I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Ro
bo
t
ics a
nd
Aut
o
m
a
t
io
n
(
I
J
RA)
Vo
l.
14
,
No
.
3
,
Sep
tem
b
er
20
25
,
p
p
.
4
2
9
~
43
7
I
SS
N:
2722
-
2
5
8
6
,
DOI
:
1
0
.
1
1
5
9
1
/i
jr
a
.
v
14
i
3
.
pp
4
2
9
-
43
7
429
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
r
a
.
ia
esco
r
e.
co
m
H
y
brid d
eep l
ea
rning
and a
ctive c
o
ntour for seg
me
nting ha
zy
ima
g
es
F
irha
n Az
ri
Ahm
a
d K
ha
irul
Anua
r
1
,
J
enev
y
J
o
ne
2
,
Ra
j
a
F
a
rha
t
ul Aiesy
a
Ra
j
a
Azha
r
3
,
Abdu
l
K
a
dir
J
um
a
a
t
1,
4
1
F
a
c
u
l
t
y
o
f
C
o
mp
u
t
e
r
a
n
d
M
a
t
h
e
ma
t
i
c
a
l
S
c
i
e
n
c
e
s
,
U
n
i
v
e
r
si
t
i
Te
k
n
o
l
o
g
i
M
A
R
A
,
S
h
a
h
A
l
a
m,
M
a
l
a
y
s
i
a
2
F
i
n
a
n
c
e
R
i
sk
C
o
st
C
o
n
t
r
o
l
,
S
h
a
n
g
r
i
-
La
R
a
sa
R
i
a
,
P
a
n
t
a
i
D
a
l
i
t
Tu
a
r
a
n
,
K
o
t
a
K
i
n
a
b
a
l
u
,
M
a
l
a
y
si
a
3
F
a
c
u
l
t
y
o
f
C
o
mp
u
t
i
n
g
a
n
d
I
n
f
o
r
mat
i
c
s,
U
n
i
v
e
r
si
t
i
M
a
l
a
y
s
i
a
S
a
b
a
h
,
K
o
t
a
K
i
n
a
b
a
l
u
,
M
a
l
a
y
si
a
4
I
n
st
i
t
u
t
e
f
o
r
B
i
g
D
a
t
a
A
n
a
l
y
t
i
c
s a
n
d
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(
I
B
D
A
A
I
)
,
U
n
i
v
e
r
si
t
i
Te
k
n
o
l
o
g
i
M
A
R
A
,
S
h
a
h
A
l
a
m
,
M
a
l
a
y
s
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
r
2
6
,
2
0
2
5
R
ev
is
ed
J
u
n
3
0
,
2
0
2
5
Acc
ep
ted
J
u
l 1
0
,
2
0
2
5
Im
a
g
e
se
g
m
e
n
tatio
n
se
e
k
s
to
d
isti
n
g
u
ish
th
e
f
o
re
g
r
o
u
n
d
fro
m
th
e
b
a
c
k
g
ro
u
n
d
f
o
r
f
u
rth
e
r
a
n
a
ly
sis.
A
re
c
e
n
t
stu
d
y
p
re
se
n
ted
a
n
e
w
a
c
ti
v
e
c
o
n
to
u
r
m
o
d
e
l
(ACM)
fo
r
ima
g
e
se
g
m
e
n
tatio
n
,
term
e
d
G
a
u
ss
ian
re
g
u
lariz
a
ti
o
n
se
lec
ti
v
e
se
g
m
e
n
t
a
ti
o
n
(G
RS
S
).
T
h
is
i
n
tera
c
ti
v
e
ACM
is
e
ffe
c
ti
v
e
fo
r
se
g
m
e
n
ti
n
g
c
e
rtain
o
b
jec
ts
in
ima
g
e
s.
Ho
we
v
e
r,
a
w
e
a
k
n
e
ss
o
f
th
e
G
RS
S
m
o
d
e
l
b
e
c
o
m
e
s a
p
p
a
re
n
t
wh
e
n
u
ti
li
z
e
d
o
n
h
a
z
y
ima
g
e
s,
a
s it
is n
o
t
in
ten
d
e
d
f
o
r
s
u
c
h
c
o
n
d
it
i
o
n
s
a
n
d
p
ro
d
u
c
e
s
in
a
d
e
q
u
a
te
o
u
tco
m
e
s.
T
h
is
p
a
p
e
r
in
tro
d
u
c
e
s
a
n
e
w
ACM
fo
r
se
g
m
e
n
ti
n
g
h
a
z
y
ima
g
e
s
th
a
t
h
y
b
rid
ize
s
a
p
re
train
e
d
d
e
e
p
lea
rn
i
n
g
m
o
d
e
l,
n
a
m
e
ly
De
h
a
z
e
Ne
t,
with
t
h
e
G
RS
S
m
o
d
e
l
.
S
p
e
c
ifi
c
a
ll
y
,
t
h
e
h
a
z
e
-
fre
e
ima
g
e
s
a
re
e
stim
a
ted
u
sin
g
De
h
a
z
e
Ne
t,
wh
ich
fu
se
s
th
e
in
fo
rm
a
ti
o
n
wit
h
t
h
e
G
RS
S
m
o
d
e
l.
Th
e
n
e
w
fo
rm
u
latio
n
,
d
e
sig
n
a
ted
a
s G
RS
S
with
De
h
a
z
e
Ne
t
(
G
DN
),
is ad
d
re
ss
e
d
v
ia t
h
e
c
a
lcu
lu
s o
f
v
a
riatio
n
s
a
n
d
e
x
e
c
u
te
d
in
M
AT
LAB
so
ftwa
re
.
T
h
e
se
g
m
e
n
tati
o
n
a
c
c
u
ra
c
y
wa
s
e
v
a
lu
a
ted
b
y
c
a
lcu
lati
n
g
Err
o
r,
Ja
c
c
a
rd
,
a
n
d
Dic
e
m
e
tri
c
s,
wh
il
e
e
fficie
n
c
y
wa
s
d
e
term
in
e
d
b
y
m
e
a
su
rin
g
p
r
o
c
e
ss
in
g
ti
m
e
.
De
sp
it
e
th
e
in
c
re
a
se
d
p
ro
c
e
ss
in
g
ti
m
e
,
n
u
m
e
rica
l
e
x
p
e
rime
n
ts
d
e
m
o
n
stra
ted
th
a
t
t
h
e
G
D
N
m
o
d
e
l
a
c
h
iev
e
d
h
ig
h
e
r
a
c
c
u
ra
c
y
,
a
s
in
d
ica
ted
b
y
t
h
e
l
o
we
r
e
rro
r
a
n
d
h
ig
h
e
r
Ja
c
c
a
rd
a
n
d
Dic
e
th
a
n
th
e
G
RS
S
m
o
d
e
l.
Th
e
G
DN
m
o
d
e
l
c
a
n
p
o
ten
ti
a
ll
y
b
e
fo
rm
u
late
d
i
n
t
h
e
v
e
c
to
r
-
v
a
lu
e
d
ima
g
e
d
o
m
a
in
in
th
e
fu
tu
re
.
K
ey
w
o
r
d
s
:
Activ
e
co
n
to
u
r
Dee
p
lear
n
in
g
Haz
y
im
ag
e
I
m
ag
e
s
eg
m
en
tatio
n
Var
iatio
n
al
lev
el
s
et
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
:
Ab
d
u
l K
ad
ir
J
u
m
aa
t
F
ac
u
lty
o
f
C
o
m
p
u
ter
an
d
Ma
th
em
atica
l Scie
n
ce
s
,
I
n
s
titu
te
f
o
r
B
ig
Data
An
aly
tics
an
d
Ar
t
if
icial
I
n
tellig
en
ce
(
I
B
DAAI
)
,
Un
iv
e
r
s
iti T
ek
n
o
lo
g
i M
AR
A
Ko
m
p
lek
s
Al
-
Kh
awa
r
izm
i,
4
0
4
5
0
Sh
ah
Alam
,
Selan
g
o
r
,
Ma
lay
s
ia
E
m
ail: a
b
d
u
lk
ad
i
r
@
tm
s
k
.
u
itm
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
Dig
ital
im
ag
e
p
r
o
ce
s
s
in
g
in
v
o
lv
es
in
ter
p
r
etin
g
an
d
ap
p
ly
i
n
g
d
ig
ital
im
ag
es
to
ex
tr
ac
t
in
f
o
r
m
atio
n
,
en
h
an
cin
g
clar
ity
a
n
d
p
r
ac
tic
ality
f
o
r
o
b
s
er
v
atio
n
.
No
tab
ly
,
n
u
m
e
r
o
u
s
tech
n
iq
u
es
ex
is
t
with
in
th
e
r
ea
lm
o
f
im
ag
e
p
r
o
ce
s
s
in
g
,
in
clu
d
in
g
im
ag
e
d
eh
az
in
g
an
d
im
ag
e
s
eg
m
en
tatio
n
.
I
n
p
ar
ticu
lar
,
i
m
ag
e
s
eg
m
en
tatio
n
in
v
o
lv
es
p
a
r
titi
o
n
in
g
o
r
d
elin
ea
tio
n
o
f
an
im
a
g
e
in
t
o
d
is
tin
ct
o
b
jects
o
r
r
eg
io
n
s
f
o
r
s
u
b
s
eq
u
en
t
p
r
o
ce
s
s
in
g
.
T
h
is
tech
n
iq
u
e
f
in
d
s
ap
p
lic
atio
n
s
in
ar
ea
s
s
u
ch
as
m
ed
ical
im
ag
e
an
aly
s
is
[
1
]
–
[
5
]
,
r
o
b
o
tics
,
p
atter
n
r
ec
o
g
n
itio
n
,
im
ag
e
u
n
d
e
r
s
tan
d
in
g
,
an
d
c
o
m
p
u
ter
v
is
io
n
[
6
]
–
[
8
]
.
An
estab
lis
h
ed
m
eth
o
d
f
o
r
im
a
g
e
s
eg
m
en
tatio
n
is
th
e
ac
tiv
e
c
o
n
to
u
r
m
o
d
el
(
AC
M)
,
wh
ich
r
elies
o
n
m
ath
em
atica
l
m
o
d
elin
g
.
Fu
r
th
er
m
o
r
e,
AC
Ms
d
em
o
n
s
tr
ate
s
ig
n
if
ican
t
ef
f
icac
y
in
p
r
o
d
u
cin
g
h
ig
h
-
q
u
ality
im
a
g
e
s
eg
m
en
tatio
n
an
d
ex
ce
l
i
n
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
3
,
Sep
tem
b
er
20
25
:
4
2
9
-
43
7
430
ex
tr
ac
tio
n
o
f
s
im
ilar
r
eg
io
n
s
[
9
]
.
T
h
e
AC
Ms
p
r
esen
t
in
v
ar
io
u
s
ap
p
licatio
n
s
[
1
0
]
–
[
1
2
]
ca
n
b
e
ca
teg
o
r
ized
in
t
o
g
lo
b
al
an
d
in
ter
ac
tiv
e
s
eg
m
e
n
tatio
n
.
G
l
o
b
a
l
s
e
g
m
e
n
t
a
ti
o
n
a
i
m
s
t
o
s
e
g
m
e
n
t
a
n
i
m
a
g
e
o
b
j
e
ct
o
v
e
r
a
w
h
o
l
e
i
m
a
g
e
.
I
n
[
7
]
,
a
l
o
c
a
l
a
n
d
g
l
o
b
a
l
A
C
M
h
y
b
r
i
d
w
as
f
o
r
m
u
l
a
t
e
d
in
a
f
u
z
z
y
t
h
e
o
r
e
t
ic
a
l
f
r
a
m
ew
o
r
k
t
o
s
e
g
m
e
n
t i
m
a
g
es
wi
t
h
i
n
te
n
s
i
t
y
i
n
h
o
m
o
g
e
n
e
i
t
y
.
M
o
r
e
o
v
e
r
,
t
o
s
e
g
m
e
n
t
m
e
d
i
ca
l
i
m
a
g
e
s
,
[
1
3
]
a
n
d
[
1
4
]
s
u
c
c
ess
f
u
l
l
y
d
e
v
e
l
o
p
e
d
a
h
y
b
r
i
d
A
C
M
a
n
d
a
s
i
m
u
l
t
a
n
e
o
u
s
s
e
g
m
e
n
t
a
ti
o
n
a
n
d
c
o
r
r
e
c
t
i
o
n
m
o
d
e
l
,
r
e
s
p
e
c
t
i
v
e
l
y
.
