T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
,
p
p
.
199
~
2
0
5
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
Fi
r
s
t G
r
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i1
.
1
5
1
2
2
199
J
o
ur
na
l ho
m
ep
a
ge
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
An it
e
ra
tive a
lg
o
rithm f
o
r
colo
r spa
ce optimiza
tion
on ima
g
e
seg
menta
tion
M
o
ura
d M
o
u
s
s
a
De
p
a
rtme
n
t
o
f
Ap
p
li
e
d
in
f
o
rm
a
ti
c
s,
Ne
two
rk
e
d
Ob
jec
ts
,
Co
n
tro
l
a
n
d
C
o
m
m
u
n
ica
ti
o
n
S
y
ste
m
s (NOCC
S
)
L
a
b
o
ra
t
o
ry
Na
ti
o
n
a
l
E
n
g
i
n
e
e
rin
g
S
c
h
o
o
l
o
f
S
o
u
ss
e
,
Un
i
v
e
rsity
o
f
S
o
u
ss
e
,
T
u
n
i
sia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
1
,
2
0
2
0
R
ev
is
ed
J
u
l 1
7
,
2
0
2
0
Acc
ep
ted
Au
g
2
9
,
2
0
2
0
Th
is
p
a
p
e
r
p
r
o
p
o
se
s,
a
n
o
v
e
l
h
y
b
rid
c
o
lo
r
c
o
m
p
o
n
e
n
t
(HCC)
iss
u
e
d
fr
o
m
a
m
o
u
n
ts
n
u
m
b
e
r
o
f
c
o
l
o
r
sp
a
c
e
with
it
e
ra
ti
v
e
m
a
n
n
e
r,
in
fa
c
t
trad
it
io
n
a
l
ima
g
e
s
o
b
tai
n
e
d
b
y
RG
B
se
n
so
r
w
e
re
n
’t
th
e
e
ffe
c
ti
v
e
wa
y
in
ima
g
e
p
ro
c
e
ss
in
g
a
p
p
li
c
a
ti
o
n
s,
f
o
r
t
h
is
p
u
rp
o
se
w
e
h
a
v
e
p
r
o
p
o
se
a
su
p
e
r
v
ise
d
a
l
g
o
rit
h
m
to
su
b
stit
u
te
RG
B
lev
e
l
b
y
h
y
b
rid
a
n
d
s
u
it
a
b
le
c
o
lo
r
sp
a
c
e
a
t
t
h
e
a
i
m
to
m
a
k
e
we
ll
re
p
re
se
n
tatio
n
o
f
t
h
e
h
a
n
d
l
e
d
a
m
o
u
n
ts
o
f
d
a
ta,
th
is
ste
p
is
e
x
trem
e
ly
imp
o
rtan
t
b
e
c
a
u
se
th
e
o
b
tain
e
d
re
su
lt
s
it
will
b
e
i
n
jec
ted
in
m
a
n
y
f
u
t
u
re
stu
d
ies
li
k
e
trac
k
in
g
,
c
las
sifica
ti
o
n
,
ste
g
a
n
o
g
ra
p
h
y
a
n
d
c
ry
p
to
g
ra
p
h
y
.
T
h
e
s
e
c
o
n
d
p
a
rt
o
f
th
is
p
a
p
e
r
c
o
n
sists
to
se
g
m
e
n
t
ima
g
e
c
o
d
e
d
in
h
y
b
ri
d
c
o
lo
r
sp
a
c
e
a
lrea
d
y
se
lec
ted
,
th
e
u
se
d
a
lg
o
rit
h
m
is
i
n
s
p
ired
fro
m
k
e
rn
e
l
fu
n
c
ti
o
n
wh
e
re
sta
ti
stica
l
d
istri
b
u
t
i
o
n
wa
s
u
se
d
t
o
m
o
d
e
l
b
a
c
k
g
ro
u
n
d
a
n
d
Ba
y
e
s
ru
le
to
m
a
k
e
d
e
c
isio
n
o
f
th
e
m
e
m
b
e
rsh
ip
o
f
e
a
c
h
p
ix
e
l,
i
n
th
is
re
se
a
rc
h
to
p
ics
we
h
a
v
e
e
x
ten
d
e
d
t
h
is
a
lg
o
rit
h
m
in
t
h
e
a
im
to
imp
ro
v
e
c
o
m
p
a
c
tn
e
ss
o
f
th
e
se
d
istr
ib
u
ti
o
n
.
Ca
u
c
h
y
b
a
c
k
g
ro
u
n
d
m
o
d
e
li
n
g
a
n
d
su
b
trac
ti
o
n
is
u
se
d
,
a
n
d
sh
o
ws
th
e
h
ig
h
a
c
c
u
ra
c
y
o
f
a
u
to
m
a
ti
c
p
lay
e
r
d
e
tec
ti
o
n
.
K
ey
w
o
r
d
s
:
C
o
lo
r
s
p
ac
e
Data
f
u
s
io
n
I
m
ag
e
s
eg
m
en
tatio
n
I
ter
ativ
e
alg
o
r
it
hm
Statis
t
ical
d
is
tr
ib
u
tio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mo
u
r
ad
M
o
u
s
s
a
Dep
ar
tm
en
t o
f
A
p
p
lied
I
n
f
o
r
m
atics
Netwo
r
k
ed
Ob
jects
,
C
o
n
tr
o
l
a
n
d
C
o
m
m
u
n
icatio
n
Sy
s
tem
s
(
NOC
C
S)
L
ab
o
r
ato
r
y
Natio
n
al
E
n
g
in
ee
r
i
n
g
Sch
o
o
l
o
f
So
u
s
s
e
,
Un
iv
er
s
ity
o
f
So
u
s
s
e
Pô
le
tech
n
o
lo
g
i
q
u
e
d
e
So
u
s
s
e,
R
o
u
te
d
e
C
ein
tu
r
e
Sah
lo
u
l
4
0
5
4
,
T
u
n
is
ia
E
m
ail:
Mo
u
r
ad
.
e
n
im
@
y
ah
o
o
.
f
r
1.
I
NT
RO
D
UCT
I
O
N
Ob
ject
d
etec
tio
n
an
d
tr
ac
k
i
n
g
[
1
,
2
]
,
s
eg
m
en
tatio
n
[3
-
5
]
,
ed
g
e
d
etec
tio
n
[
6
]
,
class
if
icatio
n
[7
-
9
]
an
d
f
ac
e
r
ec
o
g
n
itio
n
[
1
0
,
1
1
]
ar
e
t
h
e
m
ain
s
tep
s
o
f
a
co
m
p
u
ter
v
is
io
n
s
y
s
tem
f
o
r
im
ag
e
an
aly
s
is
.
T
h
e
p
u
r
p
o
s
e
o
f
s
eg
m
en
tatio
n
is
to
s
im
p
lify
an
d
o
r
c
h
an
g
e
t
h
e
r
ep
r
esen
t
atio
n
o
f
an
im
a
g
e
in
to
s
o
m
eth
in
g
th
at
is
m
o
r
e
m
ea
n
in
g
f
u
l
an
d
ea
s
ier
to
an
al
y
ze
.
C
o
lo
r
in
f
o
r
m
atio
n
is
a
cr
u
cial
way
to
in
cr
ea
s
e
s
ep
ar
a
b
i
lity
p
o
wer
o
f
m
an
y
d
is
tr
ib
u
tio
n
s
an
d
o
b
jects
in
f
r
a
m
ewo
r
k
.
G
.
H.
L
iu
et
a
l.
[
1
2
]
p
r
esen
t
im
ag
e
r
etr
iev
al
tech
n
i
q
u
e
b
ased
o
n
co
lo
r
d
if
f
er
en
ce
h
is
to
g
r
am
(
C
DH)
t
o
ex
tr
ac
t
co
lo
r
f
ea
tu
r
es
c
o
d
ed
in
∗
∗
∗
.
C
o
lo
r
in
f
o
r
m
atio
n
is
also
u
s
ed
in
s
teg
an
o
g
r
ap
h
y
f
ield
[
1
3
,
1
4
]
an
d
B
io
m
etr
ic
Au
th
en
ticatio
n
[
1
5
,
16]
,
f
o
r
in
s
tan
ce
S.
Hem
alath
a
et
a
l.
