T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
p
uti
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
24
,
No
.
1
,
Feb
r
u
ar
y
20
26
,
p
p
.
219
~
227
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
24
i
1
.
27338
219
J
o
ur
na
l ho
m
ep
a
g
e
:
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
O
pti
m
i
zed
edg
e
-
a
w
a
re f
requency
-
g
uided f
i
lteri
ng
f
o
r
robus
t
i
m
a
g
e denois
ing
I
m
a
n E
la
w
a
dy
1,
4
,
Ah
m
et
Re
ş
it
K
a
v
s
a
o
ğ
lu
2
,
İ
s
m
a
il Ra
k
ıp
K
a
ra
ş
3
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
-
El
e
c
t
r
o
n
i
c
s
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
N
a
t
u
r
a
l
S
c
i
e
n
c
e
s
,
K
a
r
a
b
u
k
U
n
i
v
e
r
si
t
y
,
K
a
r
a
b
ü
k
,
T
u
r
k
e
y
2
D
e
p
a
r
t
me
n
t
o
f
B
i
o
me
d
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
N
a
t
u
r
a
l
S
c
i
e
n
c
e
s
,
K
a
r
a
b
u
k
U
n
i
v
e
r
si
t
y
,
K
a
r
a
b
ü
k
,
T
u
r
k
e
y
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
C
o
mp
u
t
e
r
a
n
d
I
n
f
o
r
mat
i
o
n
S
c
i
e
n
c
e
s
,
K
a
r
a
b
u
k
U
n
i
v
e
r
si
t
y
,
K
a
r
a
b
u
k
,
T
u
r
k
e
y
4
De
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
L
i
e
t
u
v
o
s
I
n
ž
i
n
e
r
i
j
o
s K
o
l
e
g
i
j
a
,
K
a
u
n
a
s,
L
i
t
h
u
a
n
i
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
3
0
,
2
0
2
5
R
ev
i
s
ed
No
v
11
,
2
0
2
5
A
cc
ep
ted
Dec
8
,
2
0
2
5
T
h
e
p
ro
b
lem
o
f
d
e
n
o
isin
g
i
n
tru
si
o
n
is
sti
ll
o
f
g
re
a
t
c
o
n
c
e
rn
in
c
o
m
p
u
tati
o
n
a
l
im
a
g
in
g
b
e
c
a
u
se
o
f
th
e
trad
e
-
o
f
f
b
e
twe
e
n
n
o
ise
re
d
u
c
ti
o
n
a
n
d
im
a
g
e
stru
c
tu
re
a
n
d
d
e
tails
re
c
o
v
e
r
y
.
T
h
is
p
a
p
e
r
p
ro
p
o
se
s
a
n
o
p
t
im
ize
d
e
d
g
e
-
a
w
a
re
f
a
st
a
d
a
p
ti
v
e
g
u
id
e
d
f
il
ter
(E
-
F
A
G
F
)
c
o
m
b
in
in
g
w
a
v
e
let
-
d
o
m
a
in
d
e
c
o
m
p
o
siti
o
n
,
e
d
g
e
-
a
wa
re
n
e
ss
,
a
n
d
li
g
h
tw
e
ig
h
t
d
e
e
p
lea
rn
i
n
g
f
o
r
e
ff
i
c
ien
t
a
n
d
e
f
fe
c
ti
v
e
d
e
n
o
isin
g
.
T
h
e
b
i
o
r
th
o
g
o
n
a
l
w
a
v
e
let
tran
sf
o
r
m
is
e
m
p
lo
y
e
d
to
d
e
c
o
m
p
o
se
n
o
isy
im
a
g
e
s
in
to
l
o
w
-
a
n
d
h
ig
h
-
f
re
q
u
e
n
c
y
su
b
b
a
n
d
s
a
n
d
a
n
im
p
ro
v
e
d
e
d
g
e
-
a
tt
e
n
ti
o
n
m
a
p
f
o
r
se
lec
ti
v
e
h
ig
h
-
f
re
q
u
e
n
c
y
d
e
n
o
isin
g
.
Re
g
u
lariz
a
ti
o
n
p
a
ra
m
e
ters
a
re
e
stim
a
te
d
p
ix
e
l
-
w
ise
b
y
a
c
o
m
p
a
c
t
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
(
CNN
)
,
a
ll
o
w
in
g
sp
a
ti
a
l
-
v
a
r
y
in
g
f
il
terin
g
to
b
e
d
o
n
e
w
it
h
m
u
lt
i
-
sc
a
le
p
ro
c
e
ss
in
g
.
T
h
e
r
e
su
lt
a
n
t
E
-
F
AG
F
c
o
n
siste
n
tl
y
o
u
t
p
e
rf
o
rm
s
th
e
sta
te
o
f
th
e
a
rt
o
n
th
is
d
a
tas
e
t:
o
n
B
S
D5
0
0
f
o
r
sp
e
c
k
le
a
n
d
G
a
u
ss
ian
n
o
ise
(
p
e
a
k
sig
n
a
l
-
to
-
n
o
ise
ra
ti
o
(
P
S
NR
)
o
f
3
9
.
6
3
d
B
a
n
d
3
3
.
9
7
d
B,
re
sp
e
c
ti
v
e
ly
),
a
n
d
c
o
m
p
e
ti
ti
v
e
p
e
rf
o
r
m
a
n
c
e
f
o
r
P
o
isso
n
n
o
ise
(
3
0
.
8
4
d
B
)
a
l
a
rg
e
m
a
rg
in
c
o
m
p
a
re
d
to
th
e
re
fe
re
n
c
e
b
il
a
tera
l
a
n
d
n
o
n
-
l
o
c
a
l
m
e
a
n
s.
Ou
r
m
e
th
o
d
m
a
in
tain
s
h
ig
h
str
u
c
tu
ra
l
sim
il
a
ri
t
y
(u
p
to
0
.
9
7
in
stru
c
tu
ra
l
sim
il
a
rit
y
in
d
e
x
m
e
a
su
re
(
S
S
IM
)
)
,
ru
n
s
a
t
0
.
0
1
5
se
c
o
n
d
s
p
e
r
5
1
2
×
5
1
2
im
a
g
e
o
n
g
ra
p
h
ics
p
r
o
c
e
ss
in
g
u
n
it
(
GP
U
)
,
a
n
d
c
a
n
b
e
a
p
p
li
e
d
w
it
h
o
u
t
d
a
tas
e
t
-
sp
e
c
if
ic
train
in
g
.
T
h
e
se
re
su
lt
s
su
g
g
e
st
th
e
p
o
ss
ib
il
it
y
o
f
E
-
F
AG
F
to
a
c
h
iev
e
a
b
a
l
a
n
c
e
b
e
tw
e
e
n
c
las
si
c
a
l
e
ff
icie
n
c
y
a
n
d
lea
rn
in
g
-
b
a
se
d
a
d
a
p
tab
il
it
y
,
th
e
re
b
y
f
o
r
m
in
g
a
n
e
w
sc
e
n
a
r
io
to
c
o
m
b
in
e
f
a
st
a
n
d
re
li
a
b
le
ima
g
e
re
sto
ra
ti
o
n
f
o
r
a
c
tu
a
l
sc
e
n
a
ri
o
s.
K
ey
w
o
r
d
s
:
A
d
ap
tiv
e
f
il
ter
in
g
E
d
g
e
p
r
eser
v
atio
n
Gu
id
ed
f
ilter
I
m
ag
e
d
en
o
i
s
in
g
W
av
elet
tr
a
n
s
f
o
r
m
s
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
:
I
m
an
E
la
w
ad
y
Dep
ar
t
m
en
t o
f
E
lectr
ical
-
E
lect
r
o
n
ic
E
n
g
i
n
ee
r
i
n
g
,
Fac
u
lt
y
o
f
E
n
g
i
n
ee
r
i
n
g
a
n
d
Natu
r
a
l Sc
ie
n
ce
s
Kar
ab
u
k
U
n
iv
er
s
it
y
Kar
ab
ü
k
,
T
u
r
k
e
y
E
m
ail:
i
m
a
n
ela
w
d
y
@
k
ar
ab
u
k
.
ed
u
.
tr
1.
I
NT
RO
D
UCT
I
O
N
I
m
ag
e
d
e
n
o
is
i
n
g
is
a
f
u
n
d
a
m
e
n
tal
p
r
o
b
le
m
f
o
r
co
m
p
u
ter
v
i
s
io
n
an
d
s
i
g
n
al
p
r
o
ce
s
s
in
g
t
h
at
h
as
w
id
e
ap
p
licatio
n
s
i
n
m
an
y
f
ield
s
,
in
cl
u
d
in
g
m
ed
ical
i
m
a
g
in
g
[
1
]
,
au
to
n
o
m
o
u
s
v
eh
icles
[
2
]
,
s
atellite
an
d
ae
r
ia
l
s
en
s
in
g
[
3
]
,
an
d
lo
w
-
li
g
h
t
p
h
o
to
g
r
ap
h
y
[
4
]
.
T
h
e
b
asic
p
r
o
b
le
m
o
f
n
o
is
e
i
s
eli
m
i
n
ati
n
g
th
e
ef
f
ec
ts
ca
u
s
ed
b
y
s
en
s
o
r
li
m
itatio
n
an
d
lo
w
li
g
h
t
lev
el
ac
q
u
is
i
tio
n
,
w
h
ich
ar
e
s
u
b
s
eq
u
en
t
l
y
cr
ea
ted
d
u
r
in
g
s
ig
n
a
l
tr
an
s
m
i
s
s
io
n
w
h
ile
m
ain
tain
in
g
th
e
s
tr
u
ct
u
r
e,
tex
tu
r
e
,
an
d
s
e
m
a
n
tic
i
n
f
o
r
m
atio
n
es
s
en
tial
f
o
r
f
u
r
th
er
p
r
o
ce
s
s
in
g
[
5
]
,
[
6
]
.
C
las
s
ical
li
n
ea
r
f
i
lter
s
,
s
u
ch
a
s
th
e
m
ea
n
a
n
d
b
o
x
,
ar
e
co
m
p
u
tatio
n
all
y
s
i
m
p
le,
b
u
t
t
h
e
y
eq
u
all
y
b
lu
r
n
o
is
e
a
n
d
i
m
p
o
r
tan
t
i
n
f
o
r
m
atio
n
,
p
r
o
v
id
in
g
p
o
o
r
p
er
f
o
r
m
a
n
ce
,
f
o
r
ex
a
m
p
le,
at
t
h
e
lo
ca
tio
n
o
f
ed
g
es
[
4
]
,
[
7
]
.
N
o
n
l
in
e
a
r
m
e
th
o
d
s
,
n
am
ely
m
e
d
i
an
[
8
]
a
n
d
b
i
l
at
e
r
a
l
f
i
lt
e
r
in
g
[
9
]
,
h
av
e
b
e
e
n
u
s
e
d
f
o
r
t
h
e
i
r
c
a
p
a
b
i
li
ty
o
f
d
en
o
is
in
g
a
n
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
2
1
9
-
2
2
7
220
b
e
t
t
e
r
p
r
e
s
e
r
v
in
g
l
o
c
a
l
s
t
r
u
c
tu
r
e
s
.
B
i
l
a
t
e
r
a
l
f
i
lt
e
r
s
,
f
o
r
e
x
am
p
l
e
,
u
s
e
b
o
t
h
jo
i
n
t
s
p
a
ti
a
l
an
d
i
n
t
en
s
i
ty
-
d
o
m
a
in
w
ei
g
h
t
in
g
an
d
c
an
b
e
tt
e
r
p
r
e
s
er
v
e
e
d
g
es
th
an
p
o
s
s
i
b
l
e
r
esu
l
ts
u
s
in
g
c
o
m
p
l
e
t
ely
l
in
e
a
r
t
e
ch
n
iq
u
e
s
[
1
0
]
.
H
o
w
ev
e
r
,
s
in
c
e
th
ey
a
r
e
s
en
s
i
ti
v
e
t
o
th
e
p
a
r
am
et
e
r
s
,
th
ey
e
a
s
i
ly
o
v
e
r
s
m
o
o
t
h
d
e
t
ai
ls
am
o
n
g
t
h
e
c
o
m
p
l
ex
n
o
i
s
e
[
1
1
]
.
Do
m
ain
tr
a
n
s
f
o
r
m
s
,
e.
g
.
