I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
6
,
Decem
b
er
20
25
,
p
p
.
5
8
9
4
~
5
9
0
3
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
6
.
pp
5
8
9
4
-
5
9
0
3
5894
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Ro
bo
tic
pro
duc
t
-
ba
sed ma
nipula
ti
o
n in simula
ted
e
nv
iro
nment
J
ua
n
Ca
m
ilo
G
ua
chet
a
-
Alba
1
,
Anny
Astr
id E
s
pitia
-
Cu
bil
lo
s
2
,
Ro
bin
s
o
n J
im
enez
-
M
o
r
eno
3
1
M
e
c
h
a
n
i
c
a
l
En
g
i
n
e
e
r
i
n
g
P
r
o
g
r
a
m
,
E
n
g
i
n
e
e
r
i
n
g
F
a
c
u
l
t
y
,
U
n
i
v
e
r
si
d
a
d
e
F
e
d
e
r
a
l
d
o
R
i
o
d
e
Ja
n
e
i
r
o
,
R
i
o
d
e
Ja
n
e
i
r
o
,
B
r
a
z
i
l
2
I
n
d
u
st
r
i
a
l
E
n
g
i
n
e
e
r
i
n
g
P
r
o
g
r
a
m
,
En
g
i
n
e
e
r
i
n
g
F
a
c
u
l
t
y
,
U
n
i
v
e
r
si
d
a
d
M
i
l
i
t
a
r
N
u
e
v
a
G
r
a
n
a
d
a
,
B
o
g
o
t
á
,
C
o
l
o
mb
i
a
3
M
e
c
h
a
t
r
o
n
i
c
E
n
g
i
n
e
e
r
i
n
g
P
r
o
g
r
a
m
,
E
n
g
i
n
e
e
r
i
n
g
F
a
c
u
l
t
y
,
U
n
i
v
e
r
si
d
a
d
M
i
l
i
t
a
r
N
u
e
v
a
G
r
a
n
a
d
a
,
B
o
g
o
t
á
,
C
o
l
o
m
b
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
1
8
,
2
0
2
4
R
ev
is
ed
J
u
l 1
4
,
2
0
2
5
Acc
ep
ted
Sep
1
4
,
2
0
2
5
Be
fo
re
d
e
p
l
o
y
i
n
g
a
lg
o
rit
h
m
s
i
n
i
n
d
u
str
ial
se
tt
in
g
s,
i
t
is
e
ss
e
n
ti
a
l
t
o
v
a
li
d
a
te
th
e
m
in
v
irt
u
a
l
e
n
v
iro
n
m
e
n
ts
t
o
a
n
ti
c
ip
a
te
re
a
l
-
wo
rl
d
p
e
rfo
rm
a
n
c
e
,
id
e
n
ti
fy
p
o
ten
ti
a
l
li
m
it
a
ti
o
n
s,
a
n
d
g
u
i
d
e
n
e
c
e
ss
a
ry
o
p
ti
m
iza
ti
o
n
s.
T
h
is
st
u
d
y
p
re
se
n
ts
th
e
d
e
v
e
lo
p
m
e
n
t
a
n
d
i
n
teg
ra
ti
o
n
o
f
a
rti
ficia
l
in
telli
g
e
n
c
e
a
lg
o
r
it
h
m
s
fo
r
d
e
tec
ti
n
g
lab
e
ls
a
n
d
c
o
n
tain
e
r
fo
rm
a
ts
o
f
c
lea
n
in
g
p
ro
d
u
c
ts
u
sin
g
c
o
m
p
u
ter
v
isio
n
,
e
n
a
b
li
n
g
ro
b
o
ti
c
m
a
n
ip
u
latio
n
v
ia
a
UR5
a
rm
.
Lab
e
l
i
d
e
n
ti
f
ica
ti
o
n
is
p
e
rfo
rm
e
d
u
sin
g
th
e
sp
e
e
d
e
d
-
u
p
r
o
b
u
st
fe
a
tu
re
s
(S
URF)
a
lg
o
rit
h
m
,
e
n
su
rin
g
ro
b
u
stn
e
ss
t
o
sc
a
le
a
n
d
o
rien
ta
ti
o
n
c
h
a
n
g
e
s.
F
o
r
c
o
n
tain
e
r
re
c
o
g
n
i
ti
o
n
,
m
u
lt
ip
le
m
e
th
o
d
s
we
re
e
x
p
l
o
re
d
:
e
d
g
e
d
e
tec
ti
o
n
u
si
n
g
S
o
b
e
l
a
n
d
C
a
n
n
y
fil
ters
,
Ho
p
field
n
e
two
r
k
s
trai
n
e
d
o
n
f
il
tere
d
ima
g
e
s,
2
D
c
r
o
ss
-
c
o
rre
latio
n
,
a
n
d
fin
a
ll
y
,
a
y
o
u
o
n
ly
l
o
o
k
o
n
c
e
(
YO
LO
)
d
e
e
p
lea
rn
in
g
m
o
d
e
l.
Am
o
n
g
th
e
se
,
th
e
c
u
sto
m
-
trai
n
e
d
YO
LO
d
e
tec
to
r
p
r
o
v
i
d
e
d
t
h
e
h
i
g
h
e
st
a
c
c
u
ra
c
y
.
F
o
r
ro
b
o
ti
c
c
o
n
tro
l,
sm
o
o
th
j
o
in
t
traj
e
c
to
ries
we
re
c
o
m
p
u
ted
u
si
n
g
p
o
ly
n
o
m
ial
in
terp
o
latio
n
,
a
ll
o
win
g
th
e
UR5
r
o
b
o
t
t
o
e
x
e
c
u
te
p
ick
-
a
n
d
-
p
lac
e
o
p
e
ra
ti
o
n
s.
Th
e
e
n
t
ire
p
ro
c
e
ss
wa
s
v
a
li
d
a
ted
i
n
th
e
Co
p
p
e
li
a
S
im
sim
u
latio
n
e
n
v
iro
n
m
e
n
t,
wh
e
re
t
h
e
ro
b
o
t
su
c
c
e
ss
fu
ll
y
i
d
e
n
t
ifi
e
d
,
c
las
sified
,
a
n
d
m
a
n
ip
u
late
d
p
ro
d
u
c
ts,
d
e
m
o
n
stra
ti
n
g
t
h
e
fe
a
sib
il
it
y
o
f
th
e
p
r
o
p
o
se
d
p
ip
e
li
n
e
fo
r
fu
t
u
re
a
p
p
li
c
a
ti
o
n
s in
se
m
i
-
stru
c
tu
re
d
in
d
u
strial
c
o
n
tex
ts
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
in
tellig
en
ce
C
o
m
p
u
ter
v
is
io
n
Dee
p
lear
n
in
g
I
n
d
u
s
tr
y
5
.
0
Ma
ch
in
e
lear
n
in
g
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
An
n
y
Ast
r
id
E
s
p
itia Cu
b
illo
s
I
n
d
u
s
tr
ial
E
n
g
in
ee
r
in
g
,
Facu
lty
o
f
E
n
g
in
ee
r
i
n
g
,
Un
i
v
er
s
id
ad
Militar
Nu
ev
a
Gr
an
ad
a
C
ar
r
er
a
1
1,
101
-
8
0
,
B
o
g
o
tá,
C
o
lo
m
b
ia
E
m
ail:
an
n
y
.
esp
itia@
u
n
im
ilit
ar
.
ed
u
.
c
o
1.
I
NT
RO
D
UCT
I
O
N
Vir
tu
al
en
v
ir
o
n
m
en
ts
s
tan
d
o
u
t
as
a
m
ea
n
s
o
f
r
esear
c
h
d
e
v
elo
p
m
en
t
t
h
at
allo
ws
th
e
v
a
lid
atio
n
o
f
alg
o
r
ith
m
s
in
m
u
ltip
le
en
v
ir
o
n
m
en
ts
an
d
p
u
r
p
o
s
es
as
v
ar
ied
as
au
to
n
o
m
o
u
s
n
av
ig
atio
n
[
1
]
,
o
b
ject
r
ec
o
g
n
itio
n
an
d
lo
ca
lizatio
n
[
2
]
,
s
ec
u
r
ity
s
y
s
tem
s
in
tr
an
s
p
o
r
tatio
n
[
3
]
,
co
llab
o
r
ativ
e
p
r
o
d
u
ctio
n
en
v
ir
o
n
m
en
ts
[
4
]
,
d
r
o
n
e
o
p
er
atio
n
[
5
]
,
am
o
n
g
o
th
e
r
s
.
W
ith
in
th
ese
v
ir
tu
al
en
v
ir
o
n
m
en
ts
it
is
f
ea
s
ib
le
to
im
p
lem
e
n
t
m
ac
h
in
e
lear
n
in
g
an
d
co
m
p
u
ter
v
is
io
n
alg
o
r
ith
m
s
[
6
]
,
o
p
e
r
atin
g
with
ca
m
er
a
s
,
d
ee
p
lear
n
in
g
al
g
o
r
ith
m
s
an
d
im
ag
e
p
r
o
ce
s
s
in
g
f
o
r
p
r
o
d
u
ct
o
r
ien
ted
a
p
p
licatio
n
s
[
7
]
,
wh
e
r
e
th
is
ty
p
e
o
f
a
p
p
licatio
n
u
n
d
e
r
p
in
I
n
d
u
s
tr
y
5
.
0
[
8
]
.
A
d
v
an
ce
s
in
I
n
d
u
s
tr
y
4
.
0
[
9
]
a
n
d
5
.
0
[
1
0
]
allo
w
o
r
ien
tin
g
ef
f
o
r
ts
in
t
h
e
im
p
r
o
v
em
e
n
t
o
f
th
e
p
r
o
d
u
ctio
n
c
h
ain
,
t
h
u
s
in
v
o
lv
in
g
th
e
u
s
e
o
f
r
o
b
o
ts
in
m
an
u
f
ac
tu
r
in
g
o
p
er
atio
n
s
[
1
1
]
,
co
llab
o
r
ativ
e
wo
r
k
with
r
o
b
o
ts
[
1
2
]
an
d
o
b
ject
m
an
ip
u
latio
n
[
1
3
]
,
[
1
4
]
an
d
th
eir
in
teg
r
atio
n
with
d
if
f
e
r
en
t i
n
ter
ac
tio
n
alg
o
r
it
h
m
s
with
af
o
r
em
en
tio
n
ed
v
ir
tu
al
en
v
ir
o
n
m
en
t.
Desp
ite
th
e
g
r
o
win
g
ad
o
p
tio
n
o
f
d
ig
ital
twin
s
an
d
in
d
u
s
tr
ial
s
im
u
latio
n
,
th
er
e
is
a
lack
o
f
i
n
teg
r
ated
v
ir
tu
al
f
r
am
ew
o
r
k
s
th
at
c
o
m
b
in
e
p
r
o
d
u
ct
d
etec
tio
n
,
cl
ass
if
icatio
n
,
an
d
r
o
b
o
tic
m
a
n
ip
u
latio
n
in
s
em
i
-
s
tr
u
ctu
r
ed
en
v
ir
o
n
m
e
n
ts
[
1
5
]
.
T
h
is
is
an
im
p
o
r
tan
t
asp
ec
t
in
th
e
task
o
f
p
r
o
m
o
tin
g
tech
n
o
l
o
g
ical
p
r
o
g
r
ess
an
d
in
cu
r
s
io
n
to
im
p
r
o
v
e
th
e
s
u
p
p
ly
ch
ain
in
d
if
f
er
e
n
t
ar
ea
s
.
I
n
o
r
d
e
r
to
g
u
id
e
th
is
task
,
th
i
s
s
tu
d
y
p
r
o
p
o
s
es
a
m
eth
o
d
o
l
o
g
y
to
ap
p
ly
in
d
u
s
tr
y
5
.
0
in
s
m
all
an
d
m
ed
iu
m
p
r
o
d
u
ctio
n
co
m
p
an
ies in
v
ir
tu
al
e
n
v
ir
o
n
m
en
ts
with
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
R
o
b
o
tic
p
r
o
d
u
ct
-
b
a
s
ed
ma
n
ip
u
la
tio
n
in
s
imu
la
ted
e
n
viro
n
m
en
t
(
Ju
a
n
C
a
milo
Gu
a
c
h
eta
-
A
l
b
a
)
5895
s
et
o
f
alg
o
r
ith
m
s
r
elate
d
to
id
en
tific
atio
n
an
d
m
an
ip
u
latio
n
o
f
p
r
o
d
u
ct
th
r
o
u
g
h
m
a
ch
in
e
lear
n
in
g
an
d
co
m
p
u
ter
v
is
io
n
b
y
r
o
b
o
tic
m
an
ip
u
lato
r
.
Fo
r
its
d
ev
el
o
p
m
en
t,
p
r
o
d
u
cts
ar
e
r
ep
licated
in
a
v
ir
tu
al
en
v
ir
o
n
m
en
t,
wh
e
r
e
ed
g
e
-
b
as
ed
alg
o
r
ith
m
s
[
1
6
]
,
[
1
7
]
ar
e
u
s
ed
to
d
is
cr
im
in
ate
o
b
jects
o
f
in
ter
est,
id
e
n
tify
f
ea
tu
r
es
u
s
in
g
th
e
s
p
ee
d
e
d
-
u
p
r
o
b
u
s
t
f
ea
tu
r
es
(
SUR
F
)
alg
o
r
ith
m
[
1
8
]
,
[
1
9
]
an
d
d
ee
p
lear
n
in
g
[
2
0
]
,
[
2
1
]
f
o
r
s
u
b
s
eq
u
en
t m
an
i
p
u
latio
n
b
y
a
UR
5
r
o
b
o
tic
ar
m
[
2
2
]
,
[
2
3
]
.
