T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n Co
m
pu
t
ing
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
24
,
No
.
2
,
A
p
r
il
20
26
,
p
p
.
635
~
647
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
KA.
v
24
i
2
.
2
7
5
0
4
635
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Applica
tion o
f
t
he
t
ra
v
eling
sa
lesm
a
n problem
to o
pti
mize
skeleto
niza
tion a
nd stroke
recons
t
ruction
Alif
a
h
1
,
Dia
n Andria
na
2
,
M
u
ha
m
m
a
d Z
ulh
a
j
Alia
ns
y
a
h
3
,
L
uk
m
a
n H
a
k
im
1
,
K
ho
lid
M
urt
a
dlo
1
1
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
c
s
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
E
n
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
a
s Yu
d
h
a
r
t
a
P
a
s
u
r
u
a
n
,
P
a
su
r
u
a
n
,
I
n
d
o
n
e
s
i
a
2
R
e
s
e
a
r
c
h
C
e
n
t
e
r
f
o
r
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
C
y
b
e
r
se
c
u
r
i
t
y
,
N
a
t
i
o
n
a
l
R
e
sea
r
c
h
a
n
d
I
n
n
o
v
a
t
i
o
n
A
g
e
n
c
y
(
B
R
I
N
)
,
B
a
n
d
u
n
g
,
I
n
d
o
n
e
s
i
a
3
D
e
p
a
r
t
me
n
t
o
f
D
a
t
a
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
U
n
i
v
e
r
s
i
t
a
s
P
e
mb
a
n
g
u
n
a
n
N
a
si
o
n
a
l
V
e
t
e
r
a
n
Jaw
a
T
i
mu
r
,
S
u
r
a
b
a
y
a
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
25
,
2
0
2
5
R
ev
is
ed
Dec
27
,
2
0
2
5
Acc
ep
ted
J
an
30,
2
0
2
6
Th
e
p
re
se
rv
a
ti
o
n
o
f
T
u
ro
ts
Nu
s
a
n
tara
m
a
n
u
sc
rip
ts
writt
e
n
i
n
P
e
g
o
n
sc
ri
p
t
fa
c
e
s
sig
n
ifi
c
a
n
t
c
h
a
ll
e
n
g
e
s
d
u
e
t
o
p
h
y
sic
a
l
d
e
teri
o
ra
ti
o
n
a
n
d
t
h
e
c
o
m
p
lex
it
y
o
f
h
a
n
d
writt
e
n
st
y
les
.
T
h
is
stu
d
y
p
ro
p
o
se
s
a
n
o
v
e
l
d
ig
i
ti
z
a
ti
o
n
a
p
p
ro
a
c
h
b
a
se
d
o
n
ima
g
e
p
r
o
c
e
ss
in
g
t
o
e
x
trac
t
a
n
d
re
c
o
n
stru
c
t
h
a
n
d
writi
n
g
s
tro
k
e
s
b
y
c
o
m
b
in
i
n
g
s
k
e
leto
n
iza
ti
o
n
a
n
d
th
e
trav
e
ll
i
n
g
sa
les
m
a
n
p
ro
b
l
e
m
(TS
P
)
a
lg
o
rit
h
m
.
Th
e
n
o
v
e
lt
y
o
f
t
h
is
re
se
a
rc
h
li
e
s
in
th
e
a
p
p
li
c
a
ti
o
n
o
f
a
m
o
d
ifi
e
d
G
re
e
d
y
TS
P
a
lg
o
rit
h
m
c
a
p
a
b
le
o
f
re
c
o
g
n
izin
g
b
ra
n
c
h
i
n
g
a
n
d
c
y
c
li
c
stru
c
tu
re
s
ty
p
ica
l
o
f
Ara
b
ic
–
P
e
g
o
n
c
h
a
ra
c
ters
,
e
n
a
b
li
n
g
a
c
c
u
ra
te
r
e
c
o
n
str
u
c
ti
o
n
o
f
h
a
n
d
writ
ten
stro
k
e
se
q
u
e
n
c
e
s.
Th
e
p
ro
c
e
ss
in
v
o
l
v
e
s
p
re
p
r
o
c
e
ss
in
g
(
g
ra
y
sc
a
le,
th
re
sh
o
l
d
i
n
g
,
a
n
d
m
o
rp
h
o
l
o
g
ica
l
o
p
e
ra
ti
o
n
s),
sk
e
leto
n
e
x
trac
ti
o
n
u
sin
g
a
th
in
n
in
g
m
e
th
o
d
,
a
n
d
we
ig
h
ted
g
ra
p
h
c
o
n
stru
c
ti
o
n
b
a
se
d
o
n
Eu
c
li
d
e
a
n
d
istan
c
e
b
e
twe
e
n
sk
e
l
e
to
n
p
o
in
ts.
Th
e
p
ro
p
o
se
d
s
y
ste
m
a
c
h
iev
e
d
a
n
a
v
e
ra
g
e
p
re
c
isio
n
o
f
0
.
5
5
2
,
r
e
c
a
ll
o
f
0
.
8
1
5
,
F
1
-
s
c
o
re
o
f
0
.
6
5
7
,
a
n
d
a
c
c
u
ra
c
y
o
f
0
.
8
2
.
Th
e
se
re
su
lt
s
d
e
m
o
n
stra
te
th
e
m
e
th
o
d
’
s
e
ffe
c
ti
v
e
n
e
ss
in
d
e
tec
ti
n
g
a
n
d
re
c
o
n
stru
c
ti
n
g
c
h
a
ra
c
ter
sh
a
p
e
s
fro
m
P
e
g
o
n
m
a
n
u
sc
rip
ts.
P
ra
c
ti
c
a
ll
y
,
th
is
a
p
p
ro
a
c
h
o
ffe
rs
p
o
ten
t
ial
a
p
p
li
c
a
ti
o
n
s
in
t
h
e
a
u
to
m
a
ti
c
d
i
g
it
iza
ti
o
n
,
p
re
se
rv
a
ti
o
n
,
a
n
d
a
n
a
ly
sis
o
f
P
e
g
o
n
sc
rip
t,
c
o
n
tri
b
u
ti
n
g
to
th
e
c
o
n
se
rv
a
ti
o
n
o
f
In
d
o
n
e
sia
’
s Isla
m
ic i
n
tellec
tu
a
l
a
n
d
c
u
lt
u
ra
l
h
e
rit
a
g
e
.
K
ey
w
o
r
d
s
:
Dig
italizatio
n
Gr
ee
d
y
alg
o
r
ith
m
Peg
o
n
s
cr
ip
t
Sk
eleto
n
izatio
n
T
r
av
elin
g
s
alesm
an
p
r
o
b
lem
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
:
L
u
k
m
an
Ha
k
im
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atics E
n
g
in
ee
r
in
g
,
Facu
lty
o
f
E
n
g
in
ee
r
in
g
,
Un
iv
er
s
itas
Yu
d
h
ar
ta
Pas
u
r
u
an
J
l.
Yu
d
h
ar
ta
No
.
7
,
Kem
b
an
g
k
u
n
in
g
,
Sen
g
o
n
a
g
u
n
g
,
Pu
r
w
o
s
ar
i,
Pas
u
r
u
an
,
E
ast J
av
a
6
7
1
6
2
,
I
n
d
o
n
esia
E
m
ail:
lu
k
m
an
@
y
u
d
h
ar
ta.
ac
.
i
d
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
d
ev
el
o
p
m
en
t
o
f
I
s
lam
in
I
n
d
o
n
esia
is
clo
s
ely
lin
k
ed
to
wr
itten
tr
ad
itio
n
s
as
a
m
ed
iu
m
f
o
r
r
ec
o
r
d
in
g
,
p
r
eser
v
in
g
,
an
d
tr
an
s
m
itti
n
g
k
n
o
wled
g
e
.
Ma
n
y
wo
r
k
s
o
f
t
h
e
“
Ulam
a
Nu
s
an
tar
a
”
,
co
llectiv
ely
k
n
o
wn
as
T
u
r
o
ts
,
wer
e
wr
itten
u
s
in
g
Ar
ab
ic
Peg
o
n
s
cr
ip
t
an
ad
ap
te
d
Ar
ab
ic
wr
itin
g
s
y
s
tem
u
s
ed
f
o
r
l
o
ca
l
lan
g
u
ag
es
s
u
ch
as
J
av
an
ese,
Su
n
d
an
ese,
an
d
Ma
d
u
r
ese
[
1
]
.
As
a
f
o
r
m
o
f
I
s
lam
ic
in
tellectu
al
h
er
itag
e,
Peg
o
n
m
an
u
s
cr
ip
ts
co
n
tain
v
alu
a
b
le
h
is
to
r
ical,
r
el
ig
io
u
s
,
an
d
c
u
ltu
r
al
k
n
o
wled
g
e.
Ho
wev
er
,
an
aly
zin
g
h
a
n
d
wr
i
tten
Peg
o
n
m
an
u
s
cr
ip
ts
r
em
ain
s
a
ch
allen
g
in
g
task
.
Ar
ab
ic
-
b
ased
h
an
d
wr
itin
g
is
in
h
er
en
tly
co
m
p
lex
d
u
e
to
its
cu
r
s
iv
e
n
atu
r
e,
co
n
tex
tu
al
letter
s
h
ap
es,
o
v
er
lap
p
in
g
s
tr
o
k
es,
lo
o
p
s
,
an
d
d
iacr
itical
m
ar
k
s
[
2
]
.
T
h
e
s
e
ch
allen
g
es
ar
e
f
u
r
th
er
ex
ac
er
b
ated
b
y
th
e
p
h
y
s
ical
d
eter
i
o
r
atio
n
o
f
Nu
s
an
tar
a
m
an
u
s
cr
ip
ts
ca
u
s
ed
b
y
ag
in
g
,
h
u
m
id
ity
,
a
n
d
in
a
d
eq
u
ate
p
r
eser
v
atio
n
,
wh
ic
h
o
f
ten
r
esu
lts
in
f
ad
ed
in
k
,
b
r
o
k
en
s
tr
o
k
es,
an
d
d
a
m
ag
ed
p
ap
er
[
3
]
,
[
4
]
.
Dig
itizatio
n
h
as
th
er
ef
o
r
e
b
ec
o
m
e
an
ess
en
tial
ef
f
o
r
t
f
o
r
p
r
eser
v
in
g
T
u
r
o
ts
Nu
s
an
tar
a
m
an
u
s
cr
ip
ts
.
B
ey
o
n
d
lo
n
g
-
ter
m
co
n
s
er
v
ati
o
n
,
d
i
g
itizatio
n
im
p
r
o
v
es
ac
ce
s
s
ib
ilit
y
,
s
u
p
p
o
r
ts
s
ch
o
lar
ly
a
n
aly
s
is
,
an
d
en
a
b
les
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
1
-
1x
636
th
e
tr
an
s
m
is
s
io
n
o
f
c
u
ltu
r
al
h
er
itag
e
to
f
u
tu
r
e
g
en
e
r
atio
n
s
[
5
]
,
[
6
]
.
A
cr
itical
s
tag
e
in
th
is
p
r
o
ce
s
s
is
th
e
r
ec
o
n
s
tr
u
ctio
n
a
n
d
an
aly
s
is
o
f
h
an
d
wr
itin
g
s
tr
o
k
es,
wh
ich
a
im
s
to
r
ec
o
v
er
th
e
o
r
ig
in
al
wr
itin
g
s
tr
u
ctu
r
e
an
d
s
tr
o
k
e
s
eq
u
en
ce
f
r
o
m
s
tatic
d
i
g
ital im
ag
es
[
7
]
.
Sev
er
al
s
tu
d
ies
h
av
e
a
d
d
r
ess
ed
h
an
d
wr
itin
g
s
k
eleto
n
izatio
n
an
d
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
.
R
esear
ch
b
y
Kr
y
zh
an
o
v
s
k
ay
a
et
a
l
.
[
8
]
p
r
o
p
o
s
ed
r
ec
o
n
s
tr
u
ctin
g
p
en
s
tr
o
k
es
f
r
o
m
s
k
eleto
n
im
a
g
es
u
s
in
g
h
eu
r
is
tic
g
r
ap
h
tr
av
er
s
al,
ac
h
iev
in
g
p
ath
s
s
im
ilar
to
n
atu
r
al
h
a
n
d
wr
itin
g
b
u
t
s
tr
u
g
g
lin
g
with
b
r
an
ch
e
d
o
r
c
o
m
p
lex
ch
a
r
ac
ter
s
.
Me
an
wh
ile,
Li
et
a
l
.