A
n
o
t
h
er
p
o
p
u
l
a
r
s
t
r
a
t
e
g
y
t
o
a
c
h
i
e
v
e
s
u
c
c
e
s
s
f
u
l
i
m
a
g
e
s
e
g
m
e
n
t
a
ti
o
n
i
s
u
t
i
li
z
i
n
g
t
h
e
s
a
l
i
e
n
c
y
m
a
p
o
f
i
m
a
g
es
as
c
o
n
d
u
c
t
e
d
b
y
[
1
5
]
,
[
1
6
]
.
H
o
w
e
v
e
r
,
a
l
l
t
h
e
m
o
d
e
ls
m
a
y
p
r
o
d
u
c
e
u
n
s
a
t
i
s
f
a
ct
o
r
y
r
e
s
u
l
ts
if
t
h
e
t
a
r
g
e
t
e
d
o
b
j
e
c
t
a
p
p
r
o
x
i
m
at
e
s
t
h
e
n
e
i
g
h
b
o
r
i
n
g
o
b
j
e
c
t
o
r
im
a
g
e
n
o
i
s
e
.
H
e
n
c
e
,
t
o
a
d
d
r
e
s
s
t
h
i
s
i
s
s
u
e
,
a
n
a
lt
e
r
n
at
i
v
e
w
a
y
is
t
o
u
s
e
i
n
t
e
r
ac
t
i
v
e
im
a
g
e
s
e
g
m
e
n
t
at
i
o
n
t
e
c
h
n
i
q
u
es
.
I
n
ter
ac
tiv
e
s
eg
m
en
tatio
n
is
a
tech
n
iq
u
e
f
o
r
s
eg
m
e
n
tin
g
an
i
m
ag
e
th
at
f
o
c
u
s
es
o
n
th
e
s
elec
ted
o
b
ject
o
r
r
e
g
io
n
o
f
it
[
1
7
]
.
I
n
t
h
e
i
m
ag
e
p
r
o
ce
s
s
in
g
c
o
m
m
u
n
ity
,
in
ter
ac
tiv
e
i
m
ag
e
s
eg
m
en
tatio
n
is
also
k
n
o
wn
as
s
elec
tiv
e
im
ag
e
s
eg
m
en
tatio
n
.
I
n
ad
d
itio
n
to
r
o
b
o
tics
,
ap
p
licatio
n
s
s
u
itab
le
f
o
r
th
e
in
co
r
p
o
r
atio
n
o
f
in
ter
ac
tiv
e
s
eg
m
en
tatio
n
tech
n
iq
u
es
in
cl
u
d
e
r
esear
c
h
f
ield
s
s
u
c
h
as
m
ed
ical
im
ag
in
g
[
1
8
]
–
[
2
0
]
,
as
well
as
p
atter
n
r
ec
o
g
n
itio
n
[
2
1
]
.
T
h
e
m
o
d
els
r
eq
u
ir
e
th
e
en
d
u
s
er
to
b
e
in
ter
ac
tiv
ely
in
v
o
lv
e
d
in
d
eter
m
in
in
g
th
e
tar
g
ete
d
o
b
ject
b
y
d
ef
in
in
g
a
s
et
o
f
m
a
r
k
er
s
ar
o
u
n
d
it.
Acc
o
r
d
in
g
ly
,
th
e
m
o
d
els
will
u
tili
ze
th
e
m
a
r
k
er
s
et
to
ac
h
ie
v
e
an
ac
cu
r
ate
r
esu
lt
.
On
e
o
f
th
e
ea
r
lies
t
in
ter
ac
tiv
e
s
eg
m
en
tati
o
n
m
o
d
els
was
d
ev
elo
p
ed
in
[
2
2
]
,
wh
ich
u
tili
ze
d
a
d
is
tan
ce
f
u
n
ctio
n
co
u
p
led
w
ith
th
e
r
eg
u
la
r
izatio
n
(
to
tal
v
a
r
iatio
n
)
ter
m
.
H
o
wev
er
,
th
e
m
o
d
el
m
ay
p
r
o
d
u
ce
p
o
o
r
r
esu
lts
wh
en
th
e
o
b
ject
b
o
u
n
d
a
r
y
is
wea
k
.
T
h
u
s
,
a
o
n
e
-
lev
el
-
s
et
id
ea
with
ar
ea
c
o
n
s
tr
ain
t
was
p
r
o
p
o
s
ed
by
[
2
3
]
t
o
o
v
er
co
m
e
th
e
lim
it
atio
n
.
Alth
o
u
g
h
th
e
m
o
d
el
is
s
u
cc
ess
f
u
l,
it
r
eq
u
ir
es
s
u
b
s
tan
ti
al
co
m
p
u
tin
g
co
s
ts
.
T
h
er
ef
o
r
e,
to
en
h
an
ce
ef
f
icie
n
cy
,
t
h
e
m
o
d
el
i
n
[
2
4
]
was
p
r
o
p
o
s
ed
.
T
h
e
m
o
d
el
is
ef
f
ec
tiv
e
f
o
r
th
e
s
m
all
an
d
m
o
d
er
ate
s
ize
o
f
an
im
ag
e.
T
o
s
eg
m
en
t
lar
g
e
s
ize
o
f
im
ag
e
s
,
J
u
m
aa
t
an
d
C
h
en
[
2
5
]
s
u
cc
ess
f
u
lly
p
r
o
p
o
s
ed
a
n
ew
in
ter
ac
tiv
e
AC
M.
R
ec
en
tly
,
Saib
in
a
n
d
J
u
m
aa
t
[
1
2
]
h
av
e
s
u
cc
ess
f
u
lly
d
e
v
elo
p
ed
an
in
ter
ac
tiv
e
AC
M,
n
am
ely
Gau
s
s
ian
r
eg
u
lar
izatio
n
s
elec
tiv
e
s
eg
m
en
tatio
n
(
GR
SS
)
,
s
in
ce
th
e
p
r
ev
io
u
s
m
o
d
els
m
e
n
tio
n
ed
a
b
o
v
e
ar
e
less
ef
f
icien
t
with
r
eg
ar
d
to
s
eg
m
en
tin
g
im
ag
es
h
av
in
g
in
te
n
s
ity
in
h
o
m
o
g
e
n
eities.
Ho
wev
er
,
GR
SS
is
n
o
t
eq
u
ip
p
e
d
to
s
eg
m
en
t
im
a
g
es
af
f
ec
ted
b
y
h
az
e,
lea
d
in
g
to
s
u
b
o
p
tim
al
s
eg
m
en
tatio
n
o
u
tco
m
es.
No
te
th
at
h
az
e
in
im
ag
es
is
u
n
av
o
id
ab
le,
p
ar
ticu
lar
ly
d
u
r
in
g
th
e
ac
q
u
is
itio
n
p
h
ase
f
o
r
r
ea
l
im
ag
es.
Acc
o
r
d
in
g
to
Ali
et
a
l
.
[
2
6
]
,
it
ca
n
b
e
ch
allen
g
in
g
t
o
s
eg
m
en
t
a
r
ea
l
im
ag
e
with
th
e
p
r
esen
ce
o
f
h
az
e.
Haz
e
is
in
ter
p
r
eted
as
an
atm
o
s
p
h
er
ic
p
h
en
o
m
en
o
n
ca
u
s
ed
b
y
p
ar
ticles
s
u
ch
as
d
u
s
t,
s
m
o
k
e,
an
d
o
th
er
d
r
y
p
ar
ticl
es
s
u
s
p
en
d
ed
i
n
th
e
air
,
af
f
ec
tin
g
o
b
s
cu
r
e
v
is
ib
ilit
y
an
d
s
k
y
clar
ity
.
T
h
is
d
em
o
n
s
tr
ates
th
e
s
ig
n
if
ican
ce
o
f
th
e
im
ag
e
d
eh
az
in
g
p
r
o
ce
s
s
f
o
r
r
ea
l i
m
ag
es.
T
h
e
co
m
m
o
n
ly
u
s
ed
h
az
e
r
e
d
u
ctio
n
tech
n
iq
u
es,
s
u
ch
as
Deh
az
eNe
t
[
2
7
]
,
ca
n
en
h
a
n
ce
th
e
im
ag
e
q
u
ality
.
As
a
d
ee
p
lear
n
i
n
g
-
b
a
s
ed
m
eth
o
d
,
Deh
az
eNe
t
is
p
o
wer
f
u
l
in
r
ed
u
cin
g
im
ag
e
h
az
e.
Hen
ce
,
th
is
s
tu
d
y
p
r
esen
ts
a
n
ew
AC
M
f
o
r
s
eg
m
en
tin
g
h
az
y
im
a
g
es
th
at
h
y
b
r
id
izes
th
e
p
r
etr
ain
ed
Deh
az
eNe
t
with
th
e
GR
SS
m
o
d
el.
Sp
ec
if
ically
,
we
esti
m
ate
th
e
h
az
e
-
f
r
ee
im
ag
es
u
s
in
g
Deh
az
eNe
t
an
d
f
u
s
e
th
e
in
f
o
r
m
atio
n
with
th
e
GR
SS
m
o
d
el.
T
h
e
n
ewly
d
ev
elo
p
ed
f
o
r
m
u
latio
n
is
d
esig
n
a
ted
as
th
e
GR
SS
wi
th
Deh
az
eNe
t
(
GDN)
m
o
d
el.
T
h
e
s
u
b
s
eq
u
en
t
p
ar
ts
o
f
th
is
d
o
cu
m
e
n
t
ar
e
s
y
s
tem
atica
lly
s
tr
u
ctu
r
ed
in
to
th
r
ee
d
is
tin
ct
s
ec
tio
n
s
.
Sectio
n
2
d
escr
ib
es
th
e
m
eth
o
d
o
lo
g
y
em
p
lo
y
ed
in
th
e
s
tu
d
y
.
Me
an
wh
il
e,
Sectio
n
3
a
d
d
r
ess
es
th
e
r
esu
lts
an
d
d
is
cu
s
s
io
n
,
wh
er
ea
s
Sectio
n
4
will c
o
v
er
t
h
e
co
n
clu
s
io
n
an
d
r
ec
o
m
m
e
n
d
atio
n
s
.
2.
M
E
T
H
O
D
Sectio
n
2
d
is
cu
s
s
es
th
e
r
e
s
ea
r
ch
m
eth
o
d
o
lo
g
y
f
o
r
th
e
p
r
o
p
o
s
ed
m
o
d
el.
Ad
d
itio
n
ally
,
th
e
im
p
lem
en
tatio
n
p
r
o
ce
s
s
was
p
r
esen
ted
to
allo
w
a
p
r
ec
is
e
co
m
p
r
eh
en
s
io
n
with
r
eg
ar
d
to
h
o
w
im
ag
e
d
eh
az
i
n
g
an
d
s
eg
m
en
tatio
n
wo
r
k
.
Fig
u
r
e
1
illu
s
tr
ates th
e
m
eth
o
d
o
l
o
g
y
p
h
ase
f
lo
w
in
v
o
lv
e
d
in
th
is
r
esear
ch
.
Fig
u
r
e
1
.
Flo
w
o
f
th
e
r
esear
c
h
m
eth
o
d
o
lo
g
y
T
h
er
e
ar
e
f
o
u
r
p
h
ases
in
clu
d
ed
in
th
is
r
esear
ch
,
as
illu
s
tr
ated
in
Fig
u
r
e
1
.
T
h
e
f
ir
s
t
p
h
a
s
e
is
d
ata
ac
q
u
is
itio
n
f
r
o
m
av
ailab
le
d
at
ab
ases
.
Nex
t,
th
e
f
o
r
m
u
latio
n
o
f
th
e
G
DN
m
o
d
el
will
b
e
d
is
cu
s
s
ed
.
All
th
e
s
tep
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
I
SS
N:
2722
-
2
5
8
6
Hyb
r
id
d
ee
p
lea
r
n
in
g
a
n
d
a
ct
ive
co
n
to
u
r
fo
r
s
eg
men
tin
g
h
a
z
y
ima
g
es
(
F
ir
h
a
n
A
z
r
i A
h
ma
d
K
h
a
ir
u
l A
n
u
a
r
)
431
to
s
o
lv
e
t
h
e
p
r
o
p
o
s
ed
m
o
d
el
ar
e
e
x
p
lain
e
d
in
d
etail
i
n
t
h
is
p
h
ase
as
well.
Su
b
s
eq
u
e
n
tly
,
a
MA
T
L
AB
alg
o
r
ith
m
will
b
e
d
e
v
elo
p
ed
to
im
p
lem
en
t
th
e
m
o
d
el.