T
h
e
m
ain
o
b
jectiv
e
o
f
th
is
w
o
r
k
is
to
d
ev
elo
p
an
d
v
alid
at
e
a
h
y
b
r
id
co
m
p
u
ter
v
is
io
n
p
ip
elin
e
f
o
r
id
en
tify
in
g
a
n
d
m
a
n
ip
u
latin
g
v
is
u
ally
s
im
ilar
p
r
o
d
u
cts
u
s
in
g
an
in
d
u
s
tr
ial
r
o
b
o
tic
p
l
atf
o
r
m
in
a
f
u
lly
s
im
u
lated
en
v
ir
o
n
m
e
n
t.
T
h
e
k
ey
co
n
tr
ib
u
tio
n
s
in
clu
d
e
in
teg
r
atio
n
o
f
h
eter
o
g
en
eo
u
s
AI
te
ch
n
iq
u
es
f
o
r
o
b
ject
d
etec
tio
n
an
d
r
o
b
o
tic
co
n
tr
o
l;
im
p
lem
en
tatio
n
o
f
a
r
ea
l
-
ti
m
e
class
if
icatio
n
an
d
m
an
ip
u
latio
n
cy
cle
v
ia
th
e
C
o
p
p
eliaSim
–
MA
T
L
AB
in
ter
f
ac
e;
an
d
ev
alu
atio
n
o
f
th
e
m
eth
o
d
u
n
d
e
r
r
ea
lis
tic
co
n
d
i
tio
n
s
alig
n
ed
with
I
n
d
u
s
tr
y
5
.
0
p
r
in
cip
les.
I
n
th
e
ab
s
en
ce
o
f
s
tan
d
ar
d
ize
d
d
atasets
,
a
cu
s
to
m
s
im
u
latio
n
-
b
ased
d
ataset
was
cr
ea
ted
f
o
r
v
alid
atio
n
,
t
h
e
d
if
f
icu
lty
o
f
d
is
tin
g
u
is
h
in
g
p
r
o
d
u
cts
with
s
im
ilar
g
eo
m
etr
y
an
d
b
r
an
d
in
g
h
ig
h
lig
h
ts
th
e
ap
p
licab
ilit
y
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
to
f
u
tu
r
e
in
d
u
s
tr
ial
au
to
m
atio
n
.
2.
M
E
T
H
O
D
Me
th
o
d
o
lo
g
y
p
r
o
p
o
s
ed
is
illu
s
tr
ated
in
Fig
u
r
e
1
an
d
c
o
n
s
is
ts
o
f
f
iv
e
p
h
ases
.
T
h
e
f
ir
s
t
p
h
as
e
in
v
o
lv
es
s
elec
tin
g
th
e
ap
p
licatio
n
en
v
ir
o
n
m
en
t.
T
h
e
s
ec
o
n
d
p
h
ase
f
o
cu
s
es
o
n
d
ef
in
in
g
th
e
ar
t
if
icial
in
tellig
en
ce
alg
o
r
ith
m
f
o
r
p
r
o
d
u
ct
lab
el
id
en
tific
atio
n
.
I
n
t
h
ir
d
p
h
ase,
th
e
alg
o
r
ith
m
f
o
r
p
ac
k
a
g
e
r
ec
o
g
n
itio
n
is
estab
lis
h
ed
.
Fo
u
r
th
p
h
ase
in
v
o
l
v
es
p
r
o
g
r
a
m
m
in
g
th
e
m
o
v
em
en
t
o
f
th
e
r
o
b
o
tic
ar
m
f
o
r
p
r
o
d
u
ct
h
an
d
lin
g
.
I
n
th
e
f
in
al
p
h
ase,
th
e
alg
o
r
ith
m
s
ar
e
in
te
g
r
ated
,
an
d
co
m
p
lete
s
im
u
latio
n
is
ex
ec
u
ted
.
C
o
p
p
eliaSim
s
er
v
es
as
s
im
u
latio
n
en
v
ir
o
n
m
en
t,
wh
ile
MA
T
L
AB
is
u
s
ed
f
o
r
p
r
o
g
r
am
m
in
g
; two
s
o
f
twar
e
ar
e
co
n
n
ec
ted
th
r
o
u
g
h
th
e
r
e
m
o
te
API
.
Fig
u
r
e
1
.
Me
th
o
d
o
lo
g
y
f
lo
wch
ar
t
T
o
estab
lis
h
an
in
d
u
s
tr
ial
ap
p
licatio
n
th
at
f
r
am
es
co
n
ce
p
t
s
o
f
in
d
u
s
tr
y
5
.
0
in
s
m
all
an
d
m
ed
iu
m
-
s
ized
co
m
p
an
ies,
a
liq
u
id
cl
ea
n
in
g
p
r
o
d
u
ct
ar
e
c
h
o
s
en
,
o
r
ien
ted
to
war
d
s
m
o
s
t
co
m
m
o
n
ly
u
s
ed
in
b
o
t
h
d
o
m
esti
c
an
d
o
r
g
an
izatio
n
al
en
v
ir
o
n
m
en
ts
.
Fo
r
th
is
p
u
r
p
o
s
e,
web
s
ite
o
f
th
e
m
icr
o
co
m
p
an
y
I
n
d
u
s
tr
ias
No
v
aq
u
im
[
2
4
]
,
w
h
ich
o
f
f
er
s
a
wid
e
v
ar
iety
o
f
f
o
r
m
ats,
lab
e
ls
an
d
d
esig
n
s
o
f
d
if
f
er
e
n
t c
lean
in
g
p
r
o
d
u
cts,
was
co
n
s
u
lted
,
am
o
n
g
o
th
e
r
s
.
T
h
is
r
ev
iew
allo
ws
u
n
d
er
s
ta
n
d
in
g
p
r
o
d
u
ct
p
r
ef
e
r
en
ce
s
an
d
er
g
o
n
o
m
ics
o
f
p
ac
k
ag
in
g
.
W
i
t
h
t
h
i
s
i
n
f
o
r
m
a
t
i
o
n
,
t
h
e
s
el
e
c
t
e
d
p
r
o
d
u
c
t
s
a
r
e
f
il
t
e
r
e
d
in
t
o
t
h
e
i
r
C
AD
f
o
r
m
a
t
s
.
F
i
v
e
p
a
c
k
a
g
i
n
g
f
o
r
m
a
t
s
a
r
e
c
h
o
s
e
n
c
o
r
r
es
p
o
n
d
i
n
g
t
o
v
o
l
u
m
e
s
o
f
5
0
0
c
c
,
8
0
0
c
c
,
½
g
al
l
o
n
,
1
g
a
ll
o
n
a
n
d
2
g
a
l
lo
n
s
.
F
o
r
s
i
m
u
l
at
i
o
n
s
p
e
c
i
f
i
ca
t
i
o
n
s
,
s
i
x
p
r
o
d
u
c
t
t
y
p
es
a
r
e
s
e
le
c
t
e
d
:
d
i
s
h
w
a
s
h
e
r
,
n
e
u
t
r
a
l
c
le
a
n
e
r
,
m
u
l
ti
p
u
r
p
o
s
e
cl
e
an
e
r
,
p
o
l
y
m
e
r
i
c
wax
,
b
leac
h
an
d
u
ltra
clea
n
er
.
Fig
u
r
e
2
ex
p
o
s
es
th
e
c
o
m
p
ar
is
o
n
b
etwe
en
r
ea
l
p
r
o
d
u
cts
an
d
th
eir
r
ep
r
esen
tatio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
9
4
-
5
9
0
3
5896
Fig
u
r
e
2
(
a)
s
h
o
ws
th
e
d
if
f
er
en
t
p
r
esen
tatio
n
s
o
f
th
ese
p
r
o
d
u
cts
in
r
ea
l
p
h
o
to
g
r
ap
h
s
.
Sin
ce
5
ty
p
es
o
f
p
ac
k
ag
in
g
,
6
ty
p
es
o
f
lab
els,
7
p
r
o
d
u
ct
tex
tu
r
e
co
l
o
r
s
an
d
3
ca
p
co
lo
r
s
ar
e
u
s
ed
,
5
5
d
if
f
er
en
t
p
r
o
d
u
cts
h
av
e
b
ee
n
cr
ea
ted
f
o
r
s
im
u
latio
n
,
s
e
lectin
g
th
e
m
o
s
t r
ep
r
esen
tativ
e
an
d
r
ea
lis
tic
co
m
b
in
atio
n
s
.
I
n
Fig
u
r
e
2
(
b
)
,
th
ese
co
n
f
ig
u
r
atio
n
s
ca
n
b
e
o
b
s
er
v
e
d
in
th
e
C
o
p
p
elia
s
im
u
latio
n
e
n
v
ir
o
n
m
en
t,
s
h
o
win
g
th
eir
lab
els an
d
co
lo
r
s
.
(
a)
(
b
)
Fig
u
r
e
2
.
C
o
m
p
a
r
is
o
n
b
etwe
e
n
(
a)
r
ea
l
p
r
o
d
u
cts
[
2
4
]
an
d
(
b
)
th
eir
r
ep
r
esen
tatio
n
in
C
o
p
p
e
lia
T
o
g
en
er
ate
th
e
C
o
p
p
eliaSim
d
atab
ase,
o
b
jects
ar
e
r
o
ta
ted
in
1
0
-
d
e
g
r
ee
in
cr
e
m
en
ts
,
ca
p
tu
r
in
g
im
ag
es
at
ea
ch
s
tep
to
p
r
o
d
u
c
e
1
,
9
8
0
im
ag
es,
d
iv
id
ed
in
to
7
0
%
f
o
r
tr
ain
in
g
,
1
5
%
f
o
r
v
alid
atio
n
,
an
d
1
5
%
f
o
r
test
in
g
.
T
o
e
n
s
u
r
e
r
o
b
u
s
tn
ess
ag
ain
s
t
o
r
ien
tatio
n
,
s
ca
le,
an
d
illu
m
in
atio
n
v
ar
iatio
n
s
,
t
h
e
S
UR
F
alg
o
r
ith
m
is
u
s
ed
.
I
t
b
eg
in
s
b
y
ca
lcu
latin
g
p
o
in
ts
o
f
in
ter
est
f
r
o
m
th
e
d
et
er
m
in
an
t
o
f
th
e
s
ca
le
-
n
o
r
m
alize
d
Hess
ian
m
atr
ix
,
as
s
h
o
wn
in
(
1
)
,
wh
e
r
e
,
,
an
d
ar
e
s
ec
o
n
d
-
o
r
d
er
im
ag
e
d
er
iv
ativ
es
at
s
ca
le
σ
.
Descr
ip
to
r
s
ar
e
ex
tr
ac
ted
an
d
m
atch
ed
ag
ai
n
s
t
a
lab
el
d
atab
ase
u
s
in
g
a
n
ea
r
est
-
n
eig
h
b
o
r
r
atio
th
r
esh
o
l
d
o
f
0
.
8
,
a
u
n
iq
u
e
n
ess
co
n
s
tr
ain
t,
an
d
a
r
eg
io
n
o
f
in
ter
est
(
R
OI
)
is
d
ef
in
ed
to
im
p
r
o
v
e
ac
c
u
r
ac
y
.
T
h
e
la
b
el
with
m
o
r
e
m
atc
h
es
is
s
elec
ted
,
an
d
a
b
o
u
n
d
in
g
b
o
x
i
s
ass
ig
n
ed
to
lo
ca
lize
th
e
o
b
je
ct.
ℋ
(
,
,
σ
)
=
σ
2
(
(
,
,
σ
)
⋅
(
,
,
σ
)
−
(
,
,
σ
)
2
)
(
1
)
Fig
u
r
e
3
s
h
o
ws
th
e
e
d
g
es
o
f
p
atter
n
s
to
b
e
u
s
ed
,
Fig
u
r
e
3
(
a)
co
r
r
esp
o
n
d
s
to
a
5
0
0
cc
p
r
o
d
u
ct,
Fig
u
r
e
3
(
b
)
to
a
n
8
0
0
cc
p
r
o
d
u
ct,
Fig
u
r
e
3
(
c
)
to
a
½
g
allo
n
p
r
o
d
u
ct,
Fig
u
r
e
3
(
d
)
to
1
g
allo
n
,
an
d
Fig
u
r
e
3
(
e)
t
o
2
g
allo
n
s
,
th
ese
p
atter
n
s
s
er
v
e
as
a
r
ef
er
en
ce
in
class
if
icatio
n
s
tr
ateg
y
,
allo
win
g
to
ev
alu
a
te
ef
f
ec
tiv
en
ess
o
f
th
e
alg
o
r
ith
m
s
in
id
en
tific
atio
n
an
d
r
ec
o
g
n
itio
n
o
f
th
e
d
if
f
er
en
t
o
b
jects.
C
o
m
p
ar
is
o
n
o
f
th
ese
ed
g
es
with
p
r
o
ce
s
s
ed
im
ag
es is
ess
en
tial t
o
en
s
u
r
e
th
e
ac
cu
r
ac
y
an
d
r
o
b
u
s
tn
ess
o
f
th
e
m
eth
o
d
s
to
b
e
e
v
alu
ated
.
(
a)
(
b
)
(
c)
(
d
)
(
e)
Fig
u
r
e
3
.
