I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t
2021
,
p
p
.
3
5
7
5
~
3
5
8
3
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
1
i
4
.
pp
3
5
7
5
-
3
5
8
3
3575
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
hex
-
n
ut
s
o
rting sy
ste
m
w
ith
deep
learning
Cristia
n Al
m
a
nza
1
,
J
a
v
ier
M
a
rt
ínez
B
a
qu
er
o
2
,
Ro
bin
s
o
n J
i
m
é
nez
-
M
o
re
no
3
1,
2
F
a
c
u
lt
y
o
f
Ba
sic
S
c
ien
c
e
s a
n
d
En
g
in
e
e
rin
g
,
d
e
lo
s L
lan
o
s
U
n
iv
e
rsity
,
Co
lo
m
b
ia
3
De
p
a
rtme
n
t
o
f
M
e
c
h
a
tro
n
ic E
n
g
i
n
e
e
rin
g
,
M
il
i
tar Nu
e
v
a
G
ra
n
a
d
a
Un
iv
e
rsit
y
,
Co
lo
m
b
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
7
,
2
0
20
R
ev
i
s
ed
Dec
2
3
,
2
0
2
0
A
cc
ep
ted
J
an
1
3
,
2
0
2
1
T
h
is
a
rti
c
le
e
x
p
o
se
s
th
e
d
e
sig
n
a
n
d
im
p
lem
e
n
tatio
n
o
f
a
n
a
u
t
o
m
a
t
io
n
sy
ste
m
b
a
se
d
o
n
a
ro
b
o
ti
c
a
rm
f
o
r
h
e
x
-
n
u
t
c
las
sif
ica
ti
o
n
,
u
si
n
g
p
a
tt
e
rn
r
e
c
o
g
n
it
io
n
a
n
d
im
a
g
e
p
ro
c
e
ss
in
g
.
T
h
e
ro
b
o
ti
c
a
rm
w
o
rk
b
a
se
d
o
n
th
re
e
se
rv
o
m
o
to
rs
a
n
d
a
n
e
lec
tro
m
a
g
n
e
ti
c
e
n
d
e
ffe
c
to
r.
T
h
e
p
a
tt
e
rn
re
c
o
g
n
it
i
o
n
imp
lem
e
n
ted
a
ll
o
w
s
c
las
sify
in
g
th
re
e
d
iff
e
re
n
t
ty
p
e
s
o
f
h
e
x
-
n
u
t
th
r
o
u
g
h
d
e
e
p
lea
rn
in
g
a
lg
o
rit
h
m
s
b
a
se
d
o
n
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
a
rc
h
it
e
c
tu
re
s.
T
h
e
p
ro
p
o
se
d
m
e
th
o
d
e
x
p
o
se
s f
o
u
r
p
h
a
se
s: t
h
e
f
ir
st i
s th
e
d
e
sig
n
,
im
p
le
m
e
n
tatio
n
,
a
n
d
c
o
n
tr
o
l
o
f
a
ro
b
o
ti
c
a
rm
.
T
h
e
se
c
o
n
d
is
th
e
c
a
p
tu
re
,
c
las
sif
i
c
a
ti
o
n
,
a
n
d
im
a
g
e
tre
a
t
m
e
n
t;
th
e
th
ird
a
l
lo
w
s
g
rip
p
in
g
t
h
e
n
u
t
th
r
o
u
g
h
t
h
e
ro
b
o
t’s
in
v
e
rse
k
in
e
m
a
ti
c
.
T
h
e
f
in
a
l
p
h
a
se
is
th
e
re
-
lo
c
a
li
z
a
ti
o
n
o
f
th
e
h
e
x
-
n
u
t
i
n
th
e
re
sp
e
c
ti
v
e
c
o
n
tain
e
r.
T
h
e
a
u
to
m
a
ti
o
n
sy
ste
m
su
c
c
e
ss
f
u
ll
y
c
l
a
ss
i
fies
a
ll
th
e
ty
p
e
s
o
f
h
e
x
-
n
u
ts,
w
h
e
re
th
e
c
o
n
v
o
lu
ti
o
n
a
l
n
e
tw
o
rk
u
se
d
is
a
n
e
f
ficie
n
t
a
n
d
re
c
e
n
t
p
a
tt
e
rn
re
c
o
g
n
i
ti
o
n
m
e
th
o
d
,
w
it
h
a
n
a
c
c
u
ra
c
y
o
f
1
0
0
%
in
1
5
0
it
e
ra
ti
o
n
s.
T
h
is
d
e
v
e
lo
p
m
e
n
t
a
ll
o
w
s
f
o
r
o
b
tai
n
in
g
a
n
o
v
e
l
a
lg
o
rit
h
m
f
o
r
ro
b
o
ti
c
a
p
p
li
c
a
ti
o
n
s in
h
e
x
-
n
u
t
so
rti
n
g
.
K
ey
w
o
r
d
s
:
C
o
n
v
o
lu
tio
n
al
n
e
t
w
o
r
k
Hex
-
n
u
t c
las
s
i
f
icatio
n
I
m
ag
e
p
r
o
ce
s
s
i
n
g
R
o
b
o
tic
ar
m
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
:
R
o
b
in
s
o
n
J
i
m
én
ez
Mo
r
en
o
Me
ch
atr
o
n
ic
s
E
n
g
in
ee
r
i
n
g
P
r
o
g
r
a
m
,
Fac
u
lt
y
o
f
E
n
g
i
n
ee
r
i
n
g
Militar
Nu
e
v
a
Gr
an
ad
a
U
n
iv
e
r
s
it
y
C
ar
r
er
a
1
1
#
1
0
1
-
8
0
,
B
o
g
o
tá
D.
C
.
,
C
o
lo
m
b
ia
E
m
ail: r
o
b
in
s
o
n
.
j
i
m
en
ez
@
u
n
i
m
ilit
ar
.
ed
u
.
co
1.
I
NT
RO
D
UCT
I
O
N
P
r
o
ce
s
s
au
to
m
atio
n
is
an
ar
ea
th
at
r
o
b
o
tics
to
d
a
y
h
as
r
e
v
o
lu
t
io
n
ized
,
s
o
th
a
t
r
o
b
o
ts
o
f
d
if
f
e
r
en
t
t
y
p
e
s
s
u
c
h
a
s
a
n
t
h
r
o
p
o
m
o
r
p
h
ic
[
1
]
o
r
p
ar
allel
s
tr
u
ct
u
r
e
[
2
]
ar
e
u
s
ed
i
n
in
d
u
s
tr
ial
tas
k
s
,
s
u
c
h
as
co
llec
tin
g
a
n
d
r
elo
ca
tin
g
o
b
j
ec
ts
[
3
]
.
A
f
u
n
d
a
m
en
tal
p
ar
t
o
f
r
o
b
o
tic
ag
en
ts
'
u
s
e
i
n
au
to
m
atic
s
y
s
te
m
s
i
s
th
e
n
ee
d
f
o
r
m
ac
h
in
e
v
is
io
n
s
y
s
te
m
s
,
e
m
p
lo
y
ed
f
o
r
n
av
i
g
atio
n
tas
k
s
[
4
]
o
r
elem
e
n
ts
o
f
i
n
ter
est d
etec
tio
n
[
5
]
.
On
e
o
f
th
e
w
id
el
y
u
s
ed
m
ac
h
in
e
v
is
io
n
tech
n
iq
u
es
is
i
m
a
g
e
p
r
o
ce
s
s
in
g
alg
o
r
it
h
m
s
[
6
]
,
w
h
ich
h
as
s
ev
er
al
ap
p
licatio
n
s
at
p
atter
n
r
ec
o
g
n
iti
o
n
i
n
i
m
a
g
es.
I
n
[
7
]
,
v
ar
io
u
s
tech
n
iq
u
es
ap
p
lied
to
e
y
e
r
ec
o
g
n
itio
n
ar
e
ex
p
o
s
in
g
,
al
lo
w
in
g
t
h
e
m
to
b
e
u
s
ed
in
b
io
m
etr
ic
o
r
cli
n
ical
s
y
s
te
m
s
.
I
n
[
8
]
,
p
r
o
ce
s
s
in
g
tec
h
n
iq
u
es
to
i
m
p
r
o
v
e
th
e
a
n
al
y
s
is
in
co
lo
r
i
m
a
g
es
ar
e
ex
p
o
s
ed
.
