I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
14
,
No
.
6
,
Dec
em
b
er
2
0
2
5
,
p
p
.
4
8
9
1
~
4
9
0
1
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
6
.
p
p
4
8
9
1
-
4
9
0
1
4891
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Cla
ss
ificatio
n
sy
st
em o
f
ba
na
na
t
y
p
es a
nd
ripen
ess
le
v
els ba
sed
o
n conv
o
lutiona
l neural net
wo
rk
L
ucia
J
a
m
bo
la
,
Arsy
a
d Ra
ma
dh
a
n Da
rlis
,
Win
di M
a
la
ha
,
Dwi
Ary
a
nta
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
,
I
n
st
i
t
u
t
Te
k
n
o
l
o
g
i
N
a
s
i
o
n
a
l
B
a
n
d
u
n
g
,
B
a
n
d
u
n
g
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
23
,
2
0
2
4
R
ev
is
ed
Sep
07
,
2
0
2
5
Acc
ep
ted
No
v
08
,
2
0
2
5
Re
c
e
n
tl
y
,
th
e
a
v
a
il
a
b
il
it
y
o
f
b
a
n
a
n
a
s
in
su
p
e
rm
a
rk
e
ts
h
a
s
b
e
e
n
re
lativ
e
ly
a
b
u
n
d
a
n
t.
Ho
we
v
e
r,
m
o
st
b
u
y
e
rs
e
x
p
e
rien
c
e
p
ro
b
lem
s
c
a
teg
o
rizi
n
g
t
h
e
t
y
p
e
a
n
d
lev
e
l
o
f
r
ip
e
n
e
ss
o
f
b
a
n
a
n
a
s,
so
th
e
lev
e
l
o
f
p
u
rc
h
a
se
s
o
f
th
is
fru
it
d
e
c
re
a
se
s.
Th
is
stu
d
y
imp
lem
e
n
t
s
a
re
a
l
-
ti
m
e
sy
ste
m
th
a
t
c
a
n
a
u
to
m
a
ti
c
a
ll
y
c
las
sify
b
a
n
a
n
a
s
in
th
e
d
u
a
l
c
las
sifica
ti
o
n
b
a
se
d
o
n
ty
p
e
a
n
d
le
v
e
l
o
f
rip
e
n
e
ss
.
so
th
a
t
b
u
y
e
rs
c
a
n
c
h
o
o
se
th
e
m
b
a
se
d
o
n
th
e
ir
n
e
e
d
s.
I
n
th
is
s
tu
d
y
,
th
e
p
ro
p
o
se
d
sy
ste
m
c
o
u
l
d
c
las
sify
b
a
n
a
n
a
s
u
sin
g
a
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
(CNN
),
wh
e
re
th
e
s
y
ste
m
wa
s
i
m
p
lem
e
n
ted
in
re
a
l
-
ti
m
e
u
si
n
g
th
e
h
a
rd
wa
re
o
f
t
h
e
Je
tso
n
Na
n
o
a
s
a
p
ro
c
e
ss
in
g
u
n
i
t
a
n
d
a
c
a
m
e
ra
sy
ste
m
a
s
a
s
e
n
so
r.
Th
e
m
e
th
o
d
o
lo
g
y
a
d
o
p
ted
i
n
t
h
is
re
se
a
rc
h
in
v
o
l
v
e
s
imp
lem
e
n
t
in
g
CNN
a
rc
h
it
e
c
tu
re
s,
i.
e
.
,
Re
sN
e
t
-
1
8
a
n
d
Re
sN
e
t
-
5
0
,
u
n
d
e
r
v
a
rio
u
s
c
o
n
d
i
t
io
n
s.
Th
e
train
in
g
p
h
a
se
c
o
m
p
rise
s
6
0
e
p
o
c
h
s,
wh
il
e
tes
ti
n
g
c
o
n
si
d
e
rs
il
l
u
m
in
a
ti
o
n
p
a
ra
m
e
ters
fro
m
LE
D
li
g
h
ts
wit
h
p
o
we
r
o
f
6
wa
tt
s,
1
2
wa
tt
s,
a
n
d
2
2
wa
tt
s
u
n
d
e
r
d
istan
c
e
s
ra
n
g
in
g
fr
o
m
1
0
t
o
1
0
0
c
m
.
Th
e
re
su
l
ts
sh
o
w
th
a
t
t
h
e
sy
ste
m
c
o
u
ld
c
las
sify
t
h
e
ty
p
e
a
n
d
lev
e
l
o
f
rip
e
n
e
ss
o
f
b
a
n
a
n
a
s
in
re
a
l
-
ti
m
e
with
a
n
a
c
c
u
ra
c
y
o
f
9
3
%
wh
ic
h
is ac
h
iev
e
d
u
sin
g
t
h
e
2
2
-
wa
tt
p
o
we
r
fo
r
a
ll
ty
p
e
s a
n
d
rip
e
n
e
ss
lev
e
ls.
K
ey
w
o
r
d
s
:
B
an
an
a
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
Du
al
class
if
icatio
n
J
etso
n
Nan
o
R
ea
l
-
tim
e
im
p
lem
en
tatio
n
R
esNet
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
:
Ar
s
y
ad
R
am
ad
h
an
Dar
lis
Dep
ar
tm
en
t o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
,
I
n
s
titu
t T
ek
n
o
lo
g
i N
a
s
io
n
al
B
an
d
u
n
g
,
B
an
d
u
n
g
,
I
n
d
o
n
esia
Ph
h
.
Mu
s
to
f
a
No
.
2
3
Stre
et,
B
an
d
u
n
g
C
ity
,
W
est J
av
a
-
4
0
1
2
4
,
I
n
d
o
n
esia
E
m
ail:
ar
s
y
ad
@
iten
as.a
c.
id
1.
I
NT
RO
D
UCT
I
O
N
Fru
its
co
n
s
titu
te
a
s
ig
n
if
ican
t h
o
r
ticu
ltu
r
al
co
m
m
o
d
ity
,
p
lay
i
n
g
an
im
p
o
r
tan
t r
o
le
in
I
n
d
o
n
e
s
ia.
T
h
ese
f
r
u
its
p
lay
an
im
p
o
r
ta
n
t
r
o
le
in
b
o
d
y
m
eta
b
o
lis
m
b
ec
au
s
e
o
f
th
eir
ab
u
n
d
a
n
t
v
itam
in
an
d
m
in
er
al
co
n
ten
t.
T
h
is
asp
ec
t
s
er
v
es
as
a
b
en
ch
m
ar
k
f
o
r
citizen
s
in
cu
ltiv
atin
g
an
d
p
r
o
p
ag
atin
g
d
iv
e
r
s
e
f
r
u
it
-
y
ield
i
n
g
p
lan
t
s
p
ec
ies
[
1
]
.
Am
o
n
g
t
h
ese,
b
an
a
n
as
r
ep
r
es
en
t
a
v
a
r
ietal
f
r
u
it
with
in
th
e
ag
r
icu
ltu
r
al
s
p
h
er
e
t
h
at
b
est
o
ws
v
ar
io
u
s
h
ea
lth
b
en
ef
its
d
u
e
to
th
eir
r
ich
n
u
tr
it
io
n
al
co
m
p
o
s
itio
n
[
2
]
.
On
e
o
f
th
e
co
m
m
o
n
is
s
u
es
th
at
th
is
r
esear
ch
co
n
s
id
er
s
is
th
at
m
an
y
b
a
n
an
a
s
to
ck
s
ar
e
s
o
ld
in
m
in
im
ar
k
ets
an
d
s
u
p
e
r
m
ar
k
et
s
.
Yet,
co
n
s
u
m
e
r
s
r
em
ain
u
n
ab
le
to
d
is
tin
g
u
is
h
th
e
d
e
g
r
ee
o
f
r
ip
en
ess
o
f
th
e
b
an
an
as
o
n
th
eir
o
wn
,
p
o
ten
tially
d
ec
r
ea
s
in
g
th
eir
p
u
r
ch
asi
n
g
in
ter
est.
I
n
th
is
r
esear
ch
,
a
r
ea
l
-
tim
e
s
y
s
tem
-
b
ased
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
e
two
r
k
(
C
NN)
m
eth
o
d
o
lo
g
y
is
u
tili
ze
d
to
class
if
y
d
u
al
class
i
f
icatio
n
o
f
b
an
a
n
a,
i.e
.
,
ty
p
e
an
d
its
lev
el
o
f
r
ip
en
ess
.
T
h
is
ap
p
r
o
ac
h
en
tails
d
ir
e
ctin
g
b
an
an
a
s
p
ec
im
en
s
to
war
d
a
ca
m
er
a
to
attain
a
h
eig
h
ten
e
d
lev
el
o
f
ac
cu
r
ac
y
to
d
eter
m
in
e
an
d
class
if
y
th
e
b
an
an
a
t
y
p
e
a
n
d
r
ip
en
ess
lev
el
[
3
]
.
T
h
is
ap
p
r
o
ac
h
m
ay
allo
w
th
e
co
m
p
u
ter
to
r
e
co
g
n
ize
th
e
im
ag
e
s
im
ilar
to
t
h
e
h
u
m
an
le
v
el
in
ca
teg
o
r
izin
g
b
an
an
a
ty
p
e
an
d
r
ip
en
ess
lev
els
[
4
]
.
T
h
e
cu
r
r
en
t
tech
n
o
lo
g
ical
ad
v
an
ce
m
en
ts
h
a
v
e
f
ac
ilit
ated
h
u
m
an
ca
p
ab
ilit
y
in
d
eter
m
in
i
n
g
b
a
n
an
a
r
ip
en
ess
b
ased
o
n
co
lo
r
t
h
r
o
u
g
h
th
e
u
tili
za
tio
n
o
f
C
NN,
as
ex
p
lo
r
ed
in
th
e
i
n
v
esti
g
atio
n
b
y
Z
h
an
g
et
a
l.
[
5
]
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
9
1
-
4
9
0
1
4892
Sri
et
a
l.
[
6
]
.
An
o
th
er
s
tu
d
y
b
y
Sar
an
y
a
et
a
l.
[
7
]
was
co
n
d
u
ct
ed
.
Ho
wev
er
,
th
e
s
tu
d
ies
co
u
ld
n
o
t
d
eter
m
in
e
t
h
e
ty
p
es o
f
b
a
n
an
as a
n
d
d
id
n
o
t su
p
p
o
r
t r
ea
l
-
tim
e
im
p
lem
e
n
tatio
n
.
I
n
f
r
u
it
ty
p
e
class
if
icatio
n
,
p
r
ev
io
u
s
s
tu
d
ies
h
av
e
u
s
ed
im
a
g
e
-
b
ased
o
b
jects,
s
im
ilar
to
th
e
r
esear
ch
co
n
d
u
cte
d
b
y
Ash
r
af
et
a
l
.
[
8
]
.
Su
b
s
eq
u
en
tly
,
th
is
is
s
u
p
p
o
r
ted
b
y
Path
ak
an
d
Ma
k
w
an
a
[
9
]
ab
o
u
t
f
r
u
its
class
if
icatio
n
u
s
in
g
a
C
NN
.
A
f
u
r
th
er
co
n
tr
i
b
u
tio
n
o
f
th
is
r
esear
ch
is
th
e
class
if
icatio
n
o
f
b
an
an
a
ty
p
es u
s
in
g
a
C
NN,
wh
ich
was
co
n
d
u
cted
b
y
Sh
u
p
r
ajh
aa
et
a
l.
[
1
0
]
.
Ho
wev
er
,
n
o
n
e
o
f
th
ese
s
tu
d
ies
ca
n
d
e
ter
m
in
e
th
e
b
a
n
an
a
r
ip
en
ess
in
r
ea
l tim
e.