B
o
r
d
er
p
atter
n
s
f
o
r
c
lean
in
g
p
r
o
d
u
cts:
(
a)
5
0
0
cc
,
(
b
)
8
0
0
cc
,
(
c)
½
g
allo
n
,
(
d
)
1
g
allo
n
,
an
d
(
e)
2
g
allo
n
s
MA
T
L
AB
co
n
n
ec
ts
to
C
o
p
p
eliaSim
v
ia
its
r
em
o
te
API
,
u
s
in
g
a
v
is
io
n
s
en
s
o
r
to
ca
p
t
u
r
e
im
ag
es
f
r
o
m
th
e
s
im
u
latio
n
.
E
ac
h
im
ag
e
is
co
n
v
er
te
d
to
g
r
ay
s
ca
le,
an
d
ed
g
e
d
etec
tio
n
is
p
er
f
o
r
m
ed
u
s
in
g
th
e
So
b
el
an
d
C
an
n
y
m
eth
o
d
s
[
2
5
]
.
Fig
u
r
e
4
s
h
o
ws
th
e
r
esu
lts
f
o
r
2
-
g
allo
n
p
ac
k
ag
e:
So
b
el
in
Fig
u
r
e
4
(
a)
an
d
C
an
n
y
in
Fig
u
r
e
4
(
b
)
u
s
in
g
th
e
3
D
v
iewe
r
,
an
d
So
b
el
in
Fig
u
r
e
4
(
c
)
a
n
d
C
an
n
y
in
Fig
u
r
e
4
(
d
)
u
s
in
g
C
o
p
p
eliaSim
.
T
h
e
s
im
u
lated
im
ag
es
in
tr
o
d
u
ce
m
o
r
e
n
o
is
e,
with
C
an
n
y
d
etec
tin
g
m
o
r
e
e
d
g
es
b
u
t
also
ca
p
tu
r
i
n
g
ex
ce
s
s
iv
e
d
etail,
wh
ich
m
ay
h
in
d
er
class
if
icatio
n
.
No
tab
l
y
,
b
o
th
f
ilter
s
also
d
etec
t
th
e
lab
el
ed
g
es,
wh
ich
is
u
n
d
esira
b
le
s
in
ce
o
n
ly
th
e
o
u
te
r
co
n
to
u
r
is
r
ele
v
an
t.
T
h
is
co
n
f
ir
m
s
th
at
b
o
th
f
ilter
s
p
r
o
v
id
e
m
o
r
e
d
etail
th
a
n
n
ec
ess
ar
y
f
o
r
t
h
e
task
.
Yo
u
o
n
l
y
lo
o
k
o
n
ce
(
YOL
O)
n
etwo
r
k
is
a
d
ee
p
lear
n
in
g
m
o
d
el
k
n
o
w
n
f
o
r
its
s
p
ee
d
an
d
a
cc
u
r
ac
y
in
o
b
ject
d
etec
tio
n
,
u
s
ed
in
r
ea
l
-
tim
e
ap
p
licatio
n
s
[
2
6
]
.
I
n
th
is
s
tu
d
y
th
e
YOL
O
v
2
ar
c
h
itectu
r
e
was
c
o
n
f
ig
u
r
e
d
to
lo
ca
te
a
n
d
class
if
y
f
i
v
e
ty
p
es
o
f
o
b
jects,
o
th
er
p
r
et
r
ain
ed
d
ee
p
lear
n
in
g
m
o
d
els
co
n
s
id
er
ed
s
u
ch
as
Fas
ter
R
-
C
NN
an
d
E
f
f
icien
tDet,
wh
i
ch
ty
p
ically
o
f
f
er
h
ig
h
e
r
d
etec
tio
n
ac
cu
r
ac
y
,
esp
ec
ially
f
o
r
s
m
all
o
r
o
v
er
la
p
p
in
g
o
b
jects,
at
th
e
c
o
s
t
o
f
i
n
cr
ea
s
ed
co
m
p
u
tatio
n
al
d
em
a
n
d
.
YOL
O
v
2
was
u
ltima
tely
c
h
o
s
en
f
o
r
its
b
alan
ce
b
etwe
en
s
p
ee
d
an
d
s
u
f
f
icien
t a
cc
u
r
ac
y
with
in
th
e
co
n
s
tr
ain
ts
o
f
th
e
tar
g
eted
in
d
u
s
tr
ial
s
ce
n
ar
io
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
R
o
b
o
tic
p
r
o
d
u
ct
-
b
a
s
ed
ma
n
ip
u
la
tio
n
in
s
imu
la
ted
e
n
viro
n
m
en
t
(
Ju
a
n
C
a
milo
Gu
a
c
h
eta
-
A
l
b
a
)
5897
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
4
.
R
esu
lts
o
f
th
e
ap
p
licatio
n
o
f
f
ilter
s
o
n
th
e
2
-
g
allo
n
p
ac
k
ag
e
So
b
el
in
(
a)
an
d
C
an
n
y
in
(
b
)
u
s
in
g
t
h
e
3
D
v
iewe
r
,
an
d
So
b
el
(
c
)
an
d
C
an
n
y
in
(
d
)
u
s
in
g
C
o
p
p
eliaSim
T
h
e
m
o
tio
n
o
f
th
e
UR
5
r
o
b
o
t
f
r
o
m
it
i
s
cu
r
r
e
n
t
to
tar
g
et
p
o
s
itio
n
an
d
o
r
ien
tatio
n
is
p
r
o
g
r
am
m
e
d
u
s
in
g
p
o
ly
n
o
m
ial
in
ter
p
o
latio
n
(
tp
o
ly
)
to
g
e
n
er
ate
s
m
o
o
th
j
o
in
t
tr
ajec
to
r
ies.
I
n
v
er
s
e
k
in
e
m
atics
is
ap
p
lied
to
co
m
p
u
te
th
e
j
o
in
t
v
alu
es
co
r
r
e
s
p
o
n
d
in
g
t
o
th
e
tar
g
et
p
o
s
e.
A
t
ea
ch
s
tep
,
th
e
co
m
p
u
ted
jo
in
t
p
o
s
itio
n
s
ar
e
s
en
t
to
C
o
p
p
eliaSim
to
s
im
u
late
t
h
e
r
o
b
o
t'
s
m
o
v
em
en
t,
wh
ile
th
e
en
d
-
ef
f
ec
to
r
p
o
s
itio
n
is
r
ec
o
r
d
ed
in
a
n
o
u
tp
u
t
m
atr
ix
.
T
h
e
r
o
b
o
t
is
p
r
o
g
r
a
m
m
ed
to
class
if
y
an
o
b
ject
b
y
c
ap
tu
r
in
g
im
a
g
es
th
r
o
u
g
h
a
ca
m
er
a,
id
en
tify
i
n
g
its
lab
el
an
d
p
ac
k
ag
i
n
g
ty
p
e
u
s
in
g
SUR
F
an
d
YOL
O
a
lg
o
r
ith
m
s
,
r
esp
ec
tiv
ely
,
an
d
tr
an
s
p
o
r
tin
g
th
e
o
b
ject
to
a
d
esig
n
ated
s
h
elf
th
r
o
u
g
h
a
p
r
e
d
ef
in
ed
s
eq
u
en
ce
o
f
p
o
s
es.
O
n
ce
th
e
task
is
co
m
p
leted
,
th
e
r
o
b
o
t
r
etu
r
n
s
to
its
in
itial p
o
s
itio
n
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
Af
ter
d
ef
in
in
g
an
d
im
p
lem
en
tin
g
th
e
d
etec
tio
n
a
n
d
class
if
icatio
n
alg
o
r
ith
m
s
,
th
is
s
ec
tio
n
p
r
esen
ts
th
e
r
esu
lts
o
f
th
eir
ap
p
licatio
n
,
as
well
as
a
c
o
m
p
ar
ativ
e
an
al
y
s
is
o
f
th
eir
p
er
f
o
r
m
a
n
ce
.
T
h
e
co
d
e
im
p
lem
en
te
d
in
MA
T
L
AB
f
o
r
f
ea
tu
r
e
d
et
ec
tio
n
in
i
m
ag
es
u
s
in
g
SUR
F
alg
o
r
ith
m
is
ex
ec
u
ted
with
th
e
Fas
t
Hess
ian
alg
o
r
ith
m
.
Fig
u
r
e
5
s
h
o
ws
th
e
r
esu
lts
o
f
th
e
SUR
F
alg
o
r
ith
m
ap
p
lied
to
d
if
f
er
en
t
clea
n
in
g
p
r
o
d
u
ct
f
o
r
m
ats
,
u
s
in
g
th
e
s
ix
s
elec
ted
lab
els:
d
is
h
wash
er
,
n
eu
tr
al
clea
n
er
,
m
u
ltip
u
r
p
o
s
e
clea
n
er
,
p
o
l
y
m
e
r
ic
wax
,
b
leac
h
a
n
d
u
ltra
clea
n
er
.
T
h
ese
r
esu
lts
co
n
f
ir
m
th
e
co
r
r
ec
t p
er
f
o
r
m
an
ce
o
f
th
e
alg
o
r
ith
m
in
th
e
task
o
f
lab
el
d
etec
tio
n
an
d
class
if
icatio
n
,
en
s
u
r
in
g
th
at
th
e
s
y
s
tem
ca
n
p
r
o
p
er
l
y
id
en
t
if
y
ea
ch
p
r
o
d
u
ct
ac
co
r
d
in
g
t
o
its
co
r
r
esp
o
n
d
in
g
lab
el.
Fig
u
r
e
5
.
R
esu
lts
o
f
SUR
F a
lg
o
r
ith
m
ap
p
lied
to
d
i
f
f
er
en
t
p
r
o
d
u
cts f
o
r
m
ats with
th
eir
r
esp
ec
tiv
e
lab
els
T
h
e
Ho
p
f
ield
n
etwo
r
k
o
p
er
at
es
o
n
b
in
a
r
y
im
a
g
es
r
ep
r
esen
ted
as
b
it
m
atr
ices,
wh
er
e
ea
ch
p
ix
el
is
eith
er
0
o
r
1
.
Du
r
in
g
th
e
u
p
d
a
te
p
r
o
ce
s
s
,
it
iter
ativ
ely
ad
ju
s
t
s
th
e
b
it
v
alu
es
in
th
e
im
ag
e
u
n
til
th
ey
co
n
v
er
g
e
to
o
n
e
o
f
th
e
s
to
r
e
d
p
atter
n
s
,
th
u
s
m
in
im
izin
g
s
y
s
tem
en
er
g
y
.
T
h
is
b
eh
av
io
r
m
a
k
es
it
s
u
it
ab
le
f
o
r
task
s
lik
e
im
ag
e
r
etr
iev
al
an
d
d
e
n
o
is
in
g
.
T
h
er
ef
o
r
e,
a
Ho
p
f
ield
n
et
wo
r
k
is
tr
ain
ed
u
s
in
g
Heb
b
'
s
r
u
le
[
2
7
]
,
with
f
iv
e
r
ef
er
en
ce
im
a
g
es
p
r
o
ce
s
s
ed
with
th
e
So
b
el
f
ilter
an
d
th
en
t
h
is
n
etwo
r
k
is
u
s
ed
to
co
r
r
ec
t
a
d
is
to
r
ted
im
a
g
e,
th
e
o
b
jectiv
e
is
th
at
th
e
n
etwo
r
k
co
n
v
er
g
es to
t
h
e
r
ef
e
r
en
ce
i
m
ag
e
lik
e
th
e
test
im
ag
e.
T
h
e
alg
o
r
ith
m
was
in
teg
r
ate
d
with
C
o
p
p
elia,
en
ab
lin
g
d
i
r
ec
t
im
ag
e
ca
p
tu
r
e
f
r
o
m
th
e
s
im
u
latio
n
en
v
ir
o
n
m
en
t.
I
n
Fig
u
r
e
6
,
th
e
u
p
d
ates
f
o
r
ea
ch
iter
atio
n
o
f
th
e
Ho
p
f
ield
n
etwo
r
k
ar
e
s
h
o
wn
f
o
r
a
r
ea
l
-
tim
e
im
ag
e
ca
p
tu
r
e
d
with
th
e
So
b
el
f
ilter
ap
p
lied
to
e
n
h
an
ce
i
n
ter
ac
tio
n
with
th
e
en
v
ir
o
n
m
e
n
t.
Fig
u
r
e
6
(
a)
p
r
esen
ts
iter
atio
n
1
,
Fig
u
r
e
6
(
b
)
th
e
iter
atio
n
2
,
Fig
u
r
e
6
(
c)
th
e
iter
ati
o
n
4
an
d
Fig
u
r
e
6
(
d
)
th
e
iter
at
io
n
8
.
T
h
is
iter
ativ
e
v
is
u
aliza
tio
n
d
em
o
n
s
tr
ates
h
o
w
th
e
alg
o
r
ith
m
f
u
n
ctio
n
s
,
co
n
v
er
g
in
g
to
th
e
t
r
ain
ed
p
atter
n
.
T
h
is
in
teg
r
atio
n
is
ess
en
tial f
o
r
ac
cu
r
ate
p
r
o
d
u
ct
d
etec
tio
n
an
d
im
p
r
o
v
em
en
t
o
f
s
y
s
tem
p
er
f
o
r
m
a
n
ce
with
in
th
e
s
im
u
latio
n
.
