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.
1
6
,
No
.
1
,
Feb
r
u
ar
y
20
2
6
,
p
p
.
404
~
413
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
1
6
i
1
.
pp
4
0
4
-
4
1
3
404
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Tiny
ma
chine lea
rning
wit
h
conv
o
lutiona
l neural n
et
wo
rk f
o
r
intelligent
radia
ti
o
n mo
nitoring
in
nuclea
r installa
ti
o
n
I
s
t
o
f
a
,
G
ina
K
us
um
a
,
F
irliy
a
ni
Ra
hm
a
t
ia
Ning
s
ih
,
J
o
k
o
T
riy
a
nto
,
I
P
utu
Su
s
ila
,
At
a
ng
Su
s
ila
R
e
se
a
r
c
h
C
e
n
t
e
r
f
o
r
N
u
c
l
e
a
r
B
e
a
m A
n
a
l
y
s
i
s T
e
c
h
n
o
l
o
g
y
,
R
e
s
e
a
r
c
h
O
r
g
a
n
i
z
a
t
i
o
n
f
o
r
N
u
c
l
e
a
r
En
e
r
g
y
,
N
a
t
i
o
n
a
l
R
e
se
a
r
c
h
a
n
d
I
n
n
o
v
a
t
i
o
n
A
g
e
n
c
y
,
Ta
n
g
e
r
a
n
g
S
e
l
a
t
a
n
,
I
n
d
o
n
e
s
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
5
,
2
0
2
5
R
ev
is
ed
Sep
2
5
,
2
0
2
5
Acc
ep
ted
No
v
2
3
,
2
0
2
5
Th
is
stu
d
y
fo
c
u
se
s
o
n
d
e
v
e
l
o
p
i
n
g
a
n
i
n
telli
g
e
n
t
ra
d
iatio
n
m
o
n
it
o
r
i
n
g
s
y
ste
m
c
a
p
a
b
le
o
f
o
p
e
ra
ti
n
g
o
n
a
lo
w
-
p
o
we
r
sin
g
le
-
b
o
a
rd
c
o
m
p
u
ter
(Ra
sp
b
e
rry
P
i)
fo
r
d
e
p
lo
y
m
e
n
t
i
n
re
m
o
te
m
o
n
it
o
ri
n
g
sta
ti
o
n
s
wit
h
in
n
u
c
lea
r
fa
c
il
it
y
e
n
v
iro
n
m
e
n
ts.
T
h
e
p
r
o
p
o
se
d
sy
ste
m
u
ti
li
z
e
s
a
ra
d
io
n
u
c
li
d
e
id
e
n
ti
fica
ti
o
n
m
e
th
o
d
b
a
se
d
o
n
ti
n
y
m
a
c
h
in
e
lea
rn
in
g
(
Ti
n
y
M
L)
wit
h
a
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
(CNN
)
a
rc
h
i
tec
tu
re
.
Th
e
ra
d
io
n
u
c
li
d
e
d
a
tas
e
t
wa
s
a
c
q
u
ired
th
ro
u
g
h
m
e
a
su
re
m
e
n
ts
o
f
sta
n
d
a
rd
ra
d
iati
o
n
so
u
rc
e
s,
with
v
a
r
iatio
n
s
in
d
istan
c
e
,
e
x
p
o
su
re
ti
m
e
,
a
n
d
c
o
m
b
in
a
ti
o
n
s
o
f
m
u
lt
ip
le
so
u
rc
e
s
-
in
c
lu
d
in
g
Cs
-
1
3
7
,
Co
-
6
0
,
Cs
-
1
3
4
,
a
n
d
E
u
-
1
5
2
.
T
h
e
ra
d
iatio
n
in
ten
sity
d
a
ta
fro
m
d
e
tec
to
r
m
e
a
su
re
m
e
n
ts
we
re
stru
c
tu
re
d
in
t
o
a
re
sp
o
n
se
m
a
tri
x
a
n
d
su
b
se
q
u
e
n
tl
y
c
o
n
v
e
rted
i
n
to
a
g
r
a
y
sc
a
le
ima
g
e
d
a
tas
e
t
fo
r
m
o
d
e
l
train
in
g
.
Ke
ra
s
is
u
se
d
to
d
e
sig
n
a
n
d
trai
n
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
ls,
wh
i
le
Ten
so
r
F
lo
w
L
it
e
is
u
se
d
t
o
m
o
d
e
l
siz
e
r
e
d
u
c
ti
o
n
.
E
x
p
e
r
ime
n
tal
re
s
u
l
ts
d
e
m
o
n
s
tra
te
th
a
t
t
h
e
d
e
v
e
l
o
p
e
d
m
o
d
e
l
a
c
h
ie
v
e
s
a
n
a
c
c
u
ra
c
y
o
f
9
9
.
3
3
8
%
f
o
r
Ke
ra
s
m
o
d
e
l
tra
i
n
e
d
o
n
c
o
m
p
u
te
r
a
n
d
8
4
.
5
6
8
%
a
fte
r
d
e
p
l
o
y
m
e
n
t
o
n
t
h
e
Ra
s
p
b
e
r
r
y
P
i
.
F
u
r
t
h
e
rm
o
re
,
t
h
is
st
u
d
y
s
u
c
c
e
s
sf
u
l
l
y
d
e
s
i
g
n
e
d
a
n
d
e
m
b
e
d
d
e
d
t
h
e
T
i
n
y
M
L
m
o
d
e
l
i
n
t
o
a
n
e
n
v
ir
o
n
m
e
n
t
ra
d
ia
t
io
n
m
o
n
i
t
o
r
in
g
s
y
st
e
m
a
t
t
h
e
P
U
S
P
I
P
T
EK
n
u
c
lea
r
i
n
sta
ll
a
t
i
o
n
.
K
ey
w
o
r
d
s
:
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
Mo
n
ito
r
in
g
s
tatio
n
R
ad
io
n
u
clid
e
R
asp
b
er
r
y
Pi
T
in
y
m
ac
h
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
:
I
s
to
f
a
R
esear
ch
C
en
ter
f
o
r
Nu
clea
r
B
ea
m
An
aly
s
is
T
ec
h
n
o
lo
g
y
,
R
esear
ch
Or
g
an
izatio
n
f
o
r
N
u
clea
r
E
n
er
g
y
,
Natio
n
al
R
esear
ch
an
d
I
n
n
o
v
a
tio
n
Ag
en
cy
T
an
g
er
an
g
Selatan
,
B
an
ten
,
I
n
d
o
n
esia
E
m
ail:
is
to
0
0
1
@
b
r
in
.
g
o
.
id
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
u
tili
za
tio
n
an
d
d
ev
el
o
p
m
en
t
o
f
ar
tific
ial
in
tellig
en
ce
(
AI
)
,
f
o
r
in
s
tan
ce
m
ac
h
i
n
e
lear
n
in
g
(
ML
)
an
d
d
ee
p
lear
n
in
g
(
DL
)
,
h
av
e
ad
v
an
ce
d
s
ig
n
if
ica
n
tly
in
r
ec
en
t
y
ea
r
s
,
r
ev
o
lu
tio
n
izin
g
d
ata
an
aly
s
is
an
d
co
m
p
u
tatio
n
al
task
s
b
y
en
a
b
lin
g
ap
p
licatio
n
s
to
o
p
er
at
e
in
tellig
en
tly
[
1
]
.
R
ea
l
-
tim
e
m
o
n
ito
r
in
g
o
f
en
v
ir
o
n
m
en
tal
r
ad
iatio
n
is
an
im
p
o
r
ta
n
t
r
eq
u
i
r
em
en
t
in
n
u
clea
r
s
af
ety
an
d
r
is
k
m
iti
g
atio
n
s
y
s
tem
s
[
2
]
.
Ho
wev
er
,
a
f
r
eq
u
en
t
c
h
allen
g
e
is
th
e
ef
f
icien
cy
o
f
d
ata
t
r
an
s
m
is
s
io
n
f
r
o
m
th
e
f
ield
to
th
e
co
n
tr
o
l
ce
n
ter
,
esp
ec
ially
wh
en
th
e
s
y
s
tem
m
o
n
ito
r
s
th
e
r
ad
iatio
n
s
p
e
ctr
u
m
co
n
tin
u
o
u
s
ly
[
3
]
.
T
in
y
m
ac
h
in
e
lear
n
in
g
(
T
in
y
ML
)
p
r
o
v
id
es
an
i
n
n
o
v
ativ
e
ap
p
r
o
ac
h
b
y
d
ep
lo
y
i
n
g
ML
m
o
d
els
o
n
lo
w
-
p
o
wer
,
r
eso
u
r
ce
-
co
n
s
tr
ain
e
d
ed
g
e
d
ev
ices,
en
ab
lin
g
r
ea
l
-
ti
m
e
o
n
-
d
e
v
ice
in
f
er
en
ce
[
4
]
,
[
5
]
.
T
h
is
s
y
s
tem
s
u
p
p
o
r
ts
r
ea
l
-
tim
e
an
aly
tics
,
wh
ich
en
h
an
ce
s
d
ec
is
io
n
-
m
a
k
in
g
s
p
ee
d
an
d
im
p
r
o
v
es
th
e
o
v
e
r
all
r
esp
o
n
s
iv
en
ess
[
6
]
.
Su
c
h
ca
p
ab
ilit
ies
ar
e
p
ar
ticu
lar
ly
cr
itical
f
o
r
tim
e
-
s
en
s
itiv
e
ap
p
licatio
n
s
,
in
clu
d
in
g
au
to
n
o
m
o
u
s
v
eh
icles
[
7
]
,
h
e
alth
ca
r
e
m
o
n
ito
r
in
g
[
8
]
,
[
9
]
,
an
d
ea
r
ly
war
n
in
g
s
y
s
tem
s
[
1
0
]
,
wh
er
e
p
r
o
ce
s
s
in
g
d
elay
s
co
u
ld
lead
t
o
s
ev
er
e
co
n
s
eq
u
e
n
ce
s
.
T
h
er
ef
o
r
e,
in
th
e
co
n
tex
t
o
f
en
v
ir
o
n
m
en
tal
r
ad
iatio
n
m
o
n
ito
r
in
g
,
th
e
im
p
lem
en
tatio
n
o
f
T
i
n
y
ML
ca
n
s
er
v
e
as
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
Tin
y
ma
ch
in
e
lea
r
n
in
g
w
ith
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l n
etw
o
r
k
fo
r
in
tellig
en
t
…
(
I
s
to
fa
)
405
b
o
th
an
ea
r
ly
d
etec
tio
n
s
y
s
te
m
an
d
a
d
ec
is
io
n
-
s
u
p
p
o
r
t
to
o
l,
ef
f
ec
tiv
ely
m
itig
atin
g
th
e
p
o
ten
tial
im
p
ac
ts
o
f
r
ad
iatio
n
in
cid
e
n
ts
[
1
1
]
.
Nu
m
er
o
u
s
p
r
io
r
s
tu
d
ies
h
av
e
u
tili
ze
d
s
im
u
lated
d
ata
g
en
er
a
ted
th
r
o
u
g
h
Mo
n
te
C
ar
lo
[
1
2
]
–
[
1
4
]
a
n
d
Gea
n
t4
ap
p
licatio
n
s
[
1
5
]
–
[
1
7
]
.
As
d
em
o
n
s
tr
ated
in
th
e
s
t
u
d
y
by
Altay
e
b
et
a
l.
th
e
m
ajo
r
ity
o
f
ex
is
tin
g
r
esear
ch
lev
er
ag
in
g
g
am
m
a
-
r
a
y
s
p
ec
tr
u
m
f
o
r
r
ad
io
n
u
clid
e
i
d
en
tific
atio
n
an
d
u
tili
zin
g
s
ilico
n
p
h
o
to
m
u
ltip
lier
(
SiP
M)
[
5
]
f
o
r
th
e
s
en
s
o
r
o
r
s
cin
till
atio
n
d
etec
to
r
[
2
]
.
Ad
d
itio
n
ally
,
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
(
ML
)
tech
n
iq
u
es
[
1
8
]
,
[
1
9
]
,
in
cl
u
d
i
n
g
ar
tific
ial
n
e
u
r
al
n
etwo
r
k
(
ANN)
[
2
0
]
–
[
2
2
]
a
n
d
th
e
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN)
[
2
3
]
–
[
2
5
]
,
h
a
v
e
b
ee
n
ex
p
lo
r
ed
f
o
r
d
e
v
elo
p
i
n
g
au
t
o
m
ated
m
o
d
els.