Fin
ally
,
th
e
p
r
o
p
o
s
ed
m
o
d
el’
s
im
ag
e
s
eg
m
en
tatio
n
r
esu
lts
will b
e
co
m
p
ar
ed
to
t
h
e
ex
is
tin
g
m
o
d
el.
T
h
e
s
u
b
s
eq
u
en
t su
b
s
ec
tio
n
p
r
o
v
id
es a
th
o
r
o
u
g
h
e
x
p
lan
atio
n
o
f
ea
ch
p
h
ase.
2
.
1
.
Da
t
a
a
cquis
it
io
n
T
h
e
f
ir
s
t
p
h
ase
o
f
th
is
s
tu
d
y
i
s
to
ac
q
u
ir
e
r
ea
l
test
im
ag
es
f
r
o
m
p
u
b
licly
av
ailab
le
d
atab
a
s
es
[
2
8
]
–
[
3
2
]
.
I
n
a
d
d
itio
n
,
th
e
b
en
c
h
m
ar
k
im
a
g
es,
wh
ich
ar
e
ess
en
tial
f
o
r
co
m
p
ar
in
g
a
n
d
ev
alu
ati
n
g
d
if
f
er
e
n
t
m
o
d
els
in
th
is
s
tu
d
y
,
wer
e
also
r
etr
iev
ed
f
r
o
m
th
e
s
am
e
s
o
u
r
ce
s
.
T
h
e
im
ag
es
ar
e
r
esized
to
a
s
iz
e
o
f
256
2
5
6
p
ix
els
u
s
in
g
MA
T
L
AB
R
2
0
2
3
a
s
o
f
t
war
e.
2
.
2
.
M
o
del
f
o
r
m
ula
t
io
n
I
n
th
is
s
tu
d
y
,
th
e
GR
SS
m
o
d
e
l,
wh
ich
was
r
ec
en
tly
p
r
o
p
o
s
e
d
b
y
[
1
2
]
,
is
co
n
s
id
er
ed
.
I
n
th
e
m
o
d
el,
a
s
et
o
f
m
ar
k
er
s
,
also
k
n
o
wn
a
s
g
eo
m
etr
ical
c
o
n
s
tr
ain
t
A,
is
u
tili
ze
d
,
lo
ca
te
d
n
ea
r
th
e
tar
g
eted
o
b
ject.
T
h
e
d
is
tan
ce
en
er
g
y
ter
m
=
∫
(
)
,
wh
ich
is
f
o
r
m
u
lated
in
lev
el
s
et
f
u
n
ctio
n
(
,
)
,
in
im
ag
e
d
o
m
ain
D
is
u
s
ed
wh
er
e
th
e
E
u
clid
ea
n
d
is
tan
ce
,
o
f
ea
ch
p
ix
els
in
D
f
r
o
m
A
is
ap
p
lied
a
s
d
ef
in
ed
in
[
1
2
]
.
T
h
en
,
th
e
GR
SS
m
o
d
el
is
m
at
h
em
atica
lly
d
ef
in
e
d
as in
(
1
)
.
{
=
1
2
∫
(
0
−
(
1
(
)
+
2
(
1
−
(
)
)
)
)
2
+
∫
(
)
}
(
1
)
T
h
e
Hea
v
is
id
e
f
u
n
ctio
n
is
d
en
o
ted
b
y
(
)
,
th
e
co
n
s
tan
t
is
th
e
co
ef
f
icien
t
f
o
r
d
is
tan
ce
ter
m
,
th
e
in
ten
s
ity
av
er
ag
e
in
s
id
e
s
eg
m
e
n
tatio
n
co
n
to
u
r
is
d
en
o
te
d
b
y
1
(
,
)
=
∗
[
(
)
0
]
/
∗
(
)
,
an
d
2
(
,
)
=
∗
[
1
−
(
)
0
]
/
∗
[
1
−
(
)
]
in
d
icate
s
th
e
in
ten
s
ity
av
er
ag
e
o
u
ts
id
e
th
e
co
n
to
u
r
s
u
ch
th
at
=
−
(
2
+
2
)
/
2
2
.
T
h
e
GR
SS
m
o
d
el
is
ca
p
ab
le
o
f
ef
f
ec
tiv
el
y
s
eg
m
en
tin
g
an
o
b
ject
in
an
i
n
ten
s
ity
in
h
o
m
o
g
e
n
eity
im
ag
e.
T
h
e
s
eg
m
en
ted
co
n
to
u
r
r
esu
ltin
g
f
r
o
m
th
e
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
u
s
in
g
t
h
e
G
R
SS
m
o
d
el
ca
n
b
e
ad
ju
s
ted
in
ter
ac
tiv
ely
b
y
d
ef
i
n
in
g
a
s
u
itab
le
lo
ca
tio
n
o
f
th
e
m
ar
k
er
s
et
A
.
Ho
wev
er
,
th
e
GR
SS
m
o
d
el
is
les
s
ef
f
ec
tiv
e
in
s
eg
m
en
tin
g
im
a
g
es
with
h
az
e.
Du
s
t,
s
m
o
k
e,
a
n
d
o
th
er
d
r
y
air
b
o
r
n
e
p
ar
ticle
s
will
ca
u
s
e
d
ig
ital
im
ag
es
to
b
e
co
r
r
u
p
ted
d
u
r
in
g
th
e
im
ag
e
ca
p
tu
r
e
p
r
o
ce
s
s
,
m
ak
in
g
it
ch
allen
g
in
g
to
s
eg
m
en
t
u
s
in
g
an
im
ag
e
se
g
m
en
tatio
n
m
o
d
el.
T
o
r
ed
u
ce
im
ag
e
h
az
e,
im
ag
e
d
e
h
az
in
g
tech
n
iq
u
es
s
u
c
h
as
D
eh
az
eNe
t
[
2
7
]
a
r
e
f
r
eq
u
e
n
tly
u
s
ed
.
I
n
Deh
az
eN
et,
th
e
in
p
u
t
im
a
g
e
with
h
az
e,
0
,
ca
n
b
e
esti
m
ated
u
s
in
g
th
e
atm
o
s
p
h
er
ic
s
ca
tter
in
g
f
u
n
ctio
n
(
2
)
.
0
(
,
)
=
(
,
)
(
,
)
+
(
1
−
(
,
)
)
,
(
2
)
Her
e,
0
(
,y
)
r
ef
er
s
to
a
n
o
b
s
e
r
v
ed
in
ten
s
ity
(
i
n
p
u
t
im
ag
e
with
h
az
e)
,
(
,y
)
r
ep
r
esen
ts
a
s
ce
n
e
r
ad
ian
ce
(
d
eh
az
ed
im
a
g
e)
,
(
,
)
is
th
e
atm
o
s
p
h
er
ic
lig
h
t,
a
n
d
(
,
)
r
esem
b
les
th
e
tr
an
s
m
is
s
io
n
m
ap
d
escr
ib
in
g
th
e
lig
h
t
th
at
r
ea
ch
es
th
e
ca
m
er
a.
No
te
th
at
th
e
Deh
az
eNe
t
was
d
ev
elo
p
ed
u
tili
zin
g
a
c
o
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
etwo
r
k
i
d
ea
wh
e
r
e
th
e
in
p
u
t
is
0
wh
ile
th
e
o
u
tp
u
t
is
(
,
)
.
T
h
e
tr
ain
in
g
d
ataset
c
o
n
s
is
ts
o
f
1
0
0
,
0
0
0
s
y
n
th
etic
p
atch
es
wh
er
e
th
e
ac
tiv
atio
n
f
u
n
ctio
n
ap
p
lied
is
th
e
b
ilater
al
r
ec
tifie
d
lin
ea
r
u
n
it
(
B
R
eL
U)
.
T
h
e
lo
s
s
f
u
n
ctio
n
a
p
p
lied
is
m
ea
n
s
q
u
ar
ed
er
r
o
r
with
Sto
ch
asti
c
g
r
ad
ien
t
d
escen
t
as
th
e
o
p
tim
iz
er
.
C
o
n
s
id
er
in
g
th
e
atm
o
s
p
h
er
ic
lig
h
t,
th
e
d
eh
az
i
n
g
alg
o
r
ith
m
esti
m
ates
th
e
tr
an
s
m
is
s
io
n
m
ap
,
an
d
to
o
b
tain
th
e
s
ce
n
e
r
ad
ian
ce
(
d
eh
az
ed
im
a
g
e
)
J
,
(
2
)
is
r
ewr
itten
as (
3
)
.
(
,
)
=
(
(
,
)
−
(
1
−
(
,
)
)
)
/
(
,
)
.
(
3
)
W
ith
all
th
e
in
g
r
ed
ien
ts
,
a
n
e
w
in
ter
ac
tiv
e
AC
M
to
ef
f
ec
tiv
ely
s
eg
m
en
t
im
ag
es
with
h
az
e
ca
n
b
e
f
o
r
m
u
lated
b
y
r
ef
o
r
m
u
latin
g
th
e
r
ec
en
t
GR
SS
m
o
d
el
in
(
1
)
b
y
i
n
teg
r
atin
g
th
e
o
u
tp
u
t
f
r
o
m
th
e
p
r
etr
ain
ed
Deh
az
eNe
t
m
eth
o
d
p
r
o
p
o
s
ed
b
y
[
2
7
]
.
T
h
e
n
ew
f
o
r
m
u
latio
n
is
n
am
e
d
th
e
GDN
m
o
d
el
in
(
4
)
.
{
=
1
2
∫
(
−
(
1
+
2
(
1
−
)
)
2
+
∫
+
α
2
2
∫
0
−
(
ℎ
1
+
ℎ
2
(
1
−
)
)
2
}
,
(
4
)
Her
e,
0
d
en
o
tes
an
in
p
u
t
h
az
y
im
ag
e,
J
in
d
icate
s
th
e
o
u
t
p
u
t
o
f
Deh
az
eNe
t,
an
d
1
d
em
o
n
s
tr
ates
th
e
co
ef
f
icien
t
f
o
r
th
e
f
itti
n
g
ter
m
b
ased
o
n
th
e
o
u
tp
u
t
im
ag
e
f
r
o
m
Deh
az
eNe
t,
2
is
th
e
co
e
f
f
icie
n
t
f
o
r
th
e
f
itti
n
g
ter
m
b
ased
o
n
th
e
o
r
ig
in
al
h
az
y
in
p
u
t
im
ag
e.
T
h
e
te
r
m
s
ℎ
1
(
,
)
=
∗
[
(
)
0
]
/
∗
(
)
an
d
1
(
,
)
=
∗
[
(
)
]
/
∗
(
)
ar
e
th
e
av
e
r
ag
e
im
ag
e
in
ten
s
ities
in
s
id
e
th
e
s
eg
m
en
tatio
n
c
o
n
to
u
r
o
f
0
an
d
J,
r
esp
ec
tiv
ely
.
T
h
e
t
er
m
s
ℎ
2
(
,
)
=
∗
[
1
−
(
)
0
]
/
∗
[
1
−
(
)
]
an
d
2
(
,
)
=
∗
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
3
,
Sep
tem
b
er
20
25
:
4
2
9
-
43
7
432
[
1
−
(
)
]
/
∗
[
1
−
(
)
]
in
d
icate
th
e
av
er
ag
e
o
f
im
ag
e
in
ten
s
ities
o
u
ts
id
e
th
e
s
eg
m
en
tatio
n
co
n
to
u
r
o
f
0
an
d
J,
r
esp
ec
tiv
ely
.
B
y
ca
lcu
lu
s
o
f
v
ar
iatio
n
s
,
(
5
)
i
s
d
er
iv
ed
in
o
r
d
e
r
to
s
o
lv
e
th
e
GDN
.
(
)
{
1
[
−
1
(
)
−
2
(
1
−
(
)
)
]
(
1
−
2
)
+
2
[
0
−
ℎ
1
(
)
−
ℎ
2
(
1
−
(
)
)
]
(
ℎ
1
−
ℎ
2
)
−
}
.
(
5
)
Her
e,
th
e
g
r
a
d
ien
t d
escen
t m
et
h
o
d
ca
n
b
e
u
s
ed
to
s
o
lv
e
(
5
)
to
o
b
tain
(
6
)
.
∂
ϕ
∂
t
=
(
)
{
1
[
−
1
(
)
−
2
(
1
−
(
)
)
]
(
1
−
2
)
+
[
0
−
ℎ
1
(
)
−
ℎ
2
(
1
−
(
)
)
]
(
ℎ
1
−
ℎ
2
)
−
}
.
(
6
)
I
n
o
th
e
r
wo
r
d
s
,
th
e
m
o
d
el
GD
N
in
(
4
)
is
m
in
im
ized
b
y
s
o
l
v
in
g
(
6
)
.
