
S
upe
r
pi
xe
l
M
e
t
hods
”
,
J
o
ur
nal
of
L
at
e
x
C
la
s
s
File
s
,
V
ol
6,
N
o.
1,
D
e
c
e
m
be
r 2011.
[15]
H
a
m
i
dre
z
a
S
a
be
rka
ri
,
e
t
.
a
l
.
”
F
ul
l
y
A
ut
om
a
t
e
d
Com
pl
e
m
e
nt
a
ry
D
N
A
M
i
c
roa
rra
y
S
e
gm
e
nt
a
t
i
o
n us
i
ng a
N
ove
l
F
u
z
zy
‑
b
a
s
e
d A
l
gori
t
hm
”
,
IE
E
E
T
r
ans
ac
t
i
ons
on
Inf
or
m
at
i
on T
e
c
hnol
ogy
in
Bio
m
e
d
ic
in
e
,
V
ol
5,
Is
s
ue
3,
J
ul
-
Se
p
2015.
[16]
L
ui
s
Rue
d
a
a
nd
V
i
d
y
a
V
i
d
y
a
dh
a
ra
n,
“
A
H
i
l
l
-
c
l
i
m
bi
ng A
pproa
c
h for A
ut
om
a
t
i
c
G
ri
ddi
ng of
c
D
N
A
M
i
c
roa
rr
a
y
I
m
ag
es
”,
IA
E
S I
nt
e
r
nat
i
onal
J
ou
r
nal
of
E
l
e
c
t
r
i
c
a
l
and Com
put
e
r
E
ngi
ne
e
r
i
ng
,
v
ol
um
e
4,
N
o 6,
D
e
c
e
m
be
r 2014,
pp.
923
-
930.
[17]
B.
S
a
i
c
h
a
nda
n
a
e
t
.
a
l
.
“
H
y
p
e
rs
pe
c
t
ra
l
Im
a
ge
Cl
a
s
s
i
fi
c
a
t
i
on us
i
ng G
e
ne
t
i
c
A
l
gori
t
h
m
a
ft
e
r V
i
s
ua
l
i
z
a
t
i
on us
i
ng Im
a
g
e
F
us
i
on”
,
Int
e
r
nat
i
onal
J
our
nal
o
f
Com
put
e
r
Sc
i
e
n
c
e
A
nd T
e
c
hnol
o
gy
,
V
ol
um
e
7,
N
o.
2
,
J
une
2016,
pp.
2229
-
4333.
[18]
D
urga
P
ra
s
a
d K
ondi
s
e
t
t
y
,
D
r.
M
oha
m
m
e
d A
l
i
H
us
s
a
i
n.
“
A
Re
vi
e
w
on M
i
c
ro
a
rra
y
Im
a
ge
S
e
gm
e
n
t
a
t
i
on
M
e
t
hods
”
,
Int
e
r
nat
i
onal
J
o
ur
nal
of
Com
put
e
r
Sc
i
e
nc
e
and
I
nf
or
m
at
i
on Se
c
u
r
i
t
y
(
IJ
CSIS)
,
V
ol
.
14
,
N
o
.
12
,
D
e
c
e
m
be
r 2016
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
A
S
I
S
S
N
:
225
2
-
88
14
SL
I
C
S
u
pe
r
pi
x
e
l
B
as
e
d
Se
l
f
O
r
ga
ni
z
i
n
g
M
a
p
s
A
l
g
or
i
t
h
m
f
or
S
e
g
m
e
n
ta
tion…
(
D
u
r
ga
P
r
as
a
d
K
ond
is
e
tty
)
85
[19]
K
om
a
ng A
ri
a
na
,
e
t
.
a
l
.
,
“
Col
or Im
a
ge
S
e
gm
e
nt
a
t
i
on us
i
ng K
ohone
n S
O
M
”,
Int
e
r
nat
i
onal
J
our
n
al
of
E
ngi
n
e
e
r
i
n
g
and
T
e
c
hnol
og
y
(
IJ
E
T
)
,
V
ol
.
6,
N
o.
2,
D
e
c
e
m
be
r 2014.
[20]
K
ri
s
t
of V
a
n L
a
e
rhove
n.
e
t
.
a
l
,
“
Com
bi
ni
ng t
he
S
e
l
f
-
O
rga
ni
z
i
n
g M
a
p a
nd K
-
M
e
a
ns
Cl
us
t
e
ri
ng for O
n
-
l
i
n
e
Cl
a
s
s
i
fi
c
a
t
i
on
of
S
e
ns
or D
a
t
a
”
,
I
nt
e
r
nat
i
onal
J
ou
r
nal
of
Com
pu
t
e
r
Sc
i
e
n
c
e
and
In
f
or
m
at
i
on Se
c
ur
i
t
y
(
IJ
CSIS)
,
V
o
l
.
14,
N
o.
12,
D
e
c
e
m
be
r 2001.
[21]
M
.
N
.
M
.
S
a
p,
e
t
.
a
l
,
.
“
H
y
bri
d S
e
l
f O
rga
ni
z
i
ng
M
a
p for O
ve
rl
a
ppi
ng Cl
us
t
e
rs
”
,
Int
e
r
nat
i
onal
J
our
nal
of
Si
gnal
P
r
oc
e
s
s
i
ng,
Im
age
P
r
oc
e
s
s
i
ng a
nd P
at
t
e
r
n R
e
c
o
gni
t
i
on
,
V
ol
.
14
,
N
o.
12
,
D
e
c
e
m
be
r 2016.
[22]
J
a
nne
N
i
kki
l
,
e
t
.
a
l
,
.
“
A
na
l
y
s
i
s
a
n
d vi
s
ua
l
i
z
a
t
i
on o
f ge
ne
e
xpre
s
s
i
o
n da
t
a
us
i
ng S
O
M
”
,
N
e
ura
l
N
e
t
w
orks
15 (2002)
953
–
966 V
ol
.
14
,
N
o.
12,
D
e
c
e
m
be
r 2002.
Evaluation Warning : The document was created with Spire.PDF for Python.