Ho
we
v
er
,
lim
ited
r
esear
c
h
h
as f
o
cu
s
ed
o
n
im
p
lem
e
n
tin
g
t
h
e
tr
ain
ed
m
o
d
els with
in
T
in
y
ML
s
y
s
tem
s
f
o
r
r
ad
io
n
u
clid
e
c
lass
if
icatio
n
.
T
h
e
u
s
e
o
f
T
in
y
ML
aim
s
to
o
v
er
co
m
e
p
r
o
b
lem
s
r
elate
d
to
h
ig
h
d
ata
r
ate
tr
a
n
s
m
is
s
io
n
an
d
lim
ite
d
r
eso
u
r
ce
s
at
m
o
n
ito
r
i
n
g
s
tatio
n
s
.
W
ith
lo
ca
l
d
ata
p
r
o
ce
s
s
in
g
ca
p
a
b
ilit
ies,
T
in
y
ML
ca
n
r
ed
u
ce
th
e
n
ee
d
f
o
r
d
ata
tr
an
s
m
is
s
io
n
,
an
d
e
f
f
ici
en
tly
,
q
u
ick
ly
,
an
d
p
o
r
tab
ly
id
en
tify
r
ad
i
o
n
u
clid
es
[
2
6
]
.
I
n
th
is
s
tu
d
y
,
an
in
tellig
en
t
s
y
s
tem
-
b
ased
r
ad
ia
tio
n
m
o
n
ito
r
in
g
s
tatio
n
was
d
ev
elo
p
ed
to
id
e
n
tify
r
a
d
io
n
u
clid
es
d
ir
ec
tly
in
t
h
e
f
ield
.
W
ith
th
is
id
en
tific
atio
n
ca
p
ab
ilit
y
,
t
h
e
s
y
s
tem
ca
n
s
elec
tiv
ely
tr
an
s
m
it
d
ata
o
n
ly
wh
en
s
p
ec
tr
u
m
ir
r
eg
u
lar
ities
o
r
ab
n
o
r
m
alities
ar
e
d
etec
ted
,
th
er
eb
y
r
e
d
u
cin
g
tr
an
s
m
is
s
io
n
lo
ad
a
n
d
im
p
r
o
v
i
n
g
d
ata
co
m
m
u
n
icatio
n
ef
f
icien
cy
.
T
o
s
u
p
p
o
r
t
in
tellig
en
ce
o
n
ed
g
e
d
ev
ices,
a
T
i
n
y
ML
a
p
p
r
o
ac
h
was
im
p
lem
en
ted
to
en
ab
le
lo
ca
lized
s
p
ec
tr
u
m
a
n
a
ly
s
is
with
m
in
im
al
r
eso
u
r
ce
co
n
s
u
m
p
tio
n
.
T
h
e
f
ir
s
t
co
n
tr
i
b
u
tio
n
o
f
th
is
p
ap
er
is
b
u
ild
in
g
a
d
ataset
b
ased
o
n
r
ea
l
ex
p
er
im
en
t
in
clu
d
i
n
g
th
e
b
ac
k
g
r
o
u
n
d
en
v
ir
o
n
m
en
t
in
n
u
clea
r
in
s
tallatio
n
wh
ich
co
m
es
f
r
o
m
th
e
g
am
m
a
s
p
ec
tr
u
m
e
n
er
g
y
co
n
v
er
te
d
to
a
g
r
ay
s
ca
le
im
ag
e.
T
h
e
s
ec
o
n
d
is
d
esig
n
in
g
a
m
o
d
el
with
h
ig
h
ac
cu
r
ac
y
an
d
em
b
ed
d
in
g
it
in
a
lo
w
p
o
wer
co
n
s
u
m
p
tio
n
d
ev
ice
to
ap
p
ly
T
in
y
ML
to
r
ec
o
g
n
ize
t
h
e
ty
p
es
o
f
r
ad
io
n
u
clid
es
r
elea
s
ed
in
th
e
en
v
ir
o
n
m
en
t.
A
p
er
f
o
r
m
an
ce
ev
alu
a
tio
n
was
test
ed
to
co
m
p
ar
e
th
e
ef
f
icien
c
y
o
f
th
e
s
y
s
tem
b
ef
o
r
e
an
d
af
ter
co
n
v
e
r
s
io
n
to
T
in
y
ML
,
in
clu
d
i
n
g
asp
ec
ts
o
f
m
o
d
el
s
ize
r
ed
u
ctio
n
,
in
f
er
e
n
ce
s
p
ee
d
,
a
n
d
im
p
ac
t
o
n
id
en
tific
atio
n
q
u
ality
.
Dis
cu
s
s
io
n
s
also
f
o
cu
s
ed
o
n
th
e
ca
u
s
es
o
f
d
ata
lo
ad
r
ed
u
ctio
n
an
d
its
i
m
p
licatio
n
s
f
o
r
s
y
s
tem
r
eliab
ilit
y
.
As
a
co
n
tin
u
atio
n
,
lo
n
g
-
ter
m
in
teg
r
atio
n
o
f
T
in
y
ML
in
to
th
e
m
o
n
ito
r
in
g
s
tatio
n
will
b
e
ca
r
r
ied
o
u
t
t
o
test
th
e
s
tab
ilit
y
an
d
ad
ap
tiv
it
y
o
f
th
e
s
y
s
tem
in
m
o
r
e
co
m
p
lex
f
ield
co
n
d
itio
n
s
.
2.
M
E
T
H
O
D
T
h
e
m
eth
o
d
o
lo
g
y
im
p
lem
en
te
d
in
th
is
s
tu
d
y
co
m
m
en
ce
d
wi
th
th
e
ac
q
u
is
itio
n
o
f
g
am
m
a
-
r
ay
s
p
ec
tr
al
d
ata
f
r
o
m
a
r
a
d
iatio
n
d
etec
t
io
n
s
y
s
tem
,
s
h
o
wn
in
Fig
u
r
e
1
.
T
h
ese
s
p
ec
tr
al
d
atasets
wer
e
s
u
b
s
eq
u
en
tly
tr
an
s
f
o
r
m
ed
i
n
to
g
r
a
y
s
ca
le
im
ag
e
r
ep
r
esen
tatio
n
s
,
s
er
v
i
n
g
as
in
p
u
t f
ea
tu
r
es f
o
r
tr
ain
i
n
g
a
C
NN.
T
h
is
ap
p
r
o
ac
h
lies
in
th
e
C
NN's
p
r
o
v
en
ca
p
ab
ilit
y
to
e
x
tr
ac
t
s
p
atial
p
at
ter
n
s
an
d
f
ea
tu
r
es
f
r
o
m
two
-
d
im
en
s
io
n
al
im
ag
e
in
p
u
ts
,
m
ak
in
g
it
s
u
itab
le
f
o
r
r
ec
o
g
n
izi
n
g
s
p
ec
tr
al
s
ig
n
atu
r
es
ass
o
ciate
d
with
d
if
f
er
en
t
r
ad
io
n
u
clid
es.
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
was
in
itially
co
n
d
u
cte
d
o
n
a
p
er
s
o
n
al
co
m
p
u
ter
(
PC
)
to
o
p
tim
ize
th
e
m
o
d
el
p
ar
am
eter
s
an
d
ev
alu
ate
its
lear
n
in
g
p
er
f
o
r
m
an
ce
.
On
ce
a
s
atis
f
ac
to
r
y
lev
el
o
f
class
if
icatio
n
ac
c
u
r
ac
y
an
d
m
o
d
el
g
en
er
aliza
tio
n
was
ac
h
iev
ed
,
th
e
tr
ain
ed
C
NN
m
o
d
el
was
co
n
v
er
ted
an
d
d
ep
lo
y
e
d
o
n
to
a
R
asp
b
er
r
y
Pi,
a
lo
w
-
p
o
wer
e
d
g
e
co
m
p
u
tin
g
d
ev
ice
wh
ich
r
e
p
r
esen
t
th
e
h
a
r
d
war
e
co
n
f
ig
u
r
atio
n
o
f
an
i
n
tellig
en
t
r
ad
iatio
n
m
o
n
ito
r
in
g
s
tatio
n
.
T
h
e
e
m
b
e
d
d
ed
m
o
d
el
was
th
en
s
u
b
jecte
d
to
a
s
er
ies
o
f
f
ield
t
r
ials
d
esig
n
ed
to
s
im
u
late
r
ea
lis
tic
en
v
ir
o
n
m
e
n
tal
co
n
d
i
tio
n
s
.
T
h
ese
f
ield
ev
alu
atio
n
s
aim
ed
to
v
er
if
y
th
e
in
f
e
r
e
n
ce
ac
cu
r
ac
y
a
n
d
r
o
b
u
s
tn
ess
o
f
th
e
C
NN
wh
en
o
p
er
atin
g
in
s
itu
,
as
well
as
to
ass
e
s
s
th
e
f
ea
s
ib
ilit
y
o
f
r
ea
l
-
tim
e
s
p
ec
tr
al
class
if
icatio
n
o
n
r
eso
u
r
ce
-
c
o
n
s
tr
ain
ed
h
ar
d
war
e.
Valid
atio
n
is
cr
u
cial
in
en
s
u
r
i
n
g
t
h
e
r
eliab
ilit
y
an
d
r
esp
o
n
s
iv
en
ess
o
f
in
tellig
en
t
m
o
n
ito
r
in
g
s
y
s
tem
s
in
p
r
ac
tic
al
ap
p
licatio
n
s
.
Fig
u
r
e
1
.
W
o
r
k
f
lo
w
d
iag
r
am
o
f
PC
an
d
R
asp
b
er
r
y
Pi p
r
o
ce
s
s
Spect
ru
m
d
at
as
et
G
ray
s
cal
e
i
m
ag
e
CN
N
Mo
d
el
T
i
n
y
ML
Mod
el
D
at
a
A
cq
u
i
s
i
t
i
o
n
G
ray
s
cal
e
i
m
ag
e
A
ccu
racy
Fi
el
d
T
es
t
i
n
g
P
r
o
ces
s
i
ng
i
n
P
C
P
r
o
ces
s
i
ng
i
n
R
a
s
pberry
P
i
D
ep
l
o
y
Fe
at
u
re
C
on
s
t
ruc
t
ion
In
p
u
t
Tra
in
in
g
TF
Lit
e
C
on
v
e
rs
ion
Final
Eva
lu
at
ion
Fe
at
u
re
C
on
s
t
ruc
t
ion
In
p
u
t
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.
1
6
,
No
.
1
,
Feb
r
u
ar
y
20
2
6
:
4
0
4
-
413
406
2
.
1
.
G
ra
y
s
ca
le
i
m
a
g
e
c
o
nv
er
s
io
n a
nd
da
t
a
s
et
co
ns
t
ruct
io
n
T
h
e
ex
p
er
im
en
ts
wer
e
co
n
d
u
c
ted
b
y
v
a
r
y
in
g
s
ev
er
al
p
ar
am
e
ter
s
,
in
clu
d
in
g
th
e
ty
p
e
o
f
r
a
d
io
n
u
clid
e
,
m
ea
s
u
r
em
en
t
d
u
r
atio
n
,
an
d
th
e
d
is
tan
ce
b
etwe
en
th
e
s
o
u
r
ce
an
d
th
e
d
etec
to
r
.
T
h
ese
v
ar
iat
io
n
s
wer
e
in
ten
d
ed
to
g
en
e
r
ate
d
iv
er
s
e
b
ac
k
g
r
o
u
n
d
d
ata
th
at
clo
s
ely
r
ef
lect
ac
tu
al
en
v
ir
o
n
m
e
n
tal
co
n
d
itio
n
s
.
R
ad
iatio
n
s
ig
n
als
wer
e
ac
q
u
ir
ed
u
s
in
g
a
s
cin
till
atio
n
d
etec
to
r
,
p
r
o
d
u
cin
g
in
ten
s
ity
d
ata
th
at
wer
e
p
lo
tted
in
to
a
r
esp
o
n
s
e
m
atr
ix
.
B
ased
o
n
th
e
c
h
ar
ac
ter
is
tic
en
er
g
y
p
ea
k
s
o
f
ea
ch
r
ad
io
n
u
cli
d
e,
lab
ellin
g
was
ap
p
lied
to
th
e
r
esu
ltin
g
s
p
ec
tr
al
d
ata.
Su
b
s
eq
u
en
tly
,
th
e
lab
ell
ed
s
p
ec
tr
a
wer
e
co
n
v
er
ted
a
n
d
m
ap
p
ed
in
to
g
r
ay
s
ca
le
im
ag
e
s
(
f
ea
tu
r
e
tr
an
s
f
er
)
,
th
en
ar
r
an
g
ed
in
to
a
d
ataset
th
at
s
u
itab
le
f
o
r
ap
p
licatio
n
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
ith
m
s
in
co
m
p
u
ter
.
