I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
7
,
No
.
5
Octo
b
e
r
2
0
1
7
,
p
p
.
2
5
7
4
~
2
5
8
0
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v7
i
1
.
p
p
2
5
7
4
-
2580
2574
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e.
co
m/jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Co
m
pa
ra
tive
Ana
ly
sis
of
C
o
m
m
o
n
Edg
e Det
e
ction A
lg
o
rith
m
s
using
P
re
-
pro
cess
ing
Technique
R.
Vij
a
y
a
K
u
m
a
r
Reddy
1
,
K.
P
rudv
i R
a
j
u
2
,
M
.
J
o
g
endra
K
u
m
a
r
3
,
L.
Ra
v
i K
u
m
a
r
4
,
P
Ra
v
i P
ra
k
a
s
h
5
,
S.
Sa
i K
u
m
a
r
6
De
p
a
rtme
n
t
o
f
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
P
VP
S
IT
,
Ka
n
u
ru
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
2
1
,
201
6
R
ev
i
s
ed
A
p
r
21
,
2
0
1
7
A
cc
ep
ted
J
u
n
25
,
2
0
1
7
Ed
g
e
d
e
tec
ti
o
n
is
th
e
p
ro
c
e
ss
o
f
s
e
g
m
e
n
ti
n
g
a
n
i
m
a
g
e
b
y
d
e
tec
ti
n
g
d
isc
o
n
ti
n
u
it
i
e
s
in
b
rig
h
t
n
e
ss
.
S
e
v
e
ra
l
sta
n
d
a
rd
se
g
m
e
n
tatio
n
m
e
t
h
o
d
s
h
a
v
e
b
e
e
n
w
id
e
ly
u
se
d
f
o
r
e
d
g
e
d
e
te
c
ti
o
n
.
Ho
w
e
v
e
r,
d
u
e
to
i
n
h
e
re
n
t
q
u
a
li
ty
o
f
im
a
g
e
s,
th
e
s
e
m
e
th
o
d
s
p
ro
v
e
in
e
ff
e
c
ti
v
e
i
f
th
e
y
a
re
a
p
p
li
e
d
w
it
h
o
u
t
a
n
y
p
re
-
p
ro
c
e
ss
in
g
.
In
th
is
p
a
p
e
r,
a
n
i
m
a
g
e
p
re
-
p
ro
c
e
s
sin
g
a
p
p
ro
a
c
h
h
a
s
b
e
e
n
a
d
o
p
te
d
in
o
rd
e
r
to
g
e
t
c
e
rtain
p
a
ra
m
e
ters
th
a
t
a
re
u
se
f
u
l
to
p
e
rf
o
rm
b
e
tt
e
r
e
d
g
e
d
e
tec
ti
o
n
w
it
h
th
e
sta
n
d
a
rd
e
d
g
e
d
e
tec
ti
o
n
m
e
th
o
d
s.
T
h
e
p
ro
p
o
se
d
p
re
-
p
ro
c
e
ss
in
g
a
p
p
ro
a
c
h
in
v
o
lv
e
s
m
e
d
ian
f
il
terin
g
to
re
d
u
c
e
th
e
n
o
ise
in
im
a
g
e
a
n
d
th
e
n
e
d
g
e
d
e
tec
ti
o
n
te
c
h
n
i
q
u
e
is
c
a
rried
o
u
t.
F
in
a
ll
y
,
S
tan
d
a
rd
e
d
g
e
d
e
tec
ti
o
n
m
e
th
o
d
s
c
a
n
b
e
a
p
p
li
e
d
t
o
t
h
e
re
su
lt
a
n
t
p
re
-
p
ro
c
e
ss
in
g
im
a
g
e
a
n
d
it
s
S
im
u
latio
n
re
su
lt
s
a
re
sh
o
w
t
h
a
t
o
u
r
p
re
-
p
ro
c
e
ss
e
d
a
p
p
ro
a
c
h
w
h
e
n
u
se
d
w
it
h
a
sta
n
d
a
rd
e
d
g
e
d
e
tec
ti
o
n
m
e
th
o
d
e
n
h
a
n
c
e
s it
s p
e
rf
o
rm
a
n
c
e
.
K
ey
w
o
r
d
:
C
an
n
y
tech
n
iq
u
e
Dig
ital i
m
a
g
e
p
r
o
ce
s
s
i
n
g
Fil
ter
in
g
I
m
ag
e
ed
g
e
d
etec
tio
n
P
r
e
-
p
r
o
ce
s
s
in
g
Co
p
y
rig
h
t
©
2
0
1
7
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
R
.
Vij
ay
a
K
u
m
ar
R
ed
d
y
Dep
ar
t
m
en
t o
f
I
n
f
o
r
m
atio
n
T
e
ch
n
o
lo
g
y
,
P
VP
SI
T
,
Kan
u
r
u
,
A
P
,
I
n
d
ia
v
ij
a
y
2
.
s
r
i@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
E
d
g
e
d
etec
tio
n
i
s
b
asica
ll
y
an
i
m
a
g
e
s
eg
m
e
n
tatio
n
tec
h
n
iq
u
e
th
at
d
iv
id
es
s
p
atial
d
o
m
ain
,
o
n
w
h
ich
th
e
i
m
a
g
e
i
s
d
ef
in
ed
i
n
to
m
e
an
in
g
f
u
l
p
ar
ts
o
r
r
eg
io
n
s
[
1
]
.
T
h
e
d
is
co
n
ti
n
u
ities
ar
e
ab
r
u
p
t
ch
an
g
e
s
i
n
p
ix
e
l
in
te
n
s
it
y
a
n
d
th
e
e
d
g
es
t
h
at
c
h
ar
ac
ter
ize
b
o
u
n
d
ar
ies
o
f
o
b
j
ec
ts
in
a
n
i
m
a
g
e
a
n
d
o
b
j
ec
t
r
ec
o
g
n
i
tio
n
.
T
h
is
is
t
h
e
f
u
n
d
a
m
en
ta
l
i
m
p
o
r
ta
n
ce
o
f
s
eg
m
e
n
tat
io
n
o
f
i
m
a
g
e
i
n
p
r
o
ce
s
s
i
n
g
an
i
m
ag
e
.
E
d
g
es
t
y
p
i
ca
ll
y
o
cc
u
r
o
n
t
h
e
b
o
u
n
d
ar
y
b
et
w
ee
n
t
w
o
d
if
f
er
en
t
r
eg
io
n
s
in
a
n
i
m
ag
e,
w
h
e
r
e
th
er
e
is
a
m
o
r
e
o
r
less
ab
r
u
p
t
ch
a
n
g
e
i
n
p
ix
el
g
r
e
y
le
v
el
o
r
te
x
t
u
r
e
i
n
d
icatin
g
t
h
e
e
n
d
o
f
o
n
e
r
e
g
io
n
an
d
th
e
b
eg
i
n
in
g
o
f
a
n
o
th
er
r
e
g
i
o
n
.
E
d
g
e
d
etec
tio
n
allo
w
s
u
s
er
to
o
b
s
er
v
e
t
h
o
s
e
b
o
u
n
d
ar
ies.
E
d
g
e
d
etec
tio
n
th
er
ef
o
r
e
f
in
d
s
p
r
ac
tical
ap
p
licatio
n
s
in
m
ed
ical
i
m
a
g
in
g
,
co
m
p
u
ter
-
g
u
id
ed
s
u
r
g
er
y
,
s
atellite
i
m
ag
e
s
,
f
ac
e
r
ec
o
g
n
itio
n
,
a
n
d
f
i
n
g
er
p
r
in
t
r
ec
o
g
n
i
tio
n
,
a
u
to
m
atic
tr
af
f
ic
co
n
tr
o
lli
n
g
s
y
s
te
m
s
a
n
d
a
s
tu
d
y
o
f
an
ato
m
ical
s
tr
u
ctu
r
e
etc
.
Se
v
er
al
ed
g
e
d
etec
t
io
n
tech
n
iq
u
e
s
h
av
e
b
ee
n
d
ev
elo
p
ed
s
u
ch
as
R
o
b
er
t,
P
r
ew
itt,
So
b
el
an
d
C
a
n
n
y
.
