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.
8
, No
.
6
,
Decem
ber
201
8
, p
p.
4253
~
4257
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
8
i
6
.
pp
4253
-
42
57
4253
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Low C
om
plexit
y F
luctua
tion
M
easurem
ent in
Image
P
ro
cessing
Conside
ring Ord
er
Tareq
Kh
an
School
of Engin
ee
ring
T
ec
hnolo
g
y
,
E
aste
rn
Mic
higa
n
Univ
ersity
,
US
A
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Feb
5
, 2
01
8
Re
vised
Ju
l
8
,
201
8
Accepte
d
J
ul
28
, 2
01
8
The
stand
ard
d
e
via
ti
on
ca
n
m
eas
ure
the
spre
ad
out
of
a
s
et
of
n
um
ber
s
and
ent
rop
y
c
an
m
ea
sure
the
ran
do
m
ness.
How
eve
r,
they
do
not
conside
r
th
e
orde
r
of
the
num
ber
s.
Thi
s
ca
n
le
ad
to
m
isle
ad
i
ng
result
s
where
the
orde
r
of
the
num
ber
s
is
vit
al.
An
image
is
a
se
t
of
num
ber
s
(i.e.
pix
el
v
a
lue
s)
th
at
is
sensiti
ve
to
ord
e
r.
In
thi
s
pap
er,
a
low
complexit
y
and
eff
i
ci
en
t
m
et
hod
for
m
ea
suring
the
fl
uct
ua
ti
on
is
proposed
conside
ri
ng
the
orde
r
of
the
num
ber
s.
The
proposed
m
et
hod
sum
s
up
the
cha
nges
of
c
onsec
uti
v
e
num
b
ers
and
ca
n
be
used
in
image
proc
essing
appl
icat
ions.
Sim
ula
ti
on
show
s
tha
t
th
e
proposed
m
et
ho
d
is 8 to
33
ti
m
e
s fa
ster
tha
n
oth
e
r
relat
ed
works
.
Ke
yw
or
d:
Entr
op
y
Fluctuati
on
Im
age
p
r
ocessi
ng
Stand
a
r
d
d
evia
ti
on
Var
ia
nce
Copyright
©
201
8
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
:
Tareq K
ha
n,
School
of E
ng
i
neer
i
ng Tec
hnology,
East
ern Mi
chig
an Un
i
ver
sit
y,
118 Si
ll
H
al
l,
Yp
sil
a
nti, MI
4819
7,
USA.
Em
a
il
: t
areq
.khan@em
ic
h.
ed
u
1.
INTROD
U
CTION
Stand
a
r
d
dev
ia
ti
on
an
d
var
ia
nce
[1
]
m
easur
e
how
fa
r
a
set
of
nu
m
ber
s
are
sp
rea
d
ou
t
fr
om
their
aver
a
ge
value
and
they
giv
e
an
in
dicat
io
n
of
the
fl
uctuati
on
or
cha
otic
na
ture
of
the
nu
m
ber
s.
For
i
nst
ance,
le
t’s
con
si
der
a
n
arr
ay
of
num
ber
s
,
S
1
=
[2
,
2,
2,
2,
9,
9,
9,
9]
.
The
sta
nd
a
rd
dev
ia
ti
on
of
the
el
e
m
ents
of
S
1
is
3.74
a
nd
the v
a
riance is
14
. N
ow, let
’s
reo
r
de
r
the n
um
ber
s
o
f
S
1
and
let
’s
co
ns
ide
r
S
2
= [
2
,
9
,
2
,
9
,
2
,
9
,
2
,
9
]
.
Fo
r
S
2
,
the
st
and
a
r
d
de
viati
on
is
3.7
4
a
nd
the
va
riance
is
14
.
H
ere,
we
see
that
sta
nd
a
r
d
de
viati
on
a
nd
var
ia
nce
do
no
t
change
eve
n
if
the
num
ber
s
are
re
order
e
d.
The
num
ber
s
in
S
2
has
m
or
e
changes
with
r
espect
to
it
s
nex
t
num
ber
than
in
S
1
,
an
d
S
2
fl
uc
tuate
s
m
or
e
than
S
1
.