I
n
[
9
]
,
an
alg
o
r
i
th
m
is
ex
p
o
s
ed
to
en
h
a
n
ce
t
h
e
i
m
a
g
es
'
q
u
alit
y
,
a
h
ig
h
l
y
r
ele
v
a
n
t
f
ac
to
r
i
n
m
ed
ical
-
o
r
ien
ted
m
ac
h
i
n
e
v
i
s
io
n
ap
p
licatio
n
s
.
I
n
[
1
0
]
,
a
s
u
p
p
o
r
t
s
y
s
te
m
f
o
r
s
a
f
e
d
r
iv
in
g
b
ased
o
n
d
r
i
v
er
n
u
m
b
n
ess
d
etec
t
io
n
i
s
p
r
esen
ted
.
I
n
[
1
1
]
,
a
p
r
o
ce
s
s
in
g
a
lg
o
r
it
h
m
i
s
p
r
esen
ted
th
at
allo
w
s
t
h
e
co
u
n
ti
n
g
o
f
n
u
c
lear
ce
lls
au
to
m
atica
ll
y
,
e
v
i
d
en
cin
g
t
h
e
b
en
e
f
it
s
o
f
i
m
ag
e
p
r
o
ce
s
s
i
n
g
f
o
r
id
en
ti
f
y
i
n
g
ele
m
e
n
t
u
n
its
,
a
n
asp
ec
t u
s
ed
in
t
h
e
d
ev
elo
p
m
en
t
o
f
th
i
s
w
o
r
k
.
A
lt
h
o
u
g
h
m
o
r
e
ap
p
licatio
n
s
o
f
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
al
g
o
r
ith
m
s
[
1
2
,
1
3
]
ca
n
b
e
m
e
n
t
io
n
ed
,
th
e
d
ev
elo
p
m
en
t
p
r
esen
ted
i
n
[
1
4
]
e
m
p
lo
y
s
i
m
ag
e
p
r
o
ce
s
s
in
g
te
ch
n
iq
u
es
f
o
r
an
a
l
y
s
is
o
f
i
n
ter
n
al
s
p
ac
e
m
e
tr
ics
i
n
a
n
u
t,
h
i
g
h
li
g
h
tin
g
th
e
u
s
e
o
f
th
e
s
a
m
e
f
o
r
o
b
j
ec
ts
lo
ca
li
za
tio
n
lik
e
is
e
x
p
lo
r
ed
in
th
i
s
w
o
r
k
.
R
e
s
ea
r
ch
e
s
ar
o
u
n
d
th
e
g
e
n
e
r
al
p
r
o
p
er
ties
o
f
n
u
ts
'
u
s
e
ar
e
ex
p
o
s
ed
i
n
s
ta
te
o
f
t
h
e
ar
t
[
1
5
-
1
7
]
;
h
o
w
e
v
er
,
th
o
s
e
o
r
ien
ted
to
r
o
b
o
tic
au
to
m
atio
n
task
s
f
o
r
p
ick
an
d
p
lace
ar
e
n
o
t
r
ef
er
r
ed
.
T
h
is
r
o
b
o
tic
f
ield
task
ex
p
o
s
es
s
e
v
er
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
5
7
5
-
3583
3576
ap
p
licatio
n
s
f
o
r
o
th
er
t
y
p
es
o
f
o
b
j
ec
ts
[
1
8
,
1
9
]
.
A
s
a
p
r
o
b
lem
to
s
o
l
v
e
i
n
t
h
e
clas
s
i
f
icatio
n
o
f
a
s
p
ec
if
ic
h
e
x
-
n
u
t
k
in
d
,
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
r
e
q
u
ir
es
a
n
ad
d
itio
n
al
p
atter
n
r
e
co
g
n
itio
n
s
tag
e,
l
ik
e
n
e
u
r
al
n
et
w
o
r
k
s
,
i
n
cr
ea
s
i
n
g
th
e
co
m
p
le
x
it
y
o
f
r
o
b
o
tic
m
a
n
ip
u
lato
r
s
'
g
r
ip
p
in
g
tas
k
.
T
o
d
is
cr
i
m
in
a
te
o
b
j
ec
ts
in
i
m
ag
es,
p
a
tter
n
r
ec
o
g
n
i
tio
n
tec
h
n
iq
u
e
s
li
k
e
co
n
v
o
lu
t
io
n
al
n
et
w
o
r
k
s
[
2
0
]
ar
e
v
er
y
ef
f
icie
n
t
[
2
1
]
.
A
p
p
lic
atio
n
s
in
b
io
lo
g
ical
i
m
a
g
e
c
las
s
if
ica
tio
n
[
2
2
]
,
r
ea
l
-
ti
m
e
f
ac
e
r
ec
o
g
n
itio
n
[
2
3
]
,
o
r
g
r
asp
in
g
a
n
o
b
j
ec
t
in
a
r
o
b
o
ti
c
en
v
ir
o
n
m
e
n
t
[
2
4
]
,
g
iv
e
p
r
o
b
es
o
f
its
v
er
s
atil
it
y
[
2
5
]
.
I
n
th
i
s
w
o
r
k
,
C
NN
’
s
ar
e
u
s
ed
to
d
i
s
cr
i
m
in
ate
t
h
r
ee
clas
s
o
f
h
e
x
-
n
u
t
b
y
tr
an
s
f
er
lear
n
i
n
g
m
eth
o
d
[
2
6
]
,
u
s
in
g
a
n
e
lect
r
o
m
a
g
n
et
-
t
y
p
e
e
n
d
ef
f
ec
to
r
f
o
r
p
ick
an
d
p
lace
ta
s
k
s
,
th
r
o
u
g
h
a
r
o
b
o
tic
ar
m
a
m
ac
h
in
e
v
i
s
io
n
s
y
s
te
m
,
a
n
d
i
m
ag
e
p
r
o
ce
s
s
i
n
g
f
o
r
th
e
h
ex
-
n
u
ts
lo
ca
lizatio
n
.
As a
d
ee
p
lear
n
in
g
al
g
o
r
ith
m
,
C
N
N
is
a
r
ec
en
t
tech
n
iq
u
e
i
n
p
att
er
n
r
ec
o
g
n
itio
n
;
it i
s
b
ein
g
u
s
ed
in
th
e
r
o
b
o
t
ic
au
to
m
atiza
tio
n
p
r
o
ce
s
s
es
[
2
7
,
2
8
]
lik
e
is
ex
p
o
s
e
in
t
h
is
w
o
r
k
,
b
u
t
in
s
tate
o
f
th
e
ar
t
is
n
o
t e
x
p
lo
r
e
th
i
s
tec
h
n
iq
u
e
i
n
h
e
x
n
u
t c
la
s
s
i
f
icatio
n
w
it
h
a
r
o
b
o
tic
m
an
ip
u
lato
r
.
T
h
e
p
ap
er
is
d
iv
id
ed
in
to
f
o
u
r
s
ec
tio
n
s
.
T
h
e
f
ir
s
t
i
s
t
h
e
p
r
ev
io
u
s
i
n
tr
o
d
u
ctio
n
.
T
h
e
s
ec
o
n
d
s
ec
tio
n
ex
p
lain
s
th
e
r
esear
c
h
m
et
h
o
d
u
s
ed
,
th
e
r
o
b
o
tic
ar
m
,
its
m
at
h
e
m
a
tical
m
o
d
el,
th
e
co
n
v
o
lu
tio
n
al
n
et
w
o
r
k
ar
ch
ite
ct
u
r
e
u
s
ed
,
an
d
it
s
tr
ain
i
n
g
.
T
h
e
t
h
ir
d
s
ec
tio
n
e
x
p
o
s
es
t
h
e
r
es
u
lt
s
a
n
d
d
is
c
u
s
s
io
n
o
f
h
ex
-
n
u
t
class
i
f
icatio
n
an
d
its
a
n
al
y
s
i
s
,
an
d
f
i
n
all
y
,
s
ec
tio
n
f
o
u
r
e
x
p
o
s
es th
e
co
n
cl
u
s
io
n
s
ac
h
ie
v
ed
.
2.
P
RO
P
O
SE
D
M
E
T
H
O
D
A
t
h
r
ee
-
d
eg
r
ee
-
of
-
f
r
ee
d
o
m
r
o
b
o
tic
ar
m
w
as
i
m
p
le
m
en
t
ed
w
it
h
a
n
elec
tr
o
m
ag
n
et
-
b
ased
en
d
ef
f
ec
to
r
ca
p
ab
le
o
f
attr
ac
tin
g
th
e
n
u
t
an
d
h
o
ld
in
g
it
w
h
ile
i
t
is
r
ep
o
s
itio
n
ed
in
t
h
e
d
esire
d
p
o
s
iti
o
n
.