[
1
7
]
p
r
o
p
o
s
e
a
n
ew
im
ag
e
s
teg
an
o
g
r
ap
h
y
tec
h
n
iq
u
e
to
c
o
n
ce
a
l
s
ec
r
e
t
in
f
o
r
m
atio
n
i
n
co
lo
r
im
ag
e,
g
iv
en
th
e
u
n
s
atis
f
ac
to
r
y
r
esu
lts
with
th
e
g
r
e
y
s
ca
le
im
a
g
es,
in
f
ac
t
co
lo
r
im
ag
e
ca
n
h
o
ld
a
b
ig
am
o
u
n
t
o
f
s
ec
r
et
in
f
o
r
m
atio
n
f
o
r
th
is
r
ea
s
o
n
th
ey
h
av
e
u
s
ed
two
co
lo
r
s
p
ac
es
R
G
B
an
d
YC
b
C
r
,
ex
p
er
im
en
tal
r
esu
lts
s
h
o
w
th
at
YC
b
C
r
d
o
esn
’
t
r
eq
u
ir
es
en
o
u
g
h
co
m
p
u
tin
g
tim
e
th
an
th
e
R
G
B
o
n
e,
an
d
is
ea
s
y
f
o
r
ex
tr
ac
tin
g
s
ec
r
et
i
m
ag
es.
S.
C
h
itra
et
a
l.
[
1
8
]
p
r
o
p
o
s
e
a
tech
n
iq
u
e
in
b
io
m
etr
ic
f
ield
th
at
co
n
s
is
ts
to
f
ac
e
r
ec
o
g
n
itio
n
b
ased
o
n
s
k
in
co
lo
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
9
9
-
20
5
200
d
etec
tio
n
,
th
e
f
ir
s
t
s
tep
o
f
th
is
alg
o
r
ith
m
co
n
v
er
t
o
r
ig
in
al
im
a
g
e
in
HSV
an
d
YC
b
C
r
co
lo
r
s
p
ac
e
th
en
co
ll
ec
t
th
e
v
alu
e
o
f
H
,
S
,
Cb
,
Cr
an
d
f
in
ally
ch
ec
k
wh
eth
er
th
ese
v
alu
es a
r
e
s
atis
f
ied
with
th
e
s
p
ec
if
ied
th
r
esh
o
l
d
v
alu
es.
T
h
e
p
r
o
p
o
s
ed
p
a
p
er
is
s
u
b
d
iv
id
ed
in
to
two
p
ar
ts
th
e
f
ir
s
t
o
n
e
is
s
p
ec
if
ie
d
f
o
r
(
HC
S)
s
elec
tio
n
with
s
u
p
er
v
is
ed
an
d
iter
ativ
e
alg
o
r
i
th
m
am
o
n
g
a
wid
e
s
et
o
f
co
lo
r
lev
els
is
s
u
ed
f
r
o
m
m
an
y
co
l
o
r
s
y
s
tem
co
m
m
o
n
ly
u
s
ed
in
im
ag
e
p
r
o
ce
s
s
in
g
,
wh
en
we
h
av
e
in
tr
o
d
u
ce
m
an
y
t
o
o
ls
allo
win
g
to
s
elec
t
s
u
itab
le
h
y
b
r
id
ch
r
o
m
atic
lev
el
an
d
we
h
av
e
p
r
o
p
o
s
e
m
an
y
way
s
to
m
ak
e
d
ec
is
io
n
ab
o
u
t c
o
lo
r
co
m
p
o
n
en
ts
lik
e,
f
ir
s
tly
th
e
u
s
e
o
f
p
ar
tial
an
d
f
u
ll
d
ata
an
d
s
ec
o
n
d
ly
t
h
e
u
s
e
o
f
d
ata
f
u
s
io
n
th
e
o
r
y
.
T
h
e
s
ec
o
n
d
p
a
r
t
is
d
ed
icate
d
to
i
m
ag
e
s
eg
m
en
tatio
n
we
h
av
e
u
s
ed
a
s
tatis
t
ical
m
et
h
o
d
b
ased
o
n
C
au
ch
y
d
is
tr
ib
u
tio
n
f
o
r
m
o
d
elin
g
[
1
9
,
2
0
]
an
d
p
r
o
b
a
b
ilit
y
th
eo
r
y
b
ased
o
n
B
ay
es
r
u
les
f
o
r
m
a
k
i
n
g
d
ec
is
io
n
a
b
o
u
t
b
elo
n
g
in
g
o
f
p
ix
el
to
s
u
itab
le
clu
s
ter
s
.
At
th
e
f
ir
s
t
-
tim
e
b
u
ild
in
g
b
ac
k
g
r
o
u
n
d
m
o
d
el
an
d
in
th
e
s
ec
o
n
d
tim
e
we
d
etec
t
an
d
allo
ca
te
ea
ch
p
ix
el
to
its
clas
s
.
F
in
ally
,
we
co
n
clu
d
e
th
is
p
ap
er
b
y
ex
p
e
r
im
en
tal
an
d
d
is
cu
s
s
io
n
r
esu
lts
.
2.
I
T
E
RA
T
I
V
E
AL
G
O
R
I
T
H
M
F
O
R
CO
L
O
R
SPAC
E
S
E
L
E
C
T
I
O
N
2
.
1
.
I
ntr
o
du
ct
io
n
W
e
p
r
o
p
o
s
e
t
o
an
al
y
ze
a
m
ea
s
u
r
e
o
f
s
im
ilar
ity
o
f
o
b
jects
in
a
tr
ain
in
g
im
ag
e
f
o
r
ea
ch
co
lo
r
co
m
p
o
n
en
t
to
b
u
il
d
an
o
p
tim
al
(
HC
S)
wh
er
e
we
wa
n
t
r
ea
lize
th
e
s
eg
m
e
n
tatio
n
s
tep
.
I
t
is
ass
u
m
ed
th
at
an
o
b
ject
o
r
r
eg
i
o
n
ca
n
b
e
a
s
u
b
s
et
o
f
h
i
g
h
ly
co
lo
r
-
co
n
n
ec
ted
p
ix
els,
wh
ich
is
r
ep
r
esen
ted
b
y
th
e
c
o
lo
r
p
o
i
n
ts
in
t
h
e
c
o
lo
r
s
p
ac
e
ar
o
u
n
d
.
T
o
an
aly
ze
s
u
ch
p
r
o
p
er
ty
o
f
a
s
u
b
s
et
-
co
lo
r
,
we
p
r
o
p
o
s
e
to
m
ea
s
u
r
e
th
e
d
e
g
r
ee
o
f
c
o
n
n
ec
tio
n
th
a
t
q
u
an
tifie
s
th
e
s
p
atial
ar
r
an
g
e
m
en
t o
f
it
s
p
ix
els in
th
e
im
ag
e
lev
el.
L
et
an
d
N
s
(
p
)
ar
e
r
esp
ec
tiv
ely
a
c
o
l
o
r
o
b
ject
an
d
th
e
s
et
o
f
n
eig
h
b
o
u
r
s
p
i
x
els
o
f
a
p
i
x
el
b
elo
n
g
in
g
to
s
u
b
s
et
p
ix
els
o
f
S
.
T
h
e
co
n
n
ec
tiv
ity
d
eg
r
ee
b
etwe
en
P
an
d
S
d
en
o
ted
γ
s
(
p
)
d
ep
e
n
d
s
o
n
th
e
n
u
m
b
er
o
f
n
eig
h
b
o
r
p
ix
els
b
elo
n
g
in
g
to
N
s
(
p
)
.
T
h
e
d
eg
r
ee
o
f
c
o
n
n
ec
tio
n
DC
(
S
)
cr
iter
io
n
is
ex
p
r
ess
ed
b
y
(
1
)
:
(
)
=
∑
(
)
∈
(
∈
)
(
1
)
T
h
e
co
n
n
ec
tiv
ity
d
e
g
r
ee
DC
(
S
)
is
th
e
av
er
ag
e
n
u
m
b
er
o
f
n
eig
h
b
o
r
p
ix
els
o
f
S
wh
ich
also
b
elo
n
g
to
th
e
s
am
e
o
b
ject.
T
h
e
c
o
n
n
ec
tiv
ity
d
e
g
r
ee
is
b
etwe
en
0
an
d
1
,
wh
e
n
DC
is
ab
o
u
t
0
t
h
at
m
ea
n
s
th
e
p
ix
el
o
f
c
o
n
s
id
er
ed
clu
s
ter
ar
e
d
is
p
er
s
ed
in
tr
ea
ted
im
ag
e,
in
th
e
co
n
tr
a
r
y
ca
s
e
th
e
s
et
o
f
p
ix
el
ar
e
co
n
s
id
er
ed
as
s
tr
o
n
g
ly
c
o
n
n
ec
ted
.
2
.
2
.
E
v
a
lua
t
i
o
n o
f
c
o
nn
ec
t
iv
it
y
deg
re
e
o
f
ea
ch
co
m
po
nent
To
im
p
r
o
v
e
th
e
q
u
ality
o
f
co
lo
r
im
ag
e
s
eg
m
en
tatio
n
we
s
h
o
u
ld
s
elec
t
th
e
o
p
tim
al
co
lo
r
s
p
ac
e
[
2
1
,
2
2
]
to
r
ea
ch
ef
f
ec
tiv
e
r
esu
lts
.