H
o
w
e
v
er
,
th
is
ed
g
e
d
etec
tio
n
d
o
es
n
o
t
g
i
v
e
n
s
h
ar
p
ed
g
es a
n
d
w
h
ich
h
a
v
e
b
ee
n
h
i
g
h
l
y
s
e
n
s
iti
v
e
an
d
p
r
o
v
id
ed
p
o
o
r
p
er
f
o
r
m
an
ce
i
n
n
o
i
s
y
i
m
a
g
e
s
[
1
].
Ma
in
ai
m
o
f
p
er
f
o
r
m
i
n
g
s
e
g
m
en
tatio
n
o
n
a
g
i
v
en
i
m
ag
e
i
s
to
ex
tr
ac
t
m
a
x
i
m
u
m
i
n
f
o
r
m
atio
n
f
r
o
m
i
m
a
g
e.
T
o
r
em
o
v
e
n
o
is
e
f
r
o
m
i
m
a
g
e
s
o
m
e
f
il
ter
s
ar
e
u
s
ed
lik
e
Gau
s
s
ia
n
f
il
ter
,
m
ed
ia
n
f
ilter
,
etc
[
2
]
.
T
h
e
ed
g
e
d
etec
tio
n
is
also
u
s
ef
u
l
f
o
r
d
e
tectin
g
v
al
u
ab
le
o
r
i
m
p
o
r
ta
n
t
ch
an
g
es
i
n
t
h
e
v
al
u
e
o
f
t
h
e
i
n
ten
s
i
t
y
,
a
n
d
ca
n
b
e
ex
p
lo
r
ed
to
h
id
e
i
m
ag
e
d
is
to
r
t
io
n
s
i
n
tr
o
d
u
ce
d
i
n
t
h
e
c
h
a
n
n
el
.
T
h
is
k
i
n
d
o
f
d
etec
tio
n
is
m
o
s
tl
y
ac
h
iev
ed
u
s
i
n
g
f
ir
s
t
o
r
d
er
o
r
s
ec
o
n
d
o
r
d
er
d
er
iv
ati
v
e
o
f
i
n
t
e
n
s
it
y
v
al
u
es.
Mo
r
eo
v
er
,
o
n
e
o
f
th
e
r
ea
s
o
n
s
th
at
co
u
ld
m
a
k
e
t
h
is
v
ar
iatio
n
h
i
g
h
i
s
th
e
e
x
i
s
ten
c
e
o
f
h
i
g
h
f
r
eq
u
e
n
c
y
(
n
o
is
e)
a
t
th
at
ar
ea
.
On
ce
d
ata
is
r
ec
ei
v
ed
at
th
e
r
ec
eiv
er
,
er
r
o
r
s
ar
e
d
etec
ted
an
d
if
p
o
s
s
ib
le,
th
e
y
ar
e
also
co
r
r
ec
ted
[
3
]
.
A
n
ed
g
e
d
etec
t
io
n
f
ilter
ca
n
also
b
e
u
s
ed
to
i
m
p
r
o
v
e
t
h
e
ap
p
ea
r
an
ce
o
f
b
l
u
r
r
ed
o
r
a
n
ti
-
alia
s
ed
i
m
ag
e
s
t
r
ea
m
s
[
4
]
.
E
d
g
es
ca
n
b
e
d
etec
ted
u
s
i
n
g
v
ar
io
u
s
ed
g
e
d
etec
to
r
s
.
T
h
is
co
n
s
id
er
ed
as
th
e
id
ea
l
ed
g
e
d
etec
tio
n
al
g
o
r
ith
m
f
o
r
i
m
a
g
es
t
h
at
ar
e
co
r
r
u
p
ted
w
it
h
d
if
f
er
e
n
t
t
y
p
es
o
f
n
o
i
s
es
[
5
]
.
An
ed
g
e
is
t
y
p
icall
y
e
x
tr
ac
ted
b
y
co
m
p
u
t
in
g
t
h
e
d
er
iv
a
tiv
e
o
f
th
e
i
m
ag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
f Co
mmo
n
E
d
g
e
Dete
ctio
n
A
lg
o
r
ith
ms u
s
in
g
P
r
e
-
p
r
o
ce
s
s
in
g
Tech
n
iq
u
e
(
R
.
V
ija
ya
K
u
ma
r
R
ed
d
y
)
2575
f
u
n
ctio
n
.
So
m
e
o
f
th
e
ed
g
es
t
h
at
n
o
r
m
all
y
en
co
u
n
ter
ed
i
n
i
m
ag
e
p
r
o
ce
s
s
in
g
ar
e
s
tep
ed
g
e,
r
a
m
p
ed
g
e,
s
p
i
k
e
ed
g
e,
r
o
o
f
ed
g
e.
Step
ed
g
e
is
an
ab
r
u
p
t
in
te
n
s
i
t
y
ch
a
n
g
e.
R
a
m
p
ed
g
e
r
ep
r
esen
ts
a
g
r
ad
u
a
l
ch
a
n
g
e
i
n
in
ten
s
it
y
.
Sp
ik
e
ed
g
e
r
ep
r
esen
ts
a
q
u
ick
c
h
a
n
g
e
a
n
d
i
m
m
ed
iatel
y
r
et
u
r
n
s
to
o
r
ig
in
al
i
n
te
n
s
it
y
le
v
el.
R
o
o
f
ed
g
e
i
s
n
o
t
in
s
ta
n
ta
n
eo
u
s
o
v
er
a
s
h
o
r
t
d
is
tan
ce
[
6
]
.
C
an
n
y
’
s
ai
m
is
to
i
d
en
tify
o
p
ti
m
al
ed
g
e
d
etec
tio
n
al
g
o
r
ith
m
w
h
ic
h
r
ed
u
ce
s
th
e
p
r
o
b
ab
ilit
y
o
f
d
et
ec
tin
g
f
al
s
e
ed
g
es,
a
n
d
g
i
v
es
s
h
ar
p
ed
g
es.
Fi
n
all
y
,
i
n
t
h
i
s
m
et
h
o
d
th
e
i
m
a
g
e
i
s
p
r
o
ce
s
s
ed
an
d
th
e
r
esu
ltin
g
o
n
e
is
ap
p
lied
to
s
tan
d
ar
d
ed
g
e
d
etec
tio
n
m
et
h
o
d
f
o
r
to
en
h
an
ce
th
e
f
i
n
al
ed
g
e
d
etec
tio
n
m
et
h
o
d
p
er
f
o
r
m
an
c
e.
T
h
e
m
ai
n
ai
m
o
f
an
ed
g
e
d
etec
tio
n
m
et
h
o
d
is
to
in
v
est
i
g
atio
n
ed
g
es
o
f
a
n
i
m
a
g
e
f
o
r
d
etec
tio
n
a
n
d
lo
ca
lizatio
n
[
7
]
.
I
n
o
u
r
w
o
r
k
,
w
e
p
r
esen
ts
a
p
r
e
-
p
r
o
ce
s
s
i
n
g
a
p
p
r
o
ac
h
is
u
s
ed
to
en
h
a
n
ce
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
s
tan
d
ar
d
ed
g
e
d
etec
tio
n
m
e
th
o
d
s
.
I
n
o
u
r
ap
p
r
o
ac
h
t
h
e
i
m
a
g
e
is
p
r
e
-
p
r
o
ce
s
s
ed
w
it
h
m
ed
ian
f
ilter
i
n
g
a
n
d
th
e
n
,
a
s
tan
d
ar
d
ed
g
e
d
etec
tio
n
m
e
th
o
d
is
ap
p
lied
to
th
e
r
esu
ltan
t
s
eg
m
en
ted
i
m
a
g
e.
T
h
e
p
r
o
p
o
s
ed
p
r
e
-
p
r
o
ce
s
s
in
g
i
n
v
o
lv
e
s
co
m
p
u
tatio
n
o
f
m
ed
ia
n
f
ilter
i
n
g
o
f
i
m
a
g
e
an
d
t
h
e
n
i
m
a
g
e
s
eg
m
e
n
tatio
n
i
s
ca
r
r
ied
o
u
t.
T
h
e
r
est
o
f
th
e
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
w
s
:
I
n
Sectio
n
2
,
a
r
ev
ie
w
o
f
ed
g
e
d
etec
tio
n
m
et
h
o
d
s
is
d
escr
ib
ed
;
ad
v
an
tag
e
an
d
d
i
s
ad
v
an
tag
e
o
f
t
h
ese
m
et
h
o
d
s
ar
e
s
u
m
m
ar
ized
.