T
h
e
r
o
b
o
tic
ar
m
o
p
er
ates
b
ased
o
n
s
er
v
o
m
o
to
r
s
an
d
elec
tr
o
m
a
g
n
et
-
t
y
p
e
en
d
ef
f
ec
to
r
w
it
h
a
p
last
ic
s
h
a
f
t
t
h
at
p
r
o
tr
u
d
es
to
h
o
ld
th
e
n
u
t.
Fi
g
u
r
e
1
(
le
f
t)
il
lu
s
tr
at
es
t
h
e
te
s
t
e
n
v
ir
o
n
m
en
t
u
s
ed
,
w
h
er
e
t
h
e
b
lac
k
b
ac
k
g
r
o
u
n
d
w
h
er
e
t
h
e
n
u
ts
to
b
e
r
elo
ca
ted
w
o
u
ld
b
e
f
o
u
n
d
i
s
h
ig
h
li
g
h
ted
.
T
h
e
s
y
s
te
m
h
a
s
an
i
n
ter
f
ac
e
t
h
at
allo
w
s
h
u
m
an
i
n
ter
v
e
n
tio
n
a
n
d
co
n
tr
o
l,
th
r
o
u
g
h
t
h
e
T
eleg
r
am
a
p
p
licatio
n
[
2
9
]
ca
p
ab
le
o
f
o
p
er
atin
g
f
r
o
m
a
n
An
d
r
o
id
o
r
I
OS
ce
ll
p
h
o
n
e.
Fig
u
r
e
1
(
r
ig
h
t)
ill
u
s
tr
ate
s
th
e
lo
ca
l
co
n
tr
o
ller
b
ased
o
n
a
r
a
s
p
b
er
r
y
p
i
2
m
o
d
el
B
,
p
r
o
g
r
am
ed
in
p
y
t
h
o
n
,
t
h
e
pi
-
ca
m
er
a
V1
o
f
t
h
e
r
asp
b
er
r
y
w
as
u
s
ed
f
o
r
i
m
ag
e
ac
q
u
is
i
tio
n
,
an
d
its
GP
I
O
m
o
d
u
le
is
u
s
ed
as
a
d
r
iv
er
to
ea
ch
s
er
v
o
m
o
to
r
.
Fig
u
r
e
1
.
W
o
r
k
s
p
ac
e
an
d
co
n
tr
o
ller
cir
cu
it
A
p
h
a
s
ed
m
et
h
o
d
o
lo
g
y
w
as
u
s
ed
,
w
h
er
e
ea
c
h
p
h
ase
'
s
n
a
m
e
s
ar
e
ca
p
tu
r
e,
class
if
ica
tio
n
,
a
n
d
i
m
a
g
e
tr
ea
t
m
e
n
t.
Desi
g
n
,
i
m
p
le
m
e
n
t
atio
n
,
an
d
co
n
tr
o
l
o
f
th
e
r
o
b
o
ti
c
ar
m
,
th
e
n
u
t
's
ca
p
tu
r
e,
an
d
th
e
s
y
s
te
m
i
n
ter
f
ac
e.
Fig
u
r
e
2
s
h
o
w
i
n
g
a
f
lo
w
ch
ar
t
o
f
th
e
p
h
a
s
es
u
s
ed
in
all
t
h
e
p
r
o
ce
s
s
es.
Fo
r
th
e
f
ir
s
t
s
tag
e,
i
m
ag
e
ca
p
t
u
r
e,
d
ig
itizatio
n
an
d
tr
ea
t
m
e
n
t
,
a
s
u
p
er
v
is
o
r
y
ca
m
er
a
w
a
s
in
s
talled
i
n
th
e
w
o
r
k
ar
ea
to
ca
p
tu
r
e
t
h
e
i
n
p
u
t
i
m
a
g
es
o
f
t
h
e
s
y
s
te
m
.
St
ag
e
t
w
o
co
n
s
is
ts
o
f
t
h
e
d
esi
g
n
,
i
m
p
le
m
e
n
tatio
n
,
an
d
co
n
tr
o
l
o
f
a
r
o
b
o
tic
ar
m
,
w
h
er
e
th
e
i
m
p
le
m
e
n
tat
io
n
was
ca
r
r
ied
o
u
t
th
r
o
u
g
h
t
h
e
u
s
e
o
f
m
u
ltip
u
r
p
o
s
e
b
r
ac
k
ets,
w
h
ich
allo
w
u
s
to
f
i
x
t
h
e
s
er
v
o
m
o
to
r
s
a
n
d
s
h
ap
e
t
h
e
ar
m
.
T
h
r
ee
s
er
v
o
m
o
to
r
s
wer
e
u
s
ed
,
o
n
e
at
t
h
e
b
ase,
to
co
n
tr
o
l
th
e
"
XY"
ax
is
,
o
n
e
in
t
h
e
m
id
d
le,
w
i
th
v
er
ti
ca
l
m
o
v
e
m
e
n
t
to
co
n
tr
o
l
t
h
e
"
Z
"
ax
is
,
a
n
d
o
n
e
at
th
e
e
n
d
,
f
o
r
o
r
ien
ta
tio
n
o
f
t
h
e
en
d
ef
f
ec
to
r
.
T
h
e
f
in
i
s
h
ed
ar
m
w
a
s
p
lace
d
i
n
f
r
o
n
t
o
f
t
h
e
ca
m
er
a
s
o
t
h
at
th
e
n
u
t
ca
n
b
e
p
h
o
to
g
r
ap
h
ed
an
d
tak
en
b
y
th
e
ar
m
s
o
th
at
it
ca
n
l
ater
b
e
lo
ca
te
d
.
A
p
last
ic
co
n
tain
er
is
u
s
ed
as
a
m
ea
n
s
o
f
s
to
r
in
g
th
e
n
u
t
s
ta
k
e
n
f
r
o
m
t
h
e
w
o
r
k
ar
ea
,
as c
an
b
e
s
ee
n
i
n
Fi
g
u
r
e
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
R
o
b
o
tic
h
ex
-
n
u
t so
r
tin
g
s
ystem
w
ith
d
ee
p
lea
r
n
in
g
(
C
r
is
tia
n
A
lma
n
z
a
)
3577
Fig
u
r
e
2
.
P
r
o
p
o
s
ed
alg
o
r
ith
m
f
lo
w
ch
ar
t
Fig
u
r
e
3
.
Ma
ter
ial
s
e
m
p
lo
y
ed
I
n
Fi
g
u
r
e
4
ca
n
b
e
o
b
s
er
v
ed
th
e
s
ch
e
m
atic
m
an
ip
u
lato
r
r
ep
r
esen
tat
io
n
u
s
ed
f
o
r
th
e
i
n
v
er
s
e
k
in
e
m
at
ic
u
s
ed
to
m
o
v
e
t
h
e
r
o
b
o
t,
alo
n
g
w
it
h
t
h
e
co
o
r
d
in
ate
s
y
s
te
m
f
o
r
ea
ch
j
o
in
t.
T
h
e
an
g
le
o
f
th
e
j
o
in
t
1
an
d
t
h
e
X
’
co
m
p
o
n
e
n
t
o
f
t
h
e
f
i
n
al
p
o
i
n
t
P
w
as
o
b
tain
ed
w
it
h
t
h
e
f
r
o
n
t
al
v
ie
w
,
as
s
h
o
w
n
i
n
t
h
e
(
1
)
a
n
d
(
2
)
.
T
h
e
an
g
le
o
f
th
e
j
o
in
ts
2
an
d
3
w
a
s
o
b
tain
ed
w
it
h
th
e
f
r
o
n
tal
v
ie
w
,
b
y
g
etti
n
g
“
d
”
le
n
g
th
,
a
n
d
alf
a
a
n
d
b
eta
an
g
le
s
li
k
e
s
h
o
w
n
in
t
h
e
(
3
)
-
(
6
)
.