[
9
]
in
tr
o
d
u
ce
d
in
s
tan
ce
-
awa
r
e
s
k
eleto
n
em
b
ed
d
i
n
g
f
o
r
cu
r
v
ed
te
x
t
d
etec
tio
n
i
n
s
ce
n
e
im
ag
es,
im
p
r
o
v
in
g
d
etec
tio
n
a
cc
u
r
ac
y
b
u
t
n
o
t
f
o
c
u
s
in
g
o
n
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
.
An
o
th
er
s
tu
d
y
b
y
Diaz
et
a
l
.
[
1
0
]
m
o
d
eled
s
k
eleto
n
s
as
g
r
ap
h
s
an
d
ex
tr
ac
te
d
wr
itin
g
p
ath
s
b
ased
o
n
co
n
ti
n
u
ity
an
d
cu
r
v
atu
r
e
c
r
iter
ia,
p
r
o
d
u
cin
g
s
m
o
o
th
s
tr
o
k
e
s
eq
u
en
ce
s
.
Ho
wev
er
,
th
is
ap
p
r
o
ac
h
f
ails
to
h
an
d
le
b
r
a
n
ch
ed
o
r
m
u
lti
-
s
tr
o
k
e
ch
ar
ac
ter
s
co
m
m
o
n
l
y
f
o
u
n
d
i
n
Ar
ab
ic
-
d
e
r
iv
ed
s
cr
ip
ts
s
u
ch
as Peg
o
n
o
r
J
awi.
T
o
ad
d
r
ess
th
ese
lim
itatio
n
s
,
th
is
s
tu
d
y
p
r
o
p
o
s
es
a
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
m
eth
o
d
b
as
ed
o
n
th
e
tr
av
ellin
g
s
alesm
an
p
r
o
b
lem
(
T
SP
)
.
T
SP
i
s
well
s
u
ited
f
o
r
m
o
d
elin
g
co
n
tin
u
o
u
s
p
ath
s
an
d
h
as
th
e
p
o
ten
tial
to
p
r
o
d
u
ce
s
tr
o
k
e
r
ep
r
esen
tatio
n
s
th
at
clo
s
ely
r
e
s
em
b
le
h
u
m
an
wr
itin
g
b
eh
av
io
r
.
Nev
er
th
el
ess
,
s
tan
d
ar
d
T
SP
f
o
r
m
u
latio
n
s
r
estrict
ea
ch
n
o
d
e
to
a
s
in
g
le
v
is
it,
m
ak
in
g
th
e
m
u
n
s
u
itab
le
f
o
r
c
h
ar
ac
ter
s
wi
th
lo
o
p
s
o
r
b
r
an
ch
in
g
s
tr
u
ctu
r
es.
Similar
ly
,
G
r
ee
d
y
T
SP
s
o
lv
er
s
,
wh
ich
p
r
io
r
itize
lo
ca
l
s
h
o
r
test
d
is
tan
ce
s
,
o
f
ten
g
en
er
ate
im
p
lau
s
ib
le
s
tr
o
k
e
o
r
d
e
r
s
f
o
r
c
o
m
p
lex
h
an
d
wr
itten
ch
ar
ac
ter
s
[
1
1
]
.
T
h
er
ef
o
r
e,
th
is
r
esear
ch
i
n
tr
o
d
u
ce
s
a
m
o
d
if
ied
Gr
ee
d
y
T
SP
a
lg
o
r
ith
m
t
h
at
allo
ws
co
n
tr
o
lled
r
ev
is
itin
g
o
f
s
k
eleto
n
n
o
d
es
f
o
r
ch
ar
a
cter
s
with
lo
o
p
in
g
an
d
b
r
a
n
ch
in
g
s
tr
u
ctu
r
es,
wh
ile
m
ain
tain
in
g
s
in
g
le
-
v
is
it
co
n
s
tr
ain
ts
f
o
r
s
im
p
ler
ch
ar
a
cter
s
.
T
h
is
m
o
d
if
icati
o
n
en
ab
les
m
o
r
e
ac
c
u
r
ate
r
ec
o
n
s
tr
u
ct
io
n
o
f
h
an
d
wr
itte
n
Peg
o
n
s
tr
o
k
es a
n
d
b
etter
r
e
p
r
e
s
en
ts
th
e
o
r
ig
in
al
wr
itin
g
s
eq
u
en
ce
,
s
u
p
p
o
r
tin
g
th
e
d
ig
itizatio
n
an
d
p
r
eser
v
atio
n
o
f
Nu
s
an
tar
a
m
a
n
u
s
cr
ip
ts
.
2.
M
E
T
H
O
D
T
h
e
r
esear
ch
p
r
o
ce
s
s
co
n
s
is
ts
o
f
s
ev
er
al
s
tag
es,
in
clu
d
in
g
p
re
-
p
r
o
ce
s
s
in
g
,
s
k
eleto
n
izatio
n
,
tr
av
ellin
g
s
alesm
an
p
r
o
b
lem
,
letter
s
eg
m
en
tatio
n
b
ased
o
n
b
o
u
n
d
in
g
b
o
x
,
an
d
f
r
ee
m
an
c
h
ain
co
d
e
(
FC
C
)
.
T
h
e
f
lo
wch
ar
t
o
f
th
ese
s
tag
es is
s
h
o
wn
in
Fig
u
r
e
1
.
Fig
u
r
e
1
.
Flo
w
o
f
r
esear
ch
2
.
1
.
Da
t
a
s
et
T
h
e
p
r
o
ce
s
s
o
f
co
llectin
g
th
e
d
ataset
f
o
r
th
is
s
tu
d
y
o
r
ig
in
ated
f
r
o
m
an
a
n
cien
t
m
an
u
s
cr
ip
t
wr
itten
u
s
in
g
th
e
Peg
o
n
s
cr
ip
t,
n
am
ely
th
e
“
Kitab
Sy
air
Per
ah
u
”
.
T
h
e
in
itial
s
tag
e
b
eg
an
with
s
ca
n
n
in
g
t
h
e
m
an
u
s
cr
ip
t.
T
h
e
s
ca
n
n
ed
r
esu
lts
wer
e
th
en
p
r
o
c
ess
ed
b
y
m
an
u
all
y
cu
ttin
g
ea
c
h
p
ag
e,
wh
ich
c
o
n
s
is
ted
o
f
1
3
l
in
es
p
er
p
a
g
e.
Af
ter
th
at,
ea
ch
lin
e
was
cu
t
ag
ain
in
to
in
d
iv
id
u
al
wo
r
d
s
to
f
ac
ilit
ate
th
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
.
An
ex
am
p
le
o
f
th
e
im
ag
e
f
r
ag
m
en
ts
to
b
e
a
n
aly
ze
d
in
th
is
s
tu
d
y
ca
n
b
e
s
ee
n
in
Fig
u
r
e
2
.
Fig
u
r
e
2
.
I
m
ag
e
lin
e
c
u
to
u
ts
“
B
o
o
k
Sy
air
P
er
ah
u
”
2
.
2
.
P
re
pro
ce
s
s
ing
Pre
-
p
r
o
ce
s
s
in
g
is
p
er
f
o
r
m
e
d
to
p
r
ep
a
r
e
th
e
d
ata
s
o
th
at
it
b
ec
o
m
es
m
o
r
e
s
tr
u
ctu
r
ed
a
n
d
clea
n
er
,
f
ac
ilit
atin
g
ch
ar
ac
ter
s
h
ap
e
s
eg
m
en
tatio
n
an
d
an
aly
s
is
in
th
e
n
ex
t stag
e.
T
h
e
f
ir
s
t step
is
g
r
ay
s
ca
le
co
n
v
er
s
io
n
.
A
g
r
ay
s
ca
le
im
ag
e
r
ep
r
esen
t
s
ea
ch
p
ix
el
with
a
s
in
g
le
i
n
ten
s
ity
v
alu
e,
w
h
er
e
th
e
r
e
d
,
g
r
ee
n
,
an
d
b
l
u
e
co
m
p
o
n
en
ts
ar
e
eq
u
al,
p
r
o
d
u
ci
n
g
s
h
ad
es
b
etwe
en
b
lack
an
d
wh
ite.
Acc
o
r
d
in
g
to
Z
eg
er
et
a
l
.
[
1
2
]
,
g
r
a
y
s
ca
le
is
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
p
p
lica
tio
n
o
f th
e
tr
a
ve
lin
g
s
a
lesma
n
p
r
o
b
lem
to
o
p
timiz
e
s
ke
leto
n
iz
a
tio
n
a
n
d
s
tr
o
ke
r
ec
o
n
s
tr
u
ctio
n
(
A
lifa
h
)
637
a
d
ig
ital
im
ag
e
r
ep
r
esen
tatio
n
th
at
s
to
r
es
o
n
ly
lu
m
in
an
ce
in
t
en
s
ity
in
f
o
r
m
atio
n
,
ty
p
ically
i
n
8
-
b
it
f
o
r
m
at
with
v
alu
es
r
an
g
in
g
f
r
o
m
0
to
2
5
5
.
As
r
ep
o
r
ted
in
[
1
3
]
,
th
is
p
r
o
ce
s
s
aim
s
to
ex
tr
ac
t
im
p
o
r
tan
t
s
tr
u
ctu
r
es
o
r
o
b
jects
f
r
o
m
im
a
g
es,
p
ar
ticu
lar
ly
u
n
d
er
u
n
ev
e
n
lig
h
tin
g
co
n
d
itio
n
s
,
an
d
is
ca
r
r
ied
o
u
t u
s
in
g
(
1
)
.
=
0
,
2989
×
+
0
,
5870
×
+
0
,
1141
×
(
1
)
T
h
e
v
ar
iab
le
d
en
o
tes
th
e
g
r
ay
s
ca
le
in
ten
s
ity
v
alu
e,
wh
ile
,
,
an
d
r
ep
r
esen
t
th
e
r
ed
,
g
r
ee
n
,
an
d
b
lu
e
co
m
p
o
n
e
n
ts
o
f
th
e
R
GB
co
lo
r
m
o
d
el.
Af
ter
g
r
a
y
s
ca
le
co
n
v
er
s
io
n
,
b
in
ar
izatio
n
is
p
er
f
o
r
m
ed
u
s
in
g
th
e
Ots
u
t
h
r
esh
o
l
d
in
g
m
eth
o
d
,
an
in
ten
s
ity
-
b
ased
s
eg
m
en
tatio
n
tech
n
iq
u
e
f
o
r
g
r
ay
s
ca
le
im
ag
es
[
1
4
]
.
T
h
is
m
eth
o
d
class
if
ies
ea
ch
p
ix
el
as
f
o
r
eg
r
o
u
n
d
o
r
b
ac
k
g
r
o
u
n
d
d
e
p
en
d
in
g
o
n
wh
et
h
er
its
in
ten
s
i
ty
is
ab
o
v
e
o
r
b
elo
w
th
e
th
r
es
h
o
ld
v
alu
e
[
1
5
]
.
As
d
em
o
n
s
tr
ated
in
[
1
6
]
Ots
u
th
r
esh
o
ld
in
g
e
f
f
ec
tiv
ely
s
ep
ar
ates
im
ag
es
in
to
r
eg
io
n
s
with
clea
r
in
ten
s
ity
d
if
f
er
en
ce
s
,
im
p
r
o
v
i
n
g
s
eg
m
en
tatio
n
ac
cu
r
ac
y
with
m
in
i
m
al
co
m
p
u
tatio
n
al
co
s
t.
T
h
e
co
n
v
er
s
io
n
f
r
o
m
g
r
ay
s
ca
le
to
a
b
i
n
ar
y
im
a
g
e
is
ac
h
iev
ed
b
y
d
eter
m
i
n
in
g
a
n
o
p
tim
al
th
r
esh
o
ld
v
alu
e
u
s
in
g
(
2
)
.
(
,
)
=
{
1
0
(
,
)
>
(
,
)
≤
(
2
)
T
h
e
f
u
n
ctio
n
(
,
)
r
ep
r
esen
ts
th
e
p
i
x
el
in
ten
s
ity
at
co
o
r
d
in
ates (
,
)
,
wh
ile
d
en
o
tes th
e
th
r
esh
o
ld
v
alu
e.
Su
b
s
eq
u
en
tly
,
m
o
r
p
h
o
lo
g
ical
o
p
en
in
g
an
d
clo
s
in
g
o
p
er
atio
n
s
ar
e
ap
p
lied
.
Op
en
in
g
,
wh
ich
co
n
s
is
ts
o
f
er
o
s
io
n
f
o
llo
we
d
b
y
d
ilatio
n
u
s
in
g
th
e
s
am
e
s
tr
u
ctu
r
in
g
el
em
en
t,
is
u
s
ed
to
r
em
o
v
e
s
m
all
f
o
r
eg
r
o
u
n
d
n
o
is
e
wh
ile
p
r
eser
v
in
g
th
e
m
ain
o
b
j
ec
t
s
h
ap
e.
