T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
2
2
,
No
.
6
,
Dec
em
b
er
202
4
,
p
p
.
1
4
1
4
~1
421
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
2
2
i
6
.
25519
1414
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//telko
mn
ika
.
u
a
d
.
a
c.
i
d
To
m
a
to
lea
f
disea
se recog
nition sy
s
te
m
using
F
a
ster
R
-
CN
N
K
a
re
l O
ct
a
v
ia
nu
s
B
a
chri,
B
r
y
a
n Sa
nto
s
o
,
Du
m
a
K
ristina
Ya
nti
H
uta
pea
,
Ca
t
herine
O
liv
ia
Sere
a
t
i,
L
a
nn
y
W.
P
a
nd
j
a
it
a
n
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
A
t
ma
Jay
a
C
a
t
h
o
l
i
c
U
n
i
v
e
r
si
t
y
o
f
I
n
d
o
n
e
si
a
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
8
,
2023
R
ev
i
s
ed
J
u
l
22
,
2
0
2
4
A
cc
ep
ted
A
u
g
5
,
2024
T
h
e
o
b
jec
ti
v
e
o
f
th
is
p
a
p
e
r
is
t
o
d
e
tec
t
t
o
m
a
to
lea
f
d
ise
a
se
u
sin
g
F
a
ste
r
re
g
io
n
-
b
a
se
d
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
(
R
-
CN
N
)
.
T
h
e
to
m
a
to
lea
f
d
ise
a
se
re
c
o
g
n
it
io
n
sy
ste
m
u
ti
li
z
e
s
a
d
a
tas
e
t
c
o
n
sistin
g
o
f
h
e
a
lt
h
y
to
m
a
to
lea
v
e
s
a
n
d
e
ig
h
t
lea
f
d
ise
a
se
s,
in
c
lu
d
i
n
g
e
a
rly
b
li
g
h
t,
late
b
li
g
h
t
,
l
e
a
f
m
o
ld
,
m
o
sa
ic
v
iru
s
,
se
p
to
ria,
sp
id
e
r
m
it
e
s,
y
e
ll
o
w
l
e
a
f
c
u
rl
v
iru
s
,
a
n
d
l
e
a
f
m
in
e
r
.
T
h
e
d
a
tas
e
t
is
o
b
tai
n
e
d
f
ro
m
v
a
rio
u
s
so
u
rc
e
s,
su
c
h
a
s
Ka
g
g
l
e
,
G
o
o
g
le
Im
a
g
e
s,
Bin
g
Im
a
g
e
s,
a
n
d
Ro
b
o
f
lo
w
Un
iv
e
rse
.
P
re
-
p
ro
c
e
ss
in
g
tec
h
n
iq
u
e
s
,
in
c
lu
d
in
g
c
o
ll
a
g
e
,
ti
le,
sta
ti
c
c
r
o
p
,
a
n
d
re
siz
e
,
a
re
a
p
p
li
e
d
to
p
re
p
a
re
th
e
d
a
tas
e
t
f
o
r
train
in
g
.
Da
ta
a
u
g
m
e
n
tatio
n
m
e
th
o
d
s,
su
c
h
a
s
f
li
p
p
i
n
g
,
9
0
°
ro
tatio
n
,
e
x
p
o
su
re
a
d
ju
stm
e
n
t,
a
n
d
h
u
e
m
o
d
if
ica
ti
o
n
,
a
re
a
p
p
li
e
d
t
o
e
n
h
a
n
c
e
th
e
m
o
d
e
l
’
s
ro
b
u
stn
e
ss
a
n
d
g
e
n
e
ra
li
z
e
it
s
p
e
rf
o
rm
a
n
c
e
.
S
p
e
c
ifi
c
a
ll
y
,
w
e
im
p
le
m
e
n
ted
F
a
ste
r
R
-
CNN
a
s
p
a
rt
o
f
De
tec
tro
n
2
u
sin
g
it
s
b
a
se
m
o
d
e
ls
a
n
d
c
o
n
f
ig
u
ra
ti
o
n
s.
T
h
e
re
su
lt
s
d
e
m
o
n
stra
te
th
a
t
th
e
X
1
0
1
-
F
P
N
b
a
se
m
o
d
e
l
f
o
r
F
a
ste
r
R
-
CNN
w
it
h
th
e
d
e
f
a
u
lt
c
o
n
f
ig
u
ra
ti
o
n
s
o
f
De
tec
tro
n
2
is
e
fficie
n
t
a
n
d
g
e
n
e
ra
l
e
n
o
u
g
h
to
b
e
a
p
p
li
e
d
to
d
e
fe
c
t
d
e
tec
ti
o
n
.
T
h
is
a
p
p
ro
a
c
h
re
su
lt
s
i
n
a
n
a
v
e
ra
g
e
p
re
c
isio
n
(A
P
)
d
e
tec
ti
o
n
sc
o
re
o
f
8
7
.
0
1
%
f
o
r
v
a
li
d
a
ti
o
n
re
su
lt
s.
K
ey
w
o
r
d
s
:
Dete
ctr
o
n
2
Fas
ter
r
eg
io
n
-
b
ased
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
w
o
r
k
Ma
ch
i
n
e
lear
n
i
n
g
Ob
j
ec
t d
etec
tio
n
T
o
m
ato
lea
f
d
is
ea
s
e
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
:
Kar
el
Octa
v
ia
n
u
s
B
ac
h
r
i
Dep
ar
t
m
en
t o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
,
Fac
u
lt
y
o
f
E
n
g
i
n
ee
r
in
g
A
t
m
a
J
a
y
a
C
a
th
o
lic
U
n
i
v
er
s
it
y
o
f
I
n
d
o
n
e
s
ia
J
ak
ar
ta,
I
n
d
o
n
esia
E
m
ail:
k
ar
el.
b
ac
h
r
i@
a
t
m
aj
a
y
a
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
I
n
to
m
ato
c
u
lti
v
atio
n
,
v
ar
io
u
s
d
is
ea
s
es
ca
u
s
ed
b
y
p
est
s
s
u
c
h
as
in
s
ec
ts
,
f
u
n
g
i,
b
ac
ter
ia,
a
n
d
v
ir
u
s
e
s
[
1
]
a
r
e
o
f
ten
en
co
u
n
ter
ed
,
w
h
i
ch
ca
n
af
f
ec
t
th
e
y
ield
an
d
q
u
alit
y
o
f
to
m
ato
es.
T
h
er
e
is
also
d
is
ea
s
e
o
n
to
m
ato
leav
es
i
n
Keb
u
n
I
b
u
&
C
o
.
Fi
g
u
r
e
1
(
a)
s
h
o
w
s
t
h
e
i
m
ag
e
o
f
h
ea
lt
h
y
lea
v
es,
w
h
ile
Fi
g
u
r
e
1
(
b
)
d
is
ea
s
ed
leav
es,
w
h
ic
h
af
f
ec
ted
b
y
leaf
m
i
n
er
.
Dis
ea
s
e
s
i
n
to
m
ato
p
la
n
ts
h
a
v
e
d
if
f
er
e
n
t
c
h
ar
ac
ter
is
tic
s
f
r
o
m
o
n
e
a
n
o
th
er
.
T
h
e
ch
ar
ac
ter
is
tic
o
f
to
m
ato
d
is
e
ases
is
s
h
o
w
n
i
n
T
ab
le
1
.
H
o
w
e
v
er
,
f
ar
m
er
s
m
a
y
n
o
t
b
e
ab
le
to
id
en
tify
o
r
m
is
id
en
ti
f
y
d
is
ea
s
es
i
n
to
m
ato
p
lan
ts
,
r
esu
lti
n
g
in
p
lan
t
d
ea
t
h
.
Mo
r
eo
v
er
,
m
is
d
ia
g
n
o
s
is
o
r
m
is
id
en
ti
f
icatio
n
o
f
d
is
ea
s
es i
n
to
m
ato
p
lan
ts
ca
n
l
ea
d
to
im
p
r
o
p
er
p
esti
cid
e
ap
p
l
icatio
n
,
ca
u
s
in
g
d
a
m
a
g
e
o
r
ev
en
cr
o
p
f
ailu
r
e
[
2
]
.
T
o
a
d
d
r
e
s
s
th
is
p
r
o
b
l
em
,
an
au
to
m
a
t
e
d
s
y
s
t
em
is
r
e
q
u
i
r
e
d
t
o
d
e
t
e
c
t
p
l
an
t
d
i
s
e
as
es
.
T
h
is
c
an
b
e
s
o
lv
e
d
if
t
h
e
im
ag
es
o
f
th
e
p
l
an
t
d
i
s
e
as
e
s
c
an
b
e
c
a
p
t
u
r
e
d
an
d
d
e
t
ec
t
e
d
s
p
e
c
if
i
ca
l
ly
o
n
t
o
m
a
t
o
le
a
v
es
u
s
in
g
o
b
je
c
t
d
e
t
e
c
t
i
o
n
m
e
th
o
d
s
.
H
en
c
e
,
a
m
o
d
e
l
w
ith
h
ig
h
a
c
cu
r
a
cy
is
r
e
q
u
i
r
e
d
,
s
u
c
h
as
F
a
s
t
e
r
r
e
g
i
o
n
-
b
as
e
d
c
o
n
v
o
lu
ti
o
n
a
l
n
eu
r
a
l
n
e
t
w
o
r
k
(
R
-
C
N
N
)
,
w
h
ic
h
p
r
o
v
i
d
e
s
h
ig
h
e
r
s
p
e
e
d
an
d
a
c
c
u
r
ac
y
c
o
m
p
a
r
e
d
t
o
R
-
C
N
N
a
n
d
F
as
t
R
-
C
NN
m
o
d
e
ls
[
3
]
-
[
5
]
.
I
n
th
e
r
e
s
e
a
r
c
h
c
o
n
d
u
ct
e
d
b
y
s
e
v
e
r
al
r
e
s
e
a
r
ch
e
r
s
[
6
]
-
[
9
]
,
it
h
as
b
e
en
f
o
u
n
d
th
a
t
m
o
d
e
ls
c
at
eg
o
r
i
z
e
d
as
tw
o
-
s
ta
g
e
d
e
t
ec
t
o
r
s
a
ch
i
ev
e
h
ig
h
e
r
ac
cu
r
a
cy
c
o
m
p
a
r
e
d
t
o
o
n
e
-
s
t
ag
e
d
e
t
e
ct
o
r
s
.
I
n
p
l
an
t
d
is
e
a
s
e
d
e
t
ec
t
i
o
n
s
y
s
tem
s
,
h
ig
h
a
c
cu
r
a
cy
is
c
r
u
c
ia
l
.
T
h
e
r
ef
o
r
e
,
t
h
i
s
r
e
s
e
a
r
ch
w
il
l u
t
i
li
z
e
a
t
w
o
-
s
t
ag
e
d
e
t
e
ct
o
r
m
o
d
e
l
,
s
u
ch
a
s
F
as
t
er
R
-
C
N
N
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
To
ma
to
lea
f d
is
ea
s
e
r
ec
o
g
n
iti
o
n
s
ystem
u
s
in
g
F
a
s
ter R
-
C
N
N
(
K
a
r
el
Octa
via
n
u
s
B
a
ch
r
i
)
1415
(
a)
(
b
)
Fi
g
u
r
e
1
.
T
o
m
ato
p
lan
t i
n
K
eb
u
n
I
b
u
&
C
o
: (
a)
h
ea
lt
h
y
leav
e
s
an
d
(
b
)
a
f
f
ec
ted
leav
e
s
b
y
le
af
m
in
er
T
ab
le
1
.
C
h
ar
ac
ter
is
tics
o
f
to
m
ato
p
lan
t d
is
ea
s
e
s
[
1
0
]
,
[
1
1
]
I
mag
e
C
h
a
r
a
c
t
e
r
i
st
i
c
/
sy
mp
t
o
m
I
mag
e
C
h
a
r
a
c
t
e
r
i
st
i
c
/
sy
mp
t
o
m
I
mag
e
C
h
a
r
a
c
t
e
r
i
st
i
c
/
sy
mp
t
o
m
I
mag
e
C
h
a
r
a
c
t
e
r
i
st
i
c
/
sy
mp
t
o
m
Ea
r
l
y
b
l
i
g
h
t
C
h
a
r
a
c
t
e
r
i
z
e
d
b
y
c
o
n
c
e
n
t
r
i
c
r
i
n
g
s a
n
d
su
r
r
o
u
n
d
e
d
b
y
a
y
e
l
l
o
w
i
n
g
z
o
n
e
[
1
2
]
.
L
a
t
e
b
l
i
g
h
t
L
e
si
o
n
s a
r
e
c
h
a
r
a
c
t
e
r
i
z
e
d
b
y
b
e
i
n
g
smal
l
,
d
a
r
k
,
a
n
d
r
e
se
mb
l
i
n
g
w
a
t
e
r
-
s
o
a
k
e
d
sp
o
t
s
[
1
3
]
.
Y
e
l
l
o
w
l
e
a
f
c
u
r
l
v
i
r
u
s
U
p
w
a
r
d
c
u
r
l
i
n
g
o
f
l
e
a
f
e
d
g
e
s
(
c
u
p
p
i
n
g
)
,
t
h
i
c
k
e
n
i
n
g
o
f
l
e
a
f
v
e
i
n
s,
y
e
l
l
o
w
i
n
g
a
n
d
l
e
a
f
c
u
r
l
i
n
g
[
1
4
]
.
M
o
sai
c
v
i
r
u
s
P
r
e
se
n
c
e
o
f
mo
sai
c
p
a
t
t
e
r
n
s
a
n
d
sp
o
t
s o
n
t
o
mat
o
l
e
a
v
e
s
[
1
5
]
.