H
owe
ver,
sta
ndar
d
dev
ia
ti
on
a
nd
var
ia
nce
cannot
ac
know
le
dg
e t
his
diff
e
ren
ce
.
Let
’s
c
onsider
a
2D
case
o
f
num
ber
s.
An
im
age
is
a
2
D
ar
r
ay
of
num
ber
s i.e. p
ixel value
s
[
2
]
.
Fi
gure
1a
a
nd
Fig
ure
1
b
s
hows
t
wo
arti
fici
al
i
m
ages
w
her
e
they
ha
ve
the
sam
e
nu
m
ber
of
wh
it
e
an
d
black
pi
xels.
In
an
8
bit
gray
scal
e
i
m
age,
the
wh
it
e
pix
el
ha
s
the
value
of
255
an
d
the
bl
ack
pix
el
ha
s
the
value
of
0.
Fig.
2a
and
Fi
g.
2b
shows
tw
o
nat
ural
gr
ay
scal
e
im
ages
hav
i
ng
diff
e
re
nt
te
xtu
r
es.
The
sta
nda
rd
de
viati
on,
va
riance
,
and
ent
ropy
of
the
im
age
pixe
ls
in
Fig
ure
1
an
d
in
Fig
ure
2
a
re
s
how
n
i
n
Ta
ble
1
an
d
in
Ta
ble
2
na
m
el
y.
Fr
om
Tab
le
1
, we see t
hat the
stand
a
r
d dev
ia
ti
on
, va
riance,
and ent
ropy [3
]
-
[5
]
of
Fig
ur
e
1a
a
nd Fig
ure
1b are
exactl
y
the
sam
e.
Ho
we
ve
r,
intuit
ively
Fi
gure
1b
is
m
o
re
r
an
dom
or
unpredict
able
than
Fi
gure
1a
.
Fr
om
Table
2
,
we
se
e
that
the
sta
ndar
d
de
viati
on
,
va
riance,
a
nd
entr
op
y
of
Fi
g
ur
e
2a
a
nd
Fig
ur
e
2b
a
re
ve
r
y
cl
os
e
even
t
hough
t
he
y
hav
e
differ
ent
te
xtures.
F
or
i
ns
ta
nce,
Fi
g
ure
3a
an
d
Fi
g
ure
3b
s
how
s
the
cha
nges
of
pi
xel
values
with
res
pect
to
it
s
adj
a
cent
le
ft
pi
xel
value
f
or
row
127
f
or
Fi
gure
2a
an
d
Fig
ure
2b
nam
el
y.
He
re,
we
see that
Fig
ure
2b h
as
m
or
e fl
uctuati
ons
of pi
xel v
al
ue
s tha
n
Fi
g
ure
2a.
Q
uan
ti
ti
vely
,
av
erag
e
ab
s
olu
te
chang
e
for
Fi
gure
3a
is
9.3
4
a
nd Fig
ur
e
3b is
22.
49.
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4253
-
4257
4254
The
sta
ndar
d
st
at
ist
ic
al
m
easur
es
su
c
h
as
sta
nd
a
r
d
de
viati
on
,
va
riance
,
an
d
entr
opy
cannot
recog
nize
the
diff
e
ren
c
e
betwee
n
t
he
i
m
age
pair
s
in
Figure
1
an
d
F
igure
2
beca
use
they
do
no
t
c
on
si
der
the
or
de
r
of
the
num
ber
s. In
a
n im
age,
the orde
r
of the
nu
m
ber
s (
i.e
. p
i
xel val
ues)
a
re s
ig
nificant b
eca
us
e
changin
g
thei
r orde
r
will
change
t
he
im
age.
Th
us
if
t
hese
st
and
a
r
d
sta
ti
sti
cal
m
et
ho
ds
a
re
use
d
t
o
fi
nd
the
ra
ndomness
or
fluctuati
on
of
t
he
te
xt
ur
e
of
a
n
im
age,
they
m
ay
le
ad
to
m
i
sle
adin
g
resu
lt
s.
I
n
t
his
pap
e
r
,
an
ef
fici
ent
s
olu
ti
on
of
this
pro
blem
is
pro
po
se
d.
