I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l
E
ng
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
25
,
No
.
3
,
Ma
r
ch
20
22
,
p
p
.
1
4
4
2
~
1
4
4
9
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
25
.i
3
.
pp
1
4
4
2
-
1
4
4
9
1442
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
Efficacy
o
f
c
hili p
la
nt
disea
ses
clas
s
ificatio
n using
d
e
ep learning
:
a
prelimina
ry
stu
dy
Su
ha
na
Ro
zla
n,
M
a
rsy
it
a
H
a
na
f
i
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
a
n
d
C
o
mm
u
n
i
c
a
t
i
o
n
S
y
s
t
e
ms
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
E
n
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
i
P
u
t
r
a
M
a
l
a
y
si
a
,
S
e
l
a
n
g
o
r
,
M
a
l
a
y
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
26
,
2
0
2
1
R
ev
is
ed
Dec
7
,
2
0
2
1
Acc
ep
ted
J
an
12
,
2
0
2
2
P
lan
t
d
ise
a
se
c
las
sifica
ti
o
n
u
sin
g
d
e
e
p
lea
rn
in
g
tec
h
n
iq
u
e
s
is
a
p
o
p
u
lar
re
se
a
rc
h
a
re
a
d
u
e
to
t
h
e
n
u
m
e
ro
u
s
o
p
p
o
r
tu
n
it
ies
f
o
r
i
n
tro
d
u
c
i
n
g
a
d
v
a
n
c
e
a
n
d
ro
b
u
st
c
las
sifiers
.
Ne
v
e
rth
e
les
s,
c
las
sify
in
g
c
h
il
li
p
la
n
t
d
ise
a
se
s
a
c
c
u
ra
tely
fro
m
ima
g
e
s
u
n
d
e
r
u
n
c
o
n
tr
o
l
l
e
d
e
n
v
ir
o
n
m
e
n
t
a
n
d
v
a
rio
u
s
ima
g
in
g
c
o
n
d
i
ti
o
n
s
re
m
a
in
s
u
n
s
o
lv
e
d
d
u
e
to
t
h
e
lac
k
o
f
c
h
il
li
d
ise
a
se
ima
g
e
d
a
tas
e
ts.
In
th
is
st
u
d
y
,
t
h
e
e
ffica
c
y
o
f
th
re
e
h
ig
h
-
p
e
rf
o
rm
a
n
c
e
d
e
e
p
lea
rn
in
g
a
lg
o
rit
h
m
s,
n
a
m
e
ly
VG
G
1
6
,
In
c
e
p
ti
o
n
V3
,
a
n
d
Eff
icie
n
tNe
tB
0
,
i
n
c
las
sify
in
g
t
h
re
e
ty
p
e
s
o
f
c
h
il
l
i
lea
v
e
s
d
ise
a
se
s,
n
a
m
e
ly
u
p
wa
rd
c
u
rli
n
g
,
m
o
sa
ic/m
o
tt
li
n
g
,
a
n
d
th
e
b
a
c
teria
l
sp
o
t,
is
d
e
m
o
n
stra
ted
.
T
h
e
se
m
e
th
o
d
s
a
re
p
o
p
u
larl
y
u
se
d
fo
r
o
t
h
e
r
p
lan
t
d
ise
a
se
c
las
sifica
ti
o
n
s
d
u
e
to
th
e
i
r
e
ffe
c
ti
v
e
n
e
ss
.
Th
e
e
x
p
e
rime
n
ts
we
re
p
e
rfo
rm
e
d
o
n
th
e
3
,
0
0
0
c
h
il
li
p
lan
t
d
ise
a
se
ima
g
e
s
c
o
ll
e
c
ted
fro
m
th
r
e
e
d
iffere
n
t
field
e
n
v
iro
n
m
e
n
ts i
n
S
e
lan
g
o
r,
M
a
lay
sia
.
Th
e
ima
g
e
s
we
re
c
a
p
tu
re
d
with
a
c
o
m
p
lex
b
a
c
k
g
r
o
u
n
d
a
n
d
v
a
rio
u
s
il
l
u
m
in
a
ti
o
n
s,
a
n
g
les
,
a
n
d
d
istan
c
e
s
to
re
flec
t
t
h
e
re
a
l
-
li
fe
sc
e
n
a
rio
s.
Th
e
c
o
m
p
lex
it
y
o
f
th
e
c
o
ll
e
c
ted
i
m
a
g
e
s
wa
s
c
re
a
t
e
d
b
a
se
d
o
n
t
h
e
t
a
x
o
n
o
m
ic
in
fo
rm
a
ti
o
n
o
f
c
h
i
ll
i
lea
v
e
s
d
ise
a
se
s
a
n
d
th
e
u
n
a
v
a
il
a
b
il
it
y
o
f
c
h
i
l
li
d
ise
a
se
ima
g
e
s
u
n
d
e
r
v
a
ri
o
u
s
ima
g
in
g
c
o
n
d
i
ti
o
n
s
i
n
t
h
e
p
u
b
li
c
ly
a
v
a
il
a
b
le
p
lan
t
d
ise
a
se
d
a
tab
a
se
s.
Ex
p
e
rime
n
te
d
u
si
n
g
a
p
p
r
o
p
riate
s
p
e
c
ifi
c
a
ti
o
n
s,
th
e
m
o
d
e
ls
d
e
m
o
n
stra
ted
o
u
tstan
d
i
n
g
p
e
rfo
rm
a
n
c
e
with
m
o
re
t
h
a
n
9
5
%
a
c
c
u
ra
c
y
with
th
e
h
ig
h
e
st ac
c
u
ra
c
y
o
f
9
8
.