′
=
√
2
+
2
(
1
)
1
=
ta
n
−
1
(
2
)
=
√
(
−
1
)
2
+
′
2
(
3
)
2
=
2
2
+
3
2
+
2
2
3
c
os
3
c
os
3
=
2
−
2
2
−
3
2
2
2
3
(
4
)
s
in
3
=
±
√
1
−
2
3
(
5
)
3
=
ta
n
−
1
s
i
n
3
co
s
3
(
6
)
T
h
e
r
o
b
o
tic
ar
m
h
a
s
f
i
v
e
ca
l
i
b
r
ated
p
o
s
itio
n
s
,
t
h
e
f
ir
s
t
i
s
t
h
e
"
r
est"
p
o
s
itio
n
,
in
w
h
ich
th
e
r
o
b
o
t
w
il
l
b
e
s
tr
etch
ed
u
p
w
ar
d
s
,
t
h
e
s
ec
o
n
d
s
tate
i
s
t
h
at
o
f
"
p
ick
in
g
u
p
"
,
w
h
ich
w
ill
b
e
p
o
s
itio
n
ed
a
s
clo
s
e
to
t
h
e
n
u
t
to
b
e
g
r
ip
p
in
g
,
an
d
t
h
e
r
e
s
t,
ar
e
p
o
s
itio
n
ed
o
n
to
p
o
f
t
h
e
d
if
f
e
r
en
t
co
n
ta
in
er
s
.
Dep
en
d
i
n
g
o
n
t
h
e
t
y
p
e
o
f
n
u
t,
i
t
w
il
l b
e
p
lace
d
in
a
s
ep
ar
ate
co
n
tai
n
er
as s
h
o
w
n
i
n
F
i
g
u
r
e
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
5
7
5
-
3583
3578
Fig
u
r
e
4
.
Fro
n
tal
an
d
to
p
v
ie
w
o
f
th
e
r
o
b
o
tic
ar
m
Fig
u
r
e
5
.
R
o
b
o
t p
r
e
-
p
r
o
g
r
am
e
d
p
o
s
itio
n
s
I
n
th
e
t
h
ir
d
s
ta
g
e,
t
h
e
w
a
y
to
m
o
v
e
t
h
e
r
o
b
o
t
f
o
r
g
r
ip
th
e
n
u
t
is
e
x
p
o
s
ed
;
th
e
e
lectr
o
m
a
g
n
et
f
o
r
th
e
g
r
ip
p
er
is
e
m
p
lo
y
ed
b
ec
au
s
e
it
allo
w
s
u
s
to
ea
s
il
y
ac
t
iv
ate
an
d
d
ea
cti
v
ate
t
h
e
a
ttra
ctio
n
f
o
r
cla
m
p
i
n
g
f
er
r
o
m
ag
n
etic
ele
m
en
ts
s
u
ch
as
th
e
n
u
t,
an
d
it
d
o
es
n
o
t
r
e
q
u
ir
e
h
i
g
h
p
r
ec
is
io
n
f
o
r
g
r
ip
th
r
o
u
g
h
t
h
e
ar
m
.
T
h
e
elec
tr
o
m
ag
n
et
co
n
s
i
s
ts
o
f
a
7
0
-
tu
r
n
w
i
n
d
in
g
o
n
a
m
etal
s
c
r
e
w
,
w
i
th
2
4
-
g
a
u
g
e
co
p
p
er
w
ir
e,
w
h
ic
h
w
h
e
n
9
v
o
lts
ar
e
ap
p
lied
,
g
en
er
ates a
s
tab
le
m
ag
n
etic
f
ield
n
ec
e
s
s
ar
y
to
s
u
p
p
o
r
t a
n
u
t.
Fin
all
y
,
to
cla
s
s
i
f
y
t
h
e
h
ex
-
n
u
ts
,
a
co
n
v
o
lu
t
io
n
al
n
e
u
r
al
n
e
t
w
o
r
k
is
u
s
ed
w
h
o
s
e
ar
c
h
itect
u
r
e
ca
n
b
e
s
ee
n
i
n
T
ab
le
1
.
T
h
e
ar
ch
itectu
r
e
u
s
e
s
f
o
u
r
co
n
v
o
lu
tio
n
al
la
y
er
s
an
d
e
n
d
in
g
i
n
a
f
latte
n
la
y
er
w
it
h
th
r
ee
o
u
ts
.
T
h
e
k
er
n
el
co
lu
m
n
s
h
o
w
s
th
e
n
u
m
b
er
o
f
f
ilter
s
u
s
ed
an
d
ea
c
h
o
n
e
'
s
s
ize,
b
ased
o
n
s
q
u
ar
e
f
ilter
s
w
it
h
s
id
e
L
.
Fig
u
r
e
6
p
r
esen
ts
t
h
e
d
atab
as
e
u
s
ed
f
o
r
t
h
e
n
u
t
s
'
d
is
cr
i
m
in
atio
n
,
alth
o
u
g
h
t
h
e
d
ef
a
u
lt
b
a
ck
g
r
o
u
n
d
i
s
b
lack
,
o
th
er
b
ac
k
g
r
o
u
n
d
s
ar
e
u
s
ed
to
g
e
n
er
alize
t
h
e
lear
n
i
n
g
.
T
h
e
to
p
r
o
w
co
r
r
esp
o
n
d
s
to
th
e
t
y
p
e
1
n
u
t,
t
h
e
m
id
d
le
r
o
w
to
t
h
e
t
y
p
e
2
n
u
t,
an
d
th
e
f
i
n
al
r
o
w
to
t
h
e
t
y
p
e
3
n
u
t.
Fig
u
r
e
7
s
h
o
w
s
t
h
e
n
et
w
o
r
k
's
p
er
f
o
r
m
a
n
ce
,
w
h
ic
h
allo
w
d
is
cr
i
m
i
n
ate
t
h
e
h
ex
-
n
u
ts
,
r
ea
ch
i
n
g
1
0
0
%
p
r
ec
is
io
n
in
t
h
e
ta
s
k
,
o
n
l
y
1
5
0
iter
atio
n
s
w
a
s
n
ec
es
s
ar
y
f
o
r
th
e
n
et
w
o
r
k
lear
n
in
g
w
a
s
esta
b
lis
h
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:
2
0
8
8
-
8708
R
o
b
o
tic
h
ex
-
n
u
t so
r
tin
g
s
ystem
w
ith
d
ee
p
lea
r
n
in
g
(
C
r
is
tia
n
A
lma
n
z
a
)
3579
T
ab
le
1
.
C
o
n
v
o
lu
tio
n
al
n
et
w
o
r
k
ar
ch
itect
u
r
e.
L
A
Y
ER
K
ER
N
E
L
I
n
p
u
t
2
4
0
x
2
4
0
x
3
C
o
n
v
o
l
u
t
i
o
n
/
R
EL
U
98
8
C
o
n
v
o
l
u
t
i
o
n
/
R
EL
U
1
9
2
6
M
a
x
P
o
o
l
i
n
g
3
C
o
n
v
o
l
u
t
i
o
n
/
R
EL
U
1
9
2
3
M
a
x
P
o
o
l
i
n
g
3
C
o
n
v
o
l
u
t
i
o
n
/
R
EL
U
3
2
0
3
M
a
x
P
o
o
l
i
n
g
2
Fig
u
r
e
6
.
He
x
-
n
u
t d
atab
ase
f
o
r
C
NN
tr
ain
in
g
Fig
u
r
e
7
.
C
NN
tr
ain
i
n
g
ac
cu
r
a
c
y
Fig
u
r
e
8
s
h
o
w
s
th
e
n
et
w
o
r
k
'
s
lear
n
in
g
t
h
r
o
u
g
h
t
h
e
ac
ti
v
atio
n
s
o
f
t
h
e
la
s
t
la
y
er
,
a
n
d
it
i
s
p
o
s
s
ib
le
to
d
if
f
er
e
n
tiate
b
et
w
ee
n
th
o
s
e
f
il
ter
s
th
at
lear
n
t
h
e
b
ac
k
g
r
o
u
n
d
f
r
o
m
t
h
o
s
e
th
at
lear
n
th
e
n
u
t
s
h
ap
e.
On
ce
th
e
n
u
t
is
r
ec
o
g
n
ized
,
co
lo
r
to
g
r
a
y
s
ca
le
tr
an
s
f
o
r
m
atio
n
is
u
s
ed
a
s
al
g
o
r
ith
m
o
f
i
m
a
g
e
p
r
o
ce
s
s
in
g
[
3
0
]
,
ap
p
ly
i
n
g
th
r
es
h
o
ld
in
g
to
th
e
i
m
a
g
e
to
d
etec
t
th
e
g
eo
m
etr
ic
ce
n
ter
[
3
1
]
an
d
tak
e
t
h
is
p
i
x
el
as
a
r
e
f
er
en
ce
p
o
in
t
to
lo
ca
te
th
e
r
o
b
o
t'
s
e
n
d
ef
f
ec
to
r
o
v
er
th
e
h
ex
-
n
u
t a
s
s
h
o
w
n
i
n
F
i
g
u
r
e
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
5
7
5
-
3583
3580
Fig
u
r
e
8
.