The
pr
opos
e
d
m
et
ho
d
s
um
s
up
t
he
c
hang
es
of
c
onsecut
ive
num
ber
s
a
nd
c
a
n
recog
nize
the
diff
e
re
nt
te
xture
of
im
ages
even
th
ough
t
hey
ha
ve
sam
e
sta
ndar
d
de
viati
on
,
var
ia
nc
e,
a
nd
entr
op
y.
T
he
pro
posed
m
et
hod
can
be
us
e
d
to
ge
ner
at
e
a
f
eat
ur
e
vect
or
i
n
cl
assify
ing
t
extu
res
of
sto
ne
[6
]
,
woo
d
[
7],
batik
m
otif [
8] etc
.
(a)
(b)
Fig
ure
1.
A
rtific
ia
l im
ages h
a
ving the
sam
e n
um
ber
of whi
te
an
d blac
k pi
xels
. (a)
Pixels
d
ist
rib
uted
in
an
order
ly
fash
i
on w
it
h fe
wer
flu
ct
uations
;
(
b)
P
ixels dist
ri
bu
te
d ran
dom
l
y hav
in
g
m
or
e fl
uc
tuati
on
s
(a)
(b)
Fig
ure
2. Nat
ural
i
m
ages h
a
vin
g di
ff
e
ren
t t
e
xtures.
(a) Pixe
ls distrib
uted
havin
g fewe
r flu
ct
uations
;
(
b)
P
ixels
distrib
uted ha
vi
ng
m
or
e
fluct
uations
Table
1.
Stan
da
rd Stat
ist
ic
al
Mea
su
rem
ent o
f
Fig
ur
e
1
Fig
u
re
Stan
d
ard Deviatio
n
Variance
Entro
p
y
Fig
u
re
1a
1
2
7
.50
1
6
2
5
6
.25
1
Fig
u
re
1b
1
2
7
.50
1
6
2
5
6
.25
1
Table
2.
Stan
da
rd Stat
ist
ic
al
Mea
su
rem
ent o
f
Fig
ur
e
2
Fig
u
re
Stan
d
ard Deviatio
n
Variance
Entro
p
y
Fig
u
re
2a
3
7
.73
7
5
1
4
2
4
.1
0
7
.17
9
4
Fig
u
re
2b
3
7
.73
9
7
1
4
2
4
.3
0
7
.24
7
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
Lo
w C
omplexi
ty
Fluctu
atio
n Me
asurem
ent i
n Image
Pr
oce
ssing Co
ns
ide
r
ing
Or
der
(
T
are
q
K
han)
4255
(a)
(b)
Fig
ure
3
.
Cha
nges
of the
pix
el
v
al
ue
at r
ow
127 wit
h res
pect
to
it
s left
pix
el
.
(a
)
C
hanges
of
pix
el
values
for
Fig
ure
2a.
;
(b) C
han
ges o
f pix
el
v
al
ues
for Fi
g
ure
2b
2.
THE
PROPO
SED
METHO
D
The
pro
pose
d
fluctuati
on
m
e
asur
em
ent
m
eth
od,
f
1D
,
f
or
th
e
1D
ar
ray
is
expresse
d
in
(
1).
Her
e
x
i
is
the
i
th
el
e
m
ent
of
the
a
rr
ay
a
nd
N
is
the
siz
e
of
the
a
rr
ay
.
The
num
erator
co
ntains
the
su
m
of
the
squar
e
d
changes
of c
onsecuti
ve n
um
ber
s a
nd the
de
nom
inator
co
nta
ins the
total
nu
m
ber
o
f
c
ha
nges.
1
2
1
1
1
1
N
ii
i
D
xx
f
N
(1)
Fo
r
2D
case
,
s
uch
as
a
n
im
a
ge,
t
he
pr
opose
d
fl
uctuati
on
m
easur
em
ent
m
et
ho
d,
f
2D
,
is
ex
pr
es
sed
i
n
(2),
(3),
an
d
(4).
Her
e
(x
i
,
y
j
)
i
s
the
num
ber
at
the
i
th
col
um
n
an
d
the
j
th
r
ow.
X
a
nd
Y
a
re
t
he
total
c
olu
m
ns
a
nd
total
rows
nam
el
y.
Her
e
i
n
(
2),
f
h
re
pr
es
ents
the
fluct
uatio
n
in
the
horizo
nt
al
directi
on
an
d
f
v
in
(3)
re
pr
esents
the
fluctuati
on in
the
ver
ti
cal
directi
on.
