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.
11
,
No
.
5
,
Octo
b
er
2
0
2
1
,
p
p
.
4
4
8
9
~
4
5
0
1
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
1
i
5
.
p
p
4
4
8
9
-
4
5
0
1
4489
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
AH
P
va
lida
ted
li
t
erature r
ev
iew
of f
o
rg
ery typ
e de
pe
ndent
pa
ss
iv
e i
m
a
g
e f
o
r
g
ery detec
tion w
ith
ex
pla
ina
ble AI
K
a
ly
a
n
i K
a
da
m
1
,
Sw
a
t
i A
hir
ra
o
2
,
K
et
a
n K
o
t
ec
ha
3
1,
2
S
y
m
b
io
sis In
stit
u
te o
f
T
e
c
h
n
o
lo
g
y
,
S
y
m
b
io
sis In
tern
a
ti
o
n
a
l
(De
e
m
e
d
Un
iv
e
rsit
y
),
P
u
n
e
,
I
n
d
ia
3
He
a
d
S
CAA
I,
S
y
m
b
io
sis In
tern
a
ti
o
n
a
l
(De
e
m
e
d
Un
iv
e
rsit
y
)
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
2
6
,
2
0
2
0
R
ev
i
s
ed
Dec
2
8
,
2
0
2
0
A
cc
ep
ted
J
an
19
,
2
0
2
1
No
w
a
d
a
y
s,
a
lo
t
o
f
sig
n
i
f
ica
n
c
e
i
s
g
iv
e
n
to
w
h
a
t
w
e
re
a
d
to
d
a
y
:
n
e
ws
p
a
p
e
rs,
m
a
g
a
z
in
e
s,
n
e
w
s
c
h
a
n
n
e
ls,
a
n
d
in
tern
e
t
m
e
d
ia,
su
c
h
a
s
lea
d
i
n
g
so
c
ial
n
e
tw
o
rk
in
g
sites
li
k
e
F
a
c
e
b
o
o
k
,
In
sta
g
ra
m
,
a
n
d
Tw
it
ter
.
T
h
e
s
e
a
re
th
e
p
rim
a
r
y
w
e
ll
sp
rin
g
s
o
f
p
h
o
n
y
n
e
w
s
a
n
d
a
re
f
re
q
u
e
n
tl
y
u
ti
li
z
e
d
in
m
a
li
g
n
a
n
t
m
a
n
n
e
rs,
f
o
r
e
x
a
m
p
le,
f
o
r
h
o
rd
e
in
c
it
e
m
e
n
t.
In
t
h
e
re
c
e
n
t
d
e
c
a
d
e
,
a
trem
e
n
d
o
u
s
in
c
re
a
se
in
im
a
g
e
in
fo
rm
a
ti
o
n
g
e
n
e
ra
ti
o
n
is
h
a
p
p
e
n
i
n
g
d
u
e
to
th
e
m
a
ss
iv
e
u
se
o
f
so
c
ial
n
e
t
w
o
rk
in
g
se
r
v
ice
s.
V
a
rio
u
s
ima
g
e
e
d
it
in
g
so
f
t
wa
re
li
k
e
S
k
y
lu
m
L
u
m
in
a
r,
Co
re
l
P
a
in
tS
h
o
p
P
r
o
,
A
d
o
b
e
P
h
o
to
s
h
o
p
,
a
n
d
m
a
n
y
o
th
e
rs
a
re
u
se
d
to
c
re
a
te,
m
o
d
ify
th
e
i
m
a
g
e
s
a
n
d
v
id
e
o
s,
a
re
sig
n
if
ica
n
t
c
o
n
c
e
rn
s.
A
lo
t
o
f
e
a
rli
e
r
w
o
rk
o
f
f
o
rg
e
r
y
d
e
tec
ti
o
n
w
a
s
fo
c
u
se
d
o
n
trad
it
io
n
a
l
m
e
th
o
d
s
to
s
o
lv
e
th
e
f
o
rg
e
r
y
d
e
te
c
ti
o
n
.
Re
c
e
n
tl
y
,
De
e
p
lea
rn
i
n
g
a
lg
o
rit
h
m
s
h
a
v
e
a
c
c
o
m
p
li
sh
e
d
h
ig
h
-
p
e
rf
o
rm
a
n
c
e
a
c
c
u
ra
c
i
e
s
in
th
e
im
a
g
e
p
ro
c
e
ss
in
g
d
o
m
a
in
,
su
c
h
a
s
i
m
a
g
e
c
las
sif
i
c
a
ti
o
n
a
n
d
f
a
c
e
r
e
c
o
g
n
it
io
n
.
Ex
p
e
rts
h
a
v
e
a
p
p
li
e
d
d
e
e
p
lea
rn
in
g
tec
h
n
iq
u
e
s
to
d
e
tec
t
a
f
o
rg
e
r
y
in
th
e
im
a
g
e
to
o
.
Ho
w
e
v
e
r,
th
e
re
is
a
re
a
l
n
e
e
d
to
e
x
p
lain
w
h
y
th
e
ima
g
e
is
c
a
teg
o
rize
d
u
n
d
e
r
f
o
rg
e
d
t
o
u
n
d
e
rsta
n
d
th
e
a
lg
o
rit
h
m
’s
v
a
l
id
it
y
;
th
is
e
x
p
lan
a
ti
o
n
h
e
l
p
s
in
m
issio
n
-
c
rit
ica
l
a
p
p
li
c
a
ti
o
n
s
li
k
e
f
o
re
n
sic
.
E
x
p
lain
a
b
le
A
I
(X
A
I)
a
l
g
o
rit
h
m
s
h
a
v
e
b
e
e
n
u
se
d
to
in
terp
re
t
a
b
lac
k
b
o
x
’s
d
e
c
isio
n
in
v
a
rio
u
s
c
a
se
s.
T
h
is
p
a
p
e
r
c
o
n
tri
b
u
tes
a
su
rv
e
y
o
n
ima
g
e
f
o
rg
e
r
y
d
e
tec
ti
o
n
w
it
h
d
e
e
p
lea
rn
in
g
a
p
p
ro
a
c
h
e
s.
It
a
lso
f
o
c
u
se
s
o
n
th
e
su
rv
e
y
o
f
e
x
p
lain
a
b
le
A
I
f
o
r
ima
g
e
s.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
E
x
p
lain
ab
le
A
I
I
m
ag
e
f
o
r
g
er
y
d
etec
tio
n
I
m
ag
e
s
p
l
ic
in
g
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
:
Kal
y
a
n
i D
h
an
a
n
j
a
y
Kad
a
m
S
y
m
b
io
s
is
I
n
s
ti
tu
te
o
f
T
ec
h
n
o
lo
g
y
S
y
m
b
io
s
is
I
n
ter
n
atio
n
al
(
Dee
m
ed
Un
i
v
er
s
it
y
)
,
P
u
n
e,
I
n
d
ia
E
m
ail:
k
al
y
a
n
i
k
@
s
itp
u
n
e.
ed
u
.
in
1.
I
NT
RO
D
UCT
I
O
N
I
m
ag
e
s
ar
e
u
s
ed
i
n
al
m
o
s
t
e
v
er
y
f
ield
,
s
u
c
h
a
s
m
ed
ical
s
y
s
te
m
s
,
g
la
m
o
r
,
co
u
r
ts
,
m
ili
tar
y
,
an
d
in
d
u
s
tr
ies,
a
n
d
s
o
cial
n
et
w
o
r
k
in
g
p
lat
f
o
r
m
s
s
u
c
h
as
I
n
s
ta
g
r
a
m
,
Face
b
o
o
k
,
an
d
s
o
o
n
o
v
er
th
e
i
n
ter
n
et
[
1
]
.
R
eg
ar
d
les
s
o
f
w
h
eth
er
it i
s
t
h
e
s
p
ac
e
s
h
u
ttle e
x
p
lo
d
in
g
d
u
r
i
n
g
lau
n
c
h
,
a
m
a
n
s
tr
o
lli
n
g
o
n
th
e
m
o
o
n
,
o
r
o
f
f
icer
s
r
aisi
n
g
a
b
a
n
n
er
o
n
I
w
o
J
i
m
a
d
u
r
in
g
W
o
r
ld
W
ar
I
I
,
s
u
ch
g
r
o
u
n
d
-
b
r
ea
k
in
g
i
m
a
g
es
i
m
p
ac
t
th
e
s
o
ciet
y
[
1
]
.
T
h
e
ad
v
an
ce
m
en
t
i
n
d
i
g
ital
i
m
ag
i
n
g
s
o
f
t
w
ar
e
a
n
d
p
h
o
to
-
r
ea
li
s
t
ic
g
r
ap
h
ic
s
al
lo
w
s
p
eo
p
le
to
m
ak
e
i
m
a
g
es
m
o
r
e
r
ea
lis
tic
o
r
s
p
r
ea
d
alter
n
ativ
e
m
ea
n
in
g
s
.
I
m
ag
e
s
ca
n
b
e
f
u
s
ed
,
g
r
ap
h
ical
l
y
i
m
p
r
o
v
e
d
,
an
d
cr
ea
ted
b
y
co
m
p
u
ter
s
,
t
h
en
d
etec
ti
n
g
th
ese
co
n
tr
o
lled
i
m
a
g
es
ca
n
b
e
tr
o
u
b
leso
m
e.
T
h
e
r
ea
ln
ess
o
f
d
ig
ital
i
m
ag
e
s
b
ec
o
m
e
s
an
i
m
p
o
r
tan
t
s
t
u
d
y
ar
ea
f
o
r
r
esear
ch
a
n
d
d
ev
elo
p
m
en
t.
I
t
f
in
d
s
it
d
i
f
f
ic
u
lt
f
o
r
f
o
r
en
s
ic
ex
p
er
ts
to
id
en
ti
f
y
g
e
n
u
i
n
e
a
n
d
f
o
r
g
e
d
i
m
ag
e
s
.
I
d
en
ti
f
y
i
n
g
s
u
c
h
m
an
ip
u
lated
p
ar
ts
p
er
f
o
r
m
ed
b
y
t
h
e
f
a
k
er
’
s
m
an
ip
u
latio
n
o
p
er
atio
n
is
a
s
i
g
n
i
f
ica
n
t
w
o
r
k
.
Dete
cti
n
g
t
h
e
f
o
r
g
er
y
in
d
i
g
ital
i
m
ag
e
s
is
o
n
e
o
f
t
h
e
ch
a
llen
g
e
s
in
t
h
e
d
ig
ital
er
a.