,
w
av
elet
tech
n
iq
u
es,
esp
ec
iall
y
d
is
cr
ete
w
av
ele
t
tr
an
s
f
o
r
m
s
(
DW
T
)
,
ch
an
g
ed
th
e
p
ar
ad
ig
m
b
y
i
n
c
o
r
p
o
r
atin
g
d
if
f
er
e
n
t
f
r
eq
u
e
n
c
y
b
an
d
s
b
y
tr
a
n
s
f
o
r
m
i
n
g
i
m
a
g
es
i
n
to
m
u
lt
i
-
s
ca
le
f
r
eq
u
en
c
y
co
m
p
o
n
e
n
ts
[
1
2
]
,
[
1
3
]
.
W
av
elet
th
r
esh
o
ld
i
n
g
an
d
s
h
r
in
k
ag
e
m
et
h
o
d
s
ex
p
lo
it
th
e
n
atu
r
al
s
p
ar
s
it
y
o
f
clea
n
i
m
a
g
es
i
n
tr
an
s
f
o
r
m
ed
d
o
m
a
in
s
in
o
r
d
er
to
s
ep
ar
ate
n
o
is
e
f
r
o
m
s
i
g
n
al
[
1
4
]
.
Ho
w
e
v
er
,
th
ese
ap
p
r
o
ac
h
es
ar
e
f
r
eq
u
en
tl
y
b
ased
o
n
g
lo
b
a
l
th
r
esh
o
ld
s
o
r
h
an
d
-
d
es
i
g
n
ed
p
r
io
r
s
,
w
h
ic
h
ar
e
u
n
s
u
itab
le
f
o
r
r
ea
l
im
a
g
e
s
w
i
th
s
p
atiall
y
v
ar
ian
t
n
o
is
e
s
tati
s
ti
cs
[
1
5
]
,
[
1
6
]
.
I
n
ad
d
itio
n
,
th
e
f
ix
ed
b
ases
o
f
clas
s
ical
w
a
v
elets
m
a
y
f
ail
to
r
ep
r
esen
t
h
i
g
h
l
y
n
o
n
-
s
tatio
n
ar
y
n
a
tu
r
al
s
ce
n
e
s
[
1
7
]
.
No
n
-
lo
c
al
m
et
h
o
d
s
,
s
u
ch
as
n
o
n
-
lo
ca
l
m
ea
n
s
(
N
L
M
)
[
1
8
]
o
r
th
e
p
io
n
ee
r
in
g
b
lo
ck
m
a
tch
in
g
3
D
(
B
M3
D)
f
ilter
[
1
9
]
,
m
ak
e
u
s
e
o
f
s
el
f
-
s
i
m
i
lar
it
y
b
et
wee
n
i
m
ag
e
p
atch
e
s
.
I
n
p
ar
ticu
lar
,
B
M3
D
[
8
]
h
as
p
r
o
v
ed
its
el
f
as
a
s
tate
-
of
-
t
h
e
-
a
r
t
m
et
h
o
d
in
t
h
e
f
ield
o
f
n
atu
r
al
i
m
a
g
e
d
en
o
i
s
i
n
g
,
w
h
ic
h
is
s
u
p
er
io
r
to
th
e
r
ese
ar
ch
o
f
th
is
f
ield
in
ter
m
s
o
f
d
en
o
is
i
n
g
p
er
f
o
r
m
a
n
ce
w
it
h
th
e
co
s
t
o
f
h
i
g
h
co
m
p
u
tatio
n
al
co
m
p
le
x
it
y
a
n
d
k
ee
p
in
g
tu
n
i
n
g
i
n
m
an
y
h
y
p
er
-
p
ar
a
m
eter
s
[
2
0
]
,
[
2
1
]
.
T
h
e
ad
v
en
t
o
f
d
ata
e
x
p
lo
s
io
n
an
d
d
ee
p
lear
n
i
n
g
h
as
a
ls
o
r
ev
o
lu
tio
n
ized
th
e
d
en
o
i
s
i
n
g
f
ield
.
(
1
)
Neu
r
al
n
et
w
o
r
k
-
b
ased
(
NN)
ap
p
r
o
ac
h
es,
s
u
ch
as
th
e
d
ee
p
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
et
wo
r
k
s
(
C
NNs
)
[
2
2
]
,
FF
DNe
t
[
2
3
]
)
an
d
,
m
o
r
e
r
ec
e
n
tl
y
,
t
h
e
tr
an
s
f
o
r
m
er
-
b
ased
m
eth
o
d
s
(
S
w
in
I
R
[
2
4
]
)
,
h
av
e
ac
h
iev
ed
s
i
g
n
i
f
ican
t
d
en
o
is
in
g
p
er
f
o
r
m
a
n
ce
b
y
l
ea
r
n
in
g
m
ap
p
in
g
s
f
r
o
m
n
o
is
y
to
clea
n
i
m
ag
e
s
,
an
d
ar
e
o
f
ten
s
u
p
er
io
r
to
co
n
v
e
n
tio
n
al
m
et
h
o
d
s
o
n
co
m
m
o
n
l
y
u
s
ed
b
en
ch
m
ar
k
s
[
2
5
]
,
[
2
6
]
.
W
h
ile
th
ese
d
ee
p
m
o
d
el
s
ca
n
ac
h
ie
v
e
h
i
g
h
p
er
f
o
r
m
a
n
ce
o
n
s
e
v
er
al
tas
k
s
,
th
e
y
h
a
v
e
s
e
v
er
al
s
h
o
r
tco
m
i
n
g
s
:
t
h
e
y
n
ee
d
lar
g
e
-
s
ca
le
an
d
h
ig
h
-
acc
u
r
ate
an
n
o
tated
d
atasets
f
o
r
s
u
p
er
v
i
s
ed
tr
ain
in
g
[
2
7
]
,
m
a
y
s
u
f
f
er
f
r
o
m
a
g
e
n
er
aliza
tio
n
p
r
o
b
lem
ab
o
u
t
d
if
f
er
en
t
o
r
u
n
k
n
o
w
n
n
o
is
e
t
y
p
e
s
[
2
8
]
,
an
d
ar
e
co
m
p
u
tatio
n
al
l
y
e
x
p
en
s
i
v
e,
w
h
ich
h
i
n
d
er
s
r
ea
l
-
ti
m
e
u
s
e
[
2
9
]
.
Mo
r
eo
v
er
,
th
e
“
b
lac
k
b
o
x
”
p
r
o
p
er
ty
o
f
d
ee
p
n
et
w
o
r
k
s
p
o
s
es
in
ter
p
r
etab
ilit
y
an
d
tr
an
s
p
ar
en
c
y
c
h
alle
n
g
e
s
,
p
ar
ticu
lar
l
y
i
n
s
af
et
y
-
c
r
itical
ap
p
licatio
n
s
s
u
c
h
as h
ea
lth
ca
r
e
a
n
d
au
to
n
o
m
o
u
s
s
y
s
te
m
s
[
3
0
]
,
[
3
1
]
.
R
ec
en
t
r
esear
ch
h
as
s
h
o
w
n
a
ten
d
en
c
y
to
s
tr
i
k
e
a
b
alan
ce
b
et
w
ee
n
in
ter
p
r
etab
ilit
y
a
n
d
p
er
f
o
r
m
an
ce
b
y
co
m
b
i
n
i
n
g
cla
s
s
ical
s
i
g
n
al
p
r
o
ce
s
s
in
g
w
it
h
co
n
te
m
p
o
r
ar
y
lear
n
in
g
-
b
ased
tech
n
iq
u
es
[
3
2
]
.
Hy
b
r
id
m
et
h
o
d
s
f
u
s
e
w
av
elet
o
r
f
r
eq
u
en
c
y
-
d
o
m
ai
n
d
ec
o
m
p
o
s
it
io
n
w
it
h
NN
[
3
3
]
,
[
3
4
]
an
d
d
ev
elo
p
s
p
atiall
y
v
ar
iab
le
f
il
ter
s
b
ased
o
n
lear
n
in
g
d
ee
p
f
ea
t
u
r
es
[
3
5
]
.
T
h
e
atten
tio
n
m
ec
h
a
n
is
m
s
s
u
ch
a
s
ed
g
e
-
a
w
ar
e
an
d
f
r
eq
u
en
c
y
-
a
w
ar
e
m
o
d
u
les
h
a
v
e
ac
h
iev
ed
s
u
cc
e
s
s
f
o
r
s
alie
n
t
i
n
f
o
r
m
atio
n
f
o
c
u
s
i
n
g
an
d
lo
ca
l
r
esp
o
n
s
e
m
o
d
u
latio
n
;
h
o
w
e
v
er
,
th
e
y
m
i
g
h
t b
e
less
f
le
x
ib
le
in
c
o
p
in
g
w
it
h
ar
b
itra
r
y
i
m
a
g
e
co
n
ten
t
w
it
h
d
iv
er
s
e
n
o
is
e
d
is
tr
i
b
u
tio
n
s
[
3
6
]
–
[
3
9
]
.
Nev
er
th
e
less
,
s
o
m
e
c
h
as
m
s
c
o
n
tin
u
e
to
e
x
is
t.
Mo
s
t
‘
d
ee
p
’
d
en
o
is
er
s
ar
e
i
n
d
iv
id
u
all
y
d
ef
i
n
ed
f
o
r
s
o
m
e
n
o
i
s
e
(
Gau
s
s
ia
n
an
d
P
o
is
s
o
n
)
an
d
f
r
eq
u
en
tl
y
p
er
f
o
r
m
p
o
o
r
ly
o
n
m
ix
ed
an
d
s
p
atial
l
y
v
ar
ian
t
r
ea
l
-
n
o
i
s
y
i
m
a
g
es
[
4
0
]
,
[
4
1
]
.
C
la
s
s
ical
f
i
lter
s
,
alth
o
u
g
h
i
n
t
er
p
r
etab
le
an
d
co
m
p
u
tat
io
n
all
y
ef
f
ic
ien
t,
d
o
n
o
t
ad
j
u
s
t
th
e
ir
p
ar
am
eter
s
in
a
lo
ca
l
m
a
n
n
er
an
d
en
co
u
n
ter
d
if
f
ic
u
lt
ies
w
it
h
th
e
co
m
p
r
o
m
i
s
e
b
et
w
ee
n
ed
g
e
p
r
eser
v
atio
n
an
d
n
o
is
e
s
u
p
p
r
ess
io
n
[
4
2
]
.
Vis
io
n
tr
an
s
f
o
r
m
er
s
an
d
lar
g
e
n
et
w
o
r
k
s
ac
h
ie
v
e
i
m
p
r
ess
i
v
e
p
er
f
o
r
m
a
n
ce
;
h
o
w
e
v
er
,
th
e
y
r
eq
u
ir
e
en
o
r
m
o
u
s
co
m
p
u
t
atio
n
co
s
ts
,
m
ak
i
n
g
th
e
m
i
n
co
n
v
e
n
ie
n
t
f
o
r
ed
g
e
d
ev
ices i
n
r
ea
l
ti
m
e.
I
n
s
p
ir
ed
b
y
t
h
ese
o
b
s
er
v
atio
n
s
,
w
e
p
r
esen
t
an
o
p
ti
m
ized
ed
g
e
-
a
w
ar
e
f
ast
ad
ap
tiv
e
g
u
id
ed
f
ilter
(
E
-
F
A
GF)
,
a
n
o
v
el
al
g
o
r
ith
m
t
h
at
ai
m
s
to
co
m
b
in
e
t
h
e
i
n
t
er
p
r
etab
ilit
y
f
r
o
m
w
av
elet
tr
an
s
f
o
r
m
s
,
th
e
lo
ca
l
ad
ap
tiv
it
y
f
r
o
m
ed
g
e
-
a
w
ar
e
f
i
lter
in
g
,
an
d
t
h
e
e
x
p
r
ess
i
v
en
e
s
s
f
r
o
m
d
ee
p
lear
n
i
n
g
w
it
h
o
u
t
h
av
in
g
to
r
e
-
tr
ain
w
it
h
s
p
ec
ial
l
y
p
r
ep
ar
ed
d
atasets
o
r
m
an
u
all
y
s
et
p
ar
a
m
e
ter
s
.