Fig
u
r
e
2
ex
p
lain
s
th
e
co
n
v
er
s
io
n
o
f
g
a
m
m
a
s
p
ec
tr
u
m
to
g
r
ay
s
ca
le
im
ag
e
u
s
in
g
th
e
z
cu
r
v
e
m
eth
o
d
[
1
1
]
.
Fig
u
r
e
2
.
Gr
a
y
s
ca
le
im
ag
e
co
n
v
er
s
io
n
[
1
3
]
2
.
2
.
Desig
n a
nd
deplo
y
t
he
T
iny
M
L
m
o
del
Fig
u
r
e
3
s
h
o
ws
th
e
m
o
d
el
ar
c
h
itectu
r
e
s
tar
tin
g
with
a
n
in
p
u
t
lay
er
th
at
r
ec
eiv
es
a
3
2
×3
2
×
1
g
r
a
y
s
ca
le
im
ag
e.
First,
it
ap
p
lies
a
C
o
n
v
2
D
lay
er
with
3
2
f
ilter
s
an
d
r
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U
)
a
ctiv
atio
n
to
ex
tr
ac
t
b
asic
f
ea
tu
r
es
lik
e
ed
g
es,
f
o
l
lo
wed
b
y
a
Ma
x
Po
o
lin
g
2
D
l
ay
er
to
d
o
wn
s
am
p
le
th
e
f
ea
tu
r
e
m
ap
s
.
Nex
t,
a
C
o
n
v
2
D
lay
er
with
1
2
8
f
ilter
s
an
d
R
eL
U
ac
tiv
atio
n
ca
p
tu
r
es
m
o
r
e
co
m
p
lex
p
atter
n
s
,
ag
ain
f
o
llo
wed
b
y
a
Ma
x
Po
o
lin
g
2
D
lay
e
r
to
r
e
d
u
c
e
th
e
s
p
atial
s
ize.
T
h
e
th
ir
d
s
tep
in
v
o
lv
es
a
f
latten
lay
er
to
co
n
v
er
t
th
e
f
ea
tu
r
e
m
ap
th
at
r
ec
eiv
ed
f
r
o
m
th
e
m
ax
-
p
o
o
lin
g
lay
er
in
to
a
f
o
r
m
at
th
at
th
e
d
en
s
e
lay
er
s
ca
n
u
n
d
er
s
tan
d
.
Fin
ally
,
th
e
m
o
d
el
en
d
s
with
a
d
en
s
e
lay
er
with
a
f
ew
n
eu
r
o
n
s
eq
u
al
t
o
t
h
e
n
u
m
b
er
o
f
class
es,
u
s
in
g
So
f
tMa
x
ac
tiv
atio
n
t
o
o
u
tp
u
t c
lass
p
r
o
b
ab
ilit
ies,
in
th
is
ca
s
e
th
er
e
ar
e
f
o
u
r
class
es,
C
s
-
1
3
7
,
C
o
-
6
0
,
C
s
-
1
3
4
,
a
n
d
E
u
-
1
5
2
.
Fig
u
r
e
3
.
Mo
d
el
ar
ch
itectu
r
e
t
o
id
en
tify
r
ad
io
n
u
clid
e
Sc
o
r
e
4
Sc
o
r
e
3
Sco
r
e
2
Sc
o
r
e
1
Eu
-
152
Cs
-
134
Cs
-
137
Co
-
60
…
…
…
…
I
n
put
(3
2
,
3
2
,
1
)
Con
v2
D
(3
2
,
3
x3,
Re
L
U)
Co
n
v2
D
(1
2
8
,
3
x3,
Re
L
U)
Fl
a
tt
en
Den
se
(1
2
8
,
Re
L
U)
Dr
o
p
o
u
t(0
,
5
)
Den
se
(
n
u
m_
c
l
asse
s
,
so
f
tm
ax
)
Cl
assif
ic
a
tio
n
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
Tin
y
ma
ch
in
e
lea
r
n
in
g
w
ith
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l n
etw
o
r
k
fo
r
in
tellig
en
t
…
(
I
s
to
fa
)
407
Af
ter
th
e
m
o
d
el
is
d
esig
n
e
d
,
t
ested
an
d
v
alid
ated
,
it
is
th
e
n
em
b
ed
d
e
d
in
th
e
h
ar
d
war
e
s
y
s
tem
with
th
e
h
elp
o
f
T
en
s
o
r
Flo
w
L
ite
(
T
FLite)
.
T
FLite
i
s
u
s
ed
to
co
n
v
er
t
ML
m
o
d
els
th
at
in
itially
h
av
e
lar
g
e
m
em
o
r
y
an
d
ar
e
h
ea
v
y
to
r
u
n
,
in
to
s
m
aller
,
f
aster
,
a
n
d
m
o
r
e
ef
f
icien
t
v
er
s
io
n
s
s
o
th
at
th
e
y
ca
n
b
e
r
u
n
o
n
d
ev
ices
th
a
t
h
av
e
lim
ited
r
eso
u
r
ce
s
.
T
h
e
m
o
d
el
is
c
o
n
v
er
te
d
in
t
o
T
FLite
f
o
r
m
at
f
o
r
o
p
tim
izatio
n
,
s
o
th
e
m
o
d
el
s
ize
ca
n
b
e
s
m
aller
to
b
e
d
ep
lo
y
e
d
to
R
asp
b
er
r
y
Pi.
Af
ter
th
e
m
o
d
el
is
em
b
ed
d
e
d
in
th
e
h
ar
d
war
e,
te
s
tin
g
is
d
o
n
e
ag
ain
u
s
in
g
r
ad
io
n
u
clid
es
as
s
h
o
wn
in
Fig
u
r
e
4
.
I
m
p
lem
en
tatio
n
o
f
m
o
d
el
test
in
g
u
s
in
g
a
NaI
(
T
l)
d
etec
to
r
a
n
d
a
s
tan
d
ar
d
r
ad
iatio
n
s
o
u
r
ce
.
Da
ta
co
llectio
n
f
o
r
3
0
s
ec
o
n
d
s
with
a
d
is
tan
ce
o
f
5
0
cm
f
r
o
m
th
e
s
o
u
r
ce
to
th
e
d
etec
to
r
.
E
ac
h
test
was
ca
r
r
ie
d
o
u
t
2
0
tim
es.
T
h
e
p
o
s
itio
n
o
f
th
e
d
etec
to
r
a
n
d
s
tan
d
ar
d
s
o
u
r
ce
was
p
lace
d
p
ar
allel
as
wh
en
t
h
e
d
ataset
was
tak
en
an
d
was
also
k
ep
t
awa
y
f
r
o
m
walls
o
r
o
b
ject
s
th
at
co
u
ld
ca
u
s
e
b
ac
k
s
ca
tter
in
g
th
at
in
ter
f
er
ed
with
th
e
id
en
tific
atio
n
r
esu
lts
.
Fig
u
r
e
4
.
T
esti
n
g
o
f
T
in
y
ML
m
o
d
els em
b
ed
d
ed
in
s
y
s
tem
h
ar
d
war
e
u
s
in
g
r
ad
io
n
u
clid
e
s
tan
d
ar
d
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
F
e
a
t
ure
ex
t
r
a
ct
io
n a
nd
t
ra
ns
f
er
t
o
ima
g
e
T
h
e
tr
an
s
f
o
r
m
at
io
n
o
f
r
ad
io
n
u
cl
id
e
s
p
e
ctr
a
in
to
an
im
a
g
e
d
ata
s
et
is
il
lu
s
t
r
a
ted
in
F
ig
u
r
e
5
.
T
h
is
f
ig
u
r
e
p
r
e
s
en
ts
ex
am
p
le
s
o
f
th
e
tr
an
s
f
o
r
m
at
io
n
r
e
s
u
lt
s
f
o
r
C
s
-
1
3
7
,
C
o
-
6
0
,
C
s
-
1
3
4
,
an
d
E
u
-
1
5
2
,
o
b
ta
in
ed
wi
th
a
s
o
u
r
ce
-
to
-
d
et
ec
to
r
d
i
s
t
an
ce
o
f
2
0
cm
an
d
a
m
ea
s
u
r
e
m
en
t
ti
m
e
o
f
6
0
s
ec
o
n
d
s
.
F
ig
u
r
e
5
(
a)
s
h
o
w
s
th
e
tr
an
s
f
o
r
m
a
tio
n
o
f
th
e
g
a
m
m
a
s
p
ec
tr
u
m
o
f
C
o
-
6
0
in
t
o
a
g
r
ay
s
ca
le
im
ag
e,
d
i
s
p
lay
in
g
t
wo
d
i
s
t
in
ct
en
er
g
y
p
ea
k
s
.
F
ig
u
r
e
5
(
b
)
d
ep
ic
t
s
t
h
e
tr
an
s
f
o
r
m
a
ti
o
n
f
o
r
C
s
-
1
3
7
,
wh
i
ch
ex
h
i
b
i
t
s
a
s
in
g
l
e
p
ea
k
.
Fig
u
r
e
5
(
c)
co
r
r
e
s
p
o
n
d
s
to
C
s
-
1
3
4
,
an
d
Fig
u
r
e
5
(
d
)
to
E
u
-
1
5
2
,
b
o
th
o
f
wh
i
ch
co
n
t
ain
m
u
lt
ip
le
p
ea
k
s
.
T
h
e
g
am
m
a
s
p
e
ctr
u
m
im
ag
e
s
r
e
p
r
e
s
en
t
th
e
o
r
ig
in
a
l
s
p
ec
tr
a
o
f
e
ac
h
r
a
d
io
n
u
c
li
d
e,
in
c
lu
d
in
g
th
e
ir
c
h
ar
ac
te
r
i
s
ti
c
en
er
g
y
p
ea
k
s
.
T
h
e
s
e
o
r
ig
in
a
l
s
p
e
c
tr
a
wer
e
th
en
co
n
v
e
r
t
ed
i
n
to
n
o
r
m
al
iz
ed
o
n
e
-
d
i
m
en
s
i
o
n
al
s
p
ec
tr
a
af
t
er
b
ac
k
g
r
o
u
n
d
co
r
r
e
ct
io
n
,
b
ef
o
r
e
b
ei
n
g
tr
an
s
f
o
r
m
ed
in
to
g
r
ay
s
ca
l
e
im
ag
e
s
.
U
s
in
g
g
r
a
y
s
ca
le
im
ag
e
s
(1
-
ch
an
n
el
)
in
s
te
ad
o
f
R
G
B
(3
-
c
h
an
n
el)
f
o
r
tr
ain
in
g
c
la
s
s
i
f
i
ca
tio
n
m
o
d
el
s
r
ed
u
ce
s
co
m
p
u
t
at
io
n
al
co
m
p
lex
i
ty
an
d
m
em
o
r
y
u
s
ag
e
,
m
ak
in
g
th
e
m
m
o
r
e
ef
f
i
ci
en
t
f
o
r
d
ep
l
o
y
m
en
t
.
S
in
c
e
g
r
ay
s
ca
le
im
ag
e
s
h
av
e
o
n
ly
o
n
e
in
t
en
s
i
ty
ch
an
n
e
l,
th
e
y
r
eq
u
ir
e
l
es
s
s
t
o
r
ag
e,
f
a
s
t
er
p
r
o
ce
s
s
in
g
,
a
n
d
s
m
al
ler
m
o
d
e
l
s
i
ze
s
,
wh
i
ch
is
b
en
ef
i
ci
al
f
o
r
T
in
y
ML
ap
p
l
ic
at
io
n
s
.
T
h
ese
g
r
ay
s
ca
le
im
a
g
es
ar
e
b
o
th
th
e
d
ataset
an
d
th
e
in
p
u
t
f
o
r
th
e
C
NN
m
o
d
el
id
en
tific
ati
o
n
p
r
o
ce
s
s
.
T
h
e
co
lo
r
g
r
ad
atio
n
f
o
r
ea
c
h
p
ix
el
o
f
th
e
im
ag
e
ca
r
r
ies
in
f
o
r
m
atio
n
ab
o
u
t
th
e
v
al
u
e
o
f
r
ad
iatio
n
in
ten
s
ity
at
ce
r
tain
ch
an
n
el
p
o
s
itio
n
s
th
at
r
ep
r
esen
t
th
e
ch
ar
ac
ter
is
tics
o
f
ce
r
tain
ty
p
es
o
f
r
ad
io
n
u
clid
e
s
.