2
.
3
.
Alg
o
rit
hm
dev
elo
p
m
ent
T
h
is
p
h
ase
d
is
cu
s
s
es
th
e
im
p
lem
en
tatio
n
r
eg
ar
d
in
g
th
e
s
u
g
g
ested
m
o
d
el
in
th
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
.
T
h
er
e
ar
e
two
s
to
p
p
i
n
g
c
r
iter
ia
u
tili
ze
d
to
s
to
p
th
e
p
r
o
ce
s
s
au
to
m
atica
lly
.
Firstl
y
,
th
e
t
o
ler
an
ce
v
alu
e,
=
1
×
1
0
−
6
,
an
d
s
ec
o
n
d
l
y
,
th
e
m
ax
im
u
m
n
u
m
b
e
r
o
f
iter
atio
n
s
,
ma
x
it
=
100
.
E
q
u
atio
n
(
6
)
was
im
p
lem
en
ted
u
s
in
g
MA
T
L
AB
R
2
0
2
3
a
s
o
f
t
war
e.
T
h
e
MA
T
L
AB
Alg
o
r
it
h
m
1
p
r
o
v
id
ed
s
u
m
m
ar
izes
th
e
s
tep
s
in
v
o
lv
ed
in
th
e
GDN
m
o
d
el
im
p
lem
e
n
tatio
n
p
r
o
ce
s
s
.
Alg
o
r
ith
m
1
.
Alg
o
r
ith
m
to
im
p
lem
en
t th
e
GDN
m
o
d
el
1.
E
v
alu
ate
J
u
s
in
g
De
h
az
eNe
t o
f
E
q
u
atio
n
(
3
)
.
>> H
a
z
eI
ma
g
e=in
p
u
t ima
g
e;
>> J
=Deh
a
z
eNet(H
a
z
eI
ma
g
e
)
;
2.
Set th
e
v
alu
e
to
l,
ma
xit
,
,
,
1
,
an
d
2
.
>>
to
l=1
e
-
6
;
ma
xiter=1
0
0
;
th
eta
=8
0
;
s
ig
ma
=1
2
;
a
lp
h
a
1
=
5
,
a
lp
h
a
2
=1
;
3.
Def
in
e
th
e
m
ar
k
er
s
et
A
.
>>mx=[5
9
;
1
1
5
;
6
6
;
7
]
;
>>my=[1
1
;
9
9
;
1
9
2
;
9
9
]
;
4.
I
n
itialize
th
e
lev
el
s
et
f
u
n
ctio
n
.
>>p
h
i=d
o
u
b
le(
b
w
d
is
t(
p
o
ly2
ma
s
k(
mx,
my)
)
)
;
5.
F
o
r
iter
a
tio
n
=1
to
ma
xit
o
r
‖
+
1
−
‖
‖
‖
≤
E
v
o
lv
e
th
e
le
v
el
s
et
f
u
n
ctio
n
b
ased
o
n
(
6
)
.
R
eg
u
lar
ize
b
y
c
o
n
v
o
lv
in
g
wit
h
.
end f
o
r
>>
fo
r
iter
a
tio
n
=1
:
ma
xit
>>[
p
h
i]
=GD
N
(
I
mg
,
J,
mx,
my,
th
eta
,
s
ig
ma
,
ma
xit,
to
l)
;
>> p
h
i= c
o
n
v2
(
p
h
i,K_
s
ig
ma
,
'
s
a
me'
)
;
R
=R
esid
u
a
l(
p
h
i,o
ld
p
h
i
)
/n
o
r
m(
o
ld
p
h
i)
;
>> if
R
<to
l,
b
r
ea
k,
en
d
>> e
n
d
;
>>
fig
u
r
e,
ima
g
esc(
p
h
i)
;
c
o
l
o
r
ma
p
g
r
a
y;
2
.
4
.
P
er
f
o
r
m
a
nce
ev
a
lua
t
io
n
I
n
th
e
f
in
al
p
h
ase,
p
er
f
o
r
m
an
ce
ac
cu
r
ac
y
with
r
eg
ar
d
to
th
e
m
o
d
el
will
b
e
as
s
ess
ed
u
ti
li
zin
g
E
r
r
o
r
,
J
ac
ca
r
d
an
d
Dice
m
etr
ics
to
co
m
p
ar
e
th
e
r
esu
lts
.
T
h
e
=
1
−
(
+
)
/
(
+
+
+
)
.
Her
e,
tr
u
e
p
o
s
itiv
e
(
T
P)
r
ef
er
s
to
a
p
ix
el
(
o
r
r
eg
i
o
n
)
th
at
is
c
o
r
r
ec
tly
ex
am
i
n
ed
as
p
ar
t
o
f
t
h
e
tar
g
eted
o
b
ject.
At
th
e
s
am
e
tim
e,
th
e
tr
u
e
n
eg
ativ
e
(
T
N)
r
ep
r
esen
ts
a
p
ix
el
(
o
r
r
eg
io
n
)
th
at
is
co
r
r
ec
tly
an
aly
ze
d
as
n
o
t
b
ein
g
p
ar
t
o
f
th
e
tar
g
eted
o
b
ject.
O
n
th
e
o
th
er
h
an
d
,
t
h
e
f
alse
p
o
s
itiv
e
(
FP
)
r
ep
r
esen
ts
a
p
ix
el
(
o
r
r
eg
io
n
)
th
at
is
in
co
r
r
ec
tly
lab
eled
as
p
ar
t
o
f
th
e
tar
g
eted
o
b
ject
wh
en
it
is
a
ctu
ally
n
o
t,
wh
ile
th
e
f
alse
n
e
g
ativ
e
(
FN)
d
en
o
tes
a
p
ix
el
(
o
r
r
e
g
io
n
)
th
at
is
in
c
o
r
r
ec
tly
la
b
eled
as
n
o
t
b
ein
g
p
ar
t
o
f
th
e
tar
g
ete
d
o
b
ject
wh
en
it
ac
tu
ally
is
.
I
n
ad
d
itio
n
,
a
lo
w
v
alu
e
ap
p
r
o
a
ch
in
g
0
in
d
icate
s
th
at
th
e
m
o
d
el
ac
cu
r
ately
s
eg
m
en
ts
th
e
in
p
u
t
im
ag
es.
T
h
e
f
o
r
m
u
la
f
o
r
J
ac
ca
r
d
=
|
∩
∗
|
/
|
∪
∗
|
an
d
Dice
=
|
∩
∗
|
/
|
|
+
|
∗
|
wh
er
e
is
th
e
s
et
o
f
s
eg
m
en
ted
d
o
m
ain
s
o
f
the
ta
r
ge
te
d
ob
je
c
t
an
d
∗
is
th
e
tr
u
e
s
et
o
f
the
ta
r
ge
te
d
ob
je
c
t.
T
h
e
r
etu
r
n
v
alu
e
o
f
th
e
s
im
ilar
ity
f
u
n
ctio
n
is
b
etw
ee
n
th
e
r
an
g
e
o
f
0
an
d
1
.
No
tab
ly
,
th
e
clo
s
er
th
e
r
esu
lt v
alu
e
to
1
,
th
e
h
ig
h
er
th
e
lev
el
o
f
p
er
f
ec
tio
n
o
f
th
e
s
eg
m
en
tatio
n
.
T
h
e
e
f
f
icien
cy
co
n
ce
r
n
in
g
th
e
s
u
g
g
ested
GDN
m
o
d
el
was
an
aly
ze
d
b
y
co
m
p
u
tin
g
th
e
p
r
o
ce
s
s
in
g
t
im
e.
T
h
e
ex
p
er
im
en
ts
wer
e
co
n
d
u
cted
o
n
an
AM
D
R
y
ze
n
7
5
7
0
0
X
with
Nv
id
ia
GeFo
r
ce
1
0
7
0
,
o
p
e
r
atin
g
at
3
.
8
0
GHz
an
d
e
q
u
ip
p
ed
with
3
2
GB
o
f
R
AM
.
T
h
e
p
r
o
ce
s
s
in
g
t
im
e
was
ac
cu
r
ately
m
ea
s
u
r
ed
u
tili
zin
g
th
e
‘
tic’
an
d
‘
to
c’
f
u
n
ctio
n
s
in
MA
T
L
AB
R
2
0
2
3
a
s
o
f
twar
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
I
SS
N:
2722
-
2
5
8
6
Hyb
r
id
d
ee
p
lea
r
n
in
g
a
n
d
a
ct
ive
co
n
to
u
r
fo
r
s
eg
men
tin
g
h
a
z
y
ima
g
es
(
F
ir
h
a
n
A
z
r
i A
h
ma
d
K
h
a
ir
u
l A
n
u
a
r
)
433
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
Sectio
n
3
d
ef
in
es
th
e
n
u
m
er
ic
al
ex
p
er
im
en
ts
co
n
d
u
cte
d
an
d
th
e
f
in
d
in
g
s
o
b
tain
ed
.
I
n
th
is
s
tu
d
y
,
all
r
ea
l
test
im
ag
es
wer
e
s
eg
m
en
ted
em
p
lo
y
in
g
th
e
ex
is
tin
g
G
R
SS
m
o
d
el
[
1
2
]
as
well
as
th
e
p
r
o
p
o
s
ed
GDN
m
o
d
el.
T
h
e
p
ar
a
m
eter
s
to
l
,
ma
xit,
an
d
ar
e
s
et
to
10
−
6
,
1
0
0
,
an
d
8
0
,
r
esp
ec
tiv
ely
,
f
o
r
b
o
th
m
o
d
els.
Me
an
wh
ile,
th
e
p
ar
am
ete
r
s
1
an
d
2
ar
e
s
et
in
th
e
r
an
g
e
o
f
1
,
2
=
[
0
.
5
,
10
]
,
an
d
n
o
r
m
ally
1
>
2
f
o
r
im
ag
es
with
h
az
e
wh
ile
th
e
p
ar
am
eter
is
s
et
to
=
[
10
,
200
]
.
T
h
e
r
esu
lt
s
o
f
th
ese
s
ettin
g
s
ar
e
tab
u
lated
in
T
ab
le
1
.
T
ab
le
1
.
T
h
e
s
eg
m
en
tatio
n
r
esu
lts
Te
st
I
mag
e
s
G
R
S
S
GDN
Te
st
I
mag
e
s
G
R
S
S
GDN
1a
1b
1c
7a
7b
7c
2a
2b
2c
8a
8b
8c
3a
3b
3c
9a
9b
9c
4a
4b
4c
1
0
a
1
0
b
1
0
c
5a
5b
5c
1
1
a
1
1
b
1
1
c
6a
6b
6c
1
2
a
1
2
b
1
2
c
T
h
e
f
ir
s
t
an
d
f
o
u
r
th
co
lu
m
n
s
o
f
T
ab
le
1
d
is
p
lay
all
th
e
h
az
y
test
im
a
g
es
th
at
co
n
tain
o
b
jects
r
eq
u
ir
in
g
s
eg
m
en
tatio
n
.
T
h
e
o
b
ject
d
esig
n
ated
f
o
r
s
eg
m
en
ta
tio
n
is
in
d
icate
d
b
y
g
r
ee
n
m
a
r
k
er
s
an
d
a
y
ello
w
in
itial
co
n
to
u
r
.
T
h
e
s
ec
o
n
d
a
n
d
f
if
t
h
co
l
u
m
n
s
o
f
T
ab
le
1
d
em
o
n
s
tr
ate
th
e
r
esu
lts
u
s
in
g
th
e
GR
SS
m
o
d
el,
wh
ile
th
e
th
ir
d
an
d
last
co
lu
m
n
s
o
f
T
a
b
le
1
ar
e
th
e
r
esu
lts
f
r
o
m
th
e
GDN
m
o
d
el.
Vis
u
al
o
b
s
er
v
atio
n
r
ev
ea
ls
th
at
th
e
GR
SS
m
o
d
el’
s
s
eg
m
en
tatio
n
r
esu
lts
clea
r
ly
d
e
m
o
n
s
tr
ate
o
v
e
r
-
s
eg
m
en
tatio
n
co
m
p
ar
e
d
to
t
h
e
p
r
o
p
o
s
ed
GDN
m
o
d
el.
T
h
is
is
d
u
e
to
th
e
f
ac
t
th
at
th
e
GR
SS
m
o
d
el
ca
n
n
o
t
clea
r
ly
d
ef
i
n
e
th
e
h
az
e
in
th
e
im
ag
es
in
T
ab
le
1
(
5
b
,
6
b
,
7
b
,
8
b
,
1
1
b
)
.