Net
w
o
r
k
ac
ti
v
atio
n
s
ex
a
m
p
le
s
Fig
u
r
e
9
.
T
h
r
esh
o
ld
in
g
o
f
i
m
a
g
e
3.
RE
SU
L
T
S
A
ND
D
I
SCU
SS
I
O
N
T
ab
le
2
co
m
p
ar
es
th
e
n
u
m
b
er
o
f
n
u
ts
clas
s
i
f
ied
,
ca
p
tu
r
ed
,
an
d
co
r
r
ec
tly
lo
ca
ted
,
co
n
ce
r
n
in
g
th
e
n
u
t
s
lo
ca
ted
ex
p
o
s
ed
u
n
d
er
th
e
w
o
r
k
ar
ea
.
I
t
is
p
o
s
s
ib
le
to
in
f
er
t
h
at
th
e
r
o
b
o
tic
ar
m
f
u
l
f
illed
it
s
class
i
f
icat
io
n
tas
k
s
atis
f
ac
to
r
il
y
s
in
ce
t
h
e
r
esp
ec
tiv
e
h
e
x
-
n
u
t
s
,
w
er
e
lo
ca
ted
in
th
e
d
esire
d
p
lace
s
.
L
ik
e
w
i
s
e,
th
e
elec
tr
o
m
a
g
n
et
allo
w
ed
th
e
n
u
t
s
'
p
r
o
p
er
g
r
ip
an
d
r
elea
s
ed
th
e
m
at
th
e
ap
p
r
o
p
r
iate
tim
e
s
,
g
e
n
er
atin
g
a
r
o
b
u
s
t
an
d
r
eliab
le
s
y
s
te
m
a
s
s
h
o
w
n
i
n
Fi
g
u
r
e
1
0
.
T
ab
le
2
.
Hex
-
n
u
t c
las
s
i
f
icatio
n
N
u
t
c
l
a
ss
C
l
a
ssi
f
i
c
a
t
i
o
n
R
e
l
o
c
a
t
e
d
C
l
a
ss
1
3
3
C
l
a
ss
2
4
4
C
l
a
ss
3
2
2
D
i
sca
r
d
e
d
1
1
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
R
o
b
o
tic
h
ex
-
n
u
t so
r
tin
g
s
ystem
w
ith
d
ee
p
lea
r
n
in
g
(
C
r
is
tia
n
A
lma
n
z
a
)
3581
Fig
u
r
e
1
0
.
Hex
-
n
u
t d
eliv
er
A
s
y
s
te
m
i
n
ter
f
ac
e
w
as
u
s
ed
,
w
h
ich
ag
r
ee
d
to
d
esig
n
a
n
d
i
m
p
le
m
en
t
co
m
m
u
n
icatio
n
b
e
t
w
ee
n
th
e
T
eleg
r
am
ap
p
licatio
n
,
d
o
w
n
lo
ad
ab
le
f
r
o
m
th
e
A
p
p
Sto
r
e,
a
n
d
t
h
e
r
o
b
o
tic
s
y
s
te
m
w
h
ic
h
i
s
co
n
n
ec
ted
to
th
e
in
ter
n
e
t
as
s
h
o
w
n
i
n
Fi
g
u
r
e
1
1
.
T
h
e
s
y
s
te
m
h
as
a
p
r
ec
is
io
n
an
d
an
ac
c
u
r
ac
y
o
f
1
0
0
%,
o
b
tain
in
g
a
n
er
r
o
r
o
f
0
%,
it
is
ac
h
ie
v
ed
b
y
t
h
e
r
o
b
u
s
t
n
es
s
o
f
th
e
C
N
N
an
d
t
h
e
n
u
t
g
r
ip
'
s
s
i
m
p
licit
y
.
B
esid
es,
th
e
o
n
li
n
e
s
y
s
te
m
(
T
eleg
r
am
A
p
p
)
,
allo
w
ed
ef
f
ic
ien
t c
o
m
m
u
n
icatio
n
,
b
y
co
r
r
ec
tl
y
s
e
n
d
in
g
a
n
d
r
ec
eiv
in
g
t
h
e
r
eq
u
ested
d
ata.
Fig
u
r
e
1
1
.
T
eleg
r
a
m
r
ep
o
r
t
4.
CO
NCLU
SI
O
N
I
t
w
a
s
p
o
s
s
ib
le
to
d
esi
g
n
an
d
v
alid
ate
a
n
a
u
to
m
atic
h
e
x
-
n
u
t
cla
s
s
i
f
icat
io
n
s
y
s
te
m
,
w
h
ic
h
r
eq
u
ir
e
s
m
ac
h
in
e
v
i
s
io
n
al
g
o
r
ith
m
s
f
o
r
its
o
p
er
atio
n
.
I
t
allo
w
s
d
e
m
o
n
s
tr
ate
t
h
at
a
r
o
b
o
tic
ar
m
ca
n
f
u
lf
i
ll
clas
s
i
f
icatio
n
ass
is
tan
ce
ta
s
k
s
b
ased
o
n
th
e
i
n
teg
r
at
io
n
o
f
al
g
o
r
ith
m
s
f
o
r
co
n
tr
o
l
an
d
au
to
m
atio
n
o
p
er
atio
n
s
.
A
h
i
g
h
n
u
m
b
er
o
f
test
s
a
m
p
le
s
w
er
e
n
o
t
r
eq
u
ir
ed
b
ec
au
s
e
th
e
h
ex
-
n
u
ts
d
o
n
o
t
v
ar
y
i
n
s
ize
o
r
s
h
ap
e
f
o
r
ea
ch
class
,
w
h
ich
f
o
cu
s
th
e
s
u
cc
es
s
o
f
d
ev
elo
p
i
n
g
a
u
s
e
f
u
l
cla
s
s
i
f
icat
io
n
s
y
s
t
e
m
th
r
o
u
g
h
co
n
v
o
lu
tio
n
al
n
et
w
o
r
k
s
,
v
er
y
e
f
f
icie
n
t
f
o
r
th
is
tas
k
.
T
h
e
in
teg
r
atio
n
w
it
h
th
e
T
eleg
r
a
m
to
o
l
allo
w
s
t
h
e
u
s
er
to
m
o
n
ito
r
th
e
s
y
s
te
m
's
o
p
er
atio
n
r
e
m
o
tel
y
,
v
alid
atin
g
s
h
o
r
t r
esp
o
n
s
e
ti
m
es a
n
d
o
f
f
er
i
n
g
a
m
ea
n
s
o
f
f
ee
d
b
ac
k
to
th
e
u
s
er
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
4
,
A
u
g
u
s
t 2
0
2
1
:
3
5
7
5
-
3583
3582
ACK
N
O
WL
E
D
G
E
M
E
NT
T
h
e
au
th
o
r
s
ar
e
g
r
atef
u
l
to
Un
iv
er
s
id
ad
d
e
lo
s
lla
n
o
s
a
n
d
t
h
e
Un
i
v
er
s
id
ad
Mi
litar
N
u
e
v
a
G
r
an
ad
a
f
o
r
th
e
f
u
n
d
in
g
o
f
t
h
is
p
r
o
j
ec
t.
RE
F
E
R
E
NC
E
S
[1
]
X
.
S
.
L
i
e
t
a
l.
,
“
A
n
a
l
y
sis
a
n
d
S
im
p
li
f
ica
ti
o
n
o
f
T
h
re
e
-
Di
m
e
n
sio
n
a
l
S
p
a
c
e
V
e
c
to
r
P
W
M
f
o
r
T
h
re
e
-
P
h
a
se
F
o
u
r
-
L
e
g
In
v
e
rters
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s o
n
In
d
u
stria
l
El
e
c
tro
n
ics
,
v
o
l
.
5
8
,
p
p
.
4
5
0
-
4
6
4
,
F
e
b
.
2
0
1
1
.
[2
]
H
.
W
e
i,
Y
.
Bu
,
a
n
d
Z
.
Z
h
u
,
“
Ro
b
o
ti
c
a
rm
c
o
n
tr
o
ll
i
n
g
b
a
se
d
o
n
a
sp
ik
in
g
n
e
u
ra
l
c
ircu
it
a
n
d
sy
n
a
p
ti
c
p
las
ti
c
it
y
,
”
Bi
o
me
d
ica
l
S
i
g
n
a
l
Pr
o
c
e
ss
in
g
a
n
d
Co
n
tro
l
, v
ol
.
5
5
,
2
0
2
0
,
d
oi
:
1
0
.
1
0
1
6
/j
.
b
sp
c
.
2
0
1
9
.