Final
ly
,
f
2D
,
in
(4
)
is
the
m
ean
of
f
h
and
f
v
m
ulti
plied
by
a
facto
r
n
.
For
instance,
n
co
ul
d
be
the m
axim
u
m
p
ixel value,
i.e
. 255
f
or
an
8
-
bit
im
age.
1
2
1
11
,,
1
YX
i
j
i
j
ji
h
x
y
x
y
f
XY
(2)
1
2
1
11
,,
1
XY
i
j
i
j
ij
v
x
y
x
y
f
XY
3)
2
2
hv
D
ff
fn
(4)
3.
RESU
LT
S
Using
(
1)
,
t
he
fluctuati
on
of
S
1
is
7
and
the
fluctuati
on
of
S
2
is
49
Th
us
,
the
propose
d
m
et
ho
d
ca
n
ackno
wled
ge
the
existe
nce
of
the
fr
e
qu
e
nt
changes
in
S
2
and
giv
e
hi
gh
e
r
fluctuati
on
than
S
1
,
eve
n
thou
gh
bo
t
h
S
1
a
nd
S
2
ha
ve
the
sam
e
el
e
m
ents.
Table
3
an
d
Ta
bl
e
4
show
the
fluctua
ti
on
f
or
the
im
age
pairs
in
Fig
ure
1
a
nd
F
ig
ure
2
us
in
g
t
he
pro
pose
d
m
et
hod
nam
el
y.
Table
3
s
hows
that
the
fluct
ua
ti
on
f
or
Fig
ur
e
1b
is
m
uch
la
rg
er
t
ha
n
Fig
ure
1a
.
Fr
om
Table
4,
we
see
t
hat
the
fl
uctuati
ons
of
Fig
ure
2b
is
m
uch
la
r
ger
than
Fig
ure
2a.
Th
us
,
t
he
pro
po
s
ed
m
et
ho
d
ca
n
rec
ognize
t
he
diff
e
re
nt
am
ou
nt
of
fl
uc
tuati
on
s
bet
we
en
th
e
arti
fici
al
i
m
age
pairs
in
Fi
g
ure
1
a
nd
betw
een
the
nat
ur
al
i
m
age
pairs
in
Fig
ure
2.
On
the
oth
e
r
ha
nd
-
th
e
sta
nd
a
rd
de
via
ti
on
,
var
ia
nce,
an
d
e
ntropy
are
una
ble
to
ap
pr
eci
at
e
th
e
dif
fer
e
nt
am
ount
of
fluct
ua
ti
on
s
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4253
-
4257
4256
betwee
n
t
he
im
age p
ai
rs
in Fig
ure
1
a
s
s
how
n
in
Ta
ble 1
a
nd b
et
wee
n
t
he
i
m
age
pair
s
in
Fi
gure
2
a
s
s
ho
wn
in
Table
2.
Table
3.
Fluctu
at
ion
m
easur
e
m
ent o
f
Fig
ur
e
1
us
i
ng prop
ose
d
m
et
ho
d
Fig
u
re
Ho
rizon
tal f
lu
ctu
atio
n
(
f
h
)
Vertica
l
f
lu
ctu
atio
n
(
f
v
)
Flu
ctu
atio
n
(
f
2D
)
Fig
u
re
1a
0
1
1
2
7
.50
Fig
u
re
1b
4
.87
5
.80
1
3
6
0
.3
3
Table
4.
Fluctu
at
ion
m
easur
e
m
ent o
f
Fig
ur
e
2
us
i
ng prop
ose
d
m
et
ho
d
Fig
u
re
Ho
rizon
tal f
lu
ctu
atio
n
(
f
h
)
Vertica
l
f
lu
ctu
atio
n
(
f
v
)
Flu
ctu
atio
n
(
f
2D
)
Fig
u
re
2a
4
1
.05
3
3
.39
9
4
9
1
.2
4
Fig
u
re
2b
7
7
.68
7
6
.89
1
9
7
0
7
.70
On
e e
xisti
ng m
et
hod,
m
entione
d
in [9]
-
[
11]
, fo
r
fin
ding the fl
uctuati
on
of
a
n
im
age co
uld
be
to
us
e a
local
sta
nd
ar
d
dev
ia
ti
on
filt
er
.