Acc
o
r
d
in
g
to
Ali
et
a
l
.
[
2
6
]
,
it
ca
n
b
e
c
h
allen
g
in
g
to
s
eg
m
en
t
a
r
ea
l
im
ag
e
with
t
h
e
p
r
esen
ce
o
f
h
az
e.
T
h
e
s
eg
m
en
tatio
n
o
u
tco
m
es
f
r
o
m
GDN
in
d
icate
th
at
th
e
in
p
u
t
im
ag
es
ar
e
m
o
r
e
ef
f
ec
tiv
ely
s
eg
m
e
n
ted
.
T
h
e
m
ain
r
ea
s
o
n
f
o
r
th
is
is
th
at
th
e
d
eh
az
in
g
p
r
o
ce
s
s
,
th
at
is
t
h
e
Deh
az
eNe
t
[
2
7
]
,
was
ad
d
ed
t
o
th
e
GDN
m
o
d
el
.
I
n
p
ar
ticu
la
r
,
th
e
Deh
az
eNe
t
ca
n
r
ed
u
ce
th
e
h
az
e
in
th
e
o
r
ig
in
al
im
ag
e
[
2
7
]
.
C
o
n
s
eq
u
en
tly
,
t
h
e
s
eg
m
e
n
tatio
n
r
esu
lts
o
f
t
h
e
p
r
o
p
o
s
ed
GDN
ar
e
b
etter
co
m
p
ar
ed
to
th
e
GR
SS
m
o
d
el.
Ad
d
itio
n
ally
,
th
e
ef
f
ec
tiv
en
es
s
o
f
b
o
th
m
o
d
els
in
d
icate
d
b
y
er
r
o
r
,
d
ice,
an
d
J
ac
ca
r
d
ar
e
c
o
m
p
u
ted
as
well
to
s
u
p
p
o
r
t
th
e
r
esu
lts
f
r
o
m
v
is
u
al
o
b
s
er
v
atio
n
.
Mo
r
eo
v
er
,
t
h
e
p
r
o
ce
s
s
in
g
tim
e
is
also
r
ec
o
r
d
ed
to
m
ea
s
u
r
e
ef
f
icien
cy
.
T
a
b
le
2
tab
u
lates a
ll th
e
q
u
an
titativ
e
v
alu
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
3
,
Sep
tem
b
er
20
25
:
4
2
9
-
43
7
434
T
ab
le
2
.
T
h
e
v
alu
es o
f
er
r
o
r
,
d
ice,
J
ac
ca
r
d
,
an
d
p
r
o
ce
s
s
in
g
ti
m
e
(
tim
e)
Te
st
I
mag
e
Er
r
o
r
D
i
c
e
Jac
c
a
r
d
Ti
me
G
R
S
S
GDN
G
R
S
S
GDN
G
R
S
S
GDN
G
R
S
S
GDN
1
0
.
0
4
3
0
0
.
0
3
1
5
0
.
8
7
1
7
0
.
9
1
0
0
0
.
7
7
2
6
0
.
8
3
4
9
7
.
5
5
0
1
1
.
9
3
0
2
0
.
0
4
0
6
0
.
0
3
9
5
0
.
7
2
6
4
0
.
7
3
5
2
0
.
5
7
0
3
0
.
5
8
1
3
7
.
7
6
0
1
1
.
1
5
0
3
0
.
0
7
7
2
0
.
0
7
5
6
0
.
7
3
9
9
0
.
7
4
0
7
0
.
5
8
7
2
0
.
5
8
8
1
7
.
5
4
0
1
1
.
3
0
0
4
0
.
0
0
9
3
0
.
0
0
5
9
0
.
9
7
1
9
0
.
9
8
2
4
0
.
9
4
5
3
0
.
9
6
5
4
7
.
3
0
0
1
1
.
7
7
0
5
0
.
0
4
3
6
0
.
0
1
5
3
0
.
9
5
6
5
0
.
9
8
5
2
0
.
9
1
6
6
0
.
9
7
0
8
2
2
.
6
0
0
4
1
.
4
0
0
6
0
.
0
9
3
0
0
.
0
2
7
4
0
.
9
1
5
7
0
.
9
7
6
5
0
.
8
4
4
5
0
.
9
5
4
1
1
9
.
5
9
0
3
4
.
4
1
0
7
0
.
0
7
3
0
0
.
0
2
7
2
0
.
9
2
3
9
0
.
9
7
2
9
0
.
8
5
8
6
0
.
9
4
7
3
1
9
.
7
3
0
3
5
.
0
6
0
8
0
.
0
4
7
9
0
.
0
2
7
9
0
.
9
2
7
0
0
.
9
5
8
7
0
.
8
6
4
0
0
.
9
2
0
6
1
9
.
2
0
0
3
3
.
8
9
0
9
0
.
0
1
9
9
0
.
0
1
9
8
0
.
9
1
0
1
0
.
9
1
1
7
0
.
8
3
5
0
0
.
8
3
7
7
4
.
8
8
0
6
.
1
2
0
0
10
0
.
1
2
9
8
0
.
0
7
7
8
0
.
7
2
9
3
0
.
8
5
0
4
0
.
5
7
3
9
0
.
7
3
9
7
5
.
7
6
0
7
.
5
1
0
0
11
0
.
0
4
3
7
0
.
0
3
6
3
0
.
9
5
8
7
0
.
9
6
6
3
0
.
9
2
0
7
0
.
9
3
4
8
1
9
.
1
0
0
3
4
.
1
7
0
12
0
.
1
2
8
5
0
.
0
7
4
2
0
.
8
0
5
8
0
.
8
9
8
9
0
.
6
7
4
8
0
.
8
1
6
4
8
.
8
0
0
1
3
.
1
6
0
A
v
e
r
a
g
e
0
.
0
6
2
5
0
.
0
3
8
2
0
.
8
6
9
7
0
.
9
0
7
4
0
.
7
8
0
3
0
.
8
4
0
9
1
2
.
4
8
4
2
0
.
9
8
9
B
ased
o
n
T
ab
le
2
,
t
h
e
GDN
m
o
d
el
co
n
s
is
ten
tly
ac
h
iev
e
d
th
e
lo
west
E
r
r
o
r
v
al
u
e
ac
r
o
s
s
m
o
s
t
d
atasets
f
o
r
r
ea
l
im
ag
es.
T
h
e
m
ea
n
E
r
r
o
r
v
al
u
e
with
r
eg
ar
d
to
th
e
G
DN
m
o
d
el
is
0
.
0
3
8
2
,
wh
ile
th
e
m
ea
n
er
r
o
r
v
al
u
e
f
o
r
th
e
GR
SS
m
o
d
el
is
0
.
0
6
2
5
.
T
h
e
m
i
n
im
al
v
alu
e
o
f
th
e
E
r
r
o
r
in
d
icate
s
o
p
tim
al
ac
cu
r
ac
y
in
im
ag
e
s
eg
m
en
tatio
n
.
T
h
e
GDN
m
o
d
el
ea
r
n
ed
th
e
h
ig
h
est
Dice
v
alu
e
ac
r
o
s
s
alm
o
s
t
all
th
e
d
ata
s
ets.
T
h
e
DS
C
v
alu
e
av
er
ag
e
f
o
r
th
e
GDN
m
o
d
el
is
0
.
9
0
7
4
,
s
u
r
p
ass
in
g
th
e
GR
SS
m
o
d
el.
T
h
is
r
esu
lt
s
ig
n
if
ies
th
at
th
e
GDN
m
o
d
el
ac
h
iev
ed
th
e
h
ig
h
est
s
eg
m
en
ta
tio
n
ac
cu
r
ac
y
in
co
m
p
ar
is
o
n
t
o
th
e
ex
is
tin
g
GR
SS
m
o
d
el,
h
av
in
g
an
av
er
a
g
e
o
f
0
.
8
6
9
7
.
At
th
e
s
am
e
tim
e
,
th
e
GDN
m
o
d
el
co
n
s
is
ten
tly
ac
h
i
ev
ed
th
e
h
ig
h
est
J
ac
ca
r
d
v
alu
e
ac
r
o
s
s
m
o
s
t
o
f
th
e
d
ata.
T
h
e
m
ea
n
J
ac
ca
r
d
v
alu
e
f
o
r
th
e
GDN
m
o
d
el
is
0
.
8
4
0
9
.
T
h
e
GR
SS
m
o
d
el
y
iel
d
s
th
e
lo
west
J
ac
ca
r
d
v
alu
e
with
a
n
av
e
r
ag
e
J
ac
ca
r
d
v
alu
e
o
f
0
.
7
8
0
3
.
T
h
e
h
ig
h
e
s
t
v
alu
e
o
f
J
ac
ca
r
d
i
n
d
icate
s
a
h
ig
h
e
r
ac
cu
r
ac
y
in
im
ag
e
s
eg
m
en
tatio
n
.
T
h
ese
f
in
d
in
g
s
ar
e
p
ar
allel
with
th
e
v
is
u
al
o
b
s
er
v
atio
n
b
ased
o
n
T
ab
le
1
m
ad
e
ab
o
v
e.
Ad
d
itio
n
ally
,
th
ese
r
esu
lts
ar
e
ev
i
d
en
ce
o
f
th
e
ad
v
an
tag
es
o
f
c
o
m
b
in
in
g
t
h
e
im
ag
e
d
eh
az
i
n
g
tec
h
n
iq
u
e
with
AC
M
in
a
n
ew
p
r
o
p
o
s
ed
f
o
r
m
u
latio
n
o
f
th
e
GDN
m
o
d
el
th
at
is
ca
p
ab
le
o
f
p
r
o
d
u
cin
g
m
o
r
e
p
r
ec
is
e
s
eg
m
en
tatio
n
in
co
m
p
ar
is
o
n
t
o
th
e
o
r
ig
in
al
G
R
SS
m
o
d
el.
Ho
wev
er
,
in
co
r
p
o
r
atin
g
th
e
d
eh
az
in
g
tech
n
i
q
u
e
r
esu
lts
in
a
h
ig
h
er
co
m
p
u
tatio
n
c
o
s
t
wh
en
f
o
r
m
u
latin
g
th
e
p
r
o
p
o
s
ed
GDN
co
m
p
ar
ed
t
o
th
e
ex
is
tin
g
GR
S
S
m
o
d
el.
T
h
u
s
,
th
e
av
er
ag
e
p
r
o
ce
s
s
in
g
tim
e
f
o
r
th
e
GR
SS
m
o
d
el
is
th
e
f
astes
t
at
1
2
.
4
8
4
s
ec
o
n
d
s
,
f
o
llo
wed
b
y
th
e
GDN
m
o
d
el
at
2
0
.
9
8
9
s
ec
o
n
d
s
.
T
o
co
n
clu
d
e,
alth
o
u
g
h
th
e
p
r
o
p
o
s
ed
GDN
m
o
d
el
is
s
lo
wer
th
an
th
e
GR
SS
m
o
d
el,
th
e
ex
p
er
im
en
ts
r
ev
ea
le
d
th
at
th
e
GDN
m
o
d
el
b
ased
o
n
th
e
Deh
az
eNe
t
d
eh
az
in
g
tech
n
iq
u
e
p
r
o
d
u
ce
d
t
h
e
h
ig
h
est
ac
cu
r
ac
y
,
as
d
em
o
n
s
tr
ated
b
y
th
e
h
ig
h
er
a
v
er
ag
e
v
alu
es
o
f
J
ac
ca
r
d
,
as
well
as
Dice
v
al
u
es,
an
d
th
e
lo
west
E
r
r
o
r
v
alu
e
co
m
p
ar
ed
to
th
e
G
R
SS
m
o
d
el.
4.
CO
NCLU
SI
O
N
T
h
is
wo
r
k
r
ef
o
r
m
u
lates
th
e
GR
SS
m
o
d
el
f
o
r
h
az
y
im
ag
e
s
eg
m
en
tatio
n
,
u
s
in
g
t
h
e
D
eh
az
eNe
t
d
eh
az
in
g
ap
p
r
o
ac
h
as a
d
d
itio
n
al
f
itti
n
g
p
ar
am
eter
s
,
r
esu
ltin
g
in
a
m
o
d
if
ied
v
ar
iatio
n
k
n
o
w
n
as th
e
GR
SS
wi
th
Deh
az
eNe
t
(
GDN)
m
o
d
el.