S
e
p
t
o
r
i
a
M
a
n
y
smal
l
r
o
u
n
d
s
p
o
t
s o
f
l
i
g
h
t
b
r
o
w
n
c
o
l
o
r
w
i
t
h
w
h
i
t
e
c
e
n
t
e
r
s
[
1
5
]
.
L
e
a
f
mo
l
d
P
a
l
e
g
r
e
e
n
o
r
y
e
l
l
o
w
i
sh
p
a
t
c
h
e
s o
n
t
h
e
u
p
p
e
r
su
r
f
a
c
e
o
f
l
e
a
v
e
s [
1
3
]
.
S
p
i
d
e
r
mi
t
e
s
Y
e
l
l
o
w
i
sh
w
h
i
t
e
sp
o
t
s
[
1
6
]
.
L
e
a
f
mi
n
e
r
T
h
e
l
e
a
v
e
s
a
r
e
v
e
i
n
e
d
a
n
d
s
e
r
p
e
n
t
i
n
e
-
l
i
k
e
p
a
t
t
e
r
n
s
[
1
7
]
.
T
h
e
to
m
ato
leaf
d
is
ea
s
e
r
ec
o
g
n
i
tio
n
s
y
s
te
m
u
ti
lizes
th
e
F
aster
R
-
C
NN
w
it
h
th
e
f
ea
t
u
r
e
p
y
r
a
m
i
d
n
et
w
o
r
k
s
(
X1
0
1
-
FP
N
)
m
o
d
el
.
T
h
is
m
o
d
el
h
as
t
h
e
h
i
g
h
e
s
t
s
co
r
e
f
o
r
o
b
j
ec
t
d
etec
tio
n
a
m
o
n
g
th
e
av
a
ilab
le
m
o
d
el
s
in
t
h
e
Dete
ctr
o
n
2
lib
r
ar
y
[
1
8
]
.
Dete
ctr
o
n
2
is
an
o
b
jec
t
d
etec
tio
n
f
r
a
m
e
w
o
r
k
d
ev
e
lo
p
ed
b
y
Face
b
o
o
k
A
I
r
esear
ch
(
F
A
I
R
)
.
I
t
is
th
e
l
atest
v
er
s
io
n
o
f
th
e
p
r
ev
io
u
s
o
b
j
ec
t
d
etec
tio
n
f
r
am
e
w
o
r
k
k
n
o
w
n
as
Dete
ctr
o
n
.
Dete
ctr
o
n
2
is
b
u
i
lt
o
n
to
p
o
f
P
y
T
o
r
ch
,
a
d
ee
p
-
lear
n
in
g
f
r
a
m
e
w
o
r
k
t
h
at
p
r
o
v
id
es
co
m
p
r
eh
en
s
i
v
e
f
u
n
ctio
n
a
li
t
y
f
o
r
o
b
j
ec
t
d
etec
tio
n
r
esear
ch
an
d
d
ev
elo
p
m
en
t.
T
h
e
p
r
e
-
tr
ain
ed
m
o
d
els
a
v
ailab
le
in
t
h
e
Det
ec
tr
o
n
2
m
o
d
el
zo
o
h
av
e
b
ee
n
tr
ain
ed
o
n
th
e
co
m
m
o
n
o
b
j
ec
ts
in
co
n
tex
t
(
C
O
C
O)
d
ataset.
T
h
er
ef
o
r
e,
d
u
r
in
g
tr
ain
i
n
g
,
f
i
n
e
-
tu
n
i
n
g
is
p
er
f
o
r
m
ed
o
n
th
e
p
r
e
-
tr
ai
n
ed
m
o
d
els
.
Dete
ctr
o
n
2
o
f
f
e
r
s
s
ev
er
al
t
y
p
e
s
o
f
o
b
j
ec
t
d
etec
tio
n
,
in
cl
u
d
i
n
g
b
o
u
n
d
in
g
-
b
o
x
d
etec
tio
n
,
m
a
s
k
,
k
e
y
p
o
i
n
t
d
etec
tio
n
,
i
n
s
ta
n
ce
s
e
g
m
en
tatio
n
,
s
e
m
a
n
tic
s
eg
m
e
n
tatio
n
,
a
n
d
p
an
o
p
tic
s
eg
m
en
tatio
n
.
I
n
t
h
i
s
r
esear
ch
,
th
e
t
y
p
e
o
f
o
b
j
ec
t d
e
tectio
n
u
s
ed
is
b
o
u
n
d
i
n
g
-
b
o
x
d
etec
tio
n
.
T
h
er
e
is
ac
tiv
e
r
esear
ch
in
th
is
ar
ea
.
Fo
r
ex
a
m
p
le,
u
s
i
n
g
Fas
ter
R
-
C
NN
w
it
h
r
esid
u
al
n
et
w
o
r
k
s
(
r
es1
0
1
)
b
ased
m
o
d
el
to
d
ete
ct
h
ea
lt
h
y
to
m
ato
lea
v
es
a
n
d
f
o
u
r
lea
f
d
i
s
ea
s
es
:
p
o
w
d
er
y
m
ild
e
w
,
b
li
g
h
t,
lea
f
m
o
ld
f
u
n
g
u
s
,
an
d
to
m
ato
m
o
s
aic
v
ir
u
s
(
T
o
MV
)
,
t
h
e
r
e
s
u
l
t
in
g
m
o
d
el
a
ch
i
ev
es
m
e
an
a
v
e
r
a
g
e
p
r
e
c
i
s
i
o
n
(
m
A
P
)
o
f
9
8
.
5
4
%
[
1
9
]
.
A
l
s
o
,
u
s
in
g
an
o
b
je
c
t
d
e
t
e
c
ti
o
n
m
eth
o
d
w
i
th
Fa
s
te
r
R
-
C
NN
,
r
eg
i
o
n
-
b
as
e
d
f
u
l
ly
c
o
n
v
o
lu
ti
o
n
al
n
e
t
w
o
r
k
s
(
R
-
FC
N
)
,
an
d
s
i
n
g
l
e
s
h
o
t
d
etec
to
r
(
SSD
)
al
g
o
r
ith
m
s
m
er
g
ed
w
it
h
t
h
e
late
s
t
f
ea
tu
r
e
e
x
tr
ac
to
r
s
o
f
d
ee
p
C
NN
to
d
etec
t
to
m
a
to
leaf
d
is
ea
s
e:
lea
f
m
o
ld
,
g
r
a
y
m
o
ld
,
ca
n
k
er
,
p
lag
u
e,
m
in
er
,
lo
w
te
m
p
er
atu
r
e,
p
o
w
d
er
y
m
ild
e
w
,
w
h
itef
l
y
,
a
n
d
n
u
tr
it
io
n
al
ex
ce
s
s
.
I
n
th
is
p
r
o
p
o
s
ed
m
o
d
el,
R
-
C
N
N
w
i
th
v
i
s
u
al
g
eo
m
etr
y
g
r
o
u
p
1
6
(
VGG
-
16
)
s
h
o
w
s
b
e
tter
r
ec
o
g
n
itio
n
r
es
u
lt
s
,
w
h
ic
h
ac
h
iev
e
s
m
A
P
8
5
.
9
8
%
[
2
0
]
.
An
o
th
er
ex
a
m
p
le
is
th
e
u
s
e
o
f
C
NN
to
d
etec
t
to
m
ato
leaf
d
is
ea
s
e
o
n
a
d
ataset
co
n
s
is
tin
g
o
f
late
b
li
g
h
t,
g
r
a
y
s
p
o
t,
an
d
b
ac
ter
ial
ca
n
k
er
.
T
h
is
m
o
d
el
ac
h
iev
e
s
a
n
ac
cu
r
ac
y
o
f
8
9
% [
2
1
]
.
F
a
s
t
e
r
R
C
N
N
h
a
s
a
ls
o
b
e
en
u
s
e
d
in
t
o
m
a
t
o
d
i
s
e
a
s
e
n
o
t
u
s
in
g
t
h
e
l
e
af
im
ag
es
,
h
o
w
ev
e
r
,
th
e
r
e
a
r
e
s
o
m
e
d
i
s
t
in
ct
i
o
n
f
r
o
m
th
i
s
r
es
e
a
r
ch
,
s
o
m
e
u
s
e
th
e
b
e
r
r
y
in
s
t
e
a
d
o
f
t
h
e
l
e
av
e
s
[
2
2
]
.
I
n
o
th
e
r
r
es
e
a
r
c
h
,
th
e
s
y
s
tem
w
as
b
u
i
lt
u
s
in
g
R
e
t
in
aN
e
t
an
d
f
o
cu
s
e
d
o
n
ly
o
n
l
a
t
e
b
l
ig
h
t
,
m
o
s
a
i
c
v
i
r
us
,
t
o
m
a
t
o
s
e
p
t
o
r
i
a
l
e
af
s
p
o
t
[
2
3
]
,
w
h
i
le
o
th
er
u
s
i
n
g
r
ea
l
-
t
i
m
e
f
a
s
ter
r
eg
io
n
al
co
n
v
o
lu
t
io
n
al
n
e
u
r
al
n
et
wo
r
k
(
R
T
F
-
R
C
N
N
)
ar
ch
itectu
r
e
to
id
en
tify
f
iv
e
d
is
ea
s
e,
w
h
ic
h
ar
e
y
ello
w
lea
f
cu
r
l,
s
ep
to
r
ia
leaf
s
p
o
t,
late
b
lig
h
t,
b
ac
ter
ial
s
p
o
t,
an
d
b
lig
h
t
w
it
h
th
e
to
tal
o
f
s
a
m
p
le
o
f
o
n
l
y
1
2
5
0
0
[
2
4
]
.
T
h
is
p
ap
er
f
o
cu
s
e
s
o
n
d
ev
elo
p
i
n
g
a
p
r
ed
ictio
n
m
o
d
el
f
o
r
d
etec
tin
g
v
ar
io
u
s
t
y
p
e
s
o
f
d
is
ea
s
e
s
o
n
to
m
a
to
leav
e
s
,
in
cl
u
d
in
g
ea
r
l
y
b
li
g
h
t,
late
b
lig
h
t,
leaf
m
o
ld
,
m
o
s
aic
v
ir
u
s
,
s
ep
to
r
ia,
s
p
id
er
m
ite
s
,
y
ello
w
lea
f
cu
r
l
v
ir
u
s
,
leaf
m
i
n
er
,
an
d
h
ea
lth
y
to
m
ato
leav
es
u
s
i
n
g
co
m
b
in
ed
d
ataset
tak
e
n
f
r
o
m
v
ar
io
u
s
o
p
en
s
o
u
r
ce
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
2
,
No
.
6
,
Dec
em
b
er
20
2
4
:
1
4
1
4
-
1
421
1416
2.
M
E
T
H
O
D
T
h
is
s
ec
tio
n
d
is
c
u
s
s
es
t
h
e
m
e
th
o
d
u
s
ed
to
ac
h
ie
v
e
th
e
o
b
j
ec
tiv
e
.
T
h
e
f
ir
s
t
p
ar
t
o
f
t
h
is
s
e
ctio
n
is
t
h
e
m
et
h
o
d
u
s
ed
,
w
h
ich
is
Fa
s
ter
R
-
C
NN
X
-
1
0
1
FP
N.
T
h
e
s
ec
o
n
d
p
ar
t
o
f
th
is
s
ec
tio
n
is
d
ata
co
llectio
n
an
d
an
n
o
tatio
n
,
w
h
ich
e
x
p
lai
n
s
t
h
e
v
ar
io
u
s
s
o
u
r
ce
s
o
f
t
h
e
d
atas
et
an
d
th
e
lab
el
co
r
r
esp
o
n
d
s
t
o
th
e
d
ata
.
T
h
e
th
ir
d
p
ar
t
is
d
ata
p
r
e
p
r
o
ce
s
s
in
g
an
d
au
g
m
e
n
tatio
n
,
w
h
ic
h
p
r
ep
a
r
es
th
e
d
ata
to
b
e
r
ea
d
y
to
b
e
p
r
o
ce
s
s
ed
b
y
th
e
s
y
s
te
m
.
T
h
e
f
o
u
r
th
p
ar
t
o
f
t
h
i
s
s
ec
tio
n
e
x
p
lain
s
th
e
h
y
p
er
p
ar
a
m
eter
u
s
ed
in
tr
ai
n
i
n
g
.
T
h
e
t
esti
n
g
p
h
a
s
e
o
f
t
h
is
r
esear
ch
is
ex
p
lai
n
ed
in
t
h
e
r
e
s
u
lt a
n
d
d
is
c
u
s
s
io
n
.
2
.
1
.
F
a
s
t
er
R
-
CNN
X
-
1
0
1
F
P
N
Fas
ter
R
-
C
NN
is
a
t
w
o
-
s
ta
g
e
d
etec
to
r
m
o
d
el
th
at
u
til
izes
F
ast
R
-
C
NN
an
d
r
eg
io
n
p
r
o
p
o
s
al
n
et
w
o
r
k
(
R
P
N)
as
its
m
ain
ar
c
h
itect
u
r
e.
T
h
is
m
o
d
el
is
a
n
i
m
p
r
o
v
e
m
en
t
o
v
er
t
h
e
p
r
ev
io
u
s
m
o
d
e
l,
Fas
t
R
-
C
NN,
b
y
r
ep
lacin
g
t
h
e
s
elec
t
iv
e
s
ea
r
ch
m
et
h
o
d
w
it
h
R
P
N.
