Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
10
,
No.
4
,
A
ugus
t
2020
,
pp. 410
9
~
41
17
IS
S
N: 20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v10
i
4
.
pp
4109
-
41
17
4109
Journ
al h
om
e
page
:
http:
//
ij
ece.i
aesc
or
e.c
om/i
nd
ex
.ph
p/IJ
ECE
Comput
atio
n
al scrutiny
of im
age denoisin
g metho
d found
on
DBAMF
und
er
S
PN surr
ound
ing
Vo
r
apoj P
ata
navijit
Depa
rtment
o
f
E
le
c
tri
c
al a
nd
Ele
ct
roni
c
Eng
ineer
ing,
Fa
cul
t
y
of E
ngine
er
ing,
As
su
m
pti
on
Univer
sit
y
of
Th
ai
l
a
nd,
Th
ai
l
and
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Sep
3
, 2
019
Re
vised
Feb
2
7
,
2020
Accepte
d
Ma
r
8
, 2
020
Tra
di
ti
ona
lly
,
r
a
nk
orde
r
absolute
diffe
r
ence
(R
OA
D)
has
a
gre
at
sim
il
ar
i
t
y
ca
pa
ci
t
y
for
id
ent
if
y
ing
wheth
er
the
pixel
is
SP
N
or
noisel
ess
bec
aus
e
stat
isti
ca
l
cha
r
acte
ristic
of
ROA
D
is
desire
d
for
a
noise
ide
n
ti
f
y
i
ng
obje
ctiv
e
.
As
a
result
,
the
dec
ision
base
d
a
dapt
iv
e
m
edi
an
fil
te
r
(DBA
MF
)
tha
t
is
fou
nd
on
ROA
D
te
chni
que
has
bee
n
in
it
iall
y
proposed
for
el
iminat
i
ng
a
n
impuls
ive
noise
since
201
0.
Consequentl
y,
thi
s
an
aly
z
ed
rep
ort
foc
uses
t
o
exa
m
ine
the
sim
ilarity
c
a
pac
i
t
y
of
d
enoi
s
ing
m
et
hod
fou
nd
on
DBA
MF
for
dive
rse
SP
N
Surrounding.
In
orde
r
to
exa
m
i
ne
the
denoi
sing
ca
p
ac
i
t
y
a
n
d
i
t
s
o
b
s
t
r
u
c
t
i
o
n
o
f
t
h
e
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
,
t
h
e
f
o
u
r
o
r
i
g
i
n
a
l
d
i
g
i
t
a
l
i
m
a
g
e
s
,
c
o
m
p
r
i
s
e
d
o
f
A
i
r
p
l
a
n
e
,
P
e
p
p
e
r
,
G
i
r
l
a
n
d
L
e
n
a
,
a
r
e
e
x
a
m
i
n
e
d
i
n
t
h
e
s
e
c
o
m
p
u
t
a
t
i
o
n
a
l
s
i
m
u
l
a
t
i
o
n
s
f
o
r
S
P
N
s
u
r
r
o
u
n
d
i
n
g
b
y
i
n
i
t
i
a
l
l
y
c
o
n
t
a
m
i
n
a
t
i
n
g
t
h
e
S
P
N
w
i
t
h
d
i
v
e
r
s
e
i
n
t
e
n
s
i
t
y
.
L
a
t
e
r
,
a
l
l
c
o
n
t
a
m
i
n
a
t
e
d
d
i
g
i
t
a
l
i
m
a
g
e
s
a
r
e
d
e
n
o
i
s
e
d
b
y
t
h
e
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
.
I
n
a
d
d
i
t
i
o
n
,
t
h
e
p
r
o
p
o
s
e
d
d
e
n
o
i
s
e
d
i
m
a
g
e
,
w
h
i
c
h
i
s
c
o
m
p
u
t
e
d
b
y
t
h
i
s
D
B
A
M
F
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
,
i
s
c
o
n
f
r
o
n
t
e
d
w
i
t
h
t
h
e
o
t
h
e
r
d
e
n
o
i
s
e
d
i
m
a
g
e
s
,
w
h
i
c
h
i
s
c
o
m
p
u
t
e
d
b
y
s
t
a
n
d
a
r
d
m
e
d
i
a
n
f
i
l
t
e
r
(
S
M
F
)
,
g
a
u
s
s
i
a
n
f
i
l
t
e
r
a
n
d
a
d
a
p
t
i
v
e
m
e
d
i
a
n
f
i
l
t
e
r
(
A
M
F
)
f
o
r
d
e
m
o
n
s
t
r
a
t
i
n
g
t
h
e
D
B
A
M
F
c
a
p
a
c
i
t
y
u
n
d
e
r
s
u
b
j
e
c
t
i
v
e
m
e
a
s
u
r
e
m
e
n
t
a
s
p
e
c
t
.
Ke
yw
or
d
s
:
AMF
(ad
a
ptive
m
edian
filt
er)
Digital
im
age d
en
oisin
g
ROA
D (ra
nk
-
orde
red ab
so
l
ute
diff
e
re
nces)
SMF (s
ta
ndar
d m
edian
filt
er)
Copyright
©
202
0
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Vora
poj
Pata
na
vij
it
,
Dep
a
rtm
en
t
of
Ele
ct
rical
an
d
Ele
ct
ro
nic
Eng
ineerin
g,
Faculty
of E
ngineerin
g, As
sum
pt
ion
Un
i
versi
ty
,
PK
E Bl
dg., 2
nd Fl
r.
,
88 M
oo
8
Ba
ng N
a
-
Tra
d Km
. 2
6, Ba
ngsa
othong,
Sa
m
uth
pr
aka
rn
1054
0,
Thail
an
d.
Em
a
il
:
patanavi
j
it
@yahoo.
c
om
1.
RELE
VANT
RESEA
R
CHE
D
OF
D
E
N
OISI
NG MET
HODS F
OF S
PN
B
e
c
a
u
s
e
o
f
d
i
v
e
r
s
e
r
e
a
s
o
n
s
s
u
c
h
a
s
f
a
u
l
t
i
n
s
y
n
c
h
r
o
n
i
z
a
t
i
o
n
o
f
a
n
a
l
o
g
t
o
d
i
g
i
t
a
l
p
r
o
c
e
s
s
,
m
a
l
f
u
n
c
t
i
o
n
o
f
C
C
D
c
h
i
p
,
f
a
u
l
t
i
n
a
d
d
r
e
s
s
i
n
g
o
f
s
t
o
r
a
g
e
p
r
o
c
e
s
s
,
a
n
d
f
a
u
l
t
i
n
t
r
a
n
s
m
i
s
s
i
o
n
,
e
t
c
.
,
t
h
e
i
m
p
u
l
s
i
v
e
n
o
i
s
e
[
1
-
5
]
c
a
n
b
e
s
e
p
a
r
a
t
e
d
i
n
t
o
t
w
o
m
a
i
n
c
a
t
e
g
o
r
i
e
s
:
S
a
l
t
a
n
d
p
e
p
p
e
r
n
o
i
s
e
(
S
P
N
)
a
n
d
r
a
n
d
o
m
m
a
g
n
i
t
u
d
e
i
m
p
u
l
s
i
v
e
n
o
i
s
e
(
R
M
I
N
)
f
r
o
m
a
l
g
e
b
r
a
i
c
f
o
r
m
u
l
a
t
i
o
n
a
s
p
e
c
t
.
H
e
n
c
e
,
m
a
y
S
P
N
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
s
[
1
-
9
]
h
a
v
e
b
e
e
n
e
x
a
m
i
n
e
d
f
o
r
m
o
r
e
t
h
a
n
t
w
o
d
e
c
a
d
e
s
d
u
e
t
o
d
e
m
a
n
d
i
n
g
o
f
m
o
d
e
r
n
d
i
g
i
t
a
l
i
m
a
g
e
a
p
p
l
i
c
a
t
i
o
n
s
[
10
-
2
3
]
:
r
e
t
i
n
a
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
s
u
p
e
r
r
e
s
o
l
u
t
i
o
n
,
e
m
o
t
i
o
n
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
e
t
c
.