E
-
F
AGF
u
s
es
b
io
r
th
o
g
o
n
al
w
a
v
elets
to
b
r
ea
k
d
o
w
n
i
m
ag
e
s
an
d
d
if
f
er
en
t
ia
te
b
et
w
ee
n
n
o
is
e
an
d
s
tr
u
ct
u
r
e,
g
en
er
ate
s
ed
g
e
-
at
ten
t
io
n
m
ap
s
f
o
r
f
u
r
t
h
er
lear
n
in
g
t
h
r
o
u
g
h
co
n
v
o
lu
t
io
n
al
n
et
w
o
r
k
s
,
a
n
d
u
s
es
a
li
g
h
t
C
NN
to
p
r
ed
ict
s
p
atiall
y
v
ar
y
in
g
r
e
g
u
lar
izatio
n
p
ar
am
eter
s
.
T
h
e
m
et
h
o
d
p
r
o
v
id
es
g
o
o
d
p
er
f
o
r
m
a
n
ce
f
o
r
ad
d
itiv
es,
P
o
is
s
o
n
,
a
n
d
m
u
lti
p
licativ
e
n
o
is
e
o
n
n
atu
r
al
i
m
a
g
es
v
ia
m
u
lti
-
s
ca
l
e
g
u
id
ed
f
ilte
r
in
g
a
n
d
ad
ap
tiv
e
f
u
s
io
n
.
W
e
s
h
o
w
th
at
t
h
i
s
j
o
in
t
f
r
a
m
e
w
o
r
k
n
o
t
o
n
l
y
ac
h
iev
e
s
s
tate
-
of
-
t
h
e
-
ar
t
p
er
f
o
r
m
a
n
ce
o
n
th
e
B
SD5
0
0
d
ataset
b
u
t
also
h
as
a
v
er
y
ef
f
icie
n
t
s
p
ee
d
o
n
GP
U,
w
h
ich
f
ills
th
e
g
ap
f
o
r
r
ea
l
-
ti
m
e
h
i
g
h
-
q
u
a
lit
y
i
m
ag
e
r
esto
r
atio
n
i
n
p
r
ac
tical
ap
p
li
ca
ti
o
n
s
.
T
h
e
r
e
m
ai
n
d
er
o
f
th
is
p
ap
er
is
o
r
g
an
ized
a
s
f
o
llo
w
s
:
s
ec
tio
n
2
i
n
tr
o
d
u
ce
s
t
h
e
to
o
ls
an
d
m
ater
ial
s
u
s
ed
,
s
ec
tio
n
3
d
escr
ib
es t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
,
r
es
u
lt
s
an
d
d
is
cu
s
s
io
n
ar
e
p
r
o
v
id
ed
in
s
ec
tio
n
4
,
an
d
s
ec
tio
n
5
co
n
clu
d
es
a
n
d
g
i
v
e
s
f
u
r
t
h
er
w
o
r
k
.
2.
T
O
O
L
S AN
D
M
AT
E
RIA
L
S
W
av
elet
tr
an
s
f
o
r
m
s
h
av
e
b
ee
n
a
v
ital p
la
y
er
i
n
m
u
lti
-
r
eso
lu
t
i
o
n
an
al
y
s
i
s
an
d
i
m
a
g
e
d
en
o
is
i
n
g
,
o
w
i
n
g
to
th
eir
ca
p
ab
ilit
y
o
f
p
r
eser
v
i
n
g
f
r
eq
u
e
n
c
y
an
d
s
p
atial
i
n
f
o
r
m
atio
n
to
g
et
h
er
[
1
2
]
,
[
1
3
]
.
T
h
e
DW
T
allo
w
s
a
n
i
m
a
g
e
I
to
b
e
d
ec
o
m
p
o
s
ed
in
t
o
a
co
llectio
n
o
f
f
r
eq
u
e
n
c
y
s
u
b
-
b
an
d
s
r
ep
r
esen
ti
n
g
d
if
f
er
e
n
t
o
r
ien
ta
tio
n
s
an
d
s
ca
les.
Ma
t
h
e
m
a
ticall
y
,
f
o
r
a
s
in
g
le
-
lev
el
d
ec
o
m
p
o
s
it
io
n
:
,
=
DW
T
(
)
(
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Op
timiz
ed
ed
g
e
-
a
w
a
r
e
fr
eq
u
en
cy
-
g
u
id
e
d
filt
erin
g
fo
r
r
o
b
u
s
t ima
g
e
d
en
o
is
in
g
(
I
ma
n
E
la
w
a
d
y
)
221
w
h
er
e
is
t
h
e
lo
w
-
f
r
eq
u
e
n
c
y
(
ap
p
r
o
x
i
m
atio
n
)
co
m
p
o
n
e
n
t
an
d
=
{
ℎ
,
,
}
co
n
tain
s
th
e
h
o
r
izo
n
tal,
v
er
tical,
a
n
d
d
iag
o
n
al
d
etail
co
ef
f
icie
n
ts
,
r
esp
ec
t
iv
el
y
.
T
h
is
s
ep
ar
atio
n
allo
w
s
tar
g
eted
d
en
o
is
i
n
g
s
tr
ateg
ie
s
:
n
o
is
e
o
f
te
n
p
r
ed
o
m
in
ates
in
h
i
g
h
-
f
r
eq
u
e
n
c
y
s
u
b
b
an
d
s
,
w
h
ile
s
tr
u
ctu
r
al
i
m
a
g
e
c
o
n
ten
t
i
s
r
etai
n
ed
in
th
e
lo
w
-
f
r
eq
u
e
n
c
y
b
an
d
[
1
4
]
,
[
2
2
]
.
R
ec
o
n
s
tr
u
ctio
n
i
s
ac
h
i
ev
ed
v
ia
th
e
i
n
v
er
s
e
d
is
cr
ete
w
av
e
let
tr
an
s
f
o
r
m
(
I
DW
T
)
:
=
I
DWT
(
,
)
(
2
)
W
e
em
p
lo
y
b
io
r
th
o
g
o
n
al
w
av
elets
(
“
b
io
r
6
.
8
”
)
d
u
e
to
th
eir
s
y
m
m
etr
y
a
n
d
en
h
a
n
ce
d
en
er
g
y
co
m
p
ac
tio
n
[
1
3
]
,
alig
n
in
g
w
it
h
o
u
r
i
m
p
le
m
en
tat
io
n
in
P
y
T
o
r
ch
-
W
av
elet
s
.
2
.
1
.
E
dg
e
a
t
t
ent
i
on
Fo
r
h
ig
h
-
q
u
alit
y
i
m
a
g
e
r
esto
r
atio
n
,
ed
g
e
i
n
f
o
r
m
atio
n
i
s
es
s
en
tial
b
ec
au
s
e
to
o
m
u
ch
ed
g
e
s
m
o
o
th
in
g
d
eg
r
ad
es
p
er
ce
p
tio
n
[
9
]
,
[
2
3
]
.
Gr
ad
ien
t
o
p
er
ato
r
s
lik
e
So
b
el
o
r
P
r
ew
it
t
f
ilter
s
ar
e
co
m
m
o
n
l
y
u
s
ed
i
n
ed
g
e
d
etec
tio
n
.
T
h
e
g
r
ad
ien
t
m
a
g
n
it
u
d
e
m
ap
G
f
o
r
i
m
a
g
e
I
ca
n
b
e
d
escr
ib
ed
as f
o
llo
w
s
:
=
√
(
∗
)
2
+
(
∗
)
2
(
3
)
w
h
er
e
∗
in
d
icate
s
co
n
v
o
lu
tio
n
an
d
an
d
s
tan
d
f
o
r
t
h
e
h
o
r
izo
n
tal
an
d
v
er
tica
l
So
b
el
k
er
n
els
,
r
esp
ec
tiv
el
y
.
R
ec
en
t
l
y
,
C
NN
s
h
av
e
b
ee
n
u
s
ed
to
f
u
r
th
er
r
e
f
in
e
ed
g
e
m
ap
s
i
n
o
r
d
er
to
ad
ap
tiv
el
y
h
i
g
h
lig
h
t
s
alie
n
t
b
o
u
n
d
ar
ies
w
h
ile
r
ed
u
ci
n
g
th
e
i
m
p
ac
t
o
f
n
o
is
e
[
2
3
]
,
[
3
6
]
.
A
m
at
h
e
m
at
ical
m
o
d
el
o
f
th
e
r
esu
lta
n
t
ed
g
e
atten
tio
n
m
ap
A
is
a
s
f
o
llo
w
s
:
=
[
−
1
0
+
1
−
2
0
+
2
−
1
0
+
1
]
,
=
[
−
1
−
2
−
1
0
0
0
+
1
+
2
+
1
]
(
4
)
=
(
C
o
n
v
3
×
3
(
R
eLU
(
C
o
n
v
3
×
3
(
)
)
)
)
(
5
)
w
h
er
e
is
th
e
s
i
g
m
o
id
f
u
n
ct
i
o
n
an
d
C
o
n
v
3
×
3
d
en
o
tes
a
co
n
v
o
lu
t
i
o
n
al
la
y
er
w
it
h
a
3
×3
k
er
n
el.
W
e
u
s
e
s
ig
m
o
id
ac
tiv
atio
n
f
o
r
atten
tio
n
s
ca
li
n
g
an
d
a
lear
n
ed
ed
g
e
k
er
n
el
an
d
a
n
o
n
lin
ea
r
r
ef
i
n
e
m
e
n
t
m
o
d
u
le
to
m
a
k
e
th
is
w
o
r
k
in
o
u
r
co
d
e.
2
.
2
.
Ada
ptiv
e
re
g
ula
ri
z
a
t
io
n
R
eg
u
lar
izatio
n
is
a
ce
n
tr
al
asp
ec
t
o
f
im
a
g
e
r
esto
r
atio
n
,
w
h
er
eb
y
t
h
e
tr
ad
e
-
o
f
f
b
et
w
ee
n
f
id
e
lit
y
to
th
e
n
o
is
y
o
b
s
er
v
atio
n
a
n
d
s
m
o
o
th
n
es
s
o
r
s
p
ar
s
it
y
o
f
th
e
e
s
ti
m
at
ed
i
m
ag
e
i
s
t
u
n
ed
[
1
5
]
,
[
2
4
]
.
On
e
o
f
t
h
e
g
o
als
i
n
th
is
s
t
u
d
y
i
s
to
p
r
ed
ict
th
e
r
eg
u
lar
iz
atio
n
w
eig
h
t
lo
ca
ll
y
i
n
a
n
ad
ap
tiv
e
f
ilter
i
n
g
m
o
d
el,
s
o
th
at
t
h
e
al
g
o
r
ith
m
ca
n
ad
ap
t
to
d
if
f
er
en
t
s
i
g
n
al
an
d
n
o
is
e
co
n
d
itio
n
s
.
L
et
b
e
th
e
r
e
g
u
lar
izi
n
g
p
ar
a
m
eter
,
p
o
s
s
ib
l
y
s
p
atia
ll
y
d
ep
en
d
en
t:
=
(
)
(
6
)
w
h
er
e
is
a
lear
n
ab
le
m
ap
p
in
g
th
at
i
s
o
f
ten
m
ad
e
b
y
a
s
m
all
C
NN
th
a
t
tak
es
i
n
lo
ca
l
im
ag
e
s
tatis
tics
li
k
e
g
r
ad
ien
t
m
a
g
n
itu
d
e
o
r
v
ar
ia
n
ce
a
n
d
is
t
h
e
p
ix
e
l
lo
ca
ti
o
n
[
2
4
]
,
[
3
5
]
.
W
e
u
s
e
a
s
m
all
C
NN
ca
lled
“
Fa
s
tP
ar
a
m
Net
”
to
f
in
d
lo
ca
l
g
r
ad
ien
t
a
n
d
v
ar
ian
ce
f
ea
t
u
r
es
an
d
m
a
k
e
t
w
o
ad
ap
tiv
e
r
eg
u
l
ar
izatio
n
m
ap
s
(
o
n
e
f
o
r
lo
w
f
r
eq
u
en
c
y
a
n
d
o
n
e
f
o
r
h
ig
h
f
r
eq
u
en
c
y
)
t
h
at
ar
e
u
s
ed
i
n
th
e
f
ilter
i
n
g
p
r
o
ce
s
s
.
2
.
3
.
E
v
a
lua
t
i
o
n
m
et
ric
s
Fo
r
ev
alu
atio
n
m
etr
ics
w
e
u
s
ed
p
ea
k
s
ig
n
al
-
to
-
n
o
is
e
r
a
tio
(
P
SNR
)
an
d
s
tr
u
ct
u
r
al
s
i
m
ila
r
it
y
in
d
e
x
(
SS
I
M)
ar
e
t
w
o
co
m
m
o
n
w
a
y
s
to
m
ea
s
u
r
e
h
o
w
w
ell
d
en
o
is
i
n
g
w
o
r
k
s
.
