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.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
,
p
p
.
5
7
0
3
~
5
7
0
8
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
0
i
6
.
pp
5
7
0
3
-
5
7
0
8
5703
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m/in
d
ex
.
p
h
p
/I
JE
C
E
A new
s
w
a
r
m
in
t
ellig
ence in
for
m
a
tion tec
hnique
f
o
r
i
m
pro
v
ing
infor
m
a
tion ba
la
ncedness
on
the
s
k
in
lesio
ns seg
m
e
ntatio
n
H
.
J
.
Abd
1
,
Ah
m
a
d
S.
Abd
ul
la
h
2
,
M
uh
a
mm
ed
Sa
la
h Sa
d
i
q Al
k
a
f
a
j
i
3
1
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
Ba
b
y
lo
n
Un
iv
e
rsity
,
Ira
q
2
De
p
a
rtme
n
t
o
f
Co
m
m
u
n
ica
ti
o
n
s,
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
Un
iv
e
rsi
ty
o
f
Di
y
a
la,
Ira
q
3
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l
P
o
w
e
r
T
e
c
h
n
iq
u
e
s
En
g
in
e
e
rin
g
,
T
e
c
h
n
ic
a
l
Co
ll
e
g
e
/A
l
-
M
a
u
sa
ib
,
Al
-
F
u
ra
t
A
l
-
Aw
sa
t
Tec
h
n
ica
l
Un
i
v
e
rsit
y
,
Ira
q
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
2
0
,
2
0
1
9
R
ev
i
s
ed
Ma
y
8
,
2020
A
cc
ep
ted
Ma
y
2
3
,
2
0
2
0
M
e
th
o
d
s
o
f
im
a
g
e
p
ro
c
e
ss
in
g
c
a
n
re
c
o
g
n
ize
th
e
ima
g
e
s
o
f
m
e
lan
o
m
a
les
io
n
s
b
o
r
d
e
r
in
a
d
d
i
ti
o
n
t
o
t
h
e
d
ise
a
se
c
o
m
p
a
re
d
to
a
sk
il
led
d
e
r
m
a
to
lo
g
ist.
Ne
w
s
w
a
r
m
in
telli
g
e
n
c
e
tec
h
n
iq
u
e
d
e
p
e
n
d
s
o
n
m
e
ta
-
h
e
u
risti
c
th
a
t
is
in
d
u
strialize
d
t
o
re
so
lv
e
c
o
m
p
o
si
te
re
a
l
p
ro
b
lem
s
w
h
ich
a
re
p
ro
b
l
e
m
a
ti
c
to
e
x
p
lain
b
y
th
e
a
v
a
il
a
b
le
d
e
ter
m
i
n
isti
c
a
p
p
r
o
a
c
h
e
s.
F
o
r
a
n
a
c
c
u
ra
te
d
e
tec
ti
o
n
o
f
a
ll
se
g
m
e
n
tatio
n
a
n
d
c
las
sif
ica
ti
o
n
o
f
sk
in
les
io
n
s,
s
o
m
e
d
e
a
li
n
g
s
sh
o
u
ld
b
e
m
e
a
su
re
d
w
h
ich
c
o
n
tain
,
c
o
n
tr
a
st
b
ro
a
d
e
n
i
n
g
,
irreg
u
larity
q
u
a
n
t
it
y
,
c
h
o
ice
o
f
m
o
s
t
o
p
t
i
m
a
l
f
e
a
t
u
re
s
,
a
n
d
s
o
i
n
t
o
t
h
e
w
o
r
l
d
.
T
h
e
p
r
i
c
e
e
ss
e
n
t
i
a
l
f
o
r
t
h
e
a
c
t
i
o
n
o
f
p
r
o
g
re
ss
i
v
e
d
i
s
e
a
se
c
a
s
e
s
i
s
i
d
e
n
t
i
c
a
l
h
i
g
h
a
n
d
t
h
e
s
u
rv
i
v
a
l
p
e
r
c
e
n
t
a
g
e
i
s
l
o
w
.
M
a
n
y
e
l
e
c
t
r
o
n
i
c
d
e
rm
o
s
c
o
p
y
c
la
s
s
if
ic
a
t
i
o
n
s
a
re
a
d
v
a
n
c
e
d
d
e
p
e
n
d
o
n
t
h
e
g
r
o
u
p
i
n
g
o
f
f
o
rm
,
s
u
rf
a
c
e
a
n
d
d
y
e
f
e
a
t
u
r
e
s
t
o
f
a
c
i
l
i
ta
te
p
r
e
m
a
t
u
re
a
n
a
ly
s
is
o
f
m
a
l
ig
n
a
n
c
e
.
T
o
o
v
e
rc
o
m
e
th
is
p
ro
b
lem
a
ti
c
,
a
n
e
ff
e
c
ti
v
e
p
ro
t
o
ty
p
ica
l
f
o
r
a
c
c
u
ra
te
b
o
u
n
d
a
ry
d
e
tec
ti
o
n
a
n
d
a
rra
n
g
e
m
e
n
t
is
o
b
tai
n
a
b
le.
T
h
e
p
ro
jec
ted
c
las
sic
a
l
re
c
o
v
e
r
s
th
e
o
p
ti
m
iza
ti
o
n
se
g
m
e
n
t
o
f
a
c
c
u
ra
c
y
in
it
s
p
re
-
p
ro
c
e
ss
in
g
sta
g
e
,
a
p
p
ly
in
g
c
o
n
tras
t
im
p
ro
v
e
m
e
n
t
o
f
le
sio
n
a
re
a
c
o
m
p
a
re
d
to
th
e
c
o
n
tex
tu
a
l.
I
n
c
o
n
c
lu
sio
n
,
o
p
ti
m
ize
d
f
e
a
tu
re
s
a
re
f
u
tu
re
fe
d
in
t
o
o
f
a
rti
f
ica
l
b
e
e
c
o
lo
n
y
(
A
BC)
se
g
m
e
n
tatio
n
.
W
id
e
-
ra
n
g
in
g
re
se
a
rc
h
e
s
h
a
v
e
b
e
e
n
su
p
p
o
rted
o
u
t
o
n
f
o
u
r
d
a
tab
a
se
s n
a
m
e
d
a
s,
IS
BI
(2
0
1
6
,
2
0
1
7
,
2
0
1
8
)
a
n
d
P
H
2
.
A
lso
,
th
e
se
lec
ti
o
n
tec
h
n
i
q
u
e
o
u
tcla
ss
e
s
a
n
d
su
c
c
e
ss
f
u
ll
y
in
d
if
f
e
re
n
t
th
e
d
ism
isse
d
f
e
a
tu
re
s.
T
h
e
p
a
p
e
r
sh
o
w
s
a
d
if
fe
re
n
t
p
ro
c
e
ss
f
o
r
les
io
n
s
o
p
ti
m
a
l
se
g
m
e
n
tatio
n
th
a
t
c
o
u
l
d
b
e
f
u
n
c
ti
o
n
a
l
to
a
v
a
riatio
n
o
f
i
m
a
g
e
s
w
it
h
c
h
a
n
g
e
d
p
o
ss
e
ss
io
n
s
a
n
d
in
s
u
f
f
i
c
ie
n
c
ies
is
p
lan
n
e
d
w
it
h
m
u
lt
istep
p
re
-
p
ro
c
e
ss
in
g
sta
g
e
.
K
ey
w
o
r
d
s
:
Au
to
m
a
tic
s
e
g
m
en
ta
tio
n
De
r
m
a
to
lo
g
ists
Me
d
ian
f
il
ter
Sk
i
n
lesi
o
n
S
w
ar
m
i
n
telli
g
e
n
ce
tech
n
iq
u
e
s
Co
p
y
rig
h
t
©
2
0
2
0
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
:
H.
J
.
A
b
d
,
Dep
ar
t
m
en
t o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
,
C
o
lleg
e
o
f
E
n
g
in
ee
r
i
n
g
,
B
ab
y
l
o
n
Un
i
v
er
s
it
y
,
B
ab
y
lo
n
,
I
r
aq
E
m
ail:
E
n
g
.
h
aid
er
.
j
ab
b
e
r
@
u
o
b
ab
y
lo
n
.
ed
u
.
iq
1.