All
o
f
th
is
r
esear
ch
o
n
ly
co
n
s
id
er
s
o
n
e
ty
p
e
an
d
r
ip
e
n
ess
o
f
b
a
n
an
a,
an
d
m
o
s
t
o
f
th
em
wer
e
n
o
t
im
p
lem
en
ted
in
r
ea
l
tim
e.
R
esear
ch
co
n
d
u
cted
b
y
B
ald
o
v
in
o
et
a
l.
[
1
1
]
,
wh
ic
h
f
o
c
u
s
es
o
n
r
ea
l
-
tim
e
r
ec
o
g
n
itio
n
an
d
class
if
icatio
n
o
f
b
an
a
n
a
o
b
jects
v
ia
web
ca
m
u
s
in
g
f
ast
alg
o
r
ith
m
s
s
u
ch
as
C
NN
an
d
YOL
O,
s
tates
th
at
th
i
s
m
o
d
el
is
a
b
le
to
d
etec
t
an
d
ca
teg
o
r
ize
o
b
jects
with
a
n
ac
c
u
r
ac
y
o
f
ar
o
u
n
d
9
0
%
ac
r
o
s
s
all
class
es
o
n
a
s
tat
ic
im
ag
e
test
d
ataset.
Fo
r
r
ea
l
-
tim
e
m
o
d
el
d
etec
tio
n
,
s
am
p
les
wer
e
p
lace
d
o
n
a
b
lu
e
b
ac
k
g
r
o
u
n
d
,
c
o
n
s
id
er
in
g
t
h
at
th
e
y
ello
w
an
d
g
r
ee
n
co
l
o
r
s
in
b
an
an
a
s
am
p
les
ar
e
co
m
p
le
m
en
tar
y
p
air
s
ac
co
r
d
in
g
to
th
e
R
GB
co
lo
r
wh
ee
l.
B
an
an
a
s
am
p
les
wer
e
test
ed
b
o
th
in
g
r
o
u
p
s
an
d
i
n
d
iv
id
u
ally
.
Ho
wev
er
,
th
is
m
o
d
el
o
n
ly
ac
h
iev
ed
8
0
%
ac
cu
r
ac
y
d
u
r
in
g
r
ea
l
-
tim
e
test
in
g
o
f
in
d
i
v
id
u
al
s
am
p
les.
Ob
ject
class
if
icat
io
n
in
im
ag
es
is
o
n
e
o
f
th
e
p
r
im
ar
y
ch
alle
n
g
es
with
in
co
m
p
u
ter
v
is
io
n
.
I
ts
p
r
im
ar
y
o
b
jectiv
e
is
to
em
p
o
wer
co
m
p
u
ter
s
to
em
u
late
h
u
m
an
ca
p
ab
ilit
ies
in
in
ter
p
r
etin
g
v
is
u
al
in
f
o
r
m
atio
n
.
On
e
s
u
cc
ess
f
u
l
ap
p
r
o
ac
h
h
as
b
ee
n
th
r
o
u
g
h
th
e
ap
p
licatio
n
o
f
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
ANNs)
in
s
p
ir
ed
b
y
h
u
m
an
n
eu
r
al
n
etwo
r
k
s
,
wh
ic
h
h
av
e
f
u
r
th
er
e
v
o
lv
e
d
in
to
th
e
co
n
ce
p
t
o
f
d
ee
p
lear
n
in
g
(
D
L
)
[
1
2
]
,
[
1
3
]
.
I
n
th
e
s
tu
d
y
co
n
d
u
cted
b
y
Said
et
a
l.
[
1
4
]
,
u
tili
zin
g
th
e
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e
(
SVM)
m
eth
o
d
f
o
r
class
if
y
in
g
b
an
an
a
ty
p
es r
esu
lts
in
p
r
o
m
is
in
g
o
u
tco
m
es
[
1
4
]
,
[
1
5
]
.
H
o
wev
er
,
m
o
s
t o
f
th
e
r
esear
ch
es we
r
e
n
o
t i
m
p
lem
en
ted
in
r
ea
l
tim
e
an
d
h
as lo
w
ac
cu
r
ac
y
.
DL
r
ep
r
esen
ts
a
b
r
a
n
ch
o
f
m
a
ch
in
e
lear
n
in
g
(
ML
)
g
r
o
u
n
d
e
d
in
ANNs
[
1
6
]
.
T
h
e
f
o
u
n
d
atio
n
o
f
DL
is
p
r
im
ar
ily
r
o
o
te
d
in
ANNs
wi
t
h
in
th
e
d
o
m
ain
o
f
ML
[
1
7
]
.
Var
io
u
s
ty
p
es
o
f
n
eu
r
al
n
etwo
r
k
s
ex
is
t
with
in
th
e
r
ea
lm
o
f
DL
,
en
co
m
p
ass
in
g
ANNs,
C
NNs,
an
d
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
s
(
R
NNs)
[
1
8
]
.
A
s
u
b
s
et
o
f
DL
m
eth
o
d
o
l
o
g
ies,
th
e
C
NN,
d
er
i
v
es
f
r
o
m
th
e
p
r
in
cip
les
o
f
th
e
m
u
ltil
ay
er
p
e
r
ce
p
tr
o
n
(
ML
P)
;
h
o
wev
er
,
C
NN
is
s
p
ec
ially
en
g
in
ee
r
e
d
to
p
r
o
ce
s
s
two
-
d
im
en
s
io
n
al
d
ata
f
o
r
m
at
s
s
u
ch
as
au
d
i
o
an
d
im
ag
es
[
1
9
]
.
C
NN
is
h
ar
n
ess
ed
f
o
r
s
u
p
er
v
is
ed
lear
n
in
g
-
b
ased
class
if
icatio
n
o
f
lab
eled
d
ata
[
2
0
]
.
On
e
c
o
m
m
o
n
ly
em
p
lo
y
ed
n
eu
r
al
n
etwo
r
k
ar
ch
itectu
r
e
f
o
r
im
ag
e
d
ata
is
th
e
C
NN.
Pre
s
en
tly
,
th
e
m
o
s
t
wid
ely
ad
o
p
te
d
ap
p
r
o
ac
h
in
DL
in
v
o
lv
es
C
NNs.
At
a
ce
r
tain
d
ep
th
,
C
NN
f
alls
with
in
th
e
class
if
icatio
n
o
f
d
ee
p
n
eu
r
al
n
etwo
r
k
s
,
f
r
eq
u
en
t
ly
ap
p
lied
in
im
ag
e
d
ata
p
r
o
ce
s
s
in
g
[
2
1
]
,
[
2
2
]
.
T
h
is
s
tu
d
y
im
p
lem
e
n
ted
a
r
e
al
-
tim
e
s
y
s
tem
th
at
ca
n
au
to
m
atica
lly
class
if
y
b
an
an
as
in
th
e
d
u
al
class
if
icatio
n
b
ased
o
n
ty
p
e
an
d
lev
el
o
f
r
ip
e
n
ess
.
s
o
th
at
b
u
y
er
s
ca
n
ch
o
o
s
e
th
em
b
ased
o
n
th
eir
n
ee
d
s
.
I
n
th
is
s
tu
d
y
,
th
e
p
r
o
p
o
s
ed
s
y
s
tem
co
u
ld
class
if
y
b
an
a
n
as
u
s
in
g
a
C
NN,
wh
er
e
th
e
s
y
s
tem
was
im
p
lem
en
ted
u
s
in
g
th
e
h
ar
d
war
e
o
f
th
e
J
etso
n
Nan
o
a
s
a
p
r
o
ce
s
s
in
g
u
n
it
an
d
a
ca
m
e
r
a
s
y
s
tem
as
a
s
en
s
o
r
.
T
h
e
m
eth
o
d
o
lo
g
y
ad
o
p
ted
in
th
is
r
esear
ch
in
v
o
lv
es
im
p
lem
en
tin
g
C
NN
ar
c
h
itectu
r
es
u
n
d
er
v
ar
i
o
u
s
co
n
d
itio
n
s
.
T
h
e
R
esNet
ar
ch
itectu
r
e
co
m
p
r
is
es
s
ev
er
al
v
ar
iatio
n
s
with
v
ar
y
in
g
lay
e
r
d
ep
th
s
,
s
u
ch
as
R
esNet
-
1
8
,
R
esNet
-
3
4
,
R
esNet
-
5
0
,
R
esNet
-
1
0
1
,
an
d
R
esNet
-
1
5
2
.
I
n
th
is
s
tu
d
y
,
th
e
Py
th
o
n
p
r
o
g
r
am
m
in
g
lan
g
u
ag
e
was
em
p
l
o
y
ed
to
im
p
lem
e
n
t
C
NNs
wi
th
R
esNet
-
1
8
f
o
r
t
y
p
e
b
a
n
an
a
class
if
icatio
n
an
d
R
esNet
-
5
0
ar
ch
itectu
r
es
f
o
r
r
ip
e
n
ess
b
an
an
a
class
if
icatio
n
.
Desp
ite
th
e
en
h
an
ce
m
en
ts
th
e
R
esNet
m
o
d
e
l
b
r
in
g
s
to
th
e
in
p
u
t,
it
also
d
e
m
o
n
s
tr
ates
s
u
p
er
io
r
s
tab
ilit
y
in
h
an
d
lin
g
g
r
ad
ien
t
v
ar
iatio
n
s
d
u
r
in
g
th
e
tr
ain
in
g
p
r
o
ce
s
s
[
2
3
]
.
Fo
r
in
s
tan
ce
,
R
esNet
co
n
s
is
t
s
o
f
1
6
co
n
v
o
l
u
tio
n
al
lay
er
s
,
two
d
o
w
n
s
am
p
lin
g
lay
er
s
,
a
n
d
m
u
ltip
le
f
u
lly
co
n
n
ec
ted
(
FC
)
lay
er
s
[
2
4
]
.
2.
M
E
T
H
O
D
Fig
u
r
e
1
e
x
p
lain
s
th
e
o
v
e
r
all
wo
r
k
f
lo
w
o
f
a
b
an
a
n
a
r
ip
e
n
ess
class
if
icatio
n
s
y
s
tem
,
u
tili
zi
n
g
s
ev
er
al
d
ev
ices,
i.e
.
,
a
lap
to
p
,
a
ca
m
e
r
a,
an
d
a
J
etso
n
Nan
o
.
T
h
e
f
i
r
s
t
s
tep
i
s
co
llectin
g
a
d
ataset
with
th
r
ee
b
an
an
a
r
ip
en
ess
ca
p
tu
r
ed
th
r
o
u
g
h
a
ca
m
er
a,
s
er
v
in
g
as
in
p
u
t
f
o
r
th
e
J
etso
n
Nan
o
2
GB
.
T
h
is
p
r
o
ce
s
s
was
ex
ec
u
ted
at
d
is
tin
ct
tim
e
in
ter
v
als.
T
h
is
u
tili
ze
d
a
5
V
DC
p
o
wer
s
u
p
p
ly
to
ca
ter
t
o
th
e
J
etso
n
Nan
o
2
GB
'
s
en
er
g
y
r
eq
u
ir
em
e
n
ts
,
with
an
ad
d
itio
n
al
in
ter
n
et
co
n
n
ec
tio
n
f
ac
ilit
ated
v
ia
lo
ca
l
a
r
ea
n
etwo
r
k
(
L
AN
)
ca
b
le.
Su
b
s
eq
u
en
tly
,
a
USB
ca
b
le
e
s
t
ab
lis
h
ed
a
lin
k
b
etwe
en
th
e
J
et
s
o
n
Nan
o
2
GB
an
d
a
lap
to
p
,
wh
er
e
th
e
co
m
p
u
ter
ass
u
m
ed
th
e
r
o
le
o
f
m
an
ag
in
g
th
e
J
etso
n
Nan
o
2
GB
o
p
er
atio
n
s
.