I
n
Secti
o
n
3
.
t
h
e
P
r
o
b
lem
Fo
r
m
u
k
atio
n
f
o
r
ed
g
e
d
etec
tio
n
is
in
tr
o
d
u
ce
d
.
Sectio
n
4
,
p
r
e
s
en
t
s
e
x
p
er
i
m
e
n
t
r
esu
lts
an
d
s
h
o
w
s
a
q
u
an
tita
tiv
e
co
m
p
ar
is
o
n
b
et
w
ee
n
p
r
e
-
p
r
o
ce
s
s
ed
an
d
s
tan
d
ar
d
m
e
th
o
d
s
w
it
h
a
n
d
w
it
h
o
u
t
th
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
.
W
e
d
r
a
w
t
h
e
co
n
clu
s
io
n
i
n
Sectio
n
5
.
2.
A
RE
VI
E
W
O
N
E
D
G
E
D
E
T
E
CT
I
O
N
M
E
T
H
O
DS
E
d
g
e
d
etec
tio
n
i
s
t
h
e
p
r
o
ce
s
s
o
f
d
eter
m
i
n
i
n
g
w
h
er
e
ed
g
es
o
f
o
b
j
ec
ts
f
all
w
it
h
i
n
a
n
i
m
a
g
e.
Fig
u
r
e
1
s
h
o
w
s
s
c
h
e
m
atic
d
ia
g
r
a
m
f
o
r
th
e
s
tan
d
ar
d
ed
g
e
d
etec
tio
n
m
eth
o
d
.
T
o
in
d
en
ti
f
y
an
d
d
etec
t
ab
r
u
p
t
ch
a
n
g
e
a
t
ed
g
es,
s
e
v
er
al
o
p
er
ato
r
s
h
av
e
b
ee
n
co
n
s
tr
u
c
ted
b
ased
o
n
d
if
f
er
e
n
t
id
ea
s
.
I
n
t
h
e
f
o
llo
win
g
s
ec
tio
n
,
a
b
r
ief
r
ev
ie
w
o
n
E
d
g
e
d
ec
tectio
n
m
e
th
o
d
s
.
Fig
u
r
e
1
.
Stan
d
ar
d
ed
g
e
d
etec
to
r
.
E
x
p
lain
i
n
g
r
esear
ch
c
h
r
o
n
o
lo
g
ical,
in
c
lu
d
i
n
g
r
esear
c
h
d
esi
g
n
,
r
esear
c
h
p
r
o
ce
d
u
r
e
(
in
th
e
f
o
r
m
o
f
alg
o
r
ith
m
s
,
P
s
eu
d
o
co
d
e
o
r
o
th
er
)
,
h
o
w
to
test
an
d
d
ata
ac
q
u
is
itio
n
[1
-
3]
.
T
h
e
d
escr
ip
ti
o
n
o
f
th
e
co
u
r
s
e
o
f
r
esear
ch
s
h
o
u
ld
b
e
s
u
p
p
o
r
ted
r
ef
er
en
ce
s
,
s
o
th
e
ex
p
la
n
atio
n
ca
n
b
e
ac
ce
p
ted
s
cien
ti
f
icall
y
[
2
]
,
[
4
]
.
2
.
1
.
F
irst
O
rder
E
dg
e
Det
ec
t
o
r
T
h
e
g
r
ad
ien
t
m
eth
o
d
is
u
s
ed
t
o
d
etec
t
th
e
ed
g
es
b
y
lo
o
k
i
n
g
f
o
r
th
e
m
a
x
i
m
u
m
a
n
d
m
i
n
i
m
u
m
i
n
th
e
f
ir
s
t
d
er
i
v
ativ
e
o
f
t
h
e
i
m
ag
e.
T
h
u
s
,
co
n
s
id
er
t
h
e
t
w
o
d
i
m
e
n
s
io
n
al
f
u
n
ctio
n
f
(
x
,
y
)
to
r
ep
r
e
s
en
t
t
h
e
i
n
p
u
t
i
m
a
g
e
th
en
i
m
a
g
e
g
r
ad
ien
t
i
s
g
i
v
e
n
b
y
t
h
e
f
o
llo
w
in
g
E
q
u
atio
n
1
.
T
h
e
m
ag
n
it
u
d
e
o
f
t
h
e
g
r
a
d
ien
t
co
m
p
u
ted
b
y
E
q
u
atio
n
2
g
iv
e
s
ed
g
e
s
tr
en
g
t
h
.
T
h
e
g
r
ad
ien
t
d
ir
ec
tio
n
is
alw
a
y
s
p
er
p
en
d
icu
lar
to
th
e
d
ir
ec
tio
n
o
f
th
e
ed
g
e.
R
o
b
er
t,
So
b
el,
an
d
P
r
ew
itt
o
p
er
ato
r
s
ar
e
class
if
ied
as
s
tan
d
a
r
d
f
ir
s
t
o
r
d
er
d
e
r
iv
ativ
e
o
p
er
ato
r
s
w
h
ich
ar
e
ea
s
y
to
o
p
er
ate
b
u
t h
i
g
h
l
y
s
e
n
s
iti
v
e
to
n
o
is
e
[
1
]
.
=
[
(
,
)
]
=
[
f
x
f
y
]
=
[
(
,
)
(
,
)
]
(
1
)
T
h
e
g
r
ad
ien
t
m
a
g
n
itu
d
e
ca
n
b
e
co
m
p
u
ted
b
y
th
e
E
q
u
a
tio
n
2
:
|
∇
f
|
=
√
f
x
2
+
2
=
√
(
)
2
+
(
)
2
≃
|
|
+
|
|
(
2
)
T
h
e
g
r
ad
ien
t d
ir
ec
tio
n
ca
n
b
e
co
m
p
u
ted
b
y
th
e
E
q
u
atio
n
3
:
θ
=
ta
n
−
1
(
f
y
f
x
)
(
3
)
Ed
g
e
d
e
t
e
c
t
i
o
n
m
e
t
h
o
d
I
n
p
u
t
i
m
a
g
e
O
u
t
p
u
t
i
mag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
20
88
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
5
,
Octo
b
er
2
0
1
7
:
2
5
7
4
–
2
5
8
0
2576
2
.
1
.
1
.
Ro
bert
edg
e
d
et
ec
t
o
r
T
h
e
R
o
b
er
ts
o
p
e
r
ato
r
p
er
f
o
r
m
s
a
s
i
m
p
le,
q
u
ic
k
to
co
m
p
u
te,
T
w
o
-
Di
m
en
s
io
n
s
p
atial
g
r
ad
ien
t
m
ea
s
u
r
e
m
e
n
t o
n
a
n
i
m
a
g
e.
I
t th
u
s
h
i
g
h
li
g
h
ts
r
eg
io
n
s
o
f
h
i
g
h
s
p
atial
g
r
ad
ien
t
w
h
ic
h
o
f
te
n
c
o
r
r
esp
o
n
d
to
e
d
g
es.
I
n
its
m
o
s
t c
o
m
m
o
n
u
s
a
g
e,
t
h
e
in
p
u
t
to
th
e
o
p
er
ato
r
is
a
g
r
a
y
s
ca
le
i
m
ag
e,
a
s
is
th
e
o
u
tp
u
t.
P
ix
el
v
al
u
es a
t e
ac
h
p
o
in
t
in
th
e
o
u
tp
u
t
r
ep
r
esen
t
t
h
e
esti
m
ated
ab
s
o
lu
te
m
a
g
n
it
u
d
e
o
f
th
e
s
p
atial
g
r
ad
ie
n
t
o
f
th
e
in
p
u
t
i
m
a
g
e
at
th
at
p
o
in
t.
I
t u
s
es t
h
e
f
o
llo
w
i
n
g
2
×
2
t
w
o
k
er
n
el
s
[
8
]:
G
x
=
[
+
1
0
0
−
1
]
G
y
=
[
0
+
1
−
1
0
]
(
4
)
T
h
e
p
lu
s
f
ac
to
r
o
f
t
h
is
o
p
er
ato
r
is
its
s
i
m
p
licit
y
b
u
t
h
a
v
i
n
g
s
m
all
k
er
n
e
l
it
i
s
h
ig
h
l
y
s
en
s
iti
v
e
to
n
o
is
e
an
d
n
o
t c
o
m
p
a
tib
le
w
it
h
to
d
a
y
’
s
tec
h
n
o
lo
g
y
[
8
].