C
o
n
v
er
s
ely
,
clo
s
in
g
is
a
d
ilatio
n
f
o
llo
wed
b
y
er
o
s
io
n
o
p
er
atio
n
th
at
f
ills
s
m
all
h
o
les
an
d
co
n
n
ec
ts
f
r
ag
m
en
te
d
o
b
jects,
r
esu
ltin
g
in
s
m
o
o
th
e
r
o
b
ject
co
n
to
u
r
s
[
1
7
]
–
[
1
9
]
.
T
h
e
f
o
r
m
u
las f
o
r
m
o
r
p
h
o
lo
g
ical
o
p
en
in
g
a
n
d
clo
s
in
g
ar
e
p
r
esen
ted
in
(
3
)
an
d
(
4
)
.
∘
=
(
⊖
)
⊕
(
3
)
⨀
=
(
⊕
)
⊖
(
4
)
A
r
ep
r
esen
ts
th
e
in
p
u
t
im
ag
e
,
B
d
en
o
tes
th
e
s
tr
u
ctu
r
al
elem
en
ts
u
s
ed
in
m
o
r
p
h
o
lo
g
ical
p
r
o
ce
s
s
in
g
,
⊖
s
ig
n
if
ies th
e
er
o
s
io
n
o
p
er
atio
n
,
an
d
⊕
in
d
icate
s
th
e
d
ilatio
n
o
p
er
atio
n
.
2
.
3
.
S
k
elet
o
niza
t
io
n
Af
ter
p
r
ep
r
o
ce
s
s
in
g
,
th
e
ch
ar
a
cter
im
ag
e
is
s
k
eleto
n
ized
to
r
ed
u
ce
ea
ch
letter
to
its
m
ed
ial
ax
is
wh
ile
p
r
eser
v
in
g
its
ess
en
tial
m
o
r
p
h
o
lo
g
ical
s
tr
u
ctu
r
e.
Sk
eleto
n
iza
tio
n
ex
tr
ac
ts
a
o
n
e
-
p
i
x
el
-
wid
e
r
ep
r
esen
tatio
n
th
at
m
ain
tain
s
th
e
o
b
ject
’
s
to
p
o
lo
g
y
an
d
g
e
o
m
etr
ic
f
o
r
m
,
p
r
o
v
i
d
in
g
a
s
tr
u
ctu
r
al
b
asis
f
o
r
s
tr
o
k
e
ex
tr
ac
tio
n
f
r
o
m
s
tatic
im
ag
es
[
2
0
]
,
[
2
1
]
.
I
n
t
h
is
s
tu
d
y
,
s
k
eleto
n
izatio
n
is
p
er
f
o
r
m
ed
u
s
in
g
th
e
Z
h
a
n
g
–
Su
e
n
al
g
o
r
ith
m
,
a
p
ar
allel
th
in
n
in
g
m
eth
o
d
th
at
iter
ativ
el
y
r
em
o
v
es
p
ix
els
b
ased
o
n
lo
c
al
co
n
n
ec
tiv
ity
r
u
les
with
in
a
3
×3
n
ei
g
h
b
o
r
h
o
o
d
,
r
esu
ltin
g
in
a
c
o
n
tin
u
o
u
s
an
d
o
n
e
-
p
ix
el
-
wid
e
s
k
eleto
n
[
2
2
]
.
T
h
is
r
ep
r
esen
tatio
n
r
ed
u
ce
s
d
ata
r
ed
u
n
d
an
c
y
a
n
d
s
u
p
p
o
r
ts
ef
f
icien
t su
b
s
eq
u
en
t
p
r
o
ce
s
s
in
g
,
as f
o
r
m
ally
ex
p
r
ess
ed
in
(
5
)
.
ℎ
(
)
=
l
im
→
∞
2
(
1
(
)
)
5)
r
ep
r
esen
ts
th
e
b
in
ar
y
im
ag
e
p
r
o
d
u
ce
d
in
th
e
-
th
iter
atio
n
,
wh
ile
1
an
d
2
ar
e
th
e
two
s
u
b
-
iter
atio
n
s
o
f
th
e
th
in
n
in
g
p
r
o
ce
s
s
in
wh
ich
p
ix
els
ar
e
r
em
o
v
e
d
b
ased
o
n
lo
g
ical
r
u
les.
T
h
e
n
o
t
atio
n
l
im
→
∞
in
d
icate
s
th
at
th
e
p
r
o
ce
s
s
co
n
tin
u
es
u
n
ti
l
co
n
v
e
r
g
en
ce
,
m
ea
n
in
g
n
o
f
u
r
th
er
p
ix
el
r
em
o
v
al
o
cc
u
r
s
an
d
th
e
r
esu
lt
n
o
lo
n
g
er
ch
an
g
es.
2
.
4
.
T
SP
T
h
e
n
ex
t step
a
p
p
lies
th
e
T
SP
,
an
NP
-
h
ar
d
co
m
b
in
ato
r
ial
o
p
t
im
izatio
n
m
eth
o
d
th
at
s
ee
k
s
th
e
s
h
o
r
test
p
ath
v
is
itin
g
ea
ch
p
o
i
n
t
o
n
ce
[
2
3
]
,
[
2
4
]
.
Ho
wev
e
r
,
th
is
f
o
r
m
u
latio
n
is
n
o
t
s
u
itab
le
f
o
r
Ar
a
b
ic
letter
s
eg
m
en
tatio
n
b
ec
au
s
e
m
an
y
ch
a
r
ac
ter
s
co
n
t
ain
lo
o
p
s
an
d
b
r
a
n
ch
in
g
s
tr
u
ct
u
r
es
th
at
p
r
o
d
u
ce
n
o
n
-
lin
ea
r
s
k
eleto
n
p
ath
s
.
T
h
e
s
in
g
le
-
v
is
it c
o
n
s
tr
ain
t lim
its
ac
cu
r
ate
tr
av
er
s
al
o
f
s
u
ch
co
m
p
lex
s
h
ap
es.
C
ar
r
ab
s
et
a
l
.
[
2
5
]
,
in
tr
o
d
u
ce
d
th
e
c
ar
o
u
s
el
Gr
ee
d
y
al
g
o
r
it
h
m
,
wh
ich
allo
ws
lo
ca
l
n
o
d
e
r
ev
is
its
an
d
ac
h
iev
es
n
ea
r
-
o
p
tim
al
s
o
lu
tio
n
s
u
n
d
er
h
ig
h
r
o
u
te
co
m
p
lex
i
ty
with
lo
w
co
m
p
u
tatio
n
al
c
o
s
t.
B
u
ild
in
g
o
n
th
is
id
ea
,
th
is
s
tu
d
y
ad
o
p
ts
a
Gr
ee
d
y
-
with
-
r
ev
is
it
T
SP
ap
p
r
o
ac
h
,
e
n
ab
lin
g
s
elec
tiv
e
r
ev
is
its
f
o
r
letter
s
with
lo
o
p
s
o
r
b
r
an
ch
es,
wh
ile
m
ai
n
tain
in
g
s
i
n
g
le
-
v
is
it tr
av
er
s
al
f
o
r
s
im
p
ler
letter
s
.
C
o
m
p
ar
ativ
e
ex
p
e
r
im
en
ts
with
th
e
n
ea
r
est
n
eig
h
b
o
r
m
eth
o
d
s
h
o
w
th
at
th
e
Gr
ee
d
y
alg
o
r
i
th
m
y
ield
s
m
o
r
e
s
tab
le
tr
ajec
to
r
ies
an
d
b
etter
r
ep
r
esen
ts
n
at
u
r
al
s
tr
o
k
e
f
lo
w,
esp
ec
ially
f
o
r
c
o
m
p
lex
l
etter
s
s
u
ch
as
ح
,
ﻉ
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
1
-
1x
638
a
n
d
ﻡ
.
U
n
l
i
k
e
n
ea
r
est
n
eig
h
b
o
r
,
wh
ich
o
f
ten
ca
u
s
es
ab
r
u
p
t
ju
m
p
s
o
r
p
r
em
atu
r
e
lo
o
p
clo
s
u
r
e
,
th
e
Gr
ee
d
y
m
eth
o
d
co
n
s
id
er
s
g
lo
b
al
co
n
n
ec
tiv
ity
,
r
esu
ltin
g
in
m
o
r
e
co
h
e
r
en
t
p
ath
s
.
T
h
is
m
o
d
if
ied
T
SP
ap
p
r
o
ac
h
ef
f
ec
tiv
ely
r
ec
o
n
s
tr
u
cts
s
tr
o
k
es
in
an
cien
t
Ar
ab
ic
m
an
u
s
cr
ip
ts
b
y
p
r
ese
r
v
in
g
s
tr
o
k
e
c
o
n
tin
u
ity
,
b
r
an
c
h
in
g
b
eh
a
v
io
r
,
a
n
d
s
tr
u
ctu
r
al
co
n
n
ec
ti
v
ity
,
p
r
o
d
u
c
in
g
g
eo
m
etr
ically
ac
cu
r
ate
a
n
d
h
an
d
wr
itin
g
-
co
n
s
is
ten
t
r
esu
lts
.
Ma
th
em
atica
lly
,
T
SP
is
d
ef
in
ed
b
y
(
6
)
.
min
∑
(
,
+
1
)
,
ℎ
=
0
1
=
+
(
6
)
T
h
e
ter
m
(
,
+
1
)
d
en
o
tes
th
e
E
u
clid
e
an
d
is
tan
ce
b
etwe
en
two
co
n
s
ec
u
tiv
e
p
o
in
ts
,
wh
er
e
,
+
1
ar
e
v
is
ited
p
o
in
ts
an
d
n
is
th
e
to
tal
n
u
m
b
e
r
o
f
p
o
in
ts
.
T
h
e
Gr
ee
d
y
r
ev
is
it
alg
o
r
ith
m
wo
r
k
s
b
y
s
elec
tin
g
th
e
l
o
ca
tio
n
clo
s
est
to
th
e
cu
r
r
en
t
p
o
s
itio
n
with
o
u
t
lim
itin
g
lo
ca
tio
n
s
th
at
h
a
v
e
n
o
t
b
ee
n
v
is
ited
.
T
h
is
s
tr
ateg
y
is
r
elev
an
t
f
o
r
ca
s
es
o
f
b
r
an
ch
in
g
,
m
e
r
g
in
g
,
o
r
co
m
p
lex
c
y
clic
s
tr
u
ctu
r
es.
T
h
e
s
tep
s
o
f
th
is
alg
o
r
itm
ar
e
as f
o
llo
ws:
−
Dete
r
m
in
e
th
e
s
tar
tin
g
p
o
in
t
_
o
f
(
u
s
u
ally
s
tar
tin
g
f
r
o
m
th
e
en
d
p
o
in
t
o
f
th
e
s
k
eleto
n
)
.
−
I
n
itializatio
n
T
h
e
p
ath
is
in
itialized
as
[
_
]
,
all
p
o
in
ts
∈
ar
e
m
ar
k
ed
as
u
n
v
is
ite
d
(
[
]
=
0
)
,
th
e
s
tar
tin
g
p
o
in
t
_
is
m
ar
k
ed
as
v
is
ited
(
[
_
]
=
1
)
,
an
d
th
e
v
a
r
iab
le
cu
r
r
en
t
is
s
et
to
_
.
−
I
f
th
er
e
is
s
till
a
p
o
in
t
∈
s
u
ch
th
a
t
[
]
<
(
)
:
a.
Fin
d
th
e
clo
s
est
_
p
o
in
t
f
r
o
m
cu
r
r
en
t
u
s
in
g
E
u
clid
ea
n
d
is
tan
ce
.
T
h
e
E
u
clid
ea
n
d
is
tan
ce
f
o
r
m
u
la
ca
n
b
e
s
ee
n
in
(
7
)
.
(
,
)
=
√
(
−
)
2
+
(
−
)
2
(
7
)
b.
Ad
d
_
to
th
e
p
ath
On
ce
th
e
clo
s
est
p
o
i
n
t
is
f
o
u
n
d
,
it
is
ad
d
ed
to
th
e
p
ath
lis
t.
T
h
is
is
an
im
p
o
r
tan
t
s
tep
in
f
o
r
m
in
g
th
e
s
eq
u
en
ce
o
f
p
ath
s
th
at
r
ep
r
esen
t th
e
s
cr
atch
p
ath
.
c.
[
_
]
+
=
1
A
m
ar
k
er
in
d
icatin
g
th
at
th
e
_
p
o
in
t
h
as
alr
ea
d
y
b
ee
n
v
is
ited
.