A
t
f
i
r
s
t
,
t
h
e
S
M
F
(
S
t
a
n
d
a
r
d
M
e
d
i
a
n
F
i
l
t
e
r
)
[
8
,
2
4
]
i
s
d
i
s
c
o
v
e
r
e
d
i
n
1
9
7
5
f
o
r
e
x
c
l
u
d
i
n
g
S
P
N
(
S
a
l
t
a
n
d
P
e
p
p
e
r
N
o
i
s
e
)
a
n
d
,
l
a
t
e
r
,
i
s
b
e
c
o
m
e
o
n
e
o
f
t
h
e
m
o
s
t
c
a
p
a
b
l
e
a
n
d
p
r
a
c
t
i
c
a
l
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
s
f
r
o
m
t
h
e
f
a
c
t
t
h
a
t
t
h
i
s
m
e
t
h
o
d
i
s
l
o
w
c
o
m
p
u
t
a
t
i
o
n
a
n
d
h
i
g
h
c
a
p
a
b
i
l
i
t
y
.
T
h
e
d
e
n
o
i
s
e
d
m
e
t
h
o
d
b
a
s
e
d
o
n
G
a
u
s
s
a
i
n
f
i
l
t
e
r
[24
]
i
s
d
e
s
i
r
e
d
f
o
r
w
e
l
l
a
p
p
l
y
i
n
g
o
n
G
a
u
s
s
a
i
n
n
o
i
s
e
b
u
t
t
h
i
s
m
e
t
h
o
d
h
a
s
p
o
o
r
p
e
r
f
o
r
m
a
n
c
e
f
o
r
S
P
N
.
L
a
t
e
r
,
a
d
a
p
t
i
v
e
m
e
d
i
a
n
f
i
l
t
e
r
(
A
M
F
)
[
7
,
2
5
]
,
w
h
i
c
h
i
s
i
m
p
r
o
v
e
d
f
r
o
m
t
h
e
S
M
F
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
b
y
v
a
r
i
n
g
i
t
s
w
i
n
d
o
w
s
i
z
e
,
i
s
p
r
o
p
o
s
e
d
a
n
d
i
t
s
p
e
r
f
o
r
m
a
n
c
e
i
s
b
e
t
t
e
r
t
h
a
n
t
h
e
S
M
F
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
.
A
f
t
e
r
t
w
o
d
e
c
a
d
e
s
,
t
h
e
m
o
d
e
r
n
d
e
c
i
s
i
o
n
b
a
s
e
d
a
d
a
p
t
i
v
e
m
e
d
i
a
n
f
i
l
t
e
r
(
D
B
A
M
F
)
[
3
,
2
5
]
,
w
h
i
c
h
i
s
i
m
p
r
o
v
e
d
f
r
o
m
t
h
e
S
M
F
-
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
,
i
s
d
i
s
c
o
v
e
r
e
d
f
o
r
d
e
n
o
i
s
i
n
g
R
M
I
N
i
n
2
0
1
0
.
T
h
e
D
B
A
M
F
m
e
t
h
o
d
i
s
f
o
r
m
u
l
a
t
e
d
f
r
o
m
t
w
o
m
a
i
n
s
c
h
e
m
e
s
:
n
o
i
s
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
s
c
h
e
m
e
s
(
t
h
a
t
i
s
f
o
u
n
d
o
n
R
O
A
D
(
R
a
n
k
O
r
d
e
r
A
b
s
o
l
u
t
e
D
i
f
f
e
r
e
n
c
e
)
)
a
n
d
n
o
i
s
e
e
x
c
l
u
s
i
o
n
s
c
h
e
m
e
s
(
t
h
a
t
i
s
f
o
u
n
d
o
n
S
M
F
[
5
,
2
5
]
)
.
U
n
f
o
r
t
u
n
a
t
e
l
y
,
t
h
e
r
e
a
r
e
n
o
s
e
c
u
r
i
t
i
z
e
d
r
e
s
e
a
r
c
h
e
s
o
f
D
B
A
M
F
c
a
p
a
b
i
l
i
t
y
a
n
d
i
t
s
c
o
n
s
t
r
a
i
n
t
w
h
e
n
t
h
i
s
d
e
n
o
i
s
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
4
,
A
ugus
t
2020
:
4109
-
4117
4110
m
e
t
h
o
d
i
s
i
m
p
l
e
m
e
n
t
e
d
o
n
S
P
N
a
t
d
i
v
e
r
s
e
i
n
t
e
n
s
i
t
y
.
C
o
n
s
e
q
u
e
n
t
l
y
,
t
h
i
s
a
n
a
l
y
z
e
d
r
e
p
o
r
t
f
o
c
u
s
e
s
t
o
e
x
a
m
i
n
e
t
h
e
s
i
m
i
l
a
r
i
t
y
c
a
p
a
c
i
t
y
o
f
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
[
3
]
f
o
r
d
i
v
e
r
s
e
S
P
N
s
u
r
r
o
u
n
d
i
n
g
i
n
o
r
d
e
r
t
o
a
n
a
l
y
t
i
c
a
l
l
y
u
n
d
e
r
s
t
a
n
d
i
t
s
u
p
p
e
r
b
o
u
n
d
o
f
i
t
s
p
e
r
f
o
r
m
a
n
c
e
a
n
d
i
t
s
l
i
m
i
t
a
t
i
o
n
f
o
r
f
u
t
u
r
e
i
m
p
l
e
m
e
n
t
a
t
i
o
n
s
.
2.
THE
PR
I
M
A
RY CON
CEP
T OF
DBA
M
F
The
dist
or
te
d
portrait
is
m
a
t
hem
atical
ly
ex
plained
a
s
Y
an
d
the
portrait
intensit
y
is
m
ath
em
atical
ly
exp
la
ine
d
as
,
y
i
j
.
The
DBAMF
schem
e
[3
,
25
]
can
be
sep
arated
into
tw
o
pr
im
ary
sche
m
es:
no
is
e
recog
nizing sc
hem
e and
no
is
e re
pairin
g
sc
hem
e, w
hic
h
ca
n be c
om
pr
ehe
ns
ively
r
e
view
ed
as
upc
om
in
g.
2.1. The
prim
ary co
ncep
t of noise
reco
gn
i
z
ing s
cheme
The
perf
or
m
ing
a
rithm
etic
con
cept
of the
no
ise
r
eco
gniz
in
g schem
e can be
re
viewe
d
as.
Determ
ine
the
cal
culat
ed
s
quare
re
gion
33
W
at
33
(
3
w
)
of
the
pr
ocesse
d
portra
it
pix
el
s
at
,
ij
coor
din
at
io
n.
Determ
ine
the
abso
l
ute
difference
(
,,
s
t
i
j
Dy
)
with
norm
alizat
ion
,
so
cal
le
d
N
AD,
of
the
pr
ocesse
d
portrait
p
i
xel
with m
idd
le
m
o
st co
ordinati
on
,
ij
, whic
h
ca
n be
co
m
pr
ehe
ns
iv
el
y cl
arified as up
c
om
ing
.
,
,
,
,
t
255
s
t
i
j
i
j
s
D
y
y
y
(1)
Determ
ine
the
vecto
r
of
a
bs
ol
ute
di
ff
e
r
ence
(
,,
s
t
i
j
Dy
)
with
norm
al
iz
a
ti
on
,
s
o
cal
le
d
NRO
AD
(the
processe
d
portrait
pi
xel
with
m
idd
le
m
o
st
co
ordinati
on
,
ij
),
w
hich
a
re
al
ign
e
d
for
st
or
i
ng
only
fi
ve
unde
rm
os
t
values
f
ro
m
ei
gh
values
i
n
the
c
al
culat
ed
squa
r
e
reg
i
on.