2
.
1
.
2
.
So
bel Ed
g
e
Det
ec
t
o
r
T
h
e
So
b
el
o
p
er
at
o
r
p
er
f
o
r
m
s
a
t
wo
-
d
i
m
en
s
io
n
s
p
atial
g
r
ad
ien
t
m
ea
s
u
r
e
m
en
t
o
n
an
i
m
ag
e
an
d
s
o
e
m
p
h
a
s
izes
r
eg
io
n
s
o
f
h
i
g
h
s
p
atial
g
r
ad
ien
t
t
h
at
co
r
r
esp
o
n
d
to
ed
g
es.
H
en
ce
,
it
is
u
s
ed
to
f
in
d
t
h
e
ap
p
r
o
x
im
a
te
ab
s
o
lu
te
g
r
ad
ien
t
m
a
g
n
itu
d
e
a
t
ea
ch
p
o
in
t
i
n
a
n
in
p
u
t
g
r
a
y
s
ca
le
i
m
ag
e.
I
t
u
s
es
th
e
f
o
llo
w
i
n
g
3
×
3
t
w
o
k
er
n
e
ls
:
G
x
=
[
−
1
0
+
1
−
1
0
+
1
−
1
0
+
1
]
a
n
d
G
y
=
[
+
1
+
1
+
1
0
0
0
−
1
−
1
−
1
]
(
5
)
As
co
m
p
ar
ed
to
R
o
b
er
t
o
p
e
r
a
to
r
,
So
b
el
o
p
er
ato
r
h
as
s
lo
w
co
m
p
u
tatio
n
.
W
h
en
co
m
p
ar
ed
to
R
o
b
er
t
o
p
er
ato
r
it is
less
s
e
n
s
iti
v
e
to
n
o
is
e
.
2
.
1
.
3
.
P
re
w
it
t
E
dg
e
Det
ec
t
o
r
P
r
ew
i
tt
ed
g
e
o
p
er
ato
r
g
iv
es
b
etter
p
er
f
o
r
m
a
n
ce
th
a
n
t
h
at
o
f
So
b
el
o
p
e
r
ato
r
.
T
h
e
f
u
n
ctio
n
o
f
P
r
ew
i
tt
ed
g
e
d
etec
to
r
is
al
m
o
s
t t
h
e
s
a
m
e
as So
b
el
d
etec
to
r
b
u
t h
av
e
d
if
f
er
e
n
t k
er
n
els
:
G
x
=
[
−
1
0
+
1
−
1
0
+
1
−
1
0
+
1
]
a
n
d
G
y
=
[
+
1
+
1
+
1
0
0
0
−
1
−
1
−
1
]
(
6
)
2
.
2
.
Seco
nd
O
rd
er
E
dg
e
Det
ec
t
o
r
L
et
f
(
x
,
y
)
b
e
a
co
n
ti
n
u
o
u
s
t
w
o
-
d
i
m
e
n
s
io
n
al
s
ca
lar
f
ield
w
it
h
(
x
,
y
)
b
ei
n
g
a
p
o
i
n
t
i
n
th
e
i
m
a
g
e.
T
o
d
eter
m
in
e
t
h
e
d
er
iv
at
iv
e
s
an
o
p
er
ato
r
is
ap
p
lied
t
o
th
e
in
te
n
s
it
y
f
u
n
ctio
n
f
(
x
,
y
)
.
An
y
d
ef
i
n
itio
n
o
f
a
s
ec
o
n
d
o
r
d
er
d
er
iv
ate
m
u
s
t
b
e
ze
r
o
in
f
lat
ar
ea
s
an
d
m
u
s
t
b
e
n
o
n
z
er
o
at
th
e
o
n
s
et
an
d
en
d
o
f
a
g
r
a
y
le
v
el
s
tep
an
d
r
a
m
p
;
an
d
m
u
s
t b
e
ze
r
o
alo
n
g
r
a
m
p
s
o
f
co
n
s
ta
n
t slo
p
e
[
9
]
.
T
h
e
L
ap
lacia
n
o
p
er
ato
r
∇
2
f
o
r
a
2
D
i
m
ag
e
f
(
x
,
y
)
is
d
ef
in
ed
b
y
t
h
e
f
o
llo
w
i
n
g
E
q
u
a
tio
n
7
[8
]:
2
(
,
)
=
2
2
(
,
)
+
2
2
(
,
)
(
7
)
T
h
er
e
ar
e
s
ev
er
al
w
a
y
s
to
d
ef
i
n
e
d
ig
ital
L
ap
lacia
n
u
s
in
g
n
ei
g
h
b
o
r
h
o
o
d
s
.
I
n
A
n
y
d
ef
i
n
itio
n
,
it
h
as
to
s
ati
s
f
y
t
h
e
p
r
o
p
er
ties
o
f
a
s
ec
o
n
d
d
er
iv
ate
s
tated
in
E
q
u
atio
n
(
1
1
)
.
T
h
e
f
o
llo
w
in
g
n
o
tat
io
n
is
u
s
ed
f
o
r
s
ec
o
n
d
o
r
d
er
d
er
iv
ativ
e
i
n
th
e
x
an
d
y
d
ir
ec
tio
n
s
[
9
]
.
2
2
2
=
(
+
1
,
)
+
(
−
1
,
)
−
2
(
,
)
(
8
)
2
2
2
=
(
,
+
1
)
+
(
,
−
1
)
−
2
(
,
)
(
9
)
2
=
[
(
+
1
,
)
+
(
−
1
,
)
+
(
,
+
1
)
+
(
,
−
1
)
−
4
(
,
)
(
1
0
)
Fu
r
t
h
er
m
o
r
e,
w
h
e
n
th
e
f
ir
s
t d
e
r
iv
ati
v
e
is
at
a
m
a
x
i
m
u
m
,
t
h
e
s
ec
o
n
d
d
er
iv
ativ
e
i
s
ze
r
o
[
8
].
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
f Co
mmo
n
E
d
g
e
Dete
ctio
n
A
lg
o
r
ith
ms u
s
in
g
P
r
e
-
p
r
o
ce
s
s
in
g
Tech
n
iq
u
e
(
R
.
V
ija
ya
K
u
ma
r
R
ed
d
y
)
2577
2
.
2
.
1
.
L
a
pla
cia
n
L
ap
lacia
n
o
f
an
i
m
a
g
e
f
(
x
,
y
)
is
d
ef
i
n
ed
b
y
:
[
(
,
)
]
=
2
=
[
(
+
1
,
)
+
(
−
1
,
)
+
(
,
+
1
)
+
(
,
−
1
)
−
4
(
,
)
(
1
1
)
I
n
th
i
s
ca
s
e,
L
ap
lacia
n
k
er
n
el
g
iv
e
n
b
y
t
h
e
f
o
llo
w
in
g
eq
u
atio
n
is
ad
o
p
ted
.
=
[
0
1
0
1
−
4
1
0
1
0
]
(
12)
2
.
2
.
2
L
a
pla
cia
n o
f
g
a
us
s
ia
n
(
M
a
rr
-
H
ild
re
t
h E
dg
e
Det
ec
t
o
r)
I
t f
o
llo
w
s
th
e
f
o
llo
w
i
n
g
f
o
u
r
s
tep
s
[
1
1
]
:
1)
S
m
o
o
th
i
n
g
t
h
e
i
m
a
g
e
u
s
i
n
g
Gau
s
s
ia
n
f
ilter
.
Ga
u
s
s
ian
s
m
o
o
th
in
g
h
elp
s
el
i
m
i
n
ate
n
o
i
s
e
.
T
h
e
lar
g
er
th
e
s
i
g
m
a,
th
e
g
r
ea
ter
th
e
s
m
o
o
th
i
n
g
.
(
,
)
=
1
√
2
2
−
(
2
+
2
2
2
)
(
1
3
)
2)
E
n
h
a
n
ci
n
g
t
h
e
ed
g
e
s
u
s
in
g
L
a
p
lacia
n
o
p
er
ato
r
.
(
,
)
=
−
1
4
[
1
−
2
+
2
2
2
]
−
(
2
+
2
2
2
)
(
1
4
)
3)
E
s
ti
m
a
te
th
e
ze
r
o
cr
o
s
s
in
g
s
d
en
o
te
th
e
ed
g
e
lo
ca
tio
n
.