T
h
is
is
w
h
at
P
SN
R
m
ea
n
s
:
P
S
N
R
=
20
10
(
255
√
M
SE
)
(
7
)
w
h
er
e
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
M
SE
)
is
th
e
av
er
a
g
e
o
f
t
h
e
s
q
u
a
r
ed
d
if
f
er
en
ce
s
b
et
w
ee
n
th
e
d
en
o
is
ed
an
d
g
r
o
u
n
d
tr
u
th
i
m
a
g
es.
T
h
e
f
o
llo
w
i
n
g
i
s
SS
I
M,
w
h
ic
h
m
o
d
els
h
o
w
s
i
m
ilar
t
h
in
g
s
lo
o
k
to
p
eo
p
le:
S
S
I
M
(
,
)
=
(
2
+
1
)
(
2
+
2
)
(
2
+
2
+
1
)
(
2
+
2
+
2
)
(
8
)
w
h
er
e
,
ar
e
lo
ca
l m
ea
n
s
,
2
,
2
ar
e
v
ar
ian
ce
s
,
is
co
v
ar
ian
ce
,
a
n
d
1
,
2
ar
e
s
tab
ilizin
g
co
n
s
ta
n
t
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
2
1
9
-
2
2
7
222
3.
O
P
T
I
M
I
Z
E
D
E
-
F
AG
F
A
L
G
O
RIT
H
M
T
h
e
E
-
F
A
GF
ap
p
r
o
ac
h
is
b
as
ed
o
n
th
e
b
ac
k
g
r
o
u
n
d
to
o
ls
i
n
tr
o
d
u
ce
d
ab
o
v
e
to
r
ea
lize
f
lex
ib
le
an
d
s
ca
lab
le
i
m
a
g
e
d
en
o
is
i
n
g
tech
n
iq
u
e
s
.
T
h
e
p
r
o
ce
d
u
r
e
is
s
tr
u
c
tu
r
ed
in
t
h
e
f
o
llo
w
i
n
g
w
a
y
(
Fi
g
u
r
e
1
)
:
−
Step
1
.
F
r
eq
u
en
c
y
d
ec
o
m
p
o
s
i
tio
n
:
t
h
e
alg
o
r
it
h
m
s
tar
t
s
w
it
h
th
e
d
ec
o
m
p
o
s
i
tio
n
o
f
th
e
n
o
is
y
i
m
a
g
e
in
to
lo
w
-
a
n
d
h
i
g
h
-
f
r
eq
u
e
n
c
y
co
m
p
o
n
en
ts
in
th
e
w
a
v
elet
d
o
m
ai
n
,
as
e
x
p
lai
n
ed
i
n
s
ec
t
io
n
2
.
T
h
is
s
er
v
e
s
t
o
d
ec
o
u
p
le
co
ar
s
e
s
tr
u
ctu
r
al
k
n
o
w
led
g
e
f
r
o
m
f
i
n
e
d
etails a
n
d
n
o
is
e.
−
Step
2
.
E
d
g
e
atten
tio
n
m
ap
co
n
s
tr
u
ct
io
n
:
t
h
e
n
,
an
ed
g
e
att
en
tio
n
m
ap
is
co
n
s
tr
u
cted
b
as
ed
o
n
g
r
ad
ien
t
-
b
ased
f
ilter
in
g
an
d
s
h
allo
w
NN
r
ef
in
e
m
en
t
(
s
ee
s
ec
tio
n
2
)
.
T
h
is
m
ap
h
i
g
h
li
g
h
ts
b
o
u
n
d
ar
i
es
in
th
e
i
m
a
g
e,
en
ab
lin
g
s
u
b
s
eq
u
e
n
t
f
ilter
i
n
g
t
o
m
ai
n
tai
n
s
h
ar
p
s
tr
u
ct
u
r
es i
n
th
e
i
m
ag
e.
−
Step
3
.
A
d
ap
tiv
e
p
ar
a
m
e
ter
p
r
ed
ictio
n
:
t
o
lear
n
s
p
atia
ll
y
v
ar
y
in
g
r
e
g
u
lar
izat
io
n
p
ar
a
m
eter
s
f
o
r
ea
ch
lo
ca
l
r
eg
io
n
,
w
e
tr
ain
t
h
e
p
r
o
p
o
s
ed
l
ig
h
t
w
ei
g
h
t
NN
in
s
ec
tio
n
2
.
S
u
ch
ad
ap
tiv
e
w
e
ig
h
t
s
en
ab
le
t
h
e
f
i
lter
to
ad
ap
t
th
e
r
esp
o
n
s
e
to
d
if
f
er
en
t
tex
t
u
r
es a
n
d
lo
ca
l sig
n
al
-
to
-
n
o
i
s
e
ch
an
g
e
s
.
−
Step
4
.
M
u
lti
-
s
ca
le
ed
g
e
-
a
w
ar
e
g
u
id
ed
f
ilter
i
n
g
:
n
o
w
,
w
e
p
r
o
p
o
s
e
m
u
lti
-
s
ca
le
ed
g
e
-
a
w
ar
e
g
u
id
ed
f
ilter
i
n
g
f
o
r
ed
g
e
-
a
w
ar
e
n
o
is
e
r
ed
u
ctio
n
w
i
th
lo
w
co
m
p
le
x
it
y
.
T
h
e
a
lg
o
r
ith
m
u
ti
lizes
ed
g
e
-
a
w
ar
e
g
u
id
ed
f
ilter
i
n
g
at
s
ev
er
al
s
p
atial
s
ca
les
u
s
in
g
lo
w
-
a
n
d
h
i
g
h
-
fr
eq
u
e
n
c
y
i
n
f
o
r
m
atio
n
to
g
et
h
er
w
it
h
t
h
e
ed
g
e
atten
tio
n
m
ap
.
T
h
is
m
u
lti
-
s
ca
le
m
et
h
o
d
o
lo
g
y
in
h
er
en
t
l
y
p
r
o
v
id
es
r
o
b
u
s
t
d
e
n
o
is
i
n
g
w
h
i
le
p
r
eser
v
in
g
b
o
u
n
d
ar
y
ch
ar
ac
ter
is
tic
s
f
o
r
s
tr
u
c
tu
r
es o
f
d
if
f
er
en
t
s
izes.
−
Step
5
.
A
da
p
tiv
e
f
u
s
io
n
o
f
f
ilter
ed
o
u
tp
u
t
s
:
a
t
las
t,
t
h
e
d
en
o
is
ed
r
esu
lt
s
w
it
h
in
d
i
f
f
e
r
en
t
s
ca
les
ar
e
ad
ap
tiv
el
y
f
u
s
ed
v
ia
lear
n
ed
w
eig
h
t
s
to
g
en
er
ate
th
e
f
i
n
al
r
esto
r
ed
im
a
g
e.
T
h
is
b
len
d
in
g
in
ter
p
o
lates
b
et
w
ee
n
lo
ca
l d
etail
p
r
eser
v
ati
o
n
an
d
g
lo
b
al
s
m
o
o
t
h
n
ess
,
d
e
p
en
d
in
g
o
n
th
e
co
n
ten
t o
f
th
e
i
m
ag
e.
Fig
u
r
e
1
.
B
lo
ck
d
iag
r
a
m
o
f
t
h
e
o
p
tim
ized
E
-
F
A
G
F
i
m
a
g
e
d
en
o
is
i
n
g
f
r
a
m
e
w
o
r
k
T
h
e
f
r
a
m
e
w
o
r
k
is
b
a
s
ed
o
n
P
y
T
o
r
ch
a
n
d
u
s
es
t
h
e
P
y
T
o
r
ch
-
W
av
elet
s
lib
r
ar
y
to
p
er
f
o
r
m
DW
T
an
d
I
DW
T
.
Fas
tP
ar
am
Net
co
n
s
is
ts
o
f
th
r
ee
3
×3
co
n
v
o
lu
tio
n
al
la
y
er
s
w
it
h
So
f
tp
l
u
s
ac
ti
v
atio
n
s
,
m
a
k
i
n
g
it
co
m
p
u
tatio
n
all
y
e
f
f
icie
n
t
w
it
h
lo
w
p
ar
a
m
eter
o
v
er
h
ea
d
.
B
io
r
th
o
g
o
n
al
w
a
v
elet
s
,
s
p
ec
if
i
ca
ll
y
“
b
io
r
6
.
8
,
”
ar
e
u
s
ed
w
ith
t
h
r
ee
f
u
s
io
n
w
i
n
d
o
w
s
ize
s
(
1
5
,
1
1
an
d
7
)
in
all
th
e
ex
p
er
i
m
e
n
t
s
.
T
h
e
co
d
eb
as
e
co
n
s
is
ts
o
f
m
o
d
e
l
d
ef
in
i
tio
n
s
,
i
n
f
er
e
n
ce
s
cr
ip
ts
,
an
d
p
lo
ttin
g
u
ti
liti
es
to
en
s
u
r
e
ex
ac
t
r
ep
licatio
n
.
T
h
ese
d
esig
n
ch
o
ices
m
a
k
e
f
o
r
an
o
p
ti
m
al
tr
ad
e
-
o
f
f
b
et
w
ee
n
co
m
p
u
tatio
n
al
e
f
f
o
r
t
an
d
in
t
er
p
r
etab
ilit
y
,
en
ab
li
n
g
ex
p
er
im
en
ts
w
it
h
s
ta
n
d
ar
d
GP
U
h
ar
d
w
ar
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Op
timiz
ed
ed
g
e
-
a
w
a
r
e
fr
eq
u
en
cy
-
g
u
id
e
d
filt
erin
g
fo
r
r
o
b
u
s
t ima
g
e
d
en
o
is
in
g
(
I
ma
n
E
la
w
a
d
y
)
223
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
ex
p
er
im
e
n
t
s
w
er
e
p
er
f
o
r
m
ed
o
n
th
e
B
SD5
0
0
d
ataset
w
it
h
a
n
NVI
DI
A
R
T
X
3
0
9
0
GP
U,
u
s
in
g
P
y
t
h
o
n
3
.
9
,
an
d
P
y
T
o
r
ch
w
it
h
P
y
T
o
r
ch
-
W
a
v
elets.
W
e
test
ed
p
er
f
o
r
m
an
ce
u
n
d
er
th
r
ee
ty
p
ica
l
n
o
is
e
m
o
d
els,
i.e
.
,
Gau
s
s
ian
n
o
is
e
(
=0
.
0
3
)
,
P
o
is
s
o
n
n
o
i
s
e
(
=1
)
,
an
d
s
p
ec
k
le
n
o
is
e
(
=0
.
0
3
)
.
Qu
an
titat
iv
e
r
es
u
lt
s
ar
e
r
ep
o
r
ted
u
s
in
g
P
SNR
,
S
SIM
,
an
d
av
er
ag
e
p
er
-
i
m
a
g
e
i
n
f
er
e
n
c
e
ti
m
e.
Fo
r
a
f
air
co
m
p
ar
is
o
n
,
all
m
et
h
o
d
s
p
r
o
ce
s
s
ed
th
e
s
a
m
e
n
o
is
y
in
p
u
ts
,
an
d
co
lo
r
i
m
ag
e
s
w
e
r
e
d
en
o
is
ed
ch
an
n
el
-
w
i
s
e.
Me
d
ian
(
ME
D
)
,
b
o
x
,
b
ilater
al,
an
d
N
L
M
w
er
e
t
h
e
cla
s
s
ica
l
b
aseli
n
es.
A
co
m
p
ar
is
o
n
o
f
class
ical
a
n
d
lear
n
i
n
g
-
b
ased
d
en
o
is
i
n
g
m
et
h
o
d
s
is
s
h
o
w
n
in
T
ab
le
1
.
T
ab
le
1
.
A
v
er
ag
e
d
en
o
i
s
in
g
p
e
r
f
o
r
m
an
ce
(
P
SNR
/SS
I
M)
an
d
r
u
n
ti
m
e
p
er
5
1
2
×5
1
2
im
a
g
e
o
n
B
SD5
0
0
M
e
t
h
o
d
P
S
N
R
(
G
a
u
s
si
a
n
)
S
S
I
M
(
G
a
u
ssi
a
n
)
P
S
N
R
(
P
o
i
sso
n
)
S
S
I
M
(
P
o
i
sso
n
)
P
S
N
R
(
S
p
e
c
k
l
e
)
S
S
I
M
(
S
p
e
c
k
l
e
)
T
i
me
(
s)
B
o
x
2
5
.