T
h
e
v
is
ited
v
alu
e
is
r
ec
o
r
d
ed
s
o
th
at
th
e
a
l
g
o
r
ith
m
ca
n
tr
ac
k
v
i
s
it
v
alu
es
to
d
eter
m
in
e
wh
eth
e
r
a
p
o
in
t
n
ee
d
s
to
b
e
v
is
ited
ag
ain
(
f
o
r
e
x
am
p
le,
in
th
e
ca
s
e
o
f
b
r
a
n
ch
in
g
p
o
i
n
ts
)
o
r
if
a
s
in
g
le
v
is
it is
s
u
f
f
icien
t.
d.
cu
r
r
en
t ←
_
T
h
e
p
o
in
t
th
at
was
ju
s
t
v
is
ited
(
_
)
is
u
p
d
ated
as
th
e
c
u
r
r
e
n
t
p
o
in
t
(
)
.
T
h
is
is
im
p
o
r
tan
t
s
o
th
at
th
e
n
ex
t
p
o
in
t is ca
lcu
l
ated
f
r
o
m
th
e
latest p
o
s
itio
n
in
th
e
n
ex
t iter
atio
n
.
−
R
etu
r
n
p
ath
Descr
ip
tio
n
:
T
h
e
v
is
it
lim
it
(
)
d
ef
in
es
th
e
m
ax
im
u
m
n
u
m
b
er
o
f
allo
wab
le
v
is
its
f
o
r
ea
c
h
s
k
eleto
n
p
o
i
n
t
b
ased
o
n
its
lo
ca
l
s
tr
u
ctu
r
al
ch
ar
ac
ter
is
tics
:
f
o
r
en
d
p
o
in
ts
(
d
e
g
r
ee
=
1
)
,
(
)
=
1
s
in
ce
o
n
ly
a
s
in
g
le
v
is
it
is
r
eq
u
ir
ed
;
f
o
r
lin
ea
r
p
o
in
ts
(
d
eg
r
ee
=
2
)
,
(
)
=
1
as
n
o
r
e
v
is
its
ar
e
n
ec
ess
ar
y
in
s
tr
aig
h
t
s
eg
m
en
t
s
;
an
d
f
o
r
b
r
an
ch
i
n
g
p
o
in
ts
(
d
e
g
r
ee
≥
3
)
,
(
)
>
1
to
allo
w
r
ev
is
itin
g
th
ese
p
o
in
ts
s
o
th
at
all
co
n
n
ec
ted
b
r
a
n
ch
es
ca
n
b
e
ex
p
lo
r
e
d
,
with
th
e
v
alu
e
ty
p
ically
s
et
eq
u
al
to
o
r
s
lig
h
tly
g
r
ea
ter
t
h
an
th
e
p
o
in
t
’
s
d
eg
r
ee
.
Fu
r
th
e
r
m
o
r
e
,
(
)
=
1
ca
n
b
e
ad
ju
s
ted
ac
co
r
d
in
g
to
lo
ca
l
s
k
eleto
n
co
m
p
lex
ity
o
r
d
en
s
ity
,
wh
er
e
r
e
g
io
n
s
with
h
ig
h
co
n
n
ec
tiv
ity
o
r
cu
r
v
ed
s
tr
u
ct
u
r
es
s
u
ch
as
lo
o
p
s
m
ay
r
eq
u
ir
e
h
ig
h
er
v
is
it
lim
its
to
en
s
u
r
e
co
m
p
lete
p
at
h
r
ec
o
n
s
tr
u
ctio
n
,
wh
ile
s
im
p
ler
o
r
s
tr
aig
h
ter
r
e
g
io
n
s
ca
n
u
s
e
lo
wer
lim
its
to
av
o
id
u
n
n
e
ce
s
s
ar
y
r
ev
is
its
an
d
im
p
r
o
v
e
c
o
m
p
u
tatio
n
al
ef
f
icie
n
cy
.
2
.
5
.
L
et
t
er
s
eg
m
ent
a
t
i
o
n ba
s
ed
o
n bo
un
di
ng
bo
x
At
th
is
s
tag
e,
Ar
ab
ic
letter
s
e
g
m
en
tatio
n
is
p
er
f
o
r
m
e
d
u
s
in
g
s
k
eleto
n
izatio
n
an
d
th
e
m
o
d
if
ied
T
SP
p
ath
.
E
ac
h
letter
is
au
to
m
atic
ally
s
ep
ar
ated
f
r
o
m
a
lin
e
o
f
Ar
ab
ic
tex
t
b
y
u
tili
zin
g
s
k
elet
o
n
-
b
ased
s
u
b
-
p
ath
s
g
en
er
ated
in
th
e
p
r
ev
io
u
s
T
SP
m
o
d
if
icatio
n
p
r
o
ce
ss.
Acc
o
r
d
in
g
to
Aan
c
h
al
et
a
l
.
[
2
6
]
,
a
b
o
u
n
d
in
g
b
o
x
is
an
ax
is
-
alig
n
ed
r
ec
tan
g
le
d
ef
in
ed
b
y
x
–
y
co
o
r
d
in
ates,
wid
th
,
a
n
d
h
eig
h
t
th
at
en
cl
o
s
es
h
an
d
wr
itin
g
r
e
g
io
n
s
in
s
ca
n
n
e
d
d
o
cu
m
e
n
t
im
ag
es.
Acc
o
r
d
in
g
ly
,
th
is
s
tu
d
y
ap
p
lies
b
o
u
n
d
in
g
b
o
x
–
b
ased
letter
cr
o
p
p
in
g
as
th
e
in
itial
s
eg
m
en
tatio
n
s
tep
.
Un
l
ik
e
p
r
e
v
io
u
s
ap
p
r
o
ac
h
es,
cr
o
p
p
in
g
is
p
er
f
o
r
m
ed
f
o
r
ea
ch
s
u
b
-
p
ath
p
r
o
d
u
ce
d
b
y
th
e
m
o
d
if
ied
T
SP
alg
o
r
ith
m
,
allo
win
g
s
eg
m
en
tatio
n
to
co
n
s
id
er
b
o
th
s
p
atial
lo
ca
tio
n
a
n
d
s
tr
o
k
e
o
r
d
er
th
at
r
ef
lects
n
atu
r
al
h
a
n
d
wr
i
tin
g
s
tr
u
ctu
r
e
.
T
h
is
ap
p
r
o
ac
h
ef
f
ec
tiv
ely
s
ep
ar
ates in
d
iv
id
u
al
letter
s
f
r
o
m
th
e
b
a
ck
g
r
o
u
n
d
a
n
d
ir
r
elev
an
t e
lem
e
n
ts
s
u
ch
as n
o
is
e.
T
h
e
b
o
u
n
d
i
n
g
b
o
x
f
o
r
ea
ch
s
u
b
-
p
ath
is
co
m
p
u
ted
u
s
in
g
th
e
m
i
n
im
u
m
an
d
m
a
x
im
u
m
x
an
d
y
co
o
r
d
in
ate
v
alu
es,
as d
ef
in
ed
i
n
(
8
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
p
p
lica
tio
n
o
f th
e
tr
a
ve
lin
g
s
a
lesma
n
p
r
o
b
lem
to
o
p
timiz
e
s
ke
leto
n
iz
a
tio
n
a
n
d
s
tr
o
ke
r
ec
o
n
s
tr
u
ctio
n
(
A
lifa
h
)
639
=
min
(
)
−
,
=
ma
x
(
)
+
=
min
(
)
−
,
=
ma
x
(
)
+
(
8
)
T
h
e
s
y
m
b
o
l
is
an
ad
d
itio
n
al
m
ar
g
in
u
s
ed
to
cu
t th
e
ar
ea
s
o
th
at
it is
n
o
t to
o
n
a
r
r
o
w.
T
h
e
im
ag
e
ar
ea
in
th
e
b
o
u
n
d
in
g
b
o
x
will
b
e
cu
t
d
ir
ec
tly
f
r
o
m
th
e
co
m
b
in
ed
_
s
k
eleto
n
,
r
esu
ltin
g
i
n
a
cu
t
im
ag
e
o
f
th
e
letter
s
th
at
co
r
r
esp
o
n
d
s
to
ea
ch
s
u
b
-
p
ath
.
2
.
6
.
F
re
e
m
a
n
cha
in co
de
T
h
e
n
ex
t
s
tep
en
c
o
d
es
th
e
m
o
v
em
en
t
d
ir
ec
tio
n
b
etwe
en
s
k
e
leto
n
p
o
in
ts
u
s
in
g
t
h
e
FC
C
m
eth
o
d
f
o
r
ea
ch
s
u
b
-
p
ath
g
en
er
ate
d
b
y
t
h
e
T
SP
alg
o
r
ith
m
.
FC
C
r
ep
r
esen
ts
o
b
ject
co
n
to
u
r
s
in
b
in
ar
y
o
r
s
k
eleto
n
ized
im
ag
es
u
s
in
g
eig
h
t
d
is
cr
ete
d
ir
ec
tio
n
s
(
0
–
7
)
,
p
r
o
v
id
in
g
a
c
o
n
cise
d
e
s
cr
ip
tio
n
o
f
c
h
ar
ac
ter
s
tr
u
ctu
r
e
[
2
7
]
.
T
h
is
m
eth
o
d
is
wid
ely
u
s
ed
in
im
ag
e
co
m
p
r
ess
io
n
,
p
atter
n
r
ec
o
g
n
itio
n
,
a
n
d
s
h
ap
e
a
n
aly
s
is
b
ec
au
s
e
it
ef
f
ec
tiv
ely
ca
p
tu
r
es
d
ir
ec
tio
n
al
ch
an
g
es
[
2
8
]
.
Feth
i
et
a
l
.
[
2
9
]
a
p
p
lied
FC
C
to
ex
tr
ac
t
co
n
to
u
r
f
ea
tu
r
es
o
f
h
a
n
d
wr
itten
Ar
ab
ic
ch
a
r
ac
ter
s
an
d
p
r
o
p
o
s
ed
a
m
o
d
if
ied
FC
C
th
at
s
im
p
lifie
s
th
e
co
d
e
s
eq
u
en
ce
with
o
u
t
lo
s
in
g
ess
en
tial
s
tr
u
ctu
r
al
in
f
o
r
m
atio
n
,
lead
in
g
t
o
im
p
r
o
v
e
d
r
ec
o
g
n
itio
n
ac
cu
r
a
cy
.
I
n
th
is
s
tu
d
y
,
FC
C
is
u
s
ed
to
tr
an
s
f
o
r
m
s
k
eleto
n
p
a
th
s
in
to
n
u
m
er
ical
m
o
v
em
en
t
s
eq
u
e
n
ce
s
th
at
ca
n
b
e
s
p
atially
an
aly
ze
d
[
2
9
]
,
[
3
0
]
,
s
u
ch
as
id
en
tify
in
g
wr
itin
g
d
ir
ec
tio
n
,
lo
o
p
s
,
an
d
ch
ar
ac
ter
is
tic
tu
r
n
s
in
Ar
ab
ic
l
etter
s
.
T
h
e
eig
h
t
-
d
ir
e
ctio
n
al
s
y
s
tem
o
f
FC
C
is
i
llu
s
tr
ated
b
elo
w.
3
2
1
\
|
/
4
-
-
-
-
-
0
/ |
\
5
6
7
T
h
e
n
u
m
b
er
s
0
to
7
a
b
o
v
e
r
ep
r
esen
t
8
d
ir
ec
tio
n
s
o
f
m
o
v
em
en
t
r
elativ
e
to
a
p
o
i
n
t.
I
n
th
e
FC
C
r
ep
r
esen
tatio
n
,
ea
c
h
n
u
m
b
e
r
i
n
d
icate
s
th
e
d
ir
ec
tio
n
o
f
p
ix
el
m
o
v
em
e
n
t,
wh
er
e
0
i
n
d
icate
s
r
ig
h
t,
1
in
d
icate
s
u
p
p
er
r
ig
h
t,
2
i
n
d
icate
s
u
p
,
3
i
n
d
icate
s
u
p
p
er
lef
t,
4
in
d
icate
s
lef
t,
5
in
d
icate
s
lo
wer
lef
t,
6
i
n
d
icate
s
d
o
wn
,
an
d
7
in
d
icate
s
lo
wer
r
ig
h
t
[
3
1
]
.
2
.
7
.
E
v
a
lua
t
i
o
n
m
et
rics
E
v
alu
atio
n
is
co
n
d
u
cted
q
u
an
titativ
ely
u
s
in
g
s
tan
d
ar
d
m
u
lt
i
-
class
clas
s
if
icatio
n
m
etr
ics,
in
clu
d
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all
,
an
d
F1
-
s
co
r
e
[
3
2
]
,
wh
ich
ar
e
c
o
m
p
u
ted
b
ased
o
n
tr
u
e
p
o
s
itiv
e
(
T
P),
f
alse
p
o
s
itiv
e
(
FP
)
,
tr
u
e
n
eg
ativ
e
(
T
N)
,
a
n
d
f
alse
n
eg
ativ
e
(
FN)
v
alu
es
[
3
3
]
.