R
P
N
is
r
esp
o
n
s
ib
le
f
o
r
g
e
n
er
ati
n
g
r
e
g
io
n
p
r
o
p
o
s
als
th
a
t
p
o
ten
tiall
y
co
n
ta
in
o
b
j
ec
ts
i
n
an
i
m
a
g
e.
R
P
N
g
en
er
ates
r
eg
io
n
p
r
o
p
o
s
als
b
y
p
er
f
o
r
m
i
n
g
r
eg
r
es
s
io
n
an
d
class
i
f
icatio
n
to
d
eter
m
in
e
t
h
e
lo
ca
tio
n
an
d
p
r
esen
ce
o
f
o
b
j
ec
ts
in
ea
c
h
p
r
o
p
o
s
al
[
2
5
]
.
I
n
g
e
n
er
al,
an
o
b
j
ec
t
d
etec
to
r
co
n
s
is
t
s
o
f
th
r
ee
co
m
p
o
n
en
ts
:
th
e
b
ac
k
b
o
n
e,
n
ec
k
,
an
d
h
ea
d
.
T
h
e
b
ac
k
b
o
n
e
is
r
esp
o
n
s
ib
le
f
o
r
f
e
atu
r
e
ex
tr
ac
tio
n
f
r
o
m
th
e
i
n
p
u
t
i
m
a
g
e,
th
e
n
ec
k
is
u
s
ed
f
o
r
f
ea
tu
r
e
f
u
s
io
n
,
an
d
th
e
h
ea
d
i
s
u
s
ed
f
o
r
p
r
ed
ictio
n
[
2
6
]
.
T
h
e
Fas
ter
R
-
C
NN
X1
0
1
-
FP
N
m
o
d
els
co
n
s
i
s
t
o
f
th
r
ee
co
m
p
o
n
e
n
ts
,
w
h
ic
h
ar
e:
−
B
ac
k
b
o
n
e:
t
y
p
icall
y
,
o
b
j
ec
t
d
e
tecto
r
s
u
s
e
VG
G,
Den
s
eNe
t,
o
r
r
esid
u
al
n
et
w
o
r
k
(
R
esNet
)
a
s
th
e
b
ac
k
b
o
n
e
o
r
u
n
d
er
l
y
in
g
n
et
w
o
r
k
[
2
7
]
.
Fas
ter
R
-
C
N
N
X1
0
1
-
FP
N
u
s
e
R
esNet
as
th
eir
b
ac
k
b
o
n
e
o
r
b
ase
n
et
w
o
r
k
.
R
esNet
is
a
t
y
p
e
o
f
n
eu
r
al
n
et
w
o
r
k
i
n
tr
o
d
u
ce
d
b
y
Kai
m
i
n
g
He,
Xian
g
y
u
Z
h
an
g
,
S
h
ao
q
in
g
R
e
n
,
a
n
d
J
ian
Su
n
i
n
2
0
1
5
.
I
n
ad
d
itio
n
to
R
esNet,
t
h
er
e
ar
e
o
th
er
t
y
p
e
s
o
f
R
es
Net
ar
ch
i
tectu
r
e
s
s
u
c
h
as
R
esNet5
0
,
R
esNet1
0
1
,
an
d
R
esNet1
5
2
.
I
n
2
0
1
5
,
th
is
n
et
w
o
r
k
s
u
cc
es
s
f
u
ll
y
ac
h
ie
v
ed
f
ir
s
t
p
lace
in
th
e
I
m
a
g
eNe
t
lar
g
e
s
ca
le
v
i
s
u
al
r
ec
o
g
n
itio
n
ch
all
en
g
e
(
I
L
SV
R
C
)
an
d
C
OC
O
co
m
p
eti
tio
n
s
in
ter
m
s
o
f
i
m
a
g
e
class
i
f
icatio
n
,
o
b
j
ec
t d
etec
tio
n
,
an
d
i
m
a
g
e
s
e
g
m
e
n
tat
io
n
[
2
8
]
.
T
h
e
f
o
u
n
d
ati
o
n
o
f
t
h
e
R
e
s
Net
ar
ch
itect
u
r
e
i
s
a
co
n
v
o
l
u
tio
n
n
eu
r
al
n
e
t
w
o
r
k
(
C
NN)
th
at
c
o
n
s
is
ts
o
f
co
n
v
o
lu
tio
n
al
la
y
er
s
,
p
o
o
lin
g
la
y
er
s
,
an
d
f
u
ll
y
c
o
n
n
ec
ted
la
y
er
s
,
w
it
h
t
h
e
ad
d
itio
n
o
f
s
k
ip
co
n
n
ec
tio
n
s
in
s
o
m
e
co
n
v
o
lu
t
io
n
al
la
y
er
s
.
T
h
e
p
r
in
cip
le
b
eh
in
d
R
esNet
is
to
cr
ea
te
a
d
ee
p
er
n
et
w
o
r
k
co
m
p
ar
ed
to
o
th
er
co
n
v
e
n
tio
n
a
l
n
et
w
o
r
k
s
w
h
ile
s
i
m
u
lta
n
eo
u
s
l
y
ad
d
r
ess
in
g
t
h
e
v
an
i
s
h
in
g
g
r
ad
ien
t p
r
o
b
le
m
[
2
9
]
.
Sp
ec
if
icall
y
,
t
h
e
Fas
ter
R
-
C
NN
X1
0
1
-
FP
N
m
o
d
el
u
s
e
s
R
esNeX
t
-
101
-
3
2
x
8
d
as
its
b
ac
k
b
o
n
e.
R
esNeX
t
-
101
-
3
2
x
8
d
is
a
m
o
d
el
tr
ain
ed
u
s
i
n
g
th
e
C
a
f
f
e2
s
o
f
t
w
ar
e
b
y
r
esear
ch
er
s
at
Face
b
o
o
k
[
1
8
]
.
T
h
e
R
esNeX
t
-
1
0
1
b
lo
ck
d
iv
id
es
a
co
n
v
o
lu
tio
n
in
to
m
u
lt
ip
le
p
ar
a
llel
co
n
v
o
lu
tio
n
s
.
T
h
e
d
if
f
er
en
ce
b
etw
ee
n
th
e
R
esNeX
t b
lo
ck
a
n
d
th
e
R
es
Ne
t b
lo
ck
is
t
h
e
i
n
tr
o
d
u
ctio
n
o
f
a
n
e
w
d
i
m
en
s
io
n
ca
lled
ca
r
d
in
a
lit
y
(
t
h
e
s
ize
o
f
th
e
tr
an
s
f
o
r
m
atio
n
s
et)
in
t
h
e
R
esNeX
t
b
lo
ck
.
T
h
e
n
u
m
b
e
r
s
1
0
1
,
3
2
,
a
n
d
8
i
n
th
e
m
o
d
e
l
’
s
n
am
e
r
ef
e
r
t
o
th
e
n
u
m
b
e
r
o
f
l
ay
e
r
s
(
1
0
1
l
ay
e
r
s
)
,
c
a
r
d
i
n
al
i
ty
(
3
2
)
,
a
n
d
th
e
n
u
m
b
e
r
o
f
b
o
t
t
l
en
e
ck
w
i
d
th
(
8
)
,
r
e
s
p
e
c
t
iv
ely
[
3
0
]
.
A
c
o
m
p
a
r
i
s
o
n
b
e
tw
ee
n
th
e
R
e
s
N
e
t
b
l
o
ck
an
d
th
e
R
e
s
N
e
Xt
b
l
o
c
k
is
s
h
o
w
n
in
F
ig
u
r
e
2
.
Fig
u
r
e
2
.
L
ef
t:
a
b
lo
ck
o
f
R
e
s
Net
an
d
r
ig
h
t:
a
b
lo
ck
o
f
R
e
s
N
eXt
w
ith
ca
r
d
in
a
lit
y
=3
2
[
3
0
]
−
Nec
k
:
t
h
is
co
m
p
o
n
e
n
t
is
lo
c
ated
b
et
w
ee
n
t
h
e
b
ac
k
b
o
n
e
a
n
d
th
e
h
ea
d
.
T
h
e
F
aster
R
-
C
NN
X1
0
1
-
FP
N
m
o
d
el
u
s
es
f
ea
tu
r
e
p
y
r
a
m
id
n
et
w
o
r
k
(
FP
N)
as
th
eir
n
ec
k
.
T
h
e
ar
ch
itectu
r
e
o
f
FP
N
th
at
illu
s
tr
ated
i
n
Fig
u
r
e
3
co
n
s
is
ts
o
f
t
h
r
ee
m
ain
co
m
p
o
n
e
n
ts
:
a
b
o
tto
m
-
u
p
p
ath
w
a
y
,
a
to
p
-
d
o
w
n
p
at
h
wa
y
,
an
d
later
al
co
n
n
ec
tio
n
s
,
w
h
ic
h
co
n
n
ec
t
v
ar
io
u
s
lev
e
ls
o
f
f
ea
tu
r
e
s
(
lo
w
-
le
v
el
a
n
d
h
ig
h
-
lev
e
l
f
ea
tu
r
es)
s
o
t
h
at
ea
c
h
lev
el
i
n
t
h
e
p
y
r
a
m
id
ca
n
b
e
u
s
ed
to
d
etec
t
o
b
j
ec
ts
at
d
if
f
er
e
n
t
s
ca
les.
T
h
is
f
ea
tu
r
e
p
y
r
a
m
i
d
h
elp
s
i
m
p
r
o
v
e
d
etec
tio
n
ac
cu
r
ac
y
b
y
al
lo
w
in
g
th
e
m
o
d
el
to
id
en
ti
f
y
b
o
th
s
m
all
a
n
d
lar
g
e
o
b
j
ec
ts
[
3
1
]
.
−
Hea
d
:
th
e
h
ea
d
co
m
p
o
n
e
n
t
o
f
th
e
Fas
ter
R
-
C
N
N
X1
0
1
-
FP
N
m
o
d
el
is
a
co
m
b
in
a
tio
n
o
f
Fas
t
R
-
C
NN
an
d
R
P
N.
T
h
e
R
P
N
u
s
es th
e
f
ea
t
u
r
e
p
y
r
a
m
id
g
e
n
er
ated
b
y
FP
N
to
f
in
d
r
eg
io
n
s
t
h
at
m
a
y
co
n
ta
i
n
o
b
j
e
cts.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
To
ma
to
lea
f d
is
ea
s
e
r
ec
o
g
n
iti
o
n
s
ystem
u
s
in
g
F
a
s
ter R
-
C
N
N
(
K
a
r
el
Octa
via
n
u
s
B
a
ch
r
i
)
1417
Fig
u
r
e
3
.
FP
N
a
r
ch
itectu
r
e
[
3
1
]
2
.
2
.
Da
t
a
c
o
llect
io
n a
nd
a
nn
o
t
a
t
io
n
A
d
ataset
i
s
a
co
llectio
n
o
f
in
f
o
r
m
atio
n
u
s
ed
to
test
a
n
d
an
al
y
ze
a
s
y
s
te
m
.
I
n
th
e
co
n
tex
t
o
f
o
b
j
ec
t
d
etec
tio
n
,
th
e
r
eq
u
ir
ed
d
ataset
co
n
s
i
s
ts
o
f
a
n
n
o
tated
i
m
a
g
e
s
(
w
ith
b
o
u
n
d
i
n
g
b
o
x
es
a
n
d
l
ab
els).
T
h
e
d
ataset
co
m
p
r
is
e
s
h
ea
lt
h
y
to
m
ato
leav
es
an
d
eig
h
t
lea
f
d
is
ea
s
es:
e
ar
l
y
b
lig
h
t,
late
b
lig
h
t,
leaf
m
o
ld
,
m
o
s
aic
v
ir
u
s
,
s
ep
to
r
ia,
s
p
id
er
m
ites
,
y
ello
w
leaf
c
u
r
l
v
ir
u
s
,
an
d
lea
f
m
i
n
er
.
T
h
e
d
ataset
is
o
b
tain
ed
f
r
o
m
v
ar
io
u
s
s
o
u
r
ce
s
s
u
c
h
as
Kag
g
le,
Go
o
g
le
I
m
a
g
es,
B
in
g
I
m
a
g
es,
an
d
R
o
b
o
f
lo
w
Un
iv
er
s
e.
Data
a
n
n
o
tati
o
n
s
w
e
r
e
d
o
n
e
in
R
o
b
o
f
lo
w
.
2
.
3
.
Da
t
a
pre
-
pro
ce
s
s
ing
a
nd
a
ug
m
e
nta
t
io
n
O
n
e
o
f
th
e
d
at
a
s
et
s
n
am
e
d
Pla
n
t
V
il
l
ag
e
[
1
0
]
,
[
1
1
]
,
w
h
ic
h
c
a
n
b
e
d
o
w
n
l
o
a
d
e
d
th
r
o
u
g
h
K
a
g
g
le
,
is
i
n
ef
f
e
ct
iv
e
w
h
en
u
s
e
d
f
o
r
o
b
je
c
t
d
e
t
e
c
t
i
o
n
c
as
es
b
e
c
a
u
s
e
e
a
c
h
p
h
o
t
o
c
o
n
t
ai
n
s
o
n
ly
o
n
e
o
b
je
c
t
.
T
h
e
r
e
f
o
r
e
,
p
r
e
-
p
r
o
c
e
s
s
in
g
is
p
e
r
f
o
r
m
e
d
u
s
i
n
g
a
p
h
o
t
o
c
o
l
l
ag
e
m
et
h
o
d
.