1
0
1
6
4
0
.
[3
]
S
.
Am
in
e
a
n
d
E
.
G
.
Ha
n
n
a
,
“
Kin
e
m
a
ti
c
A
n
a
l
y
sis
o
f
HA
L
F
P
a
ra
ll
e
l
Ro
b
o
t,
”
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
rin
g
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
Rev
iew (
J
ES
T
R)
,
v
o
l.
1
2
,
n
o
.
5
,
pp.
2
0
7
–
2
1
3
,
2
0
1
9
,
doi
:
1
0
.
2
5
1
0
3
/j
e
str.
1
2
5
.
23
.
[4
]
T
.
De
w
i,
S
.
Nu
rm
a
in
i,
P
.
Rism
a
,
Y
.
Ok
tarin
a
,
a
n
d
M
.
R
o
riz
,
“
I
n
v
e
rse
k
in
e
m
a
ti
c
a
n
a
l
y
sis
o
f
4
DO
F
p
ick
a
n
d
p
lac
e
a
r
m
ro
b
o
t
m
a
n
ip
u
lato
r
u
sin
g
f
u
z
z
y
lo
g
ic
c
o
n
tro
ll
e
r
,”
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
t
e
r
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l
.
1
0
,
n
o
.
2
,
p
p
.
1
3
7
6
–
1
3
8
6
,
A
p
r.
2
0
2
0
.
[5
]
C
.
G
.
P
.
-
S
u
e
sc
u
n
,
C
.
J.
E.
-
A
ra
g
o
n
,
a
n
d
R
.
J.
-
M
o
re
n
o
,
“
Ro
b
o
t
ic
n
a
v
ig
a
ti
o
n
a
lg
o
rit
h
m
w
i
th
m
a
c
h
in
e
v
isio
n
,”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l
.
1
0
,
n
o
.
2
,
p
p
.
1
3
0
8
–
1
3
1
6
,
A
p
r.
2
0
2
0
.
[6
]
D
.
L
i,
D
.
Jia
n
g
,
R
.
Ba
o
,
L
.
Ch
e
n
,
a
n
d
M
.
K
.
Ke
rn
s
,
“
Cra
c
k
De
te
c
ti
o
n
a
n
d
Re
c
o
g
n
it
i
o
n
M
o
d
e
l
o
f
P
a
rts
Ba
s
e
d
o
n
M
a
c
h
in
e
V
isi
o
n
,
”
J
o
u
rn
a
l
o
f
E
n
g
in
e
e
rin
g
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
Rev
iew
(
J
ES
T
R)
,
v
ol
.
12
,
n
o
.
5
,
p
p
.
1
4
8
–
1
5
6
,
2
0
1
9
,
d
o
i
:
1
0
.
2
5
1
0
3
/j
e
str.
1
2
5
.
1
7
.
[7
]
D.
De
m
iro
v
ić,
E.
S
k
e
ji
ć
,
a
n
d
A
.
Š
e
ri
f
o
v
ić
–
T
rb
a
li
ć
,
“
P
e
rf
o
r
m
a
n
c
e
o
f
S
o
m
e
I
m
a
g
e
P
ro
c
e
ss
in
g
A
l
g
o
rit
h
m
s
in
T
e
n
so
rf
lo
w
,
”
2
0
1
8
2
5
th
I
n
te
rn
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
y
ste
ms
,
S
ig
n
a
ls
a
n
d
Ima
g
e
Pro
c
e
ss
in
g
(
IW
S
S
IP)
,
M
a
ri
b
o
r
,
2
0
1
8
,
p
p
.
1
–
4
,
d
o
i:
1
0
.
1
1
0
9
/IW
S
S
I
P
.
2
0
1
8
.
8
4
3
9
7
1
4
.
[8
]
Dą
b
ro
w
sk
i,
“
Co
n
tem
p
o
ra
r
y
tec
h
n
o
l
o
g
ies
a
n
d
tec
h
n
iq
u
e
s
f
o
r
p
r
o
c
e
ss
in
g
o
f
h
u
m
a
n
e
y
e
i
m
a
g
e
s
,”
2
0
1
8
S
ig
n
a
l
Pro
c
e
ss
in
g
:
Al
g
o
rit
h
ms
,
Arc
h
it
e
c
tu
re
s,
Arra
n
g
e
me
n
ts,
a
n
d
A
p
p
li
c
a
ti
o
n
s
(
S
PA
)
,
P
o
z
n
a
n
,
2
0
1
8
,
p
p
.
1
5
–
1
5,
d
o
i:
1
0
.
2
3
9
1
9
/
S
P
A
.
2
0
1
8
.
8
5
6
3
4
2
6
.
[9
]
V
.
Ja
isw
a
l,
V
.
S
h
a
rm
a
,
a
n
d
S
.
V
a
rm
a
,
“
M
M
F
O:
m
o
d
if
ied
m
o
th
f
la
m
e
o
p
ti
m
iza
ti
o
n
a
lg
o
rit
h
m
f
o
r
re
g
io
n
b
a
s
e
d
RGB
c
o
lo
r
im
a
g
e
se
g
m
e
n
tatio
n
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
m
p
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l
.
1
0
,
n
o
.
1
,
pp
.
1
9
6
–
2
0
1
,
F
e
b
.
2
0
2
0
.
[1
0
]
F
.
Kru
g
g
e
l,
“
A
S
im
p
le
M
e
a
su
re
f
o
r
A
c
u
it
y
in
M
e
d
ica
l
Im
a
g
e
s,
”
in
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Ima
g
e
Pro
c
e
ss
in
g
,
v
o
l.
2
7
,
n
o
.
1
1
,
p
p
.
5
2
2
5
–
5
2
3
3
,
No
v
.
2
0
1
8
,
d
o
i:
1
0
.
1
1
0
9
/T
IP
.
2
0
1
8
.
2
8
5
1
6
7
3
.
[1
1
]
M
.
Ya
u
ri
-
M
a
c
h
a
c
a
,
B.
M
e
n
e
se
s
-
Clau
d
i
o
,
N.
V
a
rg
a
s
-
Cu
e
n
tas
,
a
n
d
A
.
Ro
m
a
n
-
G
o
n
z
a
lez
,
“
De
si
g
n
o
f
a
V
e
h
icle
Driv
e
r
Dro
w
sin
e
ss
De
te
c
ti
o
n
S
y
s
tem
T
h
ro
u
g
h
Im
a
g
e
P
ro
c
e
ss
in
g
u
sin
g
M
a
tl
a
b
,
”
2
0
1
8
IE
EE
3
8
th
C
e
n
tra
l
Ame
ric
a
a
n
d
Pa
n
a
m
a
C
o
n
v
e
n
ti
o
n
(
CONCAPA
N
XX
XV
III)
,
S
a
n
S
a
lv
a
d
o
r
,
2
0
1
8
,
p
p
.
1
–
6
,
d
o
i:
1
0
.
1
1
0
9
/CONCA
P
A
N.2
0
1
8
.
8
5
9
6
5
1
3
.
[1
2
]
M
.
A
.
A
.
M
o
sle
h
,
A
.
A
.
AL
-
Y
a
m
n
i
,
a
n
d
A
.
G
u
m
a
e
i,
“
A
n
A
u
to
m
a
ti
c
Nu
c
lei
Ce
ll
s
Co
u
n
ti
n
g
A
p
p
ro
a
c
h
Us
in
g
Eff
e
c
ti
v
e
I
m
a
g
e
P
ro
c
e
ss
in
g
M
e
t
h
o
d
s,
”
2
0
1
9
IEE
E
4
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
ig
n
a
l
a
n
d
Ima
g
e
Pr
o
c
e
ss
in
g
(
ICS
IP)
,
W
u
x
i,
Ch
in
a
,
2
0
1
9
,
p
p
.
8
6
5
–
8
6
9
,
d
o
i:
1
0
.
1
1
0
9
/S
I
P
ROCE
S
S
.
2
0
1
9
.
8
8
6
8
7
5
3
.
[1
3
]
Z
.
A
.
A
l
w
a
n
,
H
.
M
.
F
a
rh
a
n
,
a
n
d
S
.
Q
.
M
a
h
d
i
,
“
Co
l
o
r
im
a
g
e
st
e
g
a
n
o
g
ra
p
h
y
i
n
YCb
Cr
sp
a
c
e
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
, v
ol
.
1
0
,
n
o
.
1
,
p
p
.
2
0
2
–
2
0
9
,
F
e
b
.
2
0
2
0
.
[1
4
]
S
.
Riy
a
d
i,
L
.