T
h
ese
g
r
ay
s
ca
le
im
ag
es
wer
e
tr
ain
ed
u
s
in
g
Ker
as
in
co
m
p
u
ter
,
test
ed
an
d
v
alid
ated
with
a
d
ata
r
atio
o
f
7
0
:2
0
:
1
0
.
T
h
e
r
ad
io
n
u
clid
e
id
e
n
tific
atio
n
m
o
d
el
o
b
tain
ed
f
r
o
m
th
e
in
itial
tr
ain
in
g
co
u
ld
n
o
t
b
e
d
ir
ec
tly
em
b
e
d
d
ed
in
to
a
T
in
y
ML
e
n
v
ir
o
n
m
en
t.
I
n
T
in
y
ML
,
Ker
as
is
u
s
ed
to
d
esig
n
an
d
tr
ain
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els
o
n
p
o
wer
f
u
l
s
y
s
tem
s
,
wh
ile
T
FLite
o
p
ti
m
izes
th
ese
m
o
d
els
f
o
r
d
ep
lo
y
m
en
t
o
n
r
eso
u
r
ce
-
c
o
n
s
tr
a
in
ed
ed
g
e
d
e
v
ices.
C
o
n
v
er
tin
g
a
Ker
as
m
o
d
el
to
T
FLite
r
ed
u
ce
s
m
em
o
r
y
u
s
ag
e
an
d
s
p
ee
d
s
u
p
in
f
er
en
ce
,
m
ak
in
g
it
f
ea
s
ib
le
f
o
r
lim
ited
r
an
d
o
m
-
ac
ce
s
s
m
em
o
r
y
(
R
AM
)
.
Desp
ite
th
e
m
o
d
el
m
ay
s
lig
h
tly
r
e
d
u
ce
ac
c
u
r
ac
y
,
it
is
n
ec
ess
ar
y
to
f
it
m
o
d
els
in
to
tin
y
d
ev
ices,
en
s
u
r
in
g
e
f
f
icien
t,
l
o
w
-
p
o
wer
ex
ec
u
tio
n
with
o
u
t
r
e
q
u
ir
in
g
clo
u
d
d
ep
en
d
e
n
cy
.
T
h
u
s
,
T
FLite
s
u
p
p
o
r
t T
in
y
ML
to
b
e
ap
p
licab
le
o
n
en
v
ir
o
n
m
en
t
m
o
n
ito
r
in
g
d
ev
ice
in
r
em
o
te
ar
ea
.
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.
1
6
,
No
.
1
,
Feb
r
u
ar
y
20
2
6
:
4
0
4
-
413
408
Fig
u
r
e
5
.
Featu
r
e
tr
an
s
f
er
f
r
o
m
s
p
ec
tr
u
m
v
ec
t
o
r
to
im
a
g
e
(
a
)
C
o
-
60
,
(
b
)
C
s
-
1
3
7
,
(
c)
C
s
-
134
,
an
d
(
d
)
E
u
-
1
5
2
3
.
2
.
Sy
s
t
e
m
predict
io
n a
cc
ura
cy
I
n
th
e
co
n
tex
t
o
f
T
i
n
y
ML
,
m
ac
h
in
e
lear
n
in
g
m
o
d
els
wh
ich
d
ep
lo
y
e
d
o
n
R
asp
b
er
r
y
m
u
s
t
co
m
b
in
e
th
e
f
u
n
ctio
n
o
f
Ker
as a
n
d
T
e
n
s
o
r
Flo
w
L
ite
(
T
FLite)
to
p
r
o
v
i
d
e
an
ef
f
icien
t f
r
a
m
ewo
r
k
.
Ker
as e
x
ce
ls
in
m
o
d
el
co
n
s
tr
u
ctio
n
an
d
tr
ai
n
in
g
,
w
h
ile
T
FLite
p
r
o
v
id
es
th
e
n
ec
ess
ar
y
o
p
tim
izatio
n
s
f
o
r
r
e
al
-
tim
e,
o
n
-
d
e
v
ice
in
f
er
en
ce
.
Ker
as
is
wid
ely
u
s
e
d
f
o
r
m
o
d
el
d
esig
n
an
d
tr
ain
in
g
,
wh
ile
T
FLite
i
s
a
l
ig
h
tweig
h
t
in
f
er
en
ce
en
g
in
e
w
h
i
c
h
e
n
a
b
l
es
m
o
d
e
l
e
x
e
c
u
t
i
o
n
o
n
e
d
g
e
d
e
v
i
c
e
s
.
I
t
a
l
l
o
w
s
m
o
d
e
l
s
t
r
a
i
n
e
d
w
i
t
h
Ke
r
a
s
t
o
b
e
c
o
n
v
e
r
t
e
d
i
n
t
o
a
n
e
f
f
i
c
i
e
n
t
T
F
L
it
e
f
o
r
m
a
t
,
s
i
g
n
i
f
ic
a
n
t
l
y
r
e
d
u
c
i
n
g
f
i
le
s
i
z
e
a
n
d
co
m
p
u
t
a
t
i
o
n
a
l
r
e
q
u
i
r
e
m
e
n
ts
t
h
r
o
u
g
h
o
p
t
i
m
i
z
a
t
i
o
n
s
s
u
c
h
a
s
q
u
a
n
t
iz
a
t
i
o
n
a
n
d
o
p
e
r
a
to
r
f
u
s
i
o
n
.
At
th
e
co
n
v
er
s
io
n
f
r
o
m
Ker
a
s
to
T
FLite
f
o
r
m
at,
d
ef
au
lt
s
ettin
g
is
im
p
lem
en
ted
.
T
h
is
m
ea
n
s
th
at
o
p
tim
izatio
n
an
d
q
u
an
tizatio
n
p
r
o
ce
s
s
is
n
o
t
u
tili
ze
d
to
o
b
tai
n
ef
f
icien
t
an
d
s
m
aller
f
ile
s
iz
e
m
o
d
el.
Ho
wev
er
,
ev
en
th
o
u
g
h
t
h
e
c
o
n
v
e
r
s
io
n
p
r
o
ce
s
s
d
o
es
n
o
t
co
n
d
u
ct
o
p
tim
izatio
n
an
d
q
u
an
tizatio
n
,
th
e
T
FLite
m
o
d
el
w
h
ich
r
esu
lted
is
s
till
s
m
aller
th
an
Ker
as
u
n
to
its
h
alf
-
s
ize.
Half
-
s
ize
r
esu
lt
is
ac
h
iev
ed
f
r
o
m
co
n
v
er
s
io
n
p
r
o
ce
s
s
b
ec
au
s
e
at
th
is
p
r
o
ce
s
s
,
d
ata
p
r
u
n
in
g
an
d
im
p
licit
co
m
p
r
ess
io
n
ar
e
u
tili
ze
d
.
W
h
en
co
n
v
er
t
ed
in
to
T
FLite,
th
e
u
n
em
p
lo
y
ed
p
ar
ts
at
in
f
er
en
ce
ar
e
d
is
ca
r
d
ed
,
in
clu
d
in
g
n
o
d
e
Gr
af
tr
ain
in
g
.
T
h
r
o
u
g
h
im
p
licit c
o
m
p
r
ess
io
n
,
th
e
f
lat
b
u
f
f
e
r
s
tr
u
ctu
r
e
s
to
r
es
in
f
o
r
m
atio
n
i
n
to
v
e
r
y
c
o
m
p
ac
t
f
o
r
m
with
o
u
t
Py
th
o
n
f
r
am
ew
o
r
k
o
v
er
h
ea
d
.
Oth
er
p
ar
am
eter
s
s
u
ch
as
m
etad
ata,
ch
ec
k
p
o
i
n
t,
g
r
ap
h
o
p
s
,
s
ig
n
at
u
r
e
an
d
tr
ain
i
n
g
i
n
f
o
r
m
atio
n
in
Ker
as
m
o
d
el,
a
r
e
also
ab
an
d
o
n
ed
d
u
r
in
g
th
e
c
o
n
v
er
s
io
n
p
r
o
ce
s
s
.
T
h
e
r
em
ai
n
in
g
in
f
o
r
m
atio
n
is
th
en
b
ec
o
m
e
s
tr
u
ctu
r
e
d
an
d
s
er
ialized
.
T
h
er
ef
o
r
e,
t
h
e
h
al
f
-
s
ize
co
m
p
r
ess
io
n
m
o
d
el
is
n
o
t
p
r
o
ce
s
s
ed
b
y
q
u
an
tizatio
n
f
r
o
m
f
lo
at3
2
to
o
th
er
s
f
o
r
m
at
s
u
ch
as in
t
8
,
in
t1
6
,
b
u
t
r
ath
er
th
r
o
u
g
h
d
ata
p
r
u
n
i
n
g
an
d
co
m
p
r
ess
io
n
p
r
o
ce
s
s
es.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
r
esu
lts
ex
p
lain
ed
in
Fig
u
r
e
6
s
h
o
ws
th
at
th
e
m
o
d
el
p
er
f
o
r
m
s
v
er
y
well,
s
in
ce
m
o
s
t
p
r
ed
ictio
n
s
f
all
alo
n
g
t
h
e
d
iag
o
n
al
(
c
o
r
r
ec
t
class
if
icati
o
n
s
)
.
Fro
m
th
e
co
n
f
u
s
io
n
m
at
r
ix
r
esu
lts
,
we
ca
n
ca
lcu
late
th
e
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all
a
n
d
f
1
s
co
r
e
f
o
r
ea
c
h
class
,
as
p
r
esen
ted
in
T
ab
le
1
.
T
h
e
r
esu
lts
s
h
o
w
in
a
v
e
r
y
g
o
o
d
v
al
u
e,
esp
ec
ial
ly
in
th
e
C
s
-
1
3
4
class
wh
ich
g
ets
th
e
h
ig
h
est
ac
cu
r
ac
y
r
esu
lt
o
f
9
9
.
8
9
0
%,
s
ee
n
in
Fig
u
r
e
6
(
a)
.
Ov
er
all
ac
cu
r
a
cy
f
o
r
th
e
f
o
u
r
class
es
was
9
9
.
3
3
8
%.
Fo
r
in
f
o
r
m
atio
n
,
th
e
s
ize
o
f
th
e
s
to
r
ag
e
m
em
o
r
y
o
f
t
h
e
m
o
d
el
th
at
h
a
s
b
ee
n
tr
ain
ed
u
s
in
g
Ke
r
as
is
8
MB,
wh
ile
th
e
s
ize
o
f
th
e
m
em
o
r
y
af
ter
b
ein
g
tr
an
s
f
o
r
m
ed
in
T
FLite
is
2
MB,
r
ed
u
ce
d
b
y
ap
p
r
o
x
im
ately
o
n
e
q
u
ar
ter
tim
e.
Fig
u
r
e
6
(
b
)
p
r
esen
ts
th
e
c
o
n
f
u
s
io
n
m
atr
ix
r
esu
lts
o
f
th
e
T
FLite
m
o
d
el
with
a
lo
wer
m
e
m
o
r
y
s
ize
test
ed
o
n
a
R
asp
b
er
r
y
d
ev
ice.
T
h
e
r
esu
lts
in
d
icate
a
lo
wer
ac
cu
r
ac
y
f
o
r
ea
ch
class
co
m
p
ar
e
d
to
p
r
e
v
io
u
s
test
s
.
T
h
e
h
ig
h
est
ac
cu
r
ac
y
is
ac
h
ie
v
ed
f
o
r
th
e
C
o
-
6
0
class
at
9
4
.
6
6
5
%,
wh
ile
th
e
C
s
-
1
3
4
class
r
ec
o
r
d
s
th
e
lo
west
ac
cu
r
ac
y
at
8
8
.
0
2
9
%,
as
s
h
o
wn
in
T
ab
le
1
.
T
h
e
o
v
er
al
l
m
o
d
el
a
cc
u
r
ac
y
is
8
4
.
5
6
8
%.
T
h
is
d
ec
r
e
ase
i
n
p
e
r
f
o
r
m
a
n
c
e
m
a
y
b
e
a
ttr
i
b
u
te
d
,
i
n
p
a
r
t
,
to
t
h
e
R
as
p
b
e
r
r
y
'
s
r
elat
iv
el
y
lim
ite
d
p
r
o
c
ess
i
n
g
c
a
p
a
b
il
iti
es
co
m
p
a
r
e
d
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
Tin
y
ma
ch
in
e
lea
r
n
in
g
w
ith
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l n
etw
o
r
k
fo
r
in
tellig
en
t
…
(
I
s
to
fa
)
409
to
a
s
t
an
d
ar
d
c
o
m
p
u
t
er
.