2
1
0
.
7
0
2
5
.
1
4
0
.
6
9
2
5
.
2
4
0
.
7
0
0
.
0
0
0
3
M
e
d
i
a
n
2
8
.
6
9
0
.
8
4
2
8
.
1
9
0
.
8
2
2
9
.
0
6
0
.
8
6
0
.
0
0
0
1
B
i
l
a
t
e
r
a
l
3
1
.
5
5
0
.
8
7
3
1
.
2
1
0
.
8
7
3
1
.
6
8
0
.
8
7
0
.
0
0
2
4
N
L
M
3
0
.
2
2
0
.
8
3
3
0
.
0
4
0
.
8
3
3
0
.
2
5
0
.
8
2
0
.
8
3
E
-
FA
GF (p
r
o
p
o
sed)
3
3
.
9
7
0
.
8
9
3
0
.
8
4
0
.
8
2
3
9
.
63
0
.
9
7
0
.
0
1
5
T
h
e
r
esu
lts
s
h
o
w
ed
t
h
at
t
h
e
b
ilater
al
f
il
ter
o
b
tain
ed
th
e
b
es
t
P
SNR
an
d
SS
I
M
u
n
d
er
P
o
is
s
o
n
n
o
is
e,
f
o
llo
w
ed
b
y
E
-
F
A
GF,
w
h
ic
h
w
a
s
also
q
u
ite
co
m
p
e
titi
v
e
w
it
h
o
u
t
p
ar
a
m
eter
tu
n
i
n
g
.
Mo
r
eo
v
er
,
E
-
F
A
GF
o
b
tain
ed
th
e
b
est
r
e
s
u
l
ts
u
n
d
er
s
p
ec
k
le
n
o
is
e
a
n
d
o
u
tp
er
f
o
r
m
ed
th
e
clas
s
ical
b
ilater
al
f
ilter
b
y
m
o
r
e
t
h
an
7
d
B
f
o
r
P
SNR
,
as s
h
o
w
n
i
n
Fi
g
u
r
e
s
2
(
a
)
an
d
(
b
)
w
h
i
le
p
r
eser
v
i
n
g
ed
g
es a
n
d
f
i
n
e
d
etails.
(
a)
(
b
)
Fig
u
r
e
2
.
P
er
f
o
r
m
a
n
ce
co
m
p
ar
is
o
n
u
n
d
er
d
if
f
er
en
t
n
o
is
e
t
y
p
e
s
: (
a)
SS
I
M
an
d
(
b
)
P
SNR
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
2
1
9
-
2
2
7
224
I
n
q
u
alitat
iv
e
co
m
p
ar
is
o
n
s
w
i
th
o
th
er
d
e
n
o
is
er
s
t
h
at
o
v
er
s
m
o
o
t
h
o
r
in
tr
o
d
u
ce
s
tair
ca
s
i
n
g
ar
ti
f
ac
ts
,
as illu
s
tr
ated
in
F
ig
u
r
e
3
,
E
-
F
A
G
F c
an
w
el
l p
r
eser
v
e
tex
tu
r
e
an
d
ed
g
e
in
f
o
r
m
atio
n
.
T
h
e
SS
I
M
h
is
to
g
r
a
m
s
al
s
o
d
em
o
n
s
tr
ate
t
h
at
E
-
F
AGF
in
cr
ea
s
es
th
e
av
er
ag
e
SS
I
M
a
n
d
d
ec
r
ea
s
es
q
u
alit
y
v
ar
iat
io
n
ac
r
o
s
s
th
e
d
ataset,
r
esu
lti
n
g
i
n
m
o
r
e
s
tab
le
r
es
to
r
atio
n
.
C
o
m
p
u
tatio
n
al
l
y
,
E
-
F
A
G
F
is
e
f
f
icie
n
t
s
in
ce
th
e
g
u
id
ed
f
ilter
i
n
g
co
m
p
o
n
e
n
t
s
an
d
lo
w
-
co
m
p
le
x
it
y
Fas
tP
ar
am
Net
ar
e
lo
ca
l
an
d
th
u
s
o
v
er
all
h
a
v
e
a
ti
m
e
co
m
p
lex
i
t
y
o
f
(
)
.
I
n
f
er
e
n
ce
o
n
an
R
T
X
3
0
9
0
ta
k
es
ab
o
u
t
0
.
0
1
5
s
f
o
r
a
5
1
2
×
5
1
2
im
a
g
e,
o
v
er
o
n
e
o
r
d
er
o
f
m
ag
n
it
u
d
e
f
a
s
ter
th
a
n
n
o
n
-
lo
ca
l
m
ea
n
s
an
d
w
it
h
lo
w
m
e
m
o
r
y
r
eq
u
ir
e
m
en
t
s
.
Mo
r
eo
v
er
,
n
o
h
y
p
er
p
a
r
a
m
eter
t
u
n
in
g
o
r
r
etr
ain
i
n
g
is
r
eq
u
ir
ed
w
h
e
n
s
w
itc
h
i
n
g
b
et
w
ee
n
Gau
s
s
ia
n
,
P
o
is
s
o
n
,
an
d
s
p
ec
k
le
n
o
is
e
E
-
F
AGF
r
elies
o
n
p
ix
el
-
w
is
e
r
eg
u
lar
izatio
n
o
n
t
h
e
lo
ca
l stat
is
tics
o
f
th
e
i
m
a
g
e.
Fig
u
r
e
3
.
Vis
u
al
co
m
p
ar
i
s
o
n
o
f
d
en
o
is
i
n
g
r
esu
lts
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Op
timiz
ed
ed
g
e
-
a
w
a
r
e
fr
eq
u
en
cy
-
g
u
id
e
d
filt
erin
g
fo
r
r
o
b
u
s
t ima
g
e
d
en
o
is
in
g
(
I
ma
n
E
la
w
a
d
y
)
225
5.
CO
NCLU
SI
O
N
W
e
p
r
o
p
o
s
ed
th
e
o
p
tim
iz
ed
(
E
-
FAGF
)
:
a
g
en
er
al
d
en
o
i
s
in
g
s
c
h
e
m
e,
i
n
teg
r
ati
n
g
w
a
v
elet
-
b
ased
f
r
eq
u
en
c
y
s
ep
ar
atio
n
,
ed
g
e
-
a
w
ar
e
g
u
id
an
ce
,
a
n
d
ad
ap
tiv
e
m
u
lti
-
s
ca
le
f
ilter
in
g
.
O
n
th
e
B
SD5
0
0
d
ataset,
E
-
F
A
GF
al
w
a
y
s
ac
h
ie
v
es
s
tate
-
of
-
t
h
e
-
ar
t
p
er
f
o
r
m
a
n
ce
f
o
r
Gau
s
s
ian
an
d
s
p
ec
k
le
n
o
is
e,
an
d
k
ee
p
s
co
m
p
etiti
v
e
f
o
r
P
o
is
s
o
n
n
o
is
e.
I
t
also
ac
h
iev
es
m
u
c
h
f
aster
r
u
n
ti
m
es
an
d
r
e
q
u
ir
es
n
o
r
etr
ain
in
g
f
o
r
d
i
f
f
er
en
t
n
o
is
e
t
y
p
e
s
.
W
ith
th
e
lo
ca
ll
y
ad
ap
tiv
e
E
-
F
A
GF
s
tr
u
ctu
r
e,
it
ca
n
ef
f
ic
ien
tl
y
d
ea
l
w
it
h
s
p
atiall
y
v
a
r
ian
t
n
o
is
e
an
d
i
s
f
av
o
r
ab
le
to
p
r
ac
tical
a
p
p
lica
tio
n
s
lik
e
m
ed
ical
i
m
a
g
i
n
g
,
s
cien
ti
f
ic
d
ata
ac
q
u
is
iti
o
n
,
an
d
em
b
ed
d
ed
v
is
io
n
s
y
s
te
m
s
.
A
lt
h
o
u
g
h
t
h
e
GP
U
v
er
s
io
n
i
s
in
r
ea
l
-
ti
m
e,
it
n
ee
d
s
m
o
r
e
o
p
tim
izatio
n
f
o
r
lo
w
-
p
o
w
er
o
r
m
o
b
ile
p
latf
o
r
m
s
,
s
u
ch
as
ch
a
n
n
e
l
p
r
u
n
i
n
g
,
w
ei
g
h
t
q
u
an
tizat
io
n
,
an
d
r
ep
lacin
g
f
lo
ati
n
g
-
p
o
in
t
k
er
n
el
s
w
i
th
f
i
x
ed
-
p
o
in
t.
Fo
r
f
u
tu
r
e
w
o
r
k
,
o
n
e
ca
n
ex
te
n
d
E
-
F
A
GF
f
o
r
v
id
eo
d
en
o
is
in
g
to
in
clu
d
e
s
tate
-
of
-
t
h
e
-
ar
t
atten
tio
n
m
ec
h
a
n
i
s
m
s
an
d
s
el
f
-
s
u
p
er
v
is
ed
lear
n
in
g
,
o
r
ev
en
i
n
h
y
p
er
s
p
ec
tr
al
im
a
g
es.
A
b
lati
n
g
o
n
win
d
o
w
s
ize,
w
a
v
ele
t
t
y
p
e,
an
d
co
m
p
o
n
en
t c
o
n
tr
ib
u
t
io
n
w
o
u
ld
also
h
elp
j
u
s
ti
f
y
t
h
e
d
esig
n
d
e
cis
io
n
s
at
h
a
n
d
.
F
UNDIN
G
I
NF
O
RM
AT
I
O
N
T
h
e
au
th
o
r
s
d
ec
lar
e
th
at
t
h
is
r
esear
ch
r
ec
eiv
ed
n
o
e
x
ter
n
al
f
u
n
d
i
n
g
.
AUTHO
R
CO
NT
RIB
UT
I
O
NS
ST
A
T
E
M
E
NT
T
h
is
j
o
u
r
n
al
u
s
e
s
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
i
v
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
t
h
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
lla
b
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
I
m
an
E
la
w
ad
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ah
m
et
R
e
ş
it Ka
v
s
ao
ğ
lu
✓
✓
✓
✓
✓
✓
İ
s
m
ail
R
ak
ıp
Kar
aş
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
si
s
I
:
I
n
v
e
st
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
si
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
W
e
co
n
f
ir
m
t
h
i
s
m
a
n
u
s
cr
ip
t
h
as
n
o
t
b
ee
n
p
u
b
lis
h
ed
else
wh
er
e
an
d
is
n
o
t
u
n
d
er
co
n
s
id
er
atio
n
b
y
an
o
th
er
j
o
u
r
n
al.
T
h
e
au
th
o
r
s
h
av
e
n
o
co
n
f
lict
s
o
f
i
n
ter
est to
d
ec
lar
e.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
t
u
d
y
ar
e
av
ailab
l
e
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
a
u
t
h
o
r
,
I
m
a
n
E
la
w
ad
y
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
S
.
M
u
k
s
i
m
o
v
a
,
S
.
U
m
i
r
z
a
k
o
v
a
,
S
.
M
a
r
d
i
e
v
a
,
a
n
d
Y
.
-
I
.
C
h
o
,
“
E
n
h
a
n
c
i
n
g
M
e
d
i
c
a
l
I
m
a
g
e
D
e
n
o
i
s
i
n
g
w
i
t
h
I
n
n
o
v
a
t
i
v
e
T
e
a
c
h
e
r
–
S
t
u
d
e
n
t
M
o
d
e
l
-
B
a
se
d
A
p
p
r
o
a
c
h
e
s
f
o
r
P
r
e
c
i
s
i
o
n
D
i
a
g
n
o
s
t
i
c
s
,
”
S
e
n
s
o
r
s
,
v
o
l
.
2
3
,
n
o
.
2
3
,
p
.
9
5
0
2
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
2
3
9
5
0
2
.
[
2
]
Y
.
C
u
i
,
W
.
R
e
n
,
X
.
C
a
o
,
a
n
d
A
.
K
n
o
l
l
,
“
R
e
v
i
t
a
l
i
z
i
n
g
C
o
n
v
o
l
u
t
i
o
n
a
l
N
e
t
w
o
r
k
f
o
r
I
m
a
g
e
R
e
s
t
o
r
a
t
i
o
n
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
P
a
t
t
e
r
n
A
n
a
l
y
s
i
s
a
n
d
M
a
c
h
i
n
e
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
4
6
,
n
o
.