Lat
er
,
the
sta
ti
sti
cal
m
ean
of
NRO
AD
can
be
c
om
pr
ehensi
vely
clari
fied
as
upc
om
i
ng.
5
1
5
,
,
5
1
RO
A
D
m
s
t
i
j
m
Dy
(2)
F
r
o
m
N
R
O
A
D
,
i
f
t
h
e
s
t
a
t
i
s
t
i
c
a
l
m
e
a
n
o
f
N
R
O
A
D
,
w
h
i
c
h
f
l
u
c
t
u
a
t
e
s
b
e
t
w
e
e
n
0
.
0
0
t
o
1
.
0
0
f
o
r
a
l
l
p
i
x
e
l
s
i
n
t
h
e
p
r
o
c
e
s
s
e
d
p
o
r
t
r
a
i
t
,
i
s
g
r
e
a
t
e
r
t
h
a
n
a
s
t
a
b
l
e
c
o
n
s
t
a
n
t
0
T
[3,
25
]
t
h
e
n
t
h
e
p
r
o
c
e
s
s
e
d
p
o
r
t
r
a
i
t
p
i
x
e
l
i
s
r
e
c
o
g
n
i
z
e
d
a
s
t
h
e
d
i
s
t
o
r
t
e
d
p
i
x
e
l
,
o
t
h
e
r
w
i
s
e
t
h
e
n
t
h
e
p
r
o
c
e
s
s
e
d
p
o
r
t
r
a
i
t
p
i
x
e
l
i
s
r
e
c
o
g
n
i
z
e
d
a
s
t
h
e
n
o
i
s
e
-
f
r
e
e
p
i
x
e
l
.
T
h
e
r
e
f
o
r
e
,
t
h
e
t
h
e
N
o
i
s
e
D
e
t
e
c
t
e
d
M
a
t
r
i
x
c
a
n
b
e
c
o
m
p
r
e
h
e
n
s
i
v
e
l
y
c
l
a
r
i
f
i
e
d
a
s
u
p
c
o
m
i
n
g
.
5
0
5
N
D
M
1
,
i
f
R
O
A
D
o
t
h
e
r
w
i
s
e
0
,
i
f
R
O
A
D
mm
TT
(3)
Fr
om
the
abov
e
no
ise
rec
ogni
zi
ng
schem
e
of
the
DBAM
F,
we
can
com
prehensi
vely
dis
play
this
proce
ssing
schem
e in the up
c
om
ing
f
l
owchar
t as
F
ig
ure
1
.
2.2. The
prim
ary co
ncep
t of noise
rep
airing sc
heme
The
perf
or
m
ing
a
rithm
etic
con
cept
of the
no
ise
r
e
pai
rin
g
sc
hem
e can
be
re
viewe
d
as
.
Determ
ine
the
cal
culat
ed
square
re
gion
33
W
at
33
(
3
w
)
of
ND
M
(
no
ise
detect
ed
m
at
rix)
at
,
ij
coor
din
at
io
n.
Fr
om
the
cal
culat
ed
squa
re
r
egio
n
of
N
DM
,
if
the
total
noise
-
fr
ee
pix
el
s
is
few
er
t
han
three
pi
xels
then
the d
im
ension
of the calc
ulate
d
s
quare
re
gion
33
W
is ex
pa
nd
e
d by one a
nd the
Step
2
is
reex
e
cuted.
Fr
om
the
cal
culat
ed
s
qu
a
re
reg
i
on
of
N
D
M,
if
the
total
distor
te
d
pi
xe
ls
is
m
or
e
than
tw
o
pi
xels
then
the r
e
paire
d pi
xel is e
xecu
te
d by as
upc
om
in
g
.
,,
ˆ
m
e
d
ia
n
,
i
j
i
s
j
t
F
Y
Y
s
t
W
(4)
The d
up
li
cat
ed
sh
em
e is re e
xe
cuted f
or
e
ve
r
y pixels i
n
the
processe
d p
or
tr
ai
t pixels.
Fr
om
the
a
bove
noise
re
pai
ring
sc
hem
e
of
the
DBAMF,
we
ca
n
c
om
prehensi
vely
dis
play
this
proc
essin
g
schem
e in the up
c
om
ing
f
l
owchar
t as
F
ig
ure
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Compu
t
atio
na
l
S
cr
utiny
of im
ag
e
d
e
noisi
ng
meth
od fo
und on
…
(
V
orapoj
Patan
aviji
t
)
4111
x
-
D
im
e
n
sio
n
i
Y
e
s
33
Se
t
w
ith
c
e
n
te
r
a
t
,
y
i
j
W
1
i
y
-
D
im
e
nsion
j
Y
e
s
S
ta
r
t
R
e
a
d N
oisy
I
m
a
ge
Y
1
j
No
1
jj
1
ii
End
No
,
,
,
,
t
s
t
i
j
i
j
s
D
y
y
y
5
5
,
,
1
R
O
A
D
m
s
t
i
j
m
Dy
,
is
a
n
o
ise
l
e
ss
p
ixe
l
N
o
ise
_
D
e
te
c
tion
,
0
y
i
j
ij
5
RO
A
D
4
0
m
N
o
N
o
ise
l
e
ss
Y
e
s
N
ois
y
,
i
s
a
n
o
is
y
p
ix
e
l
N
o
is
e
_
D
e
te
c
tio
n
,
1
y
i
j
ij
x
-
D
im
e
n
sio
n
i
Y
e
s
1
i
y
-
D
im
e
nsion
j
Y
e
s
S
ta
r
t
R
e
a
d N
oisy
I
m
a
ge
Y
1
j
No
1
jj
1
ii
End
No
,
is no
isy
p
ixe
l
y
i
j
Y
e
s
N
o
isy
Pixe
l
N
oise
l
e
ss
N
o
P
ixe
l
_3
N
oise
le
ss
Pix
e
ls
,,
m
e
d
ia
n
,
i
j
i
s
j
t
F
Y
X
s
t
W
3
F
W
No
1
FF
WW
Y
e
s
7
F
W
Y
e
s
No
Figure
1.
The
a
rithm
etic
con
c
ept of t
he no
ise
recog
nizing sc
hem
e
Figure
2
.
The
a
rithm
etic
con
c
ept of t
he no
ise
rep
ai
ri
ng sc
he
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
4
,
A
ugus
t
2020
:
4109
-
4117
4112
3.
THE
C
O
MP
U
TATION
AL
E
X
APLE
OF
DBA
MF CO
N
CEPT
I
n
f
i
r
s
t
c
a
s
e
s
,
t
h
i
s
p
a
r
t
c
om
p
r
e
h
e
n
s
i
v
e
l
y
r
e
vi
e
w
e
s
t
h
e
c
a
l
c
ul
a
t
i
o
n
o
f
t
h
e
e
x
a
m
p
l
e
o
f
D
B
A
M
F
n
o
i
s
e
r
e
c
o
g
n
i
z
i
n
g
s
c
h
e
m
e
a
s
i
n
F
i
gu
r
e
3
(
a
)
f
o
r
o
b
v
i
o
u
s
l
y
r
e
v
i
e
w
i
n
g
t
h
e
p
r
o
c
e
s
s
e
d
c
a
l
c
u
l
a
t
i
o
n
w
h
e
r
e
,
ij
y
is
a
d
i
s
t
o
r
e
d
p
i
x
e
l
,
w
h
i
c
h
i
s
d
i
s
t
o
r
t
e
d
b
y
f
o
r
s
a
l
t
a
n
d
p
e
p
p
e
r
n
o
i
s
e
(
,
255
ij
y
)
.
L
a
t
e
r
,
t
h
e
d
i
s
t
o
r
e
d
p
i
x
e
l
i
s
r
e
p
a
i
r
e
d
a
s
i
n
F
i
g
u
r
e
3
(
b
)
w
h
e
r
e
t
h
e
d
i
s
t
o
r
e
d
p
i
x
e
l
(
,
255
ij
y
)
i
s
r
e
p
a
i
r
e
d
t
o
b
e
t
h
e
r
e
p
a
i
r
e
d
p
i
x
e
l
(
,
118
ij
y
).