T
h
is
r
esear
ch
’
s
f
in
d
in
g
s
d
eter
m
in
ed
th
at
th
e
p
r
o
p
o
s
ed
GDN
m
o
d
el
g
en
er
ated
b
etter
im
ag
e
s
eg
m
en
tatio
n
q
u
a
lity
in
co
m
p
ar
is
o
n
to
th
e
GR
SS
m
o
d
el,
as th
e
d
e
h
az
in
g
ter
m
in
th
e
GDN
m
o
d
e
l
was
ca
p
ab
le
o
f
r
ed
u
cin
g
th
e
i
m
ag
e
h
az
e
an
d
c
o
n
s
eq
u
e
n
tly
h
elp
ed
in
g
e
n
er
atin
g
h
ig
h
s
eg
m
en
tatio
n
ac
cu
r
ac
y
.
T
h
e
o
u
tco
m
es
m
ay
b
en
ef
it
s
ev
er
al
ap
p
licatio
n
s
,
in
cl
u
d
in
g
o
b
ject
tr
ac
k
in
g
,
d
r
iv
er
less
(
a
u
t
o
n
o
m
o
u
s
)
ca
r
s
,
an
d
tr
af
f
ic
s
u
r
v
eillan
ce
.
T
h
e
p
r
im
ar
y
lim
itatio
n
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
lies
in
its
h
ig
h
p
r
o
c
ess
in
g
tim
e,
wh
ich
af
f
ec
ts
o
v
er
all
ef
f
icien
c
y
.
C
o
n
s
eq
u
en
tly
,
f
u
tu
r
e
r
esear
ch
m
ay
f
o
c
u
s
o
n
d
ev
elo
p
in
g
f
aster
o
p
tim
izatio
n
s
tr
ateg
ies to
s
o
lv
e
th
e
m
o
d
el
m
o
r
e
ef
f
icien
tly
.
A
m
o
r
e
co
m
p
r
eh
en
s
iv
e
ev
alu
atio
n
ac
r
o
s
s
d
iv
er
s
e
en
v
ir
o
n
m
e
n
ts
an
d
a
p
p
licatio
n
d
o
m
ain
s
is
al
s
o
n
ec
ess
ar
y
to
f
u
r
th
er
v
alid
at
e
th
e
m
o
d
el’
s
r
o
b
u
s
tn
ess
an
d
g
en
er
aliza
b
ilit
y
.
I
n
ad
d
itio
n
,
th
e
m
o
d
el
h
as
th
e
p
o
ten
tial
to
b
e
ex
te
n
d
ed
to
co
lo
r
im
ag
e
s
eg
m
en
tatio
n
b
y
in
co
r
p
o
r
atin
g
alter
n
ativ
e
d
eh
az
in
g
tech
n
iq
u
es
o
r
c
o
lo
r
s
p
ac
e
tr
an
s
f
o
r
m
atio
n
s
.
Alth
o
u
g
h
th
is
s
tu
d
y
c
o
n
ce
n
tr
ate
d
o
n
g
r
ay
s
ca
le
im
ag
es
to
estab
lis
h
th
e
m
o
d
el’
s
co
r
e
p
e
r
f
o
r
m
a
n
ce
,
f
u
tu
r
e
wo
r
k
will
in
v
esti
g
ate
its
ad
ap
tatio
n
to
c
o
lo
r
im
ag
es,
wh
ic
h
co
u
ld
e
n
h
an
ce
its
r
elev
an
ce
t
o
r
ea
l
-
wo
r
ld
s
ce
n
ar
i
o
s
.
W
h
ile
th
is
s
tu
d
y
f
o
cu
s
ed
o
n
s
tatic
im
ag
e
s
eg
m
en
tatio
n
,
th
e
p
r
o
p
o
s
ed
m
o
d
el
also
s
h
o
ws
p
r
o
m
is
e
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
,
in
clu
d
in
g
v
id
e
o
-
b
ased
s
eg
m
en
tatio
n
.
Fu
tu
r
e
r
esear
ch
m
a
y
ex
p
lo
r
e
i
ts
im
p
lem
en
tatio
n
in
r
ea
l
-
tim
e
s
ce
n
ar
io
s
b
y
in
teg
r
atin
g
o
p
ti
m
izatio
n
tech
n
iq
u
es
aim
ed
at
r
e
d
u
cin
g
co
m
p
u
tatio
n
al
laten
cy
.
Fu
r
th
er
m
o
r
e,
e
v
al
u
atin
g
th
e
m
o
d
el
o
n
s
eq
u
en
tia
l
v
id
eo
d
ata
wo
u
l
d
en
ab
le
ass
ess
m
en
t
o
f
its
tem
p
o
r
al
co
n
s
is
ten
cy
a
n
d
s
eg
m
e
n
t
atio
n
r
o
b
u
s
tn
ess
,
th
er
eb
y
s
u
p
p
o
r
tin
g
its
p
r
ac
tical
d
ep
lo
y
m
e
n
t in
f
ield
s
s
u
ch
as
m
ed
ical
im
ag
in
g
,
s
u
r
v
eillan
ce
,
an
d
au
t
o
n
o
m
o
u
s
s
y
s
tem
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
I
SS
N:
2722
-
2
5
8
6
Hyb
r
id
d
ee
p
lea
r
n
in
g
a
n
d
a
ct
ive
co
n
to
u
r
fo
r
s
eg
men
tin
g
h
a
z
y
ima
g
es
(
F
ir
h
a
n
A
z
r
i A
h
ma
d
K
h
a
ir
u
l A
n
u
a
r
)
435
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
t
h
o
r
s
e
x
ten
d
th
eir
s
in
c
er
e
g
r
atitu
d
e
t
o
Un
iv
er
s
iti
T
e
k
n
o
lo
g
i
MA
R
A
(
UiT
M)
f
o
r
s
u
p
p
o
r
tin
g
th
is
r
esear
ch
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
r
ec
ei
v
ed
n
o
s
p
e
cif
ic
g
r
an
t
f
r
o
m
an
y
f
u
n
d
in
g
a
g
en
cy
i
n
th
e
p
u
b
lic,
co
m
m
er
ci
al,
o
r
n
o
t
-
f
o
r
-
p
r
o
f
it secto
r
s
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Firh
an
Azr
i
Ah
m
ad
Kh
air
u
l A
n
u
ar
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
J
en
ev
y
J
o
n
e
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
aja
Far
h
atu
l
Aiesy
a
R
aja
Azh
ar
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ab
d
u
l K
ad
ir
J
u
m
aa
t
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
tu
a
li
z
a
ti
o
n
M
:
M
e
th
o
d
o
l
o
g
y
So
:
So
ftwa
re
Va
:
Va
li
d
a
ti
o
n
Fo
:
Fo
rm
a
l
a
n
a
ly
sis
I
:
I
n
v
e
stig
a
ti
o
n
R
:
R
e
so
u
rc
e
s
D
:
D
a
ta Cu
ra
ti
o
n
O
:
Wr
it
in
g
-
O
ri
g
in
a
l
Dra
ft
E
:
Wr
it
in
g
-
Re
v
iew
&
E
d
it
i
n
g
Vi
:
Vi
su
a
li
z
a
ti
o
n
Su
:
Su
p
e
rv
isi
o
n
P
:
P
ro
jec
t
a
d
m
in
istrati
o
n
Fu
:
Fu
n
d
in
g
a
c
q
u
isi
ti
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
r
eso
u
r
ce
s
u
n
d
er
p
in
n
in
g
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
p
u
b
licly
ac
ce
s
s
ib
le
in
r
ef
er
en
ce
s
[
2
8
]
–
[
3
2
]
.
RE
F
E
R
E
NC
E
S
[
1
]
F
.
A
.
S
h
e
w
a
j
o
a
n
d
K
.
A
.
F
a
n
t
e
,
“
Ti
l
e
-
b
a
s
e
d
mi
c
r
o
sc
o
p
i
c
i
ma
g
e
p
r
o
c
e
ss
i
n
g
f
o
r
ma
l
a
r
i
a
s
c
r
e
e
n
i
n
g
u
si
n
g
a
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
BM
C
Me
d
i
c
a
l
I
m
a
g
i
n
g
,
v
o
l
.
2
3
,
n
o
.
1
,
p
p
.
1
–
1
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
2
8
8
0
-
0
2
3
-
0
0
9
9
3
-
9.
[
2
]
A
.
M
.
A
n
t
e
r
a
n
d
L.
A
b
u
a
l
i
g
a
h
,
“
D
e
e
p
f
e
d
e
r
a
t
e
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
o
p
t
i
m
i
z
a
t
i
o
n
me
t
h
o
d
s
f
o
r
l
i
v
e
r
t
u
m
o
r
d
i
a
g
n
o
si
s
:
A
r
e
v
i
e
w
,
”
Arc
h
i
v
e
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
Me
t
h
o
d
s
i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
3
0
,
n
o
.
5
,
p
p
.
3
3
5
9
–
3
3
7
8
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
8
3
1
-
023
-
0
9
9
0
1
-
4.
[
3
]
S
.
S
r
i
n
i
v
a
sa
n
,
P
.
S
.
M
.
B
a
i
,
S
.
K
.
M
a
t
h
i
v
a
n
a
n
,
V
.
M
u
t
h
u
k
u
m
a
r
a
n
,
J.
C
.
B
a
b
u
,
a
n
d
L
.
V
i
l
c
e
k
o
v
a
,
“
G
r
a
d
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
t
u
mo
r
s
f
r
o
m
b
r
a
i
n
m
a
g
n
e
t
i
c
r
e
so
n
a
n
c
e
i
ma
g
e
s
u
s
i
n
g
a
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
”
D
i
a
g
n
o
st
i
c
s
,
v
o
l
.
1
3
,
n
o
.
6
,
p
.
1
1
5
3
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
s
t
i
c
s1
3
0
6
1
1
5
3
.
[
4
]
Z.
Q
.
H
a
b
e
e
b
,
B
.
V
u
k
sa
n
o
v
i
c
,
a
n
d
I
.
Q
.
A
l
-
Za
y
d
i
,
“
B
r
e
a
st
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
i
m
a
g
e
p
r
o
c
e
ssi
n
g
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
J
o
u
rn
a
l
o
f
I
m
a
g
e
a
n
d
G
r
a
p
h
i
c
s(
U
n
i
t
e
d
K
i
n
g
d
o
m
)
,
v
o
l
.
1
1
,
n
o
.
1
,
p
p
.
1
–
8
,
2
0
2
3
,
d
o
i
:
1
0
.
1
8
1
7
8
/
j
o
i
g
.
1
1
.
1
.
1
-
8.
[
5
]
I
.
M
i
sh
r
a
,
K
.
A
r
a
v
i
n
d
a
,
J
.
A
.
K
u
mar
,
C
.
K
e
e
r
t
h
i
,
R
.
D
i
v
y
a
S
h
r
e
e
,
a
n
d
S
.
S
r
i
k
u
mar,
“
M
e
d
i
c
a
l
i
ma
g
i
n
g
u
si
n
g
si
g
n
a
l
p
r
o
c
e
ss
i
n
g
:
A
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
rt
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
S
m
a
rt
E
n
e
r
g
y
,
I
C
AI
S
2
0
2
2
,
2
0
2
2
,
p
p
.
6
2
3
–
6
3
0
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
I
S
5
3
3
1
4
.
2
0
2
2
.
9
7
4
2
7
7
8
.
[
6
]
M
.
I
.
Y
e
şi
l
a
n
d
S
.
G
ö
n
c
ü
,
“
R
e
c
o
g
n
i
t
i
o
n
o
f
H
e
r
e
f
o
r
d
a
n
d
S
i
m
me
n
t
a
l
c
a
t
t
l
e
b
r
e
e
d
s v
i
a
c
o
mp
u
t
e
r
v
i
si
o
n
,
”
I
ra
n
i
a
n
J
o
u
rn
a
l
o
f
A
p
p
l
i
e
d
An
i
m
a
l
S
c
i
e
n
c
e
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
2
1
–
3
2
,
2
0
2
3
.
[
7
]
J.
F
a
n
g
,
H
.
Li
u
,
L.
Z
h
a
n
g
,
J
.
L
i
u
,
a
n
d
H
.
Li
u
,
“
R
e
g
i
o
n
-
e
d
g
e
-
b
a
s
e
d
a
c
t
i
v
e
c
o
n
t
o
u
r
s
d
r
i
v
e
n
b
y
h
y
b
r
i
d
a
n
d
l
o
c
a
l
f
u
z
z
y
r
e
g
i
o
n
-
b
a
se
d
e
n
e
r
g
y
f
o
r
i
m
a
g
e
se
g
m
e
n
t
a
t
i
o
n
,
”
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
5
4
6
,
p
p
.
3
9
7
–
4
1
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
s.
2
0
2
0
.
0
8
.
0
7
8
.
[
8
]
R
.