T
h
e
d
esig
n
ed
Ho
p
f
ield
n
etwo
r
k
was
th
en
ap
p
lied
to
th
e
1
9
8
0
im
ag
es,
in
itially
p
r
o
ce
s
s
in
g
all
im
ag
es
with
an
ed
g
e
d
etec
tio
n
f
ilter
.
T
h
is
was
d
o
n
e
to
ev
alu
ate
th
e
n
etwo
r
k
’
s
p
er
f
o
r
m
an
ce
a
n
d
v
a
lid
ate
its
ac
cu
r
ac
y
.
T
h
er
ef
o
r
e,
th
e
c
o
n
f
u
s
io
n
m
atr
ix
f
o
r
th
e
5
o
b
jects
is
p
r
esen
te
d
in
T
a
b
le
1
.
T
h
is
tab
le
r
ev
ea
ls
th
at
th
e
Ho
p
f
ield
n
etwo
r
k
is
n
o
t
an
id
ea
l
s
tr
ateg
y
,
as
a
s
in
g
le
2
D
p
atter
n
ca
n
n
o
t
r
ep
r
esen
t
t
h
e
m
o
d
el
in
all
p
o
s
s
ib
le
o
r
ien
tatio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
9
4
-
5
9
0
3
5898
Ad
d
itio
n
ally
,
t
h
e
n
etwo
r
k
s
tr
u
g
g
les
to
d
is
cr
im
in
ate
n
o
is
e
an
d
is
o
n
ly
ef
f
ec
tiv
e
f
o
r
b
i
n
a
r
y
im
a
g
es
with
f
ew
p
ix
els.
Alth
o
u
g
h
th
e
n
etwo
r
k
i
s
f
ast,
s
im
p
le
to
p
r
o
g
r
am
,
an
d
ex
ec
u
te,
it is
n
o
t f
u
n
ctio
n
al
f
o
r
th
is
ap
p
licatio
n
.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
6
.
I
te
r
ativ
e
u
p
d
ates o
f
t
h
e
Ho
p
f
ield
n
etwo
r
k
f
o
r
i
m
ag
e:
I
ter
atio
n
(
a)
1
,
(
b
)
2
,
(
c)
4
,
a
n
d
(
d
)
8
T
ab
le
1
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
class
if
icatio
n
o
f
5
o
b
jects u
s
in
g
th
e
Ho
p
f
ield
a
lg
o
r
ith
m
P
r
e
d
i
c
t
e
d
5
0
0
cc
P
r
e
d
i
c
t
e
d
8
0
0
cc
P
r
e
d
i
c
t
e
d
½
g
a
l
P
r
e
d
i
c
t
e
d
1
g
a
l
P
r
e
d
i
c
t
e
d
2
g
a
l
A
c
t
u
a
l
5
0
0
cc
1
5
0
(
3
7
.
8
8
%)
64
73
52
57
A
c
t
u
a
l
8
0
0
cc
50
1
8
0
(
4
5
.
4
5
%)
63
66
37
A
c
t
u
a
l
½
g
a
l
74
46
1
0
0
(
2
5
.
2
5
%)
79
61
A
c
t
u
a
l
1
g
a
l
66
49
71
1
4
0
(
3
5
.
3
5
%)
50
A
c
t
u
a
l
2
g
a
l
63
44
57
56
1
5
0
(
3
7
.
8
8
%)
2
D
cr
o
s
s
-
co
r
r
elatio
n
is
a
p
o
w
er
f
u
l
tech
n
iq
u
e
f
o
r
c
o
m
p
ar
i
n
g
b
in
ar
y
im
a
g
es,
allo
win
g
th
e
d
etec
tio
n
o
f
s
im
ilar
ities
an
d
p
atter
n
s
th
r
o
u
g
h
d
is
p
lace
m
en
ts
[
2
8
]
.
T
o
d
eter
m
in
e
th
e
cr
o
s
s
-
C
o
r
r
elati
o
n
f
o
r
two
b
in
ar
y
im
ag
es,
f
ir
s
t,
it
is
v
er
if
ied
th
at
th
ey
h
av
e
c
o
m
p
atib
le
d
im
en
s
io
n
s
s
o
as
n
o
t
to
g
en
er
at
e
er
r
o
r
;
th
e
n
,
it
is
ca
lcu
lated
in
MA
T
L
AB
an
d
t
h
e
m
ax
im
u
m
v
alu
e
o
f
th
e
r
esu
ltin
g
co
r
r
elatio
n
m
atr
ix
is
ex
tr
ac
ted
,
in
d
icatin
g
th
e
h
ig
h
est
d
e
g
r
ee
o
f
co
in
cid
en
ce
b
etwe
en
th
e
two
im
ag
e
s
in
q
u
esti
o
n
.
T
o
co
m
p
ar
e
th
e
cr
o
s
s
-
co
r
r
elatio
n
,
th
r
ee
ex
am
p
les
o
f
ed
g
es
ca
l
cu
lated
in
th
e
C
o
p
p
elia
s
im
u
latio
n
s
o
f
twar
e
wer
e
s
elec
t
ed
,
th
e
r
esu
lts
ar
e
p
r
esen
ted
at
T
ab
le
2
,
wh
ich
will
b
e
u
s
ed
to
q
u
an
tify
th
e
c
o
r
r
elatio
n
m
etr
ic
an
d
ev
alu
ate
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
class
if
icatio
n
s
tr
ateg
y
.
T
ab
le
2
.
C
r
o
s
s
-
co
r
r
elatio
n
m
et
r
ic
r
esu
lts
f
o
r
p
atter
n
ed
g
es
C
r
o
ss
-
c
o
r
r
e
l
a
t
i
o
n
5
0
0
c
c
8
0
0
c
c
½
g
a
l
l
o
n
1
g
a
l
l
o
n
2
g
a
l
l
o
n
s
½
g
a
l
l
o
n
0
.
1
2
0
3
2
2
0
.
1
2
1
0
4
7
0.
1
2
2
0
1
7
0
.
1
0
6
2
6
1
0
.
0
9
4
1
7
1
8
0
0
c
c
0
.
1
2
2
9
0
9
0
.
1
6
4
4
4
6
0
.
0
9
6
9
6
8
0
.
1
0
9
2
1
6
0
.
1
0
9
2
1
6
1
g
a
l
l
o
n
0
.
1
2
2
3
9
9
0
.
1
2
4
0
6
7
0
.
0
9
4
2
4
1
0
.
1
8
8
3
4
8
0
.
0
9
5
8
6
3
T
h
is
m
etr
ic
was
th
en
ap
p
lie
d
to
all
1
9
8
0
im
a
g
es,
o
n
ce
ag
ain
s
tar
tin
g
with
th
e
So
b
el
f
ilter
an
d
ca
lcu
latin
g
co
r
r
elatio
n
b
etwe
e
n
p
atter
n
s
an
d
im
ag
e
tak
en
d
i
r
ec
tly
f
r
o
m
C
o
p
p
eliaSim
.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
f
o
r
class
if
icatio
n
o
f
th
e
5
o
b
jects
is
p
r
esen
ted
in
T
ab
le
3
.
As
s
ee
n
in
th
e
tab
le,
th
e
r
esu
lts
im
p
r
o
v
ed
s
ig
n
if
ican
tly
with
th
is
m
eth
o
d
,
s
h
o
win
g
s
tr
o
n
g
p
er
f
o
r
m
an
ce
f
o
r
th
e
5
0
0
cc
an
d
1
-
g
allo
n
o
b
jects.
T
h
is
im
p
r
o
v
em
e
n
t
o
cc
u
r
r
ed
b
ec
au
s
e
th
e
s
h
ap
e
o
f
th
e
s
o
lid
s
d
o
es
n
o
t
ch
an
g
e
s
ig
n
if
ican
tly
wh
e
n
r
o
tated
.
Ho
wev
e
r
,
th
e
class
if
icatio
n
f
o
r
th
e
o
th
er
o
b
jects,
esp
ec
ially
th
e
½
g
all
o
n
an
d
2
-
g
allo
n
o
b
jects,
was
n
o
t
as
ac
cu
r
ate
d
u
e
to
th
eir
m
o
r
e
co
m
p
le
x
g
e
o
m
et
r
y
.
Alth
o
u
g
h
th
is
s
tr
ateg
y
is
s
i
m
p
le
an
d
f
ast,
it
is
o
n
ly
ef
f
ec
t
iv
e
f
o
r
cy
lin
d
r
ical
o
b
jects.
Fo
r
o
b
jects with
r
ec
ta
n
g
u
lar
p
r
is
m
s
h
ap
es,
th
is
m
eth
o
d
p
r
o
v
es in
ef
f
icie
n
t f
o
r
class
i
f
icatio
n
.
T
ab
le
3
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
class
if
icatio
n
o
f
5
o
b
jects u
s
in
g
2
D
cr
o
s
s
-
co
r
r
elatio
n
P
r
e
d
i
c
t
e
d
5
0
0
c
c
P
r
e
d
i
c
t
e
d
8
0
0
c
c
P
r
e
d
i
c
t
e
d
½
g
a
l
P
r
e
d
i
c
t
e
d
1
g
a
l
P
r
e
d
i
c
t
e
d
2
g
a
l
A
c
t
u
a
l
5
0
0
c
c
3
6
0
(
9
0
.
9
1
%)
12
15
3
6
A
c
t
u
a
l
8
0
0
c
c
28
2
4
6
(
6
2
.
1
2
%)
47
35
40
A
c
t
u
a
l
½
g
a
l
50
60
1
6
2
(
4
0
.
9
1
%)
75
49
A
c
t
u
a
l
1
g
a
l
16
14
25
3
2
1
(
8
1
.
0
6
%)
20
A
c
t
u
a
l
2
g
a
l
55
44
66
53
1
7
8
(
4
4
.
9
5
%)
B
o
th
th
e
cr
o
s
s
-
co
r
r
elatio
n
a
n
d
th
e
SUR
F
alg
o
r
ith
m
wer
e
i
n
teg
r
ated
to
p
r
ec
is
ely
d
ef
in
e
th
e
ty
p
e
o
f
o
b
ject
in
ca
m
er
a'
s
f
ield
o
f
v
i
ew
an
d
th
u
s
p
r
o
ce
ed
to
its
class
if
icatio
n
.
Fig
u
r
e
7
s
h
o
ws
th
e
m
etr
ics
o
b
tain
ed
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
R
o
b
o
tic
p
r
o
d
u
ct
-
b
a
s
ed
ma
n
ip
u
la
tio
n
in
s
imu
la
ted
e
n
viro
n
m
en
t
(
Ju
a
n
C
a
milo
Gu
a
c
h
eta
-
A
l
b
a
)
5899
an
d
d
etec
tio
n
o
f
o
b
jects
in
tw
o
ca
s
es,
u
s
in
g
im
ag
es
tak
en
d
i
r
ec
tly
f
r
o
m
th
e
C
o
p
p
elia
s
im
u
latio
n
en
v
ir
o
n
m
en
t
.
Fig
u
r
e
7
(
a)
s
h
o
ws
th
e
8
0
0
cc
co
n
tain
er
an
d
Fig
u
r
e
7
(
b
)
s
h
o
w
a
p
r
o
d
u
ct
in
½
g
allo
n
c
o
n
ta
in
er
.
T
h
i
s
f
u
s
i
o
n
o
f
t
e
c
h
n
i
q
u
e
s
al
l
o
ws
t
o
i
m
p
r
o
v
e
th
e
e
f
f
e
c
t
i
v
e
n
es
s
i
n
t
h
e
i
d
e
n
t
i
f
i
ca
t
i
o
n
a
n
d
c
l
as
s
i
f
i
c
a
ti
o
n
o
f
t
h
e
an
a
l
y
z
e
d
p
r
o
d
u
c
t
s
.
T
h
e
YOL
O
co
n
f
ig
u
r
atio
n
ai
m
ed
to
b
alan
ce
tr
ain
in
g
tim
e
an
d
d
etec
tio
n
ac
cu
r
ac
y
f
o
r
f
iv
e
d
is
tin
ct
p
r
o
d
u
ct
co
n
tain
er
s
.
T
h
e
p
r
o
ce
s
s
b
eg
an
with
lo
ad
in
g
a
p
r
etr
ain
ed
m
o
d
el
an
d
d
ef
in
in
g
th
e
n
e
two
r
k
ar
ch
itectu
r
e,
in
clu
d
in
g
an
ch
o
r
b
o
x
esti
m
atio
n
an
d
a
n
in
p
u
t
im
a
g
e
s
ize
o
f
7
2
0
×7
2
0
p
ix
els
with
th
r
ee
co
l
o
r
ch
a
n
n
els.
Du
r
i
n
g
tr
ain
in
g
,
f
ea
tu
r
e
ex
tr
ac
tio
n
l
ay
er
was
s
et
to
leak
y
R
e
LU
5
,
with
f
i
v
e
an
c
h
o
r
b
o
x
es
an
d
th
e
R
MSPr
o
p
o
p
tim
izer
.
T
h
e
k
ey
t
r
ain
in
g
p
ar
am
eter
s
in
clu
d
ed
an
i
n
itial
lear
n
in
g
r
ate
o
f
0
.
0
0
1
,
m
in
i
-
b
atch
s
ize
o
f
8
,
a
m
ax
im
u
m
o
f
1
0
ep
o
ch
s
,
a
p
i
ec
ewise
lear
n
in
g
r
ate
s
ch
ed
u
l
e,
an
d
a
d
r
o
p
p
er
io
d
o
f
5
e
p
o
ch
s
.