This
m
e
tho
d
returns
a
2D
ar
ray
wh
e
re
each
ou
tp
ut
pix
el
c
on
ta
in
s
the
sta
nd
a
r
d
dev
ia
ti
on
of
th
e
3
-
by
-
3
nei
ghbor
hood
ar
ound
the
c
orres
pond
i
ng
pix
el
in
the
input
im
age.
Sym
m
et
ric
pad
di
ng
is
us
ed
f
or
the
pix
el
s
in
the
bor
der.
Taki
ng
an
ave
rag
e
of
t
he
retu
rn
e
d
2D
arr
ay
can
in
dicat
e
the
fluctua
ti
on
of
the
im
age.
As
this
m
et
ho
d
use
s
3×
3
neig
hb
or
i
ng
pi
xels
to
cal
culat
e
the
s
ta
nd
a
rd
de
viati
on
of
eac
h
pixe
l,
the
pix
el
posit
ion
s
an
d
orders
ar
e
co
ns
ide
red
in
this
m
et
ho
d.
L
ocal
ent
ropy
filt
er
can
al
s
o
i
m
ple
m
ent
in
th
e
sam
e
way
w
her
e
e
ntropy
is
cal
culat
ed
instea
d
of
sta
nd
a
rd
de
viati
on
.
I
n
Ta
ble
5,
the
fluct
uation
is
m
easur
e
d
in
Fig
ure
2a
a
nd
Fig
ure
2b
usi
ng
local
sta
ndar
d
de
viati
on
filt
ering
,
loc
al
entropy
filt
erin
g
an
d
us
i
ng
th
e
pro
po
se
d
m
et
ho
d. He
re, we s
ee that
all
o
f
th
ese m
e
tho
ds
c
orrectl
y r
ep
o
rts
higher fluct
ua
ti
on
i
n
Fig
ure
2b tha
n
in Fig
ure
2a.
On
e
s
hortc
om
i
ng
of
t
he
loca
l
sta
nd
ar
d
devi
at
ion
filt
erin
g
and
l
ocal
entr
op
y
filt
erin
g
m
et
ho
ds
a
re
their
tim
e
co
m
plexity
.
I
n
thes
e
m
et
ho
ds
,
t
he
sta
nd
a
rd
d
e
viati
on
(
or
entr
opy)
of
9
pix
el
s
ne
eds
to b
e
cal
cu
la
te
d
for
each
pix
el
i
n
the
im
age.
H
ow
e
ve
r,
in
the
pro
po
se
d
m
et
ho
d,
only
a
difference
betwee
n
tw
o
pi
xels
ne
eds
to
be
cal
culat
ed
.
The
local
sta
ndar
d
dev
ia
ti
on
filt
ering
,
l
ocal
entr
op
y
filt
eri
ng,
an
d
the
pr
opos
e
d
m
et
ho
d
we
re
i
m
ple
m
ented
(
without
im
ple
m
enting
vecto
rizat
ion
-
f
or
pro
per
c
om
par
ison)
in
a
c
om
pu
te
r
hav
i
ng
In
te
l
Pentium
CPU
2117U
r
unni
ng
at
1.8GHz.
The
e
xec
ution
tim
es
are
sho
wn
in
Table
6.
He
re,
we
se
e
that
the
pro
po
se
d
m
et
ho
d
is
the
faste
st
hav
i
ng
the
l
ow
est
e
xec
utio
n
ti
m
e.
The
pr
opos
e
d
m
et
h
od
is
a
ppr
ox
im
a
te
ly
8
tim
es
faster
than
local
sta
nd
ard
dev
ia
ti
on
filt
ering
a
nd
a
ppr
ox
im
at
ely
33
ti
m
es
faste
r
than
l
ocal
entr
op
y
filt
ering
m
et
ho
d.
F
ourier
s
pe
ct
ra
m
et
ho
ds
[
2]
,
[12]
can
be
us
ed
to
m
easur
e
fl
uctuati
on
,
howe
ver,
the
y
hav
e
higher
co
m
pu
t
at
ion
al
c
om
ple
xity
than
[
9] an
d
the
pr
opos
e
d m
et
ho
d
-
a
nd
will
r
eq
uire
m
or
e
ex
ec
utio
n
t
i
m
e.