No
w
ad
a
y
s
,
d
ee
p
lear
n
in
g
i
s
g
ai
n
i
n
g
m
o
r
e
atten
tio
n
d
u
e
to
its
s
i
g
n
i
f
i
ca
n
t
r
es
u
lts
.
Dee
p
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
5
,
Octo
b
e
r
2
0
2
1
:
4
4
8
9
-
4501
4490
L
ea
r
n
i
n
g
al
g
o
r
ith
m
s
h
a
v
e
ac
h
iev
ed
h
i
g
h
ac
c
u
r
ac
y
,
b
u
t
o
f
f
er
in
g
d
r
a
w
b
ac
k
s
,
as
th
e
i
m
p
o
r
ta
n
t
f
ea
t
u
r
es
ca
n
n
o
t
b
e
in
ter
p
r
eted
w
it
h
t
h
e
n
u
m
b
er
s
w
h
ic
h
ar
e
g
iv
e
n
b
y
d
ee
p
lear
n
in
g
m
o
d
el.
T
h
e
s
e
m
a
n
tic
s
is
also
n
o
t
ad
d
ed
.
T
h
ese
n
et
w
o
r
k
s
ar
e
g
iv
i
n
g
a
h
ig
h
er
p
er
f
o
r
m
an
ce
w
it
h
o
u
t
t
h
e
u
n
d
er
s
ta
n
d
in
g
o
f
it
s
i
n
s
id
e
w
o
r
k
in
g
.
B
lack
b
o
x
is
s
u
es
ar
e
n
o
t
h
i
n
g
b
u
t
in
v
e
s
t
ig
ati
n
g
th
e
i
n
n
er
w
o
r
k
in
g
o
f
a
d
ee
p
lear
n
in
g
m
o
d
el
a
n
d
in
ter
p
r
etin
g
w
h
y
i
t
d
eliv
er
ed
a
g
iv
e
n
y
ield
.
T
h
es
e
p
r
o
b
lem
s
b
ec
o
m
e
m
o
r
e
cr
i
tical
w
h
en
t
h
ese
n
et
w
o
r
k
s
ar
e
u
s
ed
in
r
ea
l
-
li
f
e
ap
p
licatio
n
s
s
u
c
h
as
m
ed
ical
d
iag
n
o
s
i
s
o
r
f
o
r
en
s
ic.
E
x
p
lai
n
ab
le
A
I
e
x
p
lain
s
t
h
e
d
ec
is
io
n
s
o
f
a
n
et
w
o
r
k
.
I
n
co
m
p
u
ter
v
is
io
n
,
v
i
s
u
a
lizatio
n
n
et
w
o
r
k
s
p
r
o
v
id
e
f
a
s
cin
a
tin
g
w
a
y
s
to
v
i
s
u
a
lize
t
h
e
i
m
ag
e
’
s
ess
e
n
tia
l
f
ea
tu
r
es
,
in
ter
p
r
etin
g
t
h
e
i
m
ag
e.
I
n
t
h
is
ca
s
e,
th
e
i
n
p
u
t
d
ata
is
t
h
e
i
m
ag
e,
an
d
i
m
a
g
e
in
ter
p
r
etatio
n
is
p
er
f
o
r
m
ed
u
s
i
n
g
h
ea
t
m
ap
s
.
Hea
t
m
ap
s
g
i
v
e
t
h
e
m
o
s
t
s
i
g
n
i
f
ican
t
i
n
p
u
t
d
ata
ar
ea
s
f
o
r
t
h
e
i
n
ter
p
r
etatio
n
.
I
t
allo
w
s
t
h
e
u
s
er
to
u
n
d
er
s
ta
n
d
w
h
ich
i
m
a
g
e
p
ix
e
ls
ar
e
ass
o
ciate
d
w
it
h
th
e
e
x
p
ec
ted
class
,
an
d
it
also
ch
ec
k
s
w
h
et
h
er
th
e
d
ee
p
lear
n
in
g
m
o
d
el
f
o
cu
s
es
o
n
th
e
i
m
a
g
e
’
s
r
ea
s
o
n
ab
le
ar
ea
.
Fi
g
u
r
e
1
s
h
o
w
s
th
e
m
in
d
m
ap
o
f
t
h
e
ar
ticle.
T
h
e
ar
ticle
is
m
ai
n
l
y
d
iv
id
ed
in
to
d
ig
ital
i
m
a
g
e
f
o
r
g
er
y
d
etec
ti
o
n
,
k
e
y
w
o
r
d
an
al
y
s
i
s
u
s
in
g
v
ar
io
u
s
to
o
ls
,
d
ee
p
lear
n
in
g
f
o
r
h
an
d
li
n
g
f
o
r
g
er
y
t
y
p
e
-
d
ep
en
d
e
n
t
f
o
r
g
er
ies,
A
HP
m
o
d
el
f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
,
a
n
d
ex
p
lain
ab
le
A
I
.
Fig
u
r
e
1
.
W
o
r
k
f
l
o
w
o
f
p
ass
i
v
e
i
m
ag
e
f
o
r
g
er
y
d
etec
tio
n
w
it
h
a
f
o
cu
s
o
n
ex
p
lai
n
ab
le
AI
2.
DIGIT
AL
I
M
AG
E
F
O
RG
E
RY
DE
T
E
C
T
I
O
N
I
m
ag
e
f
o
r
g
er
y
d
etec
tio
n
i
s
cla
s
s
i
f
ied
i
n
to
t
w
o
ap
p
r
o
ac
h
es
[
1
]
:
p
ass
i
v
e
o
r
b
li
n
d
an
d
ac
tiv
e
ap
p
r
o
ac
h
.
Var
io
u
s
t
y
p
e
s
o
f
i
m
ag
e
f
o
r
g
er
i
es a
r
e
s
h
o
w
n
i
n
Fi
g
u
r
e
2.
Fig
u
r
e
2
.
Dig
ital i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
HP
va
lid
a
ted
liter
a
tu
r
e
r
ev
iew
o
f fo
r
g
ery
typ
e
d
ep
en
d
e
n
t
p
a
s
s
ive
ima
g
e
fo
r
g
ery
…
(
K
a
lya
n
i Ka
d
a
m
)
4491
2
.
1
.
Act
iv
e
a
pp
ro
a
ch
I
n
th
is
ap
p
r
o
ac
h
,
a
co
m
p
u
ter
iz
ed
s
ig
n
at
u
r
e
o
r
w
a
ter
m
ar
k
is
in
j
ec
ted
in
an
i
m
ag
e,
an
d
th
a
t
is
in
s
er
te
d
eith
er
b
y
an
in
d
i
v
id
u
a
l
o
r
b
y
t
h
e
ac
q
u
i
s
itio
n
d
ev
ice.
Dig
ital
w
ater
m
ar
k
in
g
in
s
er
ts
p
ar
ticu
lar
in
f
o
r
m
at
io
n
k
n
o
w
n
as
a
m
es
s
ag
e
d
i
g
est
i
n
s
id
e
th
e
i
m
a
g
e
w
h
i
le
ca
p
tu
r
i
n
g
it.
T
h
is
i
n
f
o
r
m
a
tio
n
i
s
th
e
n
tak
en
o
u
t
f
r
o
m
t
h
e
i
m
a
g
e
at
later
p
h
ases
to
g
et
its
au
th
e
n
ticit
y
.
T
h
is
ex
tr
icate
d
i
n
f
o
r
m
atio
n
i
s
ch
ec
k
ed
to
s
ee
if
it
v
ar
ies
o
r
n
o
t;
if
it
ch
a
n
g
es,
it
s
h
o
w
s
th
at
t
h
e
i
m
ag
e
w
as
alter
ed
af
ter
th
e
i
m
ag
e
ca
p
tu
r
i
n
g
p
r
o
ce
d
u
r
e.
I
t
is
a
t
w
o
-
p
h
a
s
e
p
r
o
ce
s
s
,
in
t
h
e
f
ir
s
t
p
h
a
s
e,
th
e
m
ess
a
g
e
-
d
ig
e
s
t
i
s
e
m
b
ed
d
ed
i
n
t
h
e
i
m
ag
e.
I
n
t
h
e
s
ec
o
n
d
p
h
a
s
e,
i.
e.
,
af
ter
r
ea
ch
i
n
g
t
h
e
i
m
a
g
e
to
it
s
s
d
esti
n
atio
n
,
th
e
m
e
s
s
a
g
e
d
ig
e
s
t
i
s
r
etr
ie
v
ed
an
d
co
m
p
ar
ed
w
it
h
t
h
e
o
b
tai
n
ed
w
ater
m
ar
k
[
1
]
.
Dig
ital
w
ater
m
ar
k
i
n
g
i
s
a
n
ef
f
ec
tiv
e
m
et
h
o
d
to
s
ec
u
r
e
th
e
i
m
a
g
e
tr
u
s
t
w
o
r
t
h
i
n
es
s
;
h
o
w
ev
er
,
d
i
f
f
er
en
t
d
if
f
ic
u
lt
ies
m
a
k
e
its
u
t
ilizatio
n
u
n
r
ea
li
s
tic.
Fe
w
ca
m
er
as
an
d
g
ad
g
ets
h
a
v
e
t
h
e
p
r
o
p
er
ty
to
in
s
er
t
w
ater
m
ar
k
i
n
th
e
i
m
a
g
e
at
th
e
ti
m
e
o
f
cr
ea
tio
n
.
So
m
e
t
y
p
es
o
f
eq
u
ip
m
en
t,
e.
g
.
,
C
an
o
n
E
OS
-
1
D,
Nik
o
n
D2
Xs,
h
a
v
e
e
m
b
ed
d
in
g
f
ea
t
u
r
es
b
u
t
ar
e
o
v
er
p
r
iced
.
T
h
e
d
is
ad
v
an
tag
e
i
s
th
at
th
i
s
m
et
h
o
d
ca
n
n
o
t
d
i
f
f
er
en
tiate
le
g
iti
m
at
e
an
d
in
v
alid
o
p
er
atio
n
s
in
t
h
e
i
m
ag
e.
L
e
g
iti
m
ate
o
p
er
atio
n
s
ar
e
p
er
f
o
r
m
ed
f
o
r
i
m
p
r
o
v
i
n
g
i
m
ag
e
q
u
alit
y
,
s
u
ch
as
s
h
ar
p
en
i
n
g
,
co
n
tr
ast
i
m
p
r
o
v
e
m
e
n
t,
an
d
s
o
o
n
.
I
n
th
e
ca
s
e
o
f
w
ater
m
ar
k
i
n
g
s
p
ec
ial
s
o
f
t
w
ar
e
o
r
h
ar
d
w
ar
e
is
n
ee
d
ed
to
in
s
er
t
m
es
s
a
g
e
d
ig
e
s
t
in
an
i
m
a
g
e.