8
3
%
b
y
In
c
e
p
t
io
n
V3
.
K
ey
w
o
r
d
s
:
C
h
ili lea
v
e
d
is
ea
s
es
C
o
m
p
lex
im
ag
es
Dee
p
lear
n
in
g
Plan
t d
is
ea
s
e
clas
s
if
icatio
n
T
r
an
s
f
er
lear
n
i
n
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
:
Ma
r
s
y
ita
Han
af
i
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
a
n
d
C
o
m
m
u
n
icatio
n
Sy
s
tem
s
E
n
g
i
n
ee
r
in
g
,
Facu
lty
o
f
E
n
g
in
ee
r
in
g
Un
iv
er
s
iti Pu
tr
a
Ma
lay
s
ia
4
3
4
0
0
UPM
Ser
d
an
g
,
Selan
g
o
r
,
Ma
lay
s
ia
E
m
ail:
m
ar
s
y
ita@
u
p
m
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
C
h
i
l
li
o
r
s
cien
tific
ally
k
n
o
wn
as
ca
p
s
icu
m
an
n
u
u
m
L
.
is
M
alay
s
ia’
s
u
tm
o
s
t
cu
ltiv
ated
cr
o
p
[
1
]
a
n
d
h
as
b
ee
n
r
ec
o
g
n
ized
as
am
o
n
g
t
h
e
to
p
ten
th
o
f
s
elf
-
s
u
f
f
icien
cy
r
atio
(
SS
R
)
in
s
elec
ted
ag
r
icu
ltu
r
al
co
m
m
o
d
ities
with
t
h
e
h
i
g
h
est
im
p
o
r
t
d
e
p
en
d
e
n
cy
r
atio
(
I
D
R
)
o
f
7
3
.
6
%
[
2
]
.
T
h
e
f
i
v
e
wi
d
ely
d
o
m
esti
ca
ted
s
p
ec
ies
p
lan
ted
as
an
n
u
al
cr
o
p
s
ar
e
ca
p
s
icu
m
an
n
u
u
m
,
ca
p
s
icu
m
f
r
u
tescen
s
e
,
ca
p
s
icu
m
ch
in
en
s
e,
ca
p
s
icu
m
b
ac
ca
tu
m
,
an
d
ca
p
s
icu
m
p
u
b
escen
ce
[
3
]
,
p
lag
u
es
an
d
d
is
e
ases
ea
s
ily
in
f
ec
t
th
ese
p
lan
t
s
.
T
h
e
ef
f
ec
ts
o
f
th
e
in
f
ec
tio
n
a
r
e
a
s
ig
n
if
ican
t
r
ed
u
ctio
n
in
ch
i
l
li
p
r
o
d
u
ctio
n
an
d
d
eter
i
o
r
atio
n
o
f
f
r
u
it
q
u
ality
,
r
esu
ltin
g
in
lo
w
r
etu
r
n
s
f
o
r
f
a
r
m
er
s
[
4
]
.
Acc
o
r
d
in
g
t
o
[
5
]
,
c
h
i
l
l
i
p
lan
t
d
is
e
ases
ar
e
m
ain
ly
d
u
e
to
th
e
i
n
f
ec
tio
n
ca
u
s
ed
b
y
p
ath
o
g
e
n
ic
m
icr
o
b
es,
n
am
ely
f
u
n
g
u
s
,
b
ac
ter
ia,
an
d
v
ir
u
s
es.
T
h
e
in
f
ec
tio
n
is
v
is
ib
le
b
u
t
n
ee
d
s
to
b
e
ex
am
in
ed
clo
s
ely
an
d
ad
e
q
u
ate
ly
c
o
n
tr
o
l
led
to
av
o
i
d
th
e
m
ass
iv
e
d
is
ea
s
es
s
p
r
ea
d
o
n
th
e
f
ar
m
.
T
h
e
c
o
n
v
en
tio
n
al
way
o
f
d
etec
tin
g
an
d
class
if
y
in
g
p
l
an
t
d
is
ea
s
es
is
tim
e
-
co
n
s
u
m
in
g
,
an
d
au
to
m
atic
d
etec
tio
n
an
d
class
if
icatio
n
ap
p
r
o
ac
h
es
h
av
e
b
ee
n
in
tr
o
d
u
ce
d
to
tack
le
th
is
p
r
o
b
lem
.
I
n
th
e
late
1
9
9
0
s
,
c
o
n
v
e
n
tio
n
al
co
m
p
u
ter
v
is
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
E
ffica
cy
o
f c
h
ili p
la
n
t d
is
ea
s
es c
la
s
s
ifica
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
:
a
p
r
elimin
a
r
y
s
tu
d
y
(
S
u
h
a
n
a
R
o
z
la
n
)
1443
tech
n
iq
u
es
wer
e
u
s
ed
to
r
eso
l
v
e
ch
illi
p
lan
t
d
is
ea
s
es’
au
to
m
atic
d
etec
tio
n
an
d
id
e
n
tific
atio
n
[
6
]
.
T
h
e
m
ajo
r
wea
k
n
ess
o
f
th
e
tr
a
d
i
tio
n
al
co
m
p
u
ter
v
is
io
n
tech
n
iq
u
e
was
th
at
it
was
o
n
ly
p
r
o
v
e
n
s
u
cc
ess
f
u
l
o
n
s
im
p
ler
a
n
d
co
n
tr
o
lled
s
etu
p
s
b
u
t stru
g
g
le
d
as th
e
o
p
e
r
atio
n
al
co
n
d
itio
n
s
ch
an
g
ed
[
7
]
.