P
h
o
t
o
c
o
ll
ag
e
is
o
n
e
m
e
th
o
d
o
f
c
o
m
b
in
in
g
m
u
lt
i
p
l
e
p
h
o
t
o
s
i
n
t
o
o
n
e
im
ag
e
.
T
h
u
s
,
th
e
p
h
o
t
o
s
i
n
th
e
d
a
t
as
e
t
a
r
e
m
e
r
g
e
d
f
r
o
m
f
o
u
r
s
e
p
a
r
a
t
e
p
h
o
t
o
s
i
n
t
o
a
s
in
g
le
p
h
o
t
o
.
I
n
ad
d
itio
n
,
p
r
e
-
p
r
o
ce
s
s
i
n
g
i
s
p
er
f
o
r
m
ed
u
s
i
n
g
t
h
e
tile
a
n
d
s
tatic
cr
o
p
m
eth
o
d
s
o
n
s
o
m
e
i
m
ag
e
s
w
it
h
ex
ce
s
s
iv
el
y
h
i
g
h
r
eso
lu
tio
n
s
.
Sin
ce
th
e
co
llected
i
m
ag
e
s
h
av
e
v
ar
y
in
g
r
eso
lu
tio
n
s
,
i
t
is
n
ec
ess
ar
y
to
s
tan
d
ar
d
ize
th
e
r
eso
l
u
tio
n
o
f
ea
ch
i
m
ag
e
i
n
t
h
e
d
ataset.
B
y
s
tan
d
ar
d
izin
g
t
h
e
r
eso
lu
tio
n
,
th
e
tr
ai
n
in
g
p
r
o
ce
s
s
w
il
l b
ec
o
m
e
f
a
s
ter
[
3
2
]
.
I
n
th
is
r
esear
ch
,
th
e
i
m
a
g
e
r
eso
lu
t
io
n
u
s
ed
f
o
r
tr
ain
i
n
g
i
s
s
et
to
5
1
2
×5
1
2
.
I
n
co
m
p
u
ter
v
i
s
io
n
,
d
ata
a
u
g
m
en
tatio
n
is
co
m
m
o
n
l
y
u
s
ed
to
ad
d
r
ess
li
m
i
ted
d
ata
is
s
u
e
s
.
I
ts
g
o
al
is
to
g
en
er
ate
a
lar
g
e,
h
i
g
h
-
q
u
a
lit
y
,
an
d
d
iv
er
s
e
tr
ain
i
n
g
d
ataset
t
h
at
w
ill e
n
ab
le
t
h
e
d
ee
p
lear
n
i
n
g
m
o
d
el
to
b
ec
o
m
e
m
o
r
e
r
o
b
u
s
t a
n
d
g
en
er
alize
well.
I
n
ad
d
itio
n
,
d
ata
au
g
m
en
ta
tio
n
is
o
n
e
o
f
t
h
e
cr
u
cia
l sta
g
e
s
in
d
ata
s
et
cr
ea
tio
n
as it h
as b
ee
n
p
r
o
v
en
to
y
ield
b
etter
ac
cu
r
ac
y
[
3
3
]
.
T
h
e
d
ata
au
g
m
e
n
tatio
n
w
as
p
er
f
o
r
m
ed
o
n
4
,
1
3
2
i
m
a
g
es,
r
e
s
u
lt
in
g
i
n
a
to
tal
o
f
1
0
,
0
4
7
i
m
ag
e
s
.
T
h
e
au
g
m
e
n
tatio
n
co
n
f
i
g
u
r
atio
n
in
clu
d
ed
f
lip
p
in
g
t
h
e
i
m
a
g
es,
ap
p
ly
in
g
a
9
0
°
r
o
tatio
n
,
ad
j
u
s
tin
g
th
e
ex
p
o
s
u
r
e,
an
d
m
o
d
i
f
y
i
n
g
t
h
e
h
u
e.
Af
ter
all
th
e
p
r
o
ce
s
s
in
g
is
co
m
p
leted
,
th
e
d
ataset
is
r
ea
d
y
to
b
e
u
s
ed
f
o
r
tr
ain
in
g
th
e
to
m
ato
leaf
d
i
s
ea
s
e
r
ec
o
g
n
itio
n
s
y
s
te
m
.
G
en
e
r
a
l
ly
,
d
a
ta
s
et
s
c
a
n
b
e
e
x
p
o
r
t
e
d
in
v
a
r
i
o
u
s
w
ay
s
an
d
f
o
r
m
at
s
.
On
e
o
f
th
e
f
o
r
m
at
s
c
o
m
m
o
n
ly
u
s
ed
i
s
c
o
m
m
o
n
o
b
je
c
t
i
n
c
o
n
t
ex
t
jav
as
c
r
i
p
t
o
b
je
c
t
n
o
t
at
i
o
n
(
C
OC
O
J
S
O
N
)
.
T
h
e
C
OC
O
J
SO
N
f
o
r
m
a
t
w
a
s
ch
o
s
e
n
b
e
c
au
s
e
th
e
D
e
t
e
ct
r
o
n
2
a
r
ch
i
t
e
c
tu
r
e
s
u
p
p
o
r
t
s
d
a
t
as
et
s
in
C
O
C
O
J
S
O
N
f
o
r
m
a
t
.
2
.
4
.
M
o
del
t
ra
ini
ng
T
h
e
o
b
j
ec
t
d
etec
tio
n
s
y
s
te
m
i
n
th
i
s
r
esear
ch
is
tr
ai
n
ed
u
s
i
n
g
t
h
e
Dete
ctr
o
n
2
f
r
a
m
e
w
o
r
k
,
w
h
ich
i
s
i
m
p
le
m
en
ted
i
n
P
y
th
o
n
an
d
e
x
ec
u
ted
o
n
Go
o
g
le
C
o
lab
.
I
n
th
e
f
ir
s
t
s
tep
,
th
e
d
o
w
n
lo
ad
o
f
Dete
ctr
o
n
2
an
d
o
th
er
n
ec
es
s
ar
y
l
ib
r
ar
y
p
ac
k
a
g
es
i
s
p
er
f
o
r
m
ed
f
o
r
tr
ain
i
n
g
th
e
o
b
j
ec
t
d
etec
tio
n
s
y
s
te
m
.
T
h
e
es
s
en
t
ial
lib
r
ar
y
p
ac
k
ag
es
th
at
n
ee
d
to
b
e
d
o
w
n
lo
ad
ed
an
d
ar
e
ass
o
ciate
d
with
t
h
e
m
ai
n
f
u
n
ctio
n
a
lit
y
in
Dete
ctr
o
n
2
in
cl
u
d
e
P
y
T
o
r
ch
,
Nu
m
P
y
,
P
y
co
co
to
o
ls
,
p
r
o
to
co
l
b
u
f
f
er
s
(
P
r
o
to
b
u
f
)
,
P
y
y
a
m
l,
an
d
cv
2
(
Op
en
C
V)
.
T
h
e
h
y
p
e
r
p
a
r
am
e
te
r
s
a
r
e
c
o
n
f
ig
u
r
e
d
d
u
r
in
g
th
e
t
r
a
in
in
g
p
r
o
c
es
s
.
Hy
p
e
r
p
a
r
am
e
t
e
r
s
a
r
e
p
a
r
am
e
t
e
r
s
u
s
e
d
t
o
c
o
n
f
ig
u
r
e
a
m
a
ch
in
e
-
l
e
a
r
n
in
g
m
o
d
el
a
n
d
a
r
e
d
e
te
r
m
in
e
d
b
e
f
o
r
e
s
t
a
r
t
in
g
th
e
t
r
ai
n
in
g
p
r
o
ce
s
s
[
3
4
]
.
T
h
e
r
e
a
r
e
m
an
y
h
y
p
e
r
p
a
r
am
et
e
r
s
th
a
t
ca
n
b
e
a
d
ju
s
t
e
d
w
h
en
t
r
a
in
i
n
g
a
F
a
s
t
e
r
R
-
C
NN
m
o
d
el
.
T
h
e
h
y
p
e
r
p
a
r
am
et
e
r
s
a
d
ju
s
t
e
d
in
t
h
i
s
t
r
a
in
in
g
a
r
e
:
a.
I
m
ag
e
s
p
er
b
atch
=4
,
w
h
ic
h
r
ep
r
esen
ts
t
h
e
n
u
m
b
er
o
f
i
m
ag
es
p
r
o
ce
s
s
ed
in
ea
ch
b
atc
h
d
u
r
in
g
t
h
e
m
o
d
e
l
tr
ain
i
n
g
[
3
5
]
.
b.
B
ase
lear
n
in
g
r
ate=
0
.
0
0
1
,
r
e
p
r
esen
t
s
th
e
b
ase
lear
n
in
g
r
ate
u
s
ed
d
u
r
in
g
t
h
e
tr
ai
n
i
n
g
p
r
o
ce
s
s
.
c.
W
ar
m
-
u
p
iter
atio
n
s
=1
0
0
0
,
th
is
p
ar
am
e
ter
d
eter
m
in
e
s
th
e
n
u
m
b
er
o
f
tr
ai
n
i
n
g
i
ter
atio
n
s
u
s
e
d
f
o
r
th
e
w
ar
m
-
u
p
s
ta
g
e
d
u
r
i
n
g
m
o
d
el
tr
ai
n
i
n
g
.
T
h
is
p
ar
a
m
eter
i
s
u
s
ed
to
lo
w
er
t
h
e
lear
n
in
g
r
ate
a
n
d
r
ed
u
ce
th
e
i
m
p
ac
t
o
f
in
s
tab
il
it
y
th
at
m
a
y
o
cc
u
r
d
u
r
i
n
g
th
e
in
i
tial tr
ai
n
i
n
g
p
h
ase
[
3
5
]
.
I
n
th
is
r
esear
ch
,
t
h
e
p
ar
a
m
eter
u
s
ed
f
o
r
th
e
w
ar
m
-
u
p
s
tag
e
is
s
et
to
1
0
0
0
,
w
h
ic
h
m
ea
n
s
t
h
at
f
r
o
m
th
e
s
tar
t
o
f
tr
ai
n
in
g
u
n
til
t
h
e
1
0
0
0
th
iter
atio
n
,
t
h
e
lear
n
in
g
r
ate
r
ed
u
ctio
n
i
s
g
r
ad
u
all
y
ap
p
lied
.
d.
Ma
x
i
m
u
m
i
ter
atio
n
r
ep
r
esen
t
s
th
e
m
a
x
i
m
u
m
n
u
m
b
er
o
f
it
er
atio
n
s
p
er
f
o
r
m
ed
d
u
r
i
n
g
m
o
d
el
tr
ain
in
g
.
I
n
th
is
r
esear
c
h
,
v
ar
io
u
s
iter
atio
n
s
w
er
e
u
s
ed
as s
h
o
w
n
i
n
F
ig
u
r
e
4
.
e.
R
eg
io
n
o
f
in
ter
est
(
R
o
I
)
b
atch
s
ize
p
er
im
ag
e=
1
2
8
,
r
ep
r
esen
tin
g
t
h
e
n
u
m
b
er
o
f
R
o
I
p
r
o
p
o
s
als
u
s
ed
in
ea
c
h
i
m
a
g
e
d
u
r
in
g
t
h
e
tr
ain
i
n
g
o
f
t
h
e
R
OI
h
ea
d
in
t
h
e
m
o
d
el.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
2
,
No
.
6
,
Dec
em
b
er
20
2
4
:
1
4
1
4
-
1
421
1418
f.
Nu
m
b
er
o
f
clas
s
es=1
0
,
w
h
er
e
9
class
es+1
clas
s
i
s
au
to
m
atic
all
y
ad
d
ed
b
y
R
o
b
o
f
lo
w
w
h
ic
h
r
ep
r
esen
t
s
th
e
class
f
o
r
all
i
m
a
g
es
[
3
5
]
.
W
h
ile
all
o
th
er
ar
e
lef
t a
t t
h
eir
d
ef
au
lt
v
al
u
es a
cc
o
r
d
in
g
to
D
etec
tr
o
n
2
’
s
s
p
ec
i
f
icatio
n
s
[
3
6
]
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
p
er
f
o
r
m
a
n
ce
test
i
n
g
in
v
o
lv
ed
ev
al
u
ati
n
g
s
e
v
er
al
m
etr
ics
co
m
m
o
n
l
y
u
s
ed
in
th
e
C
OC
O
ev
alu
a
tio
n
,
s
u
ch
as
a
v
er
ag
e
p
r
ec
is
io
n
(
A
P
)
d
an
av
er
ag
e
r
ec
all
(
A
R
)
.
T
h
e
ca
lcu
latio
n
o
f
t
h
e
A
P
an
d
A
R
m
etr
ic
is
clo
s
el
y
r
e
lated
to
th
e
i
n
ter
s
ec
tio
n
o
v
er
u
n
io
n
(
I
o
U)
m
etr
i
c.
I
o
U
is
an
ev
al
u
atio
n
m
e
th
o
d
u
s
ed
to
m
ea
s
u
r
e
th
e
ac
cu
r
ac
y
o
f
o
b
j
ec
t d
etec
tio
n
in
a
d
ataset.
A
P
in
th
e
C
O
C
O
ev
a
lu
atio
n
m
etr
ic
is
th
e
A
P
ac
r
o
s
s
all
ca
teg
o
r
ies
o
r
class
es
f
r
o
m
th
e
I
o
U
th
r
esh
o
ld
o
f
0
.
5
0
to
0
.
9
5
w
it
h
a
n
i
n
ter
v
al
o
f
0
.
0
5
[
3
7
]
.