I
m
a
g
e
f
o
r
g
er
y
is
id
en
t
if
ie
d
w
i
th
s
tati
s
tical
i
n
f
o
r
m
atio
n
.
T
h
is
i
s
d
o
n
e
b
y
d
iv
id
in
g
th
e
i
m
a
g
e
i
n
to
s
i
m
ilar
r
eg
io
n
ch
ar
ac
ter
is
t
ics;
af
ter
t
h
is
,
s
tatis
tical
m
ea
s
u
r
es
s
u
ch
a
s
m
ea
n
,
m
o
d
e,
m
ed
ia
n
,
a
n
d
r
an
g
e
o
f
p
ix
el
v
al
u
es
ar
e
i
n
s
er
ted
i
n
t
h
e
i
m
a
g
e
w
it
h
an
en
cr
y
p
t
io
n
m
et
h
o
d
[
1
]
.
T
h
en
th
i
s
in
f
o
r
m
atio
n
i
s
v
er
i
f
ied
to
ch
ec
k
its
a
u
t
h
en
t
icit
y
.
2
.
2
.
P
a
s
s
iv
e
a
pp
ro
a
ch
P
ass
iv
e
m
eth
o
d
o
lo
g
ies
[
1
]
u
s
es
s
tatis
tical
in
f
o
r
m
a
tio
n
o
f
t
h
e
i
m
ag
e
to
d
etec
t
f
o
r
g
er
ies
i
n
an
i
m
a
g
e.
T
h
ese
m
et
h
o
d
s
w
o
r
k
w
ith
o
u
t
ea
r
lier
in
f
o
r
m
atio
n
ab
o
u
t
t
h
e
i
m
a
g
e,
f
o
r
ex
a
m
p
le,
d
i
g
i
tal
w
ater
m
ar
k
s
o
r
s
ig
n
at
u
r
es.
T
h
is
m
eth
o
d
u
s
e
s
th
e
e
x
is
t
in
g
i
m
ag
e,
a
n
d
t
h
e
i
m
ag
e
’
s
m
a
n
ip
u
la
tio
n
o
p
er
atio
n
ch
a
n
g
e
s
its
p
r
o
p
er
ties
.
I
t
w
ill
m
a
k
e
t
h
e
i
m
ag
e
in
co
n
s
i
s
ten
t,
an
d
th
u
s
th
e
s
tatis
t
ical
i
m
ag
e
i
n
f
o
r
m
atio
n
is
u
s
ed
f
o
r
f
o
r
g
er
y
id
en
ti
f
icatio
n
.
I
t is
f
u
r
th
er
clas
s
if
ied
i
n
to
f
o
r
g
er
y
t
y
p
e
i
n
d
ep
en
d
en
t a
n
d
d
ep
en
d
en
t.
2
.
2
.
1
.
F
o
rg
er
y
t
y
pe
ind
epen
d
ent
T
h
ese
tech
n
iq
u
e
s
ar
e
u
s
ed
to
i
d
en
tify
t
h
e
f
o
r
g
er
ie
s
b
ased
o
n
ar
tif
ac
t
cl
u
es
le
f
t
d
u
r
i
n
g
r
eto
u
ch
in
g
a
n
d
lig
h
ti
n
g
co
n
d
itio
n
s
[
1
]
.
T
h
is
ty
p
e
o
f
f
o
r
g
er
y
u
s
e
s
s
tat
is
t
ical
d
ata
o
f
th
e
i
m
a
g
e,
an
d
s
u
c
h
i
n
v
is
ib
le
i
n
f
o
r
m
atio
n
is
m
is
u
s
ed
in
d
i
f
f
er
en
t
i
m
a
g
e
h
an
d
li
n
g
ac
ti
v
ities
,
e.
g
.
,
j
p
eg
co
n
f
i
n
i
n
g
,
i
m
a
g
e
f
ilter
i
n
g
,
co
n
t
r
ast
i
m
p
r
o
v
e
m
en
ts
,
an
d
r
esa
m
p
li
n
g
.
T
h
e
r
esa
m
p
li
n
g
p
r
o
ce
s
s
ch
a
n
g
es
t
h
e
i
m
a
g
e
b
y
p
er
f
o
r
m
i
n
g
u
p
s
a
m
p
li
n
g
an
d
d
o
w
n
s
a
m
p
li
n
g
o
p
er
atio
n
s
an
d
to
o
ls
h
an
d
li
n
g
s
u
c
h
f
o
r
g
er
ies
w
o
r
k
s
o
n
e
ith
er
p
ix
el
o
r
f
r
eq
u
e
n
c
y
d
o
m
ai
n
.
T
h
e
m
ed
ian
f
ilter
i
n
g
to
o
l
is
m
o
s
tl
y
u
s
ed
f
o
r
n
o
is
e
eli
m
in
at
io
n
a
n
d
i
m
a
g
e
i
m
p
r
o
v
e
m
en
t.
T
h
ese
m
e
th
o
d
s
g
iv
e
e
x
ce
lle
n
t
p
er
f
o
r
m
a
n
ce
,
w
h
ile
th
e
i
m
a
g
e
s
ar
e
u
n
co
m
p
r
e
s
s
ed
an
d
lar
g
e.
C
o
m
m
o
n
l
y
,
th
e
i
m
ag
e
s
ar
e
co
m
p
r
es
s
ed
w
i
th
t
h
e
J
P
E
G
c
o
m
p
r
ess
io
n
s
tr
ate
g
y
.
T
h
e
m
ai
n
g
o
al
o
f
J
P
E
G
-
b
ased
f
o
r
g
er
y
id
en
ti
f
icatio
n
tech
n
iq
u
es
is
to
f
i
n
d
o
u
t
th
e
lo
ca
tio
n
s
o
f
i
m
a
g
es
w
it
h
d
if
f
er
en
t J
P
E
G
co
m
p
r
ess
io
n
lo
ca
tio
n
s
.
Fo
r
g
er
y
t
y
p
e
d
ep
en
d
en
t
T
h
is
t
y
p
e
o
f
f
o
r
g
er
y
f
o
cu
s
es
o
n
l
y
o
n
s
p
ec
if
ic
k
i
n
d
s
o
f
f
o
r
g
er
ies
s
u
ch
as
co
p
y
m
o
v
e
an
d
i
m
ag
e
s
p
licin
g
; t
h
ese
ar
e
g
ai
n
i
n
g
m
o
r
e
atten
tio
n
as t
h
e
y
ar
e
co
m
m
o
n
l
y
p
er
f
o
r
m
ed
m
an
ip
u
latio
n
s
o
n
th
e
i
m
ag
e.
a.
C
o
p
y
m
o
v
e
T
h
is
t
y
p
e
o
f
f
o
r
g
er
y
co
p
ies
p
ar
t
o
f
th
e
i
m
ag
e
a
n
d
p
asted
in
to
th
e
s
a
m
e
i
m
a
g
e
at
a
n
o
th
er
r
eg
io
n
,
w
h
ic
h
en
tire
l
y
ch
a
n
g
es
its
m
e
an
in
g
.
T
h
is
is
u
s
u
all
y
d
o
n
e
to
m
ak
e
a
n
o
b
j
ec
t
“
d
is
ap
p
ea
r
”
f
r
o
m
t
h
e
i
m
a
g
e
b
y
h
id
in
g
it
w
it
h
a
s
eg
m
e
n
t
co
p
ie
d
f
r
o
m
a
n
o
th
er
p
ar
t
o
f
th
e
i
m
a
g
e.
T
h
is
k
i
n
d
o
f
f
o
r
g
er
y
is
v
er
y
ea
s
y
to
p
er
f
o
r
m
,
b
u
t
it
i
s
v
er
y
h
ar
d
to
d
etec
t
a
s
th
e
co
p
ied
s
ec
tio
n
co
m
e
s
f
r
o
m
t
h
e
s
a
m
e
i
m
ag
e.
I
t
s
d
i
f
f
er
e
n
t
p
r
o
p
er
ties
,
s
u
c
h
as
s
h
ad
e
p
alette,
n
o
i
s
e
co
m
p
o
n
e
n
t,
an
d
o
th
er
c
h
ar
ac
ter
is
tic
s
,
a
r
e
s
u
itab
le
to
t
h
e
r
est
o
f
t
h
e
i
m
ag
e,
w
h
i
c
h
w
i
l
l
n
o
t
b
e
r
e
c
o
g
n
i
z
e
d
v
i
a
t
h
e
s
a
m
e
m
e
t
h
o
d
s
t
h
a
t
f
i
n
d
i
n
c
o
n
s
i
s
t
e
n
c
i
e
s
i
n
s
t
a
t
i
s
t
i
c
a
l
m
e
a
s
u
r
e
s
i
n
a
n
o
t
h
e
r
p
o
r
t
i
o
n
o
f
t
h
e
i
m
ag
e.
Fig
u
r
es
3
a
n
d
4
s
h
o
w
t
h
at
n
u
m
er
o
u
s
p
ap
er
s
ar
e
p
u
b
li
s
h
e
d
in
t
h
is
ex
p
lo
r
ati
o
n
ar
ea
f
r
o
m
v
ar
io
u
s
co
u
n
tr
ies.
Data
an
al
y
s
is
is
p
er
f
o
r
m
ed
b
y
u
s
i
n
g
q
u
er
ie
s
s
u
c
h
as
‘
co
p
y
m
o
v
e
f
o
r
g
er
y
’
a
n
d
‘
co
p
y
m
o
v
e
f
o
r
g
er
y
d
etec
tio
n
u
s
i
n
g
d
ee
p
lear
n
i
n
g
’
o
n
I
E
E
E
(
ieee
x
p
lo
r
e.
o
r
g
)
,
Sco
p
u
s
(
s
co
p
u
s
.
co
m
)
,
W
OS
(
web
o
f
s
cie
n
ce
)
,
an
d
AC
M
d
atab
ases
.
T
h
e
d
u
r
atio
n
f
r
o
m
2
0
0
1
to
2
0
2
0
is
co
n
s
id
e
r
ed
f
o
r
g
ett
in
g
t
h
e
r
es
u
lt.