As
th
e
y
ea
r
s
p
ass
ed
,
au
to
m
atic
d
etec
tio
n
an
d
class
if
icatio
n
o
f
p
la
n
t
d
is
ea
s
es
u
tili
zi
n
g
im
ag
e
p
r
o
ce
s
s
in
g
an
d
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
es
h
av
e
r
ec
eiv
e
d
s
ig
n
if
ican
t
co
n
s
id
er
atio
n
a
m
o
n
g
th
e
ex
p
er
ts
o
f
th
e
s
u
b
ject.
Dee
p
lear
n
in
g
is
a
b
r
an
ch
o
f
ar
tific
ial
in
tellig
en
ce
th
at
allo
ws
m
ac
h
in
es
to
p
er
f
o
r
m
im
p
r
ess
iv
e
r
ec
o
g
n
itio
n
,
p
r
ed
ictio
n
,
an
d
f
iltra
tio
n
[
8
]
.
Ma
n
y
p
r
ac
tical
an
d
r
eliab
le
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
f
o
r
p
la
n
t
d
is
ea
s
e
clas
s
if
ica
tio
n
[
9
]
–
[
1
3
]
.
T
y
p
ically
,
th
e
class
if
icatio
n
is
p
er
f
o
r
m
e
d
ac
co
r
d
in
g
to
t
h
e
in
f
ec
ted
leaf
s
h
a
p
e
an
d
a
d
etec
tab
le
ch
an
g
e
in
th
e
leaf
co
lo
u
r
ca
u
s
ed
b
y
th
e
f
u
n
g
u
s
,
b
ac
ter
ia,
an
d
v
ir
u
s
in
f
ec
tio
n
.
T
h
e
a
p
p
licatio
n
o
f
th
e
tr
an
s
f
er
lear
n
in
g
ap
p
r
o
ac
h
f
o
r
d
ee
p
lear
n
in
g
h
as
r
ec
eiv
ed
s
ig
n
if
ican
t
atten
tio
n
.
T
r
an
s
f
er
lear
n
in
g
h
as
em
er
g
ed
as
a
p
o
wer
f
u
l
tech
n
i
q
u
e
wh
er
e
b
y
t
h
e
k
n
o
wled
g
e
g
ain
ed
f
r
o
m
th
e
lar
g
er
d
ataset
i
s
tr
an
s
f
er
r
ed
t
o
th
e
n
ew
d
ataset
[
1
4
]
,
[
1
5
]
.
I
n
s
ce
n
ar
io
s
with
in
s
u
f
f
icie
n
t
tr
ain
in
g
d
ata,
t
h
is
tech
n
iq
u
e
is
b
e
n
ef
ic
ial,
as
p
r
esen
ted
in
r
esear
ch
b
y
[
1
6
]
.
I
n
tr
an
s
f
er
lear
n
in
g
,
p
r
e
-
tr
ain
ed
m
o
d
els
ar
e
g
en
er
ally
tr
ain
e
d
o
n
a
l
ar
g
e
s
ca
le,
s
u
ch
as
I
m
ag
eNe
t
th
at
co
n
tain
s
m
illi
o
n
s
o
f
ac
tu
al
im
a
g
es.
T
h
e
ad
v
a
n
tag
e
is
th
at
th
e
lear
n
ed
f
ea
tu
r
es
ar
e
tr
a
n
s
f
er
r
ed
b
y
th
e
weig
h
ts
an
d
th
e
ar
c
h
itectu
r
e
o
b
tain
e
d
f
r
o
m
th
es
e
m
o
d
els
[
1
7
]
.
I
n
s
p
ir
ed
b
y
th
ese
f
in
d
in
g
s
,
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
e
-
tr
ain
e
d
m
o
d
el
o
f
VGG1
6
,
I
n
ce
p
tio
n
V3
,
an
d
E
f
f
icien
tNetB
0
in
class
if
y
in
g
ch
i
l
li
p
lan
t
d
is
ea
s
e
im
ag
es
ca
p
tu
r
ed
u
n
d
er
an
u
n
c
o
n
tr
o
lled
en
v
i
r
o
n
m
e
n
t
with
v
a
r
io
u
s
im
ag
i
n
g
c
o
n
d
itio
n
s
an
d
a
s
m
all
d
ataset
is
s
tu
d
ied
.
T
h
is
p
ap
er
s
h
o
ws
th
e
p
er
f
o
r
m
an
ce
o
f
th
ese
m
o
d
els
f
o
r
class
if
y
in
g
h
ig
h
ly
co
m
p
le
x
ch
i
l
li
p
lan
t
d
is
ea
s
es
im
ag
es.
T
h
e
f
i
n
d
in
g
s
in
th
is
p
a
p
er
will
c
r
ea
te
m
o
r
e
o
p
p
o
r
tu
n
ities
f
o
r
d
ev
elo
p
in
g
m
o
r
e
ac
cu
r
ate
class
if
ier
s
in
th
e
f
u
tu
r
e.
T
h
is
is
b
ec
au
s
e
th
e
ex
is
tin
g
s
tu
d
ie
s
h
av
e
o
n
l
y
s
h
o
wn
less
th
an
9
0
%
ac
cu
r
ac
y
o
n
a
p
ar
ticu
lar
ty
p
e
o
f
ch
illi
d
is
ea
s
e
[
1
8
]
,
[
1
9
]
.
T
h
is
p
a
p
er
is
o
r
g
an
ized
as
f
o
ll
o
ws.