I
NT
RO
D
UCT
I
O
N
Desp
ite
t
h
er
e
ar
e
s
ev
er
al
a
p
p
r
o
ac
h
es
o
b
tain
ab
le
f
o
r
b
io
m
ed
ical
i
m
a
g
e
s
e
g
m
e
n
tati
o
n
,
ab
o
u
t
ap
p
r
o
ac
h
es
n
ee
d
h
i
g
h
co
m
p
u
tatio
n
al
d
i
f
f
icu
l
t
y
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
s
eg
m
e
n
tatio
n
[
1
-
4
]
f
o
r
ex
a
m
p
le
th
e
tec
h
n
iq
u
e
w
h
ich
i
n
clu
d
e
s
o
f
m
ac
h
in
e
lear
n
i
n
g
,
A
r
ti
f
icia
l
s
w
ar
m
i
n
tel
lig
e
n
ce
(
ASI
)
f
o
r
ex
a
m
p
le
A
r
ti
f
ic
ial
b
ee
co
lo
n
y
(
A
B
C
)
d
esp
ite
t
h
e
f
ac
t
ap
p
r
o
x
i
m
atel
y
o
t
h
er
tech
n
iq
u
es
f
o
llo
w
a
lo
t
o
f
p
r
o
ce
d
u
r
es
f
o
r
n
e
w
alg
o
r
ith
m
s
,
s
eg
m
e
n
tatio
n
an
d
class
i
f
icatio
n
[
5
-
9
]
.
T
h
e
b
io
m
ed
ical
i
m
ag
e
s
eg
m
e
n
tatio
n
co
n
tain
s
o
f
i
m
ag
e
d
atab
ases
,
p
r
e
-
p
r
o
ce
s
s
in
g
,
f
e
atu
r
e
ex
tr
ac
tio
n
,
p
o
s
tp
r
o
ce
s
s
in
g
,
ca
talo
g
u
i
n
g
,
an
d
p
er
f
o
r
m
an
ce
e
s
ti
m
atio
n
.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
is
a
d
ev
el
o
p
m
e
n
t
o
f
ar
r
an
g
e
m
en
t
s
th
a
t
i
s
i
m
p
le
m
e
n
ted
o
n
i
m
a
g
e
d
ata,
to
co
m
p
l
ete
a
b
est
p
r
esen
tatio
n
o
f
a
d
ata
b
ase.
I
t
h
as
an
i
m
p
o
r
tan
t
in
f
l
u
e
n
ce
o
n
th
e
ac
t
o
f
th
e
s
eg
m
e
n
tatio
n
al
g
o
r
ith
m
.
Nu
m
b
er
s
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.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
:
5
7
0
3
-
5
7
0
8
5704
p
r
e
-
p
r
o
ce
s
s
in
g
s
ta
g
e,
in
b
io
m
ed
ical
i
m
a
g
e
p
r
o
ce
s
s
in
g
,
co
n
t
ain
s
o
f
i
m
ag
e
cr
o
p
p
in
g
,
f
il
ter
in
g
,
s
eg
m
e
n
tatio
n
,
g
r
ad
ien
t
o
p
er
atio
n
s
an
d
class
i
f
icatio
n
[
1
0
-
1
2
]
.
So
,
it
is
v
alu
e
to
s
tate
th
at
en
h
a
n
ci
n
g
t
h
e
s
e
g
m
e
n
tat
io
n
p
r
o
ce
s
s
w
il
l a
d
v
an
ce
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
a
d
esig
n
ated
class
i
f
ica
tio
n
m
et
h
o
d
[
1
3
]
.
I
n
ad
d
itio
n
,
p
h
y
s
ical
i
n
s
p
ec
t
io
n
b
y
a
p
r
i
m
e
d
er
m
ato
lo
g
i
s
t,
s
k
i
n
s
el
f
-
i
n
v
e
s
ti
g
atio
n
i
n
at
-
h
o
u
s
e
b
ac
k
g
r
o
u
n
d
s
i
s
ap
p
r
ec
iated
f
o
r
th
e
p
r
e
m
at
u
r
e
an
a
l
y
s
is
o
f
s
k
in
ca
n
ce
r
.
A
f
u
ll
s
k
in
s
el
f
-
ch
ec
k
co
m
p
r
i
s
es
a
co
m
p
r
eh
e
n
s
i
v
e
an
a
l
y
s
is
o
f
w
h
o
ll
y
b
o
d
y
p
o
r
tio
n
s
,
in
cl
u
d
i
n
g
t
h
e
s
ca
lp
p
ar
ts
.
I
n
to
talin
g
,
i
m
a
g
in
g
k
n
o
w
led
g
e
s
h
elp
in
ac
c
u
r
ate
an
al
y
s
is
at
an
in
itial
s
tep
,
ch
ie
f
to
b
etter
b
eh
av
io
r
an
d
o
r
g
an
izat
io
n
ap
p
r
o
ac
h
es
f
o
r
s
k
i
n
ca
n
ce
r
lesi
o
n
s
[
1
4
]
.
L
ev
el
s
till
A
h
n
et
al.
,
[
1
5
]
lectu
r
ed
th
e
p
r
o
b
lem
atic
o
f
les
io
n
s
s
u
cc
es
s
to
th
e
i
m
ag
e
b
o
r
d
er
an
d
co
m
p
leted
b
etter
o
u
tco
m
es
in
e
v
alu
a
tio
n
to
Fa
n
et
al
.
,
tr
ain
in
g
[
1
6
]
,
th
e
y
d
id
n
o
t
co
n
tr
ac
t
w
i
th
d
y
e
d
iag
r
a
m
s
a
n
d
d
ar
k
an
g
le
s
in
th
e
p
ictu
r
es.
T
h
er
ef
o
r
e,
th
e
y
p
u
r
el
y
in
d
i
f
f
er
en
t
t
h
ese
i
n
s
u
f
f
icien
cie
s
m
a
n
u
a
ll
y
p
r
ev
io
u
s
l
y
a
n
y
p
r
e
-
p
r
o
ce
s
s
i
n
g
,
w
h
ic
h
i
s
a
d
ea
d
l
y
e
f
f
o
r
t.
Fro
m
Fig
u
r
e
1
d
is
p
la
y
t
h
e
d
er
m
o
s
co
p
y
i
m
ag
e
s
w
i
t
h
v
ar
iatio
n
s
in
le
s
io
n
f
ea
t
u
r
es.
Fig
u
r
e
1
.
Der
m
o
s
co
p
y
i
m
a
g
e
s
illu
s
tr
atin
g
v
ar
iat
io
n
s
i
n
les
io
n
ch
ar
ac
ter
is
tics
[
1
7
]
Sti
m
u
lated
b
y
t
h
e
f
o
r
ag
i
n
g
p
er
f
o
r
m
a
n
ce
o
f
s
w
ar
m
,
a
n
o
v
el
k
i
n
d
o
f
o
p
ti
m
izatio
n
tech
n
iq
u
e
m
en
tio
n
ed
to
as
t
h
e
ar
tific
ial
s
w
ar
m
i
n
tell
ig
e
n
ce
s
y
s
te
m
w
a
s
tech
n
o
lo
g
ical
l
y
ad
v
an
ce
d
[
1
8
]
.
O
w
in
g
to
it
s
ca
n
d
id
n
es
s
a
n
d
r
ea
l
a
f
f
l
u
en
c
e
o
f
o
p
er
atio
n
,
ASI
s
h
a
s
b
e
en
e
x
te
n
s
i
v
el
y
u
s
ed
to
r
eso
l
v
e
s
o
m
e
k
i
n
d
s
o
f
o
p
tim
izatio
n
d
i
f
f
icu
l
ties
.
Ne
v
er
th
ele
s
s
,
li
k
e
o
th
er
s
w
ar
m
in
telli
g
e
n
ce
o
p
ti
m
izatio
n
p
r
o
ce
d
u
r
e,
lik
e,
A
B
C
p
r
o
ce
s
s
is
g
o
o
d
.
Ou
r
ar
ea
in
th
is
ar
ticle
i
s
to
m
a
k
e
k
n
o
w
n
to
an
all
-
p
u
r
p
o
s
e
tec
h
n
iq
u
e
f
o
r
o
p
ti
m
al
s
k
i
n
lesi
o
n
s
.
T
h
er
e
ar
e
s
o
m
e
p
r
o
b
l
e
m
s
in
s
c
h
e
m
in
g
s
u
ch
a
w
id
e
-
r
an
g
i
n
g
alg
o
r
it
h
m
.
I
n
r
ea
l,
d
er
m
o
s
co
p
y
p
ict
u
r
e
d
atab
ases
ar
e
co
m
p
le
x
.
Ma
x
i
m
u
m
e
x
p
ec
ted
to
b
e
af
f
o
r
d
ed
b
y
t
h
e
k
i
n
d
o
f
C
P
U
in
v
e
s
ti
g
atio
n
o
f
t
h
e
i
m
ag
e
s
d
esig
n
ated
.