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
,
co
n
v
e
r
g
es
n
ea
r
iter
atio
n
1
0
wh
er
e
th
e
lo
s
s
er
r
o
r
is
r
ed
u
ce
d
to
ze
r
o
,
s
h
o
ws
th
e
ev
o
lu
tio
n
o
f
th
e
er
r
o
r
lo
s
s
ac
r
o
s
s
ep
o
ch
s
,
h
ig
h
lig
h
ti
n
g
th
e
m
o
d
el’
s
co
n
v
er
g
e
n
ce
b
eh
av
io
r
an
d
lear
n
in
g
s
tab
ilit
y
.
Fig
u
r
e
8
s
h
o
ws
th
e
p
r
ec
is
io
n
v
s
.
r
ec
all
p
lo
ts
an
d
th
e
s
co
r
e
v
s
.
r
ec
all
p
lo
ts
f
o
r
ea
ch
o
f
th
e
f
iv
e
o
b
jects,
u
s
in
g
th
e
test
im
ag
es.
Su
b
p
lo
ts
co
r
r
esp
o
n
d
to
:
Fig
u
r
e
8
(
a)
5
0
0
cc
,
Fig
u
r
e
8
(
b
)
8
0
0
cc
,
Fig
u
r
e
8
(
c
)
½
g
allo
n
,
Fig
u
r
e
8
(
d
)
1
g
allo
n
,
an
d
Fig
u
r
e
8
(
e)
2
g
allo
n
s
.
W
h
en
an
aly
zi
n
g
th
ese
p
lo
ts
,
it
is
ev
id
en
t
th
at
th
e
m
o
d
el
p
er
f
o
r
m
ed
ad
e
q
u
ately
i
n
d
etec
tin
g
all
f
iv
e
o
b
jects
d
u
r
i
n
g
test
in
g
,
with
th
e
b
est
r
esu
l
ts
o
b
s
er
v
ed
f
o
r
t
h
e
5
0
0
m
l,
8
0
0
m
l,
an
d
1
-
g
allo
n
co
n
tain
er
s
.
(
a)
(
b
)
Fig
u
r
e
7
.
Ob
ject
d
etec
tio
n
an
d
class
if
icatio
n
r
esu
lts
u
s
in
g
cr
o
s
s
co
r
r
elatio
n
an
d
SUR
F a
lg
o
r
ith
m
f
o
r
clea
n
in
g
p
r
o
d
u
cts:
(
a)
8
0
0
cc
an
d
(
b
)
½
g
allo
n
(
a)
(
b
)
(
c)
(
d
)
(
e)
Fig
u
r
e
8
.
Acc
u
r
ac
y
a
n
d
r
ec
all
p
lo
ts
f
o
r
tr
ain
e
d
o
b
jects
: (
a)
5
0
0
cc
,
(
b
)
8
0
0
cc
,
(
c)
½
g
allo
n
,
(
d
)
1
g
allo
n
,
an
d
(
e)
2
g
allo
n
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
9
4
-
5
9
0
3
5900
T
h
ese
f
iv
e
ca
teg
o
r
ies
s
h
o
wed
h
ig
h
er
p
r
ec
is
io
n
an
d
r
ec
all
v
a
lu
es,
in
d
icatin
g
b
etter
g
en
e
r
aliza
tio
n
b
y
th
e
tr
ain
ed
m
o
d
el.
I
n
T
ab
le
4
,
th
e
co
n
f
u
s
io
n
m
atr
i
x
f
o
r
t
h
e
class
if
icatio
n
r
esu
lt
s
is
p
r
esen
ted
,
r
esu
lts
n
o
w
d
em
o
n
s
tr
ate
a
v
alid
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
.
T
h
e
n
etwo
r
k
'
s
ab
ilit
y
to
co
r
r
ec
tly
d
etec
t
o
b
jects
,
esp
ec
ially
f
o
r
th
e
5
0
0
cc
,
8
0
0
cc
,
a
n
d
1
-
g
allo
n
o
b
jects,
is
n
o
tewo
r
th
y
,
with
ac
cu
r
ac
y
p
e
r
ce
n
tag
es
a
b
o
v
e
9
2
%.
Desp
ite
th
is
,
YOL
O
m
o
d
el
h
as
s
ig
n
if
ican
tly
im
p
r
o
v
ed
class
if
icatio
n
p
er
f
o
r
m
an
ce
,
co
m
p
ar
ed
t
o
th
e
p
r
ev
io
u
s
d
etec
tio
n
ap
p
r
o
ac
h
es,
wh
er
e
o
n
l
y
th
e
5
0
0
cc
o
b
ject
was
s
u
cc
ess
f
u
lly
d
etec
ted
.
T
h
is
s
u
g
g
ests
th
at
th
e
m
o
d
el
ca
n
n
o
w
r
eliab
ly
class
if
y
m
o
s
t
o
f
th
e
o
b
jects,
th
o
u
g
h
f
u
r
th
er
t
u
n
in
g
an
d
ad
d
itio
n
al
d
ata
m
a
y
b
e
r
eq
u
ir
ed
to
en
h
a
n
ce
d
etec
tio
n
f
o
r
m
o
r
e
co
m
p
le
x
s
h
ap
es lik
e
th
e
2
-
g
allo
n
co
n
tain
e
r
.
T
ab
le
4
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
class
if
icatio
n
o
f
5
o
b
jects u
s
in
g
YOL
O
v2
a
lg
o
r
ith
m
P
r
e
d
i
c
t
e
d
5
0
0
cc
P
r
e
d
i
c
t
e
d
8
0
0
cc
P
r
e
d
i
c
t
e
d
½
g
a
l
P
r
e
d
i
c
t
e
d
1
g
a
l
P
r
e
d
i
c
t
e
d
2
g
a
l
A
c
t
u
a
l
5
0
0
cc
3
8
4
(
9
7
.
0
2
%)
4
6
5
2
A
c
t
u
a
l
8
0
0
cc
5
3
7
6
(
9
5
.
2
0
%)
5
5
5
A
c
t
u
a
l
½
g
a
l
4
6
3
4
0
(
8
6
.
1
7
%)
2
44
A
c
t
u
a
l
1
g
a
l
5
7
10
3
6
4
(
9
2
.
1
7
%)
12
A
c
t
u
a
l
2
g
a
l
5
6
52
8
3
2
5
(
8
2
.
0
7
%)
R
esu
lts
s
h
o
wn
in
T
ab
le
5
s
u
m
m
ar
ize
d
etec
tio
n
ac
cu
r
ac
y
o
f
d
if
f
er
en
t
alg
o
r
ith
m
s
im
p
le
m
en
ted
f
o
r
o
b
ject
class
if
icatio
n
.
I
t
i
s
ev
i
d
en
t
th
at
th
e
YOL
O
alg
o
r
ith
m
p
r
o
v
id
es
th
e
b
est
p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
all
o
b
ject
ty
p
es,
ac
h
iev
in
g
h
ig
h
ac
c
u
r
ac
y
,
esp
ec
ially
f
o
r
o
b
jects
with
d
is
tin
ct
s
h
ap
es.
T
h
e
Ho
p
f
iel
d
alg
o
r
ith
m
is
n
o
t
r
ec
o
m
m
en
d
ed
f
o
r
ap
p
licatio
n
s
wh
er
e
o
b
ject
s
h
ap
es
ar
e
v
er
y
s
im
ilar
,
as
its
p
er
f
o
r
m
an
ce
i
s
h
ig
h
ly
d
ep
e
n
d
en
t
o
n
th
e
c
h
o
s
en
f
ilter
a
n
d
n
u
m
b
er
o
f
p
ix
els
u
s
ed
.
T
h
e
alg
o
r
i
th
m
’
s
ac
cu
r
ac
y
ten
d
s
to
d
eg
r
ad
e
with
h
ig
h
p
ix
el
v
alu
es,
as
it
lo
s
es
co
n
v
er
g
en
c
e,
an
d
with
lo
w
p
ix
el
v
alu
es,
t
h
e
q
u
ality
o
f
th
e
ed
g
e
d
etec
ti
o
n
is
co
m
p
r
o
m
is
ed
.
W
h
ile
it
d
o
es
n
o
t
r
eq
u
ir
e
tr
ai
n
in
g
,
o
n
ly
th
e
in
itial
d
e
f
in
itio
n
o
f
p
atter
n
s
,
its
ex
ec
u
tio
n
ti
m
e
av
er
ag
e
d
2
3
7
m
s
f
o
r
th
e
im
a
g
es u
s
ed
,
m
a
k
in
g
it
r
elativ
ely
f
ast b
u
t le
s
s
ef
f
ec
ti
v
e
f
o
r
co
m
p
le
x
o
b
ject
s
h
ap
es.
T
ab
le
5
.
Dete
ctio
n
ac
c
u
r
ac
y
c
o
m
p
ar
is
o
n
o
f
alg
o
r
ith
m
s
f
o
r
cl
ass
if
y
in
g
o
b
jects
D
e
t
e
c
t
i
o
n
a
c
c
u
r
a
c
y
o
f
a
l
g
o
r
i
t
h
ms f
o
r
a
n
y
o
b
j
e
c
t
H
o
p
f
i
e
l
d
a
l
g
o
r
i
t
h
m w
i
t
h
S
o
b
e
l
f
i
l
t
e
r
2
D
C
r
o
ss
-
c
o
r
r
e
l
a
t
i
o
n
w
i
t
h
S
o
b
e
l
f
i
l
t
e
r
Y
O
LO
v
2
a
l
g
o
r
i
t
h
m
5
0
0
cc
3
7
.
8
8
%
9
0
.
9
1
%
9
7
.
0
2
%
8
0
0
cc
4
5
.
4
5
%
6
2
.
1
2
%
9
5
.
2
0
%
½
g
a
l
2
5
.
2
5
%
4
0
.
9
1
%
8
6
.
1
7
%
1
g
a
l
3
5
.
3
5
%
8
1
.
0
6
%
9
2
.
1
7
%
5
g
a
l
3
7
.
8
8
%
4
4
.
9
5
%
8
2
.
0
7
%
M
e
a
n
3
6
.
3
6
%
6
3
.
9
9
%
9
0
.
5
3
%
T
h
e
2
D
c
r
o
s
s
-
co
r
r
elatio
n
m
etr
ic
was
p
r
o
p
o
s
ed
as
a
s
im
p
le
a
lter
n
ativ
e
f
o
r
o
b
ject
class
if
icatio
n
.
T
h
is
m
eth
o
d
p
e
r
f
o
r
m
ed
well
f
o
r
cy
lin
d
r
ical
o
b
jects,
as
th
eir
o
r
ie
n
tatio
n
d
o
es
n
o
t
s
ig
n
if
ica
n
tly
al
ter
th
eir
s
ilh
o
u
ette.
Ho
wev
er
,
it
is
n
o
t
r
ec
o
m
m
e
n
d
ed
f
o
r
s
o
lid
s
with
m
o
r
e
c
o
m
p
lex
g
eo
m
etr
ies.
Desp
ite
b
ein
g
th
e
f
astes
t
in
ex
ec
u
tio
n
,
with
a
p
r
o
ce
s
s
in
g
tim
e
o
f
ju
s
t
1
8
m
s
,
its
ap
p
licati
o
n
is
lim
ited
to
s
im
p
ler
s
h
ap
es
.
Giv
en
lim
itatio
n
s
o
f
th
e
o
t
h
er
al
g
o
r
ith
m
s
,
Y
OL
O
v
2
alg
o
r
ith
m
was
c
h
o
s
en
.
I
t
le
v
er
ag
es
a
p
r
e
-
tr
ain
ed
d
ata
b
ase
o
f
th
e
en
v
ir
o
n
m
en
t
in
wh
ic
h
o
b
jects
will
b
e
u
s
ed
,
allo
win
g
f
o
r
f
o
cu
s
ed
tr
ain
in
g
o
n
o
b
ject
d
etec
tio
n
.
Alth
o
u
g
h
th
e
tr
ain
in
g
p
r
o
ce
s
s
is
tim
e
-
co
n
s
u
m
in
g
,
tak
i
n
g
1
9
3
m
in
u
tes,
an
d
its
im
p
lem
en
tatio
n
is
s
lo
wer
th
an
th
e
o
t
h
er
s
,
with
an
av
er
a
g
e
ex
ec
u
tio
n
ti
m
e
o
f
8
2
5
m
s
,
th
e
h
ig
h
ac
cu
r
ac
y
in
d
etec
tio
n
m
a
k
es
it
th
e
b
est
ch
o
ice
f
o
r
th
is
ap
p
licatio
n
.
Giv
en
th
at
th
e
p
r
im
a
r
y
o
b
je
ctiv
e
o
f
th
is
wo
r
k
was
th
e
class
if
icatio
n
o
f
o
b
jects
b
ased
o
n
th
eir
ap
p
ea
r
an
ce
,
r
ath
er
th
an
p
r
ec
is
e
lo
ca
lizatio
n
o
r
m
u
lti
-
o
b
jec
t
d
etec
tio
n
,
th
e
f
o
cu
s
was
p
lace
d
o
n
co
m
p
a
r
in
g
b
asic
y
et
r
e
p
r
esen
tativ
e
class
if
icatio
n
s
tr
ateg
ies.
T
h
is
c
h
o
ice
was
alig
n
ed
with
th
e
s
im
p
licity
o
f
th
e
s
im
u
lated
in
d
u
s
tr
ial
s
ce
n
ar
io
an
d
th
e
co
n
tr
o
lled
co
n
d
itio
n
s
o
f
t
h
e
task
.