Sectio
n
2
d
escr
ib
es
ch
i
l
li
p
lan
t
d
is
ea
s
e
tax
o
n
o
m
y
,
an
d
s
ec
tio
n
3
p
r
o
v
id
es
th
e
a
r
ch
itectu
r
e
s
o
f
th
e
u
s
ed
d
ee
p
lear
n
in
g
m
eth
o
d
s
,
m
ater
ials
,
m
eth
o
d
s
,
an
d
ex
p
er
im
en
tal
s
etu
p
,
an
d
s
ec
tio
n
4
d
is
cu
s
s
es
th
e
r
esu
lts
.
Fin
ally
,
th
e
p
ap
er
is
co
n
clu
d
e
d
in
s
ec
tio
n
5
.
2.
T
AXO
NO
M
Y
O
F
CH
I
L
I
DI
SE
AS
E
S
C
h
i
l
li
is
a
ty
p
e
o
f
p
lan
t
th
at
ca
n
b
e
ea
s
ily
af
f
ec
ted
b
y
f
u
n
g
i,
b
ac
ter
ia,
v
ir
u
s
es,
an
d
p
est
s
.
B
esid
es
,
clim
ate
ch
an
g
es
a
n
d
th
e
r
is
k
o
f
a
r
esis
tan
ce
b
r
ea
k
d
o
wn
ca
n
also
af
f
ec
t
th
e
d
u
r
ab
ilit
y
o
f
d
is
ea
s
e
r
esis
tan
ce
.
T
h
e
ex
am
p
le
o
f
th
e
f
u
n
g
i,
b
ac
ter
ia,
v
ir
u
s
es,
an
d
p
ests
co
m
m
o
n
ly
af
f
ec
ted
b
y
ch
i
l
li
p
lan
ts
[
5
]
ar
e
s
u
m
m
ar
ized
in
Fig
u
r
e
1
.
Fig
u
r
e
1
.
T
a
x
o
n
o
m
y
o
f
ch
illi
p
lan
t d
is
ea
s
es a
cc
o
r
d
in
g
to
[
5
]
I
n
th
is
s
tu
d
y
,
th
r
ee
ty
p
es
o
f
d
is
ea
s
e
wer
e
co
n
s
id
er
e
d
:
th
e
b
ac
ter
ial
s
p
o
t,
u
p
war
d
cu
r
lin
g
an
d
m
o
s
aic/m
o
ttli
n
g
,
as
s
h
o
wn
in
Fig
u
r
e
2
(
a)
,
Fig
u
r
e
2
(
b
)
,
an
d
Fig
u
r
e
2
(
c)
,
r
esp
ec
tiv
ely
.
T
h
e
b
ac
ter
ial
s
p
o
t
is
th
e
s
m
all
b
lack
s
p
o
ts
o
f
wate
r
-
s
o
ak
ed
o
n
th
e
lea
v
es
an
d
g
r
ad
u
ally
b
r
o
wn
in
g
,
co
alesce,
r
u
g
g
ed
an
d
c
r
ac
k
ed
.
I
t
is
m
ain
ly
d
u
e
to
th
e
p
at
h
o
g
e
n
th
at
is
k
n
o
wn
as
x
an
th
o
m
o
n
as
.
T
h
e
u
p
war
d
c
u
r
lin
g
d
is
ea
s
e
is
ca
u
s
ed
b
y
B
eg
o
m
o
v
ir
u
s
tr
an
s
m
itted
b
y
B
em
is
ia
wh
itef
lie
s
th
at
ca
u
s
e
d
y
ello
win
g
o
f
v
ein
s
an
d
r
ed
u
ce
d
leaf
s
ize.
T
h
e
m
o
s
aic
d
is
ea
s
e
ca
u
s
ed
th
e
leav
es to
b
e
y
ello
wed
an
d
n
ar
r
o
w,
wh
ich
is
tr
an
s
m
itted
m
ain
ly
b
y
g
r
ee
n
f
ly
a
p
h
id
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
25
,
No
.
3
,
Ma
r
ch
20
22
:
1
4
4
2
-
1
4
4
9
1444
Fig
u
r
e
2
.
Sam
p
les o
f
ch
ili p
la
n
t d
is
ea
s
es im
ag
e
u
s
ed
in
th
e
e
x
p
er
im
en
ts
,
(
a
)
b
ac
ter
ial
s
p
o
t
,
(
b
)
u
p
war
d
cu
r
lin
g
,
a
n
d
(
c
)
m
o
s
aic/m
o
ttli
n
g
3.
M
AT
E
R
I
AL
A
ND
M
E
T
H
O
DS
3
.
1
.
Chili
pla
nt
dis
ea
s
e
da
t
a
s
et
T
h
e
d
ataset
u
s
ed
in
th
is
s
tu
d
y
co
n
s
is
ts
o
f
3
,
0
0
0
im
ag
es
o
f
ca
p
s
icu
m
an
n
u
u
m
L
.
p
lan
ts
an
d
an
n
o
tated
in
to
th
r
ee
class
es
o
f
ch
i
l
li
lea
v
es
d
is
ea
s
es:
n
am
ely
th
e
u
p
w
ar
d
cu
r
lin
g
,
m
o
s
aic/m
o
ttli
n
g
an
d
b
ac
ter
ial
s
p
o
t.
T
h
e
im
ag
es
wer
e
s
elf
-
c
o
llected
u
n
d
er
an
u
n
c
o
n
tr
o
lle
d
en
v
ir
o
n
m
e
n
t
an
d
v
ar
io
u
s
illu
m
in
atio
n
s
,
v
iews,
a
n
d
d
is
tan
ce
s
to
r
ef
lect
th
e
r
ea
l
-
life
s
ce
n
ar
io
s
.