Y
o
u
sef
e
t
a
l
.
,
“
U
-
N
e
t
-
b
a
s
e
d
m
o
d
e
l
s
t
o
w
a
r
d
s
o
p
t
i
ma
l
M
R
b
r
a
i
n
i
ma
g
e
se
g
me
n
t
a
t
i
o
n
,
”
D
i
a
g
n
o
s
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
9
,
p
.
1
6
2
4
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
s
t
i
c
s1
3
0
9
1
6
2
4
.
[
9
]
K
.
C
h
e
n
,
“
I
n
t
r
o
d
u
c
t
i
o
n
t
o
v
a
r
i
a
t
i
o
n
a
l
i
m
a
g
e
-
p
r
o
c
e
ss
i
n
g
mo
d
e
l
s
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
Ma
t
h
e
m
a
t
i
c
s
,
v
o
l
.
9
0
,
n
o
.
1
,
p
p
.
1
–
8
,
2
0
1
3
,
d
o
i
:
1
0
.
1
0
8
0
/
0
0
2
0
7
1
6
0
.
2
0
1
2
.
7
5
7
0
7
3
.
[
1
0
]
G.
-
L.
Z
h
u
,
X
.
-
G
.
L
v
,
F
.
Li
,
a
n
d
X
.
-
M
.
S
u
n
,
“
N
o
n
c
o
n
v
e
x
v
a
r
i
a
t
i
o
n
a
l
a
p
p
r
o
a
c
h
f
o
r
s
i
m
u
l
t
a
n
e
o
u
s
l
y
r
e
c
o
v
e
r
i
n
g
c
a
r
t
o
o
n
a
n
d
t
e
x
t
u
r
e
i
ma
g
e
s,
”
J
o
u
r
n
a
l
o
f
El
e
c
t
ro
n
i
c
I
m
a
g
i
n
g
,
v
o
l
.
3
1
,
n
o
.
0
4
,
p
.
4
3
0
2
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
1
7
/
1
.
j
e
i
.
3
1
.
4
.
0
4
3
0
2
1
.
[
1
1
]
X
.
Li
u
,
G
.
Li
u
,
Y
.
W
a
n
g
,
G
.
L
i
,
R
.
Zh
a
n
g
,
a
n
d
W
.
P
e
n
g
,
“
A
v
a
r
i
a
t
i
o
n
a
l
l
e
v
e
l
se
t
i
ma
g
e
s
e
g
me
n
t
a
t
i
o
n
m
e
t
h
o
d
v
i
a
f
r
a
c
t
i
o
n
a
l
d
i
f
f
e
r
e
n
t
i
a
t
i
o
n
,
”
Fr
a
c
t
a
l
a
n
d
Fra
c
t
i
o
n
a
l
,
v
o
l
.
6
,
n
o
.
9
,
p
.
4
6
2
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
f
r
a
c
t
a
l
f
r
a
c
t
6
0
9
0
4
6
2
.
[
1
2
]
T.
C
.
S
a
i
b
i
n
a
n
d
A
.
K
.
Ju
m
a
a
t
,
“
V
a
r
i
a
t
i
o
n
a
l
se
l
e
c
t
i
v
e
se
g
me
n
t
a
t
i
o
n
mo
d
e
l
f
o
r
i
n
t
e
n
s
i
t
y
i
n
h
o
m
o
g
e
n
e
o
u
s
i
m
a
g
e
,
”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
E
l
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
EE
C
S
)
,
v
o
l
.
2
9
,
n
o
.
1
,
p
p
.
2
7
7
–
2
8
5
,
2
0
2
3
,
d
o
i
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
3
,
Sep
tem
b
er
20
25
:
4
2
9
-
43
7
436
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
2
9
.
i
1
.
p
p
2
7
7
-
2
8
5
.
[
1
3
]
J.
G
u
,
Z.
F
a
n
g
,
Y
.
G
a
o
,
a
n
d
F
.
Ti
a
n
,
“
S
e
g
me
n
t
a
t
i
o
n
o
f
c
o
r
o
n
a
r
y
a
r
t
e
r
i
e
s
i
ma
g
e
s
u
si
n
g
g
l
o
b
a
l
f
e
a
t
u
r
e
e
mb
e
d
d
e
d
n
e
t
w
o
r
k
w
i
t
h
a
c
t
i
v
e
c
o
n
t
o
u
r
l
o
ss
,
”
C
o
m
p
u
t
e
ri
ze
d
Me
d
i
c
a
l
I
m
a
g
i
n
g
a
n
d
G
ra
p
h
i
c
s
,
v
o
l
.
8
6
,
p
.
1
0
1
7
9
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
me
d
i
m
a
g
.
2
0
2
0
.
1
0
1
7
9
9
.
[
1
4
]
Y
.
Y
a
n
g
,
W
.
Ji
a
,
a
n
d
B
.
W
u
,
“
S
i
m
u
l
t
a
n
e
o
u
s
se
g
me
n
t
a
t
i
o
n
a
n
d
c
o
r
r
e
c
t
i
o
n
m
o
d
e
l
f
o
r
c
o
l
o
r
me
d
i
c
a
l
a
n
d
n
a
t
u
r
a
l
i
ma
g
e
s
w
i
t
h
i
n
t
e
n
si
t
y
i
n
h
o
m
o
g
e
n
e
i
t
y
,
”
V
i
su
a
l
C
o
m
p
u
t
e
r
,
v
o
l
.
3
6
,
n
o
.
4
,
p
p
.
7
1
7
–
7
3
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
3
7
1
-
019
-
0
1
6
5
1
-
4.
[
1
5
]
E.
I
q
b
a
l
,
A
.
N
i
a
z
,
A
.
A
.
M
e
m
o
n
,
U
.
A
si
m
,
a
n
d
K
.
N
.
C
h
o
i
,
“
S
a
l
i
e
n
c
y
d
r
i
v
e
n
r
e
g
i
o
n
-
e
d
g
e
-
b
a
s
e
d
t
o
p
d
o
w
n
l
e
v
e
l
s
e
t
e
v
o
l
u
t
i
o
n
r
e
v
e
a
l
s
t
h
e
a
s
y
n
c
h
r
o
n
o
u
s
f
o
c
u
s
i
n
i
m
a
g
e
s
e
g
m
e
n
t
a
t
i
o
n
,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
2
0
8
9
7
8
–
2
0
8
9
9
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
3
0
3
8
9
4
5
.
[
1
6
]
X.
-
H
.
Zh
i
a
n
d
H
.
-
B
.
S
h
e
n
,
“
S
a
l
i
e
n
c
y
d
r
i
v
e
n
r
e
g
i
o
n
-
e
d
g
e
-
b
a
se
d
t
o
p
d
o
w
n
l
e
v
e
l
se
t
e
v
o
l
u
t
i
o
n
r
e
v
e
a
l
s
t
h
e
a
sy
n
c
h
r
o
n
o
u
s
f
o
c
u
s
i
n
i
ma
g
e
s
e
g
me
n
t
a
t
i
o
n
,
”
P
a
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
,
v
o
l
.
8
0
,
p
p
.
2
4
1
–
2
5
5
,
A
u
g
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
c
o
g
.
2
0
1
8
.
0
3
.
0
1
0
.
[
1
7
]
W
.
Z
h
a
o
,
W
.
W
a
n
g
,
X
.
F
e
n
g
,
a
n
d
Y
.
H
a
n
,
“
A
n
e
w
v
a
r
i
a
t
i
o
n
a
l
m
e
t
h
o
d
f
o
r
se
l
e
c
t
i
v
e
se
g
me
n
t
a
t
i
o
n
o
f
m
e
d
i
c
a
l
i
ma
g
e
s
,
”
S
i
g
n
a
l
Pro
c
e
ssi
n
g
,
v
o
l
.
1
9
0
,
p
.
1
0
8
2
9
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
si
g
p
r
o
.
2
0
2
1
.
1
0
8
2
9
2
.
[
1
8
]
N
.
A
.
S
.
M
.
G
h
a
n
i
a
n
d
A
.
K
.
Ju
m
a
a
t
,
“
S
e
l
e
c
t
i
v
e
s
e
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
f
o
r
v
e
c
t
o
r
-
v
a
l
u
e
d
i
ma
g
e
s
,
”
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
1
,
n
o
.
2
,
p
p
.
1
4
9
–
1
7
3
,
2
0
2
2
,
d
o
i
:
1
0
.
3
2
8
9
0
/
j
i
c
t
2
0
2
2
.
2
1
.
2
.
1
.
[
1
9
]
N
.
A
.
S
.
M
.
G
h
a
n
i
,
A
.
K
.
Ju
maa
t
,
R
.
M
a
h
m
u
d
,
M
.
A
.
M
a
a
sar,
F
.
A
.
Z
u
l
k
i
f
l
e
,
a
n
d
A
.
M
.
J
a
si
n
,
“
B
r
e
a
st
a
b
n
o
r
mal
i
t
y
b
o
u
n
d
a
r
y
e
x
t
r
a
c
t
i
o
n
i
n
m
a
mm
o
g
r
a
p
h
y
i
m
a
g
e
u
si
n
g
v
a
r
i
a
t
i
o
n
a
l
l
e
v
e
l
se
t
a
n
d
s
e
l
f
-
o
r
g
a
n
i
z
i
n
g
m
a
p
(
S
O
M
)
,
”
M
a
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
1
,
n
o
.
4
,
p
.
9
7
6
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
mat
h
1
1
0
4
0
9
7
6
.
[
2
0
]
S
.
S
.
Y
a
s
i
r
a
n
e
t
a
l
.
,
“
C
o
mp
a
r
i
s
o
n
b
e
t
w
e
e
n
G
V
F
sn
a
k
e
a
n
d
ED
s
n
a
k
e
i
n
s
e
g
me
n
t
i
n
g
m
i
c
r
o
c
a
l
c
i
f
i
c
a
t
i
o
n
s,
”
i
n
I
C
C
AI
E
2
0
1
1
-
2
0
1
1
I
EEE
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
Ap
p
l
i
c
a
t
i
o
n
s
a
n
d
I
n
d
u
st
r
i
a
l
El
e
c
t
r
o
n
i
c
s
,
D
e
c
.
2
0
1
1
,
p
p
.
5
9
7
–
6
0
1
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
A
I
E.
2
0
1
1
.
6
1
6
2
2
0
4
.
[
2
1
]
N
.
A
.
K
.
Za
m
a
n
,
W
.
E.
Z.
W
.
A
.
R
a
h
man
,
A
.
K
.
J
u
ma
a
t
,
a
n
d
S
.
S
.
Y
a
si
r
a
n
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
o
f
b
r
e
a
st
a
b
n
o
r
ma
l
i
t
i
e
s
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
AI
P
C
o
n
f
e
r
e
n
c
e
P
ro
c
e
e
d
i
n
g
s
,
M
a
y
2
0
1
5
,
v
o
l
.
1
6
6
0
,
n
o
.
1
,
p
.
5
0
0
3
8
.
d
o
i
:
1
0
.
1
0
6
3
/
1
.
4
9
1
5
6
7
1
.
[
2
2
]
N
.
B
a
d
s
h
a
h
a
n
d
K
.
C
h
e
n
,
“
I
mag
e
s
e
l
e
c
t
i
v
e
s
e
g
me
n
t
a
t
i
o
n
u
n
d
e
r
g
e
o
m
e
t
r
i
c
a
l
c
o
n
st
r
a
i
n
t
s
u
si
n
g
a
n
a
c
t
i
v
e
c
o
n
t
o
u
r
a
p
p
r
o
a
c
h
,
”
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
P
h
y
s
i
c
s
,
v
o
l
.
7
,
n
o
.
4
,
p
p
.
7
5
9
–
7
7
8
,
2
0
1
0
,
d
o
i
:
1
0
.
4
2
0
8
/
c
i
c
p
.
2
0
0
9
.
0
9
.
0
2
6
.
[
2
3
]
L.
R
a
d
a
a
n
d
K
.
C
h
e
n
,
“
I
mp
r
o
v
e
d
se
l
e
c
t
i
v
e
s
e
g
me
n
t
a
t
i
o
n
mo
d
e
l
u
s
i
n
g
o
n
e
l
e
v
e
l
-
se
t
,
”
J
o
u
r
n
a
l
o
f
A
l
g
o
r
i
t
h
m
s
a
n
d
C
o
m
p
u
t
a
t
i
o
n
a
l
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
7
,
n
o
.
4
,
p
p
.
5
0
9
–
5
4
0
,
2
0
1
3
,
d
o
i
:
1
0
.
1
2
6
0
/
1
7
4
8
-
3
0
1
8
.