T
h
r
ee
d
if
f
er
en
t
s
tep
s
ar
e
p
er
f
o
r
m
ed
f
o
r
co
p
y
m
o
v
e
f
o
r
g
er
y
d
etec
tio
n
(
C
MFD)
:
s
e
p
ar
atin
g
th
e
i
m
a
g
e
in
to
b
lo
ck
s
(
o
v
er
l
y
i
n
g
/
n
o
n
-
o
v
er
l
y
i
n
g
)
,
e
x
ec
u
tin
g
th
e
i
m
ag
e
p
r
o
p
er
ty
e
x
tr
ac
tio
n
tec
h
n
iq
u
e
o
n
ea
c
h
b
lo
ck
,
co
m
p
ar
i
s
o
n
o
f
f
ea
tu
r
es.
I
n
C
MFD
m
et
h
o
d
s
[
1
]
,
th
e
p
r
in
cip
al
o
b
j
ec
tiv
e
is
to
co
m
p
ar
e
th
e
p
o
r
tio
n
s
i
n
s
id
e
a
n
i
m
ag
e
an
d
r
ec
o
g
n
ize
t
h
e
r
ep
licated
an
d
th
e
o
r
ig
in
al
r
e
g
i
o
n
.
b.
I
m
ag
e
s
p
lic
in
g
I
m
ag
e
co
m
p
o
s
iti
n
g
o
r
s
p
lici
n
g
j
o
in
s
m
o
r
e
t
h
a
n
t
w
o
i
m
ag
e
s
to
cr
ea
te
a
f
o
r
g
ed
i
m
a
g
e.
I
t
c
an
b
e
d
o
n
e
by
v
ar
io
u
s
s
o
f
t
w
ar
e
to
o
ls
s
u
ch
as
A
d
o
b
e
P
h
o
to
s
h
o
p
,
an
d
C
o
r
al
P
ain
tS
h
o
p
.
T
h
e
co
p
y
m
o
v
e
d
etec
tio
n
m
e
th
o
d
s
ca
n
n
o
t
b
e
u
s
ed
i
n
s
p
lic
in
g
a
s
it
w
o
r
k
s
o
n
m
u
ltip
le
i
m
a
g
es
;
th
e
m
an
ip
u
lated
p
ar
t
o
f
t
h
e
i
m
a
g
e
h
a
s
d
is
ti
n
ct
p
r
o
p
er
ties
c
o
m
p
ar
ed
w
it
h
t
h
e
r
est
o
f
th
e
i
m
a
g
e.
T
h
is
m
et
h
o
d
u
s
es
n
u
m
er
o
u
s
p
r
o
p
er
ties
s
u
c
h
as
DC
T
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
5
,
Octo
b
e
r
2
0
2
1
:
4
4
8
9
-
4501
4492
DW
T
co
ef
f
icien
t
s
,
B
i
-
co
h
er
e
n
ce
p
r
o
p
er
ties
,
ca
m
er
a
r
esp
o
n
s
e
o
p
er
atio
n
,
an
d
i
n
v
ar
ia
n
t
i
m
ag
e
m
o
m
e
n
ts
,
T
h
e
s
p
licin
g
m
et
h
o
d
s
co
n
s
id
er
th
e
g
ad
g
et
s
i
n
b
o
r
n
p
r
o
p
er
ties
in
s
tead
o
f
ca
p
tu
r
i
n
g
th
e
g
ad
g
et
’
s
ac
tu
a
l
att
r
ib
u
te
s
.
Fig
u
r
es
5
a
n
d
6
s
h
o
w
t
h
at
n
u
m
er
o
u
s
p
ap
er
s
ar
e
p
u
b
lis
h
e
d
in
t
h
is
e
x
p
lo
r
atio
n
ar
ea
f
r
o
m
v
ar
io
u
s
co
u
n
tr
ies.
Data
an
al
y
s
is
i
s
p
er
f
o
r
m
ed
b
y
u
s
i
n
g
q
u
er
ies
s
u
c
h
a
s
‘
i
m
a
g
e
s
p
lici
n
g
’
an
d
‘
i
m
ag
e
s
p
lic
in
g
d
etec
tio
n
u
s
i
n
g
d
ee
p
lear
n
in
g
’
o
n
I
E
E
E
(
iee
ex
p
lo
r
e.
o
r
g
)
,
Sco
p
u
s
(
s
co
p
u
s
.
co
m
)
,
W
OS(
W
eb
o
f
Scien
ce
)
,
an
d
A
C
M.
T
h
e
d
u
r
atio
n
f
r
o
m
2
0
0
1
to
2
0
2
0
is
co
n
s
id
er
ed
f
o
r
g
etti
n
g
t
h
e
r
es
u
lt.
Fig
u
r
e
3
.
L
ast 1
9
y
ea
r
s
o
f
p
u
b
licatio
n
d
ata
f
o
r
‘
co
p
y
m
o
v
e
f
o
r
g
er
y
’
i
n
s
co
p
u
s
,
I
E
E
E
,
W
OS (
w
eb
o
f
s
cien
ce
)
,
an
d
AC
M
Fig
u
r
e
4
.
P
ast 1
9
y
ea
r
s
p
u
b
lic
atio
n
co
u
n
t f
o
r
‘
co
p
y
m
o
v
e
f
o
r
g
er
y
d
etec
tio
n
u
s
in
g
d
ee
p
lear
n
in
g
’
i
n
Sco
p
u
s
,
I
E
E
E
,
W
OS (
w
eb
o
f
s
cien
ce
)
,
an
d
A
C
M
Fig
u
r
e
5
.
P
ast 1
9
y
ea
r
s
o
f
th
e
p
u
b
licatio
n
i
n
f
o
r
m
atio
n
f
o
r
‘
i
m
ag
e
s
p
licin
g
’
i
n
Sco
p
u
s
,
I
E
E
E
,
W
OS (
w
eb
o
f
Scien
ce
)
,
an
d
A
C
M
Fig
u
r
e
6
.
L
ast 1
9
y
ea
r
s
o
f
p
u
b
licatio
n
i
n
f
o
r
m
atio
n
o
n
‘
i
m
a
g
e
s
p
lici
n
g
d
etec
tio
n
u
s
in
g
d
ee
p
lear
n
in
g
’
in
Sco
p
u
s
,
I
E
E
E
,
W
OS (
w
eb
o
f
s
cien
ce
)
,
an
d
A
C
M
3.
K
E
YWO
RD
AN
AL
Y
SI
S U
S
I
NG
WO
RD
CL
O
U
D,
SCI
E
NSCAP
E
A
ND
L
E
XIM
ANC
E
R
T
O
O
L
Fig
u
r
es
7
-
9
i
n
d
icate
t
h
e
to
p
3
p
ap
er
’
s
w
o
r
d
clo
u
d
s
in
a
d
e
ep
lear
n
in
g
ap
p
r
o
ac
h
f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
an
d
e
x
p
lain
ab
le
A
I
[
2
]
.
W
o
r
d
clo
u
d
h
ig
h
lig
h
t
s
i
m
p
o
r
ta
n
t
w
o
r
d
s
i
n
t
h
is
r
e
s
ea
r
ch
f
i
e
l
d
.
T
h
e
w
o
r
d
s
w
i
t
h
s
m
a
l
l
e
r
f
o
n
t
s
a
r
e
a
n
i
n
d
i
c
a
t
i
o
n
o
f
f
u
t
u
r
e
r
e
s
e
a
r
c
h
.
A
l
l
t
h
e
w
o
r
d
c
l
o
u
d
s
a
r
e
d
r
aw
n
f
r
o
m
w
w
w
.
w
o
r
d
c
l
o
u
d
s
.
c
o
m
.
I
m
p
o
r
t
a
n
t
w
o
r
d
s
a
r
e
c
o
p
y
m
o
v
e
,
s
p
l
i
c
i
n
g
,
d
e
e
p
l
e
a
r
n
i
n
g
,
f
o
r
g
e
r
y
d
e
t
e
c
t
i
o
n
,
lo
ca
lizatio
n
o
f
f
o
r
g
er
y
,
s
e
g
m
en
tatio
n
,
class
i
f
icatio
n
,
v
is
u
aliza
tio
n
,
e
x
p
lain
ab
le,
a
n
d
X
A
I
.
L
e
s
s
i
m
p
o
r
ta
n
t
w
o
r
d
s
ar
e
i
m
a
g
e
ta
m
p
er
i
n
g
d
etec
tio
n
,
ta
m
p
er
ed
,
m
a
n
ip
u
latio
n
,
b
lo
ck
s
,
an
d
.
Fig
u
r
e
s
1
0
an
d
1
1
ar
e
d
r
a
w
n
w
it
h
th
e
Scie
n
Scap
e
to
o
l,
w
h
ic
h
s
h
o
w
s
th
e
to
p
k
e
y
w
o
r
d
s
p
er
y
ea
r
f
r
o
m
S
co
p
u
s
a
n
d
th
e
W
OS
d
atab
ase.
I
n
Fi
g
u
r
e
1
2
,
th
e
th
e
m
at
ic
g
r
ap
h
d
r
a
w
n
u
s
in
g
th
e
L
e
x
i
m
an
ce
r
to
o
l u
s
e
s
a
m
ac
h
i
n
e
l
ea
r
n
i
n
g
tech
n
iq
u
e
t
h
at
f
i
n
d
s
th
e
r
elatio
n
b
et
w
ee
n
d
if
f
er
e
n
t a
u
th
o
r
k
e
y
w
o
r
d
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
HP
va
lid
a
ted
liter
a
tu
r
e
r
ev
iew
o
f fo
r
g
ery
typ
e
d
ep
en
d
e
n
t
p
a
s
s
ive
ima
g
e
fo
r
g
ery
…
(
K
a
lya
n
i Ka
d
a
m
)
4493
Fig
u
r
e
7
.
W
o
r
d
clo
u
d
f
o
r
“
a
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
to
d
etec
tio
n
o
f
s
p
licin
g
a
n
d
co
p
y
-
m
o
v
e
f
o
r
g
er
ies i
n
i
m
a
g
es
Fig
u
r
e
8
.
W
o
r
d
clo
u
d
f
o
r
“
i
m
a
g
e
r
eg
io
n
f
o
r
g
er
y
d
etec
tio
n
:
A
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
Fig
u
r
e
9
.
W
o
r
d
clo
u
d
f
o
r
“
v
i
s
u
al
in
ter
p
r
etab
ilit
y
f
o
r
d
ee
p
lear
n
in
g
: a
s
u
r
v
e
y
∗
”
Fig
u
r
e
1
0
.
Key
w
o
r
d
s
p
er
y
ea
r
(
to
p
k
ey
w
o
r
d
s
p
er
y
ea
r
)
f
o
r
s
c
o
p
u
s
d
atab
ase
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
5
,
Octo
b
e
r
2
0
2
1
:
4
4
8
9
-
4501
4494
Fig
u
r
e
1
1
.