W
e
c
o
m
p
ar
e
a
n
u
m
b
er
o
f
c
o
lo
r
c
o
m
p
o
n
en
ts
f
r
o
m
co
n
v
en
tio
n
al
co
lo
r
s
p
ac
es
b
ased
o
n
th
e
co
n
n
ec
tiv
ity
cr
iter
i
o
n
d
ef
in
ed
in
p
r
e
v
io
u
s
s
ec
tio
n
,
an
d
we
s
elec
t
th
e
m
o
s
t
d
is
cr
im
i
n
an
t
co
m
p
o
n
e
n
ts
to
b
u
ild
th
e
h
y
b
r
id
d
is
cr
im
in
atin
g
co
lo
r
s
p
ac
e
in
wh
ich
s
eg
m
e
n
tatio
n
will
b
e
b
r
in
g
o
u
t.
W
e
s
elec
t
two
im
ag
es
co
n
tain
in
g
r
esp
ec
tiv
el
y
class
1
an
d
class
2
,
th
ese
im
ag
es
ar
e
u
s
ed
f
o
r
s
u
p
er
v
i
s
ed
lear
n
in
g
.
W
e
ev
alu
ate
f
o
r
ea
ch
co
lo
r
c
o
m
p
o
n
en
t
th
e
co
n
n
ec
ti
v
ity
d
eg
r
ee
o
f
th
e
two
im
ag
e
s
r
ep
r
esen
tin
g
t
h
e
two
e
x
is
tin
g
class
es.
Fo
r
ea
ch
co
lo
r
c
o
m
p
o
n
en
t,
we
ca
lcu
late
th
e
s
u
m
o
f
two
d
eg
r
ee
s
o
f
co
n
n
ec
tiv
ity
f
o
r
class
1
an
d
class
2
.
So
m
e
c
o
m
p
o
n
en
ts
d
o
n
o
t
d
etec
t
all
class
lev
els
o
f
co
lo
r
s
,
th
eir
d
eg
r
ee
s
o
f
co
n
n
ec
ted
n
ess
d
o
n
o
t
th
er
ef
o
r
e
r
ef
lect
th
e
d
eg
r
ee
o
f
class
co
n
n
ec
tiv
ity
,
th
ese
co
m
p
o
n
en
ts
ar
e
r
ejec
ted
lik
e
(
)
.
On
th
e
o
th
er
h
an
d
,
th
e
co
m
p
o
n
en
t
H
(
HSL
)
l
o
s
t
th
e
s
h
ap
e
o
f
th
e
o
b
ject,
th
er
e
is
a
ce
r
tain
o
v
er
-
d
etec
tio
n
,
t
h
is
co
m
p
o
n
en
t
ca
n
'
t
also
h
av
e
a
r
ea
l
d
eg
r
ee
o
f
co
n
n
ec
tiv
ity
o
f
th
e
d
etec
ted
class
.
T
h
e
b
est
d
eg
r
ee
o
f
co
n
n
e
ctiv
ity
f
o
u
n
d
is
0
.
9
2
7
1
,
t
h
is
v
alu
e
is
p
r
o
v
id
e
d
b
y
th
e
co
m
p
o
n
en
ts
f
o
llo
win
g
co
l
o
r
s
:
X
(
XYZ
)
,
L
(
Lab
)
,
a
(
Lab
)
,
b
(
Lab
)
,
I
(
Y
IQ
)
.
W
e
b
u
ild
th
e
(
HC
S)
d
im
en
s
io
n
co
n
s
is
ts
o
f
f
iv
e
co
m
p
o
n
e
n
ts
ab
o
v
e
wh
er
e
s
eg
m
e
n
tatio
n
will
b
e
m
ad
e.
I
n
o
u
r
a
p
p
r
o
ac
h
we
will
u
s
e
th
e
iter
ativ
e
alg
o
r
ith
m
b
ased
o
n
d
is
cr
i
m
in
a
n
t a
n
aly
s
is
p
o
licy
,
wh
ich
d
eter
m
in
es th
e
th
r
ee
co
m
p
o
n
e
n
ts
m
o
s
t r
elev
an
t c
o
lo
r
s
co
n
s
titu
ti
n
g
a
(
HC
S
)
.
2
.
3
.
T
ra
ini
ng
s
a
m
ple c
o
ns
t
r
uct
io
n
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
co
n
s
is
ts
to
s
ea
r
ch
th
e
e
f
f
icien
t
(
HC
S)
f
o
r
d
is
tin
g
u
is
h
in
g
two
d
o
m
in
a
n
t
class
es
o
f
p
ix
els
in
a
g
iv
en
d
ata
b
ase
i
m
ag
e.
T
o
d
eter
m
in
e
t
h
e
(
HC
S)
,
as
it’s
s
h
o
wn
in
Fig
u
r
e
1
,
w
e
tak
e
5
im
ag
es
f
o
r
ea
ch
class
o
f
p
ix
els
an
d
we
d
ef
in
e
f
o
r
ea
ch
class
Cj
,
Wi
,
j
f
r
o
m
a
s
et
o
f
Ww
r
ep
r
esen
tativ
e
p
ix
e
ls
wh
er
e
=
1
…
an
d
=
1
.
2
.
I
n
a
g
iv
en
co
lo
r
s
p
ac
e
h
a
v
in
g
d
d
im
en
s
io
n
,
we
ch
a
r
ac
te
r
ize
ea
ch
p
ix
el
,
b
elo
n
g
in
g
to
class
b
y
th
e
o
b
s
er
v
atio
n
,
=
[
,
1
,
…
,
,
,
…
,
,
]
.
T
h
e
r
o
ws
o
f
th
e
m
atr
i
x
co
r
r
esp
o
n
d
to
×
2
r
ep
r
esen
tativ
e
p
ix
els,
wh
ile
th
e
co
lu
m
n
s
co
r
r
esp
o
n
d
to
th
e
lev
els
o
f
co
lo
r
co
m
p
o
n
e
n
ts
o
f
ea
ch
p
ix
el.
W
e
ca
n
also
ex
p
r
ess
th
e
m
atr
ix
by
=
[
1
…
,
…
,
]
,
wh
er
e
=
[
1
,
1
…
,
1
,
1
,
2
…
,
,
1
]
,
co
n
tain
in
te
n
s
ities
l
ev
el
o
f
ℎ
co
lo
r
co
m
p
o
n
en
t
o
f
×
2
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
iter
a
tive
a
lg
o
r
ith
m
fo
r
co
lo
r
s
p
a
ce
o
p
timiz
a
tio
n
o
n
ima
g
e
s
eg
men
ta
tio
n
(
Mo
u
r
a
d
Mo
u
s
s
a
)
201
r
ep
r
esen
tativ
e
p
ix
els.
W
h
er
e
,
is
th
e
ℎ
in
ten
s
ity
lev
el
o
f
co
lo
r
co
m
p
o
n
en
t.
W
e
r
ep
r
esen
t
all
o
b
s
er
v
atio
n
s
in
th
e
m
a
tr
ix
f
o
r
m
:
=
(
1
,
1
1
⋯
,
1
1
1
,
2
1
⋯
,
2
1
⋮
⋮
⋮
⋮
1
,
1
⋯
,
1
1
,
2
⋯
,
2
⋮
⋮
⋮
⋮
1
,
1
⋯
,
1
1
,
2
⋯
,
2
)
Fig
u
r
e
1
.
T
r
ain
in
g
s
et
im
ag
e
f
o
r
two
d
o
m
in
an
t c
lass
es
2
.
4
.
M
ini
m
iza
t
io
n o
f
co
lo
r
c
o
m
po
nent
co
rr
ela
t
i
o
n
T
o
m
ea
s
u
r
e
th
e
lev
el
o
f
c
o
r
r
el
atio
n
b
etwe
en
th
e
co
l
o
r
co
m
p
o
n
en
ts
,
we
co
n
s
id
er
th
e
−
1
co
u
p
l
es
o
f
v
ec
to
r
s
with
an
d
o
n
e
o
f
th
e
v
ec
to
r
s
,
wh
er
e
=
1
,
.
.
.
,
−
1
.
L
et
(
,
)
,
in
(
2
)
s
h
o
ws
th
e
co
r
r
elat
io
n
m
ea
s
u
r
e
b
etwe
en
ℎ
an
d
ℎ
co
lo
r
co
m
p
o
n
e
n
t:
(
,
)
=
(
,
)
(
2
)
T
h
u
s
(
3
)
d
ef
in
es th
e
c
o
-
v
a
r
ian
ce
m
ea
s
u
r
em
en
t
(
,
)
b
etwe
en
ℎ
an
d
ℎ
lev
el:
(
,
)
=
1
∗
∑
∑
(
,
=
1
=
1
−
)
(
,
−
)
(
3
)
wh
er
e
,
,
an
d
r
ep
r
esen
t
r
esp
ec
tiv
ely
m
ea
n
s
an
d
s
tan
d
a
r
d
d
ev
iatio
n
o
f
ℎ
an
d
ℎ
co
lo
r
co
m
p
o
n
en
t,
f
o
r
ℎ
co
m
p
o
n
en
t th
e
y
ar
e
d
e
f
in
ed
b
y
(
4
)
:
=
1
∗
∑
∑
(
,
=
1
)
=
1
,
=
1
√
∗
√
∑
∑
(
,
=
1
=
1
−
)
2
(
4
)
C
o
r
r
elatio
n
v
alu
es a
r
e
r
a
n
g
ed
b
etwe
en
0
an
d
1
.