On
ce
all
co
m
p
o
n
en
ts
wer
e
in
ter
co
n
n
ec
ted
,
th
e
ca
m
er
a
ac
tiv
atio
n
en
s
u
ed
.
H
av
in
g
b
ee
n
i
n
teg
r
ated
with
t
h
e
J
etso
n
Nan
o
2
GB
,
th
e
ca
m
er
a
ass
u
m
ed
th
e
r
esp
o
n
s
ib
ilit
y
o
f
r
ea
l
-
tim
e
class
if
icatio
n
o
f
b
an
a
n
a
ty
p
e
an
d
r
ip
en
ess
lev
el.
T
h
e
J
etso
n
Nan
o
2
GB
in
itiate
d
th
e
d
ataset
co
llectio
n
p
r
o
ce
s
s
,
f
o
llo
wed
b
y
a
tr
ain
in
g
p
h
as
e
to
r
ef
in
e
th
e
s
y
s
tem
.
C
o
n
s
eq
u
e
n
tly
,
a
m
o
d
el
was
d
er
iv
ed
.
T
h
is
m
o
d
el
th
en
u
n
d
er
wen
t
test
in
g
to
a
s
ce
r
tain
th
e
ac
c
u
r
ac
y
v
alu
es
f
o
r
b
a
n
an
a
t
y
p
e
a
n
d
r
i
p
en
ess
lev
el
class
if
icatio
n
s
.
T
h
e
o
b
tai
n
ed
r
esu
lts
wer
e
s
u
b
s
eq
u
en
tly
s
h
o
wca
s
ed
o
n
a
m
o
n
ito
r
.
Fig
u
r
e
2
s
h
o
ws th
e
b
an
a
n
a
class
if
icatio
n
s
y
s
te
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fica
tio
n
s
ystem
o
f B
a
n
a
n
a
Typ
es a
n
d
r
ip
en
ess
leve
ls
b
a
s
ed
o
n
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l
…
(
Lu
cia
Ja
mb
o
l
a
)
4893
Fig
u
r
e
1
.
Sy
s
tem
b
lo
c
k
d
ia
g
r
a
m
S
t
a
r
t
B
a
n
a
n
a
O
b
j
e
c
t
P
r
o
c
e
s
s
o
f
C
l
a
s
s
i
f
y
i
n
g
t
h
e
T
y
p
e
s
a
n
d
L
e
v
e
l
s
o
f
F
r
u
i
t
M
a
t
u
r
i
t
y
T
r
a
i
n
i
n
g
T
e
s
t
i
n
g
P
r
o
c
e
s
s
C
N
N
M
o
d
e
l
f
o
r
T
y
p
e
s
a
n
d
L
e
v
e
l
s
o
f
F
r
u
i
t
M
a
t
u
r
i
t
y
P
r
o
c
e
s
s
o
f
C
l
a
s
s
i
f
y
i
n
g
t
h
e
T
y
p
e
s
a
n
d
L
e
v
e
l
s
o
f
F
r
u
i
t
M
a
t
u
r
i
t
y
T
r
a
i
n
i
n
g
E
n
d
Fig
u
r
e
2
.
Sy
s
tem
f
lo
wc
h
ar
t
T
h
e
p
r
o
g
r
am
s
tar
ts
with
ca
p
tu
r
in
g
b
an
a
n
a
s
am
p
les
th
r
o
u
g
h
a
co
n
n
ec
ted
ca
m
er
a,
w
h
ich
is
s
u
b
s
eq
u
en
tly
em
p
lo
y
ed
f
o
r
tr
ain
in
g
in
th
e
class
if
icatio
n
o
f
b
an
an
a
ty
p
es
an
d
r
ip
en
ess
lev
els.
B
ef
o
r
e
class
if
icatio
n
,
th
e
ac
q
u
ir
ed
o
b
ject
o
u
tc
o
m
es
ar
e
o
r
g
an
ized
in
to
d
atas
ets
to
f
ac
ilit
ate
th
e
tr
ain
in
g
p
r
o
ce
s
s
.
Up
o
n
co
m
p
letio
n
o
f
tr
ain
in
g
,
a
m
o
d
el
is
d
er
iv
e
d
,
s
e
r
v
in
g
as
th
e
f
o
u
n
d
atio
n
f
o
r
s
u
b
s
eq
u
en
t
class
if
icatio
n
o
f
b
a
n
an
a
ty
p
e
an
d
r
ip
e
n
ess
.
T
h
is
is
f
o
llo
wed
b
y
th
e
test
in
g
p
h
ase,
d
u
r
in
g
wh
ich
th
e
c
am
er
a
is
d
ir
ec
te
d
to
war
d
th
e
b
an
an
a
s
p
ec
im
en
s
,
y
ield
in
g
ac
cu
r
ac
y
v
alu
es
in
d
icativ
e
o
f
b
an
an
a
ty
p
e
an
d
r
ip
en
ess
lev
el.
T
h
is
r
e
s
ea
r
ch
is
co
n
d
u
cte
d
u
tili
zin
g
Go
o
g
le
C
o
lab
Pro
,
wh
ic
h
ca
n
h
elp
with
co
d
e
b
u
ild
in
g
,
t
r
ain
in
g
,
a
n
d
test
in
g
th
e
d
ata,
o
r
g
en
e
r
atin
g
th
e
v
is
u
aliza
tio
n
r
esu
lt
o
f
th
e
tr
a
in
in
g
p
r
o
ce
s
s
.
T
h
e
Py
th
o
n
p
r
o
g
r
am
m
in
g
lan
g
u
ag
e
is
u
s
ed
to
ac
h
ie
v
e
s
y
s
tem
d
ev
elo
p
m
e
n
t.
T
h
e
in
itial
s
tag
e
in
v
o
lv
es
ac
q
u
ir
in
g
d
ata
f
o
r
b
an
a
n
a
ty
p
e
r
e
co
g
n
itio
n
to
v
er
if
y
th
e
s
y
s
tem
's
ab
ilit
y
to
class
if
y
b
an
an
a
ty
p
es
ac
cu
r
atel
y
.
T
h
is
will
d
eter
m
in
e
th
e
n
ex
t
m
atu
r
ity
le
v
el
class
if
icatio
n
p
r
o
ce
s
s
.
T
h
e
d
ataset
in
th
is
s
tu
d
y
is
ca
teg
o
r
ized
in
t
o
two
g
r
o
u
p
s
:
d
ata
f
o
r
tr
ain
in
g
a
n
d
d
ata
f
o
r
test
in
g
.
T
h
e
d
ata
o
u
tco
m
es
en
co
m
p
ass
th
e
p
r
esen
tatio
n
o
f
b
a
n
an
a
t
y
p
es
an
d
ac
c
u
r
ac
y
m
etr
ics.
T
h
e
s
p
ec
if
ic
b
an
an
a
ty
p
es
e
m
p
lo
y
ed
in
t
h
is
s
tu
d
y
a
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
9
1
-
4
9
0
1
4894
en
u
m
er
ated
in
T
ab
le
1
.
I
n
th
is
s
tu
d
y
,
R
es
Net
-
1
8
was
u
s
ed
to
class
if
y
b
an
an
a
ty
p
es
d
u
e
to
its
r
elativ
ely
lo
w
co
m
p
lex
ity
,
wh
ile
R
esNet
-
5
0
was
u
s
ed
to
class
if
y
r
ip
en
ess
lev
els
d
u
e
to
th
e
n
ee
d
t
o
ac
h
i
ev
e
h
ig
h
ac
c
u
r
ac
y
ac
r
o
s
s
s
ev
er
al
r
ip
e
n
ess
class
e
s
.
I
n
r
ea
l
-
tim
e
im
p
lem
en
tatio
n
,
th
e
R
esNet
ar
ch
itectu
r
e
m
o
d
el
is
co
n
s
id
er
ed
lig
h
tweig
h
t e
n
o
u
g
h
t
o
b
e
im
p
lem
en
ted
o
n
h
ar
d
war
e.
T
ab
le
1
.
B
an
an
a
d
atas
et
To
t
a
l
d
a
t
a
se
t
1
,
0
0
0
B
a
n
a
n
a
t
y
p
e
1.
K
e
p
o
k
2.
M
u
l
i
B
a
n
a
n
a
r
i
p
e
n
e
ss
l
e
v
e
l
1.
U
n
r
i
p
e
2.
S
e
mi
-
r
i
p
e
3.
R
i
p
e
4.
P
a
r
t
i
a
l
l
y
r
i
p
e
5.
O
v
e
r
r
i
p
e
n
3.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
3
.
1
.
Da
t
a
s
et
co
lle
ct
i
o
n
B
u
ild
in
g
a
DL
s
y
s
tem
also
r
e
q
u
ir
es
a
h
u
g
e
am
o
u
n
t
o
f
d
ata,
wh
ich
is
u
s
ed
f
o
r
tr
ain
in
g
an
d
test
in
g
.
T
r
ain
in
g
d
ata
is
a
b
u
n
c
h
o
f
d
ata
co
llectio
n
co
n
s
is
tin
g
o
f
la
b
els
an
d
class
es,
wh
ich
m
ac
h
in
es
u
s
e
to
im
p
r
o
v
e
id
en
tify
in
g
ch
ar
ac
ter
is
tics
o
f
a
n
im
ag
e,
i
n
ten
d
in
g
to
estab
lis
h
a
p
att
er
n
o
r
d
ata
m
o
d
el.
On
th
e
o
th
er
s
id
e,
test
in
g
d
ata
r
ep
r
esen
ts
a
d
ataset
s
im
il
ar
ly
eq
u
ip
p
e
d
with
lab
els
o
r
class
es,
s
er
v
in
g
to
ass
es
s
th
e
e
x
ten
t
to
wh
ich
th
e
d
ev
elo
p
e
d
p
atter
n
o
r
m
o
d
el
ca
n
ac
cu
r
ately
class
if
y
th
e
ev
alu
atio
n
d
ata
[
2
5
]
.
Fig
u
r
e
3
s
h
o
w
s
a
d
ataset
ex
am
p
le
th
at
co
n
s
is
ts
o
f
2
b
a
n
an
a
ty
p
e
s
,
i.e
.
,
k
e
p
o
k
in
Fig
u
r
e
3
(
a)
a
n
d
m
u
li
in
Fig
u
r
e
3
(
b
)
.
T
h
en
,
Fig
u
r
e
4
s
h
o
ws
th
e
d
ataset
f
o
r
th
e
k
ep
o
k
b
a
n
an
a
is
s
h
o
wn
in
Fig
u
r
e
4
(
a)
wit
h
s
ev
er
al
v
ar
y
in
g
d
e
g
r
ee
s
o
f
r
ip
en
ess
lev
els,
an
d
lik
ewise
f
o
r
m
u
li
b
an
a
n
a
as
s
h
o
wn
in
Fig
u
r
e
4
(
b
)
with
s
ev
er
al
p
r
ed
eter
m
in
ed
r
ip
en
ess
s
tag
es
with
in
th
is
s
tu
d
y
.
(
a)
(
b
)
Fig
u
r
e
3
.
T
y
p
e
-
b
ased
b
an
a
n
a
d
ataset
:
(
a)
k
ep
o
k
an
d
(
b
)
m
u
l
i
(
a)
(
b
)
Fig
u
r
e
4
.