I
m
ag
e
s
o
r
ig
i
n
ated
w
i
th
m
a
n
y
ab
n
o
r
m
alitie
s
a
n
d
o
b
j
ec
ts
an
d
h
en
ce
it i
s
cr
itical
to
s
u
r
v
e
y
t
h
e
p
r
o
p
er
s
tag
es a
n
d
ap
p
r
o
ac
h
es
d
ef
in
ed
h
er
e
to
m
ed
icin
e
th
e
s
e
ir
r
eg
u
lar
itie
s
an
d
co
m
p
lete
a
n
ex
ac
t
an
a
l
y
s
i
s
.
T
h
e
o
p
ti
m
al
s
eg
m
e
n
tatio
n
w
it
h
s
u
itab
le
b
r
o
ad
-
m
in
d
ed
n
e
s
s
p
er
m
it
s
f
o
r
s
atis
f
ac
to
r
y
co
r
r
ec
tn
e
s
s
in
f
e
atu
r
e
s
e
g
m
en
tatio
n
w
h
ic
h
in
t
u
r
n
b
en
e
f
it
s
in
ta
k
e
f
u
ll
ad
v
an
tag
e
o
f
cla
s
s
i
f
icatio
n
p
r
ec
is
io
n
.
A
ll
-
p
u
r
p
o
s
e
o
u
tli
n
e
f
o
r
s
eg
m
e
n
tatio
n
m
u
s
t s
k
illed
o
f
o
v
er
co
m
i
n
g
t
h
ese
p
r
o
b
lem
s
.
2.
M
E
T
H
O
D
T
h
e
r
ea
l
co
n
tr
o
l
o
f
A
B
C
D
E
o
u
tli
n
e
i
s
co
m
p
leted
f
o
r
th
e
5
-
m
ec
h
a
n
is
m
s
w
it
h
th
e
s
u
b
s
eq
u
en
t
co
n
s
tr
u
ct
io
n
:
=
∑
5
=
1
(
1
)
w
h
er
e
w
i
an
d
p
i
ar
e
th
e
w
e
i
g
h
t
a
n
d
th
e
r
ate
co
m
p
ete
n
t
t
o
ea
ch
esti
m
atio
n
.
B
y
C
ap
d
e
h
o
u
r
at
et
al.
,
[
1
9
]
,
ac
co
r
d
in
g
to
Sto
lz
et
al.
,
[
2
0
]
.
I
m
p
o
r
tan
t
t
h
at
th
e
p
r
o
g
r
ess
o
f
t
h
e
s
k
in
ca
n
ce
r
lesi
o
n
s
o
v
er
ti
m
e
is
th
e
m
a
x
i
m
u
m
d
ata,
at
t
h
at
p
o
in
t
th
e
f
ea
tu
r
e
s
u
p
p
o
s
ed
b
y
t
h
e
p
ar
t
E
is
ap
p
lied
in
d
iv
id
u
al
f
o
r
g
r
ap
h
ic
ca
lcu
latio
n
b
y
d
er
m
ato
lo
g
i
s
t
s
at
w
h
at
ti
m
e
i
t
is
av
a
ilab
le.
T
h
e
as
s
es
s
m
en
t
o
f
A
B
C
DE
i
s
r
ea
ch
ed
r
esu
lti
n
g
(
1
)
w
it
h
t
h
e
p
o
in
ts
to
clas
s
i
f
y
t
h
e
co
lo
r
in
g
in
to
3
-
ca
s
es [
1
9
]
:
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
n
ew s
w
a
r
m
in
tellig
en
ce
in
fo
r
ma
tio
n
tech
n
iq
u
e
f
o
r
imp
r
o
vin
g
…
(
H.
J.
A
b
d
)
5705
<
4
.
75
(
Hea
lth
y
Mo
le
s
)
4
.
75
≤
≤
5
.
75
(
Su
s
p
icio
u
s
Mo
les)
>
5
.
75
(
Un
h
ea
l
th
y
Mo
les)
I
n
th
i
s
s
tu
d
y
,
w
e
ex
p
lai
n
an
a
u
to
m
at
ic
d
etec
tio
n
o
f
th
e
ASI
s
to
v
ar
iet
y
t
h
e
r
eq
u
ir
ed
in
v
es
tig
atio
n
to
s
eg
m
e
n
t
th
e
s
k
in
ca
n
ce
r
o
f
th
e
P
H
2
,
I
SIC
(
2
0
1
6
,
2
0
1
7
,
2
0
1
8
)
.
T
h
ese
ap
p
r
o
ac
h
es
ar
e
a
r
e
s
o
lu
t
io
n
-
p
r
o
tecti
v
e
w
h
ic
h
lead
s
to
s
tu
d
y
h
ig
h
-
l
ev
el
to
p
o
g
r
ap
h
ies
an
d
i
m
p
r
o
v
e
th
e
s
e
g
m
en
ta
tio
n
r
o
u
ti
n
e.
W
e
h
av
e
ap
p
lie
d
m
o
tiv
a
tio
n
f
u
n
c
tio
n
i
n
th
e
p
l
an
n
ed
s
eg
m
e
n
tatio
n
n
e
t
w
o
r
k
s
to
class
i
f
y
s
ep
ar
atel
y
p
i
x
el
i
n
t
h
e
d
er
m
o
s
co
p
y
i
m
a
g
e
in
to
2
-
p
ar
ts
(
s
k
i
n
lesi
o
n
s
an
d
n
on
-
le
s
io
n
s
)
.
T
h
e
I
SIC
2
0
1
8
d
ata
b
ases
co
n
tai
n
s
2
,
5
9
4
im
a
g
es
w
it
h
t
h
e
co
n
f
o
r
m
i
n
g
GT
d
is
g
u
is
e
s
m
ar
k
ed
b
y
p
r
o
f
ess
io
n
al
d
er
m
ato
lo
g
is
t
s
.
T
h
e
j
u
s
ti
f
icatio
n
a
n
d
ch
alle
n
g
in
g
g
r
o
u
p
s
co
m
p
r
is
e
1
,
0
0
0
im
a
g
es,
w
it
h
o
u
t
g
r
o
u
n
d
tr
u
th
.
I
n
o
u
r
e
x
p
er
i
m
e
n
tatio
n
s
,
w
e
u
s
ed
8
0
%
o
f
t
h
e
tr
ain
i
n
g
tr
ad
itio
n
al
o
f
t
h
e
I
SIC
2
0
1
8
d
ataset
f
o
r
tr
ain
i
n
g
an
d
2
0
%
f
o
r
j
u
s
tif
icatio
n
as
p
lan
n
ed
in
[
2
1
]
.
I
n
tr
y
,
I
S
B
I
2
0
1
7
d
atab
ases
w
er
e
s
ep
ar
ated
in
to
w
o
r
k
o
u
t,
j
u
s
tif
ica
tio
n
an
d
ch
a
llen
g
i
n
g
g
r
o
u
p
s
w
it
h
2
0
0
0
im
ag
e
s
.
Me
m
o
th
at
w
e
tr
ain
ed
o
u
r
t
y
p
i
ca
l
w
it
h
I
SIC
2
0
1
8
t
r
a
i
n
i
n
g
g
r
o
u
p
s
a
n
d
e
s
t
i
m
a
t
e
d
o
u
r
t
y
p
i
c
a
l
o
n
I
S
B
I
2
0
1
6
,
I
S
B
I
2
0
1
7
a
s
s
e
s
s
m
e
n
t
a
n
d
I
S
I
C
2
0
1
8
v
a
l
i
d
a
t
i
o
n
g
r
o
u
p
s
.
W
e
ex
p
er
ien
ce
d
o
u
r
ap
p
r
o
r
esu
lts
o
n
t
h
e
I
SIC 2
0
1
8
[
2
2
]
d
ata
b
ase
co
m
p
r
is
in
g
s
k
i
n
ca
n
ce
r
le
s
io
n
s
[
2
3
]
in
cl
u
d
in
g
lesi
o
n
i
m
ag
e
s
f
r
o
m
th
r
ee
m
o
d
u
le
s
.
T
h
e
ca
lcu
lati
o
n
s
tan
d
ar
d
s
ar
e
th
e
s
a
m
e
as
th
o
s
e
co
n
v
er
s
ed
i
n
th
e
o
p
ti
m
al
s
eg
m
e
n
tatio
n
s
ec
to
r
ex
ce
p
t
t
h
at
t
h
er
e
m
a
y
b
e
s
o
m
e
i
m
ag
e
s
w
h
er
e
t
h
e
tar
g
et
f
ea
t
u
r
e
d
o
es
n
o
t
o
cc
u
r
,
an
d
th
e
alg
o
r
ith
m
s
h
o
u
ld
n
o
t
b
e
r
esu
lt
all.