T
h
e
im
a
g
es
wer
e
co
llected
f
r
o
m
th
r
ee
d
if
f
er
e
n
t
f
ield
en
v
ir
o
n
m
en
ts
lo
ca
ted
at
Sij
an
g
k
an
g
Selan
g
o
r
;
C
o
m
m
u
n
ity
Ur
b
an
Fa
r
m
,
B
u
k
it
R
im
au
,
Selan
g
or
,
an
d
a
g
r
ee
n
h
o
u
s
e
at
th
e
Facu
lty
o
f
E
n
g
in
ee
r
in
g
,
Un
i
v
er
s
iti
Pu
tr
a
Ma
lay
s
ia
(
UPM)
Selan
g
o
r
.
T
h
e
im
ag
es
wer
e
c
ap
tu
r
ed
u
s
in
g
Ap
p
le
iPh
o
n
e
7
with
th
e
d
im
en
s
io
n
o
f
r
eso
lu
tio
n
3
0
2
4
×
4
0
3
2
an
d
Asu
s
Z
en
f
o
n
e
2
with
th
e
d
im
en
s
io
n
o
f
r
eso
lu
tio
n
2304
×
4
0
9
6
.
T
h
ese
im
ag
es
wer
e
cr
o
p
p
e
d
,
r
esized
an
d
f
lip
p
e
d
m
an
u
ally
u
s
in
g
Mic
r
o
s
o
f
t
Ph
o
to
s
at
th
e
in
itial
s
tag
e
to
r
ed
u
ce
th
e
b
ac
k
g
r
o
u
n
d
clu
tter
an
d
o
cc
lu
s
io
n
is
s
u
es.
Data
an
n
o
tatio
n
was
d
o
n
e
b
y
co
n
s
u
ltin
g
th
e
ex
p
er
ts
at
th
e
f
ar
m
s
an
d
cr
o
s
s
-
ch
ec
k
in
g
with
th
e
r
elate
d
p
u
b
lis
h
ed
p
ap
e
r
s
.
Fo
r
ea
ch
d
is
ea
s
e,
8
0
0
an
d
2
0
0
im
ag
es we
r
e
u
s
ed
f
o
r
tr
ain
in
g
an
d
test
in
g
,
r
esp
ec
tiv
ely
.
3
.
2
.
P
re
t
ra
ined DCN
N
m
o
del a
nd
pa
ra
m
et
er
s
I
n
th
is
s
tu
d
y
,
th
e
p
er
f
o
r
m
an
c
e
o
f
VGG1
6
,
I
n
ce
p
tio
n
V3
an
d
E
f
f
icien
tNet
B
0
in
class
if
y
in
g
ch
i
l
li
p
lan
t
d
is
ea
s
es
f
r
o
m
co
m
p
lex
im
ag
es
was
co
m
p
ar
ed
.
T
h
ese
m
o
d
els
wer
e
s
elec
ted
f
o
r
th
eir
o
u
ts
tan
d
in
g
p
er
f
o
r
m
an
ce
wh
en
class
if
ied
th
e
p
lan
t
d
is
ea
s
e
im
ag
es
f
r
o
m
th
e
I
m
ag
eNe
t
d
ataset
[
2
0
]
.
T
h
e
VGG1
6
,
as
illu
s
tr
ated
in
Fig
u
r
e
3
[
2
1
]
,
u
s
ed
a
r
ec
o
m
m
en
d
e
d
d
e
f
au
lt
in
p
u
t
im
ag
e
s
ize
o
f
2
2
4
×
2
2
4
×
3
a
n
d
1
3
co
n
v
o
l
u
tio
n
al
lay
er
s
with
a
r
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U)
ac
tiv
atio
n
f
u
n
cti
o
n
.
T
h
e
co
n
v
o
l
u
tio
n
al
lay
er
s
wer
e
f
ed
in
t
o
a
m
ax
p
o
o
lin
g
,
th
r
ee
f
u
lly
co
n
n
ec
ted
(
FC
)
lay
er
s
an
d
a
So
f
tm
ax
f
u
n
ctio
n
at
th
e
en
d
o
f
t
h
e
ar
ch
itectu
r
e.
T
h
e
last
FC
lay
er
s
wer
e
r
ep
lace
d
b
y
th
r
ee
ch
an
n
els
f
o
r
th
is
s
tu
d
y
,
in
d
icatin
g
th
e
th
r
ee
class
es
o
f
ch
i
l
li p
lan
t d
is
ea
s
es u
n
d
er
s
tu
d
y
.
Me
an
wh
ile,
I
n
ce
p
tio
n
V3
[
2
2
]
h
as
4
2
to
tal
d
ee
p
n
etwo
r
k
l
ay
er
s
with
a
g
r
id
s
ize
-
r
ed
u
ct
io
n
b
lo
c
k
b
etwe
en
th
e
in
ce
p
tio
n
m
o
d
u
le
s
b
lo
ck
s
an
d
o
n
e
au
x
iliar
y
cla
s
s
if
ier
at
th
e
th
ir
d
c
o
n
ca
ten
ate
d
tr
u
n
k
,
as
s
h
o
wn
in
Fig
u
r
e
4
.
T
h
e
r
ec
o
m
m
en
d
e
d
s
ize
o
f
th
e
in
p
u
t
im
ag
e
f
o
r
t
h
is
m
o
d
el
is
2
9
9
×2
9
9
×3
,
an
d
f
iv
e
co
n
v
o
lu
tio
n
al
lay
er
s
p
r
o
ce
s
s
th
e
in
p
u
t
im
a
g
e
with
two
m
ax
-
p
o
o
lin
g
lay
e
r
s
at
th
e
f
ir
s
t
s
tag
e.