Acc
u
r
ac
y
m
ea
s
u
r
es
o
v
er
all
class
if
icatio
n
co
r
r
ec
tn
ess
,
p
r
ec
is
io
n
ev
alu
at
es
th
e
r
eliab
ilit
y
o
f
p
o
s
itiv
e
p
r
ed
ictio
n
s
,
R
ec
all
m
ea
s
u
r
es
th
e
ab
ilit
y
to
d
etec
t
all
ac
tu
al
p
o
s
itiv
e
s
am
p
les,
a
n
d
th
e
F1
-
s
co
r
e
p
r
o
v
id
es
a
b
alan
ce
d
m
ea
s
u
r
e
b
y
co
m
b
in
in
g
p
r
ec
is
io
n
an
d
r
ec
all
,
p
ar
ticu
lar
ly
u
n
d
e
r
class
im
b
al
an
ce
co
n
d
itio
n
s
[
3
4
]
,
[
3
5
]
.
I
n
ad
d
itio
n
,
i
n
ter
s
ec
tio
n
o
v
er
u
n
i
o
n
(
I
o
U)
is
u
s
ed
to
ass
es
s
th
e
o
v
er
lap
b
etwe
en
p
r
e
d
icted
an
d
g
r
o
u
n
d
t
r
u
th
(
GT
)
r
eg
io
n
s
in
s
eg
m
en
tatio
n
task
s
.
B
ey
o
n
d
q
u
an
titativ
e
m
et
r
ics,
a
q
u
alitativ
e
n
o
d
e
-
b
ased
an
al
y
s
is
is
p
er
f
o
r
m
e
d
o
n
th
e
s
k
el
eto
n
izatio
n
r
esu
lts
to
ev
alu
ate
s
p
atial
s
tr
u
ctu
r
e
an
d
in
ter
-
n
o
d
e
co
n
n
ec
ti
v
ity
,
d
em
o
n
s
tr
atin
g
th
e
s
y
s
tem
’
s
ef
f
ec
tiv
en
ess
in
s
ep
ar
atin
g
letter
s
h
ap
es
a
n
d
d
i
ac
r
itics
.
T
o
v
er
if
y
th
at
th
e
clas
s
if
icatio
n
r
esu
lts
ar
e
n
o
t
d
u
e
t
o
r
an
d
o
m
ch
an
ce
,
a
B
in
o
m
ial
T
est
is
ap
p
lied
.
T
h
is
n
o
n
-
p
a
r
am
etr
ic
s
tatis
tical
te
s
t
d
eter
m
in
es
wh
eth
er
th
e
o
b
s
er
v
ed
class
if
icatio
n
ac
cu
r
ac
y
is
s
ig
n
if
ican
tly
h
ig
h
e
r
th
an
a
r
an
d
o
m
b
aselin
e
(
ty
p
ically
5
0
%),
th
er
eb
y
co
n
f
ir
m
in
g
th
e
v
alid
ity
o
f
th
e
s
y
s
tem
’
s
p
er
f
o
r
m
an
ce
.
T
h
e
b
i
n
o
m
ial
test
f
o
r
m
u
latio
n
is
p
r
es
en
ted
in
(
1
4
)
.
(
=
)
=
(
)
(
1
−
)
−
(
1
4
)
d
en
o
tes
th
e
to
tal
n
u
m
b
er
o
f
tr
ials
,
s
u
ch
as
th
e
n
u
m
b
er
o
f
s
am
p
les
o
r
p
r
e
d
ictio
n
s
.
r
ep
r
ese
n
ts
th
e
n
u
m
b
er
o
f
s
u
cc
ess
es,
f
o
r
ex
am
p
le
th
e
n
u
m
b
er
o
f
co
r
r
ec
t
p
r
ed
ictio
n
s
.
is
th
e
p
r
o
b
ab
ilit
y
o
f
s
u
cc
e
s
s
u
n
d
er
th
e
n
u
ll
h
y
p
o
th
esis
(
co
m
m
o
n
ly
s
et
to
0
.
5
f
o
r
a
r
an
d
o
m
b
aselin
e)
.
A
n
d
(
)
is
th
e
b
in
o
m
ial
co
ef
f
icien
t,
co
m
p
u
ted
as
!
!
(
−
)
!
.
Fo
r
a
o
n
e
-
tailed
test
,
th
e
p
-
v
al
u
e
is
ca
lcu
lated
u
s
in
g
(
1
5
)
.
=
∑
(
)
(
1
−
)
−
=
(
1
5
)
T
h
is
test
is
u
s
ed
to
d
eter
m
in
e
wh
eth
er
th
e
ac
tu
al
p
r
o
p
o
r
tio
n
o
f
s
u
cc
ess
es
is
s
ig
n
if
ican
tly
g
r
ea
ter
(
o
r
s
m
aller
,
d
ep
en
d
i
n
g
o
n
th
e
d
ir
ec
tio
n
o
f
th
e
test
)
th
an
th
e
p
r
o
b
a
b
ilit
y
a
s
s
u
m
ed
u
n
d
er
th
e
n
u
ll h
y
p
o
th
e
s
is
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
1
-
1x
640
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
P
re
-
pro
ce
s
s
ing
r
esu
lt
s
T
h
e
i
n
i
t
i
a
l
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
s
te
p
i
s
g
r
a
y
s
c
a
l
e
c
o
n
v
e
r
s
i
o
n
,
w
h
ic
h
r
e
m
o
v
e
s
c
o
l
o
r
i
n
f
o
r
m
a
t
i
o
n
a
n
d
r
e
t
a
i
n
s
o
n
l
y
p
i
x
e
l
i
n
t
e
n
s
i
t
y
t
o
f
a
c
i
li
t
ate
s
u
b
s
e
q
u
e
n
t
p
r
o
c
e
s
s
i
n
g
.
A
s
s
h
o
w
n
i
n
F
i
g
u
r
e
3
(
a
)
,
t
h
e
g
r
a
y
s
c
a
l
e
i
m
a
g
e
p
r
e
s
e
r
v
es
t
h
e
b
a
s
i
c
s
h
a
p
e
s
o
f
t
h
e
A
r
a
b
i
c
l
e
tt
e
r
s
i
n
t
h
e
S
y
ai
r
Pe
r
a
h
u
m
an
u
s
c
r
i
p
t
d
es
p
i
t
e
v
a
r
i
at
i
o
n
s
i
n
in
k
a
n
d
b
a
c
k
g
r
o
u
n
d
c
o
l
o
r
,
a
l
l
o
w
i
n
g
c
l
e
a
r
e
r
s
e
p
a
r
ati
o
n
b
e
t
w
e
e
n
t
e
x
t
a
n
d
b
a
c
k
g
r
o
u
n
d
b
a
s
e
d
o
n
i
n
t
e
n
s
i
t
y
d
i
f
f
e
r
e
n
c
e
s
.
F
o
l
l
o
w
i
n
g
t
h
i
s
s
t
e
p
,
t
h
r
es
h
o
l
d
i
n
g
i
s
a
p
p
l
i
e
d
to
c
o
n
v
e
r
t
t
h
e
g
r
a
y
s
c
a
l
e
i
m
a
g
e
i
n
t
o
a
b
i
n
a
r
y
(
b
l
a
ck
-
a
n
d
-
w
h
i
t
e
)
i
m
a
g
e
,
e
n
a
b
l
i
n
g
e
f
f
e
c
t
i
v
e
s
e
p
a
r
a
ti
o
n
o
f
l
e
t
t
e
r
o
b
j
e
c
t
s
f
r
o
m
t
h
e
b
a
c
k
g
r
o
u
n
d
.
T
h
e
O
t
s
u
t
h
r
e
s
h
o
l
d
i
n
g
r
es
u
l
ts
d
i
s
p
l
a
y
t
h
e
l
e
tt
e
r
s
i
n
w
h
i
t
e
(
v
a
l
u
e
2
5
5
)
ag
ain
s
t
a
b
l
ac
k
b
ac
k
g
r
o
u
n
d
(
v
alu
e
0
)
,
m
a
k
in
g
th
e
Ar
ab
ic
s
cr
ip
t
clea
r
ly
d
is
tin
g
u
is
h
ab
le
an
d
h
ig
h
lig
h
tin
g
th
e
cu
r
v
ed
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es o
f
Peg
o
n
ch
a
r
ac
ter
s
,
as sh
o
wn
in
Fig
u
r
e
3
(
b
)
.
(
a)
(
b
)
Fig
u
r
e
3
.
Gr
a
y
s
ca
le
an
d
th
r
esh
o
ld
in
g
r
esu
lts
:
(
a)
g
r
ay
s
ca
le
i
m
ag
e
an
d
(
b
)
b
in
ar
y
im
ag
e
a
f
ter
th
r
esh
o
ld
in
g
B
ef
o
r
e
s
k
eleto
n
izatio
n
,
an
ad
v
an
ce
d
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e
is
ap
p
lied
u
s
in
g
m
o
r
p
h
o
lo
g
ical
o
p
en
in
g
a
n
d
c
lo
s
in
g
to
im
p
r
o
v
e
letter
s
tr
u
ctu
r
e
q
u
ality
.
T
h
ese
o
p
e
r
atio
n
s
clea
n
an
d
c
o
m
p
lete
t
h
e
b
i
n
ar
y
im
a
g
e,
t
h
er
eb
y
en
h
an
cin
g
s
k
eleto
n
ex
tr
ac
tio
n
.
Op
en
in
g
r
em
o
v
es
s
m
all
n
o
is
e
in
tr
o
d
u
ce
d
d
u
r
in
g
d
ig
itizatio
n
,
wh
ile
c
lo
s
in
g
f
ill
s
s
m
all
g
ap
s
to
p
r
eser
v
e
s
tr
o
k
e
co
n
tin
u
ity
.
I
n
th
is
s
tu
d
y
,
a
2
×
2
s
q
u
a
r
e
s
t
r
u
ctu
r
in
g
elem
en
t
(
n
p
.
o
n
es((
2
,
2
)
,
n
p
.
u
in
t8
)
)
is
u
s
ed
,
as
it
e
f
f
ec
tiv
ely
r
em
o
v
es
n
o
is
e
wh
ile
p
r
eser
v
in
g
f
in
e
s
tr
o
k
e
d
etails
an
d
p
r
e
v
en
tin
g
ad
jace
n
t
letter
s
f
r
o
m
m
er
g
in
g
,
wh
ich
is
s
u
itab
le
f
o
r
Peg
o
n
m
a
n
u
s
cr
ip
t
s
with
th
in
s
tr
o
k
es
an
d
n
a
r
r
o
w
s
p
ac
in
g
.
T
h
e
p
ar
a
m
eter
s
elec
tio
n
is
d
eter
m
in
e
d
em
p
ir
ically
th
r
o
u
g
h
ex
p
er
im
en
ts
o
n
m
u
ltip
le
s
am
p
les.
Fo
r
m
an
u
s
cr
ip
ts
with
h
ig
h
e
r
n
o
is
e
lev
els,
lar
g
er
s
tr
u
ctu
r
in
g
elem
en
ts
(
e.
g
.
,
3
×3
o
r
5
×5
)
m
a
y
b
e
u
s
ed
,
p
r
o
v
id
e
d
th
at
s
tr
o
k
e
d
etails
ar
e
p
r
eser
v
ed
.
T
h
e
r
esu
lts
o
f
m
o
r
p
h
o
lo
g
ical
o
p
en
in
g
an
d
m
o
r
p
h
o
lo
g
ical
clo
s
in
g
a
r
e
s
h
o
w
n
in
Fig
u
r
es 4
(
a)
an
d
(
b
)
.
(
a)
(
b
)
Fig
u
r
e
4
.
Mo
r
p
h
o
lo
g
ical
o
p
en
i
n
g
an
d
m
o
r
p
h
o
l
o
g
ical
clo
s
in
g
r
esu
lts
:
(
a)
im
ag
e
af
ter
m
o
r
p
h
o
lo
g
ical
o
p
e
n
in
g
an
d
(
b
)
im
ag
e
a
f
ter
m
o
r
p
h
o
lo
g
i c
lo
s
in
g
3
.
2
.
S
k
elet
o
niza
t
io
n
re
s
ults
T
h
e
s
k
eleto
n
ex
tr
ac
tio
n
p
r
o
ce
s
s
in
th
is
s
tu
d
y
u
s
es
th
e
m
o
r
p
h
o
lo
g
ical
th
in
n
in
g
m
et
h
o
d
with
t
h
e
Z
h
a
n
g
-
Su
en
alg
o
r
ith
m
,
wh
ich
is
im
p
lem
en
ted
th
r
o
u
g
h
th
e
s
k
elet
o
n
ize(
)
f
u
n
ctio
n
f
r
o
m
th
e
s
k
im
ag
e.
m
o
r
p
h
o
lo
g
y
lib
r
ar
y
.