R
ip
en
ess
lev
el
-
b
ased
b
an
an
a
d
ataset
:
(
a)
k
ep
o
k
an
d
(
b
)
m
u
li
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fica
tio
n
s
ystem
o
f B
a
n
a
n
a
Typ
es a
n
d
r
ip
en
ess
leve
ls
b
a
s
ed
o
n
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l
…
(
Lu
cia
Ja
mb
o
l
a
)
4895
E
ac
h
r
ip
en
ess
lev
el
co
m
p
r
is
es
1
0
0
im
ag
es,
co
n
tr
ib
u
tin
g
to
a
d
ataset
o
f
5
0
0
im
ag
es
f
o
r
k
ep
o
k
a
n
d
5
0
0
f
o
r
m
u
li.
C
o
n
s
eq
u
en
tly
,
t
h
e
co
m
b
in
e
d
d
ataset
am
ass
ed
f
o
r
b
o
th
ty
p
es
o
f
b
an
a
n
as
an
d
ea
ch
r
ip
en
ess
s
tag
e
to
tals
1
,
0
0
0
im
ag
es.
T
h
e
am
ass
ed
d
ataset
will b
e
ca
te
g
o
r
ized
in
to
f
iv
e
d
is
tin
ct
s
tag
es o
f
r
ip
en
ess
.
T
h
ese
s
tag
es
en
co
m
p
ass
u
n
r
ip
e,
c
h
ar
ac
ter
iz
ed
b
y
th
e
b
an
a
n
a
r
etain
in
g
a
g
r
ee
n
h
u
e;
p
a
r
tially
r
ip
e,
d
is
p
lay
in
g
a
tr
an
s
itio
n
f
r
o
m
g
r
ee
n
to
a
y
ello
wis
h
tin
t;
r
ip
e,
p
o
r
t
r
ay
in
g
a
u
n
if
o
r
m
ly
y
ello
w
co
lo
r
atio
n
;
p
ar
tia
lly
o
v
er
r
ip
e,
m
a
r
k
ed
b
y
b
r
o
wn
s
p
o
ts
o
n
th
e
b
an
an
a'
s
s
u
r
f
ac
e;
an
d
f
u
lly
o
v
er
r
ip
e,
wh
er
e
th
e
b
an
a
n
a
h
as tu
r
n
ed
p
r
ed
o
m
in
an
tly
b
r
o
wn
.
3
.
2
.
T
ra
ini
ng
pro
ce
s
s
T
r
ain
in
g
d
ata
c
o
m
p
r
is
es
th
e
d
ataset
u
tili
ze
d
to
in
s
tr
u
ct
th
e
co
n
s
tr
u
cted
m
o
d
el.
T
h
e
m
o
d
el'
s
o
u
tco
m
es
ar
e
tr
ain
ed
to
alig
n
with
th
e
a
n
ticip
ated
v
alu
es.
T
h
e
y
wer
e,
co
n
v
er
s
ely
,
test
in
g
d
ata
f
u
n
ctio
n
s
to
ev
alu
ate
t
h
e
m
o
d
el
th
at
h
a
d
u
n
d
er
g
o
n
e
th
e
tr
ain
in
g
p
h
ase
.
T
ab
le
2
s
h
o
ws th
e
m
o
d
el
ar
c
h
itectu
r
e
f
o
r
th
e
tr
ain
in
g
p
r
o
ce
s
s
.
Fig
u
r
e
5
d
elin
ea
tes
th
e
test
i
n
g
s
ce
n
ar
io
in
v
o
lv
i
n
g
R
esNet
-
1
8
f
o
r
b
a
n
an
a
-
ty
p
e
class
if
i
ca
tio
n
an
d
R
esNet
-
5
0
f
o
r
r
ip
en
ess
-
lev
el
c
lass
if
icatio
n
with
in
th
e
m
o
d
el
ar
ch
itectu
r
e
a
b
o
v
e.
T
h
is
s
ce
n
a
r
io
was
d
esig
n
e
d
t
o
co
m
p
ar
e
th
e
ac
cu
r
ac
y
lev
els
an
d
lo
s
s
o
u
tco
m
es
b
as
ed
o
n
a
d
ataset
s
ize
o
f
1
,
0
0
0
in
s
tan
ce
s
.
T
h
e
o
p
tim
al
s
y
s
tem
p
er
f
o
r
m
an
ce
was
ac
h
iev
e
d
u
s
i
n
g
a
p
a
r
am
eter
co
n
f
ig
u
r
atio
n
o
f
6
0
ep
o
c
h
s
.
An
ep
o
ch
d
e
n
o
t
es
a
s
tate
in
wh
ich
th
e
en
tire
d
ataset,
s
er
v
in
g
as
in
p
u
t
f
o
r
th
e
tr
ain
i
n
g
m
o
d
el,
co
m
p
letes
a
f
u
ll
tr
av
er
s
al
th
r
o
u
g
h
th
e
n
eu
r
al
n
etwo
r
k
with
in
a
s
in
g
le
c
y
cle
[
2
6
]
.
T
a
b
le
3
ex
p
lain
s
th
e
tr
ain
in
g
r
es
u
lt,
r
ev
ea
lin
g
th
e
h
ig
h
est
ac
h
i
ev
ed
ac
c
u
r
ac
y
v
alu
e
f
o
r
b
an
a
n
a
-
ty
p
e
class
if
icatio
n
at
0
.
9
9
9
8
,
with
a
co
r
r
esp
o
n
d
in
g
lo
s
s
v
alu
e
o
f
0
.
0
0
0
8
.
A
lo
wer
lo
s
s
v
alu
e
g
en
er
ally
in
d
icate
s
a
m
o
r
e
o
p
tim
al
ac
cu
r
ac
y
v
alu
e
f
o
r
th
e
o
b
tain
ed
m
o
d
el
[
2
7
]
,
a
f
ac
to
r
cr
u
cial
f
o
r
f
u
r
th
er
v
alid
atio
n
th
r
o
u
g
h
th
e
test
in
g
p
r
o
ce
s
s
.
T
ab
l
e
2
.
T
h
e
m
o
d
el
ar
ch
itectu
r
e
N
u
mb
e
r
o
f
d
a
t
a
se
t
s
1
,
0
0
0
i
ma
g
e
s
M
o
d
e
l
R
e
sN
e
t
-
1
8
a
n
d
R
e
sN
e
t
-
50
Ep
o
c
h
60
O
p
t
i
mi
z
e
r
A
d
a
m
B
a
t
c
h
8
Le
a
r
n
i
n
g
r
a
t
e
0
.
0
0
1
Fig
u
r
e
5
.
T
r
ain
in
g
g
r
ap
h
ic
r
es
u
lt
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
9
1
-
4
9
0
1
4896
T
ab
le
3
.
Acc
u
r
ac
y
an
d
lo
s
s
ac
h
iev
ed
f
r
o
m
tr
ain
i
n
g
p
r
o
ce
s
s
D
a
t
a
s
e
t
A
c
c
u
r
a
c
y
Lo
ss
B
a
n
a
n
a
t
y
p
e
0
.
9
9
9
8
0
.
0
0
0
8
K
e
p
o
k
r
i
p
e
n
e
ss
l
e
v
e
l
0
.
9
9
6
9
0
.
0
0
0
9
M
u
l
i
r
i
p
e
n
e
ss
l
e
v
e
l
0
.
9
9
6
9
0
.
0
0
3
9
3
.
3
.
T
esting
pro
ce
ss
T
h
e
test
in
g
p
h
ase
o
f
th
is
s
tu
d
y
em
p
l
o
y
ed
th
e
Py
T
o
r
ch
m
o
d
el
with
in
th
e
C
NN
m
eth
o
d
o
lo
g
y
.
T
h
is
in
v
o
lv
e
d
ass
ess
in
g
th
e
a
cc
u
r
ac
y
o
f
t
h
e
class
if
icatio
n
o
u
tco
m
es
f
o
r
b
o
th
b
an
an
a
ty
p
e
an
d
r
ip
en
ess
lev
el
.
T
o
ex
ec
u
te
th
e
test
in
g
p
r
o
ce
s
s
,
th
e
in
itial
s
tep
in
v
o
lv
ed
ac
tiv
atin
g
th
e
ca
m
er
a
to
f
ac
ilit
ate
r
ea
l
-
tim
e
o
p
er
atio
n
.
T
h
e
ca
m
er
a
was
th
en
d
ir
ec
ted
to
w
ar
d
s
th
e
b
an
an
a
o
b
ject
f
o
r
an
aly
s
is
.
Fo
r
test
in
g
p
u
r
p
o
s
es
,
d
ev
ices
s
u
ch
as
a
lap
to
p
,
a
ca
m
e
r
a,
an
d
a
J
etso
n
Nan
o
ar
e
u
s
ed
.
T
h
e
d
etails o
f
ea
ch
d
ev
ice
ar
e
elab
o
r
ate
d
in
T
ab
le
4
.
Fig
u
r
e
6
illu
s
tr
ates th
e
b
an
an
a
ty
p
e
class
if
icatio
n
p
r
o
ce
s
s
in
v
o
lv
in
g
id
e
n
tify
in
g
a
n
d
ca
p
t
u
r
i
n
g
b
an
a
n
a
o
b
jects.
Su
b
s
eq
u
en
tly
,
ac
cu
r
ac
y
an
d
p
r
e
d
ictio
n
s
ar
e
d
is
p
lay
e
d
o
n
th
e
m
o
n
ito
r
s
cr
ee
n
.
Du
r
in
g
th
e
test
in
g
p
h
ase,
th
e
au
th
o
r
co
n
d
u
cte
d
r
ea
l
-
tim
e
test
s
f
o
r
b
o
th
b
an
an
a
ty
p
e
a
n
d
r
ip
en
ess
lev
el
class
if
icatio
n
at
a
f
ix
ed
d
is
tan
ce
o
f
3
0
cm
.
T
h
ese
test
s
wer
e
ex
ec
u
ted
u
n
d
er
v
ar
y
in
g
L
E
D
lig
h
ti
n
g
co
n
d
itio
n
s
,
with
wattag
es
o
f
6
watts
(
5
2
0
lm
)
,
1
2
watts (
1
,
2
0
0
lm
)
,
an
d
2
2
w
atts
(
2
,
4
0
0
lm
)
,
with
in
an
en
cl
o
s
ed
in
d
o
o
r
s
p
ac
e
m
ea
s
u
r
i
n
g
3
×
2
.
5
m
eter
s
.
T
ab
le
4
.
T
esti
n
g
d
ev
ice
s
p
ec
if
icatio
n
s
D
e
v
i
c
e
S
p
e
c
i
f
i
c
a
t
i
o
n
La
p
t
o
p
P
r
o
c
e
ss
o
r
:
I
n
t
e
l
C
o
r
e
i
5
-
8
2
5
0
U
,
I
n
t
e
l
U
H
D
G
r
a
p
h
i
c
s
6
2
0
R
A
M
:
8
GB
O
p
e
r
a
t
i
o
n
S
y
s
t
e
m
:
W
i
n
d
o
w
s
1
0
N
V
I
D
I
A
Jet
so
n
N
a
n
o
2
GB
P
r
o
c
e
ss
o
r
:
Q
u
a
d
-
c
o
r
e
A
R
M
®
A
5
7
G
H
z
:
2
G
B
6
4
-
LPD
D
R
4
,
2
5
.
6
G
B
/s
S
D
C
a
r
d
S
u
p
p
o
r
t
:
M
i
c
r
o
S
D
(
C
a
r
d
n
o
t
i
n
c
l
u
d
e
d
)
Lo
g
i
t
e
c
h
C
2
7
0
c
a
m
e
r
a
R
e
s
o
l
u
t
i
o
n
R
a
t
i
o
:
7
2
0
p
/
3
0
f
p
s
M
e
g
a
p
i
x
e
l
C
a
m
e
r
a
:
0
.