W
h
en
t
h
e
co
r
r
elatio
n
v
alu
e
cl
o
s
e
to
1
,
lev
e
ls
ar
e
co
n
s
id
er
ed
t
o
b
e
co
r
r
elate
d
.
T
h
e
m
a
x
im
u
m
o
f
c
o
r
r
elatio
n
b
etwe
en
th
e
ℎ
co
lo
r
c
o
m
p
o
n
en
t
a
n
d
all
o
t
h
er
−
1
co
lo
r
co
m
p
o
n
en
ts
is
co
m
p
u
ted
b
y
(
5
)
an
d
it'
s
co
n
s
id
er
ed
as
a
co
r
r
elatio
n
m
ea
s
u
r
e,
d
en
o
ted
(
)
b
etwe
en
th
e
c
o
lo
r
co
m
p
o
n
e
n
ts
o
f
th
e
(
HC
S)
:
(
)
=
=
1
−
1
(
(
,
)
)
(
5
)
W
e
co
n
s
id
er
o
n
ly
th
e
(
HC
S)
o
f
d
im
e
n
s
io
n
f
o
r
w
h
ich
th
e
co
r
r
elatio
n
m
ea
s
u
r
e
is
b
elo
w
a
g
iv
en
th
r
esh
o
ld
,
it
is
th
e
ca
n
d
i
d
ate
s
p
ac
es
th
at
will
b
e
co
n
s
id
er
ed
b
y
th
e
b
u
ild
p
r
o
ce
s
s
o
f
th
e
s
p
ac
e
m
o
s
t
d
is
cr
im
i
n
atin
g
h
y
b
r
id
c
o
lo
r
,
we
co
n
s
id
er
o
n
ly
a
lim
ited
n
u
m
b
er
o
f
h
y
b
r
id
ca
n
d
id
ate
co
lo
r
s
p
ac
e
h
av
in
g
d
im
en
s
io
n
,
wh
er
e
=
1
⋯
am
o
n
g
all
−
+
1
(
HC
S)
.
2
.
5
.
M
a
x
i
m
iza
t
io
n o
f
dis
cr
im
ina
t
ing
po
wer
W
e
p
r
esen
t
th
e
s
eq
u
en
tial
p
r
o
ce
d
u
r
e
f
o
r
s
elec
tin
g
co
lo
r
co
m
p
o
n
e
n
ts
th
at
d
eter
m
in
es
th
e
(
HC
S)
.
I
n
each
r
an
k
iter
atio
n
o
f
t
h
e
co
n
s
tr
u
ctio
n
p
r
o
ce
s
s
,
we
co
n
s
id
er
all
ca
n
d
id
ate
(
HC
S
)
o
f
d
im
e
n
s
io
n
,
wh
er
e
=
1
⋯
.
W
e
ev
alu
ate
th
eir
d
is
cr
im
in
ato
r
y
p
o
wer
s
with
a
r
elev
an
t
cr
iter
io
n
,
a
n
d
s
elec
t
th
e
b
est
o
n
e
th
at
is
th
e
(
HC
S)
h
av
in
g
d
im
en
s
io
n
lik
e
th
e
m
o
s
t
d
is
cr
i
m
in
atin
g
o
n
e,
d
en
o
ted
b
y
.
T
h
is
cr
iter
io
n
,
n
o
ted
by
(
)
m
ea
s
u
r
es
th
e
d
is
p
er
s
io
n
o
f
th
e
o
b
s
er
v
atio
n
s
r
elate
d
to
r
ep
r
esen
tativ
e
p
ix
els
in
ea
ch
h
y
b
r
id
ca
n
d
id
at
e
co
lo
r
s
p
ac
e.
On
th
e
f
ir
s
t
iter
at
io
n
=
1
,
we
co
n
s
id
er
th
e
−
d
-
d
im
e
n
s
io
n
al
ca
n
d
id
ates
s
p
ac
es
d
ef
in
ed
b
y
ea
ch
o
f
th
e
av
ailab
le
co
lo
r
c
o
m
p
o
n
e
n
ts
.
T
h
e
m
o
n
o
-
d
im
en
s
io
n
al
s
p
ac
e
s
elec
ted
is
th
e
o
n
e
th
at
m
ax
im
izes
th
e
d
is
cr
im
in
ativ
e
p
o
wer
(
1
)
.
Du
r
in
g
t
h
e
s
ec
o
n
d
iter
atio
n
o
f
t
h
e
p
r
o
ce
d
u
r
e
=
2
,
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
9
9
-
20
5
202
two
-
d
im
en
s
io
n
al
(
HC
S
)
ar
e
c
o
n
s
titu
ted
b
y
co
m
b
in
in
g
th
e
co
lo
r
c
o
m
p
o
n
en
t
is
s
u
e
f
r
o
m
1
to
ea
ch
o
f
th
e
−
1
r
em
ain
in
g
co
lo
r
co
m
p
o
n
e
n
ts
.
Am
o
n
g
2
th
ese
two
-
d
im
en
s
io
n
al
(
HC
S
)
2
wh
er
e
=
1
⋯
2
th
at
o
b
ey
t
o
th
e
co
r
r
el
atio
n
m
ea
s
u
r
e
cr
iter
io
n
,
we
s
elec
t
t
h
e
o
n
e
th
at
m
a
x
im
izes
(
2
)
,
th
is
s
elec
ted
s
p
ac
e
2
is
th
e
m
o
s
t
d
is
c
r
im
in
atin
g
(
HC
S
)
,
th
is
p
r
o
ce
d
u
r
e
is
r
e
p
r
esen
ted
b
y
Fig
u
r
e
2
,
it
was
iter
ated
u
n
til
=
in
d
icatin
g
t
h
e
d
im
e
n
s
io
n
o
f
th
e
s
u
itab
le
(
HC
S
)
.
T
h
e
r
o
b
u
s
tn
ess
p
o
wer
o
f
a
ca
n
d
id
ate
(
HC
S
)
is
ev
alu
ated
b
y
th
e
m
ea
s
u
r
e
m
en
t
o
f
co
m
p
ac
tn
ess
an
d
s
e
p
ar
ab
ilit
y
o
f
th
e
i
n
v
o
lv
ed
cl
ass
es.
W
e
u
s
e
th
e
in
tr
a
-
class
d
is
p
er
s
io
n
m
atr
ix
(
)
g
iv
en
b
y
(
6
)
.
(
)
=
1
∗
∑
∑
(
,
=
1
=
1
−
)
(
,
−
)
(
6
)
W
h
er
e
=
[
1
,
.
.
.
,
,
.
.
.
,
]
ar
e
th
e
v
ec
to
r
co
n
tain
in
g
g
r
a
v
ities
ce
n
tr
e
co
r
r
esp
o
n
d
i
n
g
to
Nw
o
b
s
er
v
atio
n
,
an
d
t
h
en
,
s
er
v
es to
e
x
p
r
ess
clu
s
ter
by
(
7
)
.
=
1
∑
,
=
1
(
7
)
Fig
u
r
e
2
.
I
te
r
ativ
e
p
r
o
ce
s
s
o
f
b
u
ild
in
g
s
u
itab
le
h
y
b
r
id
co
lo
r
s
p
ac
e
In
8
r
e
p
r
esen
ts
th
e
s
ep
a
r
ab
ilit
y
m
ea
s
u
r
em
e
n
t
b
etwe
en
clu
s
ter
s
was
ev
alu
ated
b
ase
d
o
n
i
n
ter
class
d
is
p
er
s
io
n
m
atr
ix
(
)
:
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
iter
a
tive
a
lg
o
r
ith
m
fo
r
co
lo
r
s
p
a
ce
o
p
timiz
a
tio
n
o
n
ima
g
e
s
eg
men
ta
tio
n
(
Mo
u
r
a
d
Mo
u
s
s
a
)
203
(
)
=
1
∑
(
−
)
(
−
)
=
1
(
8
)
T
h
e
d
is
cr
im
in
atin
g
p
o
wer
(
)
o
f
ea
ch
co
lo
r
co
m
p
o
n
e
n
t
was
co
m
p
u
te
d
b
y
t
r
ac
e
cr
iter
io
n
,
at
t
h
is
en
d
ca
n
b
e
ex
p
r
ess
ed
b
y
(
9
)
:
(
)
=
(
(
(
)
+
(
)
)
−
1
(
)
=
(
(
)
(
)
)
(
9
)
wh
er
e
(
)
=
(
)
+
(
)
d
esig
n
ates
t
o
ta
l
d
is
p
er
s
io
n
m
atr
ix
.