T
h
en
,
a
s
er
ies
o
f
in
ce
p
tio
n
m
o
d
u
les
p
r
o
ce
s
s
th
e
in
p
u
t
b
ef
o
r
e
f
in
ally
p
er
f
o
r
m
in
g
clas
s
if
icatio
n
u
s
in
g
a
f
u
lly
co
n
n
ec
ted
lay
er
an
d
a
So
f
tm
ax
f
u
n
ctio
n
.
E
f
f
icien
tNet
[
2
0
]
is
a
c
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
ar
ch
itectu
r
e
with
a
co
m
p
o
u
n
d
s
ca
lin
g
m
eth
o
d
th
at
u
n
if
o
r
m
ly
s
ca
le
s
all
d
ep
th
,
wid
t
h
,
an
d
r
eso
lu
tio
n
d
im
en
s
io
n
s
to
p
u
r
s
u
e
b
etter
ac
cu
r
ac
y
a
n
d
ef
f
icien
cy
.
C
u
r
r
e
n
tly
,
th
er
e
ar
e
s
ev
en
v
er
s
io
n
s
o
f
E
f
f
ici
en
tNet
n
etwo
r
k
s
,
in
wh
ich
ea
ch
v
er
s
io
n
is
th
e
u
p
g
r
a
d
ed
v
er
s
io
n
o
f
th
e
p
r
ev
i
o
u
s
,
wh
ich
is
s
ca
led
f
r
o
m
th
e
E
f
f
icien
tNetB
0
b
aselin
e
u
s
in
g
d
if
f
er
en
t c
o
m
p
o
u
n
d
co
ef
f
icien
ts
.
T
h
e
E
f
f
icien
tN
etB
0
n
etwo
r
k
co
n
s
is
ts
o
f
a
co
n
v
o
lu
tio
n
al
lay
er
,
s
ev
er
a
l
m
o
b
ile
in
v
er
ted
b
o
ttlen
ec
k
co
n
v
o
l
u
tio
n
al
(
MB
C
o
n
v
)
lay
er
s
an
d
o
p
tim
izatio
n
lay
er
s
,
as
s
h
o
wn
in
Fig
u
r
e
5
.
T
h
e
r
ec
o
m
m
en
de
d
s
ize
o
f
th
e
in
p
u
t im
ag
e
f
o
r
th
is
m
o
d
el
is
2
2
4
×
2
2
4
×
3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
E
ffica
cy
o
f c
h
ili p
la
n
t d
is
ea
s
es c
la
s
s
ifica
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
:
a
p
r
elimin
a
r
y
s
tu
d
y
(
S
u
h
a
n
a
R
o
z
la
n
)
1445
Fig
u
r
e
3
.
T
h
e
ar
ch
itectu
r
e
o
f
t
h
e
VGG1
6
n
etwo
r
k
[
2
1
]
Fig
u
r
e
4
.
T
h
e
ar
ch
itectu
r
e
o
f
I
n
ce
p
tio
n
V3
n
etwo
r
k
[
2
3
]
Fig
u
r
e
5
.
T
h
e
ar
ch
itectu
r
e
o
f
E
f
f
icien
tNetB
0
n
etwo
r
k
[
2
3
]
3
.
3
.
E
x
perim
ent
a
l set
up
T
h
e
ex
p
er
im
e
n
t
was
co
n
d
u
ct
ed
o
n
a
6
4
-
b
it
o
p
er
atin
g
s
y
s
t
em
,
an
x
6
4
-
b
ased
p
r
o
ce
s
s
o
r
r
u
n
n
in
g
o
n
I
n
tel(
R
)
C
o
r
e
(
T
M)
i5
-
1
0
2
0
0
H
C
PU
@
2
.
4
0
GHz
with
NV
I
DI
A
GeFo
r
ce
GT
X
1
6
5
0
a
n
d
8
GB
R
AM
.
All
de
ep
lear
n
in
g
m
o
d
els
wer
e
co
m
p
iled
with
GPU
s
u
p
p
o
r
t.
T
h
e
p
r
o
p
o
s
ed
ch
i
l
li
p
lan
t
d
is
ea
s
e
clas
s
if
icatio
n
i
s
s
h
o
wn
in
Fig
u
r
e
6
.
T
h
e
f
ilter
s
,
f
ea
tu
r
e
m
ap
s
,
p
o
o
lin
g
lay
er
s
an
d
h
y
p
er
p
ar
am
eter
s
o
f
th
e
VGG1
6
,
I
n
ce
p
tio
n
V
3
an
d
E
f
f
icien
tNetB
0
m
o
d
els
r
em
ain
th
e
s
am
e
s
tr
u
ctu
r
e,
as
o
b
tain
ed
f
r
o
m
Ker
as
Ap
p
licatio
n
s
API
with
I
m
ag
eNe
t
[
2
4
]
.
Nev
er
th
eless
,
a
co
m
b
in
atio
n
o
f
f
u
lly
co
n
n
ec
ted
lay
er
s
an
d
So
f
tm
ax
ac
ti
v
atio
n
was
ap
p
lied
.
T
h
is
p
ar
t
h
as
b
ee
n
co
n
v
er
ted
in
to
th
r
ee
o
u
tp
u
ts
r
ep
r
esen
tin
g
th
e
ch
illi
p
lan
t
d
is
ea
s
e
class
es
(
u
p
war
d
c
u
r
lin
g
,
b
ac
ter
ial
s
p
o
t
an
d
m
o
s
aic/m
o
ttli
n
g
)
.