Table
5.
Fluctu
at
ion
m
easur
e
m
ent o
f
Fig
ure
2
Fig
u
re
Local Stan
d
ard D
e
v
iatio
n
Filter
in
g
Local Entrop
y
Filt
ering
Prop
o
sed
M
eth
o
d
Fig
u
re
2a
1
0
.71
2
.82
9
4
9
1
.2
4
Fig
u
re
2b
2
4
.58
3
.03
1
9
7
0
7
.70
Table
6.
C
om
par
iso
n of exec
ut
ion
ti
m
e
(in
S
econds
)
Fig
u
re
Local Stan
d
ard D
e
v
iatio
n
Filter
in
g
Local Entrop
y
Filt
ering
Prop
o
sed
M
eth
o
d
Fig
u
re
2a
3
.39
1
3
.50
0
.39
Fig
u
re
2b
3
.38
1
3
.56
0
.37
4.
CONCL
US
I
O
N
Entr
op
y,
sta
nd
ard
de
viati
on,
and
va
riance
m
ay
le
ad
to
m
i
sle
adin
g
re
su
lt
s
if
they
a
re
use
d
to
m
easur
e
rand
om
ness
or
fluctuati
on.
I
n
this
pap
e
r,
a
low
com
plexity
and
fast
f
luctuat
io
n
m
ea
su
rem
ent
m
et
h
od
is
pro
po
se
d
c
on
si
der
i
ng
the
orde
r
of
the
num
ber
s
f
or
bo
t
h
1D
and
2D
cases.
The
pr
opos
e
d
m
et
ho
d
ca
n
be
us
ed
in im
age p
r
oce
ssing ap
plica
ti
on
s
s
uch as cla
ssifyi
ng textu
r
e.
REFERE
NCE
S
[1]
R.
W
it
t
e
and
J.
W
it
te
,
“
Stat
ist
ic
s
,
”
W
i
ley
,
2009.
[2]
W
.
K.
Prat
t
,
“
Digit
al Im
age Proc
essing
,”
W
i
ley
-
I
nte
rsci
enc
e
,
200
7.
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
Lo
w C
omplexi
ty
Fluctu
atio
n Me
asurem
ent i
n Image
Pr
oce
ssing Co
ns
ide
r
ing
Or
der
(
T
are
q
K
han)
4257
[3]
T.
Kva
lset
h
,
“
On
the
m
e
asure
m
ent
of
ran
dom
ne
ss
(unc
ert
a
in
t
y
):
a
m
ore
informa
ti
ve
ent
rop
y
,
”
E
ntropy
,
vo
l/
issue
:
18(5),
pp
.
159
,
2
016.
[4]
C.
Shannon,
“
A
m
at
hemati
c
al
th
eor
y
of
comm
unic
a
ti
on,
”
B
el
l
S
y
stem
Techni
ca
l
J
ournal
,
vol/is
sue:
27(4),
pp
.
623
-
656
,
1948
.
[5]
Y.
W
ua,
e
t
a
l.
,
“
Shannon
ent
rop
y
b
ase
d
ran
do
m
ness
m
ea
surem
en
t
and
t
est
f
or
image
en
cr
y
p
ti
on
,
”
Informati
on
Sci
en
ce
s,
El
se
vier
,
vol
.
2016
,
pp
.
1
–
23
,
2011
.
[6]
P.
V.
Kum
ar,
et
al.
,
“
Stone
Im
age
Cla
ss
ifi
c
at
ion
Based
o
n
Overl
a
pped
5
-
bit
T
-
Pat
te
rns
O
cc
urre
n
c
e
on
5
-
by
-
5
Sub
Im
ag
es,
”
Int
ernati
onal
Journal
of
El
e
ct
ri
cal
an
d
Computer
Engi
nee
ring
(
IJE
CE
)
,
vol
/i
ss
ue:
6
(
3
)
,
pp.
1152
-
116
0
,
2016.
[7]
A.
Fahruroz
i
,
et
al.