Fo
r
ea
c
h
r
an
k
iter
atio
n
o
f
p
r
o
p
o
s
ed
p
r
o
ce
d
u
r
e,
m
o
s
t
d
is
cr
im
in
atin
g
h
y
b
r
i
d
co
lo
r
s
p
ac
e
is
th
e
s
p
ac
e
co
n
s
titu
ted
b
y
co
m
p
o
n
en
t
s
th
at
m
ax
im
ize
th
e
(
)
cr
iter
io
n
.
3.
CO
L
O
R
I
M
AG
E
S
E
G
M
E
N
T
AT
I
O
N
P
RO
CE
DURE
T
h
e
m
a
in
id
ea
o
f
b
ac
k
g
r
o
u
n
d
s
u
p
p
r
ess
io
n
is
to
co
m
p
ar
e
ea
c
h
im
ag
e
with
a
m
o
d
el
o
f
th
e
b
ac
k
g
r
o
u
n
d
wh
ich
is
th
e
r
ef
er
en
ce
im
ag
e.
E
ac
h
p
ix
el
o
f
th
e
ca
n
d
id
ate
im
ag
e
is
co
m
p
ar
ed
with
th
e
co
r
r
e
s
p
o
n
d
in
g
p
ix
el
in
th
e
m
o
d
el.
I
n
o
u
r
wo
r
k
,
m
o
d
el
o
f
p
ix
els d
e
n
s
ity
ar
e
est
im
ated
b
y
C
au
ch
y
d
is
tr
ib
u
tio
n
as sh
o
wn
in
(
10
)
.
(
,
)
=
1
[
(
−
)
2
+
2
]
(
1
0
)
So
f
o
r
ea
ch
p
ix
el
in
th
e
b
ac
k
g
r
o
u
n
d
,
r
ef
er
r
i
n
g
to
th
e
h
is
to
r
y
o
f
its
in
ten
s
ity
v
alu
es
an
d
f
o
r
th
e
th
r
ee
d
en
s
ities
f
u
n
ctio
n
s
b
ased
o
n
C
au
ch
y
d
is
tr
ib
u
tio
n
s
f
o
r
th
r
ee
co
lo
r
co
m
p
o
n
en
ts
r
ep
r
esen
tin
g
in
ten
s
ity
v
alu
es
(
R
,
G
an
d
B
)
o
f
co
lo
r
s
p
ac
e
o
f
th
r
ee
d
im
e
n
s
io
n
,
o
u
r
h
y
b
r
id
co
l
o
r
s
p
ac
e
it
is
n
ec
ess
ar
y
to
d
eter
m
in
e
th
e
f
iv
e
p
r
o
b
ab
ilit
y
d
is
tr
ib
u
tio
n
s
.
T
o
s
lo
w
th
e
v
a
r
iatio
n
o
f
in
te
n
s
ities
p
ix
el
o
n
th
e
s
ce
n
e,
i
n
(
11
)
r
e
p
r
esen
ts
ex
p
r
ess
io
n
o
f
th
e
r
ec
u
r
s
iv
e
f
ilter
to
u
p
d
ate
r
ec
u
r
s
iv
ely
th
e
b
ac
k
g
r
o
u
n
d
m
o
d
el
b
ec
au
s
e
o
f
o
c
cu
r
r
e
d
v
ar
iatio
n
o
f
in
ten
s
ities
v
alu
es:
+
1
=
(
1
−
)
+
(
1
1
)
wh
er
e
+
1
an
d
r
ep
r
esen
t
two
s
u
cc
ess
iv
e
in
ten
s
itie
s
v
alu
es
o
f
b
ac
k
g
r
o
u
n
d
p
ix
el
an
d
ad
ap
tatio
n
co
ef
f
icien
t
is
b
etwe
en
0
an
d
1
.
C
o
n
s
id
er
i
n
g
th
e
in
ten
s
ity
v
alu
e
o
f
a
p
ix
el
at
a
g
iv
en
tim
e
,
its
m
em
b
er
s
h
ip
p
r
o
b
a
b
ilit
y
(
)
at
th
is
tim
e
ca
n
b
e
ca
lcu
lated
b
y
(
12
)
:
(
)
=
1
∑
∏
[
(
−
)
2
]
=
1
=
1
(
12
)
wh
er
e
is
th
e
d
im
e
n
s
io
n
o
f
th
e
tr
ain
in
g
s
am
p
le,
th
e
d
im
e
n
s
io
n
o
f
th
e
co
lo
r
s
p
ac
e.
Acc
o
r
d
in
g
esti
m
ated
p
r
o
b
a
b
ilit
y
an
d
a
p
p
r
o
p
r
iate
th
r
esh
o
ld
,
ea
ch
p
ix
el
is
class
if
ied
in
to
p
ix
el
o
f
b
ac
k
g
r
o
u
n
d
o
r
m
o
v
in
g
p
ix
el.
4.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S AN
D
D
I
SC
USS
I
O
N
4
.
1
.
P
a
rt
ia
l da
t
a
lea
rning
We
h
av
e
im
p
lem
en
ted
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
with
m
an
y
lear
n
in
g
s
a
m
p
le.
in
a
f
ir
s
t
s
tep
we
h
av
e
r
an
d
o
m
l
y
s
elec
ted
p
ix
els
wh
ich
r
ep
r
esen
t
v
ar
io
u
s
clu
s
ter
s
t
h
at
will
b
e
s
p
ec
if
ied
,
s
o
th
at
f
o
r
ea
ch
clu
s
ter
Cj
,
we
co
n
s
id
er
ed
a
r
an
d
o
m
ly
s
am
p
le
co
n
s
titu
ted
b
y
th
e
m
o
s
t
r
ep
r
esen
tativ
e
3
5
0
p
ix
els
o
f
f
i
v
e
im
ag
es
th
at
r
ep
r
esen
t
th
e
tr
ea
ted
class
,
wh
ich
g
iv
es
a
to
tal
am
o
u
n
ts
o
f
p
ix
els
d
esig
n
ed
b
y
=
350
∗
5
ch
o
s
en
r
a
n
d
o
m
ly
.
Af
ter
b
u
ild
in
g
th
e
lear
n
in
g
m
atr
ix
with
o
b
s
er
v
atio
n
s
in
s
ev
er
al
co
lo
r
co
m
p
o
n
en
ts
wh
ich
we
will
c
h
o
o
s
e
th
e
m
o
s
t
d
is
cr
im
in
atin
g
o
n
e.
W
e
d
en
o
te
t
h
at
th
e
r
esu
lts
o
f
t
h
is
alg
o
r
ith
m
it'
s
n
o
t
s
tab
le
in
f
ac
t th
e
s
m
all
s
am
p
le
o
f
lear
n
in
g
p
i
x
els
lead
to
u
n
c
er
tain
ly
ch
r
o
m
atic
lev
el.
T
h
e
m
o
s
t
f
r
eq
u
en
tly
o
b
tain
e
d
co
m
p
o
n
en
ts
ar
e
L
(
L
uv
)
,
L
(
Lab
)
,
Y
(
Y
UV
)
,
Y
(
Y
IQ
)
.
4
.
2
.
F
ull
da
t
a
le
a
rning
T
h
e
s
ec
o
n
d
s
tep
c
o
n
s
is
ts
to
tak
e
ac
co
u
n
t
all
p
ix
els
th
at
co
n
s
titu
te
th
e
lear
n
in
g
im
a
g
e
o
f
ea
ch
clu
s
ter
,
b
u
t p
ix
els
n
u
m
b
e
r
r
elate
d
o
f
e
ac
h
im
ag
e
th
ey
a
r
e
n
o
t u
n
if
o
r
m
.
T
h
u
s
,
we
co
n
s
id
er
Nb
th
e
lo
w
n
u
m
b
er
o
f
p
ix
els
th
at
r
ep
r
esen
t
all
lea
r
n
in
g
im
a
g
e,
tak
e
n
ac
c
o
u
n
t
th
e
s
ize
o
f
h
an
d
led
s
am
p
le
we
h
av
e
ex
ten
d
ed
t
h
e
s
ets
o
f
d
at
a
an
d
th
en
b
u
ild
Nw
p
ix
els
co
n
s
titu
ted
b
y
∗
5
f
o
r
ea
ch
clu
s
ter
Cj
.