4)
Su
b
-
p
i
x
el
lo
ca
tio
n
o
f
th
e
ed
g
e
is
d
eter
m
i
n
ed
b
y
u
s
i
n
g
li
n
ea
r
in
ter
p
o
latio
n
.
T
o
o
m
u
ch
s
m
o
o
th
in
g
m
a
y
m
a
k
e
th
e
d
etec
tio
n
o
f
ed
g
es d
i
f
f
i
cu
lt to
in
d
e
n
ti
f
y
.
2
.
2
.
3
.
Dif
f
er
ence
o
f
G
a
us
s
ia
n
T
o
r
ed
u
ce
th
e
co
m
p
u
ta
tio
n
al
r
eq
u
ir
e
m
e
n
ts
,
t
h
e
L
ap
lacia
n
o
f
G
au
s
s
ia
n
(
L
o
G
)
is
b
e
si
m
il
a
r
to
th
e
T
h
e
d
if
f
er
e
n
ce
o
f
G
a
u
s
s
ian
(
Do
G
)
.
T
h
e
w
id
t
h
o
f
t
h
e
ed
g
e
ca
n
b
e
ch
an
g
ed
b
y
ad
j
u
s
ted
σ
1
a
n
d
σ
2
v
a
lu
e
s
.
Do
G
o
p
er
ato
r
o
f
an
i
m
a
g
e
f
(
x
,
y
)
is
d
ef
in
ed
b
y
:
(
,
)
=
−
(
2
+
2
2
1
2
)
2
1
2
−
−
(
2
+
2
2
2
2
)
2
2
2
(
1
5
)
L
o
G
r
eq
u
ir
e
s
lar
g
e
co
m
p
u
tat
io
n
ti
m
e
f
o
r
a
lar
g
e
ed
g
e
d
e
tecto
r
m
as
k
.
Seco
n
d
Or
d
er
Der
iv
ati
v
e
Me
th
o
d
s
P
r
o
p
er
ties
.
T
h
e
f
o
llo
w
i
n
g
ar
e
t
h
e
m
o
s
t i
m
p
o
r
tan
t p
r
o
p
er
ties
to
co
n
s
id
er
w
h
e
n
u
s
i
n
g
th
e
s
ec
o
n
d
o
r
d
er
d
er
iv
ate
:
1)
L
ap
lacia
n
is
v
er
y
s
e
n
s
i
tiv
e
to
n
o
is
e
2)
Fals
e
an
d
m
is
s
i
n
g
ed
g
e
s
r
e
m
ai
n
s
3)
L
o
ca
lizatio
n
i
s
b
etter
th
a
n
g
r
a
d
ien
t o
p
er
ato
r
s
2
.
3
.
Ca
nn
y
edg
e
d
et
ec
t
o
r
T
h
e
C
an
n
y
ed
g
e
d
etec
tio
n
a
lg
o
r
ith
m
i
s
k
n
o
w
n
to
be
th
e
o
p
ti
m
al
ed
g
e
d
etec
to
r
alg
o
r
th
i
m
f
o
r
d
etec
tio
n
o
f
ed
g
e
in
s
e
g
m
e
n
ta
tio
n
o
f
a
i
m
a
g
e
.
I
n
t
h
is
p
ap
er
,
we
f
o
llo
w
ed
a
li
s
t
o
f
cr
iter
ia
to
i
m
p
r
o
v
e
c
u
r
r
en
t
m
et
h
o
d
s
o
f
ed
g
e
d
etec
tio
n
.
T
h
e
f
ir
s
t
an
d
m
o
s
t
o
b
v
io
u
s
is
lo
w
er
r
o
r
r
ate.
T
h
e
f
r
is
t
cr
iter
io
n
,
i
t
is
im
p
o
r
tan
t
th
at
ed
g
es
o
cc
u
r
r
i
n
g
i
n
a
n
i
m
a
g
e
s
h
o
u
ld
n
o
t
b
e
m
is
s
ed
.
T
h
e
s
ec
o
n
d
cr
iter
io
n
is
th
at
t
h
e
e
d
g
e
p
o
in
ts
b
e
w
ell
lo
ca
lized
.
I
n
o
th
er
w
o
r
d
s
,
th
e
d
is
tan
ce
b
et
w
ee
n
t
h
e
ed
g
e
p
ix
els
as
f
o
u
n
d
b
y
t
h
e
d
etec
to
r
an
d
th
e
ac
tu
al
ed
g
e
i
s
to
b
e
at
a
m
in
i
m
u
m
[
1
2
]
.
A
t
h
ir
d
cr
iter
io
n
is
to
h
av
e
s
i
n
g
l
e
ed
g
e
to
o
n
l
y
o
n
e
r
esp
o
n
s
e
o
f
a
ed
g
e
.
T
h
is
w
a
s
i
m
p
le
m
en
ted
b
ec
au
s
e
t
h
e
f
ir
s
t
t
w
o
cr
iter
io
r
s
w
er
e
n
o
t
s
u
b
s
tan
tia
l
e
n
o
u
g
h
to
co
m
p
lete
l
y
eli
m
in
at
in
g
th
e
p
o
s
s
ib
ilit
y
o
f
m
u
l
tip
le
r
esp
o
n
s
es
to
ed
g
es.
T
h
e
alg
o
r
ith
m
t
h
e
n
tr
ac
k
s
alo
n
g
th
e
to
p
o
f
th
ese
r
id
g
es
an
d
s
ets
to
ze
r
o
all
p
ix
els
th
at
ar
e
n
o
t
ac
tu
all
y
o
n
t
h
e
r
id
g
e
to
p
s
o
as
to
g
iv
e
a
t
h
i
n
li
n
e
in
t
h
e
o
u
tp
u
t,
a
p
r
o
ce
s
s
k
n
o
w
n
a
s
n
o
n
m
ax
i
m
a
l
s
u
p
p
r
ess
io
n
.
T
h
e
tr
ac
k
in
g
p
r
o
ce
s
s
ex
h
ib
its
h
y
s
ter
esis
co
n
tr
o
lled
b
y
t
w
o
th
r
esh
o
ld
s
v
a
lu
e
s
:
T
1
an
d
T
2
w
it
h
T
1
>
T
2
.
T
r
ac
k
in
g
ca
n
o
n
l
y
b
eg
i
n
at
a
p
o
in
t
o
n
a
r
id
g
e
h
i
g
h
er
t
h
an
T
1
th
r
es
h
o
ld
.
T
r
ac
k
in
g
t
h
e
n
co
n
tin
u
es
i
n
b
o
th
d
ir
ec
tio
n
s
o
u
t
f
r
o
m
t
h
at
p
o
in
t
u
n
t
il
t
h
e
h
eig
h
t
o
f
t
h
e
r
id
g
e
f
alls
b
elo
w
T
2
th
er
s
h
o
ld
.
T
h
i
s
h
y
s
ter
esi
s
h
e
lp
s
to
en
s
u
r
e
th
at
n
o
is
y
ed
g
e
s
ar
e
n
o
t
b
r
o
k
e
n
u
p
in
to
m
u
lt
ip
le
ed
g
es.
I
n
o
r
d
er
to
i
m
p
le
m
e
n
t
th
e
ca
n
n
y
ed
g
e
d
etec
to
r
alg
o
r
ith
m
th
e
s
tep
s
m
u
s
t b
e
f
o
llo
w
ed
[
10
].
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
20
88
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
5
,
Octo
b
er
2
0
1
7
:
2
5
7
4
–
2
5
8
0
2578
Step
1
:
Gau
s
s
ian
f
ilter
s
.
Step
2
:
T
ak
e
th
e
g
r
ad
ien
t o
f
t
h
e
i
m
ag
e.
Step
3
:
No
n
-
m
ax
i
m
u
m
s
u
p
p
r
ess
io
n
.
Step
4
:
T
h
e
ed
g
e
d
ir
ec
tio
n
ar
e
co
m
p
u
ted
u
s
in
g
t
h
e
g
r
ad
ien
t v
alu
es i
n
t
h
e
x
an
d
y
d
ir
ec
tio
n
s
o
f
an
2
D
i
m
a
g
e.