1
,
1
131
255
y
i
j
1,
121
255
y
i
j
1
,
1
113
255
y
i
j
,
255
y
i
j
255
,1
118
255
y
i
j
1
,
1
255
255
y
i
j
1,
0
255
y
i
j
,1
255
255
y
i
j
1
,
1
110
255
y
i
j
1
,
is a
n
oisy
d
a
ta
y
i
j
,
,
,
,
t
s
t
i
j
i
j
s
D
y
y
y
1
,
1
1
255
255
y
i
j
13
1,
255
255
y
i
j
121
1
,
1
255
255
y
i
j
113
,1
255
255
y
i
j
118
1
,
1
255
255
y
i
j
255
1,
255
255
y
i
j
0
,1
255
255
y
i
j
255
1
,
1
255
255
y
i
j
110
,
,
,
,
t
s
t
i
j
i
j
s
D
y
y
y
5
1
5,
1
N
RO
A
D
m
k
i
j
m
m
Ry
1
5
5
R
O
L
D
0
0
0
0.4863
0.5255
m
1
5
5
R
O
L
D
1
.0
1
1
8
0
.2
0
2
4
m
,
255
255
y
i
j
255
1
,
1
124
255
y
i
j
1,
134
255
y
i
j
1
,
1
142
255
y
i
j
,
0
255
y
i
j
,1
137
255
y
i
j
1
,
1
0
255
y
i
j
1,
255
255
y
i
j
,1
0
255
y
i
j
1
,
1
145
255
y
i
j
,,
s
o
r
t
0
,
0
,
0
,
0
.
4
8
6
3
,
0
.
5
2
5
5
,
0
.
5
3
7
3
,
0
.
5
5
6
9
,
0
.
5
6
8
6
,
1
k
i
j
s
t
R
y
D
,
0
.
4
8
6
3
,
0
.
5
3
7
3
,
0
,
0
.
5
2
5
5
,
0
,
1
,
0
.
5
5
6
9
,
0
,
0
.
5
6
86
st
D
,
,
,
,
t
s
t
i
j
i
j
s
D
y
y
y
1
,
1
0.4863
y
i
j
1,
0.5255
y
i
j
1
,
1
0.5569
y
i
j
,
0
y
i
j
,1
0.53
73
y
i
j
1
,
1
0
y
i
j
1,
1.00
00
y
i
j
,1
0
y
i
j
1
,
1
0.5686
y
i
j
40
1
5
2
5
5
5
RO
L
D
0
.0
3
1
4
N
o
ise
D
e
te
c
te
d
I
m
a
g
e
,
1
,
N
o
isy
Pixe
l
m
ij
If
The
n
(a)
N
o
ise
D
e
te
c
te
d
I
m
a
g
e
1
,
1
0
y
i
j
1,
0
y
i
j
1
,
1
0
y
i
j
,1
0
y
i
j
1
,
1
1
y
i
j
1,
1
y
i
j
,1
1
y
i
j
1
,
1
0
y
i
j
,
1
y
i
j
noise
l
e
ss
pixe
l
s
133
118
ˆ
,
m
e
dia
n
121
113
110
y
i
j
noi
se
l
e
ss
pi
xe
l
s
ˆ
,
m
e
di
a
n
131
,
118
,
121
,
113
,
110
y
i
j
ˆ
,
118
y
i
j
1
,
1
131
255
y
i
j
1,
121
255
y
i
j
1
,
1
113
255
y
i
j
,
255
y
i
j
255
,1
118
255
y
i
j
1
,
1
255
255
y
i
j
1,
0
255
y
i
j
,1
255
255
y
i
j
1
,
1
110
255
y
i
j
1
,
is a
n
oisy
d
a
ta
y
i
j
n
o
is
e
l
e
s
s
p
ix
e
l
s
ˆ
,
m
e
d
ia
n
1
1
0
,1
1
3
,1
1
8
,1
2
1
,1
3
1
y
i
j
Figure
3
(a)
.
T
he
c
om
pu
te
r
e
xam
ple o
f
t
he no
ise
r
ec
ognizi
ng sc
hem
e
,
(b)
T
he
c
om
pute
r
exam
ple of
the noise
rep
ai
rin
g
sc
hem
e
4.
RESU
LT
S
AND DI
SCUS
S
ION
In
t
his
exam
ining
of
the D
BA
MF
de
no
isi
ng
capacit
y,
the
c
al
culat
ion
s
of
t
war
e
i
n
this
a
na
ly
zed
repo
rt
is
MATLAB
pro
gr
am
that
i
s
run
on
w
orkstat
ion
c
om
pu
te
rs
with
the
hard
war
e
detai
l:
the
CPU
is
In
te
l
i7
-
6700H
Q
a
nd
t
he
inte
rn
al
m
e
m
or
y
is
16
GB
a
nd
al
l
w
orkst
at
io
n
c
om
pu
te
rs
sim
ulate
on
di
verse
por
trai
ts,
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Compu
t
atio
na
l
S
cr
utiny
of im
ag
e
d
e
noisi
ng
meth
od fo
und on
…
(
V
orapoj
Patan
aviji
t
)
4113
wh
ic
h
are
co
nt
ai
ned
of
Air
pl
ane,
Peppe
r,
Girl
and
Le
na
,
at
nu
m
ero
us
SPN
de
ns
it
ie
s
wh
e
re
al
l
div
erse
portrait
s
that
are
disto
rted
by
add
in
g
synt
he
siz
ed
SP
noise
.
All
distor
te
d
portrait
s
are
re
paire
d
for
obta
ini
ng
the
finest
qual
it
y
and
best
PSN
R
by
execu
ti
ng
the
im
age
den
oisi
ng
m
et
ho
d
f
ound
on
D
BAMF
for
firs
t
no
ise
recog
nizing
sc
hem
e
(in
orde
r
to
rec
ognize
wh
et
her
the
pi
xel
is
noise
-
free
or
noisy
)
a
nd,
la
te
r,
noise
rep
ai
r
schem
e (in
or
de
r
to
r
e
pair o
nly t
he
nois
y
pixe
ls).
4
.
1.
The e
xp
e
ri
menta
l i
n
ves
tig
at
io
n
of n
oise reco
gn
iz
ing
scheme
This
si
m
ulate
d
ex
per
im
ent
sect
ion
in
vestigat
es
the
op
ti
m
iz
ed
sta
ble
const
ant
0
T
fo
r
pro
vid
in
g
the
fine
st
qual
it
y
and
be
st
P
SN
R
as
s
how
n
in
Ta
ble
1
-
4.
The
sta
ble
co
nst
ant
0
T
,
wh
ic
h
f
luctuat
es
be
tw
een
0.00
t
o
1.0
0
for
al
l
pix
el
s
in
t
he
pr
ocesse
d
portrait
,
ultim
ately
i
m
pacts
to
the
de
no
isi
ng
capaci
ty
of
D
BAMF
m
et
ho
d.
C
on
se
qu
e
nt
ly
,
this
com
pu
te
r
exam
i
ning
com
pr
ehe
ns
ively
determ
ines
the
sta
ble
con
sta
nt
0
T
,
wh
i
c
h
m
ake
the
fines
t
qu
al
it
y
and
best
PSN
R
w
he
n
each
disto
rt
ed
portrait
is
execu
te
d
by
de
no
isi
ng
ca
pac
it
y
o
f
DBAMF
m
et
ho
d.
T
he
num
ero
us
di
gital
por
trai
t
s
(which
a
re
co
ntaine
d
of
Air
plane
,
Pe
pp
e
r,
Girl
a
nd
Lena
)
are
use
d
t
o
a
naly
ze
the
sta
ble
co
ns
ta
nt
0
T
by
va
ryi
ng
f
r
om
0.
00
to
0.50
at
0.0
25
in
crem
ented
ste
ps
as
disp
la
ye
d i
n T
able 1 t
o Table
4
,
r
es
pecti
vely
.