T
h
e
h
y
b
r
id
c
o
lo
u
r
c
o
m
p
o
n
en
t b
r
o
u
g
h
t
o
u
t
b
y
th
is
alg
o
r
ith
m
is
Y
(
XYZ
)
,
v
(
L
uv
)
,
S
(
HSL
)
.
4
.
3
.
Da
t
a
f
us
io
n t
heo
ry
T
h
e
m
ain
ly
i
d
ea
o
f
th
is
s
u
b
s
ec
tio
n
is
r
ed
u
cin
g
n
u
m
b
er
o
f
i
n
ter
esti
n
g
p
ix
els,
an
d
th
er
ea
f
ter
r
ed
u
c
i
n
g
th
e
lear
n
in
g
m
atr
ix
,
in
d
ee
d
th
e
co
n
s
id
er
ed
n
u
m
b
er
o
f
in
te
r
esti
n
g
p
ix
els
to
b
e
lear
n
e
d
f
o
r
ea
c
h
clu
s
ter
is
en
o
u
g
h
wid
e,
let
=
∗
5
wh
er
e
eq
u
a
l
to
1
1
2
3
,
s
in
ce
we
h
av
e
two
clu
s
ter
s
,
ea
ch
r
o
w
o
f
lear
n
in
g
m
atr
ix
co
n
tain
s
∗
5
∗
2
=
11230
p
ix
els
f
o
r
ea
ch
co
lo
r
co
m
p
o
n
e
n
t.
Fo
r
clu
s
ter
th
e
in
ter
esti
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
9
9
-
20
5
204
p
ix
els
co
r
r
esp
o
n
d
in
g
o
f
o
n
e
im
ag
e
am
o
n
g
th
e
f
iv
e
o
th
er
s
,
a
r
e
m
er
g
e
d
with
th
e
f
o
u
r
p
ix
el
s
is
s
u
ed
f
r
o
m
th
e
o
th
er
im
ag
e,
t
h
u
s
we
h
av
e
=
f
o
r
ea
ch
clu
s
ter
in
s
tead
o
f
=
∗
5
,
let
∗
2
p
ix
els
n
u
m
b
er
f
o
r
two
clu
s
ter
s
an
d
f
o
r
ea
ch
ch
r
o
m
atic
lev
el
in
s
tead
o
f
∗
5
∗
2
.
T
h
e
ess
en
tial
v
alu
e
th
at
we
s
ee
k
to
ca
lcu
late
is
th
e
lik
elih
o
o
d
f
u
n
ctio
n
(
|
)
,
wh
er
e
is
a
v
ec
to
r
co
n
s
titu
ted
b
y
th
e
ag
g
r
eg
ated
i
n
f
o
r
m
atio
n
f
r
o
m
d
if
f
er
en
t
s
o
u
r
ce
s
,
an
d
a
r
ea
l
f
ea
tu
r
e
o
f
th
e
esti
m
ated
s
y
s
tem
.
T
o
r
ea
ch
co
n
v
in
cin
g
f
u
s
io
n
r
e
s
u
lts
we
h
av
e
to
m
a
x
im
ize
th
is
lik
elih
o
o
d
f
u
n
c
tio
n
[
2
3
,
2
4
]
,
a
s
s
u
m
e
th
at
th
e
d
ata
s
o
u
r
ce
s
ar
e
co
n
d
itio
n
ally
in
d
e
p
en
d
en
t.
T
h
e
o
p
tim
al
p
a
r
am
eter
s
v
ec
to
r
∗
f
o
r
n
s
o
u
r
ce
s
is
d
ef
in
ed
b
y
(
13
)
:
∗
=
∑
−
1
∗
=
1
∑
−
1
=
1
(
1
3
)
wh
er
e
∗
is
th
e
esti
m
atio
n
m
ax
im
u
m
lik
elih
o
o
d
o
f
w
h
er
e
is
th
e
v
ar
ian
ce
r
elate
d
to
th
e
s
o
u
r
ce
m
ea
s
u
r
em
en
t
an
d
is
th
e
o
b
s
er
v
atio
n
p
r
o
v
id
ed
b
y
th
e
s
am
e
s
o
u
r
ce
.
T
h
e
ap
p
lied
p
r
o
ce
d
u
r
e
b
ased
o
n
d
ata
f
u
s
io
n
alg
o
r
ith
m
[
2
5
]
,
allo
ws
s
elec
tin
g
s
ix
h
y
b
r
id
ch
r
o
m
atic
lev
els,
(
)
,
(
)
,
(
)
,
(
)
,
(
)
an
d
(
)
.
Fin
ally
,
to
ev
alu
ate
t
h
e
ef
f
ec
t
iv
en
ess
o
f
o
u
r
p
r
o
p
o
s
ed
a
p
p
r
o
ac
h
f
o
r
h
y
b
r
id
ch
r
o
m
atic
lev
el
s
ec
tio
n
b
y
iter
ativ
e
alg
o
r
ith
m
,
we
h
av
e
u
s
ed
a
n
o
n
-
p
a
r
am
etr
ic
s
eg
m
e
n
tatio
n
tech
n
iq
u
e
b
ased
o
n
C
au
ch
y
d
is
tr
ib
u
tio
n
to
lear
n
th
e
h
is
to
r
ical
in
ten
s
ity
s
ta
te
o
f
ea
ch
p
ix
el
b
elo
n
g
i
n
g
to
b
a
ck
g
r
o
u
n
d
.
Fig
u
r
es
3
an
d
4
s
h
o
w
th
e
ef
f
ec
tiv
en
ess
o
f
o
b
tai
n
e
d
r
esu
lts
.
Fig
u
r
e
3
.
Or
ig
i
n
al
im
ag
e
o
f
th
e
f
ir
s
t c
lu
s
ter
.
(
a)
,
(
b
)
an
d
(
c)
:
s
eg
m
en
ted
im
ag
e
in
R
GB
s
p
a
ce
.
(
d
)
,
(
e
)
an
d
(
f
)
:
s
eg
m
en
tatio
n
r
esu
lts
with
f
u
ll lea
r
n
in
g
p
ix
els
.
(
g
)
,
(
h
)
,
(
i)
,
(
j)
,
(
k
)
an
d
(
l)
:
s
eg
m
en
tatio
n
r
esu
lts
with
p
ix
els f
u
s
io
n
Fig
u
r
e
4
.
Or
ig
i
n
al
im
ag
e
o
f
th
e
s
ec
o
n
d
clu
s
ter
;
(
a)
,
(
b
)
an
d
(
c)
:
s
eg
m
en
ted
im
ag
e
in
R
GB
s
p
ac
e.
(
d
)
,
(
e
)
an
d
(
f
)
:
s
eg
m
en
tatio
n
r
esu
lts
with
f
u
ll lea
r
n
in
g
p
ix
els
.
(
g
)
,
(
h
)
,
(
i)
,
(
j)
,
(
k
)
an
d
(
l)
:
s
eg
m
en
tatio
n
r
esu
lts
with
p
ix
els f
u
s
io
n
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
we
p
r
esen
ted
a
s
eg
m
en
tatio
n
m
et
h
o
d
b
y
s
tatis
tical
s
u
b
tr
ac
tio
n
o
f
th
e
b
ac
k
g
r
o
u
n
d
,
th
e
n
ew
tech
n
iq
u
e
u
s
es
th
e
C
au
ch
y
d
is
tr
ib
u
tio
n
to
m
o
d
el
an
d
s
u
b
tr
ac
t
th
e
d
en
s
ity
o
f
th
e
b
ac
k
g
r
o
u
n
d
p
ix
el
in
ten
s
ity
,
th
e
h
an
d
led
im
ag
e
ar
e
c
o
d
ed
in
h
y
b
r
id
c
o
lo
r
s
p
ac
e
b
ased
o
n
s
u
p
er
v
is
ed
an
d
iter
ativ
e
alg
o
r
ith
m
th
at
u
s
e
two
way
s
to
s
elec
t
s
ig
n
if
ican
t
ch
r
o
m
atic
lev
el,
th
e
f
ir
s
t
o
n
e
co
n
s
i
s
ts
to
lear
n
r
esp
ec
tiv
ely
th
e
p
a
r
tial
an
d
f
u
ll
am
o
u
n
t
o
f
p
ix
els,
th
e
s
ec
o
n
d
o
n
e
co
n
s
i
s
t
to
u
s
e
th
e
d
ata
f
u
s
io
n
t
h
eo
r
y
.
T
h
er
ef
o
r
e,
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
is
s
u
itab
le
f
o
r
m
an
y
ap
p
licatio
n
s
b
ased
o
n
co
lo
r
in
f
o
r
m
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
iter
a
tive
a
lg
o
r
ith
m
fo
r
co
lo
r
s
p
a
ce
o
p
timiz
a
tio
n
o
n
ima
g
e
s
eg
men
ta
tio
n
(
Mo
u
r
a
d
Mo
u
s
s
a
)
205
RE
F
E
R
E
NC
E
S
[1
]
G
.