Key
w
o
r
d
s
p
er
y
e
ar
(
to
p
k
e
y
w
o
rd
s
p
er
y
ea
r
)
f
o
r
W
OS
d
atab
ase
Fig
u
r
e
1
2
.
T
h
e
m
atic
d
iag
r
a
m
f
o
r
k
e
y
w
o
r
d
s
4.
DE
E
P
L
E
ARNI
NG
AP
P
R
O
ACH
F
O
R
CO
P
Y
M
O
V
E
AND
I
M
AG
E
SP
L
I
C
I
N
G
F
O
R
G
E
R
Y
DE
T
E
C
T
I
O
N
C
o
n
v
Nets
o
r
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
et
w
o
r
k
s
(
C
NN
s
)
[
3
]
is
o
n
e
o
f
t
h
e
m
ai
n
ca
te
g
o
r
ies
f
o
r
i
m
ag
e
r
ec
o
g
n
itio
n
an
d
class
i
f
icatio
n
s
in
n
eu
r
al
n
et
w
o
r
k
s
.
Face
r
ec
o
g
n
i
tio
n
,
an
d
d
etec
tio
n
o
f
o
b
j
e
cts
in
an
i
m
ag
e
ar
e
s
o
m
e
o
f
th
e
f
ie
ld
s
w
h
er
e
C
N
Ns
ar
e
ex
te
n
s
iv
el
y
u
s
ed
.
C
N
N
is
u
s
e
f
u
l
f
o
r
r
etr
iev
in
g
m
e
an
in
g
f
u
l
f
ea
tu
r
e
s
f
o
r
i
m
a
g
e
cla
s
s
i
f
icatio
n
.
T
h
is
m
o
d
el
[
3
]
u
s
es
C
N
N
f
o
r
id
en
ti
f
y
in
g
co
p
y
m
o
v
e
an
d
i
m
a
g
e
s
p
l
icin
g
f
o
r
g
er
y
in
a
n
i
m
a
g
e.
T
h
is
n
et
w
o
r
k
i
s
p
r
etr
ain
ed
w
it
h
t
h
e
h
elp
o
f
lab
eled
im
ag
e
s
f
r
o
m
t
h
e
tr
ain
in
g
i
m
ag
e
d
ataset.
T
h
e
s
a
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
HP
va
lid
a
ted
liter
a
tu
r
e
r
ev
iew
o
f fo
r
g
ery
typ
e
d
ep
en
d
e
n
t
p
a
s
s
ive
ima
g
e
fo
r
g
ery
…
(
K
a
lya
n
i Ka
d
a
m
)
4495
n
et
w
o
r
k
i
s
th
e
n
u
s
ed
f
o
r
ex
tr
a
ctin
g
th
e
f
ea
t
u
r
es
f
o
r
th
e
p
atc
h
es,
an
d
t
h
ese
f
ea
t
u
r
es
ar
e
co
m
b
in
ed
to
tr
ain
t
h
e
SVM
m
o
d
el.
I
n
[
4
]
,
a
t
w
o
-
p
h
a
s
e
d
ee
p
lear
n
i
n
g
ap
p
r
o
ac
h
is
e
m
p
lo
y
ed
to
lear
n
t
h
e
f
ea
t
u
r
es
f
o
r
d
etec
tin
g
i
m
a
g
e
f
o
r
g
er
y
f
o
r
a
n
i
m
a
g
e
t
h
at
co
m
es
in
v
ar
io
u
s
f
o
r
m
a
ts
.
I
n
t
h
e
f
ir
s
t
s
tag
e,
S
tac
k
A
UT
OE
NC
ODE
R
is
u
s
ed
to
lear
n
co
m
p
le
x
f
ea
t
u
r
es
f
r
o
m
e
ac
h
p
atch
.
I
n
t
h
e
s
ec
o
n
d
s
tag
e,
th
e
i
m
a
g
e
i
s
b
r
o
k
e
n
i
n
to
p
atc
h
es o
f
3
2
x
3
2
.
Fro
m
ea
ch
p
atch
,
t
h
e
co
n
te
x
tu
a
l
d
ata
is
u
s
ed
f
o
r
f
i
n
d
in
g
ta
m
p
er
i
n
g
.
I
n
f
u
t
u
r
e
w
o
r
k
,
d
ee
p
b
elief
n
et
w
o
r
k
s
ca
n
b
e
u
s
ed
f
o
r
f
ea
t
u
r
e
lear
n
i
n
g
.
T
h
i
s
w
o
r
k
[
5
]
p
r
esen
ts
a
tec
h
n
iq
u
e
f
o
r
ta
m
p
er
i
n
g
d
etec
tio
n
a
n
d
r
ec
o
g
n
izi
n
g
t
h
e
m
an
ip
u
lated
r
eg
io
n
s
f
o
r
th
e
i
m
ag
e
s
d
eliv
er
ed
th
r
o
u
g
h
t
h
e
co
m
b
i
n
atio
n
o
f
i
m
ag
e
s
ta
k
en
f
r
o
m
v
ar
io
u
s
ca
m
er
a
m
o
d
el
s
.
T
h
e
i
m
ag
e
i
s
b
r
o
k
en
i
n
to
co
lo
r
p
atch
es
o
f
6
4
x
6
4
.
I
n
th
is
,
C
NN
is
u
s
ed
to
ex
tr
ac
t
f
ea
tu
r
es
f
r
o
m
ev
er
y
p
atch
,
an
d
iter
ativ
e
al
g
o
r
ith
m
s
ar
e
u
s
ed
o
n
th
e
s
e
f
ea
t
u
r
es
to
g
r
o
u
p
th
e
m
i
n
to
p
r
is
ti
n
e
clu
s
ter
s
an
d
f
o
r
g
ed
clu
s
ter
s
.
F
u
t
u
r
e
w
o
r
k
f
o
c
u
s
es
o
n
t
h
e
u
s
a
g
e
o
f
C
NN
to
lear
n
m
o
r
e
f
ea
tu
r
es
s
u
c
h
as
b
lu
r
r
in
g
,
r
o
tati
o
n
s
,
an
d
r
esizin
g
.
T
h
is
w
o
r
k
[
6
]
u
s
es
t
w
o
d
if
f
er
en
t
m
et
h
o
d
s
f
o
r
ta
m
p
er
in
g
d
etec
tio
n
an
d
lo
ca
lizati
o
n
.
T
h
e
f
ir
s
t
m
eth
o
d
ca
lcu
late
s
r
esa
m
p
li
n
g
f
ea
t
u
r
es
o
v
er
o
v
er
l
y
in
g
ar
ea
s
.
A
d
ee
p
lear
n
in
g
n
et
w
o
r
k
an
d
G
au
s
s
ian
co
n
d
itio
n
al
r
an
d
o
m
f
ield
ar
e
u
tili
ze
d
f
o
r
cr
ea
t
in
g
t
h
e
h
ea
t
m
ap
.
I
n
th
e
n
e
x
t
m
et
h
o
d
,
th
e
r
esa
m
p
l
in
g
f
ea
t
u
r
es
ar
e
t
h
e
n
p
ass
ed
to
th
e
L
ST
M
m
o
d
el
t
o
class
i
f
y
th
e
r
eg
io
n
s
a
s
eit
h
er
f
o
r
g
ed
o
r
n
o
t.
I
n
th
e
f
u
t
u
r
e,
C
NN
a
n
d
L
ST
M
f
u
s
io
n
ca
n
b
e
u
s
ed
f
o
r
f
o
r
g
er
y
d
etec
tio
n
.
C
o
n
v
o
lu
tio
n
al
n
e
u
r
al
n
et
w
o
r
k
(
C
NN
)
[
7
]
id
en
tif
ies
t
h
e
im
ag
e
ta
m
p
er
in
g
w
it
h
au
to
m
atic
f
ea
tu
r
e
lear
n
in
g
.
T
h
is
m
o
d
el
h
as
f
i
v
e
co
n
v
o
lu
tio
n
al
la
y
er
s
.
T
h
e
n
et
w
o
r
k
is
m
ad
e
u
p
o
f
t
w
o
f
u
l
l
y
co
n
n
ec
ted
la
y
er
s
a
n
d
a
s
o
f
t
m
ax
clas
s
i
f
ier
.
I
n
th
i
s
[
8
]
,
th
e
C
NN
n
et
w
o
r
k
is
e
m
p
lo
y
ed
f
o
r
d
eter
m
in
in
g
p
atch
-
b
a
s
ed
in
p
a
in
t
in
g
o
p
er
atio
n
.
C
NN
is
u
tili
ze
d
f
o
r
lear
n
in
g
t
h
e
f
ea
t
u
r
es
o
f
t
h
e
i
m
a
g
e.
C
NN
m
o
d
el
is
tr
ai
n
ed
w
it
h
a
class
lab
el
m
atr
i
x
o
f
all
t
h
e
p
i
x
els
o
f
th
e
i
m
a
g
e.
I
n
t
h
is
w
ei
g
h
ted
cr
o
s
s
e
n
tr
o
p
y
is
u
s
ed
to
b
alan
c
e
th
e
in
p
ai
n
ted
a
n
d
u
n
i
n
p
ai
n
ted
p
ix
el
s
.
C
NN
e
n
c
o
d
er
-
d
ec
o
d
er
n
et
w
o
r
k
i
s
e
m
p
l
o
y
ed
f
o
r
p
r
ed
ictin
g
th
e
in
p
ai
n
ti
n
g
p
r
o
b
ab
ilit
y
o
f
ea
ch
p
ix
el
in
a
n
i
m
a
g
e.
I
n
t
h
i
s
w
o
r
k
[
9
]
,
illu
m
i
n
an
t
m
ap
s
a
n
d
C
NN
ar
e
u
s
ed
f
o
r
th
e
s
p
li
cin
g
id
e
n
ti
f
icatio
n
.
T
h
e
d
ee
p
an
d
tr
an
s
f
er
lear
n
i
n
g
tech
n
iq
u
e
s
ar
e
u
tili
ze
d
f
o
r
ex
tr
ac
ti
n
g
f
ea
tu
r
e
s
f
o
r
f
o
r
g
er
y
id
en
ti
f
icatio
n
.
T
h
e
class
i
f
ier
is
tr
ain
ed
w
it
h
t
h
ese
f
ea
t
u
r
es.
Af
ter
id
en
ti
f
y
i
n
g
t
h
e
i
m
a
g
e,
w
h
e
th
er
it
i
s
f
ak
e
o
r
n
o
t,
th
e
n
e
x
t
s
tep
i
s
to
lo
ca
te
th
e
m
an
ip
u
lated
r
eg
i
o
n
.