Fr
om
these
c
om
pu
te
r
exam
i
ning
of
Girl
i
n
Ta
ble
1,
the
optim
iz
ed
pre
-
sp
eci
fied
c
onsta
nt
0
T
is
ab
out
0.115
3
0.0
00
5 or be
f
luct
uat
ed fr
om
0
.075
to 0.15
0 for m
akin
g
the
f
i
nes
t DBAMF
d
e
noisi
ng ca
pacit
y
Fr
om
these
c
om
pu
te
r
exam
ining
of
Pe
pp
e
r
in
Table
2,
the
optim
iz
ed
pr
e
-
sp
eci
fied
c
ons
ta
nt
0
T
is
ab
out
0.106
9
0.0
010 o
r be f
l
uctuate
d from
0
.025 to
0.150 f
or
m
akin
g
the
f
i
nest
DBAMF
d
e
noisi
ng capacit
y.
T
a
b
l
e
1
.
T
h
e
d
e
n
o
i
s
i
n
g
c
a
p
a
c
i
t
y
i
n
t
e
r
c
o
n
n
e
c
t
i
o
n
o
f
P
S
N
R
a
n
d
t
h
e
s
t
a
b
l
e
c
o
n
s
t
a
n
t
0
T
for
a
i
r
p
l
a
n
e
i
m
a
g
e
Table
2.
T
he
deno
isi
ng capac
it
y i
nterco
nnec
ti
on
of PS
NR a
nd the
sta
ble c
on
sta
nt
0
T
for pe
pp
e
r
im
age
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
4
,
A
ugus
t
2020
:
4109
-
4117
4114
Fr
om
these
com
pu
te
r
exam
i
ning
of
Le
na
in
Table
3,
the
op
ti
m
iz
ed
pr
e
-
sp
eci
fied
co
nst
ant
0
T
is
abo
ut
0.112
5
0.0
042 o
r be f
l
uctuate
d from
0
.050 to
0.175 f
or
m
akin
g
the
f
i
nest
DBAMF
d
e
noisi
ng capacit
y.
Fr
om
these
c
om
pu
te
r
exam
ining
of
Ai
rp
la
ne
in
Ta
ble
4,
t
he
op
ti
m
iz
ed
pr
e
-
s
pecified
co
ns
ta
nt
0
T
is
ab
out
0.109
7
0.0
014 o
r be f
l
uctuate
d from
0
.050 to
0.150 f
or
m
akin
g
the
f
i
nest
DBAMF
d
e
noisi
ng capacit
y.
Table
3
. T
he
deno
isi
ng capa
c
it
y i
nterco
nnec
ti
on
of PS
NR a
nd the
s
ta
ble
c
on
sta
nt
0
T
for
gir
l
i
m
age
Table
4
. T
he
deno
isi
ng capac
it
y i
nterco
nnec
ti
on
of PS
NR a
nd the
s
ta
ble
c
on
sta
nt
0
T
for
L
E
NA im
age
4
.
2
.
The e
xp
e
ri
menta
l
in
ves
tig
at
io
n
of imag
e
den
oisin
g me
thod
found
o
n
D
B
AMF
T
h
i
s
a
n
a
l
y
z
e
d
r
e
p
o
r
t
f
o
c
u
s
e
s
t
o
e
x
a
m
i
n
e
t
h
e
c
o
m
p
u
t
a
t
i
o
n
a
l
s
c
r
u
t
i
n
y
o
f
i
m
a
g
e
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
u
n
d
e
r
S
P
N
s
u
r
r
o
u
n
d
i
n
g
.
I
n
t
h
i
s
e
x
a
m
i
n
i
n
g
o
f
t
h
e
D
B
A
M
F
d
e
n
o
i
s
i
n
g
c
a
p
a
c
i
t
y
,
f
o
u
r
a
n
a
l
y
z
e
d
i
m
a
g
e
s
,
w
h
i
c
h
a
r
e
c
o
n
t
a
i
n
e
d
o
f
A
i
r
p
l
a
n
e
,
P
e
p
p
e
r
,
G
i
r
l
a
n
d
L
e
n
a
,
a
r
e
u
s
e
d
t
o
a
n
a
l
y
z
e
d
b
y
i
n
i
t
i
a
l
l
y
a
d
d
i
n
g
s
y
n
t
h
e
s
i
z
e
d
S
P
n
o
i
s
e
f
o
r
c
r
e
a
t
i
n
g
n
u
m
e
r
o
u
s
d
i
s
t
o
r
t
e
d
p
o
r
t
r
a
i
t
s
.
L
a
t
e
r
,
a
l
l
d
i
s
t
o
r
t
e
d
p
o
r
t
r
a
i
t
s
a
r
e
r
e
p
a
i
r
e
d
f
o
r
o
b
t
a
i
n
i
n
g
t
h
e
f
i
n
e
s
t
q
u
a
l
i
t
y
a
n
d
b
e
s
t
P
S
N
R
b
y
e
x
e
c
u
t
i
n
g
t
h
e
i
m
a
g
e
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
.
F
r
o
m
t
h
e
a
b
o
v
e
e
x
a
m
i
n
i
n
g
,
t
h
e
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
s
b
y
a
p
p
l
y
i
n
g
A
M
F
[7
,
2
5
]
a
n
d
D
B
A
M
F
p
r
o
d
u
c
e
t
h
e
f
i
n
e
s
t
q
u
a
l
i
t
y
a
n
d
b
e
s
t
P
S
N
R
t
h
a
n
o
t
h
e
r
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
s
f
o
r
i
n
s
t
a
n
t
S
M
F
a
n
d
G
a
u
s
s
i
a
n
f
i
l
t
e
r
.
H
o
w
e
v
e
r
,
t
h
e
D
B
A
M
F
h
a
s
m
a
r
g
i
n
a
l
l
y
i
m
p
r
o
v
e
d
t
h
a
n
A
M
F
f
r
o
m
t
h
a
t
f
a
c
t
t
h
a
t
t
h
e
D
B
A
M
F
i
s
i
n
i
t
i
a
l
l
y
d
e
v
e
l
o
p
e
d
s
o
l
e
l
y
f
o
r
r
a
n
d
o
m
m
a
g
n
i
t
u
d
e
i
m
p
u
l
s
i
v
e
n
o
i
s
e
(
R
M
I
N
)
b
u
t
t
h
e
A
M
F
i
s
d
e
v
e
l
o
p
e
d
s
o
l
e
l
y
f
o
r
S
P
N
.
F
r
o
m
t
h
e
s
e
c
o
m
p
u
t
e
r
e
x
a
m
i
n
i
n
g
o
f
t
h
e
d
e
n
o
i
s
i
n
g
c
a
p
a
c
i
t
y
i
n
T
a
b
l
e
5
(
a
)
f
o
r
L
e
n
a
a
n
d
P
e
p
p
e
r
a
n
d
T
a
b
l
e
5
(
b
)
f
o
r
G
i
r
l
a
n
d
A
i
r
p
l
a
n
e
,
a
l
t
h
r
o
g
h
t
h
e
D
B
A
M
F
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
i
s
o
r
i
g
i
n
a
l
l
y
d
e
s
i
r
e
d
f
o
r
R
V
I
N
,
t
h
e
D
B
A
M
F
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
c
a
n
p
r
o
v
i
d
e
t
h
e
f
i
n
e
r
e
s
u
l
t
s
(
t
h
e
d
e
n
o
i
s
e
d
i
m
a
g
e
s
w
i
t
h
h
i
g
h
q
u
a
l
i
t
y
)
.