I
n
th
is
co
n
te
x
t,
th
e
im
p
lem
e
n
tatio
n
o
f
m
eth
o
d
s
s
u
ch
as
SUR
F,
2
D
cr
o
s
s
-
co
r
r
elatio
n
,
an
d
YOL
O
c
o
n
f
ig
u
r
e
d
m
ain
ly
f
o
r
s
in
g
le
-
o
b
ject
class
if
icatio
n
—
p
r
o
v
ed
ef
f
ec
tiv
ely
.
T
h
e
r
esu
lts
o
b
tain
ed
v
alid
ate
th
e
a
p
p
licab
ilit
y
o
f
th
ese
s
tr
ateg
ies
in
s
tr
u
ctu
r
ed
en
v
ir
o
n
m
en
ts
,
with
th
e
ad
d
e
d
b
en
ef
it
o
f
r
ap
id
ex
e
cu
tio
n
tim
es.
W
h
ile
m
o
r
e
ad
v
an
ce
d
d
ee
p
lea
r
n
in
g
m
o
d
els
c
o
u
ld
o
f
f
er
im
p
r
o
v
e
d
p
er
f
o
r
m
an
ce
in
m
o
r
e
c
o
m
p
lex
s
ce
n
es,
th
e
cu
r
r
en
t
ap
p
r
o
ac
h
p
r
o
v
id
es
a
r
eliab
le
an
d
c
o
m
p
u
tatio
n
ally
ef
f
icien
t
s
o
lu
tio
n
s
u
itab
le
f
o
r
p
r
o
t
o
ty
p
i
n
g
an
d
ex
p
e
r
im
en
tatio
n
in
s
em
i
-
co
n
tr
o
lled
s
ettin
g
s
.
Af
ter
in
teg
r
atin
g
all
th
e
alg
o
r
ith
m
s
,
th
e
s
im
u
latio
n
en
v
ir
o
n
m
en
t
was
u
p
d
ated
with
th
e
co
r
r
esp
o
n
d
in
g
3
D
m
o
d
els
o
f
th
e
co
n
tain
er
s
,
en
ab
lin
g
a
co
m
p
lete
test
o
f
th
e
d
etec
tio
n
,
class
if
icatio
n
,
an
d
m
an
ip
u
latio
n
p
ip
elin
e.
Fig
u
r
e
9
illu
s
tr
ates
th
e
f
in
al
s
etu
p
,
wh
er
e
th
e
UR
5
r
o
b
o
t
id
en
tifie
s
ea
ch
o
b
je
ct
u
s
in
g
th
e
v
is
io
n
s
y
s
tem
,
class
if
ie
s
it b
ased
o
n
th
e
tr
ain
ed
m
o
d
els,
an
d
ex
ec
u
t
es a
p
ick
-
an
d
-
p
lace
task
ac
co
r
d
in
g
to
th
e
ass
ig
n
ed
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
R
o
b
o
tic
p
r
o
d
u
ct
-
b
a
s
ed
ma
n
ip
u
la
tio
n
in
s
imu
la
ted
e
n
viro
n
m
en
t
(
Ju
a
n
C
a
milo
Gu
a
c
h
eta
-
A
l
b
a
)
5901
ca
teg
o
r
y
.
T
h
is
in
teg
r
atio
n
c
o
n
f
ir
m
s
th
e
c
o
h
er
e
n
ce
b
etwe
e
n
v
is
u
al
r
ec
o
g
n
itio
n
an
d
r
o
b
o
tic
co
n
tr
o
l
with
in
a
f
u
lly
s
im
u
lated
in
d
u
s
tr
ial
s
ce
n
ar
io
.
Fig
u
r
e
9
.
Simu
latio
n
en
v
ir
o
n
m
en
t w
ith
in
teg
r
ated
s
o
lid
s
f
o
r
tr
an
s
p
o
r
t a
n
d
h
a
n
d
lin
g
4.
CO
NCLU
SI
O
N
T
h
is
wo
r
k
d
em
o
n
s
tr
ated
v
ia
b
ilit
y
o
f
co
m
b
in
in
g
class
ical
im
ag
e
p
r
o
ce
s
s
in
g
m
eth
o
d
s
with
d
ee
p
lear
n
in
g
tech
n
iq
u
es
f
o
r
o
b
ject
class
if
icatio
n
an
d
r
o
b
o
tic
m
a
n
ip
u
latio
n
in
a
s
im
u
lated
in
d
u
s
tr
ial
co
n
tex
t.
W
h
ile
tr
ad
itio
n
al
ap
p
r
o
ac
h
es
s
u
ch
a
s
SU
R
F,
ed
g
e
f
ilter
s
,
Ho
p
f
ield
n
etwo
r
k
s
,
an
d
c
r
o
s
s
-
co
r
r
el
atio
n
o
f
f
er
e
d
f
ast
ex
ec
u
tio
n
an
d
s
im
p
licity
,
th
ey
s
h
o
wed
clea
r
lim
itatio
n
s
in
ac
cu
r
ac
y
an
d
ad
ap
tab
ilit
y
wh
e
n
h
an
d
lin
g
co
m
p
le
x
s
h
ap
es
o
r
n
o
is
y
en
v
ir
o
n
m
en
t
s
.
I
n
co
n
tr
ast,
YOL
O
v
2
m
o
d
el
ex
h
ib
ited
s
ig
n
if
ican
tly
h
i
g
h
er
class
if
icatio
n
p
er
f
o
r
m
an
ce
an
d
r
o
b
u
s
tn
ess
.
Giv
en
th
e
s
im
p
licity
o
f
th
e
s
im
u
lated
s
etu
p
an
d
th
e
g
o
al
o
f
co
m
p
ar
in
g
class
if
icatio
n
s
tr
ateg
ies
r
ath
er
th
an
f
u
ll
o
b
ject
d
etec
tio
n
p
ip
elin
es,
th
e
a
p
p
lied
m
eth
o
d
s
p
r
o
v
ed
e
f
f
ec
tiv
e
,
co
n
f
ir
m
in
g
th
at
b
asic
tech
n
iq
u
es
ca
n
b
e
u
s
ef
u
l
f
o
r
s
tr
u
ctu
r
ed
en
v
ir
o
n
m
en
ts
wh
er
e
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
ar
e
lim
ited
.
T
o
s
tr
en
g
th
en
co
n
tr
ib
u
tio
n
an
d
s
itu
ate
it
with
in
th
e
cu
r
r
en
t
s
tate
o
f
th
e
a
r
t,
f
u
tu
r
e
wo
r
k
m
u
s
t
in
clu
d
e
co
m
p
ar
is
o
n
s
with
m
o
r
e
r
ec
en
t
p
r
etr
ai
n
ed
m
o
d
els
s
u
ch
as
YOL
O
v
5
,
Fas
ter
R
-
C
N
N,
o
r
E
f
f
icien
tDet.
T
h
ese
m
o
d
els
ar
e
lik
el
y
to
o
f
f
er
im
p
r
o
v
ed
p
er
f
o
r
m
an
ce
in
t
er
m
s
o
f
g
en
er
aliza
tio
n
an
d
d
e
tectio
n
u
n
d
er
m
o
r
e
ch
allen
g
in
g
co
n
d
itio
n
s
.
Mo
r
e
o
v
er
,
e
x
p
an
d
in
g
th
e
d
ataset,
test
in
g
with
r
ea
l
-
wo
r
ld
s
en
s
o
r
s
,
an
d
e
v
alu
atin
g
ag
ain
s
t
s
tan
d
ar
d
ized
b
en
ch
m
ar
k
s
will
b
e
es
s
en
tial
to
v
alid
ate
s
y
s
tem
’
s
s
ca
lab
ilit
y
a
n
d
r
elev
an
ce
.
T
h
e
m
eth
o
d
o
l
o
g
y
la
y
s
a
r
ep
lica
b
le
f
o
u
n
d
atio
n
f
o
r
in
teg
r
atin
g
co
m
p
u
ter
v
is
io
n
an
d
r
o
b
o
tic
co
n
tr
o
l
in
I
n
d
u
s
tr
y
5
.
0
-
in
s
p
ir
ed
en
v
ir
o
n
m
en
ts
an
d
h
ig
h
lig
h
ts
im
p
o
r
tan
ce
o
f
s
tr
ateg
ie
s
th
at
co
n
s
id
er
b
o
th
ac
cu
r
ac
y
an
d
co
m
p
u
tatio
n
al
ef
f
icien
cy
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
s
wo
u
ld
lik
e
to
th
a
n
k
th
e
Un
iv
e
r
s
id
ad
Militar
Nu
ev
a
Gr
an
a
d
a,
f
r
o
m
wh
er
e
t
h
ey
ar
e
f
u
ll
-
tim
e
ass
o
ciate
p
r
o
f
ess
o
r
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Pro
d
u
ct
d
er
iv
e
d
f
r
o
m
th
e
r
ese
ar
ch
p
r
o
ject
titl
ed
"
Mejo
r
a
mien
to
d
e
p
r
o
ce
s
o
s
d
e
co
n
tr
o
l
d
e
in
ve
n
ta
r
io
y
a
lis
ta
mien
to
d
e
p
e
d
id
o
s
en
el
ma
r
co
d
e
la
i
n
d
u
s
tr
ia
4
.
0
"
I
NV
-
I
NG
-
3
9
7
0
f
in
a
n
ce
d
b
y
th
e
v
ice
-
r
ec
to
r
f
o
r
r
esear
ch
o
f
t
h
e
Un
iv
er
s
id
ad
Militar
Nu
ev
a
Gr
an
ad
a,
y
ea
r
2
0
2
4
.
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
J
u
an
C
am
ilo
Gu
ac
h
eta
-
Alb
a
✓
✓
✓
✓
✓
✓
✓
✓
An
n
y
Ast
r
id
E
s
p
itia
-
C
u
b
illo
s
✓
✓
✓
✓
✓
✓
✓
R
o
b
in
s
o
n
J
im
én
ez
-
Mo
r
en
o
✓
✓
✓
✓
✓
✓
✓
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
8
9
4
-
5
9
0
3
5902
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
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
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
i
n
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
aila
b
le
o
n
r
eq
u
est
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
au
th
o
r
.
RE
F
E
R
E
NC
E
S
[
1
]
H.
-
W
.
Le
e
,
M
.
-
I
.
R
o
h
,
Y
.
-
M
.
C
h
o
,
a
n
d
J.
-
H
.
P
a
r
k
,
“
D
a
t
a
a
sso
c
i
a
t
i
o
n
f
o
r
a
u
t
o
n
o
mo
u
s
s
h
i
p
s
b
a
se
d
o
n
v
i
r
t
u
a
l
s
i
mu
l
a
t
i
o
n
e
n
v
i
r
o
n
m
e
n
t
,
”
O
c
e
a
n
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
2
8
1
,
p
.
1
1
4
6
4
6
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
o
c
e
a
n
e
n
g
.
2
0
2
3
.
1
1
4
6
4
6
.
[
2
]
M
.
M
.
A
l
a
m
a
n
d
S
.
M
.
M
.
R
a
h
m
a
n
,
“
A
f
f
i
n
e
t
r
a
n
sf
o
r
mat
i
o
n
o
f
v
i
r
t
u
a
l
3
D
o
b
j
e
c
t
u
s
i
n
g
2
D
l
o
c
a
l
i
z
a
t
i
o
n
o
f
f
i
n
g
e
r
t
i
p
s
,
”
V
i
rt
u
a
l
Re
a
l
i
t
y
& In
t
e
l
l
i
g
e
n
t
H
a
r
d
w
a
re
,
v
o
l
.
2
,
n
o
.
6
,
p
p
.
5
3
4
–
5
5
5
,
D
e
c
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
v
r
i
h
.
2
0
2
0
.
1
0
.
0
0
1
.
[
3
]
H
.
L
i
e
t
a
l
.
,
“
V
i
r
t
u
a
l
p
o
i
n
t
t
r
a
c
k
i
n
g
m
e
t
h
o
d
f
o
r
o
n
l
i
n
e
d
e
t
e
c
t
i
o
n
o
f
r
e
l
a
t
i
v
e
w
h
e
e
l
-
r
a
i
l
d
i
sp
l
a
c
e
m
e
n
t
o
f
r
a
i
l
w
a
y
v
e
h
i
c
l
e
s,”
Re
l
i
a
b
i
l
i
t
y
E
n
g
i
n
e
e
r
i
n
g
&
S
y
st
e
m
S
a
f
e
t
y
,
v
o
l
.
2
4
6
,
p
.
1
1
0
0
8
7
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
e
ss
.
2
0
2
4
.
1
1
0
0
8
7
.
[
4
]
Y
.
D
i
n
g
a
n
d
X
.
L
u
o
,
“
A
v
i
r
t
u
a
l
c
o
n
st
r
u
c
t
i
o
n
v
e
h
i
c
l
e
s
a
n
d
w
o
r
k
e
r
s
d
a
t
a
set
w
i
t
h
t
h
r
e
e
-
d
i
me
n
s
i
o
n
a
l
a
n
n
o
t
a
t
i
o
n
s,”
En
g
i
n
e
e
ri
n
g
Ap
p
l
i
c
a
t
i
o
n
s
o
f
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
3
3
,
p
.
1
0
7
9
6
4
,
J
u
l
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
n
g
a
p
p
a
i
.
2
0
2
4
.
1
0
7
9
6
4
.
[
5
]
H
.
Zh
a
n
g
,
T
.
Li
,
Y
.
L
i
,
a
n
d
Z.