T
h
e
f
u
tu
r
e
w
o
r
k
co
n
s
id
er
s
lo
ca
lizatio
n
m
ap
s
f
o
r
th
e
id
en
tif
icat
io
n
o
f
f
o
r
g
ed
r
eg
io
n
s
in
a
n
i
m
a
g
e.
I
n
[
1
0
]
,
a
f
u
ll
y
co
n
v
o
lu
t
io
n
al
n
et
w
o
r
k
is
u
s
ed
f
o
r
d
etec
tin
g
t
h
e
i
m
ag
e
s
p
lici
n
g
.
T
h
e
Sin
g
le
ta
s
k
FC
N
is
tr
ai
n
ed
w
i
th
a
s
u
r
f
ac
e
lab
el
th
a
t
id
en
ti
f
i
es
ev
er
y
p
ix
el
i
n
t
h
e
i
m
a
g
e
as
r
ea
l
o
r
n
o
t.
SF
C
N
g
en
er
ate
s
a
co
ar
s
e
lo
ca
lizatio
n
o
u
tp
u
t.
MF
C
N
is
tr
ai
n
ed
o
n
th
e
s
u
p
er
f
icial
lab
el
an
d
b
o
u
n
d
ar
y
,
w
h
ic
h
s
h
o
w
s
th
at
th
e
p
ix
el
i
s
r
elate
d
t
o
th
e
s
p
liced
r
eg
io
n
’
s
b
o
r
d
er
.
T
h
e
ed
g
e
e
n
h
an
ce
d
MFC
N
u
s
e
s
a
s
u
r
f
ac
e
lab
el
a
n
d
ed
g
e
p
r
o
b
a
b
ilit
y
m
ap
,
w
h
ic
h
is
b
etter
th
an
SF
C
N
an
d
th
e
MFC
N
ap
p
r
o
ac
h
.
I
n
th
is
w
o
r
k
[
1
1
]
,
C
NN
b
ased
m
et
h
o
d
s
ar
e
u
s
ed
f
o
r
h
a
n
d
li
n
g
co
p
y
m
o
v
e
f
o
r
g
er
y
d
etec
tio
n
.
C
NN
w
i
th
C
F
A
(
co
lo
r
f
ilte
r
ar
r
ay
)
f
ea
t
u
r
es
ar
e
u
s
ed
f
o
r
f
i
n
d
in
g
a
n
d
lo
ca
lizin
g
th
e
m
a
n
ip
u
lated
r
eg
io
n
.
T
h
e
C
F
A
in
ter
p
o
latio
n
t
ec
h
n
iq
u
e
d
ev
elo
p
s
in
ter
r
elatio
n
s
h
ip
an
d
co
n
s
is
te
n
c
y
b
et
w
ee
n
t
h
e
p
i
x
els.
T
h
er
ef
o
r
e,
in
co
n
s
is
te
n
c
y
i
n
m
an
ip
u
lated
ar
ea
s
ca
n
b
e
u
s
ed
f
o
r
id
en
ti
f
y
i
n
g
t
h
e
f
o
r
g
ed
r
eg
io
n
s
.
R
-
C
NN
[
1
2
]
is
e
m
p
lo
y
ed
f
o
r
f
i
n
d
in
g
m
an
ip
u
lated
r
e
g
io
n
s
i
n
f
o
r
g
ed
i
m
ag
e
s
.
T
h
is
n
et
w
o
r
k
p
r
o
v
id
es
t
w
o
s
tr
ea
m
s
;
o
n
e
i
s
R
GB
i
n
w
h
ic
h
f
ea
tu
r
es
ar
e
e
x
tr
ac
ted
f
r
o
m
R
GB
i
m
a
g
e
to
d
etec
t
ta
m
p
er
in
g
s
u
c
h
as
u
n
n
at
u
r
al
alter
ed
b
o
u
n
d
ar
ies,
an
d
s
tr
o
n
g
co
n
tr
ast
d
i
f
f
er
e
n
ce
.
T
h
e
s
ec
o
n
d
is
th
e
n
o
is
e
s
tr
ea
m
,
w
h
ic
h
tak
e
s
ad
v
an
ta
g
e
o
f
o
b
tai
n
in
g
n
o
is
e
f
ea
t
u
r
es
an
d
id
e
n
ti
f
y
t
h
e
i
n
co
n
s
i
s
te
n
c
y
w
i
th
r
ea
l
a
n
d
m
a
n
ip
u
lated
r
eg
io
n
s
.
T
h
e
b
ilin
ea
r
p
o
o
lin
g
la
y
er
i
s
t
h
en
u
s
ed
f
o
r
f
u
s
i
n
g
th
e
s
e
f
ea
t
u
r
es
r
etr
iev
ed
f
r
o
m
t
w
o
s
tr
ea
m
s
.
I
n
th
is
[
1
3
]
,
d
ee
p
n
eu
r
al
n
et
w
o
r
k
a
n
d
co
n
d
itio
n
al
r
an
d
o
m
f
ield
ar
e
u
s
ed
to
id
en
ti
f
y
th
e
i
m
a
g
e
’
s
s
p
liced
r
eg
io
n
,
w
h
ic
h
d
o
es
n
o
t
n
ee
d
an
y
p
r
io
r
d
ata.
T
h
r
ee
u
n
iq
u
e
v
ar
iatio
n
s
o
f
FC
N
w
er
e
u
tili
ze
d
to
i
m
p
r
o
v
e
th
e
ac
c
u
r
ac
y
.
Di
s
co
v
er
y
o
f
s
m
al
l
m
a
n
ip
u
lated
o
b
j
ec
ts
is
d
if
f
ic
u
lt
as
d
o
w
n
-
s
a
m
p
li
n
g
a
ctio
n
r
ed
u
ce
s
th
e
r
ea
l
i
m
a
g
e
m
ea
s
u
r
e
m
e
n
t
s
an
d
m
ak
e
s
s
m
al
ler
o
b
j
ec
ts
co
n
s
id
er
ab
ly
h
ar
d
er
to
id
en
ti
f
y
.
An
o
th
er
is
s
u
e
i
s
th
e
o
v
er
f
i
tti
n
g
i
s
s
u
e,
w
h
ic
h
w
i
ll
b
e
ad
d
r
ess
ed
in
f
u
tu
r
e
w
o
r
k
.
An
o
th
er
f
u
tu
r
e
w
o
r
k
i
n
cl
u
d
es
th
e
o
p
ti
m
izatio
n
o
f
t
h
e
n
et
w
o
r
k
f
o
r
s
p
lici
n
g
d
etec
tio
n
.
T
h
is
f
r
a
m
e
w
o
r
k
[
1
4
]
is
u
s
ed
f
o
r
th
e
d
etec
t
io
n
o
f
f
o
r
g
ed
i
m
ag
e
s
.
I
n
t
h
is
n
et
w
o
r
k
,
t
h
e
i
m
a
g
e
i
s
s
p
li
t
in
to
p
atch
es
.
T
h
en
r
esa
m
p
li
n
g
f
ea
t
u
r
es
ar
e
ex
tr
ac
ted
f
r
o
m
th
ese
p
atch
es.
T
h
e
Hilb
er
t
cu
r
v
e
f
in
d
s
o
u
t
th
e
s
eq
u
en
c
in
g
o
f
p
atc
h
es
w
h
ic
h
ar
e
s
u
p
p
lied
to
L
ST
M.
T
h
e
s
a
m
p
lin
g
f
ea
t
u
r
es
d
etec
t
t
h
e
c
o
r
r
elatio
n
b
et
w
ee
n
p
atch
es.
T
h
e
en
co
d
er
is
u
s
ed
f
o
r
f
i
n
d
in
g
o
u
t
t
h
e
s
p
atial
lo
c
atio
n
o
f
a
m
a
n
ip
u
lated
r
eg
io
n
.
Sp
atial
f
ea
tu
r
es
f
r
o
m
t
h
e
en
co
d
er
an
d
o
u
t
f
ea
t
u
r
es
f
r
o
m
L
ST
M
ar
e
c
o
m
b
i
n
e
d
f
o
r
d
etec
tin
g
f
o
r
g
er
y
in
a
n
im
ag
e.
T
h
e
d
ec
o
d
er
m
o
d
el
g
i
v
e
s
t
h
e
f
i
n
er
r
ep
r
esen
tatio
n
o
f
th
e
s
p
atial
m
ap
,
wh
ich
o
f
f
er
s
th
e
alter
ed
r
eg
io
n
in
an
i
m
ag
e.
An
i
m
p
r
o
v
ed
m
a
s
k
R
-
C
N
N
m
o
d
e
l
[
1
5
]
(
r
eg
io
n
al
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
w
o
r
k
)
i
s
p
r
o
p
o
s
ed
w
ith
a
So
b
el
f
ilter
to
r
ec
o
g
n
ize
th
e
alter
ed
a
n
d
u
n
al
ter
ed
r
eg
io
n
’
s
d
is
ti
n
cti
v
e
f
ea
t
u
r
es.
T
h
is
n
e
t
w
o
r
k
h
a
n
d
les
t
wo
t
y
p
e
s
o
f
f
o
r
g
e
r
i
e
s
,
s
u
c
h
a
s
c
o
p
y
m
o
v
e
a
n
d
s
p
l
i
c
i
n
g
.
P
i
x
e
l
w
i
s
e
i
n
f
o
r
m
a
t
i
o
n
i
s
u
s
e
d
f
o
r
t
r
a
i
n
i
n
g
t
h
e
m
o
d
e
l
,
a
n
d
S
o
b
e
l
f
i
l
t
e
r
s
u
s
e
t
h
e
e
d
g
e
’
s
i
n
f
o
r
m
a
t
i
o
n
t
o
i
d
e
n
t
i
f
y
t
h
e
m
a
n
i
p
u
l
a
t
e
d
b
o
u
n
d
a
r
i
e
s
.
F
u
l
l
y
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
m
o
d
e
l
(
F
C
N
N
)
[
1
6
]
is
u
s
ed
f
o
r
i
m
a
g
e
s
p
lici
n
g
d
etec
t
io
n
.
I
t
d
is
tin
g
u
is
h
es
t
h
e
alter
i
n
g
o
f
an
i
m
ag
e
a
s
w
ell
as
r
ec
o
g
n
ize
s
th
e
f
o
r
g
er
y
o
f
s
p
liced
r
eg
io
n
s
.
T
h
e
FC
N
g
iv
e
s
m
is
cla
s
s
i
f
icatio
n
w
h
e
n
th
er
e
ar
e
n
u
m
er
o
u
s
p
eo
p
le
ar
e
th
er
e
in
th
e
n
o
n
-
s
p
liced
r
eg
io
n
o
f
an
i
m
ag
e.