M.
Az
iza
h
,
F
.
Ha
k
im
,
C
.
Da
m
a
rjati,
S
.
P
u
sp
i
ta,
a
n
d
E
.
S
o
f
ian
i
,
“
Esti
m
a
ti
o
n
o
f
tertiary
d
e
n
ti
n
th
ick
n
e
ss
o
n
p
u
lp
c
a
p
p
in
g
trea
tme
n
t
w
it
h
d
ig
i
tal
ima
g
e
p
ro
c
e
ss
in
g
tec
h
n
o
lo
g
y
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
ter
E
n
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l
.
1
0
,
n
o
.
1
,
p
p
.
5
2
1
–
5
2
9
,
F
e
b
.
2
0
2
0
.
[1
5
]
C.
L
in
,
S
.
L
in
,
C.
Hw
a
n
g
,
H.
T
u
,
C.
Ch
e
n
,
a
n
d
C.
W
e
n
g
,
“
Re
a
l
-
T
i
m
e
I
m
a
g
e
-
Ba
se
d
De
f
e
c
t
In
sp
e
c
ti
o
n
S
y
s
te
m
o
f
In
tern
a
l
T
h
re
a
d
f
o
r
Nu
t,
”
i
n
IEE
E
T
r
a
n
sa
c
ti
o
n
s
o
n
In
s
tr
u
m
e
n
ta
ti
o
n
a
n
d
M
e
a
su
re
me
n
t
,
v
o
l.
6
8
,
n
o
.
8
,
p
p
.
2
8
3
0
–
2
8
4
8
,
A
u
g
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/T
IM
.
2
0
1
8
.
2
8
7
2
3
1
0
.
[1
6
]
Z.
X
u
tao
,
G
.
F
e
n
g
,
L
.
Ya
n
,
H.
L
i
x
in
,
a
n
d
S
.
L
in
q
i,
“
Re
se
a
rc
h
o
n
n
o
n
-
c
o
n
tac
t
m
e
a
su
re
m
e
n
t
m
e
th
o
d
o
f
b
a
ll
sc
re
w
n
u
t
b
a
se
d
o
n
sp
e
c
tral
c
o
n
f
o
c
a
l
p
ri
n
c
ip
le,”
2
0
1
7
1
3
th
IEE
E
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
El
e
c
tro
n
ic
M
e
a
su
re
me
n
t
&
In
stru
me
n
ts (
ICEM
I)
,
Ya
n
g
z
h
o
u
,
2
0
1
7
,
p
p
.
4
9
3
–
4
9
8
,
d
o
i:
1
0
.
1
1
0
9
/
ICEM
I.
2
0
1
7
.
8
2
6
5
8
6
7
.
[1
7
]
F
.
Kra
e
m
e
r,
M
.
Kle
in
,
a
n
d
M
.
Oe
c
h
sn
e
r,
“
F
a
ti
g
u
e
stre
n
g
th
o
f
m
e
tri
c
ste
e
l
sc
re
w
s
d
e
p
e
n
d
in
g
o
n
p
r
e
-
lo
a
d
a
n
d
n
u
t
ty
p
e
,
”
En
g
in
e
e
rin
g
Fa
i
lu
re
An
a
ly
sis
,
v
o
l.
1
1
2
,
2
0
2
0
,
A
rt.
n
o
.
1
0
4
4
8
4
,
d
o
i:
1
0
.
1
0
1
6
/
j.
e
n
g
f
a
il
a
n
a
l.
2
0
2
0
.
1
0
4
4
8
4
.
[1
8
]
Ch
a
n
g
-
G
u
a
n
g
Z
.
,
Hu
a
-
X
i
Z
.
,
a
n
d
Hu
-
T
ian
F
.
,
“
Ex
p
e
rim
e
n
tal
a
n
a
l
y
sis
o
f
th
e
we
a
r
c
o
e
ff
icie
n
t
o
f
d
o
u
b
le
-
n
u
t
b
a
ll
sc
re
w
s,
”
W
e
a
r
,
2
0
2
0
,
p
p
.
4
4
6
–
4
4
7
,
d
o
i:
1
0
.
1
0
1
6
/j
.
w
e
a
r.
2
0
2
0
.
2
0
3
2
0
1
.
[1
9
]
J
.
O.
P
in
z
ó
n
-
A
re
n
a
s
a
n
d
R
.
Jim
é
n
e
z
-
M
o
re
n
o
,
“
Ob
jec
t
so
rti
n
g
i
n
a
n
e
x
ten
d
e
d
w
o
rk
a
re
a
u
sin
g
c
o
ll
a
b
o
ra
ti
v
e
ro
b
o
ti
c
s
a
n
d
DA
G
-
CNN
,
”
AR
PN
J
o
u
rn
a
l
o
f
E
n
g
i
n
e
e
rin
g
a
n
d
A
p
p
l
ied
S
c
ien
c
e
s
,
v
o
l.
1
5
,
n
o
.
2
,
Ja
n
.
2
0
2
0
.
[2
0
]
J.
J.
v
a
n
V
u
u
re
n
,
L
.
T
a
n
g
,
I.
A
l
-
Ba
h
a
d
ly
,
a
n
d
K.
A
rif
,
“
T
o
w
a
rd
s
th
e
a
u
t
o
n
o
m
o
u
s
ro
b
o
ti
c
g
rip
p
i
n
g
a
n
d
h
a
n
d
l
in
g
o
f
n
o
v
e
l
o
b
jec
ts,”
2
0
1
9
1
4
th
IEE
E
Co
n
fer
e
n
c
e
o
n
In
d
u
stri
a
l
El
e
c
t
ro
n
ics
a
n
d
Ap
p
l
ica
ti
o
n
s
(
ICIEA
)
,
X
i'
a
n
,
Ch
in
a
,
2
0
1
9
,
p
p
.
1
0
0
6
–
1
0
1
0
,
d
o
i:
1
0
.
1
1
0
9
/ICIE
A
.
2
0
1
9
.
8
8
3
3
6
9
1
.
[2
1
]
I
.
Ra
fe
g
a
s,
M
.
V
a
n
re
ll
,
L
.
A
.
A
le
x
a
n
d
re
,
a
n
d
G
.
A
ria
s,
“
Un
d
e
rsta
n
d
in
g
train
e
d
CNN
s
b
y
in
d
e
x
in
g
n
e
u
ro
n
se
lec
ti
v
it
y
,
”
Pa
tt
e
rn
Rec
o
g
n
i
ti
o
n
L
e
tt
e
rs
,
2
0
1
9
.
[2
2
]
H
.
Id
e
,
T
.
Ko
b
a
y
a
sh
i,
K
.
W
a
tan
a
b
e
,
a
n
d
T
.
Ku
rit
a
,
“
Ro
b
u
st
p
r
u
n
in
g
f
o
r
e
f
f
i
c
ien
t
CNN
s,
”
Pa
tt
e
rn
Rec
o
g
n
it
i
o
n
L
e
tt
e
rs
,
v
ol
.
1
3
5
,
p
p
.
90
–
9
8
,
2
0
2
0
.
[2
3
]
J
.
Qin
,
W
.
P
a
n
,
X
.
X
ian
g
,
Y
.
Tan
,
a
n
d
G
.
Ho
u
,
“
A
b
io
lo
g
ica
l
i
m
a
g
e
c
las
sif
i
c
a
ti
o
n
m
e
th
o
d
b
a
se
d
o
n
im
p
ro
v
e
d
CNN
,
”
Eco
lo
g
ica
l
I
n
fo
rm
a
ti
c
s
,
v
ol
.
5
8
,
2
0
2
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:
2
0
8
8
-
8708
R
o
b
o
tic
h
ex
-
n
u
t so
r
tin
g
s
ystem
w
ith
d
ee
p
lea
r
n
in
g
(
C
r
is
tia
n
A
lma
n
z
a
)
3583
[2
4
]
P
ra
n
a
v
K
.
B
.
a
n
d
M
a
n
ik
a
n
d
a
n
J
.
,
“
De
sig
n
a
n
d
Ev
a
lu
a
ti
o
n
o
f
a
Re
a
l
-
T
i
m
e
F
a
c
e
Re
c
o
g
n
it
io
n
S
y
ste
m
u
sin
g
Co
n
v
o
l
u
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
s,
”
Pro
c
e
d
ia
Co
mp
u
ter
S
c
ien
c
e
,
v
ol
.
1
7
1
,
pp.
1
6
5
1
–
1
6
5
9
,
2
0
2
0
.
[2
5
]
S
.