Fig
u
r
e
2
A
b
lo
ck
d
iag
r
a
m
o
f
th
e
s
tep
s
in
th
e
to
tal
p
r
o
ce
s
s
co
m
p
lica
ted
is
ill
u
s
tr
ated
.
Fig
u
r
e
2
.
T
h
e
f
lo
w
d
iag
r
a
m
o
f
th
e
s
tep
s
i
n
th
e
e
n
tire
p
r
o
ce
s
s
co
m
p
lica
ted
is
ill
u
s
tr
ated
E
s
ti
m
a
tio
n
o
f
th
e
o
u
tco
m
es
o
f
m
ela
n
o
m
a
le
s
io
n
s
s
e
g
m
en
tati
o
n
p
r
o
ce
d
u
r
e
ca
n
b
e
s
i
m
ilar
l
y
m
u
ltip
ar
t
an
d
ar
g
u
ab
le
in
r
elatio
n
s
o
f
r
ec
o
r
d
in
g
th
e
f
i
n
es
t
m
et
h
o
d
.
T
h
e
b
o
u
n
d
ar
y
o
f
s
k
in
ca
n
ce
r
lesi
o
n
s
e
g
m
e
n
tatio
n
d
if
f
er
s
ex
te
n
s
iv
el
y
a
m
o
n
g
p
r
o
f
es
s
io
n
al
s
.
E
s
ti
m
atio
n
ap
p
r
o
ac
h
es c
a
n
b
e
ass
e
s
s
m
e
n
t i
s
d
ep
en
d
in
g
o
n
a
s
co
r
e
o
n
co
n
d
itio
n
t
h
at
b
y
h
u
m
a
n
f
o
cu
s
s
es.
T
h
e
ex
p
er
i
m
e
n
tal
f
o
r
m
at
f
o
r
co
u
n
tin
g
m
u
s
t
b
e
co
r
r
ec
tl
y
clea
r
f
o
r
e
x
a
m
p
le,
th
e
co
u
n
ti
n
g
tech
n
iq
u
e
is
d
ep
e
n
d
ab
le
a
m
o
n
g
c
h
an
g
ed
s
u
b
j
ec
ts
.
A
n
i
m
p
ar
tial a
s
s
e
s
s
m
en
t is
d
ep
en
d
in
g
o
n
er
r
o
r
q
u
an
tit
ies
d
esi
g
n
ed
w
h
ile
r
el
atin
g
th
e
a
u
to
m
ated
b
o
r
d
er
w
ith
t
h
e
n
o
r
m
al
m
a
n
u
al
m
ela
n
o
m
a
le
s
io
n
r
eg
io
n
s
ed
g
e.
Sev
er
al
o
b
j
ec
tiv
e
ass
es
s
m
en
t p
r
o
ce
d
u
r
es a
r
e
d
em
ar
ca
t
ed
in
th
e
w
o
r
k
s
[
2
4
]
.
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.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
:
5
7
0
3
-
5
7
0
8
5706
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
o
esti
m
ate
o
u
r
m
et
h
o
d
an
d
th
e
p
r
o
j
ec
ted
o
p
tim
al
s
eg
m
e
n
tatio
n
tec
h
n
iq
u
e,
w
e
ap
p
l
y
f
o
u
r
v
ar
io
u
s
d
atab
ases
.
T
h
r
ee
o
f
th
ese
d
atab
ases
ar
e
d
esig
n
a
ted
f
r
o
m
(
I
S
B
I
)
in
2
0
1
6
,
2
0
1
7
an
d
2
0
1
8
p
ar
ts
.
T
h
e
p
ictu
r
es
in
I
SB
I
(
2
0
1
6
,
2
0
1
7
)
h
av
e
an
ex
ce
s
s
iv
e
q
u
a
n
tit
y
o
f
m
atc
h
i
n
g
m
atter
s
(
i
n
telli
g
e
n
t
m
ater
ia
ls
)
.
So
,
w
e
tr
y
a
n
d
ex
a
m
in
at
io
n
o
u
r
s
y
s
te
m
o
n
ea
ch
o
f
th
e
s
e
d
atab
ases
.
A
d
d
itio
n
al
d
atab
ase
t
h
at
h
a
s
b
ee
n
e
x
t
en
s
i
v
el
y
ap
p
lied
i
n
d
er
m
o
s
co
p
ic
s
t
u
d
y
is
P
H2
,
th
e
co
d
es
ar
e
ex
ec
u
ted
in
M
A
T
L
A
B
R
2
0
1
7
b
,
an
d
all
ex
p
er
im
en
tatio
n
s
ar
e
d
o
n
e
o
n
r
e
m
o
v
al
a
n
d
an
y
n
o
is
e.
T
o
esti
m
a
te
t
h
e
s
e
g
m
en
ta
tio
n
tech
n
iq
u
e
q
u
an
titati
v
el
y
,
w
e
in
cl
u
d
e
DS
C
,
j
ac
ca
r
d
s
i
m
ila
r
it
y
i
n
d
e
x
(
J
SI)
,
a
n
d
th
e
p
ar
am
eter
s
ar
e
A
cc
u
r
ac
y
(
A
cc
)
,
Sen
s
iti
v
it
y
(
Sen
s
)
,
an
d
Sp
ec
if
icit
y
(
S
p
ec
)
as
d
esig
n
ated
in
[
2
5
]
.
T
h
e
o
u
tco
m
e
s
o
f
c
h
a
llen
g
in
g
th
e
p
r
o
j
ec
ted
o
p
tim
a
l
s
e
g
m
e
n
tat
io
n
m
et
h
o
d
o
n
ch
an
g
ed
d
er
m
o
s
co
p
ic
d
atab
ases
ar
e
ass
o
ciate
d
to
t
h
e
o
th
er
u
p
-
to
-
d
ate
m
et
h
o
d
s
a
n
d
r
ep
o
r
ted
in
T
a
b
les
1
-
4
.
W
h
e
n
ac
co
m
p
li
s
h
ed
o
n
th
e
P
H2
,
o
u
r
m
e
th
o
d
o
u
tcla
s
s
es
t
h
e
o
th
er
ap
p
r
o
ac
h
es
in
T
ab
le
1
.
P
H2
is
a
co
m
p
ar
ati
v
el
y
m
i
n
o
r
an
d
ea
s
y
d
atab
ase
th
at
all
its
p
ictu
r
e
s
w
er
e
tak
e
n
b
y
a
co
n
s
ta
n
t
ac
h
iev
e
m
en
t
ar
r
an
g
e
m
en
t,
an
d
it
is
n
o
r
m
al
to
r
ea
ch
esti
m
atio
n
m
etr
ics
as
h
i
g
h
as
i
n
f
o
r
m
ed
v
al
u
es
i
n
T
ab
le
1.
T
h
e
I
SB
I
2
0
1
6
d
atab
ase,
o
u
tco
m
e
s
ar
e
d
escr
ib
ed
f
o
r
b
o
th
p
h
y
s
ical
ac
ti
v
it
y
a
n
d
ch
a
llen
g
in
g
g
r
o
u
p
s
.
A
s
u
n
d
er
s
to
o
d
in
T
ab
le
2
,
o
u
r
alg
o
r
ith
m
o
u
tclas
s
es
t
h
e
o
th
er
f
a
il
e
d
d
e
t
ec
t
i
o
n
-
b
a
s
e
d
a
lg
o
r
i
th
m
s
[
1
5
,
1
6
]
an
d
o
u
r
p
r
ev
i
o
u
s
te
c
h
n
i
q
u
e
in
Z
am
an
i
e
t
al
.
[
2
6
]
.
A
o
n
th
e
I
S
B
I
2
0
1
6
t
r
a
in
g
r
o
u
p
.
I
t
s
r
o
u
t
in
e
i
s
a
li
t
tle
o
f
p
o
o
r
e
r
q
u
al
i
ty
th
an
th
a
t
o
f
Yu
an
e
t
a
l
.
[
2
7
]
,
w
h
i
ch
is
d
e
p
e
n
d
in
g
o
n
D
C
N
Ns
.
Ass
es
s
m
en
t
o
f
t
h
e
I
SB
I
2
0
1
6
a
s
s
es
s
m
en
t
g
r
o
u
p
is
in
f
o
r
m
ed
i
n
T
ab
le
2
.
I
n
an
o
th
er
w
a
y
,
w
e
u
s
e
o
u
r
tec
h
n
iq
u
e
o
n
th
e
I
SB
I
(
2
0
1
7
,
2
0
1
8
)
ex
am
g
r
o
u
p
a
n
d
cu
r
r
en
t t
h
e
o
u
tco
m
es i
n
T
ab
le
s
3
an
d
4
.