T
h
e
i
m
p
r
o
v
ed
FC
N
is
u
s
ed
to
ca
p
tu
r
e
th
e
d
if
f
er
en
t
f
ea
tu
r
e
s
o
f
s
p
liced
an
d
n
o
n
-
s
p
lice
d
r
eg
io
n
s
.
T
h
e
i
m
p
r
o
v
ed
C
NN
n
et
w
o
r
k
is
tr
ai
n
ed
w
ith
au
th
e
n
tic
a
n
d
alter
ed
i
m
a
g
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
5
,
Octo
b
e
r
2
0
2
1
:
4
4
8
9
-
4501
4496
5.
ANALY
T
I
CAL
H
I
E
RAR
C
H
I
CAL
P
RO
C
E
SS
I
N
G
(
AH
P
)
M
O
DE
L
F
O
R
I
M
AG
E
F
O
RG
E
RY
DE
T
E
C
T
I
O
N
An
al
y
tical
Hier
ar
c
h
ical
P
r
o
ce
s
s
is
a
m
u
lti
-
cr
iter
ia
d
ec
is
io
n
-
m
ak
in
g
m
at
h
e
m
a
tical
tec
h
n
iq
u
e.
I
t
h
el
p
s
to
m
a
k
e
as
s
o
ciatio
n
s
a
m
o
n
g
q
u
alitati
v
e
an
d
q
u
a
n
titat
iv
e
at
tr
ib
u
tes
an
d
i
s
al
m
o
s
t
u
s
ed
in
ev
er
y
ar
ea
.
I
n
th
i
s
p
ap
er
,
th
e
A
HP
ca
lcu
latio
n
is
d
o
n
e
b
y
u
s
i
n
g
an
o
n
li
n
e
to
o
l
(
h
ttp
s
://b
p
m
s
g
.
co
m
/ah
p
/)
.
T
h
is
p
r
o
ce
s
s
f
o
llo
w
s
f
i
v
e
s
tep
s
,
w
h
ic
h
ar
e
g
iv
e
n
b
e
lo
w
:
i)
H
ier
ar
ch
y
m
o
d
eli
n
g
,
ii)
P
r
io
r
ities
estab
lis
h
m
e
n
t
;
ii
i)
Dec
is
io
n
r
elate
d
to
o
v
er
all
p
r
io
r
ities
o
f
h
ier
ar
c
h
y
;
iv
)
C
h
ec
k
i
n
g
co
n
s
is
te
n
c
y
;
an
d
v)
J
u
d
g
m
e
n
t p
r
ep
ar
atio
n
.
I
n
A
HP
,
n
u
m
er
ical
r
a
n
k
i
n
g
[
1
7
]
r
an
g
in
g
f
r
o
m
1
to
9
co
m
p
ar
es
attr
ib
u
tes
at
v
ar
io
u
s
le
v
els
i
n
t
h
e
h
ier
ar
ch
y
b
y
co
n
s
id
er
in
g
alt
er
n
ativ
e
s
.
F
ig
u
r
e
2
d
ec
id
es
d
if
f
er
e
n
t
le
v
el
s
o
f
d
ec
is
io
n
s
f
o
r
t
h
e
a
n
al
y
tica
l
h
ier
ar
ch
y
p
r
o
ce
s
s
o
f
d
i
g
ital
i
m
ag
e
f
o
r
g
er
y
d
etec
tio
n
.
L
iter
a
tu
r
e
s
h
o
w
s
t
h
at
p
as
s
iv
e
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
i
s
m
o
r
e
s
i
g
n
i
f
ica
n
t
th
a
n
th
e
ac
ti
v
e
ap
p
r
o
ac
h
;
th
er
ef
o
r
e,
in
lev
el
1
p
ass
iv
e
ap
p
r
o
ac
h
is
u
s
ed
to
m
a
k
e
a
d
ec
is
io
n
.
Af
ter
lo
o
k
i
n
g
at
h
ier
ar
ch
y
m
o
d
elin
g
,
T
ab
le
s
1
-
3
i
n
d
icate
s
p
r
io
r
itizin
g
a
ttrib
u
te
s
f
o
r
p
er
f
o
r
m
i
n
g
A
HP
o
n
i
m
a
g
e
f
o
r
g
er
y
.
T
h
e
c
o
lo
r
c
o
d
e
in
co
lu
m
n
s
P
an
d
Q
o
f
T
ab
le
1
-
3
s
h
o
w
s
th
e
n
u
m
b
er
o
f
co
m
p
ar
is
o
n
s
th
at
m
u
s
t
b
e
m
ad
e
b
et
w
ee
n
P
an
d
Q.
B
y
estab
lis
h
i
n
g
r
elev
a
n
ce
w
ith
s
ca
le
(
1
-
9
)
,
th
e
y
ello
w
co
lo
r
in
d
icate
s
s
ig
n
i
f
ica
n
c
e
a
m
o
n
g
t
h
e
co
m
p
ar
ed
attr
ib
u
te
s
.
T
ab
le
4
-
6
s
h
o
w
co
n
s
i
s
ten
c
y
ch
ec
k
s
f
o
r
co
n
s
id
er
ed
lev
els
o
f
a
f
ir
s
t a
n
d
s
ec
o
n
d
s
et
o
f
i
m
p
o
r
t
an
t
i
m
ag
e
f
o
r
g
er
y
d
etec
tio
n
attr
ib
u
te
s
.
T
h
ese
tab
les
s
h
o
w
t
h
e
p
r
io
r
it
y
a
n
d
r
an
k
o
f
a
g
r
o
u
p
o
f
i
m
p
o
r
tan
t a
ttrib
u
tes,
w
it
h
th
e
h
ig
h
e
s
t p
r
io
r
it
y
in
d
icati
n
g
th
e
m
o
s
t i
m
p
o
r
tan
t a
ttrib
u
te.
I
n
th
e
s
ec
o
n
d
le
v
el,
th
e
f
o
r
g
er
y
t
y
p
e
-
d
ep
en
d
en
t
tec
h
n
iq
u
e
is
co
n
s
id
er
ed
.
L
iter
atu
r
e
s
h
o
w
s
th
at
t
h
ese
f
o
r
g
er
ies
e
m
p
h
ases
o
n
l
y
o
n
p
a
r
ticu
lar
t
y
p
e
s
o
f
f
o
r
g
er
ie
s
,
an
d
th
ese
ar
e
o
b
tain
i
n
g
m
o
r
e
r
ec
o
g
n
i
tio
n
a
s
t
h
e
y
ar
e
f
r
eq
u
en
tl
y
p
er
f
o
r
m
ed
m
a
n
ip
u
l
atio
n
s
i
n
t
h
e
i
m
a
g
e.
Fi
g
u
r
es
1
3
-
1
5
s
h
o
w
r
a
n
k
lev
e
ls
1
,
2
an
d
3
,
r
esp
ec
tiv
el
y
.
Fig
u
r
e
1
6
s
u
m
m
ar
izes
th
e
AHP
i
m
p
le
m
en
ta
tio
n
r
esu
lt
f
o
r
th
e
u
n
d
er
tak
e
n
r
e
s
ea
r
ch
p
r
o
b
le
m
,
w
h
ic
h
g
i
v
es
s
ig
n
i
f
ica
n
t a
ttrib
u
te
s
at
ev
er
y
l
ev
el.
T
ab
le
1
.
Fo
r
m
u
lat
io
n
o
f
p
r
io
r
ities
f
o
r
lev
e
l 1
(
t
y
p
es o
f
f
o
r
g
er
y
d
etec
tio
n
s
u
c
h
as a
cti
v
e
an
d
p
ass
iv
e
ap
p
r
o
ac
h
)
-
f
ir
s
t
s
et
o
f
s
i
g
n
i
f
ican
t a
ttrib
u
te
s
f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
(
1
co
m
p
ar
is
o
n
)
F
i
r
st
se
t
o
f
a
t
t
r
i
b
u
t
e
s
Eq
u
a
l
l
y
i
mp
o
r
t
a
n
t
H
o
w
mu
c
h
i
mp
o
r
t
a
n
t
?
P
Q
1
2
3
4
5
6
7
8
9
1
P
a
ssi
v
e
A
p
p
r
o
a
c
h
A
c
t
i
v
e
A
p
p
r
o
a
c
h
T
ab
le
2
.
C
o
n
s
is
te
n
c
y
c
h
ec
k
f
o
r
lev
el
1
-
t
h
e
f
ir
s
t set o
f
s
ig
n
i
f
i
ca
n
t a
ttrib
u
te
s
f
o
r
i
m
ag
e
f
o
r
g
e
r
y
d
etec
tio
n
C
o
n
si
s
t
e
n
c
y
c
h
e
c
k
f
o
r
l
e
v
e
l
1
-
t
h
e
f
i
r
s
t
se
t
o
f
si
g
n
i
f
i
c
a
n
t
a
t
t
r
i
b
u
t
e
s
+
-
1
2
P
r
i
o
r
i
t
y
R
a
n
k
P
a
ssi
v
e
A
p
p
r
o
a
c
h
8
3
.
3
%
1
0
.
0
%
0
.
0
%
1
5
.
0
0
A
c
t
i
v
e
A
p
p
r
o
a
c
h
1
6
.
7
%
2
0
.
0
%
0
.
0
%
0
.
2
0
1
C
o
mp
a
r
i
so
n
s
1
P
r
i
n
c
i
p
a
l
e
i
g
e
n
v
a
l
u
e
2
.
0
0
0
C
o
n
si
s
t
e
n
c
y
c
h
e
c
k
0
.
0
%
Ei
g
e
n
v
e
c
t
o
r
so
l
u
t
i
o
n
1
i
t
e
r
a
t
i
o
n
s,
d
e
l
t
a
7
.
7
E
-
34
T
ab
le
3
.
Fo
r
m
u
lat
io
n
o
f
p
r
io
r
ities
f
o
r
lev
e
l 2
(
ty
p
e
s
o
f
p
ass
i
v
e
f
o
r
g
er
y
d
etec
tio
n
s
u
c
h
as
f
o
r
g
er
y
t
y
p
e
-
d
ep
en
d
en
t a
n
d
f
o
r
g
er
y
t
y
p
e
i
n
d
ep
en
d
en
t)
-
t
h
e
s
ec
o
n
d
s
et
o
f
s
ig
n
if
ican
t a
ttrib
u
tes
f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
(
1
co
m
p
ar
is
o
n
)
S
e
c
o
n
d
se
t
o
f
a
t
t
r
i
b
u
t
e
s
Eq
u
a
l
l
y
i
mp
o
r
t
a
n
t
H
o
w
mu
c
h
i
mp
o
r
t
a
n
t
?