9
F
o
c
u
s T
y
p
e
s
:
f
i
x
e
d
f
o
c
u
s
Fig
u
r
e
6
.
R
ea
l
-
tim
e
p
r
o
p
o
s
ed
s
y
s
tem
test
in
g
Fig
u
r
e
7
s
h
o
ws
th
e
class
if
icatio
n
r
esu
lt
;
wh
er
e
k
ep
o
k
b
an
a
n
a
ty
p
es
in
Fig
u
r
e
7
(
a)
an
d
m
u
li
b
an
an
a
ty
p
e
in
Fig
u
r
e
7
(
b
)
,
wh
ich
ar
e
ca
teg
o
r
ized
in
to
th
r
ee
illu
m
in
atio
n
lev
els:
6
watts,
1
2
w
atts
,
an
d
2
2
watts.
T
h
e
s
y
s
tem
d
e
m
o
n
s
tr
ates
ef
f
e
ctiv
e
class
if
icatio
n
o
f
k
ep
o
k
ty
p
e,
ac
h
ie
v
in
g
a
r
em
ar
k
a
b
le
ac
cu
r
ac
y
r
ate
o
f
1
0
0
%
u
n
d
er
th
e
2
2
-
watt
ca
teg
o
r
y
.
N
o
tab
ly
,
th
e
k
e
p
o
k
ty
p
e
with
in
th
e
6
-
watt
ca
teg
o
r
y
ex
h
ib
its
a
s
lig
h
t
v
ar
ian
ce
o
f
1%
-
2
%
at
1
0
an
d
1
5
s
ec
o
n
d
s
,
attr
ib
u
ted
to
th
e
lo
wer
illu
m
in
atio
n
p
o
wer
o
f
6
watts.
C
o
n
v
er
s
ely
,
less
er
ac
cu
r
ac
y
is
o
b
s
er
v
ed
in
t
h
e
6
-
watt
ca
te
g
o
r
y
at
th
e
2
0
-
s
ec
o
n
d
m
ar
k
f
o
r
th
e
m
u
lti
-
ty
p
e.
T
h
e
2
2
-
watt
ca
teg
o
r
y
,
h
o
wev
e
r
,
s
u
cc
ess
f
u
lly
ac
h
iev
es
ac
cu
r
at
e
class
if
icatio
n
,
attain
in
g
a
p
er
f
ec
t
ac
cu
r
ac
y
s
co
r
e
o
f
1
0
0
%.
I
n
s
u
m
m
atio
n
,
it
ca
n
b
e
d
e
d
u
ce
d
th
at
th
e
l
o
wer
ac
cu
r
ac
y
v
alu
es,
as
r
e
f
lecte
d
b
y
th
e
o
v
er
all
av
er
a
g
e
ac
cu
r
ac
y
,
alig
n
with
t
h
e
6
-
watt
illu
m
in
atio
n
ca
teg
o
r
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fica
tio
n
s
ystem
o
f B
a
n
a
n
a
Typ
es a
n
d
r
ip
en
ess
leve
ls
b
a
s
ed
o
n
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l
…
(
Lu
cia
Ja
mb
o
l
a
)
4897
(
a)
(
b
)
Fig
u
r
e
7
.
B
an
an
a
ty
p
e
class
if
icatio
n
r
esu
lt u
n
d
er
v
a
r
io
u
s
la
m
p
p
o
wer
:
(
a)
k
e
p
o
k
an
d
(
b
)
m
u
li
Fig
u
r
e
8
illu
s
tr
ates
th
e
ac
cu
r
ac
y
v
alu
es
o
f
k
e
p
o
k
at
th
e
u
n
r
ip
e
r
ip
en
ess
lev
el
u
n
d
er
th
e
6
-
watt
illu
m
in
atio
n
ca
teg
o
r
y
,
d
is
p
lay
in
g
a
4
%
d
if
f
er
e
n
ce
f
r
o
m
o
th
er
ca
teg
o
r
ies,
with
a
n
a
v
er
ag
e
ac
cu
r
ac
y
o
f
9
4
%
.
Fo
r
th
e
p
ar
tially
r
ip
e
m
u
li,
th
e
6
-
watt
ca
teg
o
r
y
f
ails
to
ac
h
iev
e
ef
f
ec
tiv
e
r
ip
en
ess
d
etec
tio
n
,
y
ield
in
g
a
n
av
e
r
ag
e
ac
cu
r
ac
y
o
f
1
2
%.
Ad
d
itio
n
all
y
,
ac
cu
r
ac
y
r
em
ain
s
b
elo
w
2
0
%
ac
r
o
s
s
d
is
tan
ce
s
s
p
an
n
in
g
1
0
cm
to
1
0
0
cm
d
u
e
to
in
ad
eq
u
ate
lig
h
tin
g
co
n
d
itio
n
s
.
I
n
co
n
t
r
ast,
th
e
1
2
-
watt
an
d
2
2
-
watt
ca
teg
o
r
ies
y
ield
a
v
er
a
g
e
ac
cu
r
ac
y
v
alu
es
o
f
9
5
%
an
d
1
0
0
%,
r
esp
ec
tiv
ely
.
T
h
e
s
y
s
tem
e
x
ce
ls
in
class
if
icatio
n
,
p
ar
ticu
lar
l
y
u
n
d
er
t
h
e
1
2
-
watt
an
d
2
2
-
watt
ca
teg
o
r
ies,
ac
h
iev
in
g
n
o
ta
b
le
ac
cu
r
ac
y
r
ates.
T
h
en
,
Fig
u
r
e
9
s
h
o
ws
th
e
class
if
icatio
n
o
f
s
em
i
-
r
ip
e
k
ep
o
k
r
ip
e
n
ess
lev
els
u
n
d
er
th
e
6
-
watt
illu
m
in
atio
n
ca
teg
o
r
y
,
in
d
icatin
g
a
lack
o
f
ac
cu
r
ac
y
d
etec
ti
o
n
,
as
r
ef
lecte
d
b
y
an
av
er
ag
e
ac
cu
r
ac
y
o
f
2
5
%.
Fu
r
th
er
m
o
r
e
,
ac
cu
r
ac
y
le
v
els
at
v
ar
y
in
g
d
is
tan
ce
s
f
all
b
elo
w
2
5
%.
C
o
n
v
er
s
ely
,
with
in
th
e
1
2
-
watt
an
d
2
2
-
watt
ca
teg
o
r
ies,
th
e
a
v
er
ag
e
ac
cu
r
a
cy
v
alu
es
s
o
ar
to
9
7
%
an
d
9
3
%,
r
esp
ec
tiv
ely
,
s
ig
n
if
y
in
g
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
.
T
h
is
v
ar
ian
ce
ca
n
b
e
attr
ib
u
te
d
to
th
e
in
s
u
f
f
icien
cy
o
f
illu
m
i
n
atio
n
with
in
th
e
6
-
watt
ca
teg
o
r
y
.
C
o
n
v
e
r
s
ely
,
th
e
p
ar
tially
r
ip
e
m
u
li
ex
h
i
b
its
r
o
b
u
s
t
r
esu
lts
,
b
o
asti
n
g
an
av
er
a
g
e
ac
cu
r
ac
y
o
f
9
8
%
u
n
d
er
th
e
6
-
watt
illu
m
in
atio
n
ca
teg
o
r
y
.
No
tab
l
y
,
th
is
s
y
s
tem
ex
ce
ls
in
clas
s
if
y
in
g
s
em
i
-
r
ip
e
m
u
li
s
p
ec
im
en
s
ac
r
o
s
s
th
e
6
-
watt,
1
2
-
watt,
an
d
22
-
watt
ca
teg
o
r
ie
s.
Fig
u
r
e
8
.
Un
r
ip
e
class
if
icatio
n
u
n
d
er
v
ar
io
u
s
lam
p
p
o
wer
Fig
u
r
e
9
.
Sem
i
-
r
i
p
e
class
if
icatio
n
r
esu
lt u
n
d
er
v
ar
io
u
s
lam
p
p
o
wer
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
5
10
15
20
25
30
A
c
c
u
r
a
c
y
(
%)
T
i
me
(
s)
6
W
a
t
t
1
2
W
a
t
t
2
2
W
a
t
t
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
5
10
15
20
25
30
A
cc
u
r
ac
y
(%)
T
i
me
(
s)
6
Wa
t
t
1
2
W
a
t
t
2
2
W
a
t
t
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
5
10
15
20
25
30
A
c
c
u
r
a
c
y
(
%)
T
i
me
(
s)
0%
20%
40%
60%
80%
100
%
10
20
30
40
50
60
70
80
90
100
A
c
c
ur
ac
y
(
%)
D
i
st
an
c
e
(
c
m
)
Ke
po
k
6
W
a
t
t
Ke
po
k
12
W
at
t
Ke
po
k
22
W
at
t
M
ul
i
6
W
at
t
M
ul
i
12
W
at
t
M
ul
i
22
W
at
t
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
5
10
15
20
25
30
A
c
c
u
r
a
c
y
(
%)
T
i
me
(
s)
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
10
20
30
40
50
60
70
80
90
1
0
0
A
c
c
u
r
a
c
y
(
%)
D
i
st
a
n
c
e
(
c
m)
K
e
p
o
k
6
W
a
t
t
K
e
p
o
k
1
2
W
a
t
t
K
e
p
o
k
2
2
W
a
t
t
M
u
l
i
6
W
a
t
t
M
u
l
i
1
2
W
a
t
t
M
u
l
i
2
2
W
a
t
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
9
1
-
4
9
0
1
4898
As
s
h
o
wn
in
Fig
u
r
e
1
0
,
th
e
2
2
-
watt
ca
teg
o
r
y
d
e
m
o
n
s
tr
ate
s
an
ex
ce
p
tio
n
al
ac
cu
r
ac
y
tr
en
d
wh
en
ex
am
in
ed
at
in
ter
v
als
o
f
5
to
3
0
s
ec
o
n
d
s
,
r
esu
ltin
g
i
n
a
r
em
ar
k
ab
ly
ac
c
u
r
ate
o
f
1
0
0
%
f
o
r
b
o
th
k
ep
o
k
an
d
m
u
li
.
W
ith
in
th
e
1
2
-
watt
ca
teg
o
r
y
,
d
u
r
in
g
th
e
r
ip
e
r
ip
en
ess
lev
el,
th
e
av
er
ag
e
ac
cu
r
ac
y
v
alu
es
f
o
r
k
ep
o
k
an
d
m
u
li
s
tan
d
at
9
1
%
an
d
8
4
%,
r
esp
ec
t
iv
ely
.
C
o
n
v
er
s
ely
,
u
n
d
e
r
th
e
6
-
watt
ca
teg
o
r
y
,
ac
c
u
r
ac
y
d
im
i
n
is
h
es
to
an
av
er
ag
e
o
f
4
1
%
f
o
r
k
ep
o
k
an
d
6
2
%
f
o
r
m
u
li.
T
h
is
d
is
cr
ep
an
cy
ca
n
b
e
attr
ib
u
ted
to
th
e
im
p
ac
t
o
f
lig
h
ti
n
g
co
n
d
itio
n
s
,
af
f
ec
tin
g
th
e
ac
cu
r
ac
y
o
f
th
e
class
if
icatio
n
p
r
o
ce
s
s
.
Fu
r
th
er
m
o
r
e,
th
e
class
if
icatio
n
ac
cu
r
ac
y
d
ec
r
ea
s
es
as
th
e
d
is
tan
ce
b
etwe
en
th
e
class
if
ied
b
an
a
n
a
o
b
jects
an
d
th
e
ca
m
er
a
in
cr
ea
s
es.