1
2
,
p
p
.
9
4
2
3
–
9
4
3
8
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
T
P
A
M
I
.
2
0
2
4
.
3
4
1
9
0
07.
[
3
]
K
.
Z
h
a
n
g
,
W
.
Z
u
o
,
Y
.
C
h
e
n
,
D
.
M
e
n
g
,
a
n
d
L
.
Z
h
a
n
g
,
“
B
e
y
o
n
d
a
G
a
u
s
s
i
a
n
D
e
n
o
i
se
r
:
R
e
s
i
d
u
a
l
L
e
a
r
n
i
n
g
o
f
D
e
e
p
C
N
N
f
o
r
I
m
a
g
e
D
e
n
o
i
s
i
n
g
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
I
m
a
g
e
P
r
o
c
e
s
s
i
n
g
,
v
o
l
.
2
6
,
n
o
.
7
,
p
p
.
3
1
4
2
–
3
1
5
5
,
J
u
l
.
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
T
I
P
.
2
0
1
7
.
2
6
6
2
2
0
6
.
[
4
]
A
.
B
u
a
d
e
s
,
B
.
C
o
l
l
,
a
n
d
J
.
-
M
.
M
o
r
e
l
,
“
A
N
o
n
-
L
o
c
a
l
A
l
g
o
r
i
t
h
m
f
o
r
I
m
a
g
e
D
e
n
o
i
s
i
n
g
,
”
i
n
2
0
0
5
I
E
E
E
C
o
m
p
u
t
e
r
S
o
c
i
e
t
y
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
(
C
V
P
R
’
05)
,
I
E
E
E
,
2
0
0
5
,
p
p
.
6
0
–
6
5
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
0
5
.
3
8
.
[
5
]
C
.
T
o
m
a
s
i
a
n
d
R
.
M
a
n
d
u
c
h
i
,
“
B
i
l
a
t
e
r
a
l
f
i
l
t
e
r
i
n
g
f
o
r
g
r
a
y
a
n
d
c
o
l
o
r
i
m
a
g
e
s,
”
S
i
x
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
(
I
E
E
E
C
a
t
.
N
o
.
9
8
C
H
3
6
2
7
1
)
,
B
o
m
b
a
y
,
I
n
d
i
a
,
1
9
9
8
,
p
p
.
8
3
9
-
8
4
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
V
.
1
9
9
8
.
7
1
0
8
1
5
.
[
6
]
S
.
G
.
C
h
a
n
g
,
B
.
Y
u
,
a
n
d
M
.
V
e
t
t
e
r
l
i
,
“
A
d
a
p
t
i
v
e
w
a
v
e
l
e
t
t
h
r
e
s
h
o
l
d
i
n
g
f
o
r
i
m
a
g
e
d
e
n
o
i
s
i
n
g
a
n
d
c
o
m
p
r
e
s
s
i
o
n
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
I
m
a
g
e
P
r
o
c
e
s
s
i
n
g
,
v
o
l
.
9
,
n
o
.
9
,
p
p
.
1
5
3
2
–
1
5
4
6
,
2
0
0
0
,
d
o
i
:
1
0
.
1
1
0
9
/
8
3
.
8
6
2
6
3
3
.
[
7
]
K
.
H
e
,
J
.
S
u
n
,
a
n
d
X
.
T
a
n
g
,
“
G
u
i
d
e
d
I
m
a
g
e
F
i
l
t
e
r
i
n
g
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
P
a
t
t
e
r
n
A
n
a
l
y
s
i
s
a
n
d
M
a
c
h
i
n
e
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
3
5
,
n
o
.
6
,
p
p
.
1
3
9
7
–
1
4
0
9
,
J
u
n
.
2
0
1
3
,
d
o
i
:
1
0
.
1
1
0
9
/
T
P
A
M
I
.
2
0
1
2
.
2
1
3
.
[
8
]
R
.
A
.
K
a
r
i
m
,
N
.
W
.
A
r
s
h
a
d
,
a
n
d
Y
.
A
.
W
a
h
a
b
,
“
C
o
n
t
r
a
s
t
m
o
d
i
f
i
c
a
t
i
o
n
f
o
r
p
r
e
-
e
n
h
a
n
c
e
m
e
n
t
p
r
o
c
e
s
s
i
n
m
u
l
t
i
-
c
o
n
t
r
a
s
t
r
u
b
e
o
s
i
s
i
r
i
d
i
s
i
m
a
g
e
s
,
”
T
E
L
K
O
M
N
I
K
A
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
2
1
,
n
o
.
4
,
p
p
.
8
4
6
–
8
5
7
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
2
9
2
8
/
t
e
l
k
o
m
n
i
k
a
.
v
2
1
i
4
.
2
2
2
5
1
.
[
9
]
Z
.
W
a
n
g
,
A
.
C
.
B
o
v
i
k
,
H
.
R
.
S
h
e
i
k
h
,
a
n
d
E
.
P
.
S
i
m
o
n
c
e
l
l
i
,
“
I
m
a
g
e
q
u
a
l
i
t
y
a
s
s
e
s
s
me
n
t
:
f
r
o
m
e
r
r
o
r
v
i
s
i
b
i
l
i
t
y
t
o
s
t
r
u
c
t
u
r
a
l
s
i
m
i
l
a
r
i
t
y
,
”
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
2
1
9
-
2
2
7
226
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
I
m
a
g
e
P
r
o
c
e
s
s
i
n
g
,
v
o
l
.
1
3
,
n
o
.
4
,
p
p
.
6
0
0
–
6
1
2
,
A
p
r
.
2
0
0
4
,
d
o
i
:
1
0
.
1
1
0
9
/
T
I
P
.
2
0
0
3
.
8
1
9
8
6
1
.
[
1
0
]
S
.
G
u
o
,
Z
.
Y
a
n
,
K
.
Z
h
a
n
g
,
W
.
Z
u
o
,
a
n
d
L
.
Z
h
a
n
g
,
“
T
o
w
a
r
d
C
o
n
v
o
l
u
t
i
o
n
a
l
B
l
i
n
d
D
e
n
o
i
s
i
n
g
o
f
R
e
a
l
P
h
o
t
o
g
r
a
p
h
s
,
”
i
n
2
0
1
9
I
E
E
E
/
C
V
F
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
(
C
V
P
R
)
,
I
E
E
E
,
J
u
n
.
2
0
1
9
,
p
p
.
1
7
1
2
–
1
7
2
2
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
9
.
0
0
1
8
1
.
[
1
1
]
J.
-
M
.
G
u
o
,
H
.
P
r
a
s
e
t
y
o
,
a
n
d
K
.
W
o
n
g
,
“
H
a
l
f
t
o
n
i
n
g
-
b
a
s
e
d
B
l
o
c
k
T
r
u
n
c
a
t
i
o
n
C
o
d
i
n
g
i
m
a
g
e
r
e
s
t
o
r
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
V
i
s
u
a
l
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
I
m
a
g
e
R
e
p
r
e
s
e
n
t
a
t
i
o
n
,
v
o
l
.
3
5
,
p
p
.
1
9
3
–
1
9
7
,
F
e
b
.
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
v
c
i
r
.
2
0
1
5
.
1
2
.
0
1
6
.
[
1
2
]
Z
.
X
i
a
,
J
.
L
i
,
W
.
J
i
a
o
,
C
.
Z
h
a
n
g
,
J
.
Z
h
a
n
g
,
a
n
d
G
.
X
i
e
,
“
W
F
C
-
B
S
N
:
W
a
v
e
l
e
t
f
u
s
i
o
n
-
b
a
s
e
d
c
o
n
d
i
t
i
o
n
a
l
b
l
i
n
d
-
s
p
o
t
n
e
t
w
o
r
k
f
o
r
se
l
f
-
s
u
p
e
r
v
i
s
e
d
f
o
r
w
a
r
d
s
o
n
a
r
d
e
n
o
i
s
i
n
g
,
”
E
n
g
i
n
e
e
r
i
n
g
A
p
p
l
i
c
a
t
i
o
n
s
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
5
7
,
p
.
1
1
1
2
5
5
,
O
c
t
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
n
g
a
p
p
a
i
.
2
0
2
5
.
1
1
1
2
5
5
.
[
1
3
]
F.
B
.
C
o
t
t
e
r
,
U
s
e
s
o
f
c
o
m
p
l
e
x
w
a
v
e
l
e
t
s
i
n
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
,
P
h
.
D
.
d
i
s
s
e
r
t
a
t
i
o
n
,
U
n
i
v
.
o
f
C
a
m
b
r
i
d
g
e
,
C
a
m
b
r
i
d
g
e
,
U
.
K
.
,
A
u
g
.
2
0
1
9
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
r
e
p
o
s
i
t
o
r
y
.
c
a
m
.
a
c
.
u
k
/
h
a
n
d
l
e
/
1
8
1
0
/
3
0
6
6
6
1
.
[
1
4
]
C
.
T
i
a
n
,
M
.
Z
h
e
n
g
,
W
.
Z
u
o
,
B
.
Z
h
a
n
g
,
Y
.
Z
h
a
n
g
,
a
n
d
D
.
Z
h
a
n
g
,
“
M
u
l
t
i
-
s
t
a
g
e
i
m
a
g
e
d
e
n
o
i
s
i
n
g
w
i
t
h
t
h
e
w
a
v
e
l
e
t
t
r
a
n
s
f
o
r
m
,
”
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
,
v
o
l
.
1
3
4
,
p
.
1
0
9
0
5
0
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
c
o
g
.
2
0
2
2
.
1
0
9
0
5
0
.
[
1
5
]
S
.
A
n
w
a
r
a
n
d
N
.
B
a
r
n
e
s
,
“
R
e
a
l
I
m
a
g
e
D
e
n
o
i
s
i
n
g
W
i
t
h
F
e
a
t
u
r
e
A
t
t
e
n
t
i
o
n
,
”
i
n
2
0
1
9
I
E
E
E
/
C
V
F
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
(
I
C
C
V
)
,
I
E
E
E
,
O
c
t
.
2
0
1
9
,
p
p
.
3
1
5
5
–
3
1
6
4
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
V
.
2
0
1
9
.
0
0
3
2
5
.
[
1
6
]
J
.
S
u
n
a
n
d
W
.
W
a
n
g
,
“
S
t
u
d
y
o
n
U
n
d
e
r
w
a
t
e
r
I
m
a
g
e
D
e
n
o
i
s
i
n
g
A
l
g
o
r
i
t
h
m
B
a
s
e
d
o
n
W
a
v
e
l
e
t
T
r
a
n
s
f
o
r
m
,
”
J
o
u
r
n
a
l
o
f
P
h
y
s
i
c
s
:
C
o
n
f
e
r
e
n
c
e
S
e
r
i
e
s
,
v
o
l
.
8
0
6
,
n
o
.
1
,
p
.
0
1
2
0
0
6
,
F
e
b
.
2
0
1
7
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/
8
0
6
/
1
/
0
1
2
0
0
6
.
[
1
7
]
A
.
A
k
g
u
n
d
o
g
d
u
,
S
.
K
u
r
t
,
N
.
K
i
l
i
c
,
O
.
N
.
U
c
a
n
,
a
n
d
N
.
A
k
a
l
i
n
,
“
D
i
a
g
n
o
s
i
s
o
f
R
e
n
a
l
F
a
i
l
u
r
e
D
i
s
e
a
se
U
s
i
n
g
A
d
a
p
t
i
v
e
N
e
u
r
o
-
F
u
z
z
y
I
n
f
e
r
e
n
c
e
S
y
s
t
e
m
,
”
J
o
u
r
n
a
l
o
f
M
e
d
i
c
a
l
S
y
s
t
e
m
s
,
v
o
l
.
3
4
,
p
p
.
1
0
0
3
–
1
0
0
9
,
D
e
c
.
2
0
1
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
9
1
6
-
0
0
9
-
9
3
1
7
-
2.
[
1
8
]
M
.
W
o
l
t
e
r
,
F
.
B
l
a
n
k
e
,
J
.
G
a
r
c
k
e
,
a
n
d
C
.
T
.
H
o
y
t
,
“
p
t
w
t
-
T
h
e
P
y
T
o
r
c
h
W
a
v
e
l
e
t
T
o
o
l
b
o
x
,
”
J
o
u
r
n
a
l
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
R
e
s
e
a
r
c
h
,
v
o
l
.