W
e
n
,
“
V
i
r
t
u
a
l
e
l
e
c
t
r
o
m
a
g
n
e
t
i
c
e
n
v
i
r
o
n
m
e
n
t
m
o
d
e
l
i
n
g
b
a
s
e
d
d
a
t
a
a
u
g
m
e
n
t
a
t
i
o
n
f
o
r
d
r
o
n
e
s
i
g
n
a
l
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
a
n
d
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
,
n
o
.
4
,
p
p
.
3
0
8
–
3
2
0
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
i
i
x
d
.
2
0
2
3
.
0
8
.
0
0
2
.
[
6
]
Y
.
S
h
a
h
e
d
i
,
M
.
Z
a
n
d
i
,
a
n
d
M
.
B
i
mak
r
,
“
A
c
o
m
p
u
t
e
r
v
i
s
i
o
n
s
y
s
t
e
m
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
f
o
r
p
r
e
d
i
c
t
i
o
n
o
f
p
h
y
s
i
c
o
c
h
e
m
i
c
a
l
c
h
a
n
g
e
s
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
c
o
a
t
e
d
sw
e
e
t
c
h
e
r
r
y
,
”
H
e
l
i
y
o
n
,
v
o
l
.
1
0
,
n
o
.
2
0
,
p
.
e
3
9
4
8
4
,
O
c
t
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
2
4
.
e
3
9
4
8
4
.
[
7
]
M
.
D
u
t
t
a
a
n
d
A
.
G
a
n
g
u
l
y
,
“
I
n
c
r
e
men
t
a
l
-
b
a
s
e
d
Y
O
LO
V
3
m
o
d
e
l
w
i
t
h
h
y
p
e
r
-
p
a
r
a
m
e
t
e
r
o
p
t
i
mi
z
a
t
i
o
n
f
o
r
p
r
o
d
u
c
t
i
m
a
g
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
i
n
E
-
c
o
mm
e
r
c
e
se
c
t
o
r
,
”
Ap
p
l
i
e
d
S
o
f
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
6
5
,
p
.
1
1
2
0
2
9
,
N
o
v
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
s
o
c
.
2
0
2
4
.
1
1
2
0
2
9
.
[
8
]
J.
Le
n
g
e
t
a
l
.
,
“
U
n
l
o
c
k
i
n
g
t
h
e
p
o
w
e
r
o
f
i
n
d
u
s
t
r
i
a
l
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
t
o
w
a
r
d
s
I
n
d
u
s
t
r
y
5
.
0
:
i
n
s
i
g
h
t
s,
p
a
t
h
w
a
y
s
,
a
n
d
c
h
a
l
l
e
n
g
e
s,”
J
o
u
rn
a
l
o
f
M
a
n
u
f
a
c
t
u
ri
n
g
S
y
st
e
m
s
,
v
o
l
.
7
3
,
p
p
.
3
4
9
–
3
6
3
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
ms
y
.
2
0
2
4
.
0
2
.
0
1
0
.
[
9
]
M
.
Jasi
u
l
e
w
i
c
z
-
K
a
c
z
mar
e
k
a
n
d
A
.
G
o
l
a
,
“
M
a
i
n
t
e
n
a
n
c
e
4
.
0
t
e
c
h
n
o
l
o
g
i
e
s
f
o
r
s
u
st
a
i
n
a
b
l
e
m
a
n
u
f
a
c
t
u
r
i
n
g
-
a
n
o
v
e
r
v
i
e
w
,
”
I
F
AC
-
Pa
p
e
rsO
n
L
i
n
e
,
v
o
l
.
5
2
,
n
o
.
1
0
,
p
p
.
9
1
–
9
6
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
f
a
c
o
l
.
2
0
1
9
.
1
0
.
0
0
5
.
[
1
0
]
E.
V
a
l
e
t
t
e
,
H
.
B
r
i
l
E
l
-
H
a
o
u
z
i
,
a
n
d
G
.
D
e
mes
u
r
e
,
“
I
n
d
u
st
r
y
5
.
0
a
n
d
i
t
s t
e
c
h
n
o
l
o
g
i
e
s
:
a
s
y
s
t
e
m
a
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
u
p
o
n
t
h
e
h
u
m
a
n
p
l
a
c
e
i
n
t
o
I
o
T
-
a
n
d
C
P
S
-
b
a
se
d
i
n
d
u
s
t
r
i
a
l
s
y
st
e
ms,
”
C
o
m
p
u
t
e
rs
&
I
n
d
u
s
t
ri
a
l
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
8
4
,
p
.
1
0
9
4
2
6
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
i
e
.
2
0
2
3
.
1
0
9
4
2
6
.
[
1
1
]
K
.
W
a
n
,
C
.
L
i
,
F
.
-
S
.
L
o
,
a
n
d
P
.
Z
h
e
n
g
,
“
A
v
i
r
t
u
a
l
r
e
a
l
i
t
y
-
b
a
se
d
i
m
m
e
r
si
v
e
t
e
l
e
o
p
e
r
a
t
i
o
n
s
y
s
t
e
m
f
o
r
r
e
m
o
t
e
h
u
ma
n
-
r
o
b
o
t
c
o
l
l
a
b
o
r
a
t
i
v
e
m
a
n
u
f
a
c
t
u
r
i
n
g
,
”
M
a
n
u
f
a
c
t
u
r
i
n
g
L
e
t
t
e
rs
,
v
o
l
.
4
1
,
p
p
.
4
3
–
5
0
,
O
c
t
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
mf
g
l
e
t
.
2
0
2
4
.
0
9
.
0
0
8
.
[
1
2
]
M
.
A
n
d
e
r
ss
o
n
a
n
d
A
.
S
y
b
e
r
f
e
l
d
t
,
“
I
m
p
r
o
v
e
d
i
n
t
e
r
a
c
t
i
o
n
w
i
t
h
c
o
l
l
a
b
o
r
a
t
i
v
e
r
o
b
o
t
s
-
e
v
a
l
u
a
t
i
o
n
o
f
e
v
e
n
t
-
s
p
e
c
i
f
i
c
h
a
p
t
i
c
f
e
e
d
b
a
c
k
i
n
v
i
r
t
u
a
l
r
e
a
l
i
t
y
,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
3
2
,
p
p
.
1
0
5
5
–
1
0
6
4
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
4
.
0
1
.
1
0
4
.
[
1
3
]
S
.
M
u
s
t
a
r
y
,
M
.
A
.
K
a
s
h
e
m
,
M
.
A
.
C
h
o
w
d
h
u
r
y
,
a
n
d
M
.
M
.
R
a
n
a
,
“
M
a
t
h
e
mat
i
c
a
l
m
o
d
e
l
a
n
d
e
v
a
l
u
a
t
i
o
n
o
f
d
y
n
a
mi
c
st
a
b
i
l
i
t
y
o
f
i
n
d
u
st
r
i
a
l
r
o
b
o
t
m
a
n
i
p
u
l
a
t
o
r
:
u
n
i
v
e
r
sal
r
o
b
o
t
,
”
S
y
s
t
e
m
s
a
n
d
S
o
f
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
6
,
p
.
2
0
0
0
7
1
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
sas
c
.
2
0
2
3
.
2
0
0
0
7
1
.
[
1
4
]
T.
T
o
n
e
r
,
M
.
S
a
e
z
,
D
.
M
.
T
i
l
b
u
r
y
,
a
n
d
K
.
B
a
r
t
o
n
,
“
O
p
p
o
r
t
u
n
i
t
i
e
s
a
n
d
c
h
a
l
l
e
n
g
e
s
i
n
a
p
p
l
y
i
n
g
r
e
i
n
f
o
r
c
e
m
e
n
t
l
e
a
r
n
i
n
g
t
o
r
o
b
o
t
i
c
man
i
p
u
l
a
t
i
o
n
:
a
n
i
n
d
u
s
t
r
i
a
l
c
a
s
e
st
u
d
y
,
”
M
a
n
u
f
a
c
t
u
ri
n
g
L
e
t
t
e
rs
,
v
o
l
.
3
5
,
p
p
.
1
0
1
9
–
1
0
3
0
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
mf
g
l
e
t
.
2
0
2
3
.
0
8
.
0
5
5
.
[
1
5
]
A
.
Ef
e
n
d
i
,
Y
.
-
H
.
S
h
a
o
,
a
n
d
C
.
-
Y
.
H
u
a
n
g
,
“
T
e
c
h
n
o
l
o
g
i
c
a
l
d
e
v
e
l
o
p
m
e
n
t
a
n
d
o
p
t
i
mi
z
a
t
i
o
n
o
f
p
u
sh
i
n
g
a
n
d
g
r
a
s
p
i
n
g
f
u
n
c
t
i
o
n
s
i
n
r
o
b
o
t
a
r
ms:
a
r
e
v
i
e
w
,
”
Me
a
su
reme
n
t
,
v
o
l
.
2
4
2
,
p
.
1
1
5
7
2
9
,
J
a
n
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
a
s
u
r
e
me
n
t
.
2
0
2
4
.
1
1
5
7
2
9
.
[
1
6
]
V
.
P
r
i
y
a
n
k
a
,
Y
.
S
.
R
a
m
a
,
K
.
S
r
a
v
a
n
i
,
a
n
d
B
.
K
a
v
y
a
,
“
I
mp
l
e
m
e
n
t
a
t
i
o
n
o
f
s
o
b
e
l
e
d
g
e
d
e
t
e
c
t
i
o
n
w
i
t
h
i
m
a
g
e
p
r
o
c
e
ssi
n
g
o
n
F
P
G
A
,
”
i
n
2
0
2
4
2
n
d
W
o
rl
d
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
&
C
o
m
p
u
t
i
n
g
(
W
C
O
N
F)
,
J
u
l
.
2
0
2
4
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
W
C
O
N
F
6
1
3
6
6
.
2
0
2
4
.
1
0
6
9
2
3
0
1
.
[
1
7
]
L.
L.
V
a
i
s
h
n
a
v
i
,
M
.
B
h
a
v
a
n
a
,
D
.
C
.
M
.
M
a
n
i
d
e
e
p
,
N
.
N
,
a
n
d
V
.
V
e
n
u
g
o
p
a
l
,
“
A
f
a
st
a
n
d
a
c
c
u
r
a
t
e
o
b
j
e
c
t
c
o
u
n
t
i
n
g
u
s
i
n
g
s
o
b
e
l
e
d
g
e
d
e
t
e
c
t
i
o
n
s
y
st
e
m
f
o
r
r
e
a
l
–
t
i
m
e
st
a
r
g
a
z
i
n
g
,
”
i
n
2
0
2
4
Asi
a
P
a
c
i
f
i
c
C
o
n
f
e
r
e
n
c
e
o
n
I
n
n
o
v
a
t
i
o
n
i
n
T
e
c
h
n
o
l
o
g
y
(
A
PC
I
T
)
,
J
u
l
.
2
0
2
4
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
A
P
C
I
T6
2
0
0
7
.
2
0
2
4
.
1
0
6
7
3
4
6
2
.
[
1
8
]
H
.
G
a
o
,
X
.
P
e
n
g
,
X
.
P
a
n
g
,
a
n
d
X
.
F
e
n
g
,
“
A
n
e
f
f
i
c
i
e
n
t
t
a
r
g
e
t
r
e
c
o
g
n
i
t
i
o
n
met
h
o
d
b
a
s
e
d
o
n
m
i
x
e
d
b
i
n
a
r
y
S
U
R
F
d
e
s
c
r
i
p
t
o
r
w
i
t
h
r
e
g
i
o
n
s
e
g
m
e
n
t
a
t
i
o
n
,
”
i
n
2
0
2
4
4
3
r
d
C
h
i
n
e
se
C
o
n
t
r
o
l
C
o
n
f
e
r
e
n
c
e
(
C
C
C
)
,
Ju
l
.
2
0
2
4
,
p
p
.
7
3
3
9
–
7
3
4
4
,
d
o
i
:
1
0
.
2
3
9
1
9
/
C
C
C
6
3
1
7
6
.
2
0
2
4
.
1
0
6
6
2
6
0
5
.
[
1
9
]
L.
O
u
y
a
n
g
,
R
.
Lu
,
Y
.
Y
e
,
M
.
X
i
n
g
,
a
n
d
Q
.
C
a
i
,
“
F
e
a
t
u
r
e
p
o
i
n
t
e
x
t
r
a
c
t
i
o
n
a
n
d
ma
t
c
h
i
n
g
b
a
se
d
o
n
i
m
p
r
o
v
e
d
S
U
R
F
a
l
g
o
r
i
t
h
m,
”
i
n
2
0
2
3
C
h
i
n
a
Au
t
o
m
a
t
i
o
n
C
o
n
g
r
e
ss (C
AC
)
,
N
o
v
.
2
0
2
3
,
p
p
.
1
8
4
6
–
1
8
5
1
,
d
o
i
:
1
0
.
1
1
0
9
/
C
A
C
5
9
5
5
5
.
2
0
2
3
.
1
0
4
5
0
3
5
7
.
[
2
0
]
R
.
H
a
mza
h
,
L.
A
n
g
,
R
.
R
o
s
l
a
n
,
N
.
H
.
I
b
r
a
h
i
m
T
e
o
,
K
.
A
b
d
u
l
S
a
ma
d
,
a
n
d
K
.
A
.