T
h
er
ef
o
r
e,
A
P
r
ep
r
esen
ts
th
e
AP
ca
lc
u
lated
at
1
0
d
if
f
er
e
n
t
I
o
U
th
r
es
h
o
ld
s
(
0
.
5
0
,
0
.
5
5
,
0
.
6
0
,
.
.
.
,
0
.
9
5
)
.
I
n
ad
d
itio
n
,
it
is
p
o
s
s
ib
le
to
ev
alu
ate
it
u
s
in
g
i
n
d
iv
id
u
al
I
OU
v
al
u
e
s
,
w
it
h
t
h
e
m
o
s
t
u
s
ed
v
al
u
es
b
ei
n
g
5
0
%
an
d
7
5
%.
T
h
ese
ar
e
r
ep
o
r
ted
as
A
P
5
0
an
d
A
P
7
5
,
r
esp
ec
tiv
el
y
.
A
P
is
also
ca
lcu
lated
ac
r
o
s
s
s
ca
les
at
d
if
f
er
en
t
I
o
Us
(
0
.
5
0
:0
.
9
5
)
.
T
h
ese
ca
teg
o
r
ies
ar
e
d
iv
id
ed
in
to
th
r
ee
d
if
f
er
e
n
t
s
izes:
A
P
s
f
o
r
s
m
a
ll
o
b
j
ec
ts
(
ar
ea
<3
2
2
)
,
A
P
m
f
o
r
m
ed
iu
m
o
b
j
ec
ts
(
3
2
2
<a
r
ea
<9
6
2
)
,
an
d
A
P
l
f
o
r
lar
g
e
o
b
j
ec
ts
(
ar
ea
>9
6
2
).
W
h
ile
th
e
AR
is
ca
lc
u
lated
at
d
if
f
er
e
n
t
m
a
x
i
m
u
m
d
etec
t
io
n
s
p
er
im
a
g
e.
T
h
ese
ca
teg
o
r
ies
a
r
e
d
iv
id
ed
to
b
e
A
R
m
ax
=1
,
AR
m
a
x
=1
0
,
an
d
AR
m
ax
=1
0
0
.
L
i
k
e
th
e
A
P
m
etr
ic,
AR
is
also
ca
lc
u
l
ated
ac
r
o
s
s
s
ca
les
at
d
if
f
er
e
n
t
I
o
Us
(
0
.
5
0
:0
.
9
5
)
.
T
h
ese
ca
teg
o
r
ies
ar
e
d
iv
id
ed
in
to
th
r
ee
d
if
f
er
en
t
s
ize
s
:
A
R
s
m
a
l
l,
A
R
m
ed
i
u
m
,
an
d
AR
lar
g
e.
So
m
e
t
y
p
e
s
o
f
A
P
an
d
AR
m
etr
ics i
n
th
e
C
O
C
O
e
v
alu
a
tio
n
m
etr
ic
ar
e
s
h
o
w
n
i
n
T
ab
le
2
.
Fig
u
r
e
4
s
h
o
w
s
t
h
e
A
P
a
g
ai
n
s
t
iter
atio
n
s
u
s
i
n
g
Fa
s
ter
R
-
C
NN
X1
0
1
-
FP
N.
I
t
ca
n
b
e
o
b
s
er
v
ed
th
at
d
u
r
in
g
th
e
e
v
al
u
atio
n
o
f
t
h
e
v
alid
atio
n
d
ata,
t
h
e
m
o
d
el
w
i
th
t
h
e
b
est
A
P
is
t
h
e
m
o
d
el
o
b
tain
ed
d
u
r
in
g
th
e
3
7
0
0
0
th
iter
atio
n
w
it
h
an
AP
o
f
8
7
.
0
1
.
Fo
r
a
m
o
r
e
co
m
p
r
eh
en
s
i
v
e
v
ie
w
,
all
C
O
C
O
ev
alu
a
tio
n
m
etr
ic
s
,
in
cl
u
d
in
g
A
P
an
d
av
er
ag
e
r
ec
all,
f
o
r
th
e
m
o
d
el
at
iter
atio
n
3
7
,
0
0
0
is
s
h
o
w
n
i
n
T
ab
le
3
.
T
h
e
m
o
d
el
p
er
f
o
r
m
ed
w
ell
i
n
t
h
e
n
o
n
-
d
ata
s
et
test
i
m
ag
e
s
.
T
h
e
m
o
d
el
ca
n
d
etec
t
th
e
h
ea
l
th
y
lea
v
es
as
s
h
o
w
n
i
n
Fi
g
u
r
e
5
(
a)
,
as
all
th
e
h
ea
lt
h
y
lea
v
es
ar
e
r
ec
o
g
n
i
ze
d
h
ea
lth
y
b
y
th
e
m
o
d
el.
Fi
g
u
r
e
5
(
b
)
s
h
o
w
s
th
at
th
e
m
o
d
el
ca
n
r
ec
o
g
n
ize
leaf
m
o
ld
d
is
ea
s
ed
,
as
all
d
is
ea
s
ed
m
o
ld
s
ar
e
r
ec
o
g
n
ized
lea
f
m
o
ld
w
it
h
v
ar
io
u
s
d
eg
r
ee
o
f
ce
r
tain
ties
,
w
h
ile
o
n
e
h
ea
lt
h
y
lea
f
ar
e
r
ec
o
g
n
ized
h
ea
lth
y
,
e
v
e
n
th
o
u
g
h
w
it
h
o
u
t
1
0
0
%
d
eg
r
ee
o
f
ce
r
tain
t
y
.
Fi
g
u
r
e
5
(
c)
s
h
o
w
s
t
h
at
th
e
m
o
d
el
ca
n
r
ec
o
g
n
ized
1
leaf
w
ith
lea
f
m
in
er
d
is
ea
s
e
a
m
o
n
g
h
ea
lt
h
y
lea
v
e
s
.
Fig
u
r
e
5
(
d
)
s
h
o
w
s
th
at
t
h
e
m
o
d
el
ca
n
r
ec
o
g
n
ize
m
u
ltip
le
leaf
d
is
ea
s
e
s
.
T
ab
le
2
.
C
OC
O
ev
al
u
atio
n
m
e
tr
ics
[
3
8
]
AP
A
r
e
a
M
a
x
i
m
u
m
d
e
t
e
c
t
i
o
n
I
o
U
A
P
a
c
r
o
ss
sca
l
es
A
r
e
a
M
a
x
i
m
u
d
e
t
e
c
t
i
o
n
I
o
U
AP
A
ll
1
0
0
0
.
5
0
:
0
.
9
5
A
P
s
S
mal
l
o
b
j
e
c
t
(
a
r
e
a
<
3
2
2
)
1
0
0
0
.
5
0
:
0
.
9
5
A
P
5
0
A
ll
1
0
0
0
.
5
0
A
P
m
M
e
d
i
u
m o
b
j
e
c
t
(
3
2
2
<
a
r
e
a
<
9
6
2
)
1
0
0
0
.
5
0
:
0
.
9
5
A
P
7
5
A
ll
1
0
0
0
.
7
5
A
P
l
L
a
r
g
e
o
b
j
e
c
t
(
a
r
e
a
>
9
6
2
)
1
0
0
0
.
5
0
:
0
.
9
5
AR
A
r
e
a
M
a
x
i
m
u
m
d
e
t
e
c
t
i
o
n
I
o
U
A
R
a
c
r
o
ss
s
c
a
l
e
s
A
r
e
a
M
a
x
i
m
u
m
d
e
t
e
c
t
i
o
n
I
o
U
A
R
max
=
1
A
ll
1
0
.
5
0
:
0
.
9
5
A
R
s
S
mal
l
o
b
j
e
c
t
(
a
r
e
a
<
3
2
2
)
1
0
0
0
.
5
0
:
0
.
9
5
A
R
max
=
1
0
A
ll
10
0
.
5
0
:
0
.
9
5
A
R
m
M
e
d
i
u
m o
b
j
e
c
t
(
3
2
2
<
a
r
e
a
<
9
6
2
)
1
0
0
0
.
5
0
:
0
.
9
5
A
R
max
=
1
0
0
A
ll
1
0
0
0
.
5
0
:
0
.
9
5
A
R
l
L
a
r
g
e
o
b
j
e
c
t
(
a
r
e
a
>
9
6
2
)
1
0
0
0
.
5
0
:
0
.
9
5
Fig
u
r
e
4.
E
v
alu
atio
n
r
esu
lts
o
n
v
alid
atio
n
u
s
i
n
g
A
P
m
etr
ic
a
g
ain
s
t
iter
atio
n
s
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
To
ma
to
lea
f d
is
ea
s
e
r
ec
o
g
n
iti
o
n
s
ystem
u
s
in
g
F
a
s
ter R
-
C
N
N
(
K
a
r
el
Octa
via
n
u
s
B
a
ch
r
i
)
1419
T
ab
le
3
.
C
OC
O
ev
al
u
atio
n
m
e
tr
ics u
s
in
g
Fa
s
ter
R
-
C
N
N
X1
0
1
-
FP
N
(
iter
atio
n
3
7
,
0
0
0
)
I
o
U
A
r
e
a
M
a
x
d
e
t
s
AP
I
o
U
A
r
e
a
M
a
x
d
e
t
s
AR
0
.
5
0
:
0
.
9
5
A
ll
1
0
0
8
7
.
0
1
0
.
5
0
:
0
.
9
5
A
ll
1
5
9
.
2
0
.
5
0
A
ll
1
0
0
9
5
.
1
0
.
5
0
:
0
.
9
5
A
ll
10
8
9
.
8
0
.
7
5
A
ll
1
0
0
9
2
.
9
5
0
.
5
0
:
0
.
9
5
A
ll
1
0
0
9
0
.
7
0
.
5
0
:
0
.
9
5
S
mal
l
1
0
0
3
1
.
2
5
0
.
5
0
:
0
.
9
5
S
mal
l
1
0
0
58
0
.
5
0
:
0
.
9
5
M
e
d
i
u
m
1
0
0
5
8
.
1
7
0
.
5
0
:
0
.
9
5
M
e
d
i
u
m
1
0
0
6
5
.
9
0
.
5
0
:
0
.
9
5
L
a
r
g
e
1
0
0
8
9
.
8
0
.
5
0
:
0
.
9
5
L
a
r
g
e
1
0
0
9
3
.
1
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
5
.
R
esu
lts
o
n
te
s
t i
m
ag
es: (
a)
h
ea
lth
y
lea
v
es [
3
9
]
,
(
b
)
d
is
ea
s
ed
leav
es [
4
0
]
,
(
c)
h
ea
lth
y
a
n
d
d
is
ea
s
ed
leav
es [
4
1
]
,
an
d
(
d
)
d
is
ea
s
ed
leav
es i
n
K
eb
u
n
I
b
u
&
C
o
4.
CO
NCLU
SI
O
N
T
h
e
to
m
ato
leaf
d
is
ea
s
e
r
ec
o
g
n
itio
n
s
y
s
te
m
,
b
ased
o
n
th
e
Fas
ter
R
-
C
N
N
X1
0
1
-
FPN
m
o
d
el
ar
ch
itect
u
r
e
u
s
i
n
g
th
e
Dete
ctr
o
n
2
f
r
a
m
e
w
o
r
k
,
w
a
s
tr
ain
ed
o
n
a
d
ataset
co
n
s
is
ti
n
g
o
f
1
0
,
0
4
7
i
m
ag
e
s
co
n
tai
n
i
n
g
v
ar
io
u
s
d
i
s
ea
s
e
s
.
T
h
e
m
o
d
el
w
a
s
tr
ain
ed
f
o
r
3
7
,
0
0
0
iter
atio
n
s
,
r
es
u
lti
n
g
i
n
t
h
e
b
est
A
P
o
f
8
7
.
0
1
.
T
h
is
s
y
s
te
m
ca
n
d
etec
t
v
ar
io
u
s
d
is
ea
s
es
i
n
to
m
ato
leav
es,
in
cl
u
d
in
g
e
ar
l
y
b
lig
h
t,
late
b
lig
h
t,
leaf
m
o
ld
,
m
o
s
a
ic
v
ir
u
s
,
s
ep
to
r
ia,
s
p
id
er
m
ites
,
y
ello
w
leaf
c
u
r
l
v
ir
u
s
,
leaf
m
in
er
,
an
d
h
ea
lt
h
y
to
m
ato
leav
e
s
.
Alth
o
u
g
h
t
h
e
m
o
d
el
p
r
o
v
id
es
s
atis
f
ac
to
r
y
r
esu
l
ts
,
t
h
e
to
m
ato
leaf
d
is
ea
s
e
r
ec
o
g
n
itio
n
s
y
s
te
m
ca
n
b
e
f
u
r
th
er
d
ev
elo
p
ed
to
d
etec
t
o
th
er
d
is
ea
s
es
o
n
to
m
ato
leav
es
as
w
ell
as
s
y
m
p
to
m
s
o
f
n
u
tr
ie
n
t
d
ef
icie
n
c
y
.
A
d
d
itio
n
a
ll
y
,
th
e
to
m
ato
lea
f
d
is
ea
s
e
r
ec
o
g
n
i
tio
n
s
y
s
te
m
ca
n
also
b
e
i
m
p
le
m
en
ted
in
to
m
o
b
ile
ap
p
licatio
n
s
.
ACK
NO
WL
E
D
G
M
E
NT
S
T
h
e
au
th
o
r
s
w
o
u
ld
li
k
e
to
ex
p
r
ess
t
h
eir
s
in
ce
r
e
g
r
atit
u
d
e
an
d
ap
p
r
ec
iatio
n
to
th
e
I
n
s
titu
te
f
o
r
R
esear
ch
a
n
d
C
o
m
m
u
n
i
t
y
De
v
elo
p
m
e
n
t,
At
m
a
J
a
y
a
C
at
h
o
l
ic
Un
iv
er
s
it
y
o
f
I
n
d
o
n
esia
f
o
r
p
r
o
v
id
in
g
in
ter
n
al
r
esear
ch
g
r
an
t
0
2
6
9
.