Ne
v
e
r
th
e
less
,
f
o
r
en
v
i
r
o
n
m
en
tal
m
o
n
it
o
r
i
n
g
s
t
ati
o
n
a
p
p
lic
ati
o
n
s
,
t
h
is
le
v
el
o
f
a
cc
u
r
ac
y
r
e
m
ai
n
s
a
cc
e
p
t
a
b
le
,
alt
h
o
u
g
h
f
u
r
th
e
r
i
m
p
r
o
v
e
m
e
n
ts
a
r
e
p
la
n
n
ed
t
o
e
n
h
an
ce
t
h
e
m
o
d
e
l’
s
p
e
r
f
o
r
m
a
n
ce
.
3
.
3
.
E
v
a
lua
t
i
o
n o
f
t
he
T
iny
M
L
m
o
del e
m
bedd
ed
i
n ha
rdwa
re
T
h
e
m
o
s
t
ac
cu
r
ate
ar
ch
itectu
r
al
m
o
d
el
was
in
teg
r
ated
in
to
th
e
R
asp
b
er
r
y
Pi
h
ar
d
war
e
s
y
s
tem
an
d
ev
alu
ated
th
r
o
u
g
h
d
ir
ec
t
test
in
g
with
r
ad
io
n
u
clid
es,
as
illu
s
tr
ated
in
Fig
u
r
e
4
.
T
h
e
e
x
p
e
r
im
en
tal
p
r
o
ce
d
u
r
e
in
v
o
lv
ed
p
ar
titi
o
n
i
n
g
th
e
d
ata
ac
co
r
d
i
n
g
to
r
ad
i
o
n
u
clid
e
ty
p
e.
T
ab
le
2
d
is
p
lay
s
th
e
test
r
e
s
u
lts
,
wh
ich
r
ev
ea
l
v
ar
iatio
n
s
in
ac
cu
r
ac
y
ac
r
o
s
s
d
if
f
er
en
t
class
es.
T
h
e
ac
cu
r
ac
y
r
esu
lts
o
f
th
e
C
NN
m
o
d
el
te
s
t
in
T
in
y
ML
u
s
in
g
d
ir
ec
t
m
ea
s
u
r
em
e
n
t
d
ata
f
r
o
m
th
e
d
etec
to
r
ar
e
lo
we
r
co
m
p
ar
ed
to
th
e
tr
ain
in
g
d
ataset.
T
h
is
d
if
f
er
en
ce
is
ca
u
s
ed
b
y
s
ev
er
al
f
ac
to
r
s
r
el
ated
to
d
ata
ch
ar
ac
ter
is
tics
,
t
h
e
r
ea
l
en
v
ir
o
n
m
en
t,
an
d
th
e
lim
itatio
n
s
o
f
th
e
T
in
y
ML
p
latf
o
r
m
.
Dir
ec
t
m
e
asu
r
em
en
t
d
ata
u
s
u
ally
h
as
h
ig
h
er
v
a
r
iab
ilit
y
d
u
e
to
n
o
is
e
f
r
o
m
th
e
s
en
s
o
r
,
en
v
ir
o
n
m
en
tal
f
lu
ctu
atio
n
s
,
o
r
h
ar
d
war
e
in
co
n
s
is
ten
cies.
I
f
th
e
tr
ai
n
in
g
d
ataset
d
o
e
s
n
o
t
in
clu
d
e
t
h
ese
v
ar
iatio
n
s
,
th
e
m
o
d
el
h
as d
if
f
icu
lty
r
ec
o
g
n
izin
g
p
atter
n
s
in
r
ea
l d
ata.
I
n
ad
d
itio
n
,
tr
ain
in
g
d
atasets
ar
e
o
f
ten
tak
en
f
r
o
m
co
n
tr
o
lled
co
n
d
itio
n
s
,
an
d
th
er
ef
o
r
e
less
r
ep
r
esen
tativ
e
o
f
th
e
ac
tu
al
c
o
n
d
itio
n
s
u
n
d
er
wh
ich
d
ir
ec
t
m
ea
s
u
r
em
en
ts
ar
e
m
a
d
e.
Fo
r
ex
am
p
le,
v
a
r
iatio
n
s
in
r
ad
io
n
u
clid
e
ac
tiv
ity
,
en
v
ir
o
n
m
en
tal
in
ter
f
er
en
ce
,
o
r
d
if
f
er
en
t
d
etec
to
r
ch
ar
ac
ter
is
tics
ar
e
n
o
t
alwa
y
s
r
ep
r
esen
ted
in
t
h
e
tr
ai
n
in
g
d
ataset.
I
f
th
e
d
ataset
lac
k
s
th
ese
v
ar
iatio
n
s
,
t
h
e
C
NN
m
o
d
el
ten
d
s
to
s
im
p
ly
m
em
o
r
ize
p
atter
n
s
f
r
o
m
th
e
t
r
ain
in
g
d
ata
with
o
u
t
b
ein
g
a
b
le
to
g
en
er
alize
to
n
ew
d
ata
.
T
h
is
ca
n
also
b
e
ex
ac
er
b
ated
if
t
h
e
m
o
d
el
s
u
f
f
er
s
f
r
o
m
o
v
er
f
itti
n
g
,
wh
e
r
e
th
e
m
o
d
el
f
o
c
u
s
es
to
o
m
u
ch
o
n
s
p
ec
if
ic
p
atter
n
s
in
th
e
tr
ain
in
g
d
ataset
an
d
is
u
n
a
b
le
to
h
an
d
le
v
ar
iatio
n
s
in
th
e
d
ir
ec
t m
ea
s
u
r
em
en
t
d
ata.
(
a)
(
b
)
Fig
u
r
e
6
.
C
o
n
f
u
s
io
n
m
atr
i
x
(
a)
u
s
in
g
“Ke
r
as”
m
o
d
el
a
n
d
(
b
)
u
s
in
g
“T
FLite”
m
o
d
el
T
ab
le
1
.
Mo
d
el
p
er
f
o
r
m
an
ce
b
ased
o
n
Ker
as tr
ain
in
g
v
s
T
FLite
tr
ain
in
g
M
a
t
r
i
x
e
v
a
l
u
a
t
i
o
n
M
o
d
e
l
t
r
a
i
n
e
d
u
s
i
n
g
K
e
r
a
s
M
o
d
e
l
t
r
a
i
n
e
d
u
s
i
n
g
TFLi
t
e
Co
-
60
Cs
-
1
3
7
Cs
-
1
3
4
Eu
-
152
Co
-
60
Cs
-
1
3
7
Cs
-
1
3
4
Eu
-
152
A
c
c
u
r
a
c
y
9
9
.
5
5
9
9
9
.
4
0
5
9
9
.
8
9
0
9
9
.
8
2
4
9
4
.
6
6
5
9
2
.
6
1
5
8
8
.
0
2
9
9
3
.
8
2
7
P
r
e
c
i
s
i
o
n
9
9
.
4
6
7
9
8
.
5
9
5
9
9
.
6
4
8
9
9
.
6
4
7
1
0
0
.
0
0
0
7
7
.
8
7
9
7
1
.
7
7
6
1
0
0
.
0
0
0
R
e
c
a
l
l
9
8
.
7
6
5
9
9
.
0
3
0
9
9
.
9
1
2
9
9
.
6
4
7
7
8
.
6
6
0
9
8
.
4
1
3
8
5
.
8
9
1
7
5
.
3
0
9
F
1
sc
o
r
e
9
9
.
1
1
5
9
8
.
8
1
2
9
9
.
7
8
0
9
9
.
6
4
7
8
8
.
0
5
5
8
6
.
9
5
0
7
8
.
2
0
2
8
5
.
9
1
5
M
a
c
r
o
-
p
r
e
c
i
si
o
n
9
9
.
3
3
9
8
7
.
4
1
4
M
a
c
r
o
-
r
e
c
a
l
l
9
9
.
3
3
8
8
4
.
5
6
8
M
a
c
r
o
-
F1
9
9
.
3
3
7
8
4
.
7
8
1
O
v
e
r
a
l
l
A
c
c
u
r
a
c
y
9
9
.
3
3
8
8
4
.
5
6
8
I
n
ad
d
itio
n
,
tr
ain
in
g
d
atasets
ar
e
o
f
ten
tak
en
f
r
o
m
co
n
tr
o
lled
co
n
d
itio
n
s
,
an
d
th
er
ef
o
r
e
less
r
ep
r
esen
tativ
e
o
f
th
e
ac
tu
al
c
o
n
d
itio
n
s
u
n
d
er
wh
ich
d
ir
ec
t
m
ea
s
u
r
em
en
ts
ar
e
m
a
d
e.
Fo
r
ex
am
p
le,
v
a
r
iatio
n
s
in
r
ad
io
n
u
clid
e
ac
tiv
ity
,
en
v
ir
o
n
m
en
tal
in
ter
f
er
en
ce
,
o
r
d
if
f
er
en
t
d
etec
to
r
ch
ar
ac
ter
is
tics
ar
e
n
o
t
alwa
y
s
r
ep
r
esen
ted
in
t
h
e
tr
ai
n
in
g
d
ataset.
I
f
th
e
d
ataset
lac
k
s
th
ese
v
ar
iatio
n
s
,
t
h
e
C
NN
m
o
d
el
ten
d
s
to
s
im
p
ly
m
em
o
r
ize
p
atter
n
s
f
r
o
m
th
e
t
r
ain
in
g
d
ata
with
o
u
t
b
ein
g
a
b
le
to
g
en
er
alize
to
n
ew
d
ata
.
T
h
is
ca
n
also
b
e
ex
ac
er
b
ated
if
t
h
e
m
o
d
el
s
u
f
f
er
s
f
r
o
m
o
v
er
f
itti
n
g
,
wh
e
r
e
th
e
m
o
d
el
f
o
c
u
s
es
to
o
m
u
ch
o
n
s
p
ec
if
ic
p
atter
n
s
in
th
e
tr
ain
in
g
d
ataset
an
d
is
u
n
a
b
le
to
h
an
d
le
v
ar
iatio
n
s
in
th
e
d
ir
ec
t m
ea
s
u
r
em
en
t
d
ata.
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.
1
6
,
No
.
1
,
Feb
r
u
ar
y
20
2
6
:
4
0
4
-
413
410
T
ab
le
2
.
R
esu
lts
o
f
test
in
g
m
o
d
el
-
4
o
n
R
asp
b
er
r
y
with
d
etec
to
r
m
ea
s
u
r
em
en
t
d
ata
D
a
t
a
s
e
t
N
u
mb
e
r
o
f
T
e
s
t
s
T
r
u
e
P
r
e
d
i
c
t
i
o
n
F
a
l
se
P
r
e
d
i
c
t
i
o
n
A
c
c
u
r
a
c
y
(
%)
Co
-
60
20
14
6
70
Cs
-
1
3
7
20
17
3
85
Cs
-
1
3
4
20
16
4
80
Eu
-
152
20
17
3
85
Acc
o
r
d
in
g
t
o
T
ab
le
2
,
it
ca
n
b
e
an
aly
ze
d
th
at
th
e
h
ig
h
ac
c
u
r
ac
y
o
f
C
s
-
1
3
4
is
m
o
s
t
lik
ely
d
u
e
t
o
its
s
p
ec
tr
al
ch
ar
ac
ter
is
tics
th
at
r
esem
b
le
th
e
b
ac
k
g
r
o
u
n
d
,
m
ak
i
n
g
it
ea
s
ier
f
o
r
th
e
m
o
d
el
to
r
ec
o
g
n
ize
th
e
p
atter
n
co
m
p
ar
ed
to
o
th
er
m
o
r
e
co
m
p
lex
r
a
d
io
n
u
cli
d
es.
I
n
ad
d
itio
n
,
b
ias
in
t
h
e
tr
ain
in
g
d
ataset,
s
u
ch
as
th
e
d
o
m
in
an
ce
o
f
d
ata
th
at
r
esem
b
les
th
e
b
ac
k
g
r
o
u
n
d
o
r
s
im
p
l
e
p
atter
n
s
,
ca
n
im
p
r
o
v
e
th
e
p
r
ed
ictio
n
o
f
C
s
-
1
3
4
.
W
h
en
C
s
-
1
3
4
ap
p
ea
r
s
in
co
m
b
in
atio
n
with
o
th
er
r
a
d
io
n
u
clid
es,
its
s
tab
le
co
n
tr
ib
u
tio
n
h
elp
s
th
e
m
o
d
el
r
ec
o
g
n
ize
t
h
e
o
v
er
all
p
atter
n
b
etter
.