,
“
W
ood
Cla
ss
ifi
cation
Based on E
dge
Det
ec
t
i
ons
and
Te
xtur
e F
ea
ture
s Sel
ec
t
i
on,
”
Int
ernati
on
al
Journal
of
Elec
t
rical
and
Computer
Eng
ine
ering
(
IJE
CE)
,
vol
/i
ss
ue:
6
(
5
)
,
pp
.
216
7
-
2175,
2016
.
[8]
I.
Nurhai
da
,
e
t
al.
,
“
T
ext
ure
Fu
sion
for
Bat
ik
Motif
Ret
ri
eval
S
y
stem,
”
In
te
rn
ati
onal
Journal
of
El
e
ct
ri
cal
an
d
Computer
Engi
n
ee
ring (
IJE
C
E)
,
vol
/i
ss
ue:
6
(
6
)
,
p
p.
3174
-
3187
,
2
016.
[9]
Standa
rd
devi
a
ti
on
fil
ters
,
2018.
htt
p://s
pat
i
al
-
ana
l
y
st
.
net/ILWI
S/htm
/i
lwisapp/
f
il
te
r_
t
y
pes_stan
dar
d_deviati
on_
fil
ters
.
htm
.
[10]
Loc
a
l
stand
ard
d
evi
a
ti
on
of
imag
e,
2018
.
h
tt
ps:/
/
ww
w.m
at
hwork
s.c
om
/h
el
p/
images/re
f/stdf
il
t
.
html
.
[11]
S.
S.
Singh,
et
.
a
l
,
“
Loc
a
l
cont
r
ast
enha
n
c
ement
u
sing
loc
al
stand
a
rd
devi
ation,
”
Inte
rnational
Journal
of
Computer
Appl
ic
a
ti
ons (
0975
–
888)
,
vol
/i
s
sue:
47
(
15
)
,
pp
.
31
-
35,
2012
.
[12]
M.
Coggins
and
A.
K.
Jain,
“A
spati
al
fil
t
ering
appr
oac
h
to
te
xture
an
aly
si
s,”
Pat
t
ern
Re
c
ognit
ion
Letters
,
vol
/i
ss
ue:
3
(
3
)
,
p
p.
195
–
203
,
198
5.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Dr.
Ta
r
eq
Khan
recei
ved
his
Ph.D.
degr
ee
f
ro
m
the
depa
r
tme
nt
of
E
le
c
tri
c
al
and
Com
pute
r
Engi
ne
eri
ng
of
Univer
sit
y
of
Sa
skatc
hewa
n
,
Ca
nada
.
Dr
.
Khan
i
s
now
an
As
sist
ant
Profess
or
in
the
School
of
En
gine
er
ing
Techn
olog
y
of
Ea
stern
Michi
gan
Univ
ersity
,
US
A.
To
dat
e
,
D
r.
Khan
has
aut
hore
d
(a
nd
co
-
aut
hore
d)
2
books,
1
bo
ok
cha
pte
r
,
18
pee
r
-
rev
ie
wed
j
ourna
ls
and
26
int
ern
at
ion
al
co
nfe
ren
c
e
pape
rs.
He
cur
ren
tly
h
as
2
U
S
pat
ents
gra
nte
d.
In
a
ddit
ion
to
his
ac
ad
emic
r
ese
ar
ch,
h
e
h
as
al
so
industrial
experie
nc
es
on
embedde
d
s
y
stem
p
roje
c
ts
such
as
designi
ng
pr
e
-
p
ai
d
el
e
ct
r
ic
i
t
y
a
nd
gas
m
eteri
ng
sy
st
em,
aut
om
at
ic
m
et
er
re
adi
n
g
(AM
R),
dat
a
ac
cusa
ti
on
and
m
onit
oring
s
y
ste
m
s
et
c.
His
rese
arc
h
intere
sts
inc
lude
image
pro
ce
ss
ing,
sm
art
hom
e,
embedde
d
s
y
stems
ta
rg
eting
healthcare
a
ppl
icati
ons,
Internet
of
Thi
ngs
(
IoT)
,
m
ac
h
ine
le
arn
ing,
and
caps
ule
endosc
op
y
.
He
is
a
m
ember
of
the
Insti
tute
of
El
e
ct
r
ic
a
l
a
nd
El
ectroni
cs
Engi
ne
ers
(IE
EE).
Evaluation Warning : The document was created with Spire.PDF for Python.