F
r
o
m
t
h
e
s
e
i
n
v
e
r
s
t
i
g
a
t
i
o
n
,
t
h
e
D
B
A
M
F
m
e
t
h
o
d
a
n
d
a
d
a
p
t
i
v
e
m
e
d
i
a
n
f
i
l
t
e
r
(
A
M
F
)
c
a
n
p
r
o
d
u
c
e
t
h
e
d
e
n
o
i
s
e
d
i
m
a
g
e
w
i
t
h
f
i
n
e
r
q
u
a
l
i
t
y
a
n
d
h
i
g
h
P
S
N
R
,
w
h
i
c
h
i
s
c
o
n
f
r
o
n
t
e
d
w
i
t
h
t
h
e
o
t
h
e
r
d
e
n
o
i
s
e
d
i
m
a
g
e
s
,
w
h
i
c
h
i
s
c
o
m
p
u
t
e
d
b
y
s
t
a
n
d
a
r
d
m
e
d
i
a
n
f
i
l
t
e
r
(
S
M
F
)
a
n
d
G
a
u
s
s
i
a
n
F
i
l
t
e
r
.
D
u
e
t
o
c
i
r
c
u
m
s
p
e
c
t
i
o
n
o
f
s
h
e
e
t
o
f
p
a
p
e
r
,
s
o
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Compu
t
atio
na
l
S
cr
utiny
of im
ag
e
d
e
noisi
ng
meth
od fo
und on
…
(
V
orapoj
Patan
aviji
t
)
4115
g
r
a
p
h
i
c
a
l
r
e
s
u
l
t
s
(
o
f
L
e
n
a
i
m
a
g
e
a
t
1
0
%
a
n
d
2
0
%
)
o
f
t
h
e
d
e
n
o
i
s
e
d
D
B
A
M
F
m
e
t
h
o
d
a
n
d
o
t
h
e
r
d
e
n
o
i
s
e
d
m
e
t
h
o
d
s
a
r
e
s
h
o
w
n
i
n
F
i
g
u
r
e
4
.
Table
5
(a
). T
he
an
al
ysi
s r
e
po
rt of the
d
e
nois
ing
ca
pacit
y
of
D
BAM
F
unde
r
SP
N su
rro
unding
SPN
PNSR (d
B)
Operated i
mages
Noise densi
t
y
LR i
m
age
Noise supp
ressing
tech
nique
Median (3x
3)
Mean (3x3)
AMF
DBA MF
Lena
(256x256)
D=0.05
18.7139
31.6421
22.4181
36.0907
40.0318
D=0.10
15.6564
30.7076
19.3812
35.3032
34.4817
D=0.15
13.8274
29.2982
17.5385
33.7454
29.3024
D=0.20
12.6389
27.6257
16.3208
32.1558
27.1381
D=0.25
11.6783
25.4101
15.3526
29.8105
24.7729
D=0.30
10.8971
23.6811
14.5829
27.9141
22.9755
D=0.35
10.2240
20.8127
13.8785
25.6654
21.0287
D=0.40
9.6481
19.0080
13.2479
23.7903
19.5071
D=0.45
9.0745
16.8389
12.6598
21.5949
17.8200
D=0.50
8.6553
15.4758
12.2146
20.5725
16.1734
D=0.55
8.2118
13.8573
11.7609
19.4896
14.1775
D=0.60
7.7813
12.3280
11.2939
18.1747
11.7120
D=0.65
7.4884
11.3251
11.0012
17.7283
10.3838
D=0.70
7.1697
10.2861
10.6509
17.1153
8.9514
D=0.75
6.8497
9.1271
10.2599
16.5388
7.5456
D=0.80
6.5846
8.3331
10.0057
16.4554
7.0520
D=0.85
6.3241
7.5344
9.7338
16.4230
6.4819
D=0.90
6.0604
6.8241
9.4356
16.5352
6.1742
Pepper
(256x256)
D=0.05
18.4752
32.2578
22.1408
37.1145
37.5975
D=0.10
15.3798
30.6116
19.0677
36.0391
32.8628
D=0.15
13.5570
28.8470
17.2234
33.6095
28.5840
D=0.20
12.3593
26.5888
15.9804
31.6485
25.9117
D=0.25
11.3929
24.2073
14.9986
29.4205
23.6700
D=0.30
10.6242
22.0663
14.1748
26.7650
21.7606
D=0.35
9.9742
20.3774
13.5209
25.5249
20.3507
D=0.40
9.3998
18.4321
12.9076
23.4995
18.4004
D=0.45
8.8599
16.6168
12.3275
21.7177
17.0967
D=0.50
8.3843
14.8506
11.8117
20.2203
15.3313
D=0.55
7.9930
13.4655
11.3720
19.0894
13.5815
D=0.60
7.6189
12.0128
10.9563
18.1116
11.9462
D=0.65
7.2684
10.8920
10.5158
17.3657
10.3517
D=0.70
6.9246
9.7704
10.2039
16.5923
8.9034
D=0.75
6.6418
8.8751
9.8955
16.2338
7.8407
D=0.80
6.3710
8.0166
9.252.
16.0896
7.0426
D=0.85
6.1097
7.2402
9.2949
16.0498
6.5095
D=0.90
5.8582
6.5767
9.0214
16.2932
6.2081
T
able
5
(
b
).
T
he
an
al
ysi
s r
e
po
rt of the
d
e
nois
ing
ca
pacit
y o
f
D
BAM
F
unde
r
SP
N su
rro
unding
SPN
PNSR (d
B)
Operated i
mages
Noise densi
t
y
LR i
m
age
Noise supp
ressing
tech
nique
Median (3x
3)
Mean (3x3)
AMF
DBA MF
Girl
(256x256)
D=0.05
16.4490
32.4867
20.0454
37.5895
39.2037
D=0.10
13.6890
31.5583
17.2530
36.9197
35.9413
D=0.15
11.9287
27.6179
15.3515
34.818
32.0931
D=0.20
10.6567
25.5153
13.9593
32.0437
29.3961
D=0.25
9.5498
22.9614
12.7148
29.6074
26.069
D=0.30
8.8677
20.7738
11.9599
27.6930
23.2608
D=0.35
8.0984
18.4410
11.0501
34.9709
20.5962
D=0.40
7.5798
1635146
10.4543
23.3736
18.4867
D=0.45
7.0728
14.8145
9.8471
21.8119
16.9252
D=0.50
6.5712
13.0319
9.2367
20.1712
15.1631
D=0.55
6.2085
11.8226
8.7895
19.2184
13.6912
D=0.60
5.8609
10.4981
8.3590
18.4518
12.2929
D=0.65
5.4832
3.1396
7.8712
17.2740
10.9353
D=0.70
5.1311
8.0463
7.4271
16.7334
9.5595
D=0.75
4.8712
7.1994
7.0814
16.2921
8.703
D=0.80
4.5674
6.2520
6.6881
16.2795
7.5002
D=0.85
4.3054
5.4218
6.3340
16.5924
6.5222
D=0.90
4.0573
4.7465
5.9986
16.7463
5.2931
Airplane
(256x256)
D=0.05
17.9498
31.4106
21.5802
36.6063
36.5023
D=0.10
14.8320
29.6532
18.4426
34.6311
31.5421
D=0.15
13.1197
28.3176
16.6870
33.5561
29.9163
D=0.20
11.8045
26.4356
15.3181
31.3844
26.2166
D=0.25
10.9272
24.4147
14.3866
29.5029
25.0114
D=0.30
10.0510
21.8862
13.4526
27.1347
22.1793
D=0.35
9.4325
19.6835
12.7646
26.0118
20.5242
D=0.40
8.8735
17.6412
12.1397
23.0147
18.5963
D=0.45
8.3344
15.8686
11.5224
21.2768
17.0442
D=0.50
7.8600
14.2697
11.0091
19.6201
15.4295
D=0.55
7.4696
12.8823
10.5769
18.6408
13.9249
D=0.60
7.0920
11.5290
10.1202
17.6586
12.3281
D=0.65
6.7276
10.4080
9.7008
16.9400
10.9416
D=0.70
6.4028
9.3041
9.3238
16.2514
9.8385
D=0.75
6.1274
8.3797
9.0020
15.9223
8.6814
D=0.80
5.8647
7.5835
8.6893
15.7428
7.9942
D=0.85
5.5768
6.7043
8.3346
15.8098
6.9951
D=0.90
5.3335
6.0278
8.0381
16.0834
6.3296
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
4
,
A
ugus
t
2020
:
4109
-
4117
4116
(a)
(b)
Figure
4. The
a
naly
sis re
port
of noise
rep
ai
ri
ng m
et
ho
d of t
h
e
DBAMF
5.