Ou
r
s
tr
ateg
ic
p
r
o
ce
d
u
r
e
r
ea
ch
es
av
er
ag
e
J
ac
v
al
u
e
o
f
(
9
6
.
9
8
,
9
4
.
1
2
,
8
8
.
0
0
an
d
9
0
.
0
0
)
,
J
SI
o
f
(
9
4
.
5
5
,
8
9
.
1
1
,
7
6
.
2
7
an
d
7
9
.
8
8
)
o
n
6
0
0
i
m
ag
e
s
o
f
t
h
e
I
SB
I
2
0
1
7
an
d
I
SB
I
2
0
1
8
test
g
r
o
u
p
,
t
h
e
v
ar
ian
ce
s
f
r
o
m
o
th
er
h
ig
h
er
co
m
p
eti
to
r
s
in
T
ab
le
4
ar
e
ass
ess
m
e
n
t
p
ar
a
m
e
ter
s
h
a
v
e
t
h
e
s
a
m
e
d
r
if
t.
T
h
e
o
u
tco
m
es
d
e
m
o
n
s
tr
ate
t
h
at
th
e
p
r
o
j
ec
ted
tech
n
iq
u
e
ca
n
r
ea
lize
a
g
o
o
d
s
e
g
m
e
n
tat
io
n
r
o
u
ti
n
e,
eq
u
i
v
ale
n
t
to
t
h
e
o
th
er
p
r
in
cip
le
o
f
ap
p
r
o
ac
h
es,
w
h
ic
h
r
e
g
u
lar
l
y
ar
e
d
ep
en
d
in
g
o
n
D
C
NN
s
an
d
A
B
C
al
g
o
r
ith
m
.
No
te
th
at
t
h
e
I
SB
I
2
0
1
6
,
I
SB
I
2
0
1
7
an
d
I
SB
I
2
0
1
8
p
h
y
s
ical
ac
tiv
it
y
g
r
o
u
p
s
w
er
e
ap
p
lied
to
tr
ain
th
e
f
ailed
p
lo
ts
g
iv
en
i
n
T
a
b
les
2
-
4
.
Fig
u
r
e
3
illu
s
tr
atio
n
s
t
h
e
s
e
g
m
en
tatio
n
o
u
t
co
m
es
a
n
d
th
eir
s
alien
c
y
p
lo
t
s
(
th
e
m
id
d
le
r
o
w
)
.
T
h
e
g
r
ee
n
ch
ar
ac
ter
izes
t
h
e
GT
class
if
i
ca
tio
n
,
r
ed
f
o
r
t
h
e
cu
r
v
e
o
f
th
e
le
s
io
n
’
s
p
r
i
m
ar
y
co
v
er
,
a
n
d
th
e
b
lu
e
c
u
r
v
e
s
d
is
p
la
y
th
e
b
o
u
n
d
ar
ies o
f
t
h
e
d
is
tin
g
u
is
h
ed
last
s
e
g
m
e
n
tatio
n
th
r
o
u
g
h
lev
e
l g
r
o
u
p
p
r
o
g
r
ess
io
n
.
T
ab
le
1
.
R
esu
lts
o
f
p
h
2
d
ataset
s
eg
m
e
n
tatio
n
s
T
e
c
h
n
i
q
u
e
A
c
c
S
p
e
c
S
e
n
Jac
JS
I
P
e
n
n
i
s
i
[
2
8
]
8
9
.
4
9
7
.
0
1
7
1
.
1
3
-
-
B
a
r
a
t
a
[
2
9
]
-
-
-
9
1
.
4
-
A
h
n
[
1
5
]
9
3
.
6
7
-
7
8
.
7
6
8
9
.
3
4
-
Z
a
man
i
[
2
6
]
9
6
.
5
8
9
8
.
0
1
9
2
.
9
4
9
2
.
9
0
8
5
.
8
8
Jah
a
n
i
f
a
r
[
3
0
]
9
7
.
3
3
9
8
.
5
5
9
7
.
5
2
9
5
.
2
2
9
2
.
6
7
P
r
o
p
o
se
d
9
8
.
2
3
9
8
.
4
5
9
7
.
8
9
9
6
.
9
8
9
4
.
5
5
T
ab
le
2
.
R
esu
lts
o
f
I
SB
I
2
0
1
6
tr
ain
s
et
s
e
g
m
en
tatio
n
s
T
e
c
h
n
i
q
u
e
A
c
c
S
p
e
c
S
e
n
Jac
JS
I
F
a
n
[
2
2
]
9
1
.
4
0
-
7
1
.
1
3
8
1
.
1
1
-
A
h
n
[
2
3
]
-
-
-
8
3
.
5
5
-
Z
a
man
i
[
2
6
]
9
3
.
5
1
9
8
.
0
7
8
3
.
2
8
9
.
9
0
8
2
.
0
8
Jah
a
n
i
f
a
r
[
3
0
]
9
4
.
2
3
9
8
.
4
5
8
8
.
7
2
9
1
.
7
2
8
5
.
5
7
P
r
o
p
o
se
d
9
5
.
3
8
9
8
.
0
0
9
0
.
9
5
9
4
.
1
2
8
9
.
1
1
T
ab
le
3
.
R
esu
lts
o
f
I
SB
I
2
0
1
7
test
s
et
s
e
g
m
en
tatio
n
s
T
e
c
h
n
i
q
u
e
A
c
c
S
p
e
c
S
e
n
Jac
JS
I
Y
u
a
n
[
2
7
]
9
3
.
4
0
9
7
.
0
3
8
2
.
5
8
4
.
0
9
7
6
.
5
A
h
n
[
1
5
]
9
3
.
4
9
4
.
3
2
8
0
.
1
8
4
.
7
7
5
.
8
Z
a
man
i
[
2
6
]
9
3
.
5
1
9
5
.
6
0
8
2
.
2
2
8
5
.
9
8
8
1
.
0
0
Jah
a
n
i
f
a
r
[
3
0
]
9
3
.
2
3
9
8
.
5
0
8
1
.
3
3
8
3
.
9
2
7
4
.
8
0
P
r
o
p
o
se
d
9
4
.
4
0
9
8
.
2
0
9
2
.
6
2
8
8
.
0
0
7
6
.
2
7
T
ab
le
4
.
R
esu
lts
o
f
I
SB
I
2
0
1
8
test
s
et
s
e
g
m
en
tatio
n
s
T
e
c
h
n
i
q
u
e
A
c
c
S
p
e
c
S
e
n
Jac
JS
I
Y
u
a
n
[
2
7
]
9
4
.
6
6
9
6
.
3
9
8
3
.
5
0
8
5
.
0
0
7
7
.
5
5
A
h
n
[
1
5
]
9
3
.
2
2
9
4
.
0
0
8
2
.
1
8
8
3
.
7
7
7
7
.
1
1
Z
a
man
i
[
2
6
]
9
3
.
8
8
9
5
.
0
0
8
3
.
7
5
8
5
.
0
0
8
2
.
4
4
Jah
a
n
i
f
a
r
[
3
0
]
9
3
.
0
0
9
7
.
4
4
8
2
.
3
0
8
4
.
9
0
7
8
.
9
7
P
r
o
p
o
se
d
9
4
.
0
0
9
8
.
6
5
9
1
.
7
6
9
0
.
0
0
7
9
.
8
8
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
n
ew s
w
a
r
m
in
tellig
en
ce
in
fo
r
ma
tio
n
tech
n
iq
u
e
f
o
r
imp
r
o
vin
g
…
(
H.
J.
A
b
d
)
5707
F
ig
u
re
3
.
S
a
m
p
li
n
g
v
a
lu
a
ti
o
n
o
f
th
e
p
ro
p
o
se
d
se
g
m
e
n
tatio
n
tec
h
n
iq
u
e
.
T
h
e
(
a
an
d
b
)
r
o
w
i
n
F
i
g
u
r
e
3
s
h
o
w
s
f
o
u
r
e
x
a
m
p
le
s
o
f
r
is
k
y
ca
s
e
s
w
it
h
t
h
eir
co
n
f
o
r
m
i
n
g
t
h
e
A
B
C
tech
n
iq
u
e
t
h
r
o
u
g
h
m
o
d
er
n
AS
I
w
it
h
m
ed
ia
n
f
i
lter
.
T
h
e
(
c
an
d
d
)
r
o
w
i
n
F
i
g
u
r
e
3
illu
s
tr
ate
s
th
at
d
ata
f
o
r
f
o
u
r
s
i
m
p
le
ca
s
e
s
.