7
/I
I
I
/L
P
PM
-
P
M.
1
0
.
0
1
/
0
3
/2
0
2
3
th
at
s
u
p
p
o
r
ted
th
is
ac
tiv
it
y
.
T
h
is
r
esear
ch
g
r
an
t
p
la
y
ed
a
cr
u
cial
r
o
le
in
th
e
s
u
cc
e
s
s
f
u
l
co
m
p
let
io
n
o
f
t
h
is
r
esear
c
h
.
A
d
d
itio
n
all
y
,
t
h
e
au
th
o
r
s
w
o
u
ld
also
lik
e
to
ac
k
n
o
w
led
g
e
an
d
th
a
n
k
Mr
.
A
r
ie
Har
tan
to
,
K
eb
u
n
I
b
u
&
C
o
f
o
r
th
eir
s
u
p
p
o
r
t
an
d
co
o
p
er
atio
n
,
w
h
ich
m
ad
e
th
is
r
esear
c
h
p
o
s
s
ib
le.
RE
F
E
R
E
NC
E
S
[
1
]
H
.
N
a
z
k
i
,
S
.
Y
o
o
n
,
A
.
F
u
e
n
t
e
s,
a
n
d
D
.
S
.
P
a
r
k
,
“
U
n
s
u
p
e
r
v
i
se
d
i
m
a
g
e
t
r
a
n
sl
a
t
i
o
n
u
si
n
g
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s
f
o
r
i
mp
r
o
v
e
d
p
l
a
n
t
d
i
se
a
se
r
e
c
o
g
n
i
t
i
o
n
,
”
C
o
m
p
u
t
e
rs
a
n
d
El
e
c
t
r
o
n
i
c
s
i
n
A
g
ri
c
u
l
t
u
re
,
v
o
l
.
1
6
8
,
p
.
1
0
5
1
1
7
,
J
a
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
a
g
.
2
0
1
9
.
1
0
5
1
1
7
.
[
2
]
M
.
A
.
A
z
i
m,
M
.
K
.
I
s
l
a
m,
M
.
M
.
R
a
h
ma
n
,
a
n
d
F
.
Ja
h
a
n
,
“
A
n
e
f
f
e
c
t
i
v
e
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
me
t
h
o
d
f
o
r
r
i
c
e
l
e
a
f
d
i
se
a
se
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
T
e
l
k
o
m
n
i
k
a
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
El
e
c
t
ro
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
1
9
,
n
o
.
2
,
p
p
.
4
6
3
–
4
7
0
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
2
9
2
8
/
T
EL
K
O
M
N
I
K
A
.
v
1
9
i
2
.
1
6
4
8
8
.
[
3
]
R
.
G
i
r
s
h
i
c
k
,
J
.
D
o
n
a
h
u
e
,
T
.
D
a
r
r
e
l
l
,
a
n
d
J
.
M
a
l
i
k
,
“
R
i
c
h
F
e
a
t
u
r
e
H
i
e
r
a
r
c
h
i
e
s
f
o
r
A
c
c
u
r
a
t
e
O
b
j
e
c
t
D
e
t
e
c
t
i
o
n
a
n
d
S
e
m
a
n
t
i
c
S
e
g
m
e
n
t
a
t
i
o
n
,
”
i
n
P
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
E
E
E
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
(
C
V
P
R
)
,
2
0
1
4
,
p
p
.
5
8
0
–
5
8
7
.
[
4
]
R
.
G
i
r
sh
i
c
k
,
“
F
a
st
R
-
C
N
N
,
”
a
rX
i
v
p
r
e
p
ri
n
t
a
r
Xi
v
:
1
5
0
4
.
0
8
0
8
3
,
2
0
1
5
.
[
5
]
S
.
R
e
n
,
K
.
H
e
,
R
.
G
i
r
sh
i
c
k
,
a
n
d
J.
S
u
n
,
“
F
a
st
e
r
R
-
C
N
N
:
T
o
w
a
r
d
s
r
e
a
l
-
t
i
me
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
w
i
t
h
r
e
g
i
o
n
p
r
o
p
o
sal
n
e
t
w
o
r
k
s,”
Ad
v
a
n
c
e
s i
n
N
e
u
r
a
l
I
n
f
o
rm
a
t
i
o
n
Pro
c
e
ssi
n
g
S
y
s
t
e
m
s
,
p
p
.
9
1
–
9
9
,
2
0
1
5
.
[
6
]
M
.
C
a
r
r
a
n
z
a
-
G
a
r
c
í
a
,
J
.
T
o
r
r
e
s
-
M
a
t
e
o
,
P
.
L
a
r
a
-
B
e
n
í
t
e
z
,
a
n
d
J
.
G
a
r
c
í
a
-
G
u
t
i
é
r
r
e
z
,
“
O
n
t
h
e
p
e
r
f
o
r
man
c
e
o
f
o
n
e
-
st
a
g
e
a
n
d
t
w
o
-
st
a
g
e
o
b
j
e
c
t
d
e
t
e
c
t
o
r
s
i
n
a
u
t
o
n
o
mo
u
s
v
e
h
i
c
l
e
s
u
s
i
n
g
c
a
me
r
a
d
a
t
a
,
”
R
e
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
1
–
2
3
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s1
3
0
1
0
0
8
9
.
[
7
]
J.
H
u
a
n
g
e
t
a
l
.
,
“
S
p
e
e
d
/
a
c
c
u
r
a
c
y
t
r
a
d
e
-
o
f
f
s
f
o
r
mo
d
e
r
n
c
o
n
v
o
l
u
t
i
o
n
a
l
o
b
j
e
c
t
d
e
t
e
c
t
o
r
s,”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
3
0
t
h
I
EEE
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
,
C
VPR
2
0
1
7
,
I
EEE,
Ju
l
.
2
0
1
7
,
p
p
.
3
2
9
6
–
3
3
0
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
7
.
3
5
1
.
[
8
]
L
.
A
.
V
a
r
g
a
,
S
.
K
o
c
h
,
a
n
d
A
.
Z
e
l
l
,
“
C
o
mp
r
e
h
e
n
si
v
e
A
n
a
l
y
si
s
o
f
t
h
e
O
b
j
e
c
t
D
e
t
e
c
t
i
o
n
P
i
p
e
l
i
n
e
o
n
U
A
V
s,”
Re
m
o
t
e
S
e
n
s
i
n
g
,
v
o
l
.
1
4
,
n
o
.
2
1
,
p
.
5
5
0
8
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s
1
4
2
1
5
5
0
8
.
[
9
]
S
.
S
r
i
v
a
s
t
a
v
a
,
A
.
V
.
D
i
v
e
k
a
r
,
C
.
A
n
i
l
k
u
m
a
r
,
I
.
N
a
i
k
,
V
.
K
u
l
k
a
r
n
i
,
a
n
d
V
.
P
a
t
t
a
b
i
r
a
m
a
n
,
“
C
o
mp
a
r
a
t
i
v
e
a
n
a
l
y
si
s
o
f
d
e
e
p
l
e
a
r
n
i
n
g
i
mag
e
d
e
t
e
c
t
i
o
n
a
l
g
o
r
i
t
h
ms
,
”
J
o
u
r
n
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
8
,
n
o
.
1
,
p
.
6
6
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
0
5
3
7
-
0
2
1
-
0
0
4
3
4
-
w.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
2
,
No
.
6
,
Dec
em
b
er
20
2
4
:
1
4
1
4
-
1
421
1420
[
1
0
]
K
a
g
g
l
e
,
“
P
l
a
n
t
V
i
l
l
a
g
e
D
a
t
a
se
t
,
”
k
a
g
g
l
e
.
c
o
m
.
h
t
t
p
s:
/
/
w
w
w
.
k
a
g
g
l
e
.
c
o
m/
d
a
t
a
s
e
t
s/
e
mm
a
r
e
x
/
p
l
a
n
t
d
i
se
a
se
(
a
c
c
e
sse
d
J
u
l
.
0
6
,
2
0
2
3
)
.
[
1
1
]
D
.
P
.
H
u
g
h
e
s
a
n
d
M
.
S
a
l
a
t
h
e
,
“
A
n
o
p
e
n
a
c
c
e
ss
r
e
p
o
si
t
o
r
y
o
f
i
mag
e
s
o
n
p
l
a
n
t
h
e
a
l
t
h
t
o
e
n
a
b
l
e
t
h
e
d
e
v
e
l
o
p
me
n
t
o
f
mo
b
i
l
e
d
i
se
a
se
d
i
a
g
n
o
st
i
c
s,
”
2
0
1
5
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
a
r
x
i
v
.
o
r
g
/
a
b
s/
1
5
1
1
.
0
8
0
6
0
[
1
2
]
R
.
C
h
a
e
r
a
n
i
a
n
d
R
.
E.
V
o
o
r
r
i
p
s,
“
T
o
mat
o
e
a
r
l
y
b
l
i
g
h
t
(
A
l
t
e
r
n
a
r
i
a
so
l
a
n
i
)
:
T
h
e
p
a
t
h
o
g
e
n
,
g
e
n
e
t
i
c
s,
a
n
d
b
r
e
e
d
i
n
g
f
o
r
r
e
si
st
a
n
c
e
,”
J
o
u
rn
a
l
o
f
G
e
n
e
ra
l
Pl
a
n
t
P
a
t
h
o
l
o
g
y
,
v
o
l
.
7
2
,
n
o
.
6
,
p
p
.
3
3
5
–
3
4
7
,
D
e
c
.
2
0
0
6
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
0
3
2
7
-
0
0
6
-
0
2
9
9
-
3.
[
1
3
]
S
.
Z
.
M
.
Z
a
k
i
,
M
.
A
.
Z
u
l
k
i
f
l
e
y
,
M
.
M
.
S
t
o
f
a
,
N
.
A
.
M
.
K
a
mar
i
,
a
n
d
N
.
A
.
M
o
h
a
me
d
,
“
C
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
t
o
m
a
t
o
l
e
a
f
d
i
se
a
se
s
u
s
i
n
g
mo
b
i
l
e
n
e
t
v
2
,
”
I
AE
S
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
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
.
9
,
n
o
.
2
,
p
p
.
2
9
0
–
2
9
6
,
J
u
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
9
.
i
2
.
p
p
2
9
0
-
2
9
6
.
[
1
4
]
Z
.
Y
a
n
,
A
.
M
.
A
.
W
o
l
t
e
r
s,
J.
N
a
v
a
s‐
C
a
st
i
l
l
o
,
a
n
d
Y
.
B
a
i
,
“
T
h
e
g
l
o
b
a
l
d
i
me
n
si
o
n
o
f
t
o
mat
o
y
e
l
l
o
w
l
e
a
f
c
u
r
l
d
i
se
a
se
:
C
u
r
r
e
n
t
s
t
a
t
u
s
a
n
d
b
r
e
e
d
i
n
g
p
e
r
sp
e
c
t
i
v
e
s,”
M
i
c
r
o
o
r
g
a
n
i
sm
s
,
v
o
l
.
9
,
n
o
.
4
,
d
o
i
:
1
0
.
3
3
9
0
/
m
i
c
r
o
o
r
g
a
n
i
sms
9
0
4
0
7
4
0
.
[
1
5
]
C
.
K
.
T
e
t
t
e
y
e
t
a
l
.
,
“
T
o
mat
o
mo
t
t
l
e
mo
sai
c
v
i
r
u
s:
C
h
a
r
a
c
t
e
r
i
z
a
t
i
o
n
,
r
e
si
st
a
n
c
e
g
e
n
e
e
f
f
e
c
t
i
v
e
n
e
ss,
a
n
d
q
u
i
n
t
u
p
l
e
x
R
T
-
P
C
R
d
e
t
e
c
t
i
o
n
sy
st
e
m,”
J
o
u
rn
a
l
o
f
I
n
t
e
g
r
a
t
i
v
e
A
g
ri
c
u
l
t
u
re
,
v
o
l
.
2
1
,
n
o
.
9
,
p
p
.
2
6
4
1
–
2
6
5
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
i
a
.
2
0
2
2
.
0
7
.
0
2
0
.
[
1
6
]
A
.
T
.
N
i
e
u
w
e
n
h
u
i
z
e
n
,
J.
K
o
o
l
,
H
.
K
.
S
u
h
,
a
n
d
J
.
H
e
mm
i
n
g
,
“
A
u
t
o
mat
e
d
sp
i
d
e
r
mi
t
e
d
a
mag
e
d
e
t
e
c
t
i
o
n
o
n
t
o
mat
o
l
e
a
v
e
s
i
n
g
r
e
e
n
h
o
u
se
s,”
Ac
t
a
H
o
r
t
i
c
u
l
t
u
r
a
e
,
n
o
.
1
2
6
8
,
p
p
.
1
6
5
–
1
7
2
,
Ja
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
7
6
6
0
/
A
c
t
a
H
o
r
t
i
c
.
2
0
2
0
.
1
2
6
8
.
2
1
.
[
1
7
]
S
.
M
u
st
a
p
h
a
,
A
.
K
.
M
u
sa
,
a
n
d
A
.
S
.