T
h
is
m
eth
o
d
wo
r
k
s
iter
ativ
ely
b
y
s
elec
tiv
ely
r
e
m
o
v
in
g
e
d
g
e
p
ix
els
with
o
u
t
alter
in
g
th
e
m
ain
to
p
o
lo
g
ical
s
tr
u
ctu
r
e
o
f
th
e
letter
s
,
r
esu
ltin
g
in
a
o
n
e
-
p
i
x
el
s
k
eleto
n
th
at
r
ep
r
esen
ts
th
e
m
ed
i
al
ax
is
o
f
ea
c
h
letter
s
tr
o
k
e.
T
h
e
s
k
eleto
n
izatio
n
r
es
u
lts
ca
n
b
e
s
ee
n
in
Fig
u
r
e
5
.
Fig
u
r
e
5
.
Sk
eleto
n
izatio
n
r
esu
l
ts
Fig
u
r
e
5
s
h
o
ws
th
at
th
e
s
k
eleto
n
(
f
r
am
e
lin
es)
s
u
cc
ess
f
u
lly
f
o
l
lo
ws
th
e
m
ain
co
n
to
u
r
s
o
f
Ar
a
b
ic
letter
s
,
in
clu
d
in
g
cu
r
v
es a
n
d
co
n
n
ec
tio
n
s
b
etwe
en
letter
s
.
T
h
e
s
tr
o
k
es o
f
th
e
letter
s
h
av
e
b
ee
n
s
u
cc
ess
f
u
lly
r
ed
u
ce
d
to
th
e
m
ed
ial
ax
is
,
b
u
t
s
till
r
etain
th
e
s
h
ap
e
an
d
d
ir
ec
tio
n
o
f
th
e
o
r
ig
in
al
s
tr
o
k
es.
T
h
is
s
h
o
ws
th
at
th
e
m
o
r
p
h
o
l
o
g
ical
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
p
p
lica
tio
n
o
f th
e
tr
a
ve
lin
g
s
a
lesma
n
p
r
o
b
lem
to
o
p
timiz
e
s
ke
leto
n
iz
a
tio
n
a
n
d
s
tr
o
ke
r
ec
o
n
s
tr
u
ctio
n
(
A
lifa
h
)
641
th
in
n
in
g
m
et
h
o
d
u
s
in
g
th
e
Z
h
an
g
-
Su
en
alg
o
r
ith
m
is
ef
f
ec
ti
v
e
in
ex
tr
ac
tin
g
im
p
o
r
tan
t
f
ea
tu
r
es
f
r
o
m
co
m
p
lex
letter
s
h
ap
es.
3
.
3
.
T
SP
re
s
ults
I
n
th
is
s
tu
d
y
,
th
e
T
SP
alg
o
r
ith
m
is
ap
p
lied
to
s
k
eleto
n
ized
P
eg
o
n
w
o
r
d
im
ag
es
to
g
en
e
r
ate
s
eq
u
en
tial
s
tr
o
k
e
p
ath
s
th
at
clo
s
ely
r
ec
o
n
s
tr
u
ct
th
e
o
r
ig
in
al
h
a
n
d
wr
iti
n
g
.
T
h
e
Gr
ee
d
y
r
e
v
is
it
alg
o
r
ith
m
en
ab
les
th
e
T
SP
p
ath
to
f
o
llo
w
co
m
p
lex
letter
s
tr
u
ctu
r
es,
in
clu
d
in
g
lo
o
p
s
an
d
b
r
an
ch
es,
wh
ile
p
r
eser
v
i
n
g
co
n
n
ec
tiv
ity
b
etwe
en
s
tr
o
k
e
s
eg
m
en
ts
.
E
ac
h
r
esu
ltin
g
p
ath
is
v
is
u
alize
d
as
a
d
ir
e
cted
g
r
ap
h
to
a
n
aly
ze
its
co
r
r
e
s
p
o
n
d
en
ce
with
t
h
e
o
r
i
g
in
al
letter
s
h
ap
es.
Fo
r
ev
alu
atio
n
,
o
n
e
wo
r
d
c
o
n
t
ain
in
g
m
u
ltip
le
l
o
o
p
in
g
letter
s
was
s
elec
ted
f
r
o
m
ea
ch
s
eg
m
en
ted
lin
e
to
f
ac
ilit
ate
clea
r
o
b
s
er
v
atio
n
an
d
an
aly
s
is
o
f
th
e
g
en
e
r
ated
p
ath
s
.
T
h
e
d
etailed
T
SP
p
ath
r
esu
lts
ar
e
s
h
o
wn
in
Fig
u
r
e
6
(
a)
,
wh
ile
h
ig
h
lig
h
ted
lo
o
p
h
a
n
d
lin
g
an
d
r
ev
is
itin
g
b
eh
av
io
r
a
r
e
illu
s
tr
ated
in
Fig
u
r
e
6
(
b
)
.
(
a)
(
b
)
Fig
u
r
e
6
.
T
SP
r
esu
lts
:
(
a)
o
v
er
all
g
en
er
ated
T
SP
p
ath
a
n
d
(
b
)
d
etail
v
iew
o
f
lo
o
p
tr
a
v
er
s
al
an
d
r
ev
is
itin
g
b
eh
av
io
r
I
t
ca
n
b
e
s
ee
n
th
at
th
e
T
SP
p
ath
g
en
e
r
ated
h
as
th
e
ab
ilit
y
t
o
f
o
llo
w
th
e
en
tire
s
tr
u
ctu
r
e
o
f
th
e
letter
with
o
u
t
b
r
ea
k
in
g
th
e
p
ath
,
ev
en
wh
en
th
e
s
tr
u
ctu
r
e
is
cir
cu
l
ar
o
r
b
r
a
n
ch
ed
.
Po
in
ts
v
is
ited
m
o
r
e
th
an
o
n
ce
ar
e
m
ar
k
ed
in
p
u
r
p
le,
in
d
icatin
g
th
at
r
ev
is
iti
n
g
o
cc
u
r
s
in
th
o
s
e
ar
ea
s
d
u
e
to
th
e
co
m
p
lex
ity
o
f
th
e
letter
s
tr
u
ctu
r
e.
T
h
is
m
eth
o
d
is
ab
le
to
cr
ea
te
p
ath
s
th
at
m
o
r
e
clo
s
ely
r
esem
b
l
e
th
e
h
an
d
-
d
r
awn
s
tr
o
k
es
th
at
f
o
r
m
letter
s
.
Un
lik
e
co
n
v
en
tio
n
al
T
SP
,
wh
ich
o
n
ly
allo
ws o
n
e
v
is
it p
er
p
o
in
t.
3
.
4
.
L
et
t
er
s
eg
m
ent
a
t
i
o
n r
esu
lt
s
Af
ter
o
b
tain
in
g
t
h
e
T
SP
p
ath
s
an
d
s
u
b
-
p
at
h
s
,
letter
s
eg
m
en
ta
tio
n
is
p
er
f
o
r
m
e
d
u
s
in
g
b
o
u
n
d
in
g
b
o
x
es
f
o
r
ea
ch
s
u
b
-
p
at
h
.
T
h
e
p
r
o
ce
s
s
s
tar
ts
b
y
ex
tr
ac
tin
g
s
k
eleto
n
p
o
in
ts
an
d
g
en
er
atin
g
a
r
ig
h
t
-
to
-
l
ef
t
T
SP
p
ath
b
ased
o
n
E
u
clid
ea
n
d
is
tan
ce
.
T
h
e
T
SP
p
ath
is
th
en
d
iv
id
ed
in
to
s
u
b
-
p
ath
s
u
s
in
g
th
e
s
p
lit_
ts
p
_
p
ath
_
b
y
_
d
is
tan
ce
(
)
f
u
n
ctio
n
,
wh
er
e
a
n
ew
s
u
b
-
p
a
th
is
cr
ea
ted
if
th
e
d
is
tan
ce
b
etwe
en
co
n
s
ec
u
tiv
e
p
o
in
ts
ex
ce
ed
s
a
m
ax
im
u
m
th
r
esh
o
ld
(
e.
g
.
,
3
p
ix
els).
Fo
r
ea
ch
s
u
b
-
p
ath
,
a
b
o
u
n
d
i
n
g
b
o
x
is
co
m
p
u
te
d
u
s
in
g
th
e
m
in
im
u
m
an
d
m
a
x
im
u
m
x
an
d
y
c
o
o
r
d
in
ates,
en
ab
lin
g
in
d
iv
id
u
al
letter
s
to
b
e
c
r
o
p
p
ed
in
to
s
ep
ar
ate
im
ag
e
s
eg
m
e
n
ts
.
T
h
ese
s
eg
m
en
ted
letter
im
ag
es
ar
e
s
av
e
d
in
a
d
esig
n
ated
d
ir
ec
to
r
y
,
with
a
“
la
b
el.
tx
t
”
f
ile
p
r
ep
ar
e
d
f
o
r
s
u
b
s
eq
u
en
t
la
b
elin
g
.
An
ex
am
p
le
o
f
th
e
b
o
u
n
d
in
g
b
o
x
–
b
ased
letter
s
eg
m
en
tatio
n
r
esu
lts
is
s
h
o
wn
in
Fig
u
r
e
s
7
(
a)
to
(
e)
.
(
a)
(
b
)
(
c)
(
d
)
(
e)
Fig
u
r
e
7
.
L
etter
s
eg
m
en
tatio
n
r
esu
lts
:
(
a)
f
ir
s
t seg
m
en
ted
letter
,
(
b
)
s
ec
o
n
d
s
eg
m
e
n
ted
lette
r
,
(
c)
th
i
r
d
s
eg
m
en
ted
letter
,
(
d
)
f
o
u
r
th
s
e
g
m
en
ted
letter
,
a
n
d
(
e
)
f
if
th
s
eg
m
en
ted
letter
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
1
-
1x
642
Fig
u
r
e
7
s
h
o
ws
th
at
t
h
is
m
eth
o
d
is
ef
f
ec
tiv
e
in
f
o
llo
win
g
th
e
co
m
p
lex
s
h
a
p
es
o
f
Ar
a
b
ic
s
cr
ip
t
th
r
o
u
g
h
s
k
eleto
n
p
ath
s
an
d
s
ep
ar
atin
g
letter
s
b
ased
o
n
th
e
b
o
u
n
d
in
g
b
o
x
o
f
ea
ch
T
SP
s
u
b
-
p
ath
.
T
h
e
test
r
esu
lts
ar
e
d
is
p
lay
ed
in
th
e
f
o
r
m
o
f
letter
s
eg
m
en
t
v
is
u
ali
za
tio
n
s
an
d
ex
tr
ac
ted
im
ag
e
f
iles
.
T
h
ese
tw
o
f
iles
s
er
v
e
as
th
e
b
asis
f
o
r
th
e
letter
r
ec
o
g
n
itio
n
s
tag
e
in
th
e
n
ex
t
p
r
o
ce
s
s
.
3
.
5
.
F
CC
re
s
ults
I
n
th
is
s
tu
d
y
,
F
C
C
is
u
s
e
d
t
o
an
a
l
y
z
e
r
e
c
o
n
s
t
r
u
c
te
d
l
e
t
te
r
s
t
r
o
k
e
s
b
y
e
x
a
m
i
n
i
n
g
t
h
e
s
h
a
p
e
a
n
d
d
i
r
e
c
t
i
o
n
o
f
p
e
n
m
o
v
e
m
e
n
t
.
A
f
t
e
r
s
k
e
l
et
o
n
i
z
a
t
i
o
n
a
n
d
T
S
P
-
b
as
e
d
p
a
th
t
r
a
c
k
i
n
g
,
m
u
l
t
i
p
l
e
s
u
b
-
p
a
t
h
s
r
e
p
r
e
s
e
n
t
i
n
g
l
e
t
te
r
s
k
e
l
e
t
o
n
s
a
r
e
o
b
t
a
i
n
e
d
.
E
ac
h
s
u
b
-
p
a
t
h
i
s
t
h
e
n
co
n
v
er
ted
in
to
a
FC
C
th
r
o
u
g
h
th
e
f
o
llo
win
g
s
tep
s
.
First,
ea
ch
s
u
b
-
p
ath
is
p
r
o
ce
s
s
ed
s
eq
u
en
tially
as
a
s
er
ies
o
f
(
x
,
y
)
s
k
eleto
n
c
o
o
r
d
in
ates.