2
5
,
n
o
.
8
0
,
p
p
.
1
–
7
,
2
0
2
4
.
[
1
9
]
K
.
D
a
b
o
v
,
A
.
F
o
i
,
V
.
K
a
t
k
o
v
n
i
k
,
a
n
d
K
.
E
g
i
a
z
a
r
i
a
n
,
“
I
m
a
g
e
D
e
n
o
i
s
i
n
g
b
y
S
p
a
r
s
e
3
-
D
T
r
a
n
s
f
o
r
m
-
D
o
m
a
i
n
C
o
l
l
a
b
o
r
a
t
i
v
e
F
i
l
t
e
r
i
n
g
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
I
m
a
g
e
P
r
o
c
e
s
s
i
n
g
,
v
o
l
.
1
6
,
n
o
.
8
,
p
p
.
2
0
8
0
–
2
0
9
5
,
A
u
g
.
2
0
0
7
,
d
o
i
:
1
0
.
1
1
0
9
/
T
I
P
.
2
0
0
7
.
9
0
1
2
3
8
.
[
2
0
]
J
.
L
e
h
t
i
n
e
n
e
t
a
l
.
,
“
N
o
i
s
e
2
N
o
i
se
:
L
e
a
r
n
i
n
g
I
m
a
g
e
R
e
s
t
o
r
a
t
i
o
n
w
i
t
h
o
u
t
C
l
e
a
n
D
a
t
a
,
”
P
r
o
c
e
e
d
i
n
g
s
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
R
e
s
e
a
r
c
h
,
v
o
l
.
8
0
,
p
p
.
2
9
6
5
–
2
9
7
4
,
2
0
1
8
.
[
2
1
]
Z
.
Z
h
a
n
g
e
t
a
l
.
,
“
R
e
a
l
-
T
i
m
e
C
o
n
t
r
o
l
l
a
b
l
e
D
e
n
o
i
s
i
n
g
f
o
r
I
m
a
g
e
a
n
d
V
i
d
e
o
,
”
i
n
2
0
2
3
I
E
E
E
/
C
V
F
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
(
C
V
P
R
)
,
I
E
E
E
,
J
u
n
.
2
0
2
3
,
p
p
.
1
4
0
2
8
–
1
4
0
3
8
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
5
2
7
2
9
.
2
0
2
3
.
0
1
3
4
8
.
[
2
2
]
K
.
Z
h
a
n
g
,
W
.
Z
u
o
,
a
n
d
L
.
Z
h
a
n
g
,
“
F
F
D
N
e
t
:
T
o
w
a
r
d
a
F
a
s
t
a
n
d
F
l
e
x
i
b
l
e
S
o
l
u
t
i
o
n
f
o
r
C
N
N
-
B
a
s
e
d
I
m
a
g
e
D
e
n
o
i
s
i
n
g
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
I
m
a
g
e
P
r
o
c
e
s
s
i
n
g
,
v
o
l
.
2
7
,
n
o
.
9
,
p
p
.
4
6
0
8
–
4
6
2
2
,
S
e
p
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
T
I
P
.
2
0
1
8
.
2
8
3
9
8
9
1
.
[
2
3
]
Z
.
H
.
S
h
a
h
e
t
a
l
.
,
“
I
m
a
g
e
r
e
s
t
o
r
a
t
i
o
n
i
n
f
r
e
q
u
e
n
c
y
s
p
a
c
e
u
s
i
n
g
c
o
m
p
l
e
x
-
v
a
l
u
e
d
C
N
N
s
,
”
F
r
o
n
t
i
e
r
s
i
n
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
7
,
p
.
1
3
5
3
8
7
3
,
S
e
p
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
8
9
/
f
r
a
i
.
2
0
2
4
.
1
3
5
3
8
7
3
.
[
2
4
]
W
.
B
o
o
n
p
o
o
k
e
t
a
l
.,
“
D
e
e
p
L
e
a
r
n
i
n
g
S
e
m
a
n
t
i
c
S
e
g
m
e
n
t
a
t
i
o
n
f
o
r
L
a
n
d
U
s
e
a
n
d
L
a
n
d
C
o
v
e
r
T
y
p
e
s
U
s
i
n
g
L
a
n
d
s
a
t
8
I
m
a
g
e
r
y
,
”
I
S
P
R
S
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
G
e
o
-
I
n
f
o
r
m
a
t
i
o
n
,
v
o
l
.
1
2
,
n
o
.
1
,
p
.
1
4
,
J
a
n
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
i
j
g
i
1
2
0
1
0
0
1
4
.
[
2
5
]
P
.
A
r
b
e
l
á
e
z
,
M
.
M
a
i
r
e
,
C
.
F
o
w
l
k
e
s
,
a
n
d
J
.
M
a
l
i
k
,
“
C
o
n
t
o
u
r
D
e
t
e
c
t
i
o
n
a
n
d
H
i
e
r
a
r
c
h
i
c
a
l
I
m
a
g
e
S
e
g
me
n
t
a
t
i
o
n
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
P
a
t
t
e
r
n
A
n
a
l
y
s
i
s
a
n
d
M
a
c
h
i
n
e
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
3
3
,
n
o
.
5
,
p
p
.
8
9
8
–
9
1
6
,
M
a
y
2
0
1
1
,
d
o
i
:
1
0
.
1
1
0
9
/
T
P
A
M
I
.
2
0
1
0
.
1
6
1
.
[
2
6
]
H
.
C
h
i
h
a
o
u
i
a
n
d
P
.
F
a
v
a
r
o
,
“
M
a
s
k
e
d
a
n
d
S
h
u
f
f
l
e
d
B
l
i
n
d
S
p
o
t
D
e
n
o
i
s
i
n
g
f
o
r
R
e
a
l
-
W
o
r
l
d
I
m
a
g
e
s
,
”
i
n
2
0
2
4
I
E
E
E
/
C
V
F
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
(
C
V
P
R
)
,
I
E
E
E
,
J
u
n
.
2
0
2
4
,
p
p
.
3
0
2
5
–
3
0
3
4
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
5
2
7
3
3
.
2
0
2
4
.
0
0
2
9
2
.
[
2
7
]
F
.
W
a
n
g
,
M
.
Z
h
e
n
g
,
K
.
M
a
,
a
n
d
X
.
H
u
,
“
R
e
s
a
m
p
l
i
n
g
a
p
p
r
o
a
c
h
f
o
r
i
m
b
a
l
a
n
c
e
d
d
a
t
a
c
l
a
s
s
i
f
i
c
a
t
i
o
n
b
a
s
e
d
o
n
c
l
a
s
s
i
n
s
t
a
n
c
e
d
e
n
s
i
t
y
p
e
r
f
e
a
t
u
r
e
v
a
l
u
e
i
n
t
e
r
v
a
l
s
,
”
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
6
9
2
,
p
.
1
2
1
5
7
0
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
s
.
2
0
2
4
.
1
2
1
5
7
0
.
[
2
8
]
M
.
D
a
v
i
e
s
e
t
a
l
.
,
“
L
o
i
h
i
:
A
N
e
u
r
o
mo
r
p
h
i
c
M
a
n
y
c
o
r
e
P
r
o
c
e
s
s
o
r
w
i
t
h
O
n
-
C
h
i
p
L
e
a
r
n
i
n
g
,
”
I
E
E
E
M
i
c
r
o
,
v
o
l
.
3
8
,
n
o
.
1
,
p
p
.
8
2
–
9
9
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
M
M
.
2
0
1
8
.
1
1
2
1
3
0
3
5
9
.
[
2
9
]
X
.
C
a
o
,
Z
.
L
i
u
,
H
.
P
a
r
k
,
W
.
R
e
n
,
S
.
S
u
n
,
a
n
d
R
.
W
a
n
g
,
“
E
n
sI
R
:
A
n
E
n
s
e
m
b
l
e
A
l
g
o
r
i
t
h
m
f
o
r
I
m
a
g
e
R
e
s
t
o
r
a
t
i
o
n
v
i
a
G
a
u
s
s
i
a
n
M
i
x
t
u
r
e
M
o
d
e
l
s
,
”
i
n
A
d
v
a
n
c
e
s
i
n
N
e
u
r
a
l
I
n
f
o
r
m
a
t
i
o
n
P
r
o
c
e
s
s
i
n
g
S
y
s
t
e
m
s
3
7
,
S
a
n
D
i
e
g
o
,
C
a
l
i
f
o
r
n
i
a
,
U
S
A
:
N
e
u
r
a
l
I
n
f
o
r
ma
t
i
o
n
P
r
o
c
e
s
s
i
n
g
S
y
s
t
e
m
s
F
o
u
n
d
a
t
i
o
n
,
I
n
c
.
(
N
e
u
r
I
P
S
)
,
2
0
2
4
,
p
p
.
1
3
3
4
8
7
–
1
3
3
5
1
7
.
d
o
i
:
1
0
.
5
2
2
0
2
/
0
7
9
0
1
7
-
4
2
4
3
.
[
3
0
]
H
.
L
e
e
,
H
.
C
h
o
i
,
K
.
S
o
h
n
,
a
n
d
D
.
M
i
n
,
“
K
N
N
L
o
c
a
l
A
t
t
e
n
t
i
o
n
f
o
r
I
m
a
g
e
R
e
s
t
o
r
a
t
i
o
n
,
”
i
n
2
0
2
2
I
E
E
E
/
C
V
F
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
(
C
V
P
R
)
,
I
E
E
E
,
J
u
n
.
2
0
2
2
,
p
p
.
2
1
2
9
–
2
1
3
9
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
5
2
6
8
8
.
2
0
2
2
.
0
0
2
1
8
.
[
3
1
]
O
.
R
o
n
n
e
b
e
r
g
e
r
,
P
.
F
i
s
c
h
e
r
,
a
n
d
T
.
B
r
o
x
,
“
U
-
n
e
t
:
C
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
s
f
o
r
b
i
o
m
e
d
i
c
a
l
i
m
a
g
e
s
e
g
m
e
n
t
a
t
i
o
n
,
”
i
n
M
e
d
i
c
a
l
I
m
a
g
e
C
o
m
p
u
t
i
n
g
a
n
d
C
o
m
p
u
t
e
r
-
A
s
s
i
s
t
e
d
I
n
t
e
r
v
e
n
t
i
o
n
–
M
I
C
C
A
I
2
0
1
5
,
N
.
N
a
v
a
b
,
J
.
H
o
r
n
e
g
g
e
r
,
W
.
W
e
l
l
s
,
a
n
d
A
.
F
r
a
n
g
i
,
E
d
s
.
,
C
h
a
m
:
S
p
r
i
n
g
e
r
,
2
0
1
5
,
p
p
.
2
3
4
–
2
4
1
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
2
4
5
7
4
-
4
_
2
8
.
[
3
2
]
Y
.
Z
.
B
a
l
a
,
P
.
A
b
d
u
l
S
a
m
a
t
,
K
.
Y
.
S
h
a
r
i
f
,
a
n
d
N
.
M
a
n
s
h
o
r
,
“
T
h
e
i
n
f
l
u
e
n
c
e
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
o
n
t
h
e
p
r
e
d
i
c
t
i
v
e
p
e
r
f
o
r
m
a
n
c
e
o
f
c
r
o
s
s
-
p
r
o
j
e
c
t
d
e
f
e
c
t
p
r
e
d
i
c
t
i
o
n
:
e
m
p
i
r
i
c
a
l
a
n
a
l
y
s
i
s
,
”
T
E
L
K
O
M
N
I
K
A
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
2
2
,
n
o
.
4
,
p
p
.
8
3
0
–
8
3
7
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
1
2
9
2
8
/
t
e
l
k
o
m
n
i
k
a
.
v
2
2
i
4
.
2
5
9
1
6
.
[
3
3
]
Y
.
R
a
o
,
W
.
Z
h
a
o
,
Z
.
Z
h
u
,
J
.
L
u
,
a
n
d
J
.
Z
h
o
u
,
“
G
l
o
b
a
l
F
i
l
t
e
r
N
e
t
w
o
r
k
s
f
o
r
I
m
a
g
e
C
l
a
s
s
i
f
i
c
a
t
i
o
n
,
”
A
d
v
a
n
c
e
s
i
n
N
e
u
r
a
l
I
n
f
o
r
m
a
t
i
o
n
P
r
o
c
e
s
s
i
n
g
S
y
s
t
e
m
s
,
v
o
l
.
2
,
p
p
.