M
.
Ba
sa
v
a
ra
j,
e
t
a
l.
,
“
Cro
w
d
An
o
m
a
ly
De
tec
ti
o
n
Us
i
n
g
M
o
ti
o
n
Ba
se
d
S
p
a
ti
o
-
Tem
p
o
ra
l
F
e
a
tu
re
An
a
ly
sis
Co
n
ten
t
,
"
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
,
v
ol
.
7
,
n
o
.
3
,
pp.
737
-
7
4
7
,
2
0
1
7
.
[2
]
G
.
Ted
d
y
S
u
r
y
a
,
e
t
a
l.
,
"
Art
ifi
c
ial
Ne
u
ra
l
Ne
two
rk
Ba
se
d
F
a
st
Ed
g
e
De
tec
ti
o
n
Al
g
o
r
it
h
m
f
o
r
M
RI
M
e
d
ica
l
Im
a
g
e
s
,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
7,
n
o
.
1
,
p
p
.
1
2
3
-
1
3
0
,
2
0
1
7
.
[3
]
S.
Riy
a
n
t
o
,
B.
Ali
Rid
h
o
,
S.
In
d
r
a
Ad
ji
,
A.
Ad
a
m
S
h
id
q
u
l
,
"
Au
t
o
m
a
ti
c
Ca
rd
iac
S
e
g
m
e
n
tatio
n
Us
in
g
Tri
a
n
g
le
a
n
d
Op
ti
c
a
l
F
lo
w
,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
E
n
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ie
n
c
e
,
v
o
l
.
8
,
n
o
.
2
,
p
p
.
315
-
3
2
6
,
2
0
1
7
.
[4
]
Q.
Xia
o
q
u
n
,
"
Im
a
g
e
S
e
g
m
e
n
tatio
n
Re
se
a
rc
h
Ba
se
d
o
n
G
A
a
n
d
Im
p
ro
v
e
d
Otsu
Al
g
o
ri
th
m
,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
,
v
o
l
.
7
,
n
o
.
2
,
p
p
.
5
3
3
-
5
4
1
,
2
0
1
7
.
[5
]
G
.
T
im
a
r
,
e
t
a
l
.
,
"
A
n
a
l
o
g
i
c
p
re
p
r
o
c
e
s
s
i
n
g
a
n
d
s
e
g
m
e
n
t
a
t
i
o
n
a
l
g
o
r
i
t
h
m
s
f
o
r
o
f
f
l
i
n
e
h
a
n
d
w
r
i
t
i
n
g
r
e
c
o
g
n
i
t
i
o
n
,
"
P
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
0
2
7
t
h
I
E
E
E
I
n
t
e
r
n
a
t
i
o
n
a
l
W
o
r
k
s
h
o
p
o
n
C
e
l
l
u
l
a
r
N
e
u
r
a
l
N
e
tw
o
r
k
s
a
n
d
T
h
e
i
r
A
p
p
l
i
c
a
t
i
o
n
s
,
2
0
0
3
.
[6
]
M.
M
o
u
ss
a
,
H.
El
Ou
n
i
,
A.
Do
u
ik
,
"
E
d
g
e
De
tec
ti
o
n
Ba
se
d
o
n
F
u
z
z
y
L
o
g
ic
a
n
d
H
y
b
ri
d
Ty
p
e
s o
f
S
h
a
n
n
o
n
En
tro
p
y
,
"
J
o
u
rn
a
l
o
f
Circ
u
it
s,
S
y
ste
ms
,
a
n
d
Co
mp
u
ter
s
,
v
o
l
.
2
9
,
n
o
.
1
4
,
2
0
2
0
.
[7
]
S
.
S
a
ich
a
n
d
a
n
a
,
K.
S
ri
n
iv
a
s,
R.
Kira
n
Ku
m
a
r,
"
Im
a
g
e
F
u
sio
n
i
n
Hy
p
e
rsp
e
c
tral
Im
a
g
e
Clas
sifica
ti
o
n
u
si
n
g
G
e
n
e
ti
c
Alg
o
rit
h
m
,
"
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
E
n
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
2
,
n
o
.
3
,
p
p
.
7
0
3
-
7
1
1
,
2
0
1
6
.
[8
]
P.
P
.
Ag
u
n
g
,
B.
A
g
u
s,
P
.
S
.
B
ib
,
"
M
o
d
e
li
n
g
S
i
n
g
u
lar
Va
lu
e
De
c
o
m
p
o
siti
o
n
a
n
d
K
-
M
e
a
n
s
o
f
Co
re
Im
a
g
e
i
n
Clas
ifi
c
a
ti
o
n
o
f
P
o
ten
ti
a
l
N
ic
k
e
l,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
,
v
o
l.
1
3
,
n
o
.
3
,
p
p
.
5
6
1
-
5
6
7
,
2
0
1
5
.
[9
]
T.
Ro
n
č
e
v
ić,
M
.
Bra
o
v
ić,
D.
S
ti
p
a
n
iče
v
,
"
No
n
-
P
a
ra
m
e
tri
c
Co
n
tex
t
-
Ba
se
d
Ob
jec
t
Clas
sifica
ti
o
n
in
Im
a
g
e
s,
"
J
o
u
rn
a
l
o
f
In
fo
rm
a
t
io
n
T
e
c
h
n
o
l
o
g
y
a
n
d
C
o
n
tro
l,
v
o
l
.
4
6
,
n
o
.
1
,
pp
.
8
6
-
9
9
,
2
0
1
7
.
[1
0
]
M
.
F
a
d
d
ly
,
H.
Afd
a
ll
y
n
a
,
S
.
S
a
if
u
lIzw
a
n
,
"
F
a
c
ial
Re
c
o
g
n
i
ti
o
n
i
n
M
u
lt
imo
d
a
l
Bi
o
m
e
tri
c
s S
y
ste
m
fo
r
F
in
g
e
r
Disa
b
led
Ap
p
li
c
a
n
ts,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
mp
u
ter
S
c
ien
c
e
,
v
o
l
.
6
,
n
o
.
3
,
p
p
.
6
3
8
-
6
4
5
,
2
0
1
7
.
[1
1
]
M.
M
o
u
ss
a,
M.
Ha
m
il
a
,
A.
Do
u
ik
,
"
A
N
o
v
e
l
F
a
c
e
Re
c
o
g
n
it
io
n
Ap
p
r
o
a
c
h
Ba
se
d
on
G
e
n
e
ti
c
Alg
o
rit
h
m
Op
ti
m
iza
ti
o
n
,
"
S
tu
d
ies
in
In
f
o
rm
a
ti
c
s
a
n
d
Co
n
tro
l
,
v
o
l.
2
7
,
n
o
.
1
,
p
p
.
1
2
7
-
1
3
4
,
2
0
1
8
.
[1
2
]
G.
H.
Li
u
,
J.
Y
.
Ya
n
g
,
"
Co
n
ten
t
-
b
a
se
d
ima
g
e
re
tri
e
v
a
l
u
si
n
g
c
o
lo
r
d
iffere
n
c
e
h
ist
o
g
ra
m
,
"
P
a
tt
e
rn
Rec
o
g
.
,
v
o
l.
4
6
,
n
o
.
1
,
p
p
.
1
8
8
-
1
9
8
,
2
0
1
3
.
[1
3
]
S
.
Ad
it
y
a
Ku
m
a
r,
S
.
G
a
n
d
h
a
rb
a
,
"
A
Re
v
iew
o
n
LS
B
S
u
b
st
it
u
t
io
n
a
n
d
P
VD
Ba
se
d
Im
a
g
e
S
teg
a
n
o
g
ra
p
h
y
Tec
h
n
iq
u
e
s,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
2,
n
o
.
3
,
p
p
.
7
1
2
-
7
1
9
,
2
0
1
6
.
[
1
4
]
P
.
M
o
h
a
m
m
a
d
R
a
s
o
u
l
,
H
.
A
l
i
,
"
B
l
i
n
d
S
t
e
g
a
n
o
g
r
a
p
h
y
i
n
C
o
l
o
r
I
m
a
g
e
s
b
y
D
o
u
b
l
e
W
a
v
e
l
e
t
T
r
a
n
s
f
o
r
m
a
n
d
I
m
p
r
o
v
e
d
A
r
n
o
l
d
T
r
a
n
s
f
o
r
m
,
"
I
n
d
o
n
e
s
i
a
n
J
o
u
r
n
a
l
o
f
E
l
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
,
n
o
.
2
,
p
p
.
5
8
6
-
6
0
0
,
2
0
1
6
.
[1
5
]
B.
Be
lea
n
,
M
.
S
trez
a
,
S
.
Cr
isa
n
,
S
.