I
n
w
h
o
ll
y
p
ict
u
r
es th
e
r
ed
,
b
lu
e,
an
d
g
r
ee
n
o
u
tl
in
es c
h
ar
ac
ter
ize
th
e
b
o
r
d
er
s
o
f
GT
.
4.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
T
h
e
m
eth
o
d
s
o
b
j
ec
tiv
e
at
a
u
t
o
m
a
tin
g
t
h
e
c
h
ec
k
u
p
o
f
s
k
i
n
ca
n
ce
r
lesi
o
n
s
,
co
n
v
e
n
tio
n
a
ll
y
co
m
p
leted
b
y
d
er
m
ato
lo
g
i
s
ts
,
ar
e
th
e
s
e
d
a
y
s
lib
er
ated
b
y
t
h
e
d
ee
p
-
lear
n
in
g
-
b
u
il
t
ap
p
r
o
ac
h
es.T
h
ese
ap
p
r
o
ac
h
es
ar
e
th
e
r
ec
o
r
d
p
r
ec
is
e
an
d
s
ca
la
b
le,
th
en
th
e
y
n
ee
d
lar
g
e
tr
ain
i
n
g
d
atab
ases
a
n
d
t
h
u
s
t
h
eir
ap
p
licab
ilit
y
i
n
d
er
m
ato
lo
g
y
is
co
o
p
er
ated
b
y
th
e
s
co
p
e
o
f
t
h
e
p
u
b
licl
y
d
er
m
ato
lo
g
ical
d
atab
ases
,
w
h
ic
h
ar
e
f
r
eq
u
e
n
tl
y
s
lig
h
t
an
d
co
m
p
r
is
e
b
lo
ck
i
n
g
s
.
W
e
p
r
esen
t
a
r
eso
lu
t
io
n
f
o
r
th
is
p
r
o
b
lem
atic
t
h
at
d
ep
en
d
o
n
c
ar
ef
u
l
i
n
f
o
r
m
atio
n
d
ec
o
n
ta
m
i
n
atio
n
t
h
at
eli
m
i
n
at
es
co
m
m
u
n
al
o
b
s
tr
u
c
tio
n
s
f
r
o
m
d
er
m
o
s
co
p
ic
p
ictu
r
es
a
n
d
i
n
cr
ea
s
e
t
h
at
ap
p
lies
th
e
m
o
d
er
n
p
r
o
ce
d
u
r
e
o
f
s
w
ar
m
in
te
lli
g
en
ce
tec
h
n
iq
u
e
in
f
o
r
m
atio
n
g
en
er
atio
n
to
en
h
a
n
ce
in
f
o
r
m
a
tio
n
b
alan
ce
d
n
es
s
.
T
h
e
y
p
r
o
v
e
th
e
ef
f
icien
c
y
o
f
o
u
r
s
y
s
te
m
o
n
th
e
s
k
i
n
lesi
o
n
s
s
e
g
m
e
n
tatio
n
ta
s
k
.
T
h
e
ass
ess
m
e
n
t
o
f
th
e
r
es
u
lt
is
d
ep
en
d
in
g
o
n
th
e
to
ta
l
(
A
cc
,
Se
n
,
Sp
ec
,
J
ac
an
d
J
SI)
o
f
th
e
s
e
g
m
en
tat
io
n
a
n
d
ca
talo
g
u
in
g
o
n
th
e
ch
ec
k
u
p
g
r
o
u
p
.
I
n
th
i
s
f
ield
,
it
i
s
v
er
y
s
i
g
n
i
f
ica
n
t
n
o
t
to
er
r
o
r
m
ali
g
n
a
n
ce
d
esp
ite
t
h
e
f
ac
t
b
ein
g
ca
p
ab
le
o
f
s
u
i
tab
l
y
r
ec
o
g
n
ize
b
en
ig
n
les
io
n
s
a
s
m
u
c
h
as
p
r
o
b
a
b
le.
T
h
e
g
o
al
is
to
m
ar
k
h
ig
h
es
t
Sen
w
h
er
ea
s
ad
j
u
s
tin
g
to
r
is
e
Sp
ec
,
th
u
s
r
aisi
n
g
th
e
t
o
tal
A
cc
.
T
h
e
w
o
r
k
co
n
f
ir
m
atio
n
s
a
d
if
f
er
e
n
t
p
r
o
ce
s
s
f
o
r
lesi
o
n
s
o
p
ti
m
al
s
e
g
m
en
tatio
n
th
at
co
u
ld
b
e
f
u
n
ctio
n
al
to
a
v
ar
iati
o
n
o
f
i
m
a
g
es
w
i
t
h
ch
an
g
ed
p
o
s
s
es
s
io
n
s
a
n
d
in
s
u
f
f
icie
n
cies i
s
p
r
o
p
o
s
ed
w
it
h
m
u
ltis
tep
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e.
RE
F
E
R
E
NC
E
S
[1
]
Ka
ra
b
o
g
a
D,
A
k
a
y
B,
“
A
c
o
m
p
a
ra
ti
v
e
stu
d
y
o
f
a
rti
f
ici
a
l
b
e
e
c
o
lo
n
y
a
lg
o
rit
h
m
,
”
Ap
p
l
M
a
th
Co
mp
u
t,
v
o
l
.
2
1
4
,
p
p
.
1
0
8
-
1
1
3
,
2
0
0
9
.
[2
]
A
ri
f
in
A
Z,
e
t
a
l,
,
“
A
u
to
m
a
ti
c
i
m
a
g
e
slice
m
a
r
k
in
g
p
ro
p
a
g
a
ti
o
n
o
n
se
g
m
e
n
tatio
n
o
f
d
e
n
tal
CBC
T
,
”
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
ti
o
n
,
Co
m
p
u
t
in
g
,
El
e
c
tro
n
ics
a
n
d
Co
n
tro
l)
,
v
o
l
.
17
,
n
o
9
,
p
p
.
3
2
1
8
-
3
2
2
5
,
20
1
9.
[3
]
J.
M
u
th
u
sw
a
m
y
,
B.
K
a
n
m
a
n
i,
“
Op
ti
m
iza
ti
o
n
Ba
se
d
L
iv
e
r
Co
n
to
u
r
Ex
trac
ti
o
n
o
f
A
b
d
o
m
in
a
l
CT
I
m
a
g
e
s,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
),
v
o
l.
8
,
n
o
.
6
,
p
p
.
5
0
6
1
-
5
0
7
0
,
2
0
1
8
.
[4
]
C
a
o
Y,
L
u
Y,
P
a
n
X
,
S
u
n
N,
“
A
n
i
m
p
ro
v
e
d
g
lo
b
a
l
b
e
st
g
u
id
e
d
a
rti
f
ici
a
l
b
e
e
c
o
lo
n
y
a
lg
o
rit
h
m
f
o
r
c
o
n
ti
n
u
o
u
s
o
p
ti
m
iza
ti
o
n
p
ro
b
lem
s,”
Clu
ste
r Co
mp
u
t
.
,
v
o
l
.
2
2
,
p
p
.
3
0
1
1
-
3
0
1
9
,
2
0
1
8
.
[5
]
M.
A
lb
k
o
sh
,
e
t
a
l.
,
“
Op
ti
m
iza
ti
o
n
o
f
d
isc
re
te
wa
v
e
let
tran
s
f
o
r
m
fe
a
tu
re
s
u
sin
g
a
rti
f
icia
l
b
e
e
c
o
lo
n
y
a
lg
o
rit
h
m
f
o
r
tex
tu
re
im
a
g
e
c
las
si
f
ic
a
ti
o
n
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
E
n
g
i
n
e
e
rin
g
(
I
J
ECE
),
v
o
l
.
9
,
no
.
6
,
p
p
.
5
2
5
2
-
5
2
6
2
,
2
0
1
9
.
[6
]
Ka
ra
b
o
g
a
D,
B
a
stu
rk
B,
“
On
th
e
p
e
rf
o
r
m
a
n
c
e
o
f
a
rti
f
icia
l
b
e
e
c
o
lo
n
y
(
ABC)
a
l
g
o
ri
th
m
,
”
Ap
p
l
S
o
ft
Co
mp
u
t
,
v
o
l.
8
,
p
p
.
6
8
7
-
6
9
7
,
2
0
0
8
.
[7
]
M
e
z
u
ra
-
M
o
n
t
e
s
E,
Ce
ti
n
a
-
Do
m
ín
g
u
e
z
O,
“
E
m
p
iri
c
a
l
a
n
a
l
y
sis
o
f
a
m
o
d
if
ied
a
rti
f
icia
l
b
e
e
c
o
lo
n
y
f
o
r
c
o
n
stra
in
e
d
n
u
m
e
rica
l
o
p
ti
m
iza
ti
o
n
,
”
Ap
p
l
M
a
th
C
o
mp
u
t,
v
o
l.