All
p
r
e
-
t
r
ain
ed
m
o
d
el
s
wer
e
s
et
to
f
r
o
ze
n
lay
er
s
to
av
o
id
Ke
r
as
f
r
o
m
u
p
d
atin
g
th
eir
weig
h
ts
d
u
r
in
g
th
e
tr
ain
i
n
g
.
Oth
er
s
ettin
g
s
i
n
clu
d
e
b
atch
n
o
r
m
aliza
tio
n
a
n
d
b
atch
s
ize.
B
ased
on
[
2
5
]
,
ea
c
h
p
ix
el
v
alu
e
o
f
t
h
e
im
ag
es
was
d
iv
id
ed
b
y
2
5
5
f
o
r
b
atch
n
o
r
m
aliza
tio
n
,
an
d
th
e
s
elec
ted
b
atch
s
ize
was
3
2
.
B
atch
n
o
r
m
aliza
tio
n
co
u
l
d
o
v
er
co
m
e
th
e
p
r
o
b
lem
o
f
i
n
ter
n
al
c
o
v
ar
iate
s
h
if
t,
wh
ich
ca
n
im
p
ed
e
th
e
tr
ain
in
g
o
f
d
ee
p
n
eu
r
al
n
et
wo
r
k
s
.
Sto
ch
asti
c
g
r
ad
ie
n
t
d
e
s
ce
n
t
(
SGD)
was
u
s
ed
as
th
e
o
p
tim
izer
d
u
e
t
o
its
h
ig
h
p
er
f
o
r
m
a
n
ce
[
2
6
]
,
wh
ile
th
e
lear
n
in
g
r
ate
o
f
0
.
0
0
0
1
wa
s
ad
o
p
ted
b
ased
o
n
[
1
6
]
.
T
h
e
ep
o
ch
is
s
et
to
5
0
,
an
d
th
e
s
elec
tio
n
is
b
ased
o
n
s
ev
er
al
tr
ials
,
s
u
ch
as
1
0
,
3
0
,
5
0
an
d
1
0
0
ep
o
ch
s
.
T
h
e
r
esu
lts
h
av
e
s
h
o
wn
th
at
5
0
ep
o
ch
s
h
av
e
p
r
o
d
u
ce
d
h
ig
h
ac
cu
r
ac
y
an
d
b
etter
p
r
o
ce
s
s
in
g
s
tab
ilit
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
25
,
No
.
3
,
Ma
r
ch
20
22
:
1
4
4
2
-
1
4
4
9
1446
T
h
e
in
p
u
t
im
ag
es
wer
e
d
iv
i
d
e
d
in
to
two
s
ets,
8
0
%
f
o
r
tr
ain
i
n
g
an
d
2
0
%
f
o
r
test
in
g
,
as
r
ec
o
m
m
en
d
e
d
by
[
2
7
]
.
T
h
e
im
ag
es
wer
e
r
esized
ac
co
r
d
in
g
to
th
e
m
o
d
el
’
s
d
ef
au
lt
s
ize,
2
2
4
×2
2
4
p
ix
e
ls
f
o
r
VGG1
6
a
n
d
E
f
f
icien
tNetB
0
,
an
d
2
9
9
×2
9
9
p
ix
els
f
o
r
I
n
ce
p
tio
n
V3
.
T
h
e
m
eth
o
d
s
wer
e
tr
ain
e
d
with
two
tr
ain
in
g
s
ets,
wh
er
e
th
e
f
ir
s
t
s
et
co
n
s
is
ts
o
f
o
r
ig
in
al
im
ag
es
an
d
th
e
s
ec
o
n
d
s
etco
n
s
is
ts
o
f
au
g
m
en
ted
im
ag
es.
B
o
th
s
ets
co
n
s
is
t o
f
th
e
s
am
e
am
o
u
n
t o
f
im
ag
es th
at
is
2
4
0
0
im
ag
es.
I
n
th
e
s
ec
o
n
d
tr
ain
in
g
s
et,
th
e
im
ag
es we
r
e
s
h
ea
r
ed
at
an
an
g
le
o
f
0
.
2
d
eg
r
ee
s
,
z
o
o
m
ed
at
0
.
2
m
a
g
n
if
icatio
n
an
d
h
o
r
iz
o
n
tal
f
lip
p
ed
u
s
in
g
I
m
ag
ed
atag
en
er
at
o
r
in
Ker
asap
p
licatio
n
.
I
m
a
g
e
d
ata
g
en
er
ato
r
wo
r
k
s
r
an
d
o
m
ly
in
r
ea
l
-
tim
e
,
with
th
e
n
u
m
b
er
o
f
im
ag
es
r
em
ai
n
in
g
th
e
s
am
e.
T
h
e
au
g
m
en
tatio
n
p
ar
am
eter
s
s
elec
tio
n
was
d
ec
i
d
ed
b
ased
o
n
th
e
o
b
s
er
v
atio
n
f
r
o
m
a
f
ew
tr
ials
,
wh
er
e
th
e
f
ea
tu
r
es o
f
th
e
d
is
ea
s
e
ca
n
b
e
v
is
u
alize
d
u
s
in
g
th
ese
p
ar
am
eter
s
.
Fig
u
r
e
6
.
T
h
e
p
r
o
p
o
s
ed
c
h
ili p
lan
t d
is
ea
s
e
class
if
icat
io
n
f
r
a
m
ewo
r
k
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
elec
ted
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
was
ev
alu
ated
b
ased
o
n
ac
cu
r
ac
y
,
r
ec
all,
p
r
ec
is
io
n
,
an
d
F1
-
s
co
r
e
.