Ne
x
t,
d
ir
ec
tio
n
co
d
es a
r
e
ass
ig
n
ed
f
o
r
ea
ch
p
a
ir
o
f
co
n
s
ec
u
tiv
e
p
o
i
n
ts
u
s
in
g
th
e
d
ir
ec
tio
n
_
co
d
e(
)
f
u
n
ctio
n
wh
en
th
e
p
o
in
ts
ar
e
d
ir
ec
t
8
-
co
n
n
ec
ted
n
eig
h
b
o
r
s
.
I
f
co
n
s
e
cu
tiv
e
p
o
in
ts
ar
e
n
o
t
ad
jace
n
t,
s
tep
wis
e
in
ter
p
o
latio
n
is
ap
p
lied
b
y
in
cr
em
en
tally
m
o
v
in
g
to
war
d
th
e
tar
g
et
p
o
in
t
b
ased
o
n
th
e
s
ig
n
o
f
co
o
r
d
i
n
ate
d
if
f
er
e
n
ce
s
,
co
n
v
er
tin
g
ea
ch
s
tep
in
to
a
Fre
em
an
d
ir
ec
tio
n
c
o
d
e.
A
m
ax
im
u
m
s
tep
lim
it is
im
p
o
s
ed
to
av
o
i
d
in
f
i
n
ite
lo
o
p
s
in
in
v
alid
p
ath
s
.
All
g
en
er
ated
d
ir
ec
ti
o
n
co
d
es
ar
e
s
to
r
ed
as
an
a
r
r
ay
r
ep
r
ese
n
tin
g
th
e
FC
C
f
o
r
ea
ch
s
u
b
-
p
ath
.
An
ex
am
p
le
o
f
FC
C
ex
tr
ac
tio
n
f
o
r
th
e
wo
r
d
“
Per
ah
u
”
is
s
h
o
wn
in
Fig
u
r
e
8
.
I
n
Fig
u
r
e
8
,
t
h
e
s
m
all
wh
ite
n
u
m
b
er
s
o
n
ea
ch
p
ath
in
d
icate
th
e
d
i
r
ec
tio
n
o
f
m
o
v
e
m
en
t
b
ased
o
n
t
h
e
FC
C
.
Me
an
wh
ile,
th
e
b
lu
e
an
d
y
ello
w
d
o
ts
in
d
icate
th
e
s
tar
tin
g
an
d
en
d
in
g
p
o
in
ts
o
f
ea
c
h
s
u
b
-
p
ath
.
Fig
u
r
e
8
.
FC
C
g
en
er
ated
f
r
o
m
th
e
wo
r
d
,
s
tar
t f
r
o
m
t
h
e
to
p
r
i
g
h
t to
th
e
lef
t
3
.
6
.
E
v
a
lua
t
i
o
n
m
et
rics r
esu
lt
s
Af
ter
co
m
p
letin
g
all
p
r
e
-
p
r
o
ce
s
s
in
g
,
s
eg
m
en
tatio
n
,
an
d
T
SP
-
b
ased
p
r
o
ce
s
s
in
g
s
tag
es,
an
ev
alu
atio
n
is
co
n
d
u
cte
d
to
ass
es
s
th
e
s
y
s
te
m
’
s
ab
ilit
y
to
d
etec
t
an
d
s
eg
m
en
t
letter
s
.
T
h
e
ev
alu
atio
n
co
m
p
ar
es
th
e
ex
p
ec
ted
n
u
m
b
er
o
f
ch
ar
ac
ter
s
GT
wit
h
th
e
n
u
m
b
e
r
o
f
ch
ar
ac
ter
s
d
e
tecte
d
b
y
th
e
s
y
s
tem
(
p
r
e
d
ictio
n
)
.
GT
v
alu
es
ar
e
o
b
tain
ed
m
an
u
ally
f
r
o
m
r
ef
e
r
e
n
ce
d
ata,
wh
ile
p
r
ed
ictio
n
s
ar
e
d
er
iv
e
d
f
r
o
m
th
e
s
eg
m
en
tatio
n
r
esu
lts
p
r
o
d
u
ce
d
b
y
th
e
alg
o
r
ith
m
.
I
n
th
is
s
tu
d
y
,
GT
d
ata
ar
e
ta
k
en
f
r
o
m
th
e
f
i
r
s
t
p
ag
e
o
f
th
e
Sy
air
Per
ah
u
m
a
n
u
s
cr
ip
t
b
y
r
an
d
o
m
ly
s
elec
tin
g
th
r
ee
lin
es
(
r
asm
)
an
d
ev
alu
atin
g
t
h
em
m
an
u
ally
.
Prio
r
to
lin
e
-
lev
el
e
v
alu
atio
n
,
a
p
r
elim
in
ar
y
w
o
r
d
-
lev
el
ex
p
er
im
e
n
t
is
p
er
f
o
r
m
e
d
u
s
in
g
th
e
wo
r
d
“Per
ah
u
”,
as
d
is
cu
s
s
ed
in
th
e
p
r
ev
io
u
s
s
ec
tio
n
.
A
d
etaile
d
b
r
ea
k
d
o
wn
o
f
th
e
ev
alu
atio
n
r
esu
lts
f
o
r
“Per
ah
u
”
is
p
r
esen
te
d
in
T
ab
le
1
.
T
h
en
,
f
o
r
a
d
etailed
ex
p
lan
atio
n
o
f
ea
ch
E
v
alu
atio
n
v
alu
e
p
er
r
o
w,
p
lease
r
ef
er
to
T
ab
le
2
.
As
a
co
m
p
lem
en
t
to
th
e
v
is
u
al
an
aly
s
is
,
T
ab
le
3
d
is
p
lay
s
a
q
u
an
titativ
e
ev
alu
atio
n
o
f
th
e
s
eg
m
en
tatio
n
r
esu
lts
o
n
ea
c
h
lin
e
o
f
t
h
e
m
a
n
u
s
cr
ip
t,
wh
ich
in
clu
d
es
th
e
n
u
m
b
er
o
f
G
T
c
h
ar
ac
ter
s
,
d
etec
ted
ch
a
r
ac
ter
s
(
DT
)
,
co
r
r
ec
tly
s
eg
m
e
n
ted
c
h
ar
ac
ter
s
T
P,
FP
,
FN,
as
well
as
p
er
f
o
r
m
an
ce
m
etr
ics
s
u
ch
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e
an
d
I
o
U.
T
ab
le
1
.
E
v
alu
atio
n
r
esu
lts
o
f
th
e
wo
r
d
“Per
ah
u
”
F
i
g
u
r
e
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
A
c
c
u
r
a
c
y
I
o
U
0
.
6
0
0
0
.
7
5
0
0
.
6
6
7
0
.
7
5
0
.
7
0
0
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
p
p
lica
tio
n
o
f th
e
tr
a
ve
lin
g
s
a
lesma
n
p
r
o
b
lem
to
o
p
timiz
e
s
ke
leto
n
iz
a
tio
n
a
n
d
s
tr
o
ke
r
ec
o
n
s
tr
u
ctio
n
(
A
lifa
h
)
643
T
ab
le
2.
E
v
alu
atio
n
r
esu
lts
p
er
r
o
w
O
r
i
g
i
n
a
l
i
ma
g
e
P
r
o
c
e
ss
e
d
r
e
s
u
l
t
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
A
c
c
u
r
a
c
y
I
o
U
0
.
5
9
5
0
.
8
1
5
0
.
6
8
7
0
.
8
2
0
.
5
2
4
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
A
c
c
u
r
a
c
y
I
o
U
0
.
5
5
0
0
.
7
8
6
0
.
6
4
7
0
.
7
9
0
.
5
4
3
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
A
c
c
u
r
a
c
y
I
o
U
0
.
5
1
2
0
.
8
4
6
0
.
6
3
8
0
.
8
5
0
.
5
2
0
T
ab
le
3
.
Deta
iled
ev
al
u
atio
n
r
esu
lts
o
f
s
eg
m
en
tatio
n
s
y
s
tem
I
mag
e
GT
DT
TP
FP
FN
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
S
c
o
r
e
A
c
c
u
r
a
c
y
I
o
U
P
0
1
-
1
27
37
22
15
5
0
.
5
9
5
0
.
8
1
5
0
.
6
8
7
0
.
8
2
0
.
5
2
4
P
0
1
-
9
28
40
22
18
6
0
.
5
5
0
0
.
7
8
6
0
.
6
4
7
0
.
7
9
0
.
5
4
3
P
0
1
-
10
26
43
22
21
4
0
.
5
1
2
0
.
8
4
6
0
.
6
3
8
0
.
8
5
0
.
5
2
0
A
v
e
r
a
g
e
81
1
2
0
66
54
15
0
.
5
5
2
0
.
8
1
5
0
.
6
5
7
0
.
8
2
0
.
5
2
9
As
a
r
esu
lt,
th
e
e
v
a
l
u
a
t
i
o
n
m
e
t
r
i
c
s
s
h
o
w
t
h
at
t
h
e
s
y
s
t
e
m
is
q
u
i
te
g
o
o
d
a
t
r
e
c
o
g
n
i
z
i
n
g
t
h
e
n
u
m
b
e
r
o
f
l
e
t
t
e
r
s
b
a
s
e
d
o
n
t
h
e
n
u
m
b
e
r
d
e
t
e
c
t
e
d
.
H
o
w
e
v
e
r
,
b
e
c
a
u
s
e
t
h
e
e
v
al
u
a
t
i
o
n
o
n
l
y
c
o
n
s
i
d
e
r
s
t
h
e
s
u
i
t
ab
i
l
i
t
y
o
f
t
h
e
n
u
m
b
e
r
w
i
t
h
o
u
t
v
e
r
i
f
y
i
n
g
t
h
e
c
o
r
r
e
c
t
n
es
s
o
f
t
h
e
i
d
e
n
ti
t
y
o
r
p
o
s
i
t
i
o
n
o
f
th
e
l
e
t
t
e
r
s
,
t
h
e
r
e
s
u
lt
s
d
o
n
o
t
f
u
l
ly
r
e
f
l
e
c
t
t
h
e
o
v
e
r
a
l
l
a
c
c
u
r
a
c
y
o
f
t
h
e
s
y
s
t
e
m
.
T
h
i
s
f
o
c
u
s
w
as
c
h
o
s
e
n
b
e
c
a
u
s
e
t
h
e
r
ese
a
r
c
h
p
h
a
s
e
i
s
s
t
i
ll
f
o
c
u
s
e
d
o
n
r
e
c
o
n
s
t
r
u
c
t
i
n
g
b
as
ic
s
t
r
o
k
e
f
o
r
m
s
t
h
r
o
u
g
h
s
k
e
l
et
o
n
iz
a
t
i
o
n
an
d
p
ath
o
p
tim
izatio
n
,
r
ath
er
th
an
o
n
f
u
ll c
h
ar
ac
ter
r
ec
o
g
n
itio
n
.
T
o
ev
alu
ate
th
e
tem
p
o
r
al
ac
cu
r
ac
y
o
f
th
e
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
s
eq
u
en
ce
,
th
is
s
tu
d
y
ad
d
s
n
o
d
es
at
ea
ch
p
o
in
t o
f
t
h
e
T
SP
tr
ajec
to
r
y
.
E
a
ch
n
o
d
e
r
ep
r
esen
ts
th
e
alg
o
r
ith
m
’
s
tr
av
er
s
al
s
eq
u
en
ce
s
o
th
at
th
e
r
ec
o
n
s
tr
u
ctio
n
p
r
o
ce
s
s
ca
n
co
n
s
id
er
b
o
th
th
e
s
p
atial
d
im
en
s
io
n
an
d
th
e
tem
p
o
r
al
s
eq
u
en
ce
o
f
th
e
s
tr
o
k
e
s
.
T
h
e
r
esu
lts
o
f
th
e
n
o
d
e
ad
d
itio
n
s
tr
ateg
y
in
th
e
T
SP
tr
ajec
to
r
y
a
r
e
s
h
o
wn
in
F
ig
u
r
e
9
(
a)
,
wh
ich
p
r
esen
ts
th
e
o
v
e
r
all
wo
r
d
-
lev
el
tr
ajec
to
r
y
,
an
d
Fig
u
r
e
9
(
b
)
,
wh
ich
p
r
o
v
id
es a
d
etailed
v
iew
o
f
th
e
r
ec
o
n
s
tr
u
cte
d
s
tr
o
k
e
s
eq
u
en
ce
.
(
a)
(
b
)
Fig
u
r
e
9
.
I
m
p
lem
en
tatio
n
o
f
a
d
d
in
g
n
o
d
es o
n
th
e
T
SP
p
ath
:
(
a)
g
lo
b
al
tr
ajec
to
r
y
a
f
ter
n
o
d
e
in
s
er
tio
n
an
d
(
b
)
m
a
g
n
if
ied
v
iew
illu
s
tr
atin
g
th
e
r
ef
in
e
d
an
d
co
n
tin
u
o
u
s
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
T
h
e
ad
d
itio
n
o
f
s
eq
u
en
tial
n
o
d
e
n
u
m
b
er
in
g
alo
n
g
th
e
T
SP
p
at
h
is
in
ten
d
ed
to
r
ec
o
r
d
th
e
tr
a
v
er
s
al
o
r
d
er
(
tem
p
o
r
al
o
r
d
er
)
o
f
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
.