A
l
i
y
u
,
“
A
ss
e
ssm
e
n
t
o
f
l
e
a
f
mi
n
e
r
L
i
r
i
o
m
y
z
a
sp
p
(
d
i
p
t
e
r
a
:
A
g
r
o
m
y
z
i
d
a
e
)
d
a
mag
e
o
n
t
o
ma
t
o
a
n
d
t
h
e
y
i
e
l
d
o
u
t
p
u
t
i
n
I
l
o
r
i
n
,
S
o
u
t
h
e
r
n
G
u
i
n
e
a
S
a
v
a
n
n
a
h
,
N
i
g
e
r
i
a
,
”
A
g
ro
s
e
a
rc
h
,
v
o
l
.
2
0
,
n
o
.
2
,
p
p
.
1
–
9
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
4
3
1
4
/
a
g
r
o
sh
.
v
2
0
i
2
.
1
.
[
1
8
]
Y
.
W
u
,
A
.
K
i
r
i
l
l
o
v
,
F
.
M
a
ss
a
,
W
.
-
Y
.
L
o
,
a
n
d
R
.
G
i
r
sh
i
c
k
,
“
D
e
t
e
c
t
r
o
n
2
mo
d
e
l
z
o
o
,
”
g
i
t
h
u
b
.
c
o
m
,
2
0
1
9
.
h
t
t
p
s:
/
/
g
i
t
h
u
b
.
c
o
m/
f
a
c
e
b
o
o
k
r
e
se
a
r
c
h
/
d
e
t
e
c
t
r
o
n
2
(
a
c
c
e
sse
d
J
u
l
.
0
6
,
2
0
2
3
)
.
[
1
9
]
Y
.
Z
h
a
n
g
,
C
.
S
o
n
g
,
a
n
d
D
.
Z
h
a
n
g
,
“
D
e
e
p
L
e
a
r
n
i
n
g
-
B
a
se
d
O
b
j
e
c
t
D
e
t
e
c
t
i
o
n
I
mp
r
o
v
e
m
e
n
t
f
o
r
T
o
mat
o
D
i
se
a
se
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
5
6
6
0
7
–
5
6
6
1
4
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
2
9
8
2
4
5
6
.
[
2
0
]
A
.
F
u
e
n
t
e
s,
S
.
Y
o
o
n
,
S
.
C
.
K
i
m,
a
n
d
D
.
S
.
P
a
r
k
,
“
A
r
o
b
u
st
d
e
e
p
-
l
e
a
r
n
i
n
g
-
b
a
se
d
d
e
t
e
c
t
o
r
f
o
r
r
e
a
l
-
t
i
me
t
o
mat
o
p
l
a
n
t
d
i
se
a
se
s
a
n
d
p
e
st
s re
c
o
g
n
i
t
i
o
n
,
”
S
e
n
s
o
rs (
S
w
i
t
z
e
rl
a
n
d
)
,
v
o
l
.
1
7
,
n
o
.
9
,
p
.
2
0
2
2
,
S
e
p
.
2
0
1
7
,
d
o
i
:
1
0
.
3
3
9
0
/
s
1
7
0
9
2
0
2
2
.
[
2
1
]
S
.
A
d
h
i
k
a
r
i
,
E.
K
C
,
L
.
B
a
l
k
u
mari
,
B
.
S
h
r
e
st
h
a
,
a
n
d
B
.
B
a
i
j
u
,
“
T
o
mat
o
P
l
a
n
t
D
i
se
a
se
s De
t
e
c
t
i
o
n
S
y
st
e
m U
si
n
g
I
mag
e
P
r
o
c
e
ssi
n
g
,
”
Re
se
a
rc
h
g
a
t
e
.
N
e
t
,
v
o
l
.
I
,
p
p
.
8
1
–
8
6
,
2
0
1
8
.
[
2
2
]
Q
.
W
a
n
g
a
n
d
F
.
Q
i
,
“
T
o
mat
o
d
i
se
a
s
e
s
r
e
c
o
g
n
i
t
i
o
n
b
a
se
d
o
n
f
a
st
e
r
R
C
N
N
,
”
i
n
Pro
c
e
e
d
i
n
g
s
-
1
0
t
h
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
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
i
n
Me
d
i
c
i
n
e
a
n
d
Ed
u
c
a
t
i
o
n
,
I
T
ME
2
0
1
9
,
I
EEE,
A
u
g
.
2
0
1
9
,
p
p
.
7
7
2
–
7
7
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
T
M
E.
2
0
1
9
.
0
0
1
7
6
.
[
2
3
]
I
.
K
u
mar,
A
.
B
i
s
h
t
,
a
n
d
J.
R
a
w
a
t
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
t
o
ma
t
o
p
l
a
n
t
d
i
se
a
se
u
si
n
g
f
a
s
t
e
r
R
-
C
N
N
a
n
d
R
e
t
i
n
a
N
e
t
,
”
C
o
n
v
e
r
g
e
n
c
e
o
f
C
l
o
u
d
C
o
m
p
u
t
i
n
g
,
AI
,
a
n
d
A
g
ri
c
u
l
t
u
r
a
l
S
c
i
e
n
c
e
,
p
p
.
3
0
6
–
3
2
7
,
2
0
2
3
,
d
o
i
:
1
0
.
4
0
1
8
/
9
7
9
-
8
-
3
6
9
3
-
0
2
0
0
-
2
.
c
h
0
1
6
.
[
2
4
]
M
.
A
l
r
u
w
a
i
l
i
,
M
.
H
.
S
i
d
d
i
q
i
,
A
.
K
h
a
n
,
M
.
A
z
a
d
,
A
.
K
h
a
n
,
a
n
d
S
.
A
l
a
n
a
z
i
,
“
R
TF
-
R
C
N
N
:
A
n
A
r
c
h
i
t
e
c
t
u
r
e
f
o
r
R
e
a
l
-
T
i
me
T
o
mat
o
P
l
a
n
t
L
e
a
f
D
i
se
a
se
s
D
e
t
e
c
t
i
o
n
i
n
V
i
d
e
o
S
t
r
e
a
mi
n
g
U
si
n
g
F
a
st
e
r
-
R
C
N
N
,
”
Bi
o
e
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
9
,
n
o
.
1
0
,
p
.
5
6
5
,
O
c
t
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
b
i
o
e
n
g
i
n
e
e
r
i
n
g
9
1
0
0
5
6
5
.
[
2
5
]
R
.
B
u
d
i
a
r
sa
,
R
.
W
a
r
d
o
y
o
,
a
n
d
A
.
M
u
s
d
h
o
l
i
f
a
h
,
“
F
a
c
e
r
e
c
o
g
n
i
t
i
o
n
f
o
r
o
c
c
l
u
d
e
d
f
a
c
e
w
i
t
h
m
a
sk
r
e
g
i
o
n
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
a
n
d
f
u
l
l
y
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
:
a
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
El
e
c
t
ri
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
3
,
n
o
.
5
,
p
p
.
5
6
6
2
–
5
6
7
3
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
3
i
5
.
p
p
5
6
6
2
-
5
6
7
3
.
[
2
6
]
S
.
B
o
u
r
a
y
a
a
n
d
A
.
B
e
l
a
n
g
o
u
r
,
“
D
e
e
p
L
e
a
r
n
i
n
g
b
a
se
d
N
e
c
k
M
o
d
e
l
s
f
o
r
O
b
j
e
c
t
D
e
t
e
c
t
i
o
n
:
A
R
e
v
i
e
w
a
n
d
a
B
e
n
c
h
mark
i
n
g
S
t
u
d
y
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ad
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
2
,
n
o
.
1
1
,
p
p
.
1
6
1
–
1
6
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
J
A
C
S
A
.
2
0
2
1
.
0
1
2
1
1
1
9
.
[
2
7
]
A
.
R
a
h
i
m,
A
.
M
a
q
b
o
o
l
,
a
n
d
T
.
R
a
n
a
,
“
M
o
n
i
t
o
r
i
n
g
so
c
i
a
l
d
i
st
a
n
c
i
n
g
u
n
d
e
r
v
a
r
i
o
u
s
l
o
w
l
i
g
h
t
c
o
n
d
i
t
i
o
n
s
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
a
n
d
a
si
n
g
l
e
mo
t
i
o
n
l
e
ss t
i
me
o
f
f
l
i
g
h
t
c
a
me
r
a
,
”
PLo
S
O
N
E
,
v
o
l
.
1
6
,
p
.
e
0
2
4
7
4
4
0
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
o
n
e
.
0
2
4
7
4
4
0
.
[
2
8]
C
.
R
.
d
e
L
i
ma,
S
.
G
.
K
h
a
n
,
S
.
H
.
S
h
a
h
,
a
n
d
L
.
F
e
r
r
i
,
“
M
a
s
k
r
e
g
i
o
n
-
b
a
se
d
C
N
N
s
f
o
r
c
e
r
v
i
c
a
l
c
a
n
c
e
r
p
r
o
g
r
e
ssi
o
n
d
i
a
g
n
o
si
s
o
n
p
a
p
sme
a
r
e
x
a
mi
n
a
t
i
o
n
s,”
H
e
l
i
y
o
n
,
v
o
l
.
9
,
n
o
.
1
1
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
2
3
.
e
2
1
3
8
8
.
[
2
9
]
I
.
V
a
l
o
v
a
,
C
.
H
a
r
r
i
s,
N
.
G
u
e
o
r
g
u
i
e
v
a
,
a
n
d
T
.
M
a
i
,
“
I
n
-
B
e
t
w
e
e
n
L
a
y
e
r
s M
o
d
u
l
a
r
R
e
si
d
u
a
l
N
e
u
r
a
l
N
e
t
w
o
r
k
f
o
r
t
h
e
C
l
a
ssi
f
i
c
a
t
i
o
n
o
f
I
mag
e
s,”
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
.
1
8
5
,
p
p
.
2
2
3
–
2
3
0
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
1
.
0
5
.
0
2
3
.
[
3
0
]
S
.
X
i
e
,
R
.
G
i
r
sh
i
c
k
,
P
.
D
o
l
l
á
r
,
Z
.
T
u
,
a
n
d
K
.
H
e
,
“
A
g
g
r
e
g
a
t
e
d
r
e
si
d
u
a
l
t
r
a
n
sf
o
r
mat
i
o
n
s
f
o
r
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s,”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
3
0
t
h
I
EEE
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
si
o
n
a
n
d
P
a
t
t
e
rn
Re
c
o
g
n
i
t
i
o
n
,
C
VP
R
2
0
1
7
,
I
EEE,
Ju
l
.
2
0
1
7
,
p
p
.
5
9
8
7
–
5
9
9
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
7
.
6
3
4
.
[
3
1
]
T
.
Y
.
L
i
n
,
P
.
D
o
l
l
á
r
,
R
.
G
i
r
sh
i
c
k
,
K
.
H
e
,
B
.
H
a
r
i
h
a
r
a
n
,
a
n
d
S
.
B
e
l
o
n
g
i
e
,
“
F
e
a
t
u
r
e
p
y
r
a
mi
d
n
e
t
w
o
r
k
s
f
o
r
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
,
”
i
n
Pro
c
e
e
d
i
n
g
s
-
3
0
t
h
I
E
EE
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
si
o
n
a
n
d
P
a
t
t
e
rn
R
e
c
o
g
n
i
t
i
o
n
,
C
VP
R
2
0
1
7
,
I
EEE,
J
u
l
.
2
0
1
7
,
p
p
.
9
3
6
–
944
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
7
.
1
0
6
.
[
3
2
]
S
.
S
a
p
o
n
a
r
a
a
n
d
A
.
El
h
a
n
a
s
h
i
,
“
I
mp
a
c
t
o
f
I
mag
e
R
e
si
z
i
n
g
o
n
D
e
e
p
L
e
a
r
n
i
n
g
D
e
t
e
c
t
o
r
s
f
o
r
T
r
a
i
n
i
n
g
T
i
me
a
n
d
M
o
d
e
l
P
e
r
f
o
r
man
c
e
,
”
i
n
L
e
c
t
u
re
N
o
t
e
s
i
n
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
,
2
0
2
2
,
p
p
.
1
0
–
17
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
030
-
9
5
4
9
8
-
7
_
2
.
[
3
3
]
C
.
N
.
V
a
s
c
o
n
c
e
l
o
s
a
n
d
B
.
N
.
V
a
s
c
o
n
c
e
l
o
s,
“
Ex
p
e
r
i
me
n
t
s
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
d
e
r
mo
sco
p
y
i
mag
e
a
n
a
l
y
si
s,”
P
a
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
L
e
t
t
e
rs
,
v
o
l
.
1
3
9
,
p
p
.
9
5
–
1
0
3
,
N
o
v
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
r
e
c
.
2
0
1
7
.
1
1
.
0
0
5
.
[
3
4
]
V
.
C
.
S
.
R
.
R
a
y
a
v
a
r
a
p
u
a
n
d
A
.
M
a
h
a
p
a
t
r
o
,
“
N
L
O
S
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
d
mi
t
i
g
a
t
i
o
n
i
n
U
W
B
p
o
si
t
i
o
n
i
n
g
w
i
t
h
b
a
g
g
i
n
g
-
b
a
se
d
e
n
se
mb
l
e
d
c
l
a
ssi
f
i
e
r
s,”
A
n
n
a
l
e
s
d
e
s
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
/
A
n
n
a
l
s
o
f
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
7
7
,
n
o
.
5
–
6
,
p
p
.
2
6
7
–
2
8
0
,
J
u
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
2
2
4
3
-
0
2
1
-
0
0
8
8
4
-
6.
[
3
5
]
A
.
S
u
l
j
o
v
i
c
,
S
.
C
a
k
i
c
,
T
.
P
o
p
o
v
i
c
,
a
n
d
S
.