No
tab
ly
,
th
is
s
y
s
tem
ef
f
ec
tiv
ely
a
n
d
ac
cu
r
ately
class
if
i
es
r
ip
e
s
tates
f
o
r
b
o
th
k
ep
o
k
an
d
m
u
li
u
n
d
e
r
th
e
1
2
-
watt
an
d
2
2
-
watt
illu
m
in
atio
n
ca
te
g
o
r
ies.
Fig
u
r
e
1
1
p
r
o
v
id
es
in
s
ig
h
t
in
to
th
e
g
r
ap
h
r
ep
r
esen
tin
g
th
e
p
ar
t
ially
r
ip
e
k
ep
o
k
an
d
m
u
li
ac
c
u
r
ac
y
lev
els
u
n
d
er
d
if
f
er
e
n
t
illu
m
in
atio
n
c
o
n
d
itio
n
s
.
W
ith
in
th
e
6
-
watt
ca
t
eg
o
r
y
,
t
h
e
av
er
a
g
e
ac
cu
r
ac
y
v
a
lu
es
f
o
r
k
ep
o
k
an
d
m
u
li
ar
e
6
2
%
an
d
5
2
%,
r
esp
ec
tiv
ely
.
I
n
t
h
e
1
2
-
watt
ca
teg
o
r
y
,
th
ese
v
alu
es
im
p
r
o
v
e
to
an
av
er
ag
e
o
f
8
4
%
f
o
r
k
ep
o
k
a
n
d
7
5
%
f
o
r
m
u
li.
T
h
e
2
2
-
watt
ca
teg
o
r
y
d
em
o
n
s
tr
ates
a
r
em
ar
k
ab
le
av
er
a
g
e
ac
cu
r
a
cy
o
f
9
9
%
f
o
r
b
o
t
h
ty
p
es
o
f
b
a
n
an
as.
No
tab
ly
,
v
ar
iatio
n
s
in
ac
cu
r
ac
y
ac
r
o
s
s
th
e
6
-
watt,
1
2
-
watt,
a
n
d
2
2
-
watt
c
ateg
o
r
ies
s
tem
f
r
o
m
d
if
f
er
en
ce
s
in
lig
h
tin
g
co
n
d
itio
n
s
.
Ad
d
itio
n
ally
,
th
e
p
ar
tially
r
ip
e
s
tate
o
f
b
an
a
n
as
ex
h
ib
it
s
s
u
b
tle
d
if
f
er
en
ce
s
in
ap
p
ea
r
a
n
ce
co
m
p
ar
ed
t
o
f
u
lly
r
ip
e
o
n
es,
ch
ar
ac
ter
ize
d
b
y
b
r
o
wn
s
p
o
ts
.
T
h
e
s
y
s
tem
ef
f
ec
tiv
ely
ac
h
iev
es
ac
cu
r
ate
cl
ass
if
icatio
n
,
p
ar
ticu
lar
ly
in
th
e
2
2
-
watt
ca
teg
o
r
y
,
wh
er
e
it
attain
s
a
n
o
tab
le
s
u
cc
ess
r
ate
o
f
9
9
%
in
d
elin
ea
tin
g
p
a
r
tially
r
ip
e
b
a
n
a
n
a
s
tates.
An
d
th
e
last
r
esu
lt,
Fig
u
r
e
1
2
,
illu
s
tr
ates
th
e
p
r
o
p
o
s
e
d
s
y
s
tem
'
s
ef
f
ec
tiv
e
f
u
n
ctio
n
ality
,
with
h
ig
h
ac
cu
r
a
cy
ac
h
iev
ed
in
class
if
y
in
g
th
e
r
ip
en
ess
lev
els o
f
o
v
e
r
r
ip
e
k
e
p
o
k
an
d
m
u
li.
Fig
u
r
e
1
0
.
R
ip
e
class
if
icatio
n
r
esu
lt u
n
d
er
v
ar
io
u
s
lam
p
p
o
wer
Fig
u
r
e
1
1
.
Par
tially
r
ip
e
class
if
icatio
n
r
esu
lt u
n
d
er
v
ar
io
u
s
lam
p
p
o
wer
Fo
r
k
ep
o
k
,
u
n
d
er
th
e
6
-
watt
illu
m
in
atio
n
ca
teg
o
r
y
,
an
a
v
er
ag
e
ac
cu
r
ac
y
o
f
9
9
%
is
ac
h
iev
ed
;
th
e
1
2
-
watt
ca
teg
o
r
y
d
em
o
n
s
tr
ates
a
p
er
f
ec
t
1
0
0
%
ac
c
u
r
ac
y
r
ate,
wh
ile
t
h
e
2
2
-
watt
ca
te
g
o
r
y
m
ain
tain
s
an
eq
u
ally
im
p
r
ess
iv
e
av
er
ag
e
a
cc
u
r
ac
y
o
f
1
0
0
%.
Similar
ly
,
f
o
r
m
u
li,
th
e
6
-
watt
ca
teg
o
r
y
attain
s
an
av
er
a
g
e
ac
cu
r
ac
y
o
f
9
5
%,
th
e
1
2
-
watt
ca
teg
o
r
y
r
ea
ch
es
1
0
0
%
ac
cu
r
ac
y
,
an
d
th
e
2
2
-
watt
ca
teg
o
r
y
s
ec
u
r
es
an
ac
cu
r
ac
y
r
ate
o
f
9
8
%.
I
n
tr
ig
u
in
g
ly
,
th
e
s
y
s
tem
co
n
tin
u
es
to
u
p
h
o
l
d
its
p
er
f
o
r
m
a
n
ce
ev
e
n
wh
en
th
e
s
p
ec
im
en
s
ar
e
pos
itio
n
ed
at
d
is
tan
ce
s
b
etwe
en
3
0
cm
an
d
6
0
cm
.
I
n
b
o
th
k
ep
o
k
an
d
m
u
li
ca
s
es,
th
e
ac
c
u
r
ate
class
if
icatio
n
r
em
ain
s
co
n
s
is
ten
t,
b
o
asti
n
g
a
n
av
er
ag
e
ac
cu
r
ac
y
o
f
1
0
0
%
with
in
th
is
d
is
tan
ce
r
an
g
e.
T
h
ese
r
esu
lts
s
h
o
w
th
at
illu
m
in
atio
n
im
p
ac
ts
ac
cu
r
ac
y
.
T
h
is
is
b
ec
au
s
e
th
e
lev
el
o
f
illu
m
in
atio
n
in
cr
ea
s
es th
e
ca
m
er
a'
s
ab
ilit
y
to
b
etter
r
ec
o
g
n
ize
o
b
jects.
T
h
is
will in
f
lu
en
ce
th
e
u
s
er
s
y
s
tem
to
co
n
s
id
er
u
s
in
g
h
ig
h
er
-
wattag
e
lig
h
ts
wh
en
d
i
s
p
lay
in
g
b
an
a
n
as
o
f
v
ar
io
u
s
v
ar
ieties an
d
r
ip
en
ess
lev
els.
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
5
10
15
20
25
30
A
cc
u
r
ac
y
(%)
T
i
me
(
s)
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
10
20
30
40
50
60
70
80
90
1
0
0
A
c
c
u
r
a
c
y
(
%)
D
i
st
a
n
c
e
(
c
m)
K
e
p
o
k
6
W
a
t
t
K
e
p
o
k
1
2
W
a
t
t
K
e
p
o
k
2
2
W
a
t
t
M
u
l
i
6
W
a
t
t
M
u
l
i
1
2
W
a
t
t
M
u
l
i
2
2
W
a
t
t
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
5
10
15
20
25
30
A
c
c
u
r
a
c
y
(
%)
T
i
me
(
s)
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
10
20
30
40
50
60
70
80
90
1
0
0
A
c
c
u
r
a
c
y
(
%)
D
i
st
a
n
c
e
(
c
m)
K
e
p
o
k
6
W
a
t
t
K
e
p
o
k
1
2
W
a
t
t
K
e
p
o
k
2
2
W
a
t
t
M
u
l
i
6
W
a
t
t
M
u
l
i
1
2
w
a
t
t
M
u
l
i
2
2
W
a
t
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
C
la
s
s
i
fica
tio
n
s
ystem
o
f B
a
n
a
n
a
Typ
es a
n
d
r
ip
en
ess
leve
ls
b
a
s
ed
o
n
c
o
n
vo
lu
tio
n
a
l n
e
u
r
a
l
…
(
Lu
cia
Ja
mb
o
l
a
)
4899
Fig
u
r
e
1
2
.
Ov
er
r
ip
e
class
if
icat
io
n
r
esu
lt u
n
d
er
v
a
r
io
u
s
lam
p
p
o
wer
4.
CO
NCLU
SI
O
N
I
n
th
is
s
tu
d
y
,
th
e
p
r
o
p
o
s
ed
s
y
s
tem
co
u
ld
class
if
y
b
an
a
n
a
ty
p
e
an
d
lev
el
o
f
r
ip
en
ess
in
r
ea
l
-
tim
e
co
n
d
itio
n
s
u
s
in
g
a
C
NN
an
d
J
etso
n
Nan
o
as
h
ar
d
war
e.
T
h
e
r
esu
lt
o
f
th
e
class
if
icatio
n
p
r
o
c
ess
b
ased
o
n
b
an
an
a
ty
p
e
with
in
th
e
ca
te
g
o
r
ies o
f
6
watts,
1
2
watts,
an
d
2
2
watts ex
h
ib
it e
lev
ated
ac
c
u
r
ac
y
v
alu
es,
with
an
av
er
a
g
e
ac
cu
r
ac
y
r
ate
b
etwe
en
8
0
%
to
1
0
0
%.
T
h
e
ass
ess
m
en
t
o
f
r
ip
en
ess
lev
els
in
th
e
1
2
-
watt
a
n
d
2
2
-
watt
ca
teg
o
r
ies
y
ield
s
av
er
ag
e
ac
cu
r
ac
y
v
alu
es
ex
ce
ed
in
g
8
0
%,
wh
ile
th
e
6
-
w
att
ca
teg
o
r
y
d
em
o
n
s
tr
ates
a
lo
wer
ac
cu
r
ac
y
r
esu
lt,
f
allin
g
b
elo
w
6
0
%.
No
tab
ly
,
t
h
e
lu
m
in
a
n
ce
o
f
t
h
e
s
y
s
tem
s
ig
n
if
ican
tly
i
n
f
lu
en
ce
s
th
e
s
y
s
t
em
'
s
p
er
f
o
r
m
an
ce
.
Hig
h
er
illu
m
in
atio
n
d
u
r
i
n
g
test
in
g
co
r
r
elate
s
with
elev
ated
ac
cu
r
ac
y
lev
els,
as
o
b
s
er
v
ed
in
th
e
1
2
-
watt
an
d
22
-
watt
s
ettin
g
s
.
Fu
r
th
e
r
m
o
r
e,
th
e
im
p
ac
t
o
f
d
is
tan
ce
d
u
r
in
g
t
esti
n
g
is
ev
id
en
t,
r
e
v
ea
lin
g
h
ig
h
er
ac
c
u
r
ac
y
v
alu
es
with
in
th
e
r
a
n
g
e
o
f
3
0
cm
to
5
0
cm
,
av
e
r
ag
in
g
ab
o
v
e
8
5
%.
C
o
n
v
er
s
ely
,
ac
cu
r
ac
y
d
im
i
n
is
h
es
as
th
e
d
is
tan
ce
in
cr
ea
s
es.