P
Q
1
2
3
4
5
6
7
8
9
1
F
o
r
g
e
r
y
t
y
p
e
d
e
p
e
n
d
e
n
t
F
o
r
g
e
r
y
t
y
p
e
i
n
d
e
p
e
n
d
e
n
t
T
ab
le
4
.
C
o
n
s
is
te
n
c
y
c
h
ec
k
in
g
f
o
r
lev
el
2
-
t
h
e
s
ec
o
n
d
s
e
t o
f
s
ig
n
if
ican
t a
ttrib
u
tes
f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
(
1
co
m
p
ar
is
o
n
)
C
o
n
si
s
t
e
n
c
y
c
h
e
c
k
f
o
r
l
e
v
e
l
2
–
t
h
e
se
c
o
n
d
se
t
o
f
si
g
n
i
f
i
c
a
n
t
a
t
t
r
i
b
u
t
e
s
+
-
1
2
P
r
i
o
r
i
t
y
R
a
n
k
F
o
r
g
e
r
y
t
y
p
e
d
e
p
e
n
d
e
n
t
8
7
.
5
%
1
0
.
0
%
0
.
0
%
1
0
.
1
4
2
8
5
7
F
o
r
g
e
r
y
t
y
p
e
i
n
d
e
p
e
n
d
e
n
t
1
2
.
5
%
2
0
.
0
%
0
.
0
%
7
1
C
o
mp
a
r
i
so
n
s
1
P
r
i
n
c
i
p
a
l
e
i
g
e
n
v
a
l
u
e
2
.
0
0
0
C
o
n
si
s
t
e
n
c
y
c
h
e
c
k
0
.
0
%
Ei
g
e
n
v
e
c
t
o
r
so
l
u
t
i
o
n
1
i
t
e
r
a
t
i
o
n
s,
d
e
l
t
a
0
.
0
E+
0
I
n
t
h
e
t
h
i
r
d
l
e
v
e
l
,
e
q
u
a
l
p
r
i
o
r
i
t
i
e
s a
r
e
g
i
v
e
n
t
o
b
o
t
h
c
o
p
y
mo
v
e
a
n
d
sp
l
i
c
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
HP
va
lid
a
ted
liter
a
tu
r
e
r
ev
iew
o
f fo
r
g
ery
typ
e
d
ep
en
d
e
n
t
p
a
s
s
ive
ima
g
e
fo
r
g
ery
…
(
K
a
lya
n
i Ka
d
a
m
)
4497
T
ab
le
5
.
Fo
r
m
u
lat
io
n
o
f
p
r
io
r
ities
f
o
r
lev
e
l 3
(
ty
p
e
s
o
f
f
o
r
g
er
y
t
y
p
e
-
d
ep
en
d
en
t s
u
c
h
as c
o
p
y
m
o
v
e
an
d
s
p
licin
g
)
-
th
e
t
h
ir
d
s
et
o
f
s
ig
n
i
f
ican
t a
ttrib
u
te
s
f
o
r
i
m
ag
e
f
o
r
g
er
y
d
etec
tio
n
(
1
co
m
p
ar
is
o
n
)
T
h
i
r
d
se
t
o
f
a
t
t
r
i
b
u
t
e
s
Eq
u
a
l
l
y
i
mp
o
r
t
a
n
t
H
o
w
mu
c
h
i
mp
o
r
t
a
n
t?
P
Q
1
2
3
4
5
6
7
8
9
1
c
o
p
y
mo
v
e
sp
l
i
c
i
n
g
T
ab
le
6
.
C
o
n
s
is
te
n
c
y
c
h
ec
k
in
g
f
o
r
lev
el
3
-
t
h
e
t
h
ir
d
s
et
o
f
s
i
g
n
i
f
ican
t a
ttrib
u
tes
f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
(
1
co
m
p
ar
is
o
n
)
C
o
n
si
s
t
e
n
c
y
c
h
e
c
k
f
o
r
l
e
v
e
l
3
-
t
h
e
t
h
i
r
d
se
t
o
f
s
i
g
n
i
f
i
c
a
n
t
a
t
t
r
i
b
u
t
e
s
+
-
1
2
P
r
i
o
r
i
t
y
R
a
n
k
C
o
p
y
mo
v
e
5
0
.
0
%
1
0
.
0
%
0
.
0
%
1
1
.
0
0
S
p
l
i
c
i
n
g
5
0
.
0
%
1
0
.
0
%
0
.
0
%
1
.
0
0
1
C
o
mp
a
r
i
so
n
s
1
P
r
i
n
c
i
p
a
l
e
i
g
e
n
v
a
l
u
e
2
.
0
0
0
C
o
n
si
s
t
e
n
c
y
c
h
e
c
k
0
.
0
%
Ei
g
e
n
v
e
c
t
o
r
so
l
u
t
i
o
n
1
i
t
e
r
a
t
i
o
n
s,
d
e
l
t
a
0
.
0
E+
0
Fig
u
r
e
1
3
.
R
an
k
le
v
el
1
-
th
e
f
ir
s
t set o
f
s
ig
n
i
f
ica
n
t a
ttrib
u
te
s
f
o
r
im
a
g
e
f
o
r
g
er
y
d
etec
tio
n
Fig
u
r
e
1
4
.
R
an
k
le
v
el
2
-
th
e
s
e
co
n
d
s
et
o
f
s
i
g
n
if
ica
n
t a
ttrib
u
t
es f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
5
,
Octo
b
e
r
2
0
2
1
:
4
4
8
9
-
4501
4498
Fig
u
r
e
1
5
.
R
an
k
le
v
el
3
-
th
e
t
h
i
r
d
s
et
o
f
s
ig
n
i
f
ica
n
t a
ttrib
u
te
s
f
o
r
i
m
a
g
e
f
o
r
g
er
y
d
etec
tio
n
Fig
u
r
e
1
6
.
Su
m
m
ar
izatio
n
o
f
A
HP
i
m
p
le
m
e
n
tatio
n
r
esu
l
t f
o
r
th
e
u
n
d
er
tak
en
r
esear
c
h
p
r
o
b
le
m
6.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
O
N
E
XP
L
A
I
NAB
L
E
AI
Ma
ch
i
n
e
L
ea
r
n
i
n
g
is
f
r
eq
u
e
n
tl
y
u
s
ed
i
n
r
esear
ch
a
n
d
d
ev
elo
p
m
e
n
t
a
n
d
,
m
ai
n
l
y
t
h
r
o
u
g
h
Dee
p
L
ea
r
n
i
n
g
an
d
n
e
u
r
al
n
et
w
o
r
k
ac
h
ie
v
e
m
en
t.
No
w
ad
a
y
s
,
m
ac
h
in
e
lear
n
in
g
n
et
w
o
r
k
s
a
r
e
o
v
er
g
r
o
w
i
n
g
f
o
r
m
ak
in
g
p
r
ed
ictio
n
s
in
cr
itica
l
co
n
tex
t
s
;
d
i
f
f
er
e
n
t
A
I
s
tak
e
h
o
ld
er
s
r
eq
u
ir
e
tr
an
s
p
ar
en
c
y
i
n
p
r
ed
ictio
n
s
[
1
8
]
.
T
h
is
is
ac
h
ie
v
ed
b
y
u
til
izi
n
g
in
ter
p
r
etab
le
n
et
w
o
r
k
s
o
r
g
r
o
w
i
n
g
n
e
w
s
tr
ateg
ie
s
a
n
d
u
s
i
n
g
a
lo
ca
l
ap
p
r
o
x
im
a
tio
n
f
o
r
in
ter
p
r
etatio
n
.
X
A
I
h
elp
s
t
h
e
clie
n
ts
t
o
u
n
d
er
s
ta
n
d
an
d
tr
u
s
t
a
n
et
w
o
r
k
u
s
i
n
g
v
ar
io
u
s
in
ter
p
r
etab
le
n
et
w
o
r
k
s
.
A
I
e
x
h
ib
ited
to
b
e
i
m
p
o
r
tan
t
i
n
d
is
co
v
er
i
n
g
n
e
w
u
s
es
f
o
r
e
x
is
ti
n
g
d
r
u
g
s
,
d
etec
tin
g
ca
n
ce
r
in
ti
s
s
u
es,
d
is
tin
g
u
is
h
i
n
g
h
ea
r
t
ar
r
h
y
th
m
i
a,
an
d
a
n
ticip
ati
n
g
h
y
p
o
g
l
y
ce
m
ia
in
p
eo
p
le
w
i
th
d
iab
etes a
t
an
ea
r
lier
t
i
m
e
t
h
an
th
at
o
f
t
h
e
clin
ical
i
n
d
u
s
tr
y
.
Ma
ch
i
n
e
L
ea
r
n
in
g
,
A
r
ti
f
icial
I
n
telli
g
e
n
ce
,
Neu
r
al
Net
w
o
r
k
s
,
an
d
Dee
p
L
ea
r
n
in
g
m
o
d
el
s
h
a
v
e
r
ec
en
t
l
y
s
h
o
w
n
th
eir
m
ed
ical
f
ield
p
er
f
o
r
m
a
n
ce
,
b
u
t
j
u
s
ti
f
icat
io
n
s
a
s
s
i
s
ti
n
g
t
h
e
o
u
tp
u
ts
o
f
a
n
et
w
o
r
k
r
e
m
ain
s
ig
n
i
f
ica
n
t,
e.
g
.
,
in
t
h
e
p
r
ec
is
io
n
m
ed
icin
e
f
i
eld
.
Dee
p
L
ea
r
n
in
g
i
s
also
u
s
e
d
in
o
th
er
d
o
m
ain
s
,
e.
g
.
,
s
elf
-
d
ir
ec
ted
v
e
h
icles i
n
t
r
an
s
p
o
r
t s
y
s
te
m
s
,
s
a
f
et
y
,
a
n
d
ec
o
n
o
m
ics
.
H
u
m
an
s
u
s
i
n
g
t
h
es
e
n
et
w
o
r
k
s
m
u
s
t b
e
tr
u
s
t
w
o
r
t
h
y
,
i
n
ter
p
r
etab
le,
an
d
tr
ac
tab
le.
Nu
m
er
o
u
s
r
esear
ch
an
d
d
e
v
elo
p
m
e
n
t
p
ap
er
s
s
h
o
w
t
h
at
X
A
I
h
a
s
Evaluation Warning : The document was created with Spire.PDF for Python.