CONCL
US
I
O
N
T
h
i
s
a
n
a
l
y
z
e
d
r
e
p
o
r
t
f
o
c
u
s
e
s
t
o
e
x
a
m
i
n
e
t
h
e
s
i
m
i
l
a
r
i
t
y
c
a
p
a
c
i
t
y
o
f
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
f
o
r
d
i
v
e
r
s
e
S
P
N
S
u
r
r
o
u
n
d
i
n
g
.
I
n
o
r
d
e
r
t
o
e
x
a
m
i
n
e
t
h
e
d
e
n
o
i
s
i
n
g
c
a
p
a
c
i
t
y
a
n
d
i
t
s
o
b
s
t
r
u
c
t
i
o
n
o
f
t
h
e
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
,
t
h
e
f
o
u
r
o
r
i
g
i
n
a
l
d
i
g
i
t
a
l
i
m
a
g
e
s
,
c
o
m
p
r
i
s
e
d
o
f
A
i
r
p
l
a
n
e
,
P
e
p
p
e
r
,
G
i
r
l
a
n
d
L
e
n
a
,
a
r
e
e
x
a
m
i
n
e
d
i
n
t
h
e
s
e
c
o
m
p
u
t
a
t
i
o
n
a
l
s
i
m
u
l
a
t
i
o
n
f
o
r
S
P
N
s
u
r
r
o
u
n
d
i
n
g
b
y
i
n
i
t
i
a
l
l
y
c
o
n
t
a
m
i
n
a
t
i
n
g
t
h
e
S
P
N
w
i
t
h
d
i
v
e
r
s
e
i
n
t
e
n
s
i
t
y
.
T
h
e
f
i
r
s
t
c
o
n
s
t
r
i
b
u
t
i
o
n
o
f
t
h
i
s
r
e
p
o
r
t
i
s
t
h
e
o
p
t
i
m
i
z
e
d
s
t
a
b
l
e
c
o
n
s
t
a
n
t
,
w
h
i
c
h
i
s
d
e
t
e
r
m
i
n
e
d
f
r
o
m
c
o
m
p
u
t
e
r
s
i
m
u
l
a
t
i
o
n
a
t
S
P
N
s
u
r
r
o
u
n
d
i
n
g
.
L
a
t
e
r
,
t
h
e
s
e
c
o
n
d
c
o
n
s
t
r
i
b
u
t
i
o
n
o
f
t
h
i
s
r
e
p
o
r
t
i
s
t
h
e
o
v
e
r
a
l
l
c
a
p
a
c
i
t
y
o
f
d
e
n
o
i
s
i
n
g
m
e
t
h
o
d
f
o
u
n
d
o
n
D
B
A
M
F
,
c
o
n
f
r
o
n
t
e
d
w
i
t
h
t
h
e
o
t
h
e
r
d
e
n
o
i
s
e
d
i
m
a
g
e
s
(
S
M
F
a
n
d
g
a
u
s
s
i
a
n
f
i
l
t
e
r
a
n
d
a
d
a
p
t
i
v
e
m
e
d
i
a
n
f
i
l
t
e
r
(
A
M
F
)
)
u
n
d
e
r
t
h
e
S
P
N
w
i
t
h
d
i
v
e
r
s
e
i
n
t
e
n
s
i
t
y
.
ACKN
OWLE
DGE
MENTS
The researc
h p
roject
was fu
nded
b
y
Assum
ption U
niv
e
rsity
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Compu
t
atio
na
l
S
cr
utiny
of im
ag
e
d
e
noisi
ng
meth
od fo
und on
…
(
V
orapoj
Patan
aviji
t
)
4117
REFERE
NCE
S
[1]
A.
Aw
ad,
“
Removal
of
fixe
d
-
val
ued
impuls
e
noise
base
d
on
proba
bil
i
t
y
o
f
exi
stence
of
t
he
image
pixel,
”
Inte
rnational
Jo
urnal
of El
e
ct
ri
c
al
&
Co
mputer
Engi
ne
ering,
vol
.
8
,
no
.
4
,
pp
.
21
06
-
2114,
2018
.
[2]
H.
Hw
ang
and
R.
A.
Hadda
d,
“
Adapti
ve
m
edi
a
n
fil
te
rs
new
a
lg
orit
hm
s
and
resu
lt
s,”
IE
EE
Tr
ans
act
ions
of
Imag
e
Proce
ss
ing,
vo
l.
4,
pp
.
4
,
pp
.
499
-
502,
1994
.
[3]
Kornkam
ol
Thakulsukana
nt
and
Vorapoj
Pata
n
a
vij
it,
“
Sim
ula
ti
o
n
eva
lu
at
ion
of
noise
suppress
ing
te
chn
ique
b
ase
d
on
dec
ision
bas
ed
ada
p
ti
ve
m
e
dia
n
filter
for
s
al
t
&
p
eppe
r
no
ise,
”
16
th
Inte
r
nati
onal
Conf
ere
nce
on
El
e
ct
ric
al
Engi
ne
ering/
E
lectronic
s,
Computer,
Te
lecomm
unic
ati
ons
and
Info
rm
ati
on
Technology,
pp
.
585
-
588
,
2019
.
[4]
K.
Arun
Sai,
K.
Ravi
,
”An
eff
i
ci
en
t
filter
ing
technique
for
den
oising
col
our
i
m
age
s,”
Int
erna
ti
onal
Journal
o
f
El
e
ct
rica
l
and
C
omputer
Engi
n
e
ering
(
IJE
CE)
,
v
ol.
8
,
no
.
5
,
pp
.
3
604
-
3608,
2018
.
[5]
R.
H.
Chan
,
C
-
W
Ho
and
M.
Nikil
ova
,
“
Salt
&
peppe
r
no
ise
removal
b
y
m
ed
ia
n
-
t
y
p
e
noise
d
et
e
ct
ors
and
d
etail
-
pre
serving
r
egulari
z
at
ion
,
”
IE
EE Tr
ansacti
ons of
I
mage
Proc
essing,
vol
.
14
,
no
.
10
,
pp
.
1479
-
1485
,
2005.
[6]
V
.
P
a
t
a
n
a
v
i
j
i
t
,
“
T
h
e
b
i
l
a
t
e
r
a
l
d
e
n
o
i
s
i
n
g
p
e
r
f
o
r
m
a
n
c
e
i
n
f
l
u
e
n
c
e
o
f
w
i
n
d
o
w
,
”
S
p
a
t
i
a
l
a
n
d
R
a
d
i
o
m
e
t
r
i
c
V
a
r
i
a
n
c
e
,
2
0
1
5
.
[7]
V.
Pata
navijit,
“
Perform
anc
e
an
aly
s
is
of
denoi
si
ng
al
gorit
hm
ba
sed
on
ada
pti
v
e
m
edi
an
fil
t
er
un
der
uns
y
stemat
i
c
int
ensity
impuls
e
and
sal
t
&
pep
per
noise
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
E
l
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
C
o
n
g
r
e
s
s
,
pp.
1
-
4
,
2018.
[8]
W
.
K.
Prat
t, “M
edi
an
filter
ing,
”
Tech.
Re
p
.
,
Image
Proc
.
Inst
.
,
Un
iv.
South
ern
C
al
i
fornia
,
Los
Ange
le
s,
1975
.
[9]
Yiqiu
Dong,
R
a
y
m
ond
H.
Ch
an,
and
Shufan
g
Xu,
“
A
det
e
c
ti
on
sta
ti
sti
c
for
ran
dom
-
val
ued
impuls
e
noise
,
”
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
.
4
.
p
p
.