T
h
en
,
it
s
h
o
wed
th
at
illu
m
in
atio
n
im
p
ac
ts
a
cc
u
r
ac
y
d
u
e
to
th
e
lev
el
o
f
illu
m
i
n
atio
n
in
cr
ea
s
es
th
e
ca
m
er
a'
s
ab
ilit
y
to
b
etter
r
ec
o
g
n
ize
o
b
jects.
T
h
is
s
y
s
tem
will
b
e
u
s
ef
u
l
in
h
elp
in
g
s
u
p
er
m
ar
k
et
o
f
f
icer
s
an
d
b
u
y
er
s
in
au
t
o
m
atica
lly
class
if
y
in
g
t
h
e
ty
p
e
an
d
lev
el
o
f
r
ip
e
n
ess
o
f
b
a
n
an
as.
I
n
t
h
e
f
u
tu
r
e
,
th
e
ty
p
e
o
f
b
an
a
n
a,
esp
ec
ially
im
p
o
r
ted
b
an
a
n
as,
i.e
.
,
C
av
en
d
is
h
,
n
ee
d
s
to
b
e
in
cr
ea
s
ed
as
a
d
ataset,
an
d
th
e
ap
p
licatio
n
o
f
th
e
b
an
an
a
class
if
icatio
n
s
y
s
tem
in
r
ea
l
en
v
ir
o
n
m
en
ts
n
ee
d
s
to
b
e
d
ev
elo
p
ed
.
I
n
ad
d
itio
n
,
th
e
u
s
e
o
f
th
e
s
am
e
s
y
s
tem
ca
n
b
e
a
p
p
li
ed
to
o
th
er
f
r
u
its
o
r
v
e
g
etab
les,
s
o
th
at
th
e
s
m
ar
t
m
ar
k
et
will
b
e
a
v
er
y
g
o
o
d
d
ev
el
o
p
m
en
t
in
th
e
f
u
tu
r
e.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
wo
r
k
was
s
u
p
p
o
r
ted
b
y
R
esear
ch
Gr
an
t
f
o
r
m
Kem
d
ik
tis
ain
tek
in
1
2
5
/C
3
/DT
.
0
5
.
0
0
/PL
/2
0
2
5
an
d
2
3
/SP
K/R
ek
to
r
at/I
ten
as/VI/2
0
2
5
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
All
au
th
o
r
s
h
av
e
co
n
tr
ib
u
ted
s
i
g
n
if
ican
tly
to
th
e
c
o
n
ce
p
t
u
aliza
tio
n
,
m
eth
o
d
o
lo
g
y
,
r
esear
ch
a
n
d
wr
itin
g
o
f
th
e
p
a
p
er
,
a
n
d
th
ey
h
a
v
e
a
ll
r
ea
d
an
d
ag
r
ee
d
to
th
e
c
u
r
r
en
t
v
er
s
io
n
o
f
th
e
m
an
u
s
cr
ip
t
.
Usi
n
g
th
e
jo
u
r
n
al
C
o
n
tr
ib
u
to
r
R
o
les T
ax
o
n
o
m
y
(
C
R
ed
iT)
,
th
e
co
n
t
r
ib
u
tio
n
s
ar
e
s
u
m
m
ar
is
ed
as f
o
llo
ws.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
L
u
cia
J
am
b
o
la
✓
✓
✓
✓
✓
✓
✓
✓
✓
A
r
s
y
a
d
R
a
m
a
d
h
a
n
D
a
r
l
i
s
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
W
in
d
i M
alah
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
Dwi
Ar
y
an
ta
✓
✓
✓
✓
✓
✓
✓
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
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
5
10
15
20
25
30
A
c
c
u
r
a
c
y
(
%)
T
i
me
(
s)
K
e
p
o
k
6
W
a
t
t
K
e
p
o
k
1
2
w
a
t
t
K
e
p
o
k
2
2
W
a
t
t
M
u
l
i
6
w
a
t
t
M
u
l
i
1
2
w
a
t
t
M
u
l
i
2
2
W
a
t
t
0%
2
0
%
4
0
%
6
0
%
8
0
%
1
0
0
%
10
20
30
40
50
60
70
80
90
1
0
0
A
c
c
u
r
a
c
y
(
%)
D
i
st
a
n
c
e
(
c
m)
K
e
p
o
k
6
W
a
t
t
K
e
p
o
k
1
2
W
a
t
t
K
ep
o
k
2
2
W
a
t
t
M
u
l
i
6
W
at
t
M
u
l
i
1
2
W
a
t
t
M
u
l
i
2
2
W
a
t
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
9
1
-
4
9
0
1
4900
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
av
ailab
ilit
y
is
n
o
t
ap
p
lica
b
le
to
th
is
p
ap
er
as n
o
n
ew
d
at
a
wer
e
cr
ea
ted
o
r
an
aly
ze
d
in
t
h
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
R
.
K
.
Y
a
d
a
v
a
n
d
J.
C
.
D
a
g
a
r
,
“
I
n
n
o
v
a
t
i
o
n
s
i
n
u
t
i
l
i
z
a
t
i
o
n
o
f
p
o
o
r
-
q
u
a
l
i
t
y
w
a
t
e
r
f
o
r
su
s
t
a
i
n
a
b
l
e
a
g
r
i
c
u
l
t
u
r
a
l
p
r
o
d
u
c
t
i
o
n
,
”
i
n
I
n
n
o
v
a
t
i
v
e
S
a
l
i
n
e
A
g
ri
c
u
l
t
u
re
,
N
e
w
D
e
l
h
i
:
S
p
r
i
n
g
e
r
I
n
d
i
a
,
2
0
1
6
,
p
p
.
2
1
9
–
2
6
3
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
81
-
3
2
2
-
2
7
7
0
-
0
_
1
1
.
[
2
]
B
.
S
i
n
g
h
,
J
.
P
.
S
i
n
g
h
,
A
.
K
a
u
r
,
a
n
d
N
.
S
i
n
g
h
,
“
B
i
o
a
c
t
i
v
e
c
o
mp
o
u
n
d
s
i
n
b
a
n
a
n
a
a
n
d
t
h
e
i
r
a
ss
o
c
i
a
t
e
d
h
e
a
l
t
h
b
e
n
e
f
i
t
s
–
a
r
e
v
i
e
w
,
”
Fo
o
d
C
h
e
m
i
st
r
y
,
v
o
l
.
2
0
6
,
p
p
.
1
–
1
1
,
S
e
p
.
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
f
o
o
d
c
h
e
m.
2
0
1
6
.
0
3
.
0
3
3
.
[
3
]
N
.
E
.
-
B
e
n
d
a
r
y
,
E.
E
.
H
a
r
i
r
i
,
A
.
E.
H
a
ssan
i
e
n
,
a
n
d
A
.
B
a
d
r
,
“
U
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
e
v
a
l
u
a
t
i
n
g
t
o
ma
t
o
r
i
p
e
n
e
s
s,”
Ex
p
e
rt
S
y
st
e
m
s w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
4
2
,
n
o
.
4
,
p
p
.
1
8
9
2
–
1
9
0
5
,
M
a
r
.
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
1
4
.
0
9
.
0
5
7
.
[
4
]
F
.
M
.
A
.
M
a
z
e
n
a
n
d
A
.
A
.
N
a
sh
a
t
,
“
R
i
p
e
n
e
ss
c
l
a
s
si
f
i
c
a
t
i
o
n
o
f
b
a
n
a
n
a
s
u
s
i
n
g
a
n
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
A
ra
b
i
a
n
J
o
u
r
n
a
l
f
o
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
4
4
,
n
o
.
8
,
p
p
.
6
9
0
1
–
6
9
1
0
,
A
u
g
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
3
6
9
-
0
1
8
-
0
3
6
9
5
-
5.
[
5
]
Y
.
Zh
a
n
g
,
J.
L
i
a
n
,
M
.
F
a
n
,
a
n
d
Y
.
Z
h
e
n
g
,
“
D
e
e
p
i
n
d
i
c
a
t
o
r
f
o
r
f
i
n
e
-
g
r
a
i
n
e
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
b
a
n
a
n
a
’
s
r
i
p
e
n
i
n
g
s
t
a
g
e
s,
”
EU
R
AS
I
P
J
o
u
rn
a
l
o
n
I
m
a
g
e
a
n
d
Vi
d
e
o
Pro
c
e
ss
i
n
g
,
v
o
l
.
2
0
1
8
,
n
o
.
1
,
D
e
c
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
3
6
4
0
-
0
1
8
-
0
2
8
4
-
8.
[
6
]
M
.
K
.
S
r
i
,
K
.
S
a
i
k
r
i
s
h
n
a
,
a
n
d
V
.
V
.
K
u
m
a
r
,
“
C
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
r
i
p
e
n
i
n
g
o
f
b
a
n
a
n
a
f
r
u
i
t
u
s
i
n
g
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
s,”
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
4
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
:
I
n
n
o
v
a
t
i
v
e
Ad
v
a
n
c
e
m
e
n
t
i
n
E
n
g
i
n
e
e
r
i
n
g
&
T
e
c
h
n
o
l
o
g
y
(
I
AET)
2
0
2
0
,
2
0
2
0
,
p
p
.
1
-
6
,
d
o
i
:
1
0
.
2
1
3
9
/
ssr
n
.
3
5
5
8
3
5
5
.
[
7
]
N
.
S
a
r
a
n
y
a
,
K
.
S
r
i
n
i
v
a
sa
n
,
a
n
d
S
.
K
.
P
.
K
u
m
a
r
,
“
B
a
n
a
n
a
r
i
p
e
n
e
ss
st
a
g
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
:
a
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
J
o
u
rn
a
l
o
f
Am
b
i
e
n
t
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
H
u
m
a
n
i
z
e
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
3
,
n
o
.
8
,
p
p
.
4
0
3
3
–
4
0
3
9
,
A
u
g
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
2
6
5
2
-
0
2
1
-
0
3
2
6
7
-
w.
[
8
]
S
.
A
sh
r
a
f
,
I
.
K
a
d
e
r
y
,
M
.
A
.
A
.
C
h
o
w
d
h
u
r
y
,
T.
Z.
M
a
h
b
u
b
,
a
n
d
R
.
M
.
R
a
h
m
a
n
,
“
F
r
u
i
t
i
m
a
g
e
c
l
a
s
si
f
i
c
a
t
i
o
n
u
s
i
n
g
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
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
S
o
f
t
w
a
r
e
I
n
n
o
v
a
t
i
o
n
,
v
o
l
.
7
,
n
o
.
4
,
p
p
.
5
1
–
7
0
,
O
c
t
.
2
0
1
9
,
d
o
i
:
1
0
.
4
0
1
8
/
I
JS
I
.
2
0
1
9
1
0
0
1
0
3
.
[
9
]
R
.
P
a
t
h
a
k
a
n
d
H
.
M
a
k
w
a
n
a
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
o
f
f
r
u
i
t
s
u
s
i
n
g
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
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
mo
d
e
l
s,”
J
o
u
rn
a
l
o
f
Ma
n
a
g
e
m
e
n
t
I
n
f
o
rm
a
t
i
o
n
a
n
d
D
e
c
i
si
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
2
4
,
n
o
.
3
,
p
p
.
1
–
1
2
,
2
0
2
1
.
[
1
0
]
T.
S
h
u
p
r
a
j
h
a
a
,
J.
M
.
R
a
j
,
P
.
S
.
K
u
m
a
r
,
K
.
N
.
S
h
e
e
b
a
,
a
n
d
K
.
D
h
a
y
a
l
i
n
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
b
a
se
d
m
o
b
i
l
e
a
p
p
l
i
c
a
t
i
o
n
f
o
r
v
a
r
i
e
t
a
l
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
d
r
i
p
e
n
e
ss g
r
a
d
i
n
g
o
f
t
r
a
d
i
t
i
o
n
a
l
I
n
d
i
a
n
b
a
n
a
n
a
v
a
r
i
e
t
i
e
s,”
i
n
2
0
2
3
S
e
c
o
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
El
e
c
t
r
i
c
a
l
,
El
e
c
t
r
o
n
i
c
s
,
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s
(
I
C
EEICT)
,
A
p
r
.