Zh
u
,
X
.
Z
h
e
n
g
,
M
.
Xu
,
Z.
Z
e
n
g
,
a
n
d
H.
Zh
a
n
g
,
“
A
Ro
b
o
ti
c
S
e
m
a
n
ti
c
G
r
a
sp
in
g
M
e
th
o
d
f
o
r
P
ick
-
a
n
d
-
p
lac
e
T
a
s
k
s,
”
2
0
1
9
Ch
in
e
se
Au
to
ma
ti
o
n
Co
n
g
re
ss
(
CAC)
,
H
a
n
g
z
h
o
u
,
Ch
i
n
a
,
2
0
1
9
,
p
p
.
4
1
3
0
–
4
1
3
6
,
d
o
i:
1
0
.
1
1
0
9
/CA
C4
8
6
3
3
.
2
0
1
9
.
8
9
9
6
3
2
8
.
[2
6
]
A
.
A
.
A
b
d
u
lm
u
n
e
m
,
Z
.
A
.
A
b
u
ti
h
e
e
n
,
H
.
J.
A
leq
a
b
ie
,
“
Re
c
o
g
n
it
i
o
n
o
f
c
o
r
o
n
a
v
iru
s
d
ise
a
se
(COV
I
D1
9
)
u
sin
g
d
e
e
p
lea
rn
in
g
n
e
tw
o
rk
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
ol
.
1
1
,
n
o
1
,
pp
.
3
6
5
-
374
,
F
e
b
.
2
0
2
1
.
[2
7
]
J.
L
in
,
L
.
Zh
a
o
,
Q.
Wan
g
,
R
.
W
a
rd
,
a
n
d
Z.
J
.
W
a
n
g
,
“
DT
-
L
ET
:
D
e
e
p
tran
s
f
e
r
l
e
a
rn
in
g
b
y
e
x
p
lo
ri
n
g
w
h
e
re
to
tran
sf
e
r,
”
Ne
u
ro
c
o
mp
u
ti
n
g
,
v
ol
.
3
9
0
,
p
p
.
99
–
1
0
7
,
2
0
2
0
.
[2
8
]
P
.
Us
e
c
h
e
,
R
.
Jim
e
n
e
z
-
M
o
re
n
o
,
a
n
d
J
.
M
.
Ba
q
u
e
ro
,
“
A
lg
o
rit
h
m
o
f
d
e
tec
ti
o
n
,
c
las
sif
ica
ti
o
n
a
n
d
g
rip
p
in
g
o
f
o
c
c
lu
d
e
d
o
b
jec
ts
b
y
CNN
tec
h
n
i
q
u
e
s
a
n
d
Ha
a
r
c
las
si
f
iers
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
m
p
u
t
e
r
En
g
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
ol
.
1
0
,
n
o
.
5
,
pp
.
4
7
1
2
-
4
7
2
0
,
Oc
t
.
2
0
2
0
.
[2
9
]
R
.
Jim
é
n
e
z
-
M
o
re
n
o
,
J
.
O
.
P
in
z
ó
n
-
A
re
n
a
s,
a
n
d
C
.
G
.
P
a
c
h
ó
n
-
S
u
e
sc
ú
n
,
“
A
ss
ist
a
n
t
ro
b
o
t
th
r
o
u
g
h
d
e
e
p
lea
rn
in
g
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
, v
ol
.
1
0
,
n
o
.
1
,
pp
.
1
0
5
3
-
1
0
6
2
,
F
e
b
.
2
0
2
0
.
[3
0
]
T.
S
u
ti
k
n
o
,
L
.
Ha
n
d
a
y
a
n
i,
D
.
S
t
iaw
a
n
,
M
.
A
.
Riy
a
d
i,
a
n
d
I
.
M
.
I
.
S
u
b
r
o
to
,
“
W
h
a
tsA
p
p
,
V
i
b
e
r
a
n
d
T
e
leg
ra
m
w
h
ich
is
b
e
st
f
o
r
In
sta
n
t
M
e
ss
a
g
in
g
?
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
6
,
n
o
.
3
,
p
p
.
9
0
9
–
9
1
4
,
Ju
n
.
2
0
1
6
,
doi
:
1
0
.
1
1
5
9
1
/
ij
e
c
e
.
v
6
i3
.
1
0
2
7
1
.
[3
1
]
C.
S
a
ra
v
a
n
a
n
,
“
Co
l
o
r
Im
a
g
e
to
Gra
y
sc
a
le
I
m
a
g
e
Co
n
v
e
rsio
n
,
”
2
0
1
0
S
e
c
o
n
d
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
C
o
mp
u
ter
En
g
i
n
e
e
rin
g
a
n
d
A
p
p
li
c
a
ti
o
n
s
,
Ba
li
Isla
n
d
,
2
0
1
0
,
p
p
.
1
9
6
–
1
9
9
,
d
o
i:
1
0
.
1
1
0
9
/ICC
EA
.
2
0
1
0
.
1
9
2
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Cr
istia
n
Al
m
a
n
z
a
L
e
o
n
is
a
n
El
e
c
tro
n
ic
En
g
in
e
e
r,
g
ra
d
u
a
ted
f
ro
m
th
e
d
e
lo
s
L
lan
o
s
Un
iv
e
rsit
y
,
Co
lo
m
b
ia
in
2
0
1
8
a
n
d
w
it
h
g
re
a
t
in
tere
sts
a
n
d
sk
il
ls
in
th
e
f
ield
o
f
r
e
se
a
r
c
h
,
m
o
re
sp
e
c
i
f
ica
ll
y
in
a
re
a
s su
c
h
a
s ro
b
o
t
ics
,
e
lec
tro
n
ic
c
ircu
it
s a
n
d
p
ro
g
ra
m
m
in
g
Ro
b
i
n
s
o
n
J
i
m
é
n
e
z
-
M
o
r
e
n
o
is
a
n
El
e
c
tro
n
ic
En
g
in
e
e
r
g
ra
d
u
a
te
d
f
ro
m
th
e
F
ra
n
c
isc
o
Jo
sé
d
e
Ca
ld
a
s
District
Un
iv
e
rsity
in
2
0
0
2
.
He
re
c
e
iv
e
d
a
M
.
S
c
.
in
En
g
in
e
e
rin
g
f
ro
m
th
e
Na
ti
o
n
a
l
Un
iv
e
rsit
y
o
f
Co
lo
m
b
ia
in
2
0
1
2
a
n
d
P
h
.
D
i
n
En
g
i
n
e
e
rin
g
a
t
th
e
F
ra
n
c
isc
o
Jo
sé
d
e
Ca
ld
a
s
District
Un
iv
e
rsit
y
in
2
0
1
8
.
His
c
u
rre
n
t
w
o
rk
in
g
a
s
a
ss
istan
t
p
ro
f
e
ss
o
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
re
se
a
rc
h
f
o
c
u
se
s
o
n
th
e
u
se
o
f
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s
f
o
r
o
b
jec
t
re
c
o
g
n
it
i
o
n
a
n
d
im
a
g
e
p
ro
c
e
ss
in
g
f
o
r
ro
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
in
e
in
tera
c
ti
o
n
.
J
a
v
ier
Ed
u
a
r
d
o
M
a
r
tin
e
z
B
a
q
u
e
r
o
is
a
n
El
e
c
tro
n
ic
E
n
g
in
e
e
r
g
ra
d
u
a
ted
f
ro
m
d
e
lo
s
L
lan
o
s
Un
iv
e
rsit
y
in
2
0
0
2
.
P
o
sg
ra
d
u
a
te
d
in
El
e
c
tro
n
ic
In
stru
m
e
n
tatio
n
f
ro
m
S
a
n
to
T
o
m
a
s
Un
iv
e
rsit
y
i
n
2
0
0
4
a
n
d
M
sc
in
Ed
u
c
a
ti
v
e
T
e
c
h
n
o
lo
g
y
a
t
A
u
to
n
o
m
a
o
f
Bu
c
a
ra
m
a
n
g
a
Un
iv
e
rsit
y
in
2
0
1
3
.
His
c
u
rre
n
t
w
o
rk
in
g
a
s
a
ss
istan
t
p
ro
f
e
ss
o
r
o
f
d
e
lo
s
L
ln
a
o
s
Un
iv
e
rsity
a
n
d
re
se
a
r
c
h
f
o
c
u
se
s
o
n
In
stru
m
e
n
tatio
n
,
A
u
to
m
a
ti
o
n
a
n
d
Co
n
tr
o
l.
Evaluation Warning : The document was created with Spire.PDF for Python.