F
a
r
i
z
a
A
b
u
S
a
ma
h
,
“
C
o
mp
a
r
i
n
g
m
o
d
i
f
i
e
d
Y
o
l
o
V
5
a
n
d
f
a
st
e
r
r
e
g
i
o
n
a
l
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
p
e
r
f
o
r
m
a
n
c
e
f
o
r
r
e
c
y
c
l
e
w
a
st
e
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
i
n
2
0
2
4
I
EEE
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
u
t
o
m
a
t
i
c
C
o
n
t
ro
l
a
n
d
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
(
I
2
C
AC
I
S
)
,
J
u
n
.
2
0
2
4
,
p
p
.
4
1
5
–
4
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
2
C
A
C
I
S
6
1
2
7
0
.
2
0
2
4
.
1
0
6
4
9
8
3
5
.
[
2
1
]
V
.
S
.
M
u
r
a
v
i
e
v
a
n
d
P
.
E.
Z
h
g
u
t
o
v
,
“
Th
e
i
n
f
l
u
e
n
c
e
o
f
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
o
n
t
h
e
a
c
c
u
r
a
c
y
o
f
o
b
j
e
c
t
l
o
c
a
l
i
z
a
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
t
h
e
Y
O
LO
d
e
t
e
c
t
o
r
,
”
i
n
2
0
2
4
2
6
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
D
i
g
i
t
a
l
S
i
g
n
a
l
Pro
c
e
ss
i
n
g
a
n
d
i
t
s
Ap
p
l
i
c
a
t
i
o
n
s
(
D
S
PA)
,
M
a
r
.
2
0
2
4
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
D
S
P
A
6
0
8
5
3
.
2
0
2
4
.
1
0
5
1
0
0
6
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
R
o
b
o
tic
p
r
o
d
u
ct
-
b
a
s
ed
ma
n
ip
u
la
tio
n
in
s
imu
la
ted
e
n
viro
n
m
en
t
(
Ju
a
n
C
a
milo
Gu
a
c
h
eta
-
A
l
b
a
)
5903
[
2
2
]
A
.
B
r
e
y
,
J.
J.
Q
u
i
n
t
a
n
a
,
M
.
D
i
a
z
,
a
n
d
M
.
A
.
F
e
r
r
e
r
,
“
S
mart
p
h
o
n
e
-
b
a
s
e
d
c
o
n
t
r
o
l
sy
st
e
m
f
o
r
u
n
i
v
e
r
s
a
l
r
o
b
o
t
U
R
5
e
:
a
t
o
o
l
f
o
r
r
o
b
o
t
i
c
s
e
d
u
c
a
t
i
o
n
,
”
i
n
2
0
2
4
XVI
C
o
n
g
res
o
d
e
T
e
c
n
o
l
o
g
í
a
,
Ap
re
n
d
i
za
j
e
y
E
n
se
ñ
a
n
z
a
d
e
l
a
El
e
c
t
r
ó
n
i
c
a
(
T
A
EE)
,
J
u
n
.
2
0
2
4
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
TA
EE
5
9
5
4
1
.
2
0
2
4
.
1
0
6
0
4
9
9
4
.
[
2
3
]
J.
V
i
l
l
a
l
o
b
o
s,
I
.
Y
.
S
a
n
c
h
e
z
,
a
n
d
F
.
M
a
r
t
e
l
l
,
“
S
t
a
t
i
st
i
c
a
l
c
o
m
p
a
r
i
s
o
n
o
f
D
e
n
a
v
i
t
-
H
a
r
t
e
n
b
e
r
g
b
a
se
d
i
n
v
e
r
se
k
i
n
e
mat
i
c
s
o
l
u
t
i
o
n
s
o
f
t
h
e
U
R
5
r
o
b
o
t
i
c
m
a
n
i
p
u
l
a
t
o
r
,
”
i
n
2
0
2
1
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
E
l
e
c
t
ri
c
a
l
,
C
o
m
p
u
t
e
r,
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
Me
c
h
a
t
r
o
n
i
c
s
En
g
i
n
e
e
ri
n
g
(
I
C
E
C
C
ME
)
,
O
c
t
.
2
0
2
1
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
EC
C
M
E5
2
2
0
0
.
2
0
2
1
.
9
5
9
1
1
0
4
.
[
2
4
]
I
n
d
u
st
r
i
a
s
N
o
v
a
q
u
i
m
S
A
S
,
“
C
l
e
a
n
i
n
g
p
r
o
d
u
c
t
s
t
a
i
l
o
r
e
d
t
o
y
o
u
r
n
e
e
d
s
,
”
(
i
n
S
p
a
n
i
sh
)
.
I
n
d
u
st
r
i
a
s
N
o
v
a
q
u
i
m
S
A
S
,
h
t
t
p
s
:
/
/
w
w
w
.
n
o
v
a
q
u
i
m.
c
o
m/
p
r
o
d
u
c
t
o
s/
c
a
t
e
g
o
r
i
a
/
l
i
mp
i
e
z
a
-
g
e
n
e
r
a
l
/
(
a
c
c
e
ss
e
d
J
u
l
.
0
2
,
2
0
2
5
)
.
[
2
5
]
D
.
R
.
D
.
V
a
r
m
a
a
n
d
R
.
P
r
i
y
a
n
k
a
,
“
P
e
r
f
o
r
ma
n
c
e
m
o
n
i
t
o
r
i
n
g
o
f
n
o
v
e
l
i
r
i
s
d
e
t
e
c
t
i
o
n
s
y
st
e
m
u
si
n
g
s
o
b
e
l
a
l
g
o
r
i
t
h
m
i
n
c
o
m
p
a
r
i
s
o
n
w
i
t
h
c
a
n
n
y
a
l
g
o
r
i
t
h
m b
y
m
i
n
i
mi
z
i
n
g
t
h
e
m
e
a
n
sq
u
a
r
e
e
r
r
o
r
,
”
i
n
2
0
2
2
3
r
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
t
e
l
l
i
g
e
n
t
En
g
i
n
e
e
ri
n
g
a
n
d
Ma
n
a
g
e
m
e
n
t
(
I
C
I
E
M)
,
A
p
r
.
2
0
2
2
,
p
p
.
5
0
9
–
5
1
2
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
EM
5
4
2
2
1
.
2
0
2
2
.
9
8
5
3
1
2
7
.
[
2
6
]
H
.
L
i
u
,
“
F
i
n
e
-
g
r
a
i
n
e
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
n
d
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
o
f
mo
t
i
o
n
i
n
j
u
r
y
i
ma
g
e
s
b
a
se
d
o
n
i
mp
r
o
v
e
d
Y
O
LO
a
l
g
o
r
i
t
h
m,”
i
n
2
0
2
4
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
D
a
t
a
S
c
i
e
n
c
e
a
n
d
N
e
t
w
o
r
k
S
e
c
u
r
i
t
y
(
I
C
D
S
N
S
)
,
J
u
l
.
2
0
2
4
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
D
S
N
S
6
2
1
1
2
.
2
0
2
4
.
1
0
6
9
0
8
6
1
.
[
2
7
]
A
.
K
a
p
o
o
r
,
N
.
S
i
n
g
h
,
a
n
d
V
.
M
o
t
i
l
a
l
,
“
M
o
d
e
l
l
i
n
g
s
o
c
i
a
l
n
e
t
w
o
r
k
s
u
si
n
g
m
o
d
i
f
i
e
d
h
o
p
f
i
e
l
d
n
e
u
r
a
l
n
e
t
w
o
r
k
a
n
d
i
d
e
n
t
i
f
y
l
e
a
d
e
r
s
h
i
p
t
r
a
i
t
s,”
i
n
2
0
1
9
S
e
c
o
n
d
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
A
d
v
a
n
c
e
d
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
P
a
r
a
d
i
g
m
s
(
I
C
AC
C
P)
,
F
e
b
.
2
0
1
9
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
C
C
P
.
2
0
1
9
.
8
8
8
2
8
9
2
.
[
2
8
]
Y
.
K
i
n
o
sh
i
t
a
,
K
.
Y
a
ma
o
k
a
,
a
n
d
H
.
K
i
y
a
,
“
M
a
x
i
m
i
z
a
t
i
o
n
o
f
2
D
c
r
o
ss
-
c
o
r
r
e
l
a
t
i
o
n
b
a
s
e
d
o
n
a
u
x
i
l
i
a
r
y
f
u
n
c
t
i
o
n
met
h
o
d
f
o
r
i
ma
g
e
a
l
i
g
n
me
n
t
,
”
i
n
2
0
2
3
As
i
a
Pa
c
i
f
i
c
S
i
g
n
a
l
a
n
d
I
n
f
o
rm
a
t
i
o
n
Pr
o
c
e
ssi
n
g
Asso
c
i
a
t
i
o
n
A
n
n
u
a
l
S
u
m
m
i
t
a
n
d
C
o
n
f
e
r
e
n
c
e
(
APS
I
PA
AS
C
)
,
O
c
t
.
2
0
2
3
,
p
p
.
2
0
4
3
–
2
0
4
7
,
d
o
i
:
1
0
.
1
1
0
9
/
A
P
S
I
P
A
A
S
C
5
8
5
1
7
.
2
0
2
3
.
1
0
3
1
7
3
7
1
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
J
u
a
n
Ca
m
il
o
G
u
a
c
h
e
ta
-
Alb
a
re
c
e
iv
e
d
h
is
B.
S
.
in
m
e
c
h
a
tro
n
ics
e
n
g
in
e
e
ri
n
g
fr
o
m
Un
iv
e
rsid
a
d
M
il
it
a
r
Nu
e
v
a
G
ra
n
a
d
a
in
2
0
2
0
a
n
d
h
is
M
.
S
c
.
i
n
m
e
c
h
a
tro
n
ics
e
n
g
i
n
e
e
rin
g
fro
m
Un
iv
e
rsid
a
d
M
il
it
a
r
Nu
e
v
a
G
ra
n
a
d
a
in
2
0
2
2
.
Cu
rre
n
t
ly
,
h
e
is
wo
rk
i
n
g
t
o
wa
rd
th
e
P
h
.
D.
i
n
m
e
c
h
a
n
ica
l
e
n
g
in
e
e
rin
g
a
t
U
n
iv
e
rsid
a
d
e
F
e
d
e
ra
l
d
o
Ri
o
d
e
Ja
n
e
iro
.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
ro
b
o
ti
c
s,
m
u
lt
i
-
a
g
e
n
t
s
y
ste
m
s
a
n
d
o
p
ti
m
iza
ti
o
n
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ju
a
n
.
c
a
m
il
o
@m
e
c
a
n
ica
.
c
o
p
p
e
.
u
fr
j.
b
r.
Ann
y
As
tr
id
Espi
ti
a
-
Cub
il
lo
s
p
e
rfo
rm
e
d
h
e
r
u
n
d
e
r
g
ra
d
u
a
te
st
u
d
ies
in
i
n
d
u
strial
e
n
g
in
e
e
rin
g
in
t
h
e
Un
iv
e
rsid
a
d
M
il
it
a
r
Nu
e
v
a
G
ra
n
a
d
a
in
2
0
0
2
a
n
d
M
.
S
c
.
in
i
n
d
u
strial
e
n
g
in
e
e
rin
g
fr
o
m
th
e
U
n
iv
e
rsid
a
d
d
e
L
o
s
An
d
e
s
in
2
0
0
6
.
S
h
e
is
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
o
n
in
d
u
strial
e
n
g
in
e
e
rin
g
p
ro
g
ra
m
a
t
Un
iv
e
rsid
a
d
M
i
li
tar
Nu
e
v
a
G
ra
n
a
d
a
,
Bo
g
o
tá,
C
o
lo
m
b
ia.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
n
n
y
.
e
sp
it
ia@
u
n
imili
tar.ed
u
.
c
o
.
Ro
b
in
so
n
J
i
m
é
n
e
z
-
Mo
r
e
n
o
is
a
n
e
le
c
tro
n
ic
e
n
g
i
n
e
e
r
g
ra
d
u
a
ted
fro
m
Un
iv
e
rsid
a
d
Distrit
a
l
F
ra
n
c
isc
o
J
o
sé
d
e
Ca
ld
a
s
i
n
2
0
0
2
.
He
re
c
e
iv
e
d
a
M
.
S
c
.
i
n
e
n
g
in
e
e
rin
g
fro
m
Un
i
v
e
rsid
a
d
Na
c
io
n
a
l
d
e
Co
lo
m
b
ia
in
2
0
1
2
a
n
d
P
h
.
D.
i
n
e
n
g
i
n
e
e
rin
g
a
t
U
n
iv
e
rsid
a
d
Distrit
a
l
F
ra
n
c
isc
o
Jo
sé
d
e
Ca
l
d
a
s
in
2
0
1
8
.
He
is
a
ss
o
c
iate
p
ro
fe
s
so
r
o
f
Un
iv
e
rsid
a
d
M
il
it
a
r
Nu
e
v
a
G
ra
n
a
d
a
a
n
d
h
is
re
se
a
rc
h
fo
c
u
se
s
o
n
th
e
u
se
o
f
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s
f
o
r
o
b
jec
t
re
c
o
g
n
it
io
n
a
n
d
ima
g
e
p
ro
c
e
ss
in
g
f
o
r
r
o
b
o
ti
c
a
p
p
li
c
a
ti
o
n
s
su
c
h
a
s
h
u
m
a
n
-
m
a
c
h
i
n
e
in
tera
c
ti
o
n
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ro
b
i
n
so
n
.
ji
m
e
n
e
z
@u
n
imili
tar.ed
u
.
c
o
.
Evaluation Warning : The document was created with Spire.PDF for Python.