2
1
8
,
n
o
.
2
2
,
p
p
.
1
0
9
4
3
-
1
0
9
7
3
,
2
0
1
2
.
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.
10
,
No
.
6
,
Dec
em
b
er
2
0
2
0
:
5
7
0
3
-
5
7
0
8
5708
[8
]
M
a
L
,
Zh
u
Y,
Z
h
a
n
g
D,
Ni
u
B,
“
A
h
y
b
rid
a
p
p
ro
a
c
h
t
o
a
rti
f
icia
l
b
e
e
c
o
l
o
n
y
a
lg
o
rit
h
m
,
”
Ne
u
ra
l
Co
mp
u
t
A
p
p
l
,
v
o
l.
2
7
,
n
o
.
2
,
p
p
.
3
8
7
-
4
0
9
,
2
0
1
6
.
[9
]
Kira
n
M
.
S
,
Ha
k
li
H,
G
u
n
d
u
z
M
,
Ug
u
z
H,
“
A
rti
f
icia
l
b
e
e
c
o
lo
n
y
a
lg
o
rit
h
m
w
it
h
v
a
riab
le
se
a
rc
h
stra
teg
y
f
o
r
c
o
n
ti
n
u
o
u
s
o
p
t
im
iza
ti
o
n
,
”
I
n
f
S
c
i,
v
o
l.
3
0
0
,
p
p
.
1
4
0
-
1
5
7
,
2
0
1
5
.
[
1
0
]
Al
-
m
a
s
n
i
M
.
A
,
A
l
-
a
n
t
a
r
i
M
.
A
,
C
h
o
i
M
-
T
,
H
a
n
S
-
M
,
K
i
m
T
-
S
(
2
0
1
8
)
S
k
i
n
l
e
s
i
o
n
s
e
g
m
e
n
t
a
t
i
o
n
i
n
d
e
r
m
o
s
c
o
p
y
i
m
a
g
e
s
v
i
a
d
e
e
p
f
u
l
l
r
e
s
o
l
u
t
i
o
n
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
s
,
”
C
o
m
p
u
t
M
e
t
h
o
d
s
P
r
o
g
B
i
o
m
e
d
,
v
o
l
.
1
6
2
,
p
p
.
2
2
1
-
2
3
1
,
2
0
1
8
.
[1
1
]
V
a
ish
a
li
Ja
isw
a
l,
V
a
rsh
a
S
h
a
rm
a
,
S
u
n
it
a
V
a
rm
a
,
“
A
n
im
p
le
m
e
n
ta
ti
o
n
o
f
n
o
v
e
l
g
e
n
e
ti
c
b
a
se
d
c
lu
st
e
rin
g
a
lg
o
rit
h
m
f
o
r
c
o
lo
r
im
a
g
e
se
g
m
e
n
tatio
n
,
”
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
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.
1
7
,
n
o
.
3
,
p
p
.
1
4
6
1
-
1
4
6
7
,
20
19
.
[1
2
]
Ka
n
g
D,
Ki
m
S
,
P
a
rk
S
,
“
F
lo
w
-
g
u
id
e
d
h
a
ir
re
m
o
v
a
l
f
o
r
a
u
to
m
a
ted
sk
in
les
io
n
id
e
n
ti
f
ica
ti
o
n
,
”
M
u
lt
i
me
d
T
o
o
ls A
p
p
l
,
v
o
l.
7
7
,
n
o
.
8
,
p
p
.
9
8
9
7
-
9
9
0
8
,
2
0
1
8
.
[
1
3
]
S
a
n
k
a
r
a
n
S
,
e
t
a
l
.
,
“
A
c
o
m
p
a
r
a
t
i
v
e
a
s
s
e
s
s
m
e
n
t
o
f
s
e
g
m
e
n
t
a
t
i
o
n
s
o
n
s
k
i
n
l
e
s
i
o
n
t
h
r
o
u
g
h
v
a
r
i
o
u
s
e
n
t
r
o
p
y
a
n
d
s
i
x
s
i
g
m
a
t
h
r
e
s
h
o
l
d
s
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
S
M
A
C
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
V
i
s
i
o
n
a
n
d
B
i
o
-
E
n
g
i
n
e
e
r
i
n
g
,
p
p
.
1
7
9
-
1
8
8
,
2
0
1
8
.
[1
4
]
Ja
y
a
n
th
i
M
u
th
u
sw
a
m
y
,
B
Ka
n
m
a
n
i,
“
Op
ti
m
iz
a
ti
o
n
Ba
se
d
L
iv
e
r
Co
n
t
o
u
r
Ex
trac
ti
o
n
o
f
A
b
d
o
m
in
a
l
CT
I
m
a
g
e
s,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
),
v
o
l
8
,
n
o
6
,
p
p
.
5
0
6
1
-
5
0
7
0
,
2
0
1
8
.
[
1
5
]
E.
A
h
n
,
J.
Kim
,
L
.
Bi,
A
.
Ku
m
a
r,
C.
L
i,
M
.
F
u
lh
a
m
,
a
n
d
D.
D.
F
e
n
g
,
“
S
a
li
e
n
c
y
-
b
a
s
e
d
L
e
sio
n
S
e
g
m
e
n
tatio
n
v
ia
B
a
c
k
g
r
o
u
n
d
D
e
t
e
c
t
i
o
n
i
n
D
e
r
m
o
s
c
o
p
i
c
I
m
a
g
e
s
,
”
I
E
E
E
J
.
B
i
o
m
e
d
.
H
e
a
l
.
i
n
f
o
r
m
a
t
i
c
s
,
v
o
l
.
2
1
,
n
o
.
6
,
p
p
.
1
6
8
5
-
1
6
9
3
,
2
0
1
7
.
[1
6
]
H.
F
a
n
,
F
.
X
ie,
Y.
L
i,
Z.
Jia
n
g
,
a
n
d
J.
L
iu
,
“
A
u
to
m
a
ti
c
se
g
m
e
n
tatio
n
o
f
d
e
rm
o
sc
o
p
y
i
m
a
g
e
s
u
sin
g
sa
li
e
n
c
y
c
o
m
b
in
e
d
w
it
h
Otsu
t
h
re
sh
o
l
d
,
”
Co
mp
u
t
.
Bi
o
l.
M
e
d
.
,
v
o
l.
8
5
,
p
p
.
7
5
-
8
5
,
2
0
1
7
.
[1
7
]
IS
IC,
"
IS
IC
A
rc
h
iev
e
:
T
h
e
In
tern
a
ti
o
n
a
l
S
k
in
Im
a
g
in
g
Co
ll
a
b
o
ra
ti
o
n
:
M
e
lan
o
m
a
P
ro
jec
t,
"
IS
IC,
5
Ja
n
2
0
1
6
.
[
On
li
n
e
]
.
A
v
a
il
a
b
l
e
:
h
tt
p
s://
isic
-
a
rc
h
iv
e
.
c
o
m
/#
.
[1
8
]
S
.
A
.
Y
a
se
a
r,
a
n
d
K.
R.
Ku
-
M
a
h
a
m
u
d
,
“
No
n
-
d
o
m
in
a
ted
so
rti
n
g
Ha
rris’s
h
a
w
k
m
u
lt
i
-
o
b
jec
ti
v
e
o
p
ti
m
iz
e
r
b
a
se
d
o
n
re
fe
re
n
c
e
p
o
in
t
a
p
p
ro
a
c
h
,
”
In
d
o
n
e
sia
J
o
u
rn
a
l
o
f
El
e
c
trica
l
E
n
g
in
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l
.
1
5
,
n
o
.
3
,
p
p
.
1
6
0
3
-
1
6
1
4
,
20
19
.
[1
9
]
G
e
r
m
á
n
Ca
p
d
e
h
o
u
ra
t
,
A
n
d
ré
s
Co
re
z
,
A
n
a
b
e
ll
a
Ba
z
z
a
n
o
,
Ro
d
rig
o
A
lo
n
so
,
a
n
d
P
a
b
l
o
M
u
sé
,
“
T
o
wa
rd
a
c
o
m
b
in
e
d
to
o
l
to
a
ss
ist
d
e
rm
a
to
lo
g
ists
in
m
e
lan
o
m
a
d
e
tec
ti
o
n
f
ro
m
d
e
rm
o
sc
o
p
ic
im
a
g
e
s
o
f
p
ig
m
e
n
ted
sk
in
les
io
n
s,”
Pa
tt
e
rn
Rec
o
g
n
it
io
n
L
e
tt
e
rs
,
v
o
l
.