1
1
1
2
-
1
1
2
0
,
2007
[10]
Om
Praka
sh
Verm
a
and
Niti
n
Sharm
a,
“
Inte
nsi
t
y
pr
ese
rving
ca
s
t
removal
in
co
l
or
images
using
par
ticl
e
sw
arm
opti
m
iz
ation,
”
I
nte
rnational
Jou
rnal
of El
e
ct
ri
ca
l
&
Co
mputer
E
ngine
ering
,
v
ol
.
7,
n
o
.
5
,
pp
.
258
1
-
2595
,
2017
.
[11]
Madina
Ham
ia
n
e
and
Fat
ema
Saee
d
,
“
SV
M
cl
assific
at
ion
of
M
RI
bra
in
images
for
computer
-
a
ss
iste
d
dia
gnosis,
Inte
rnatio
nal
Jo
urnal
of El
e
ct
ri
c
al
and
Comput
er
Engi
n
ee
ring
,
v
o
l.
7
,
n
o
.
5
,
pp
.
25
55
-
2564
,
2017
.
[12]
A.
S.
M.
Shafi,
et
.
al
.
,
“
Dec
om
positi
on
of
co
lo
r
wave
let
with
highe
r
orde
r
st
a
ti
stical
te
xtur
e
a
nd
convol
uti
on
a
l
neur
al
n
et
work
f
ea
tur
es
set
base
d
cl
assifi
ca
t
ion
of
col
ore
ctal
pol
y
p
s
from
vide
o
en
doscop
y
,
”
In
te
rn
ati
onal
Journal
of
E
le
c
tric
al
and
Computer
Eng
i
nee
ring
,
v
ol
.
10
,
n
o.
3,
pp.
2986
-
2996
,
2020
.
[13]
S.
Bagc
hi,
et.al,
“
Im
age
proc
essing
and
m
ac
hine
le
arn
ing
te
chn
iq
ues
used
in
com
pute
r
-
a
ide
d
dete
ct
ion
s
y
st
em
fo
r
m
am
m
ogra
m
sc
ree
ning
-
A
rev
iew
,
”
Inte
rnat
iona
l
Journal
of
El
e
ct
rical
and
Computer
Engi
n
ee
ri
ng
,
v
ol.
10
,
n
o
.
3,
pp.
2336
-
2348
,
2020.
[14]
C
é
s
a
r
G
.
P
a
c
h
ó
n
-
S
u
e
s
c
ú
n
,
C
a
r
l
o
s
J
.
E
n
c
i
s
o
-
A
r
a
g
ó
n
,
R
o
b
i
n
s
o
n
J
i
m
é
n
e
z
-
M
o
r
e
n
o
,
“
R
o
b
o
t
i
c
n
a
v
i
g
a
t
i
o
n
a
l
g
o
r
i
t
h
m
w
i
t
h
m
a
c
h
i
n
e
v
i
s
i
o
n
,
”
Int
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
E
l
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
0
,
n
o
.
2
,
p
p
.
1
3
0
8
-
1
3
1
6
,
2
0
2
0
.
[15]
D.
Kesrara
t,
et
.
a
l,
“
A
novel
el
e
m
ent
ar
y
spa
ti
a
l
expa
nding
sche
m
e
form
on
sisr
m
et
hod
with
modi
f
y
ing
geman
&
m
cc
lure
func
t
ion
,
”
TEL
KOMNIK
A
Tel
ec
omm
unication, Com
puti
n
g,
E
lectronic
s an
d
Control,
v
ol
.
1
7,
n
o
.
5,
2019.
[16]
V.
Pata
nav
ij
i
t,
“
Denoising
per
fo
rm
anc
e
an
aly
s
is
of
ada
pt
ive
de
cis
ion
base
d
inve
r
se
dista
nc
e
weig
hte
d
interpol
at
io
n
(DBID
W
I)
al
gorit
hm
for
sa
lt
a
nd
peppe
r
noise
,
”
Indon
esian
J
ournal
of
El
e
ct
r
ic
al
Engi
n
ee
rin
g
and
Compute
r
Sci
en
ce,
vol
.
15
,
no.
2,
pp.
804
-
8
13,
2019
.
[17]
V.
Kishoreba
bu
1,
e
t.
a
l.,
“
An
ad
apt
iv
e
de
ci
sion
base
d
interpol
at
i
on
sche
m
e
for
th
e
removal
of
hig
h
density
sa
lt
an
d
peppe
r
no
ise in
i
m
age
s,
”
EURA
S
IP
Journal
on
I
mage
and
V
ide
o
Proce
ss
ing,
vo
l.
2017,
no
.
1
,
201
7.
[18]
V.
Pata
nav
ij
i
t,
e
t.
a
l,
“
The
st
at
ist
ic
a
l
anal
y
s
is
of
ran
dom
-
val
ued
i
m
pulse
noise
de
te
c
ti
on
t
ec
hniqu
es
base
d
on
th
e
loc
a
l
image
ch
ara
c
te
rist
ic
:
ROA
D,
ROLD
and
RORD
,
”
Ind
onesian
Journal
of
E
le
c
tric
al
Engi
ne
ering
an
d
Computer
Scien
ce
,
vol
.
15
,
no
.
1
,
pp
.
749
-
803,
20
19.
[19]
Y.
Dong,
et
.
al
,
“
A
det
ec
t
ion
stat
i
stic
for
ran
dom
-
val
ued
impuls
e
noise,
”
IE
EE
Tr
ansacti
ons
on
Image
Proce
ss
ing,
vol.
16
,
no
.
4
,
pp
.
1112
-
1120
,
20
07.
[20]
H.
Yu,
e
t.al
,
“
An
eff
icient
pr
oce
dure
for
r
e
m
oving
ran
dom
-
val
ued
impuls
e
noise
in
imag
es,
”
I
EE
E
Sign
al
Proce
ss
ing
Let
ter
s,
vol. 15, pp. 9
22
-
925,
2008
.
[21]
V.
Ja
y
araj
and
D.
Ebe
n
ez
er
,
“
A
New
Sw
it
chi
ng
-
BasedMe
dian
Filt
eri
ng
Sche
m
e
and
Algorithm
for
Removal
o
f
High
-
Densit
y
Sa
lt
and
Pepp
er
No
ise
in
Im
age
s,
”
EURA
SIP
J
ASP
,
2010.
[22]
Ali
S.
Aw
ad
,
“
Standa
rd
Devi
at
io
n
for
Obtai
ning
the
Optimal
Dire
ction
in
the
Removal
of
Im
pul
s
e
Noise,
”
IE
E
E
Signal
Let
te
rs
,
2
011.
[23]
V.
Pata
nav
ij
i
t
and
K.
Tha
ku
l
sukana
nt,
“
Sim
ula
t
ed
Evaluati
on
of
New
Swi
tc
hing
B
ase
d
Media
n
Filt
er
f
or
Suppress
ing
SPN
and
RVIN
,
”
Indone
sian
Journal
of
Elec
tri
cal
Engi
ne
ering
and
Computer
Sci
en
ce
,
Vol
.
15,
No.
2
,
Aug.
2019.
[24]
R.
C.
Gonza
lez a
nd
R.
E. Woods
,
“
Digit
al Im
age
Proce
ss
ing,
”
Pre
nti
c
e
Hall
,
2nd
e
dit
ion,
NJ,
USA,
2002.
[25]
V.
R.
Vijay
kum
ar
and
P.
Joth
ib
asu,
“
Dec
ision
Based
Adatpv
iv
e
Medi
an
Fil
te
r
to
R
emove
Blo
tc
hes,
Scra
tc
hes
,
Strea
ks,
Strip
es
and
Im
pulse
Noise
in
Im
age
s,
”
2010
IEE
E
In
te
r
nati
onal
Conf
ere
nce
on
Image
Proce
ss
ing,
Hon
g
Kong,
pp.
117
-
1
20
,
Sept
ember
2
010.
Evaluation Warning : The document was created with Spire.PDF for Python.