S
a
n
d
i
,
“
D
e
t
e
c
t
i
o
n
o
f
P
l
a
n
t
D
i
s
e
a
s
e
s
U
s
i
n
g
L
e
a
f
I
m
a
g
e
s
a
n
d
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
,
”
2
0
2
2
2
1
s
t
I
n
t
e
r
n
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
I
N
F
O
T
E
H
-
J
A
H
O
R
I
N
A
,
I
N
F
O
T
E
H
2
0
2
2
-
P
r
o
c
e
e
d
i
n
g
s
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
F
O
T
E
H
5
3
7
3
7
.
2
0
2
2
.
9
7
5
1
2
4
5
.
[
3
6
]
A
P
I
D
o
c
u
m
e
n
t
a
t
i
o
n
,
“
d
e
t
e
c
t
r
o
n
2
.
c
o
n
f
i
g
,
”
d
e
t
e
c
t
r
o
n
2
.
r
e
a
d
t
h
e
d
o
c
s
.
i
o
.
h
t
t
p
s
:
/
/
d
e
t
e
c
t
r
o
n
2
.
r
e
a
d
t
h
e
d
o
c
s
.
i
o
/
e
n
/
l
a
t
e
s
t
/
m
o
d
u
l
e
s
/
c
o
n
f
i
g
.
h
t
m
l
(
a
c
c
e
s
s
e
d
J
u
l
.
0
5
,
2
0
2
3
)
.
[
3
7
]
A
.
W
e
n
d
a
,
I
.
P
e
r
man
a
,
Y
u
smar
,
a
n
d
N
.
N
.
K
u
r
n
i
a
w
a
t
i
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
p
a
d
d
y
l
e
a
f
d
i
se
a
se
s
b
a
se
d
o
n
t
e
x
t
u
r
e
a
n
a
l
y
si
s
o
f
b
l
o
b
s
a
n
d
c
o
l
o
r
se
g
me
n
t
a
t
i
o
n
,
”
T
e
l
k
o
m
n
i
k
a
(
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
C
o
m
p
u
t
i
n
g
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
n
t
r
o
l
)
,
v
o
l
.
1
8
,
n
o
.
4
,
p
p
.
2
0
1
8
–
2
0
2
6
,
A
u
g
.
2
0
2
0
,
d
o
i
:
1
0
.
1
2
9
2
8
/
T
EL
K
O
M
N
I
K
A
.
V
1
8
I
4
.
1
4
6
1
4
.
[
3
8
]
C
O
C
O
,
“
C
O
C
O
Ev
a
l
u
a
t
i
o
n
,
”
c
o
c
o
d
a
t
a
se
t
.
o
rg
.
h
t
t
p
s:
/
/
c
o
c
o
d
a
t
a
se
t
.
o
r
g
/
#
d
e
t
e
c
t
i
o
n
-
e
v
a
l
(
a
c
c
e
s
se
d
Ju
l
.
0
6
,
2
0
2
3
)
.
[
3
9
]
S
.
N
.
U
.
o
f
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
“
j
so
n
2
t
x
t
D
a
t
a
se
t
,
”
u
n
i
v
e
r
se.
r
o
b
o
f
l
o
w
.
c
o
m
.
h
t
t
p
s:
/
/
u
n
i
v
e
r
se
.
r
o
b
o
f
l
o
w
.
c
o
m/
se
o
u
l
-
n
a
t
i
o
n
a
l
-
u
n
i
v
e
r
si
t
y
-
of
-
sci
e
n
c
e
-
a
n
d
-
t
e
c
h
n
o
l
o
g
y
-
d
g
c
y
c
/
j
so
n
2
t
x
t
-
w
j
5
8
i
(
a
c
c
e
sse
d
Ju
l
.
0
6
,
2
0
2
3
)
.
[
4
0
]
R
i
s
i
n
g
s
u
n
,
“
t
o
ma
t
o
_
d
i
se
a
se
D
a
t
a
se
t
,
”
u
n
i
v
e
rse.
r
o
b
o
f
l
o
w
.
c
o
m
.
h
t
t
p
s:
/
/
u
n
i
v
e
r
se
.
r
o
b
o
f
l
o
w
.
c
o
m/
r
i
si
n
g
su
n
-
b
9
r
l
6
/
t
o
ma
t
o
_
d
i
se
a
se
-
b
c
3
y
e
(
a
c
c
e
sse
d
S
e
p
t
.
03
,
2
0
2
3
).
[
4
1
]
V
.
N
.
U
n
i
v
e
r
si
t
y
,
“
d
a
t
a
se
t
_
s
p
l
i
t
D
a
t
a
s
e
t
,
”
u
n
i
v
e
rse.
r
o
b
o
f
l
o
w
.
c
o
m
.
(
a
c
c
e
sse
d
Ju
l
.
0
6
,
2
0
2
3
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
To
ma
to
lea
f d
is
ea
s
e
r
ec
o
g
n
iti
o
n
s
ystem
u
s
in
g
F
a
s
ter R
-
C
N
N
(
K
a
r
el
Octa
via
n
u
s
B
a
ch
r
i
)
1421
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
K
a
r
e
l
O
c
ta
v
ia
n
u
s
B
a
c
h
r
i
is
a
m
e
m
b
e
r
o
f
A
t
m
a
Ja
y
a
Ca
th
o
li
c
Un
iv
e
rsity
o
f
In
d
o
n
e
sia
.
He
o
b
tai
n
e
d
h
is
D
o
c
to
ra
te
in
In
telli
g
e
n
t
I
n
stru
m
e
n
tatio
n
in
2
0
1
9
.
His
re
se
a
rc
h
in
tere
sts
a
re
in
telli
g
e
n
t
in
str
u
m
e
n
tatio
n
a
n
d
sy
ste
m
o
n
c
h
i
p
(S
o
C).
He
w
a
s
a
lso
in
v
o
lv
e
d
i
n
se
v
e
r
a
l
a
rti
f
icia
l
in
telli
g
e
n
c
e
p
ro
j
e
c
ts
to
d
e
sig
n
c
o
g
n
it
iv
e
in
stru
m
e
n
tatio
n
sy
ste
m
s.
Cu
rre
n
tl
y
,
h
e
is
w
o
rk
in
g
o
n
in
telli
g
e
n
t
m
o
n
i
to
rin
g
a
n
d
c
o
n
tr
o
l
sy
ste
m
s
f
o
r
a
g
ricu
lt
u
ra
l
s
u
sta
in
a
b
il
it
y
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
a
r
e
l.
b
a
c
h
ri@atm
a
ja
y
a
.
a
c
.
id
.
B
r
y
a
n
S
a
n
to
s
o
is
a
st
u
d
e
n
t
a
t
th
e
E
lec
tri
c
a
l
En
g
in
e
e
rin
g
,
A
tma
Ja
y
a
Ca
th
o
li
c
Un
iv
e
rsit
y
o
f
In
d
o
n
e
sia
.
He
h
a
s
a
stro
n
g
in
tere
st
in
m
a
c
h
in
e
lea
rn
in
g
,
p
r
o
g
ra
m
m
a
b
le
lo
g
ic
c
o
n
tro
ll
e
r
(
P
L
C),
a
n
d
i
n
tern
e
t
o
f
th
in
g
s
(
Io
T
).
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
b
ry
a
n
.
2
0
1
9
0
4
5
2
0
0
1
5
@s
t
u
d
e
n
t.
a
t
m
a
ja
y
a
.
a
c
.
id
.
Dum
a
K
r
isti
n
a
Ya
n
ti
H
u
t
a
p
e
a
is
c
u
rre
n
tl
y
a
f
a
c
u
lt
y
m
e
m
b
e
r
a
t
A
tm
a
Ja
y
a
Ca
th
o
li
c
U
n
iv
e
rsity
o
f
In
d
o
n
e
si
a
in
Ja
k
a
rta.
He
r
e
x
p
e
rti
se
li
e
s
in
n
e
u
ro
sig
n
a
l
p
ro
c
e
ss
in
g
,
b
io
m
e
d
ica
l
e
n
g
in
e
e
rin
g
,
a
n
d
tele
c
o
m
m
u
n
ica
ti
o
n
.
S
h
e
c
o
m
p
lete
d
h
e
r
b
a
c
h
e
lo
r
’
s
d
e
g
re
e
in
e
lec
tri
c
a
l
e
n
g
in
e
e
rin
g
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
In
d
o
n
e
sia
i
n
2
0
0
1
.
S
h
e
t
h
e
n
p
u
rs
u
e
d
f
u
rth
e
r
stu
d
ies
a
t
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
P
ET
RON
A
S
in
M
a
la
y
sia
,
e
a
rn
in
g
a
M
a
ste
r
o
f
S
c
ien
c
e
(M
.
S
c
.
)
d
e
g
re
e
in
2
0
1
0
a
n
d
a
Do
c
to
ra
te
in
2
0
1
8
.
Du
rin
g
h
e
r
d
o
c
to
ra
l
st
u
d
ies
in
M
a
lay
sia
,
sh
e
w
a
s
a
ss
ig
n
e
d
a
s
a
re
s
e
a
rc
h
sc
ien
ti
st
a
t
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
P
ET
R
ON
A
S
f
o
r
a
p
e
rio
d
o
f
3
y
e
a
rs,
f
ro
m
2
0
1
5
t
o
2
0
1
8
.
A
f
ter
c
o
m
p
le
ti
n
g
h
e
r
d
o
c
t
o
ra
l
st
u
d
ies
a
n
d
h
e
r
ro
le
a
s
a
re
se
a
rc
h
sc
ien
ti
st
,
sh
e
jo
in
e
d
t
h
e
f
a
c
u
lt
y
o
f
th
e
M
a
st
e
r
’
s P
ro
g
ra
m
in
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
a
t
A
tm
a
Ja
y
a
Ca
th
o
li
c
Un
iv
e
rsit
y
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
u
m
a
.
k
risti
n
a
@a
t
m
a
ja
y
a
.
a
c
.
id
.
Ca
th
e
r
in
e
O
li
v
ia
S
e
r
e
a
ti
is
a
lec
tu
re
r
a
n
d
re
se
a
rc
h
e
r
a
t
th
e
A
tma
Ja
y
a
Ca
th
o
li
c
Un
iv
e
rsit
y
o
f
In
d
o
n
e
sia
.
Ca
t
h
e
rin
e
’
s
i
n
tere
st
su
b
jec
ts
o
f
re
se
a
rc
h
a
re
El
e
c
tro
n
ic
In
stru
m
e
n
tatio
n
S
y
ste
m
s a
n
d
S
y
ste
m
o
n
Ch
i
p
(S
o
C)
.
S
h
e
w
a
s
a
lso
in
v
o
lv
e
d
in
se
v
e
ra
l
re
se
a
r
c
h
p
ro
jec
ts
to
d
e
sig
n
c
o
g
n
it
iv
e
in
stru
m
e
n
tatio
n
sy
ste
m
s.
S
o
m
e
o
f
th
e
m
a
re
b
u
il
d
i
n
g
a
so
f
t
w
a
re
c
o
g
n
it
iv
e
in
terp
re
tatio
n
o
f
sh
ip
m
o
v
e
m
e
n
ts,
f
o
r
In
d
o
n
e
sia
n
m
a
r
in
e
se
c
u
rit
y
p
u
rp
o
se
s,
a
n
d
c
o
g
n
it
iv
e
e
lec
tro
c
a
rd
io
g
ra
p
h
(E
CG
)
d
e
si
g
n
.
Cu
rre
n
tl
y
,
h
e
r
re
se
a
rc
h
p
ro
jec
t
is
f
o
c
u
sin
g
o
n
d
e
sig
n
in
g
th
e
a
rc
h
it
e
c
tu
re
o
f
Co
g
n
it
iv
e
p
ro
c
e
ss
o
rs.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
c
a
th
e
rin
e
.
o
li
v
ia@
a
t
m
a
ja
y
a
.
a
c
.
id
.
La
n
n
y
W
.
Pa
n
d
ja
it
a
n
is
a
lec
tu
re
r
a
t
th
e
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
A
t
m
a
Ja
y
a
Ca
th
o
li
c
U
n
iv
e
rsity
o
f
In
d
o
n
e
sia
.
S
h
e
c
o
m
p
lete
d
h
e
r
B
a
c
h
e
lo
r
’
s
d
e
g
re
e
a
t
IT
B
F
a
c
u
lt
y
o
f
In
d
u
strial
T
e
c
h
n
o
lo
g
y
,
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
in
1
9
7
3
,
IT
B
M
a
ste
r
’
s
P
ro
g
ra
m
,
S
T
EI,
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
in
1
9
9
4
,
a
n
d
D
o
c
to
ra
l
P
ro
g
ra
m
a
t
IT
B,
S
T
EI,
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
in
2
0
0
0
.
S
h
e
is
o
n
e
o
f
th
e
P
r
o
f
e
ss
o
rs
o
f
En
g
in
e
e
rin
g
El
e
c
tri
c
a
l
a
t
A
t
m
a
Ja
y
a
C
a
th
o
li
c
Un
iv
e
rsit
y
o
f
In
d
o
n
e
sia
.
He
r
a
re
a
o
f
e
x
p
e
rti
se
is
in
te
ll
ig
e
n
t
sy
st
e
m
s
b
a
se
d
e
lec
tro
n
ic
in
stru
m
e
n
tatio
n.
S
h
e
c
a
n
b
e
c
o
n
ta
c
ted
a
t
e
m
a
il
:
m
m
.
lan
n
y
@a
t
m
a
ja
y
a
.
a
c
.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.