Eme
rich
,
"
D
o
rsa
l
Ha
n
d
Ve
in
P
a
tt
e
rn
An
a
ly
sis
a
n
d
Ne
u
ra
l
Ne
two
r
k
s
fo
r
Bi
o
m
e
tri
c
Au
th
e
n
t
ica
ti
o
n
,
"
S
tu
d
ies
in
In
f
o
r
ma
ti
c
s
a
n
d
C
o
n
tr
o
l
,
v
o
l
.
2
6
,
n
o
.
3
,
p
p
.
3
0
5
-
3
1
4
,
2
0
1
7
.
[1
6
]
A.
M
ise
v
ičiu
s,
E.
S
tan
e
v
ičie
n
ė
,
"
A
Ne
w
Hy
b
ri
d
G
e
n
e
ti
c
Alg
o
rit
h
m
fo
r
t
h
e
G
re
y
P
a
tt
e
rn
Q
u
a
d
ra
ti
c
As
sig
n
m
e
n
t
P
ro
b
lem
,
"
J
o
u
r
n
a
l
o
f
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
a
n
d
C
o
n
tr
o
l
,
v
o
l.
4
7
,
n
o
.
3
,
p
p
.
5
0
3
-
5
2
0
,
2
0
1
8
.
[1
7
]
S
.
He
m
a
lath
a
,
U.
D.
Ac
h
a
ry
a
,
A.
Re
n
u
k
a
,
"
Co
m
p
a
riso
n
o
f
S
e
c
u
re
a
n
d
Hig
h
Ca
p
a
c
it
y
C
o
lo
r
Im
a
g
e
S
teg
a
n
o
g
ra
p
h
y
Tec
h
n
iq
u
e
s In
R
g
b
A
n
d
Yc
b
c
r
D
o
m
a
in
s,
"
In
t
.
J
.
o
f
A
d
v
.
In
fo
.
T
e
c
h
,
v
o
l.
3
,
n
o
.
3
,
p
p
.
2
8
6
-
2
9
1
,
2
0
1
3
.
[1
8
]
S
.
Ch
it
ra
a
n
d
G
.
Ba
lak
rish
n
a
n
,
"
Co
m
p
a
ra
ti
v
e
S
tu
d
y
f
o
r
Two
C
o
lo
r
S
p
a
c
e
s
HSCb
Cr
a
n
d
YCb
C
r
in
S
k
in
Co
l
o
r
De
tec
ti
o
n
,
"
A
p
p
l
ied
M
a
t
h
e
ma
ti
c
a
l
S
c
ien
c
e
,
v
o
l.
6
,
p
p
.
4
2
2
9
-
4
2
3
8
,
2
0
1
2
.
[1
9
]
L.
W
.
Re
n
,
M
.
S
.
Ab
d
Ra
h
m
a
n
,
A.
M
o
h
d
Ari
ffin
,
"
Clas
sifica
ti
o
n
o
f
P
a
rti
a
l
Disc
h
a
rg
e
S
o
u
rc
e
s
u
sin
g
S
tatisti
c
a
l
Ap
p
ro
a
c
h
,
"
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
,
v
o
l.
6
,
n
o
.
3
,
p
p
.
5
3
7
-
5
4
3
,
2
0
1
7
.
[2
0
]
G
.
He
,
H.
L
v
,
B
.
L
i,
Y
.
L
i,
"
Hy
p
e
rso
n
ic
Ve
h
icle
Trac
k
in
g
b
a
se
d
o
n
Im
p
r
o
v
e
d
Cu
r
re
n
t
S
tat
isti
c
a
l
M
o
d
e
l,
"
T
EL
KOM
NIKA
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
,
v
o
l.
1
1
,
n
o
.
1
1
,
p
p
.
6
3
0
9
-
6
3
1
4
,
2
0
1
3
.
[2
1
]
W.
Yu
tan
,
L.
Wen
b
in
,
P
.
S
h
u
a
i,
K.
Jia
n
g
m
in
g
,
"
S
e
g
m
e
n
tatio
n
M
e
t
h
o
d
o
f
Li
n
g
w
u
Lo
n
g
J
u
ju
b
e
s
Ba
se
d
o
n
L*
a
*
b
*
Co
lo
r
S
p
a
c
e
,
"
T
E
L
K
OM
NIKA
I
n
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
E
n
g
in
e
e
rin
g
,
v
o
l.
1
1
,
n
o
.
9
,
p
p
.
5
3
4
4
-
5
3
5
1
,
2
0
1
3
.
[2
2
]
Z.
Xin
b
o
,
L.
K
u
n
p
e
n
g
,
W.
Xia
o
l
in
g
,
Y.
Ch
a
n
g
h
o
n
g
,
"
M
o
v
i
n
g
S
h
a
d
o
w
Re
m
o
v
a
l
Alg
o
rit
h
m
Ba
se
d
o
n
HSV
C
o
l
o
r
S
p
a
c
e
,
"
T
E
L
KOM
NIKA
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
r
in
g
,
v
o
l.
1
2
,
n
o
.
4
,
p
p
.
2
7
69
-
2
7
7
5
,
2
0
1
4
.
[2
3
]
N.
El
fe
ll
y
,
J.
Y.
Die
u
lo
t
,
M
.
Be
n
re
jeb
,
P
.
B
o
rn
e
,
"
M
u
lt
imo
d
e
l
C
o
n
tr
o
l
De
sig
n
Us
in
g
Un
su
p
e
rv
ise
d
Clas
sifiers
,
"
S
tu
d
ies
in
In
f
o
rm
a
ti
c
s a
n
d
Co
n
tro
l
,
v
o
l.
2
1
,
n
o
.
1
,
p
p
.
1
0
1
-
1
0
8
,
2
0
1
2
.
[2
4
]
R.
Ath
i
lak
sh
m
i,
R.
Ra
jav
e
l
,
S
.
G
.
Ja
c
o
b,
"
F
u
si
o
n
F
e
a
tu
re
S
e
lec
ti
o
n
:
Ne
w
In
si
g
h
ts
in
t
o
F
e
a
tu
re
S
u
b
s
e
t
De
tec
ti
o
n
i
n
Bio
lo
g
ica
l
Da
ta M
in
in
g
,
"
S
t
u
d
ies
in
In
fo
rm
a
t
ics
a
n
d
C
o
n
tro
l
,
v
o
l
.
2
8
,
n
o
.
3
,
p
p
.
3
2
7
-
3
3
6
,
2
0
1
9
.
[2
5
]
B.
Ra
v
it
e
ja,
e
t
a
l
.
,
"
A
Ne
w
M
e
th
o
d
o
l
o
g
y
o
f
Hie
ra
rc
h
ica
l
Im
a
g
e
F
u
si
o
n
in
F
ra
m
e
wo
rk
f
o
r
H
y
p
e
rsp
e
c
tral
Im
a
g
e
S
e
g
m
e
n
tatio
n
,
"
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
mp
u
ter
S
c
ien
c
e
,
v
o
l.
6
,
n
o
.
1
,
p
p
.
5
8
-
6
5
,
2
0
1
7
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
R
Mo
u
r
a
d
M
o
u
ss
a
wa
s
b
o
r
n
i
n
Be
n
Aro
u
s,
T
u
n
isia
in
1
9
7
7
.
He
re
c
e
iv
e
d
t
h
e
M
a
ste
r
d
e
g
re
e
in
Au
to
m
a
ti
c
fro
m
th
e
"
Eco
le
Na
ti
o
n
a
led
I
n
g
n
ieu
rs d
e
M
o
n
a
stir"
i
n
2
0
0
6
a
n
d
t
h
e
P
h
.
D.
d
e
g
re
e
in
El
e
c
tri
c
a
l
En
g
e
n
e
e
rin
g
fro
m
th
e
s
a
m
e
sc
h
o
o
l
i
n
2
0
1
1
.
He
is
c
u
rre
n
t
ly
a
n
As
sista
n
t
P
ro
fe
ss
o
r
i
n
S
ig
n
a
l
a
n
d
Im
a
g
e
P
ro
c
e
ss
in
g
a
t
S
c
ien
c
e
s
fa
c
u
lt
y
o
f
G
a
fsa
.
He
is
a
lso
a
Re
se
a
rc
h
e
r
a
t
th
e
Ne
two
rk
e
d
Ob
jec
ts,
C
o
n
tr
o
l
a
n
d
Co
m
m
u
n
ica
ti
o
n
S
y
ste
m
s
(NO
CCS
)
Re
se
a
rc
h
Lab
o
ra
to
r
y
,
Na
ti
o
n
a
l
E
n
g
in
e
e
rin
g
S
c
h
o
o
l
o
f
S
o
u
ss
e
.
His
re
se
a
rc
h
is
re
late
d
to
I
m
a
g
e
a
n
d
v
id
e
o
P
r
o
c
e
ss
in
g
,
M
a
c
h
in
e
lea
rn
i
n
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.