Ho
wev
e
r
,
d
if
f
er
en
c
es
in
ac
cu
r
ac
y
b
etwe
en
r
ad
i
o
n
u
clid
es
ca
n
also
b
e
in
f
lu
en
ce
d
b
y
n
o
is
e,
d
ata
im
b
alan
ce
,
o
r
lack
o
f
v
ar
iatio
n
in
th
e
tr
ain
in
g
d
ataset,
wh
ich
r
e
q
u
ir
e
im
p
r
o
v
em
en
ts
to
im
p
r
o
v
e
m
o
d
el
g
en
e
r
aliza
tio
n
.
E
n
v
ir
o
n
m
en
tal
f
ac
t
o
r
s
also
p
lay
a
s
ig
n
if
ican
t
r
o
le
in
r
ed
u
cin
g
ac
cu
r
ac
y
.
R
ea
l
-
wo
r
ld
d
ir
ec
t
m
ea
s
u
r
em
en
ts
ar
e
o
f
ten
a
f
f
ec
ted
b
y
co
n
d
itio
n
s
s
u
ch
as
tem
p
er
atu
r
e,
h
u
m
id
ity
,
o
r
elec
tr
o
m
ag
n
etic
in
ter
f
er
en
ce
.
T
h
ese
co
n
d
itio
n
s
ar
e
n
o
t a
lway
s
ca
p
tu
r
ed
in
th
e
tr
ain
in
g
d
ataset,
s
o
th
e
m
o
d
el
ca
n
n
o
t a
d
ap
t w
ell.
T
h
e
s
ig
n
al
v
ar
iatio
n
s
p
r
o
d
u
ce
d
b
y
th
e
r
ad
iatio
n
d
etec
to
r
in
d
ir
ec
t
m
ea
s
u
r
em
en
ts
ca
n
also
d
if
f
er
s
ig
n
if
ican
tly
f
r
o
m
th
e
tr
ain
in
g
d
ata,
m
a
k
in
g
th
e
m
o
d
el'
s
p
r
ed
ictio
n
s
less
ac
cu
r
ate.
T
h
e
test
ed
T
in
y
ML
s
y
s
tem
,
i
m
p
lem
en
ted
o
n
a
R
asp
b
er
r
y
Pi,
h
as
b
ee
n
in
teg
r
ated
with
ad
d
itio
n
al
m
icr
o
co
n
tr
o
ller
co
m
p
o
n
en
ts
with
in
an
en
v
ir
o
n
m
en
tal
m
o
n
ito
r
in
g
s
tatio
n
,
as
illu
s
tr
ate
d
in
Fig
u
r
e
7
.
T
h
is
s
tatio
n
en
co
m
p
ass
es
n
o
t
o
n
l
y
a
m
o
d
el
f
o
r
m
o
n
ito
r
in
g
r
a
d
i
o
n
u
clid
e
r
elea
s
es
b
u
t
also
s
en
s
o
r
s
f
o
r
c
o
llectin
g
m
eteo
r
o
lo
g
ical
a
n
d
h
u
m
id
ity
d
ata.
T
o
s
u
p
p
o
r
t
en
e
r
g
y
au
to
n
o
m
y
in
r
em
o
te
o
r
o
f
f
-
g
r
i
d
lo
c
atio
n
s
,
th
e
s
y
s
tem
is
eq
u
ip
p
e
d
with
s
o
lar
p
a
n
els.
I
n
th
e
f
u
tu
r
e,
th
is
s
tatio
n
is
in
te
n
d
ed
to
b
e
d
ep
lo
y
e
d
in
is
o
late
d
ar
ea
s
to
m
o
n
ito
r
r
ad
io
n
u
clid
e
d
is
p
er
s
io
n
ca
r
r
ie
d
b
y
win
d
f
r
o
m
v
a
r
io
u
s
s
o
u
r
ce
s
.
Fig
u
r
e
7
.
R
asp
b
er
r
y
p
i
h
ar
d
wa
r
e
in
teg
r
ated
i
n
to
en
v
ir
o
n
m
en
t
al
m
o
n
ito
r
in
g
s
tatio
n
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
d
em
o
n
s
tr
ates
th
e
s
u
cc
ess
f
u
l
im
p
lem
en
tatio
n
o
f
a
T
i
n
y
ML
m
o
d
el
f
o
r
r
ea
l
-
tim
e
class
if
icatio
n
o
f
r
ad
io
n
u
clid
es
in
an
em
b
ed
d
e
d
en
v
ir
o
n
m
e
n
tal
m
o
n
ito
r
in
g
s
y
s
tem
.
Ker
as
an
d
T
en
s
o
r
Flo
w
L
it
e
s
er
v
e
co
m
p
lem
e
n
tar
y
r
o
les
in
th
e
d
e
v
elo
p
m
e
n
t
an
d
d
ep
l
o
y
m
en
t
o
f
T
in
y
ML
m
o
d
els
o
n
d
e
v
ices
lik
e
th
e
R
asp
b
er
r
y
Pi.
T
h
e
o
p
tim
ized
m
o
d
el
b
ased
o
n
r
esu
lt
ac
h
iev
e
d
h
ig
h
ac
cu
r
ac
y
o
f
9
9
.
3
3
8
%
t
r
ain
ed
u
s
in
g
Ker
as
,
an
d
8
4
.
5
6
8
%
tr
ain
e
d
u
s
in
g
T
FLite.
Fo
r
th
e
r
ea
l
m
ea
s
u
r
em
en
t
u
s
in
g
th
e
h
ar
d
war
e,
t
h
e
h
ig
h
est
ac
cu
r
ac
y
o
b
tain
ed
8
5
%
f
o
r
E
u
-
1
5
2
class
.
T
h
is
in
teg
r
ated
s
y
s
tem
n
o
t
o
n
ly
m
o
n
ito
r
s
r
a
d
io
ac
tiv
e
r
elea
s
es
b
u
t
also
tr
ac
k
s
wea
th
er
an
d
h
u
m
id
it
y
d
ata,
en
h
an
cin
g
en
v
ir
o
n
m
en
t
al
s
u
r
v
eillan
ce
ca
p
ab
ilit
ies.
W
ith
s
o
lar
-
p
o
wer
ed
en
er
g
y
au
to
n
o
m
y
,
th
e
s
o
lu
tio
n
is
s
u
itab
le
f
o
r
r
em
o
te
d
e
p
lo
y
m
en
ts
,
en
a
b
lin
g
ea
r
ly
d
etec
tio
n
o
f
r
a
d
io
n
u
clid
e
d
is
p
er
s
io
n
v
ia
win
d
cu
r
r
en
ts
.
Fu
tu
r
e
wo
r
k
will
f
o
cu
s
o
n
s
ca
lin
g
th
e
s
y
s
tem
f
o
r
wid
er
g
eo
g
r
ap
h
ic
co
v
er
ag
e
,
im
p
r
o
v
in
g
m
o
d
el
r
o
b
u
s
tn
ess
with
ad
d
itio
n
al
d
ata,
a
n
d
in
te
g
r
atin
g
wir
eless
s
en
s
o
r
n
etwo
r
k
s
f
o
r
r
ea
l
-
tim
e
d
ata
tr
an
s
m
is
s
io
n
.
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
Tin
y
ma
ch
in
e
lea
r
n
in
g
w
ith
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l n
etw
o
r
k
fo
r
in
tellig
en
t
…
(
I
s
to
fa
)
411
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
was f
u
n
d
e
d
b
y
Natio
n
al
R
esear
ch
an
d
I
n
n
o
v
a
tio
n
Ag
en
cy
o
f
I
n
d
o
n
esia.
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
t
r
ib
u
to
r
R
o
les
T
a
x
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
i
d
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
I
s
to
f
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
Gin
a
Ku
s
u
m
a
✓
✓
✓
✓
✓
✓
Firliy
an
i
R
ah
m
atia
Nin
g
s
ih
✓
✓
✓
✓
✓
✓
J
o
k
o
T
r
iy
a
n
to
✓
✓
✓
I
Pu
tu
Su
s
ila
✓
✓
✓
✓
✓
✓
Atan
g
Su
s
ila
✓
✓
✓
✓
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
T
h
e
au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
ter
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
Au
th
o
r
s
s
tate
f
o
r
d
ata
av
a
ilab
ilit
y
th
at
s
u
p
p
o
r
ts
th
e
f
i
n
d
i
n
g
s
o
f
th
is
s
tu
d
y
a
r
e
av
ailab
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
]
I
.
H
.
S
a
r
k
e
r
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
:
A
l
g
o
r
i
t
h
ms,
r
e
a
l
-
w
o
r
l
d
a
p
p
l
i
c
a
t
i
o
n
s a
n
d
r
e
sea
r
c
h
d
i
r
e
c
t
i
o
n
s,”
S
N
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
,
n
o
.
3
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
2
9
7
9
-
0
2
1
-
0
0
5
9
2
-
x.
[
2
]
S
.
M
.
G
a
l
i
b
,
P
.
K
.
B
h
o
w
m
i
k
,
A
.
V
.
A
v
a
c
h
a
t
,
a
n
d
H
.
K
.
L
e
e
,
“
A
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
f
o
r
a
u
t
o
ma
t
e
d
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
r
a
d
i
o
i
s
o
t
o
p
e
s
u
s
i
n
g
N
a
I
g
a
mm
a
-
r
a
y
s
p
e
c
t
r
a
,
”
N
u
c
l
e
a
r
E
n
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
5
3
,
n
o
.
1
2
,
p
p
.
4
0
7
2
–
4
0
7
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
t
.
2
0
2
1
.
0
6
.
0
2
0
.
[
3
]
M
.
S
o
o
r
i
,
B
.
A
r
e
z
o
o
,
a
n
d
R
.
D
a
st
r
e
s,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
i
n
a
d
v
a
n
c
e
d
r
o
b
o
t
i
c
s,
a
r
e
v
i
e
w
,
”
C
o
g
n
i
t
i
v
e
R
o
b
o
t
i
c
s
,
v
o
l
.
3
,
n
o
.
M
a
r
c
h
,
p
p
.
5
4
–
7
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
g
r
.
2
0
2
3
.
0
4
.
0
0
1
.
[
4
]
P
.
P
.
R
a
y
,
“
A
r
e
v
i
e
w
o
n
Ti
n
y
M
L:
S
t
a
t
e
-
of
-
t
h
e
-
a
r
t
a
n
d
p
r
o
sp
e
c
t
s,”
J
o
u
rn
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
-
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
4
,
n
o
.
4
,
p
p
.
1
5
9
5
–
1
6
2
3
,
A
p
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
s
u
c
i
.
2
0
2
1
.
1
1
.
0
1
9
.
[
5
]
M
.
A
l
t
a
y
e
b
,
M
.
Ze
n
n
a
r
o
,
a
n
d
E.
P
i
e
t
r
o
s
e
m
o
l
i
,
“
T
i
n
y
M
L
G
a
mm
a
r
a
d
i
a
t
i
o
n
c
l
a
ss
i
f
i
e
r
,
”
N
u
c
l
e
a
r
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
5
5
,
n
o
.
2
,
p
p
.
4
4
3
–
4
5
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
t
.
2
0
2
2
.
0
9
.
0
3
2
.
[
6
]
A
.
E
l
h
a
n
a
sh
i
,
P
.
D
i
n
i
,
S
.
S
a
p
o
n
a
r
a
,
a
n
d
Q
.
Zh
e
n
g
,
“
A
d
v
a
n
c
e
me
n
t
s
i
n
T
i
n
y
M
L:
a
p
p
l
i
c
a
t
i
o
n
s,
l
i
mi
t
a
t
i
o
n
s,
a
n
d
i
m
p
a
c
t
o
n
I
o
T
d
e
v
i
c
e
s,
”
E
l
e
c
t
r
o
n
i
c
s (S
w
i
t
zer
l
a
n
d
)
,
v
o
l
.
1
3
,
n
o
.
1
7
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s1
3
1
7
3
5
6
2
.
[
7
]
M
.
d
e
P
r
a
d
o
e
t
a
l
.
,
“
R
o
b
u
s
t
i
f
y
i
n
g
t
h
e
d
e
p
l
o
y
m
e
n
t
o
f
t
i
n
y
M
L
mo
d
e
l
s
f
o
r
a
u
t
o
n
o
mo
u
s
mi
n
i
-
v
e
h
i
c
l
e
s,
”
Pro
c
e
e
d
i
n
g
s
-
I
E
EE
I
n
t
e
r
n
a
t
i
o
n
a
l
S
y
m
p
o
si
u
m
o
n
C
i
rc
u
i
t
s
a
n
d
S
y
s
t
e
m
s
,
v
o
l
.
2
0
2
1
-
M
a
y
,
p
p
.
1
–
1
6
,
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
1
0
4
1
3
3
9
.
[
8
]
M
.
B
h
a
m
a
r
e
,
P
.