3
2
,
n
o
.
1
6
,
p
p
.
2
1
8
7
-
2
1
9
6
,
2
0
1
1
.
[2
0
]
W
S
to
lz,
e
t
a
l,
“
A
b
c
d
ru
le
o
f
d
e
r
m
a
to
sc
o
p
y
-
a
n
e
w
p
ra
c
ti
c
a
l
m
e
th
o
d
f
o
r
e
a
rl
y
re
c
o
g
n
it
io
n
o
f
m
a
li
g
n
a
n
t
-
m
e
lan
o
m
a
,
”
Eu
ro
p
e
a
n
J
o
u
rn
a
l
o
f
De
rm
a
t
o
lo
g
y
,
v
o
l.
4
,
n
o
.
7
,
p
p
.
5
2
1
-
5
2
7
,
1
9
9
4
.
[2
1
]
S
a
li
d
o
J.
A
.
A
.
,
“
Co
n
ra
d
o
Ru
iz
J
r.
Us
in
g
d
e
e
p
lea
rn
in
g
to
d
e
tec
t
m
e
lan
o
m
a
in
d
e
rm
o
sc
o
p
y
i
m
a
g
e
s
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
M
a
c
h
i
n
e
L
e
a
rn
in
g
a
n
d
Co
mp
u
ti
n
g
,
v
o
l.
8
,
n
o
.
1
,
p
p
.
6
1
-
6
8
,
2
0
1
8
.
[2
2
]
P
.
T
sc
h
a
n
d
l
,
C.
R
o
se
n
d
a
h
l,
a
n
d
H.
Kitt
ler,
“
T
h
e
HA
M
1
0
0
0
0
d
a
tas
e
t,
a
larg
e
c
o
ll
e
c
ti
o
n
o
f
m
u
lt
i
-
so
u
rc
e
s
d
e
rm
a
to
sc
o
p
ic i
m
a
g
e
s o
f
c
o
m
m
o
n
p
ig
m
e
n
ted
sk
in
les
io
n
s,”
S
c
i.
D
a
ta
,
v
o
l.
5
,
2
0
1
8
.
[2
3
]
N.
C.
F
.
C
o
d
e
ll
a
,
e
t
a
l.
.
“
S
k
in
les
io
n
a
n
a
ly
sis
to
w
a
rd
m
e
lan
o
m
a
d
e
tec
ti
o
n
:
A
c
h
a
ll
e
n
g
e
a
t
th
e
2
0
1
7
I
n
tern
a
ti
o
n
a
l
s
y
m
p
o
siu
m
o
n
b
io
m
e
d
ica
l
i
m
a
g
in
g
(IS
BI),
h
o
ste
d
b
y
th
e
in
tern
a
ti
o
n
a
l
sk
in
im
a
g
in
g
c
o
ll
a
b
o
ra
ti
o
n
(IS
IC),
”
2
0
1
8
IEE
E
1
5
th
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m o
n
Bi
o
me
d
ica
l
Ima
g
i
n
g
(
IS
BI
2
0
1
8
)
,
W
a
sh
in
g
to
n
,
DC,
p
p
.
1
6
8
-
1
7
2
,
2
0
1
8
.
[2
4
]
M
a
e
n
T
a
k
ru
ri,
M
o
h
a
m
e
d
Kh
a
led
A
b
u
M
a
h
m
o
u
d
,
A
d
e
l
A
l
-
Ju
m
a
il
y
,
“
P
S
O
-
S
VM
h
y
b
rid
sy
ste
m
f
o
r
m
e
lan
o
m
a
d
e
tec
ti
o
n
f
ro
m
h
isto
-
p
a
th
o
lo
g
ica
l
im
a
g
e
s,
”
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
i
n
e
e
rin
g
(
IJ
ECE
),
v
o
l.
9
,
n
o
.
4
,
p
p
.
2
9
4
1
-
2
9
4
9
,
2
0
1
9
.
[2
5
]
N.
C.
F
.
Co
d
e
ll
a
,
e
t
a
l.
,
“
S
k
in
L
e
sio
n
A
n
a
ly
sis
T
o
wa
rd
M
e
lan
o
m
a
De
tec
ti
o
n
:
A
Ch
a
ll
e
n
g
e
a
t
th
e
2
0
1
7
In
ter
n
a
ti
o
n
a
l
S
y
m
p
o
siu
m
o
n
Bio
m
e
d
ica
l
Im
a
g
in
g
(IS
BI),
Ho
ste
d
b
y
th
e
In
te
rn
a
ti
o
n
a
l
S
k
in
Im
a
g
in
g
C
o
ll
a
b
o
ra
ti
o
n
(
IS
IC),
”
IEE
E
1
5
th
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m o
n
Bi
o
me
d
ica
l
Ima
g
i
n
g
(
IS
BI
2
0
1
8
),
p
p
.
1
6
8
-
1
7
2
,
2
0
1
8
.
[2
6
]
N.
Za
m
a
n
i
Taje
d
d
in
a
n
d
B.
M
o
h
a
m
m
a
d
Zad
e
h
A
sl,
“
A
G
e
n
e
ra
l
A
l
g
o
rit
h
m
f
o
r
A
u
to
m
a
ti
c
Les
io
n
S
e
g
m
e
n
t
a
ti
o
n
in
De
r
m
o
sc
o
p
y
I
m
a
g
e
s,”
2
3
rd
Ir
a
n
i
a
n
Co
n
fer
e
n
c
e
o
n
Bi
o
me
d
ic
a
l
E
n
g
i
n
e
e
rin
g
a
n
d
1
st
In
ter
n
a
ti
o
n
a
l
Ira
n
ia
n
Co
n
fer
e
n
c
e
o
n
Bi
o
me
d
ica
l
En
g
in
e
e
rin
g
(
ICBM
E),
p
p
.
1
3
4
-
1
3
9
,
2
0
1
6
.
[2
7
]
Y.
Yu
a
n
,
M
.
C
h
a
o
,
a
n
d
Y.
C.
L
o
,
“
A
u
to
m
a
ti
c
S
k
in
L
e
sio
n
S
e
g
m
e
n
tatio
n
Us
in
g
De
e
p
F
u
ll
y
Co
n
v
o
lu
ti
o
n
a
l
Ne
tw
o
rk
s
w
it
h
Ja
c
c
a
rd
Dista
n
c
e
,
”
IEE
E
T
r
a
n
s.
M
e
d
.
Ima
g
in
g
,
v
o
l
.
3
6
,
n
o
.
9
,
p
p
.
1
8
7
6
-
1
8
8
6
,
2
0
1
7
.
[2
8
]
A
.
P
e
n
n
isi,
e
t
a
l
,
“
S
k
in
les
io
n
im
a
g
e
s
e
g
m
e
n
tatio
n
u
si
n
g
De
lau
n
a
y
T
rian
g
u
latio
n
f
o
r
m
e
lan
o
m
a
d
e
te
c
ti
o
n
,
”
Co
mp
u
t
.
M
e
d
.
Ima
g
i
n
g
Gr
a
p
h
.
,
v
o
l.
5
2
,
p
p
.
8
9
-
1
0
3
,
2
0
1
6
.
[2
9
]
C.
Ba
ra
ta,
e
t
a
l
,
“
T
w
o
S
y
ste
m
s
fo
r
th
e
De
tec
ti
o
n
o
f
M
e
lan
o
m
a
s
i
n
De
rm
o
sc
o
p
y
I
m
a
g
e
s
Us
in
g
Te
x
tu
re
a
n
d
C
o
lo
r
F
e
a
tu
re
s,”
IEE
E
S
y
st.
J
.
,
v
o
l.
8
,
n
o
.
3
,
p
p
.
9
6
5
-
9
7
9
,
2
0
1
3
.
[3
0
]
M
.
Ja
h
a
n
if
a
r,
N.
Za
m
a
n
i
T
a
jed
d
in
,
B.
M
o
h
a
m
m
a
d
z
a
d
e
h
A
sl
a
n
d
A
.
G
o
o
y
a
,
"
S
u
p
e
rv
ise
d
S
a
li
e
n
c
y
M
a
p
Driv
e
n
S
e
g
m
e
n
tatio
n
o
f
L
e
sio
n
s
in
De
rm
o
sc
o
p
ic
Im
a
g
e
s,
"
in
IEE
E
J
o
u
rn
a
l
o
f
Bi
o
me
d
ic
a
l
a
n
d
He
a
lt
h
I
n
fo
r
ma
ti
c
s
,
v
o
l.
2
3
,
n
o
.
2
,
p
p
.
5
0
9
-
5
1
8
,
M
a
r
.
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.