Step
5
:
E
d
g
e
d
ir
ec
tio
n
th
at
ca
n
b
e
tr
ac
ed
in
an
i
m
ag
e.
Step
6
:
H
y
s
ter
esi
s
.
2
.
4
.
M
er
it
s
a
nd
de
m
er
it
s
o
f
s
t
a
nd
a
rd
edg
e
det
ec
t
io
n
m
et
h
o
ds
E
ac
h
ed
g
e
d
etec
tio
n
m
eth
o
d
h
as its
Me
r
its
a
n
d
De
m
er
it
s
.
T
a
ble 1
s
u
m
m
ar
izes t
h
e
m
ai
n
M
er
its
an
d
De
m
er
its
o
f
ea
ch
m
et
h
o
d
[
11
].
T
ab
le
1
.
Me
r
its
an
d
d
em
er
its
o
f
ed
g
e
d
etec
to
r
io
n
m
et
h
o
d
s
.
M
e
t
h
o
d
M
e
r
i
t
s
D
e
me
r
i
t
s
S
o
b
e
l
,
P
r
e
w
i
t
t
,
R
o
b
e
r
t
s
S
i
mp
l
i
c
i
t
y
i
n
d
e
t
e
c
t
i
o
n
o
f
e
d
g
e
s a
n
d
t
h
e
i
r
o
r
i
e
n
t
a
t
i
o
n
s
S
e
n
si
t
i
v
i
t
y
t
o
n
o
i
se
d
a
t
a
a
n
d
i
n
a
c
c
u
r
a
t
e
Z
e
r
o
c
r
o
ssi
n
g
(
L
a
p
l
a
c
i
a
n
,
2
n
d
d
i
r
e
c
t
i
o
n
a
l
d
e
r
i
v
a
t
i
v
e
)
D
e
t
e
c
t
i
o
n
o
f
e
d
g
e
s a
n
d
t
h
e
i
r
o
r
i
e
n
t
a
t
i
o
n
s a
r
e
h
a
v
i
n
g
f
i
x
e
d
c
h
a
r
a
c
t
e
r
i
st
i
c
s
i
n
a
l
l
d
i
r
e
c
t
i
o
n
s o
f
i
mag
e
R
e
sp
o
n
d
i
n
g
t
o
so
me
o
f
t
h
e
e
x
i
st
i
n
g
e
d
g
e
s,
se
n
si
t
i
v
i
t
y
t
o
n
o
i
se
L
a
p
l
a
c
i
a
n
o
f
G
a
u
ssi
a
n
(
L
o
G
)
-
(
M
a
r
r
-
H
i
l
d
r
e
t
h
)
F
i
n
d
i
n
g
t
h
e
c
o
r
r
e
c
t
p
l
a
c
e
s o
f
e
d
g
e
s,
a
n
d
t
e
st
i
n
g
w
i
d
e
r
a
r
e
a
a
r
o
u
n
d
t
h
e
p
i
x
e
l
f
o
r
e
d
g
e
s
M
a
l
f
u
n
c
t
i
o
n
i
n
g
a
t
t
h
e
c
o
r
n
e
r
s,
c
u
r
v
e
s a
n
d
w
h
e
r
e
t
h
e
g
r
a
y
l
e
v
e
l
i
n
t
e
n
s
i
t
y
f
u
n
c
t
i
o
n
v
a
r
i
e
s.
D
u
e
t
o
L
a
p
l
a
c
i
a
n
f
i
l
t
e
r
t
h
e
o
r
i
e
n
t
a
t
i
o
n
o
f
e
d
g
e
a
r
e
c
a
n
n
o
t
b
e
f
i
n
d
.
G
a
u
ssi
a
n
(
C
a
n
n
y
,
S
h
e
n
-
C
a
st
a
n
)
U
si
n
g
p
r
o
b
a
b
i
l
i
t
y
f
o
r
f
i
n
d
i
n
g
e
r
r
o
r
r
a
t
e
,
l
o
c
a
l
i
z
a
t
i
o
n
o
f
r
e
sp
o
n
se
.
I
mp
r
o
v
i
n
g
si
g
n
a
l
t
o
n
o
i
se
r
a
t
i
o
f
o
r
b
e
t
t
e
r
d
e
t
e
c
t
i
o
n
.
s
p
e
c
i
a
l
l
y
i
n
n
o
i
se
c
o
n
d
i
t
i
o
n
s.
C
o
mp
l
e
x
c
o
mp
u
t
a
t
i
o
n
s,
f
a
l
se
z
e
r
o
c
r
o
ssi
n
g
,
t
i
me
c
o
n
su
mi
n
g
3.
P
RO
B
L
E
M
F
O
R
M
UL
AT
I
O
N
Fig
u
r
e
2
s
h
o
w
s
s
ch
e
m
at
ic
d
iag
r
a
m
f
o
r
th
e
p
r
o
p
o
s
ed
p
r
e
-
p
r
o
ce
s
s
ed
ed
g
e
d
etec
tio
n
m
et
h
o
d
.
I
n
th
i
s
m
et
h
o
d
th
e
i
m
ag
e
is
b
ee
n
p
re
-
p
r
o
ce
s
s
ed
in
f
ir
s
t
s
tep
an
d
t
h
e
n
i
n
s
ec
o
n
d
s
tep
t
h
e
r
e
s
u
l
ted
i
m
a
g
e
i
s
ap
p
lied
to
s
tan
d
ar
d
ed
g
e
d
etec
tio
n
m
et
h
o
d
.
T
h
e
p
r
o
p
o
s
ed
p
r
e
-
p
r
o
ce
s
s
in
g
i
n
v
o
lv
e
s
co
m
p
u
tat
io
n
o
f
m
ed
ian
f
ilter
i
n
g
o
f
i
m
a
g
e
an
d
th
e
n
i
m
a
g
e
s
e
g
m
e
n
tatio
n
is
ca
r
r
ied
o
u
t.
Fig
u
r
e
2
.
P
r
o
p
o
s
ed
tech
n
iq
u
e
f
o
r
ed
g
e
d
etec
tio
n
T
h
is
p
r
e
-
p
r
o
ce
s
s
in
g
ca
n
b
e
i
m
p
le
m
en
ted
i
n
th
e
f
o
llo
w
i
n
g
t
wo
s
tep
s
.
Step
1
: p
r
e
-
p
r
o
ce
s
s
in
g
1.
C
o
n
v
er
t th
e
o
r
ig
i
n
al
i
m
a
g
e
Fi
g
u
r
e
3
to
g
r
a
y
s
ca
le
as s
h
o
w
n
i
n
Fi
g
u
r
e
4
2.
C
alcu
latio
n
to
tal
ed
g
e
p
i
x
els i
d
en
tifie
d
as
e
d
g
es
o
f
g
r
a
y
s
ca
le
i
m
ag
e
b
e
f
o
r
e
ap
p
ly
t
h
e
m
ed
ia
n
f
i
lter
3.
Fil
ter
th
e
g
r
a
y
s
ca
le
i
m
a
g
e
u
s
i
n
g
a
p
r
e
-
p
r
o
ce
s
s
in
g
tec
h
n
iq
u
e
i.
e.
m
ed
ia
n
f
ilter
̂
(
,
)
=
(
,
)
{
(
,
)
}
(
1
6
)
4.
C
alcu
latio
n
t
o
tal
ed
g
e
p
i
x
els i
d
en
tifie
d
as e
d
g
es
o
f
g
r
a
y
s
ca
le
i
m
ag
e
a
f
ter
ap
p
l
y
i
n
g
t
h
e
m
ed
i
an
f
i
lter
.
Ste
p
2
: E
d
g
e
d
etec
tio
n
m
et
h
o
d
an
d
c
o
m
p
ar
s
io
n
1.
Fin
d
t
h
e
d
if
f
er
en
ce
b
et
w
ee
n
th
e
o
r
ig
in
al
g
r
a
y
s
ca
le
i
m
a
g
e
an
d
m
ed
ian
f
ilter
g
r
a
y
s
ca
le
i
m
a
g
e
2.
Usi
n
g
t
h
e
No
is
e
R
ed
u
ce
d
R
ati
o
(
NR
R
)
.
ca
lcu
l
ate
t
h
e
co
m
p
ar
i
s
o
n
b
et
w
ee
n
t
h
e
m
3.