V
.
K
u
l
k
a
r
n
i
,
R
.
R
a
n
e
,
S
.
B
o
b
d
e
,
a
n
d
R
.
P
a
t
a
n
k
a
r
,
“
T
i
n
y
M
L
a
p
p
l
i
c
a
t
i
o
n
s
a
n
d
u
s
e
c
a
s
e
s
f
o
r
h
e
a
l
t
h
c
a
r
e
,
”
T
i
n
y
ML
f
o
r
E
d
g
e
I
n
t
e
l
l
i
g
e
n
c
e
i
n
I
o
T
a
n
d
L
P
WA
N
N
e
t
w
o
r
k
s
,
v
o
l
.
1
5
,
n
o
.
1
,
p
p
.
3
3
1
–
3
5
3
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
B
9
7
8
-
0
-
44
-
322202
-
3
.
0
0
0
1
9
-
1.
[
9
]
W
.
D
i
n
g
,
J
.
Y
u
,
C
.
Z
h
e
n
g
,
P
.
F
u
,
a
n
d
Q
.
H
u
a
n
g
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
n
o
n
i
n
v
a
si
v
e
q
u
a
n
t
i
f
i
c
a
t
i
o
n
o
f
si
n
g
l
e
-
i
ma
g
i
n
g
sess
i
o
n
d
y
n
a
mi
c
P
ET
-
C
T
i
n
o
n
c
o
l
o
g
y
,
”
I
EEE
T
r
a
n
sa
c
t
i
o
n
s
o
n
Me
d
i
c
a
l
I
m
a
g
i
n
g
,
v
o
l
.
4
1
,
n
o
.
2
,
p
p
.
3
4
7
–
3
5
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
T
M
I
.
2
0
2
1
.
3
1
1
2
7
8
3
.
[
1
0
]
D
.
S
a
n
s
o
n
e
a
n
d
A
.
Z
h
u
,
“
U
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
o
c
r
e
a
t
e
a
n
e
a
r
l
y
w
a
r
n
i
n
g
s
y
st
e
m
f
o
r
w
e
l
f
a
r
e
r
e
c
i
p
i
e
n
t
s,”
O
x
f
o
rd
Bu
l
l
e
t
i
n
o
f
Ec
o
n
o
m
i
c
s
a
n
d
S
t
a
t
i
st
i
c
s
,
v
o
l
.
8
5
,
n
o
.
5
,
p
p
.
9
5
9
–
9
9
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
1
1
/
o
b
e
s.1
2
5
5
0
.
[
1
1
]
D
.
L
i
a
n
g
e
t
a
l
.
,
“
R
a
p
i
d
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
m
b
a
s
e
d
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
n
a
l
s
o
f
N
u
c
l
e
a
r
En
e
r
g
y
,
v
o
l
.
1
3
3
.
E
l
se
v
i
e
r
L
t
d
,
p
p
.
4
8
3
–
4
9
0
,
N
o
v
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
n
u
c
e
n
e
.
2
0
1
9
.
0
5
.
0
5
1
.
[
1
2
]
Y
.
W
a
n
g
e
t
a
l
.
,
“
Ex
p
l
a
i
n
a
b
l
e
r
a
d
i
o
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
m
b
a
s
e
d
o
n
t
h
e
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
c
l
a
ss
a
c
t
i
v
a
t
i
o
n
map
p
i
n
g
,
”
N
u
c
l
e
a
r
En
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
5
4
,
n
o
.
1
2
,
p
p
.
4
6
8
4
–
4
6
9
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
t
.
2
0
2
2
.
0
8
.
0
1
1
.
[
1
3
]
S
.
Q
i
e
t
a
l
.
,
“
R
a
d
i
o
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
m
e
t
h
o
d
f
o
r
N
a
I
l
o
w
-
c
o
u
n
t
g
a
mm
a
-
r
a
y
s
p
e
c
t
r
a
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
N
u
c
l
e
a
r
En
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
5
4
,
n
o
.
1
,
p
p
.
2
6
9
–
2
7
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
t
.
2
0
2
1
.
0
7
.
0
2
5
.
[
1
4
]
G
.
D
a
n
i
e
l
,
F
.
C
e
r
a
u
d
o
,
O
.
Li
m
o
u
si
n
,
D
.
M
a
i
e
r
,
a
n
d
A
.
M
e
u
r
i
s,
“
A
u
t
o
mat
i
c
a
n
d
r
e
a
l
-
t
i
me
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
r
a
d
i
o
n
u
c
l
i
d
e
s
i
n
G
a
mm
a
-
R
a
y
S
p
e
c
t
r
a
:
a
n
e
w
me
t
h
o
d
b
a
s
e
d
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
t
r
a
i
n
e
d
w
i
t
h
s
y
n
t
h
e
t
i
c
d
a
t
a
s
e
t
,
”
I
E
EE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
N
u
c
l
e
a
r
S
c
i
e
n
c
e
,
v
o
l
.
6
7
,
n
o
.
4
,
p
p
.
6
4
4
–
6
5
3
,
A
p
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
T
N
S
.
2
0
2
0
.
2
9
6
9
7
0
3
.
[
1
5
]
H
.
L.
Li
u
,
H
.
B
.
Ji
,
J.
M
.
Z
h
a
n
g
,
C
.
L
.
Z
h
a
n
g
,
J
.
L
u
,
a
n
d
X
.
H
.
F
e
n
g
,
“
A
n
o
v
e
l
a
p
p
r
o
a
c
h
f
o
r
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
f
r
o
m
a
g
a
mm
a
-
r
a
y
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.
1
6
,
No
.
1
,
Feb
r
u
ar
y
20
2
6
:
4
0
4
-
413
412
e
n
e
r
g
y
s
p
e
c
t
r
u
m b
a
s
e
d
o
n
i
m
a
g
e
d
e
s
c
r
i
p
t
o
r
t
r
a
n
sf
e
r
r
i
n
g
f
o
r
r
a
d
i
o
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
N
u
c
l
e
a
r
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
i
q
u
e
s
,
v
o
l
.
3
3
,
n
o
.
1
2
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
1
3
6
5
-
0
2
2
-
0
1
1
5
0
-
7.
[
1
6
]
S
.
X
.
Ze
n
g
e
t
a
l
.
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
l
o
w
-
c
o
u
n
t
,
l
o
w
-
r
e
s
o
l
u
t
i
o
n
g
a
mm
a
sp
e
c
t
r
a
l
r
a
d
i
o
n
u
c
l
i
d
e
u
si
n
g
2
D
C
N
N
-
B
i
LST
M
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
N
u
c
l
e
a
r E
n
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
5
7
,
n
o
.
1
2
,
D
e
c
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
t
.
2
0
2
5
.
1
0
3
8
5
4
.
[
1
7
]
C
.
L
i
e
t
a
l
.
,
“
A
n
e
w
r
a
d
i
o
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
m
e
t
h
o
d
f
o
r
l
o
w
-
c
o
u
n
t
e
n
e
r
g
y
sp
e
c
t
r
a
w
i
t
h
m
u
l
t
i
p
l
e
r
a
d
i
o
n
u
c
l
i
d
e
s,
”
A
p
p
l
i
e
d
Ra
d
i
a
t
i
o
n
a
n
d
I
so
t
o
p
e
s
,
v
o
l
.
1
8
5
,
J
u
l
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
p
r
a
d
i
s
o
.
2
0
2
2
.
1
1
0
2
1
9
.
[
1
8
]
G
.
K
u
s
u
ma,
R
.
M
.
S
a
r
y
a
d
i
,
S
.
K
.
W
i
j
a
y
a
,
S
.
S
o
e
k
i
r
n
o
,
P
.
P
r
a
j
i
t
n
o
,
a
n
d
I
.
P
.
S
u
si
l
a
,
“
R
a
d
i
o
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
a
l
y
s
i
s
u
si
ng
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
G
EA
N
T
4
si
mu
l
a
t
i
o
n
,
”
AI
P
C
o
n
f
e
r
e
n
c
e
Pr
o
c
e
e
d
i
n
g
s
,
v
o
l
.
2
3
8
1
,
n
o
.
N
o
v
e
m
b
e
r
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
6
3
/
5
.
0
0
6
7
5
9
3
.
[
1
9
]
H
.
C
.
L
e
e
e
t
a
l
.
,
“
R
a
d
i
o
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
b
a
s
e
d
o
n
e
n
e
r
g
y
-
w
e
i
g
h
t
e
d
a
l
g
o
r
i
t
h
m
a
n
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
l
i
e
d
t
o
a
m
u
l
t
i
-
a
r
r
a
y
p
l
a
st
i
c
s
c
i
n
t
i
l
l
a
t
o
r
,
”
N
u
c
l
e
a
r
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
n
o
.
M
a
y
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
t
.
2
0
2
3
.
0
7
.
0
0
5
.
[
2
0
]
C
.
B
o
b
i
n
,
O
.
B
i
c
h
l
e
r
,
V
.
Lo
u
r
e
n
ç
o
,
C
.
T
h
i
a
m,
a
n
d
M
.
Th
é
v
e
n
i
n
,
“
R
e
a
l
-
t
i
me
r
a
d
i
o
n
u
c
l
i
d
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
i
n
γ
-
e
mi
t
t
e
r
m
i
x
t
u
r
e
s
b
a
s
e
d
o
n
s
p
i
k
i
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
A
p
p
l
i
e
d
R
a
d
i
a
t
i
o
n
a
n
d
I
s
o
t
o
p
e
s
,
v
o
l
.
1
0
9
,
p
p
.
4
0
5
–
4
0
9
,
M
a
r
.
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
p
r
a
d
i
s
o
.
2
0
1
5
.
1
2
.
0
2
9
.
[
2
1
]
M
.
K
a
mu
d
a
a
n
d
C
.
J.
S
u
l
l
i
v
a
n
,
“
A
n
a
u
t
o
m
a
t
e
d
i
s
o
t
o
p
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
d
q
u
a
n
t
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
i
s
o
t
o
p
e
mi
x
t
u
r
e
s
i
n
l
o
w
-
r
e
so
l
u
t
i
o
n
g
a
m
ma
-
r
a
y
s
p
e
c
t
r
a
,
”
Ra
d
i
a
t
i
o
n
P
h
y
s
i
c
s
a
n
d
C
h
e
m
i
st
r
y
,
v
o
l
.
1
5
5
,
p
p
.
2
8
1
–
2
8
6
,
F
e
b
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
a
d
p
h
y
sc
h
e
m.
2
0
1
8
.
0
6
.
0
1
7
.
[
2
2
]
M
.
K
a
m
u
d
a
,
J.
S
t
i
n
n
e
t
t
,
a
n
d
C
.
J
.
S
u
l
l
i
v
a
n
,
“
A
u
t
o
m
a
t
e
d
i
so
t
o
p
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
m
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s,”
I
EE
E
T
ra
n
s
a
c
t
i
o
n
s
o
n
N
u
c
l
e
a
r
S
c
i
e
n
c
e
,
v
o
l
.
6
4
,
n
o
.
7
,
p
p
.
1
8
5
8
–
1
8
6
4
,
J
u
l
.
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
TN
S
.
2
0
1
7
.
2
6
9
3
1
5
2
.
[
2
3
]
A
.
S
h
a
r
ma
,
E.
V
a
n
s,
D
.
S
h
i
g
e
mi
z
u
,
K
.
A
.
B
o
r
o
e
v
i
c
h
,
a
n
d
T.
Ts
u
n
o
d
a
,
“
D
e
e
p
I
n
si
g
h
t
:
A
met
h
o
d
o
l
o
g
y
t
o
t
r
a
n
s
f
o
r
m
a
n
o
n
-
i
ma
g
e
d
a
t
a
t
o
a
n
i
ma
g
e
f
o
r
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
a
r
c
h
i
t
e
c
t
u
r
e
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
rt
s
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
1
–
7
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
1
9
-
4
7
7
6
5
-
6.
[
2
4
]
R
.
S
.
S
a
e
e
d
a
n
d
B
.
K
.
O
.
C
h
a
b
o
r
A
l
w
a
w
i
,
“
A
b
i
n
a
r
y
c
l
a
ssi
f
i
c
a
t
i
o
n
m
o
d
e
l
o
f
C
O
V
I
D
-
1
9
b
a
se
d
o
n
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
Bu
l
l
e
t
i
n
o
f
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
3
,
p
p
.
1
4
1
3
–
1
4
1
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
2
i
3
.
4
8
3
2
.
[
2
5
]
F
.
Te
m
p
o
l
a
,
R
.
W
a
r
d
o
y
o
,
A
.
M
u
sd
h
o
l
i
f
a
h
,
R
.