7
.
4
.
5
0
9
.
[
2
4
]
J.
S
p
e
n
c
e
r
a
n
d
K
.
C
h
e
n
,
“
A
c
o
n
v
e
x
a
n
d
s
e
l
e
c
t
i
v
e
v
a
r
i
a
t
i
o
n
a
l
mo
d
e
l
f
o
r
i
mag
e
s
e
g
me
n
t
a
t
i
o
n
,
”
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
M
a
t
h
e
m
a
t
i
c
a
l
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
6
,
p
p
.
1
4
5
3
–
1
4
7
2
,
2
0
1
5
,
d
o
i
:
1
0
.
4
3
1
0
/
C
M
S
.
2
0
1
5
.
v
1
3
.
n
6
.
a
5
.
[
2
5
]
A
.
K
.
J
u
ma
a
t
a
n
d
K
.
C
h
e
n
,
“
A
r
e
f
o
r
mu
l
a
t
e
d
c
o
n
v
e
x
a
n
d
sel
e
c
t
i
v
e
v
a
r
i
a
t
i
o
n
a
l
i
m
a
g
e
se
g
m
e
n
t
a
t
i
o
n
m
o
d
e
l
a
n
d
i
t
s
f
a
st
mu
l
t
i
l
e
v
e
l
a
l
g
o
r
i
t
h
m,
”
N
u
m
e
r
i
c
a
l
M
a
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
2
,
p
p
.
4
0
3
–
4
3
7
,
J
u
n
.
2
0
1
9
,
d
o
i
:
1
0
.
4
2
0
8
/
n
mt
m
a
.
O
A
-
2
0
1
7
-
0
1
4
3
.
[
2
6
]
H
.
A
l
i
,
A
.
S
h
e
r
,
M
.
S
a
e
e
d
,
a
n
d
L.
R
a
d
a
,
“
A
c
t
i
v
e
c
o
n
t
o
u
r
i
ma
g
e
se
g
me
n
t
a
t
i
o
n
mo
d
e
l
w
i
t
h
d
e
-
h
a
z
i
n
g
c
o
n
st
r
a
i
n
t
s,
”
I
ET
I
m
a
g
e
Pro
c
e
ssi
n
g
,
v
o
l
.
1
4
,
n
o
.
5
,
p
p
.
9
2
1
–
9
2
8
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
4
9
/
i
e
t
-
i
p
r
.
2
0
1
8
.
5
9
8
7
.
[
2
7
]
B
.
C
a
i
,
X
.
X
u
,
K
.
Ji
a
,
C
.
Q
i
n
g
,
a
n
d
D
.
Ta
o
,
“
D
e
h
a
z
e
N
e
t
:
A
n
e
n
d
-
to
-
e
n
d
s
y
st
e
m
f
o
r
si
n
g
l
e
i
ma
g
e
h
a
z
e
r
e
mo
v
a
l
,
”
I
EE
E
T
ra
n
s
a
c
t
i
o
n
s
o
n
I
m
a
g
e
Pro
c
e
ss
i
n
g
,
v
o
l
.
2
5
,
n
o
.
1
1
,
p
p
.
5
1
8
7
–
5
1
9
8
,
N
o
v
.
2
0
1
6
,
d
o
i
:
1
0
.
1
1
0
9
/
TI
P
.
2
0
1
6
.
2
5
9
8
6
8
1
.
[
2
8
]
V
.
T
.
P
h
a
m
,
“
H
o
w
t
o
u
s
e
D
e
t
e
c
t
r
o
n
2
,
”
G
i
t
H
u
b
.
h
t
t
p
s
:
/
/
g
i
t
h
u
b
.
c
o
m
/
f
a
c
e
b
o
o
k
r
e
s
e
a
r
c
h
/
d
e
t
e
c
t
r
o
n
2
/
i
s
s
u
e
s
/
7
9
5
(
a
c
c
e
s
s
e
d
F
e
b
.
0
5
,
2
0
2
4
)
.
[
2
9
]
J.
Li
,
J
.
Z
h
a
n
g
,
a
n
d
D
.
Ta
o
,
“
D
e
e
p
a
u
t
o
ma
t
i
c
n
a
t
u
r
a
l
i
ma
g
e
ma
t
t
i
n
g
,
”
C
o
R
R
,
v
o
l
.
a
b
s/
2
1
0
7
.
0
,
2
0
2
1
.
[
3
0
]
Y
.
Q
i
a
o
e
t
a
l
.
,
“
A
t
t
e
n
t
i
o
n
-
g
u
i
d
e
d
h
i
e
r
a
r
c
h
i
c
a
l
st
r
u
c
t
u
r
e
a
g
g
r
e
g
a
t
i
o
n
f
o
r
i
mag
e
m
a
t
t
i
n
g
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
EE
E
C
o
m
p
u
t
e
r
S
o
c
i
e
t
y
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
,
2
0
2
0
,
p
p
.
1
3
6
7
3
–
1
3
6
8
2
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
4
2
6
0
0
.
2
0
2
0
.
0
1
3
6
9
.
[
3
1
]
H
.
L
i
,
J.
C
a
i
,
T
.
N
.
A
.
N
g
u
y
e
n
,
a
n
d
J.
Z
h
e
n
g
,
“
A
b
e
n
c
h
mar
k
f
o
r
s
e
man
t
i
c
i
ma
g
e
s
e
g
m
e
n
t
a
t
i
o
n
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
I
EEE
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
Mu
l
t
i
m
e
d
i
a
a
n
d
Ex
p
o
,
2
0
1
3
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
M
E
.
2
0
1
3
.
6
6
0
7
5
1
2
.
[
3
2
]
M
.
Y
.
C
h
e
n
e
t
a
l
.
,
“
A
u
t
o
ma
t
i
c
C
h
i
n
e
se
f
o
o
d
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
d
q
u
a
n
t
i
t
y
e
s
t
i
m
a
t
i
o
n
,
”
i
n
S
I
G
G
RAPH
Asi
a
2
0
1
2
T
e
c
h
n
i
c
a
l
Br
i
e
f
s,
S
A
2
0
1
2
,
2
0
1
2
.
d
o
i
:
1
0
.
1
1
4
5
/
2
4
0
7
7
4
6
.
2
4
0
7
7
7
5
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Fi
r
h
a
n
Az
r
i
Ah
m
a
d
K
h
a
iru
l
Anu
a
r
is
a
stu
d
e
n
t
o
f
th
e
B.
S
c
.
d
e
g
re
e
in
M
a
th
e
m
a
ti
c
s
fro
m
Un
i
v
e
rsiti
Te
k
n
o
lo
g
i
M
ARA
(UiTM
)
S
h
a
h
A
lam
.
His
re
se
a
rc
h
in
tere
sts
re
v
o
lv
e
a
r
o
u
n
d
m
a
th
e
m
a
ti
c
a
l
m
o
d
e
li
n
g
a
n
d
ima
g
e
p
r
o
c
e
ss
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
2
0
2
3
1
0
4
9
3
9
@s
tu
d
e
n
t.
u
it
m
.
e
d
u
.
m
y
.
J
e
n
e
v
y
J
o
n
e
o
b
tain
e
d
h
e
r
B.
S
c
.
d
e
g
re
e
in
M
a
t
h
e
m
a
ti
c
s
fro
m
U
n
i
v
e
rsiti
Te
k
n
o
lo
g
i
M
ARA
(UiTM
)
S
h
a
h
Ala
m
.
S
h
e
is
c
u
rre
n
tl
y
w
o
rk
i
n
g
a
s
a
fi
n
a
n
c
e
risk
c
o
st
c
o
n
tro
l
a
ss
istan
t
a
t
S
h
a
n
g
ri
-
La
Ra
sa
Ria,
Ko
ta
Ki
n
a
b
a
lu
.
He
r
re
se
a
rc
h
in
tere
sts
re
v
o
l
v
e
a
ro
u
n
d
m
a
th
e
m
a
ti
c
a
l
m
o
d
e
li
n
g
a
n
d
ima
g
e
p
r
o
c
e
ss
in
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
jen
e
v
y
jo
n
e
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
I
SS
N:
2722
-
2
5
8
6
Hyb
r
id
d
ee
p
lea
r
n
in
g
a
n
d
a
ct
ive
co
n
to
u
r
fo
r
s
eg
men
tin
g
h
a
z
y
ima
g
es
(
F
ir
h
a
n
A
z
r
i A
h
ma
d
K
h
a
ir
u
l A
n
u
a
r
)
437
Ra
ja
F
a
r
h
a
tu
l
Aie
sy
a
R
a
ja
Az
h
a
r
o
b
tain
e
d
h
e
r
B.
S
c
.
d
e
g
re
e
in
M
a
th
e
m
a
ti
c
s
fro
m
Un
i
v
e
rsiti
Tek
n
o
lo
g
i
M
AR
A
(UiTM
)
S
h
a
h
Ala
m
a
n
d
is
c
u
r
re
n
tl
y
p
u
rsu
i
n
g
a
M
a
ste
r
o
f
Co
m
p
u
ter
S
c
ien
c
e
(Co
m
p
u
tati
o
n
a
l
In
telli
g
e
n
c
e
)
a
t
Un
i
v
e
rsiti
M
a
lay
sia
S
a
b
a
h
.
He
r
stu
d
ies
fo
c
u
s
o
n
c
o
m
p
u
tatio
n
a
l
i
n
telli
g
e
n
c
e
tec
h
n
iq
u
e
s,
i
n
c
lu
d
in
g
n
a
tu
re
-
i
n
sp
ired
c
o
m
p
u
ti
n
g
,
a
rti
ficia
l
n
e
u
ra
l
n
e
two
r
k
s,
a
n
d
m
u
lt
i
-
o
b
j
e
c
ti
v
e
o
p
ti
m
iza
ti
o
n
.
He
r
re
se
a
rc
h
in
tere
sts
re
v
o
l
v
e
a
ro
u
n
d
m
a
th
e
m
a
ti
c
a
l
m
o
d
e
li
n
g
a
n
d
ima
g
e
p
ro
c
e
ss
in
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
a
ies
y
a
7
7
@
y
a
h
o
o
.
c
o
m
.
Abd
u
l
K
a
d
ir
J
u
m
a
a
t
o
b
tai
n
e
d
t
h
e
B.
S
c
.
a
n
d
M
.
S
c
.
d
e
g
re
e
s
in
M
a
th
e
m
a
ti
c
s
fro
m
Un
iv
e
rsiti
Te
k
n
o
lo
g
i
M
ARA
(Ui
TM
)
S
h
a
h
Ala
m
,
M
a
lay
sia
,
a
n
d
t
h
e
Un
iv
e
rsit
y
o
f
Li
v
e
r
p
o
o
l
i
n
th
e
Un
it
e
d
Ki
n
g
d
o
m
a
wa
rd
e
d
h
im
a
P
h
.
D.
d
e
g
re
e
,
p
a
rti
c
u
larl
y
in
a
p
p
li
e
d
m
a
th
e
m
a
ti
c
s
(m
a
th
e
m
a
ti
c
a
l
ima
g
in
g
m
e
th
o
d
s)
.
P
re
se
n
tl
y
,
h
e
h
o
l
d
s
t
h
e
p
o
sit
io
n
o
f
se
n
i
o
r
lec
tu
re
r
a
t
th
e
S
c
h
o
o
l
o
f
M
a
t
h
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
F
a
c
u
lt
y
o
f
C
o
m
p
u
ter
a
n
d
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
a
n
d
a
re
se
a
rc
h
fe
ll
o
w
a
t
th
e
In
stit
u
te
f
o
r
Big
Da
ta
An
a
ly
ti
c
s
a
n
d
Artifi
c
i
a
l
In
telli
g
e
n
c
e
(IBDA
AI)
in
UiTM
S
h
a
h
Ala
m
,
M
a
lay
sia
.
His
re
se
a
rc
h
in
tere
sts
e
n
c
o
m
p
a
ss
v
a
rio
u
s
field
s
su
c
h
a
s
ima
g
e
/sig
n
a
l
p
r
o
c
e
ss
in
g
,
a
rti
fici
a
l
in
telli
g
e
n
c
e
,
c
o
m
p
u
ter
v
isio
n
,
b
io
m
e
tri
c
s,
m
e
d
ica
l
ima
g
e
a
n
d
a
n
a
l
y
sis,
a
n
d
p
a
tt
e
r
n
re
c
o
g
n
it
io
n
.
His e
m
a
il
a
d
d
re
ss
is ab
d
u
l
k
a
d
ir@t
m
sk
.
u
it
m
.
e
d
u
.
m
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.