9
8
0
–
9
9
3
,
2
0
2
1
.
[
3
4
]
Z
.
L
i
u
e
t
a
l
.
,
“
S
w
i
n
T
r
a
n
s
f
o
r
m
e
r
:
H
i
e
r
a
r
c
h
i
c
a
l
V
i
s
i
o
n
T
r
a
n
s
f
o
r
me
r
u
s
i
n
g
S
h
i
f
t
e
d
W
i
n
d
o
w
s
,
”
i
n
2
0
2
1
I
E
E
E
/
C
V
F
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
(
I
C
C
V
)
,
I
E
E
E
,
O
c
t
.
2
0
2
1
,
p
p
.
9
9
9
2
–
1
0
0
0
2
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
V
4
8
9
2
2
.
2
0
2
1
.
0
0
9
8
6
.
[
3
5
]
S
.
T
u
a
n
d
M
.
Z
h
a
n
g
,
“
Th
e
B
i
g
D
a
t
a
M
o
d
e
l
f
o
r
U
r
b
a
n
R
o
a
d
L
a
n
d
U
s
e
P
l
a
n
n
i
n
g
I
s
B
a
se
d
o
n
a
N
e
u
r
a
l
N
e
t
w
o
r
k
A
l
g
o
r
i
t
h
m
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
N
e
u
r
o
s
c
i
e
n
c
e
,
v
o
l
.
2
0
2
2
,
p
p
.
1
–
1
2
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
2
7
2
7
5
1
2
.
[
3
6
]
J
.
X
u
,
H
.
L
i
,
Z
.
L
i
a
n
g
,
D
.
Z
h
a
n
g
,
a
n
d
L
.
Z
h
a
n
g
,
“
R
e
a
l
-
wo
r
l
d
N
o
i
sy
I
m
a
g
e
D
e
n
o
i
s
i
n
g
:
A
N
e
w
B
e
n
c
h
m
a
r
k
,
”
a
r
X
i
v
p
r
e
p
r
i
n
t
,
2
0
1
8
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
8
0
4
.
0
2
6
0
3
.
[
3
7
]
H
.
J
i
a
n
g
,
A
.
L
u
o
,
H
.
F
a
n
,
S
.
H
a
n
,
a
n
d
S
.
L
i
u
,
“
L
o
w
-
L
i
g
h
t
I
m
a
g
e
E
n
h
a
n
c
e
m
e
n
t
w
i
t
h
W
a
v
e
l
e
t
-
B
a
s
e
d
D
i
f
f
u
s
i
o
n
M
o
d
e
l
s
,
”
A
C
M
T
r
a
n
s
a
c
t
i
o
n
s
o
n
G
r
a
p
h
i
c
s
,
v
o
l
.
4
2
,
n
o
.
6
,
p
p
.
1
–
1
4
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
4
5
/
3
6
1
8
3
7
3
.
[
3
8
]
C
.
W
a
n
g
e
t
a
l
.
,
“
H
o
w
P
o
w
e
r
f
u
l
P
o
t
e
n
t
i
a
l
o
f
A
t
t
e
n
t
i
o
n
o
n
I
m
a
g
e
R
e
s
t
o
r
a
t
i
o
n
?
,
”
a
r
X
i
v
p
r
e
p
r
i
n
t
,
2
0
2
4
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
2
4
0
3
.
1
0
3
3
6
.
[
3
9
]
M
.
S
.
S
t
u
a
r
t
,
“
A
d
a
p
t
a
b
l
e
a
n
d
t
r
u
s
t
w
o
r
t
h
y
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
h
u
m
a
n
a
c
t
i
v
i
t
y
r
e
c
o
g
n
i
t
i
o
n
f
r
o
m
b
i
o
e
l
e
c
t
r
i
c
s
i
g
n
a
l
s
,
”
V
i
r
g
i
n
i
a
C
o
m
m
o
n
w
e
a
l
t
h
U
n
i
v
e
r
s
i
t
y
,
2
0
2
4
.
d
o
i
:
1
0
.
2
5
7
7
2
/
B
G
1
A
-
K
4
6
3
.
[
4
0
]
Y
.
P
o
i
r
i
e
r
-
G
i
n
t
e
r
a
n
d
J
.
-
F
.
L
a
l
o
n
d
e
,
“
R
o
b
u
s
t
U
n
s
u
p
e
r
v
i
s
e
d
S
t
y
l
e
G
A
N
I
ma
g
e
R
e
s
t
o
r
a
t
i
o
n
,
”
i
n
2
0
2
3
I
E
E
E
/
C
V
F
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
(
C
V
P
R
)
,
I
E
E
E
,
J
u
n
.
2
0
2
3
,
p
p
.
2
2
2
9
2
–
2
2
3
0
1
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
5
2
7
2
9
.
2
0
2
3
.
0
2
1
3
5
.
[
4
1
]
J
.
X
i
a
o
e
t
a
l
.
,
“
F
r
e
q
u
e
n
c
y
-
A
w
a
r
e
G
u
i
d
a
n
c
e
f
o
r
B
l
i
n
d
I
m
a
g
e
R
e
s
t
o
r
a
t
i
o
n
v
i
a
D
i
f
f
u
s
i
o
n
M
o
d
e
l
s
,
”
i
n
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
A
.
D
e
l
B
u
e
,
C
.
C
a
n
t
o
n
,
J
.
P
o
n
t
-
T
u
s
e
t
,
a
n
d
T
.
T
o
m
m
a
s
i
,
E
d
s
.
,
C
h
a
m
:
S
p
r
i
n
g
e
r
,
2
0
2
5
,
p
p
.
7
1
–
8
7
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
9
1
8
3
8
-
4
_
5
.
[
4
2
]
J
.
H
o
u
,
Z
.
Z
h
u
,
J
.
H
o
u
,
H
.
L
i
u
,
H
.
Z
e
n
g
,
a
n
d
H
.
Y
u
a
n
,
“
G
l
o
b
a
l
s
t
r
u
c
t
u
r
e
‑
a
w
a
r
e
d
i
f
f
u
s
i
o
n
p
r
o
c
e
s
s
f
o
r
l
o
w
‑
l
i
g
h
t
i
m
a
g
e
e
n
h
a
n
c
e
m
e
n
t
,
”
i
n
A
d
v
a
n
c
e
s
i
n
N
e
u
r
a
l
I
n
f
o
r
m
a
t
i
o
n
P
r
o
c
e
s
s
i
n
g
S
y
s
t
e
m
s
,
”
i
n
A
d
v
a
n
c
e
s
i
n
N
e
u
r
a
l
I
n
f
o
r
m
a
t
i
o
n
P
r
o
c
e
s
s
i
n
g
S
y
s
t
e
m
s
,
2
0
2
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
Op
timiz
ed
ed
g
e
-
a
w
a
r
e
fr
eq
u
en
cy
-
g
u
id
e
d
filt
erin
g
fo
r
r
o
b
u
s
t ima
g
e
d
en
o
is
in
g
(
I
ma
n
E
la
w
a
d
y
)
227
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
I
m
a
n
Ela
w
a
d
y
is
a
re
se
a
rc
h
e
r
a
n
d
a
c
a
d
e
m
ic
in
th
e
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
E
n
g
in
e
e
rin
g
a
t
Ka
ra
b
ü
k
Un
iv
e
rsity
.
He
h
o
l
d
s
a
D
o
c
to
ra
te
a
n
d
h
a
s
a
n
a
c
a
d
e
m
ic
b
a
c
k
g
ro
u
n
d
i
n
t
h
e
a
re
a
o
f
E
lec
tri
c
a
l
E
n
g
in
e
e
rin
g
a
n
d
C
o
m
p
u
ter
S
c
ien
c
e
.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
sig
n
a
l
a
n
d
im
a
g
e
p
ro
c
e
ss
in
g
,
d
ig
it
a
l
c
o
m
m
u
n
ica
ti
o
n
s,
a
n
d
t
h
e
o
p
ti
m
iza
ti
o
n
a
lg
o
ri
th
m
s.
He
h
a
s
p
u
b
li
sh
e
d
a
v
a
riet
y
o
f
p
a
p
e
rs
c
o
v
e
rin
g
a
re
a
s
ra
n
g
in
g
f
ro
m
n
o
ise
-
re
d
u
c
ti
o
n
o
v
e
r
w
irele
ss
c
h
a
n
n
e
ls
b
y
m
e
a
n
s
o
f
v
e
c
to
r
q
u
a
n
ti
z
a
ti
o
n
c
o
m
p
re
ss
io
n
a
n
d
f
il
terin
g
to
b
re
a
st
c
a
n
c
e
r
d
e
tec
ti
o
n
w
it
h
d
e
e
p
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s
a
n
d
Ra
sp
b
e
r
ry
P
i
3
b
o
a
rd
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ima
n
e
la
w
d
y
@
k
a
ra
b
u
k
.
e
d
u
.
tr.
Ahm
e
t
Re
şit
K
a
v
s
a
o
ğ
lu
re
c
e
iv
e
d
Ph
.
D
.
i
n
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
a
t
S
a
k
a
r
y
a
Un
iv
e
rsit
y
,
T
u
rk
e
y
.
H
e
is
n
o
w
A
ss
o
c
iate
P
ro
f
e
ss
o
r
in
th
e
Bio
m
e
d
ica
l
En
g
in
e
e
rin
g
De
p
a
rtme
n
t
o
f
Ka
ra
b
ü
k
Un
iv
e
rsity
/
T
u
rk
e
y
.
His
re
se
a
rc
h
in
tere
sts
a
re
in
b
io
m
e
d
ica
l
sig
n
a
l
p
ro
c
e
ss
in
g
,
m
a
c
h
in
e
lea
rn
i
n
g
a
n
d
h
u
m
a
n
-
c
o
m
p
u
ter
in
tera
c
ti
o
n
.
He
is
t
h
e
a
u
t
h
o
r
o
f
m
a
n
y
jo
u
r
n
a
l
p
a
p
e
rs
a
n
d
c
o
n
f
e
re
n
c
e
p
a
p
e
rs
in
th
o
se
f
ield
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
a
v
s
a
o
g
lu
@k
a
ra
b
u
k
.
e
d
u
.
tr.
İs
m
a
il
R
a
k
ıp
K
a
r
a
ş
re
c
e
i
v
e
d
h
is
B
.
Sc
.
f
ro
m
S
e
lçu
k
Un
iv
e
rsity
,
M
.
Sc
.
f
ro
m
G
e
b
z
e
In
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
,
a
n
d
P
h
.
D
.
in
G
IS
a
n
d
Re
m
o
te
S
e
n
sin
g
f
ro
m
Yıld
ız
T
e
c
h
n
ica
l
Un
iv
e
rsit
y
.
He
tea
c
h
e
s
in
th
e
C
o
m
p
u
ter
En
g
in
e
e
rin
g
De
p
a
rtm
e
n
t
o
f
th
e
Ka
ra
b
ü
k
Un
iv
e
rsit
y
a
n
d
h
e
is
t
h
e
h
e
a
d
o
f
3
D
-
G
e
o
In
f
o
rm
a
ti
c
s
Re
s
e
a
rc
h
G
ro
u
p
.
His
re
se
a
rc
h
to
p
ic
is
g
e
o
i
n
f
o
rm
a
ti
c
s
a
n
d
g
e
o
g
ra
p
h
ic i
n
f
o
rm
a
ti
o
n
sy
ste
m
s
(G
IS
),
3
D
G
IS
,
sp
a
ti
a
l
d
a
ta
a
n
a
l
y
sis
a
n
d
m
a
c
h
in
e
lea
rn
in
g
a
p
p
li
c
a
ti
o
n
in
g
e
o
sp
a
ti
a
l
p
e
rsp
e
c
ti
v
e
.
He
h
a
s
p
u
b
li
sh
e
d
m
o
re
th
a
n
1
0
0
p
u
b
li
c
a
ti
o
n
s
a
n
d
c
u
rre
n
tl
y
a
c
ti
v
e
l
y
e
n
g
a
g
e
s
in
d
iff
e
re
n
t
n
a
ti
o
n
a
l
a
n
d
i
n
tern
a
ti
o
n
a
l
p
r
o
jec
ts,
su
p
p
o
rted
b
y
T
ÜBİ
TA
K an
d
EU.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
is
m
a
il
.
k
a
r
a
s@
k
a
ra
b
u
k
.
e
d
u
.
tr
.
Evaluation Warning : The document was created with Spire.PDF for Python.