2
0
2
3
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
EEI
C
T5
6
9
2
4
.
2
0
2
3
.
1
0
1
5
7
5
7
8
.
[
1
1
]
R
.
G
.
B
a
l
d
o
v
i
n
o
,
R
.
A
.
U
.
L
i
m,
P
.
R
.
R
.
S
a
l
v
a
d
o
r
,
a
n
d
E
.
A
.
P
.
Ti
a
mz
o
n
,
“
R
e
a
l
-
t
i
me
b
a
n
a
n
a
r
i
p
e
n
e
ss
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
u
si
n
g
y
o
l
o
v
8
,
”
i
n
2
0
2
4
9
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Me
c
h
a
t
ro
n
i
c
s
E
n
g
i
n
e
e
ri
n
g
(
I
C
O
M)
,
I
EEE,
A
u
g
.
2
0
2
4
,
p
p
.
2
1
9
–
2
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
O
M
6
1
6
7
5
.
2
0
2
4
.
1
0
6
5
2
4
3
8
.
[
1
2
]
O
.
I
.
A
b
i
o
d
u
n
,
A
.
Ja
n
t
a
n
,
A
.
E.
O
m
o
l
a
r
a
,
K
.
V
.
D
a
d
a
,
N
.
A
.
M
o
h
a
m
e
d
,
a
n
d
H
.
A
r
sh
a
d
,
“
S
t
a
t
e
-
of
-
t
h
e
-
a
r
t
i
n
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
a
p
p
l
i
c
a
t
i
o
n
s:
a
s
u
r
v
e
y
,
”
H
e
l
i
y
o
n
,
v
o
l
.
4
,
n
o
.
1
1
,
N
o
v
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
1
8
.
e
0
0
9
3
8
.
[
1
3
]
D
.
M
i
c
h
e
l
sa
n
t
i
,
Z
.
-
H
.
Ta
n
,
S
.
S
i
g
u
r
d
sso
n
,
a
n
d
J.
Je
n
se
n
,
“
D
e
e
p
-
l
e
a
r
n
i
n
g
-
b
a
s
e
d
a
u
d
i
o
-
v
i
su
a
l
s
p
e
e
c
h
e
n
h
a
n
c
e
m
e
n
t
i
n
p
r
e
se
n
c
e
o
f
l
o
m
b
a
r
d
e
f
f
e
c
t
,
”
S
p
e
e
c
h
C
o
m
m
u
n
i
c
a
t
i
o
n
,
v
o
l
.
1
1
5
,
p
p
.
3
8
–
5
0
,
D
e
c
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
sp
e
c
o
m.
2
0
1
9
.
1
0
.
0
0
6
.
[
1
4
]
N
.
S
.
M
.
S
a
i
d
,
H
.
M
a
d
z
i
n
,
S
.
K
.
A
l
i
,
a
n
d
N
.
S
.
B
e
n
g
,
“
C
o
m
p
a
r
i
so
n
o
f
c
o
l
o
r
-
b
a
s
e
d
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
m
e
t
h
o
d
s
i
n
b
a
n
a
n
a
l
e
a
f
d
i
s
e
a
s
e
s
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
S
V
M
a
n
d
K
-
NN
,”
I
n
d
o
n
e
s
i
a
n
J
o
u
r
n
a
l
o
f
E
l
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
4
,
n
o
.
3
,
p
p
.
1
5
2
3
-
1
5
3
3
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
2
4
.
i
3
.
p
p
1
5
2
3
-
1
5
3
3
.
[
1
5
]
R
.
K
o
sas
i
h
,
S
.
S
u
d
a
r
y
a
n
t
o
,
a
n
d
A
.
F
a
h
r
u
r
o
z
i
,
“
C
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
si
x
b
a
n
a
n
a
r
i
p
e
n
e
s
s l
e
v
e
l
s b
a
s
e
d
o
n
st
a
t
i
st
i
c
a
l
f
e
a
t
u
r
e
s
o
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ad
v
a
n
c
e
s
i
n
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
2
,
n
o
.
4
,
p
p
.
3
1
7
-
326
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
a
s
.
v
1
2
.
i
4
.
p
p
3
1
7
-
3
2
6
.
[
1
6
]
R
.
C
h
e
l
l
a
p
p
a
,
S
.
T
h
e
o
d
o
r
i
d
i
s,
a
n
d
A
.
v
a
n
S
c
h
a
i
k
,
“
A
d
v
a
n
c
e
s
i
n
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
”
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
EEE
,
v
o
l
.
1
0
9
,
n
o
.
5
,
p
p
.
6
0
7
–
6
1
1
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
JP
R
O
C
.
2
0
2
1
.
3
0
7
2
1
7
2
.
[
1
7
]
B
.
R
.
H
u
ss
e
i
n
,
O
.
A
.
M
a
l
i
k
,
W
.
-
H
.
O
n
g
,
a
n
d
J.
W
.
F
.
S
l
i
k
,
“
R
e
c
o
n
st
r
u
c
t
i
o
n
o
f
d
a
m
a
g
e
d
h
e
r
b
a
r
i
u
m
l
e
a
v
e
s
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s f
o
r
i
m
p
r
o
v
i
n
g
c
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
,
”
Ec
o
l
o
g
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
6
1
,
M
a
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
c
o
i
n
f
.
2
0
2
1
.
1
0
1
2
4
3
.
[
1
8
]
P
.
D
h
r
u
v
a
n
d
S
.
N
a
sk
a
r
,
“
I
mag
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
u
si
n
g
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
(
C
N
N
)
a
n
d
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
(
R
N
N
)
:
a
r
e
v
i
e
w
,
”
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
I
n
f
o
r
m
a
t
i
o
n
Pr
o
c
e
ss
i
n
g
,
2
0
2
0
,
p
p
.
3
6
7
–
381
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
981
-
15
-
1
8
8
4
-
3
_
3
4
.
[
1
9
]
N
.
S
i
n
g
h
a
n
d
H
.
S
a
b
r
o
l
,
“
C
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
-
a
n
e
x
t
e
n
s
i
v
e
a
r
e
n
a
o
f
d
e
e
p
l
e
a
r
n
i
n
g
.
a
c
o
mp
r
e
h
e
n
s
i
v
e
s
t
u
d
y
,
”
Ar
c
h
i
v
e
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
Me
t
h
o
d
s i
n
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
2
8
,
n
o
.
7
,
p
p
.
4
7
5
5
–
4
7
8
0
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
8
3
1
-
0
2
1
-
0
9
5
5
1
-
4.
[
2
0
]
S
.
K
.
N
i
a
z
i
a
n
d
Z
.
M
a
r
i
a
m
,
“
R
e
c
e
n
t
a
d
v
a
n
c
e
s
i
n
ma
c
h
i
n
e
-
l
e
a
r
n
i
n
g
-
b
a
se
d
c
h
e
mo
i
n
f
o
r
m
a
t
i
c
s
:
a
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
M
o
l
e
c
u
l
a
r
S
c
i
e
n
c
e
s
,
v
o
l
.
24
,
n
o
.
14
,
J
u
l
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
i
j
ms
2
4
1
4
1
1
4
8
8
.
[
2
1
]
T.
I
t
o
e
t
a
l
.
,
“
D
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
i
n
c
o
r
p
o
r
a
t
i
n
g
c
n
n
a
n
d
mf
f
o
r
i
t
e
m
-
b
a
se
d
f
a
s
h
i
o
n
r
e
c
o
mm
e
n
d
a
t
i
o
n
,
”
K
n
o
w
l
e
d
g
e
M
a
n
a
g
e
m
e
n
t
a
n
d
Ac
q
u
i
si
t
i
o
n
f
o
r I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s (PK
A
W
2
0
2
1
)
,
2
0
2
1
,
p
p
.
4
6
–
57
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
6
9
8
8
6
-
7
_
4
.
[
2
2
]
L
.
A
l
z
u
b
a
i
d
i
e
t
a
l
.
,
“
R
e
v
i
e
w
o
f
d
e
e
p
l
e
a
r
n
i
n
g
:
c
o
n
c
e
p
t
s,
C
N
N
a
r
c
h
i
t
e
c
t
u
r
e
s
,
c
h
a
l
l
e
n
g
e
s,
a
p
p
l
i
c
a
t
i
o
n
s
,
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
s
,
”
J
o
u
rn
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
8
,
M
a
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
0
5
3
7
-
021
-
0
0
4
4
4
-
8
.
[
2
3
]
B
.
Li
a
n
d
D
.
Li
m
a
,
“
F
a
c
i
a
l
e
x
p
r
e
ss
i
o
n
r
e
c
o
g
n
i
t
i
o
n
v
i
a
r
e
s
n
e
t
-
5
0
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
g
n
i
t
i
v
e
C
o
m
p
u
t
i
n
g
i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
,
p
p
.
5
7
–
6
4
,
Ju
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
c
c
e
.
2
0
2
1
.
0
2
.
0
0
2
.
[
2
4
]
X
.
O
u
e
t
a
l
.
,
“
M
o
v
i
n
g
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
m
e
t
h
o
d
v
i
a
r
e
sn
e
t
-
1
8
w
i
t
h
e
n
c
o
d
e
r
–
d
e
c
o
d
e
r
st
r
u
c
t
u
r
e
i
n
c
o
m
p
l
e
x
s
c
e
n
e
s,”
I
EEE
A
c
c
e
ss
,
v
o
l
.
7
,
p
p
.
1
0
8
1
5
2
–
1
0
8
1
6
0
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
3
1
9
2
2
.
[
2
5
]
G
.
V
.
H
o
r
n
e
t
a
l
.
,
“
T
h
e
i
n
a
t
u
r
a
l
i
s
t
s
p
e
c
i
e
s
c
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
d
e
t
e
c
t
i
o
n
d
a
t
a
s
e
t
,
”
i
n
2
0
1
8
I
E
EE/
C
VF
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
Pa
t
t
e
rn
Re
c
o
g
n
i
t
i
o
n
,
I
EEE,
Ju
n
.
2
0
1
8
,
p
p
.
8
7
6
9
–
8
7
7
8
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
8
.
0
0
9
1
4
.
[
2
6
]
S
.
A
f
a
q
a
n
d
S
.
R
a
o
,
“
S
i
g
n
i
f
i
c
a
n
c
e
o
f
e
p
o
c
h
s
o
n
t
r
a
i
n
i
n
g
a
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
c
i
e
n
t
i
f
i
c
&
T
e
c
h
n
o
l
o
g
y
Re
se
a
rc
h
,
v
o
l
.
9
,
n
o
.
6
,
p
p
.
4
8
5
–
4
8
8
,
2
0
2
0
.
[
2
7
]
C
.
Ta
n
t
i
t
h
a
m
t
h
a
v
o
r
n
,
S
.
M
c
I
n
t
o
s
h
,
A
.
E.
H
a
ssa
n
,
a
n
d
K
.
M
a
t
s
u
m
o
t
o
,
“
A
n
e
mp
i
r
i
c
a
l
c
o
m
p
a
r
i
s
o
n
o
f
mo
d
e
l
v
a
l
i
d
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
f
o
r
d
e
f
e
c
t
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
s
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
S
o
f
t
w
a
re
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
4
3
,
n
o
.
1
,
p
p
.
1
–
1
8
,
Jan
.
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
TSE
.
2
0
1
6
.
2
5
8
4
0
5
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.