A
p
p
l
y
t
h
e
ab
o
v
e
s
tep
s
to
d
if
f
e
r
en
t e
d
g
e
d
etec
tio
n
al
g
o
r
ith
m
s
.
P
r
o
p
o
se
d
p
r
e
-
p
r
o
c
e
ssi
n
g
E
d
g
e
d
e
t
e
c
t
i
o
n
m
e
t
h
o
d
I
n
p
u
t
i
m
a
g
e
O
u
t
p
u
t
i
mag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
C
o
mp
a
r
a
tive
A
n
a
lysi
s
o
f Co
mmo
n
E
d
g
e
Dete
ctio
n
A
lg
o
r
ith
ms u
s
in
g
P
r
e
-
p
r
o
ce
s
s
in
g
Tech
n
iq
u
e
(
R
.
V
ija
ya
K
u
ma
r
R
ed
d
y
)
2579
F
ig
u
re
3
.
Orig
i
n
a
l
im
a
g
e
s
F
ig
u
re
4
.
G
ra
y
ima
g
e
s
4.
RE
SU
L
T
S
T
h
is
s
ec
tio
n
p
r
esen
t
s
a
co
m
p
ar
is
o
n
b
et
w
ee
n
d
i
f
f
er
en
t
ed
g
e
d
etec
tio
n
m
o
d
els.
Fo
r
th
i
s
p
u
r
p
o
s
e
MA
T
L
A
B
8
.
1
(
R
2
0
1
3
a)
p
r
o
g
r
a
m
h
a
s
b
ee
n
ad
o
p
ted
to
in
v
esti
g
ate
th
e
d
i
f
f
er
en
ce
b
et
wee
n
ed
g
e
d
etec
t
io
n
m
o
d
el
s
.
T
h
is
co
m
p
ar
is
o
n
b
et
w
ee
n
R
o
b
er
ts
,
P
r
e
w
it
t,
s
o
b
el,
lo
g
a
n
d
C
an
n
y
ed
g
e
d
etec
tio
n
m
et
h
o
d
s
i
s
d
o
n
e
u
s
i
n
g
g
r
o
u
n
d
tr
u
t
h
o
f
i
m
a
g
e
s
.
T
h
e
p
er
f
o
r
m
a
n
ce
p
ar
a
m
ete
r
s
u
s
ed
ar
e
E
d
g
e
P
ix
el
s
B
ef
o
r
e
P
r
e
-
p
r
o
ce
s
s
in
g
(EP
B
),
E
d
g
e
P
ix
els
Af
ter
P
r
e
-
p
r
o
ce
s
s
i
n
g
(
E
P
A
)
an
d
No
is
e
R
ed
u
ce
d
R
atio
(
NR
R
)
.
T
h
e
o
r
ig
in
a
l
i
m
a
g
e
s
ar
e
u
s
ed
i
n
s
i
m
u
latio
n
a
n
d
its
g
r
o
u
n
d
tr
u
th
a
n
d
th
e
o
u
tp
u
t
o
f
s
t
an
d
ar
d
an
d
p
r
e
-
p
r
o
ce
s
s
ed
m
e
t
h
o
d
s
m
e
n
tio
n
ed
in
T
ab
le
2
.
EP
B
=
(
1
∑
∑
B
ij
M
j
=
1
N
i
=
1
)
∗
100
(
1
7
)
EP
A
=
(
1
∑
∑
A
ij
M
j
=
1
N
i
=
1
)
∗
100
(
1
8
)
Her
e
B
ij
an
d
A
ij
i
s
E
d
g
e
p
ix
el
id
en
ti
f
ied
b
e
f
o
r
e
an
d
Af
ter
p
r
e
-
p
r
o
ce
s
s
in
g
,
N
a
n
d
M
ar
e
r
o
w
an
d
clo
u
m
n
s
ize.
No
is
e
R
ed
u
ce
d
R
at
io
(
NR
R
)
=
E
P
A
-
EP
B
.
(
1
9
)
T
ab
le
2
.
C
o
m
p
ar
is
o
n
b
et
w
ee
n
th
e
p
r
o
p
o
s
ed
p
r
e
-
p
r
o
ce
s
s
ed
an
d
s
tan
d
ar
d
ed
g
e
d
etec
tio
n
m
et
h
o
d
s
Ed
g
e
d
e
t
e
c
t
i
o
n
m
e
t
h
o
d
S
t
a
n
d
a
r
d
Pre
-
p
r
o
c
e
ss
e
d
N
o
i
se
r
e
d
u
c
e
d
r
a
t
i
o
(
N
R
R
)
.
R
o
b
e
r
t
S
a
mp
l
e
i
mag
e
9
8
.
8
9
9
8
9
8
.
9
9
2
6
0
.
0
9
2
8
L
e
n
a
9
6
.
6
5
1
8
9
6
.
7
5
8
4
0
.
1
0
6
6
M
a
n
d
r
i
l
l
9
7
.
8
8
5
0
9
7
.
4
2
0
9
0
.
4
6
4
0
P
r
e
w
i
t
t
S
a
mp
l
e
i
mag
e
9
8
.
6
6
0
3
9
8
.
8
2
3
7
0
.
1
6
3
4
L
e
n
a
9
6
.
4
4
8
0
9
7
.
1
2
1
8
0
.
6
7
3
8
M
a
n
d
r
i
l
l
9
5
.
8
8
2
2
9
6
.
8
4
6
2
0
.
9
6
3
9
S
o
b
e
l
S
a
mp
l
e
i
mag
e
9
8
.
6
3
3
8
9
8
.
7
9
7
9
0
.
1
6
4
0
L
e
n
a
9
6
.
4
2
3
1
9
7
.
1
1
0
1
0
.
6
8
7
0
M
a
n
d
r
i
l
l
9
5
.
7
8
4
5
9
6
.
7
4
0
1
0
.
9
5
5
5
L
a
p
l
a
c
i
a
n
o
f
g
a
u
ss
i
a
n
(
L
o
G
)
S
a
mp
l
e
i
mag
e
9
6
.
8
3
2
2
9
7
.
0
7
1
1
0
.
2
3
8
9
L
e
n
a
9
4
.
1
7
2
8
9
4
.
5
4
4
8
0
.
3
7
1
9
M
a
n
d
r
i
l
l
8
8
.
7
2
3
7
9
0
.
2
0
2
9
1
.
4
7
9
1
C
a
n
n
y
S
a
mp
l
e
i
mag
e
9
5
.
2
5
6
1
9
5
.
5
8
3
3
0
.
3
2
7
1
L
e
n
a
9
1
.
9
7
6
3
9
3
.
1
9
7
9
1
.
2
2
1
6
M
a
n
d
r
i
l
l
8
4
.
4
6
1
8
8
6
.
9
5
7
6
2
.
4
9
5
8
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
20
88
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
5
,
Octo
b
er
2
0
1
7
:
2
5
7
4
–
2
5
8
0
2580
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
w
e
h
a
v
e
p
r
ese
n
t
ed
a
p
r
e
-
p
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
in
o
r
d
er
to
en
h
a
n
ce
t
h
e
p
er
f
o
r
m
an
ce
o
f
co
m
m
o
n
l
y
u
s
ed
ed
g
e
d
etec
tio
n
m
et
h
o
d
s
.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
co
n
s
is
ts
o
f
co
m
p
u
t
atio
n
o
f
th
e
m
ed
ia
n
f
ilter
i
n
g
to
r
ed
u
ce
t
h
e
n
o
is
e
i
n
i
m
a
g
e
a
n
d
t
h
en
i
m
a
g
e
s
e
g
m
en
tatio
n
is
ca
r
r
ied
o
u
t
.
Fin
a
ll
y
,
a
s
ta
n
d
ar
d
ed
g
e
d
etec
tio
n
m
et
h
o
d
ca
n
b
e
ap
p
lied
to
th
e
r
esu
l
tan
t
s
e
g
m
e
n
te
d
i
m
ag
e
s
.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
h
as
b
ee
n
s
u
m
m
ar
ized
in
s
tep
s
an
d
d
e
m
o
n
s
tr
ated
b
y
a
n
i
m
a
g
e
ex
a
m
p
l
e.
E
x
p
er
im
e
n
tal
r
es
u
lts
h
av
e
s
h
o
w
n
t
h
at
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
is
u
s
ed
w
ith
s
ta
n
d
ar
d
ed
g
e
d
etec
tio
n
m
e
th
o
d
s
to
en
h
a
n
ce
th
e
ir
p
er
f
o
r
m
an
ce
.