E
ac
h
n
o
d
e
is
ass
ig
n
ed
a
g
lo
b
al
s
eq
u
e
n
ce
n
u
m
b
er
a
cr
o
s
s
all
s
u
b
-
p
ath
s
,
allo
win
g
ev
er
y
s
k
eleto
n
p
o
in
t
to
b
e
r
ep
r
e
s
en
ted
n
o
t
o
n
ly
s
p
a
tially
b
u
t
also
tem
p
o
r
ally
b
ased
o
n
its
v
is
it
o
r
d
er
.
T
h
is
en
ab
les ev
alu
atio
n
f
r
o
m
t
wo
p
er
s
p
ec
tiv
es:
−
Sp
atial
a
cc
u
r
ac
y
,
w
h
ich
ass
ess
es
wh
eth
er
th
e
T
SP
p
ath
f
u
lly
co
v
er
s
th
e
ch
ar
ac
ter
s
h
ap
e
wit
h
o
u
t
m
is
s
in
g
s
tr
o
k
e
co
m
p
o
n
en
ts
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
1
-
1x
644
−
T
em
p
o
r
al
a
cc
u
r
ac
y
,
w
h
ich
e
v
a
lu
ates w
h
eth
er
t
h
e
r
ec
o
n
s
tr
u
ct
ed
s
tr
o
k
e
o
r
d
er
m
atch
es
n
atu
r
a
l h
an
d
w
r
itin
g
p
atter
n
s
,
s
u
ch
as
c
o
r
r
ec
t
tr
a
v
er
s
al
d
ir
ec
tio
n
o
r
lo
o
p
o
r
d
er
in
g
.
T
h
u
s
,
n
o
d
e
n
u
m
b
e
r
in
g
f
u
n
ctio
n
s
as
a
tim
e
-
s
tep
in
d
icato
r
in
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
.
T
o
f
u
r
t
h
er
v
alid
ate
th
e
ev
alu
a
tio
n
r
esu
lts
,
a
B
in
o
m
ial
T
est
is
ap
p
lied
to
co
n
f
ir
m
th
at
th
e
o
b
s
er
v
ed
ac
cu
r
ac
y
is
n
o
t
d
u
e
to
c
h
an
ce
a
n
d
ex
ce
ed
s
th
e
r
an
d
o
m
b
aselin
e
(
5
0
%).
T
h
e
h
y
p
o
th
eses
ar
e
d
ef
in
ed
as:
h
ip
o
tesi
s
n
o
l
(
₀
)
;
a
cc
u
r
ac
y
≤
0
.
5
0
a
n
d
h
i
p
o
tesi
s
alter
n
atif
(
₁
)
;
a
cc
u
r
ac
y
>
0
.
5
0
.
Fo
r
th
e
wo
r
d
“Per
ah
u
”
,
th
e
o
b
tain
e
d
p
-
v
alu
e
is
0
.
0
6
2
5
0
.
Alth
o
u
g
h
s
lig
h
tly
ab
o
v
e
th
e
0
.
0
5
s
ig
n
if
ican
ce
th
r
esh
o
ld
,
th
is
r
esu
lt
in
d
icate
s
th
at
th
e
m
o
d
el
’
s
ac
cu
r
ac
y
is
m
ar
g
in
all
y
b
u
t
co
n
s
is
ten
tly
h
ig
h
er
th
an
r
an
d
o
m
g
u
ess
in
g
,
s
u
p
p
o
r
tin
g
th
e
claim
th
at
th
e
p
r
o
p
o
s
ed
m
eth
o
d
d
em
o
n
s
tr
ate
s
m
ea
n
in
g
f
u
l
g
en
er
aliza
tio
n
ca
p
ab
ilit
y
.
3
.
7
.
Str
o
k
e
re
c
o
ns
t
ruct
io
n e
v
a
lua
t
io
n
T
o
co
m
p
lem
en
t
th
e
c
h
ar
ac
ter
c
o
u
n
t
-
b
ased
ev
alu
atio
n
,
th
is
s
tu
d
y
ad
d
ed
a
q
u
an
titativ
e
e
v
alu
a
tio
n
o
f
t
h
e
ac
cu
r
ac
y
o
f
th
e
s
tr
o
k
e
p
ath
s
g
en
er
ated
b
y
th
e
T
SP
alg
o
r
ith
m
.
T
h
e
e
v
alu
atio
n
was
co
n
d
u
cted
u
s
in
g
two
m
ai
n
m
etr
ics:
a
v
er
ag
e
p
ath
d
e
v
iatio
n
(
APD)
,
wh
ich
m
ea
s
u
r
es
th
e
s
p
atial
ac
cu
r
ac
y
o
f
th
e
s
tr
o
k
e
p
ath
s
,
an
d
tem
p
o
r
al
s
eq
u
en
ce
ac
c
u
r
ac
y
(
T
SA)
,
wh
ich
m
ea
s
u
r
es
t
h
e
co
r
r
esp
o
n
d
e
n
ce
o
f
th
e
s
tr
o
k
e
s
eq
u
en
ce
to
th
e
o
r
i
g
in
al
wr
itin
g
p
atter
n
.
T
h
e
r
esu
lts
o
f
th
is
q
u
a
n
titativ
e
ev
alu
atio
n
ar
e
p
r
esen
ted
in
T
ab
le
4
.
T
ab
le
4
.
Qu
a
n
titativ
e
ac
cu
r
ac
y
ass
es
s
m
en
t o
f
r
ec
o
n
s
tr
u
cted
s
tr
o
k
e
p
ath
s
I
mag
e
A
P
D
(
p
x
)
TSA
(
%)
P
0
1
-
1
1
.
9
2
8
6
,
3
%
P
0
1
-
9
2
.
1
5
8
2
,
5
%
P
0
1
-
10
2
.
0
3
8
8
.
1
%
A
v
e
r
a
g
e
2
.
0
3
8
5
.
6
%
T
h
e
ev
alu
atio
n
r
esu
lts
in
T
ab
le
4
s
h
o
w
th
at
th
e
T
SP
r
ec
o
n
s
tr
u
ctio
n
p
ath
h
as
lo
w
s
p
atial
d
ev
iatio
n
,
with
an
av
er
a
g
e
APD
v
alu
e
o
f
2
.
0
3
p
ix
els,
in
d
icatin
g
th
at
th
e
s
tr
o
k
e
s
h
ap
e
is
clo
s
e
to
th
e
g
r
o
u
n
d
-
tr
u
th
p
atter
n
.
I
n
ad
d
itio
n
,
th
e
T
SA
v
alu
e
o
f
8
5
.
6
%
in
d
icate
s
th
at
m
o
s
t
o
f
t
h
e
tr
av
er
s
al
s
eq
u
en
ce
o
f
th
e
T
SP
p
ath
f
o
llo
ws
th
e
ac
tu
al
letter
wr
itin
g
p
atter
n
.
T
h
u
s
,
th
e
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
i
s
n
o
t
o
n
ly
v
is
u
ally
co
r
r
ec
t,
b
u
t
also
q
u
an
titativ
ely
m
ea
s
u
r
ab
le
in
ter
m
s
o
f
b
o
t
h
s
h
ap
e
an
d
wr
itin
g
s
eq
u
e
n
ce
.
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
s
u
cc
ess
f
u
lly
d
em
o
n
s
tr
ates
a
s
k
eleto
n
-
b
ased
im
ag
e
p
r
o
ce
s
s
in
g
f
r
am
ewo
r
k
f
o
r
Peg
o
n
s
cr
ip
t
th
at
in
teg
r
ates
th
in
n
i
n
g
-
b
ased
s
k
eleto
n
izatio
n
,
a
m
o
d
if
ied
T
SP
,
an
d
Gr
ee
d
y
r
e
v
is
it
o
p
tim
izatio
n
to
r
e
co
n
s
tr
u
ct
h
an
d
wr
itten
s
tr
o
k
e
p
ath
s
.
T
h
e
p
r
o
p
o
s
ed
a
p
p
r
o
ac
h
ef
f
ec
t
iv
ely
h
an
d
les
co
m
p
lex
Ar
ab
ic
letter
s
tr
u
ctu
r
es
in
v
o
lv
in
g
lo
o
p
s
an
d
b
r
an
ch
es,
en
ab
lin
g
ac
cu
r
ate
s
tr
o
k
e
tr
a
v
er
s
al
an
d
letter
s
eg
m
en
tatio
n
u
s
in
g
E
u
clid
ea
n
-
d
is
tan
ce
–
b
ased
s
u
b
-
p
at
h
s
ep
a
r
atio
n
an
d
b
o
u
n
d
in
g
b
o
x
cr
o
p
p
in
g
.
E
x
p
er
im
en
tal
r
esu
lts
s
h
o
w
th
at
th
e
s
y
s
tem
ac
h
iev
es
an
av
er
a
g
e
p
r
ec
is
io
n
o
f
0
.
5
5
2
,
r
ec
all
o
f
0
.
8
1
5
,
F1
-
s
co
r
e
o
f
0
.
6
5
7
,
an
d
ac
cu
r
ac
y
o
f
8
2
%,
in
d
icatin
g
r
eliab
le
p
er
f
o
r
m
an
ce
i
n
d
etec
ti
n
g
an
d
s
eg
m
en
tin
g
Peg
o
n
lette
r
s
b
ased
o
n
c
h
ar
ac
ter
co
u
n
t.
T
o
f
u
r
t
h
er
v
alid
ate
th
e
ef
f
ec
tiv
en
ess
o
f
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
,
a
q
u
an
titativ
e
ev
al
u
atio
n
u
s
in
g
APD
an
d
T
SA
was
co
n
d
u
cted
.
T
h
e
r
esu
l
ts
s
h
o
w
a
lo
w
av
er
ag
e
APD
o
f
2
.
0
3
p
ix
els
an
d
a
h
ig
h
T
SA
o
f
8
5
.
6
%,
co
n
f
ir
m
in
g
th
at
th
e
r
ec
o
n
s
tr
u
cted
p
ath
s
cl
o
s
ely
f
o
llo
w
b
o
t
h
th
e
s
p
atial
s
h
ap
e
an
d
tem
p
o
r
al
wr
itin
g
s
eq
u
en
ce
o
f
th
e
o
r
ig
in
al
h
an
d
wr
itin
g
.
T
h
ese
f
in
d
in
g
s
d
e
m
o
n
s
tr
ate
th
at
th
e
p
r
o
p
o
s
ed
T
S
P
-
b
ased
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
i
s
n
o
t
o
n
ly
v
is
u
ally
co
h
er
en
t b
u
t a
ls
o
q
u
an
titativ
ely
ac
cu
r
ate.
Alth
o
u
g
h
th
e
cu
r
r
e
n
t e
v
alu
atio
n
f
o
cu
s
es o
n
s
eg
m
en
tatio
n
an
d
s
tr
o
k
e
r
ec
o
n
s
tr
u
ctio
n
r
ath
er
th
an
f
u
ll
ch
ar
ac
ter
r
ec
o
g
n
itio
n
,
th
e
r
esu
lts
p
r
o
v
id
e
a
s
tr
o
n
g
f
o
u
n
d
ati
o
n
f
o
r
f
u
tu
r
e
w
o
r
k
th
at
in
co
r
p
o
r
ates
c
h
ar
ac
ter
cla
s
s
if
icatio
n
an
d
s
p
atial
c
o
n
tex
t
an
aly
s
is
to
s
u
p
p
o
r
t
co
m
p
lete
Peg
o
n
m
an
u
s
cr
ip
t
tr
an
s
cr
ip
tio
n
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
ar
e
g
r
atef
u
l
f
o
r
th
e
f
u
n
d
i
n
g
s
u
p
p
o
r
t
f
r
o
m
Un
iv
er
s
itas
Yu
d
h
ar
ta
Pas
u
r
u
an
an
d
R
esear
ch
C
en
ter
f
o
r
Ar
tific
ial
I
n
tellig
e
n
ce
an
d
C
y
b
er
s
ec
u
r
ity
,
Nati
o
n
al
R
esear
ch
an
d
I
n
n
o
v
atio
n
Ag
en
cy
(
B
R
I
N)
,
B
an
d
u
n
g
wh
ic
h
h
as c
o
n
t
r
ib
u
te
d
to
th
e
im
p
lem
en
tatio
n
o
f
th
i
s
r
esear
ch
.
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
lla
b
o
r
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.