R
o
si
h
a
n
,
a
n
d
L
.
S
u
mary
a
n
t
i
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
o
f
c
l
o
v
e
t
y
p
e
s
u
si
n
g
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
a
l
g
o
r
i
t
h
m
w
i
t
h
o
p
t
i
m
i
z
i
n
g
h
y
p
e
r
p
a
r
a
mt
e
r
s,”
B
u
l
l
e
t
i
n
o
f
E
l
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
4
4
4
–
4
5
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
3
i
1
.
5
5
3
3
.
[
2
6
]
D
.
L.
D
u
t
t
a
a
n
d
S
.
B
h
a
r
a
l
i
,
“
Ti
n
y
M
L
m
e
e
t
s
I
o
T:
a
c
o
mp
r
e
h
e
n
s
i
v
e
s
u
r
v
e
y
,
”
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
,
v
o
l
.
1
6
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
o
t
.
2
0
2
1
.
1
0
0
4
6
1
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Is
to
fa
is
a
re
se
a
rc
h
e
r
a
t
t
h
e
Re
se
a
rc
h
Ce
n
ter
f
o
r
N
u
c
lea
r
Be
a
m
An
a
ly
sis
Tec
h
n
o
l
o
g
y
,
Re
se
a
rc
h
Org
a
n
iza
ti
o
n
fo
r
Nu
c
lea
r
En
e
rg
y
,
Na
ti
o
n
a
l
Re
se
a
rc
h
a
n
d
In
n
o
v
a
ti
o
n
Ag
e
n
c
y
,
I
n
d
o
n
e
sia
.
He
o
b
tain
e
d
h
is
b
a
c
h
e
lo
r’s
d
e
g
re
e
i
n
n
u
c
lea
r
e
n
g
i
n
e
e
rin
g
fro
m
G
a
d
ja
h
M
a
d
a
Un
i
v
e
rsity
,
In
d
o
n
e
sia
,
i
n
2
0
0
2
.
Afte
r
th
a
t,
h
e
c
o
n
ti
n
u
e
d
h
is
stu
d
y
i
n
th
e
m
a
g
ister
sc
ien
c
e
p
ro
g
ra
m
a
t
th
e
P
h
y
sic
s
De
p
a
rtme
n
t,
Un
i
v
e
rsitas
In
d
o
n
e
sia
,
g
ra
d
u
a
ti
n
g
i
n
2
0
2
4
.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
e
lec
tro
n
ics
,
n
u
c
lea
r
in
stru
m
e
n
tati
o
n
,
e
m
b
e
d
d
e
d
sy
ste
m
s,
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
isto
0
0
1
@b
r
in
.
g
o
.
i
d
.
G
in
a
K
u
sum
a
is
a
r
e
se
a
r
c
h
e
r
a
t
th
e
Re
se
a
r
c
h
Ce
n
ter
fo
r
Nu
c
lea
r
Be
a
m
An
a
ly
si
s
Tec
h
n
o
l
o
g
y
,
Re
se
a
rc
h
Org
a
n
iza
ti
o
n
fo
r
Nu
c
lea
r
En
e
r
g
y
,
Na
ti
o
n
a
l
Re
se
a
rc
h
a
n
d
I
n
n
o
v
a
ti
o
n
Ag
e
n
c
y
,
In
d
o
n
e
sia
.
He
wa
s
b
o
rn
in
Bo
g
o
r,
Wes
t
Ja
v
a
.
He
o
b
tain
e
d
h
is
b
a
c
h
e
lo
r’s
d
e
g
re
e
in
n
u
c
lea
r
tec
h
n
o
p
h
y
sic
s
fro
m
t
h
e
P
o
l
y
tec
h
n
ic
I
n
stit
u
te
o
f
N
u
c
lea
r
Tec
h
n
o
l
o
g
y
–
BATAN
Yo
g
y
a
k
a
rta
in
2
0
1
3
a
n
d
t
h
e
n
c
o
n
ti
n
u
e
d
h
is
st
u
d
y
in
th
e
m
a
g
is
ter
sc
ien
c
e
p
ro
g
ra
m
a
t
th
e
P
h
y
sic
s De
p
a
rtme
n
t,
Un
i
v
e
rsitas
In
d
o
n
e
sia
,
g
ra
d
u
a
ti
n
g
in
2
0
2
1
.
His res
e
a
r
c
h
field
s a
re
n
u
c
lea
r
in
stru
m
e
n
tati
o
n
,
e
lec
tro
n
ics
,
e
m
b
e
d
d
e
d
s
y
ste
m
s,
d
e
sk
t
o
p
p
r
o
g
ra
m
m
in
g
,
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
g
in
a
0
0
4
@
b
rin
.
g
o
.
i
d
.
Fi
r
li
y
a
n
i
Ra
h
m
a
ti
a
Nin
g
si
h
is
a
re
se
a
rc
h
e
r
a
t
t
h
e
Re
se
a
rc
h
C
e
n
ter
fo
r
Nu
c
lea
r
Be
a
m
An
a
ly
sis
Tec
h
n
o
l
o
g
y
.
S
h
e
wa
s
b
o
rn
i
n
S
u
ra
b
a
y
a
,
Eas
t
Ja
v
a
.
S
h
e
g
ra
d
u
a
ted
fro
m
th
e
b
a
c
h
e
lo
r’s p
r
o
g
ra
m
in
n
u
c
lea
r
e
n
g
in
e
e
rin
g
a
t
Un
i
v
e
rsitas
G
a
d
jah
M
a
d
a
in
2
0
1
5
.
Afte
r
t
h
a
t,
s
h
e
c
o
n
ti
n
u
e
d
h
e
r
st
u
d
y
in
th
e
m
a
g
is
ter
sc
ien
c
e
p
ro
g
ra
m
a
t
t
h
e
P
h
y
si
c
s
De
p
a
rtme
n
t,
Un
i
v
e
rsitas
In
d
o
n
e
sia
,
g
ra
d
u
a
ti
n
g
in
2
0
2
3
.
S
h
e
is
c
u
rre
n
tl
y
a
c
ti
v
e
a
s
a
n
a
ss
istan
t
re
se
a
rc
h
e
r
a
t
th
e
Na
ti
o
n
a
l
Re
se
a
rc
h
a
n
d
In
n
o
v
a
ti
o
n
Ag
e
n
c
y
o
f
t
h
e
Re
p
u
b
li
c
o
f
I
n
d
o
n
e
sia
.
He
r
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
p
h
o
to
n
b
e
a
m
tec
h
n
o
l
o
g
y
,
ra
d
iati
o
n
m
e
a
su
re
m
e
n
t,
a
rti
ficia
l
in
telli
g
e
n
c
e
,
a
n
d
ra
d
iati
o
n
ima
g
in
g
tec
h
n
iq
u
e
s.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
firl
0
0
1
@b
ri
n
.
g
o
.
id
.
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
Tin
y
ma
ch
in
e
lea
r
n
in
g
w
ith
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l n
etw
o
r
k
fo
r
in
tellig
en
t
…
(
I
s
to
fa
)
413
J
o
k
o
Tr
iy
a
n
t
o
is
a
re
se
a
rc
h
e
r
a
t
th
e
Re
se
a
rc
h
Ce
n
ter
fo
r
N
u
c
lea
r
Be
a
m
An
a
ly
si
s
Tec
h
n
o
l
o
g
y
.
He
wa
s
b
o
r
n
i
n
W
o
n
o
g
iri
,
Ce
n
tral
Ja
v
a
.
He
g
ra
d
u
a
ted
fr
o
m
t
h
e
u
n
d
e
rg
ra
d
u
a
te
p
ro
g
ra
m
in
n
u
c
lea
r
e
n
g
i
n
e
e
rin
g
a
t
Un
i
v
e
rsitas
G
a
d
jah
M
a
d
a
in
1
9
9
5
.
Afte
r
t
h
a
t,
h
e
c
o
n
ti
n
u
e
d
h
is
stu
d
ies
i
n
t
h
e
m
a
ste
r
o
f
e
n
g
i
n
e
e
rin
g
p
ro
g
ra
m
a
t
th
e
De
p
a
rtme
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
ri
n
g
,
Un
iv
e
rsity
o
f
In
d
o
n
e
sia
,
g
ra
d
u
a
t
in
g
in
2
0
1
2
.
He
is
c
u
rre
n
tl
y
a
c
t
iv
e
a
s
a
re
se
a
rc
h
e
r
a
t
th
e
Na
ti
o
n
a
l
Re
se
a
rc
h
a
n
d
In
n
o
v
a
ti
o
n
A
g
e
n
c
y
o
f
t
h
e
Re
p
u
b
li
c
o
f
I
n
d
o
n
e
sia
.
His
re
se
a
rc
h
i
n
tere
sts
in
c
lu
d
e
n
u
c
lea
r
in
stru
m
e
n
tati
o
n
a
n
d
c
o
n
tr
o
l
sy
ste
m
s,
ra
d
iatio
n
m
e
a
su
re
m
e
n
t,
a
n
d
e
m
b
e
d
d
e
d
sy
ste
m
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
jo
k
o
0
1
9
@
b
rin
.
g
o
.
id
.
I
Pu
tu
S
u
sil
a
is
a
re
se
a
rc
h
e
r
a
t
th
e
Re
se
a
rc
h
Ce
n
ter
f
o
r
N
u
c
lea
r
Be
a
m
An
a
ly
sis
Tec
h
n
o
l
o
g
y
,
Re
se
a
rc
h
Org
a
n
iza
ti
o
n
fo
r
Nu
c
lea
r
En
e
rg
y
,
Na
ti
o
n
a
l
Re
se
a
rc
h
a
n
d
In
n
o
v
a
ti
o
n
Ag
e
n
c
y
,
I
n
d
o
n
e
sia
.
He
re
c
e
iv
e
d
h
is
d
o
c
to
ra
l
d
e
g
re
e
fr
o
m
T
o
h
o
k
u
Un
iv
e
rsit
y
,
Ja
p
a
n
,
in
2
0
0
9
.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
so
ftwa
re
d
e
v
e
lo
p
m
e
n
t
,
ima
g
e
p
ro
c
e
ss
in
g
,
a
n
d
e
m
b
e
d
d
e
d
sy
ste
m
s
in
b
io
m
e
d
ica
l
a
n
d
h
e
a
lt
h
re
se
a
rc
h
.
He
is
p
a
rti
c
u
larly
in
tere
ste
d
in
d
e
v
e
lo
p
in
g
h
e
a
lt
h
c
a
re
e
q
u
ip
m
e
n
t,
in
c
lu
d
i
n
g
h
a
rd
wa
re
,
so
ftwa
re
,
a
n
d
o
p
e
n
-
so
u
rc
e
a
p
p
li
c
a
ti
o
n
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ip
u
t
0
0
1
@b
ri
n
.
g
o
.
i
d
.
Ata
n
g
S
u
sila
is
a
re
se
a
rc
h
e
r
a
t
th
e
Re
se
a
r
c
h
Ce
n
ter
fo
r
Nu
c
lea
r
Be
a
m
An
a
ly
sis
Tec
h
n
o
l
o
g
y
,
Re
se
a
rc
h
Org
a
n
iza
ti
o
n
fo
r
Nu
c
lea
r
En
e
rg
y
,
Na
ti
o
n
a
l
Re
se
a
rc
h
a
n
d
In
n
o
v
a
ti
o
n
Ag
e
n
c
y
,
I
n
d
o
n
e
sia
.
He
o
b
tain
e
d
h
is
b
a
c
h
e
lo
r’s
d
e
g
re
e
i
n
n
u
c
lea
r
e
n
g
i
n
e
e
rin
g
fro
m
G
a
d
ja
h
M
a
d
a
Un
iv
e
rsity
,
In
d
o
n
e
sia
,
in
1
9
8
8
.
Afte
r
th
a
t,
h
e
c
o
n
ti
n
u
e
d
h
is
stu
d
y
i
n
th
e
m
a
ste
r’s
p
ro
g
ra
m
in
sc
ien
c
e
s
a
n
d
e
n
g
i
n
e
e
rin
g
a
t
t
h
e
Un
iv
e
rsit
y
o
f
Ts
u
k
u
b
a
,
Ja
p
a
n
,
g
ra
d
u
a
ti
n
g
in
2
0
0
1
.
His
re
se
a
r
c
h
in
tere
sts
in
c
lu
d
e
e
lec
tro
n
ics
,
n
u
c
lea
r
in
stru
m
e
n
tatio
n
,
e
m
b
e
d
d
e
d
sy
ste
m
s,
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
tan
0
0
1
@b
r
i
n
.
g
o
.
i
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.