R
es
u
lts
al
s
o
s
h
o
w
n
th
at
ed
g
e
d
etec
tio
n
w
i
th
th
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
ap
p
r
o
ac
h
p
r
o
v
id
es
th
e
b
etter
p
er
f
o
r
m
an
ce
a
m
o
n
g
o
th
er
s
tan
d
ar
d
m
et
h
o
d
s
.
T
h
e
ac
cu
r
ac
y
i
s
in
cr
ea
s
e
w
h
e
n
w
e
ar
e
w
o
r
k
in
g
w
it
h
i
m
a
g
es
h
av
in
g
m
o
r
e
r
eg
io
n
a
l
d
if
f
er
e
n
ce
in
t
h
e
g
r
e
y
le
v
el.
6.
ACK
NO
WL
E
D
G
E
M
E
NT
Au
t
h
o
r
is
t
h
an
k
f
u
l
to
all
A
s
s
i
tan
t
P
r
o
f
ess
o
r
,
Dep
ar
t
m
e
n
t
t.
o
f
IT
,
f
o
r
g
iv
in
g
co
n
tin
u
es
s
u
p
p
o
r
t
an
d
en
co
u
r
ag
e
m
e
n
t
to
ca
r
r
y
o
u
t
t
h
is
w
o
r
k
.
A
u
th
o
r
s
ar
e
also
t
h
a
n
k
f
u
l
to
th
e
r
e
v
ie
w
er
f
o
r
cr
iticall
y
g
o
in
g
t
h
r
o
u
g
h
th
e
m
a
n
u
s
cr
ip
t a
n
d
g
iv
in
g
v
a
l
u
ab
le
s
u
g
g
esti
o
n
s
f
o
r
th
e
i
m
p
r
o
v
e
m
e
n
t o
f
m
a
n
u
s
cr
ip
t.
RE
F
E
R
E
NC
E
S
[1
]
M
u
k
e
sh
,
K.
a
n
d
Ro
h
in
i
,
S
.
A
lg
o
rit
h
m
a
n
d
T
e
c
h
n
iq
u
e
o
n
V
a
rio
u
s
Ed
g
e
De
tec
ti
o
n
:
A
S
u
rv
e
y
.
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
(
S
IPI
J
),
4
,
6
5
-
75
,
2
0
1
3
.
[2
]
W
.
X
.
Ka
n
g
,
Q.
Q.
Ya
n
g
,
R.
R.
L
ian
g
,
“
T
h
e
Co
m
p
a
ra
ti
v
e
Re
s
e
a
rc
h
o
n
Im
a
g
e
S
e
g
m
e
n
tatio
n
A
lg
o
rit
h
m
s”
,
IEE
E
Co
n
fer
e
n
c
e
o
n
ET
C
S
,
p
p
.
7
0
3
-
7
0
7
,
2
0
0
9
.
[3
]
Qu
rb
a
n
A
M
e
m
o
n
,
“
Em
b
e
d
d
in
g
A
u
th
e
n
ti
c
a
ti
o
n
a
n
d
Disto
r
ti
o
n
Co
n
c
e
a
lme
n
t
in
Im
a
g
e
s
-
A
No
isy
Ch
a
n
n
e
l
P
e
rsp
e
c
ti
v
e
”
In
te
rn
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
El
e
c
trica
l
En
g
i
n
e
e
rin
g
,
C
o
mp
u
ter
S
c
ien
c
e
a
n
d
In
f
o
rm
a
ti
c
s
(
EE
CS
I
2
0
1
4
)
,
Yo
g
y
a
k
a
rta,
In
d
o
n
e
sia
,
2
0
-
2
1
A
u
g
u
st 2
0
1
4
.
[4
]
Ra
fa
e
l
C.
G
o
n
z
a
lez
,
Rich
a
rd
E.
W
o
o
d
s,
S
tev
e
n
L
.
Ed
d
in
s,
“
Dig
it
a
l
I
m
a
g
e
P
ro
c
e
ss
in
g
Us
in
g
M
ATLA
B,
”
S
e
c
o
n
d
Ed
it
i
o
n
,
G
a
tes
m
a
rk
P
u
b
l
ish
i
n
g
,
2
0
0
9
.
[5
]
Ca
n
n
y
,
J.
A
Co
m
p
u
tatio
n
a
l
A
p
p
r
o
a
c
h
to
Ed
g
e
De
tec
ti
o
n
.
T
r
a
n
sa
c
ti
o
n
s
o
n
Pa
t
ter
n
An
a
lys
is
a
n
d
M
a
c
h
i
n
e
In
tellg
e
n
c
e
,
3
,
6
7
9
-
6
9
7
,
1
9
8
6
.
[6
]
M
.
Jo
g
e
n
d
ra
K
u
m
a
r
,
G
V
S
Ra
j
Ku
m
a
r
,
R.
V
ij
a
y
Ku
m
a
r
Re
d
d
y
“
Re
v
ie
w
On
I
m
a
g
e
S
e
g
m
e
n
tatio
n
T
e
c
h
n
iq
u
e
s
”
,
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
S
c
ien
ti
fi
c
Res
e
a
rc
h
En
g
in
e
e
rin
g
&
T
e
c
h
n
o
lo
g
y
(
IJ
S
RE
T
),
3
,
p
p
.
9
9
2
-
9
9
7
,
2
0
1
4
.
[7
]
M
u
a
m
m
e
r
C
a
tak
.
A
No
n
li
n
e
a
r
Dire
c
ti
o
n
a
l
De
riv
a
ti
v
e
S
c
h
e
m
e
f
o
r
Ed
g
e
De
tec
ti
o
n
.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
2
,
n
o
.
4
,
p
p
.
5
6
3
-
5
7
0
,
2
0
1
2
.
[8
]
Ra
n
i,
V.
a
n
d
S
h
a
rm
a
,
D.
A
S
tu
d
y
o
f
Ed
g
e
-
D
e
tec
ti
o
n
M
e
th
o
d
s.
I
n
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
S
c
ien
c
e
,
En
g
i
n
e
e
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
Res
e
a
rc
h
(
IJ
S
ET
R
)
,
1
,
6
2
-
65
,
2
0
1
2
.
[9
]
G
o
n
z
a
lez
,
R.
C.
a
n
d
W
o
o
d
s,
R.
E
.
Dig
it
a
l
Im
a
g
e
P
ro
c
e
ss
in
g
.
2
n
d
E
d
it
io
n
,
P
re
n
ti
c
e
Ha
ll
,
Up
p
e
r
S
a
d
d
le
Riv
e
r
,
2
0
0
2
.
[1
0
]
IKG
D
P
u
tra,
E
Erd
iaw
a
n
,
"
Hig
h
p
e
rf
o
rm
a
n
c
e
p
a
lm
p
rin
t
id
e
n
ti
f
ica
ti
o
n
sy
ste
m
b
a
se
d
o
n
tw
o
d
im
e
n
sio
n
a
l
g
a
b
o
r
,
”
T
EL
KOM
NIKA
T
e
lec
o
mm
u
n
ic
a
ti
o
n
C
o
mp
u
ti
n
g
El
e
c
tro
n
ics
a
n
d
Co
n
tro
l.
,
v
o
l.
8
,
n
o
.
3
,
p
p
.
3
0
9
-
3
1
8
,
2
0
1
0
.
[1
1
]
S
a
lu
ja,
S
.
,
S
i
n
g
h
,
A
.
K.
a
n
d
A
g
ra
w
a
l,
S
.
A
S
tu
d
y
o
f
Ed
g
e
-
D
e
tec
ti
o
n
M
e
th
o
d
s.
I
n
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Ad
v
a
n
c
e
d
Res
e
a
rc
h
in
C
o
mp
u
ter
a
n
d
C
o
mm
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
(
IJ
A
RCCE
),
2
,
9
9
4
-
9
9
9
,
2
0
1
3
.
[1
2
]
G
o
n
z
a
lez
&
W
o
o
d
s,
Dig
it
a
l
Im
a
g
e
P
r
o
c
e
ss
in
g
,
3
rd
e
d
it
i
o
n
,
P
re
n
ti
c
e
Ha
ll
In
d
ia,
2
0
0
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.