A
cc
u
r
ac
y
is
th
e
n
u
m
b
e
r
o
f
co
r
r
ec
tly
id
en
tifie
d
s
am
p
les
,
an
d
r
ec
a
ll
is
th
e
n
u
m
b
e
r
o
f
p
o
s
itiv
e
s
am
p
les
th
at
ar
e
ac
cu
r
ately
id
en
tifie
d
.
Me
an
w
h
ile,
p
r
ec
is
io
n
is
th
e
m
ea
s
u
r
em
en
t
o
f
ac
cu
r
atel
y
id
en
tifie
d
s
am
p
les
am
o
n
g
al
l
th
e
tr
u
e
s
am
p
les,
a
n
d
th
e
F1
-
s
co
r
er
ep
r
esen
ts
a
h
ar
m
o
n
ic
m
ea
n
b
etwe
en
s
en
s
itiv
ity
an
d
p
r
ec
is
io
n
.
T
h
e
ex
p
er
im
en
ts
wer
e
co
n
d
u
cted
o
n
two
d
atasets
,
wh
er
e
th
e
f
ir
s
t
d
ataset
co
n
s
is
t
o
f
o
r
ig
in
al
im
a
g
es
an
d
au
g
m
en
t
ed
im
ag
es
i
n
th
e
s
ec
o
n
d
d
ata
s
et.
T
h
e
r
esu
lts
in
Fig
u
r
e
7
(
a
)
,
Fig
u
r
e
7
(
b
)
an
d
Fig
u
r
e
7
(
c)
s
h
o
w
th
at
E
f
f
icien
tNetB
0
o
u
tp
er
f
o
r
m
e
d
VGG1
6
an
d
I
n
ce
p
tio
n
V3
,
b
u
t
in
Fig
u
r
e
8
(
a)
,
Fig
u
r
e
8
(
b
)
an
d
Fig
u
r
e
8
(
c)
,
it
is
s
h
o
wn
t
h
at
I
n
ce
p
tio
n
V3
o
u
tp
e
r
f
o
r
m
e
d
VGG1
6
a
n
d
E
f
f
icien
tNetB
0
.
I
t
is
also
o
b
s
er
v
ed
th
at
f
o
r
b
o
th
ca
s
es,
VGG1
6
a
n
d
I
n
ce
p
tio
n
V3
r
eq
u
ir
ed
less
co
m
p
u
t
atio
n
tim
e
co
m
p
ar
ed
t
o
E
f
f
icien
tNetB
0
to
r
ea
ch
ab
o
v
e
9
0
%.
(
a)
(
b
)
(
c)
Fig
u
r
e
7
.
T
h
e
ac
cu
r
ac
y
p
r
o
d
u
c
ed
b
y
(
a)
VGG1
6
,
(
b
)
I
n
ce
p
tio
n
V3
,
an
d
(
c)
E
f
f
icien
tNetB
0
u
s
in
g
o
r
ig
in
al
im
ag
es f
o
r
tr
ain
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
E
ffica
cy
o
f c
h
ili p
la
n
t d
is
ea
s
es c
la
s
s
ifica
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
:
a
p
r
elimin
a
r
y
s
tu
d
y
(
S
u
h
a
n
a
R
o
z
la
n
)
1447
(
a)
(
b
)
(
c)
Fig
u
r
e
8
.
T
h
e
ac
cu
r
ac
y
p
r
o
d
u
c
ed
b
y
(
a)
VGG1
6
,
(
b
)
I
n
ce
p
tio
n
V3
,
an
d
(
c)
E
f
f
icien
tNetB
0
u
s
in
g
au
g
m
en
te
d
im
ag
es f
o
r
tr
ain
i
n
g
Nev
er
th
eless
,
th
e
ac
cu
r
ac
y
o
f
m
o
d
els tr
ain
ed
b
y
o
r
ig
i
n
al
im
a
g
es d
id
n
o
t d
i
f
f
er
m
u
ch
wh
en
tr
ain
ed
o
n
au
g
m
en
ted
im
a
g
es.
At
th
e
5
0
ep
o
ch
s
,
it
is
s
h
o
wn
th
at
th
e
I
n
ce
p
tio
n
V3
h
as
p
r
o
d
u
ce
d
t
h
e
h
ig
h
est
ac
cu
r
ac
y
o
f
9
8
.
8
3
%
o
n
au
g
m
en
ted
im
a
g
es
wh
ile
E
f
f
icien
tNetB
0
h
ig
h
est
ac
cu
r
ac
y
o
f
9
7
.
6
7
%
o
n
o
r
ig
in
al
im
a
g
es,
r
esp
ec
tiv
ely
.
T
h
e
o
b
tain
ed
a
cc
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F1
-
s
co
r
e
f
o
r
th
e
VGG1
6
,
I
n
ce
p
tio
n
V
3
an
d
E
f
f
icien
tNetB
0
ar
e
s
h
o
wn
in
T
ab
le
1
.
I
t
is
s
h
o
wn
th
at
th
e
p
er
f
o
r
m
a
n
ce
o
f
E
f
f
icien
tNetB
0
an
d
I
n
ce
p
tio
n
V3
is
ap
p
r
o
x
im
ately
s
im
ilar
f
o
r
b
o
t
h
ca
s
es,
th
at
is
,
wh
e
n
o
r
ig
in
al
o
r
au
g
m
en
ted
im
ag
es
tr
ai
n
th
e
m
o
d
els.
H
o
wev
er
,
E
f
f
icien
tNetB
0
h
as
b
etter
p
er
f
o
r
m
an
ce
wh
en
tr
ain
e
d
b
y
o
r
ig
in
al
im
ag
es,
b
u
t
o
p
p
o
s
ite
to
I
n
ce
p
tio
n
V3
,
th
e
m
o
d
el
p
er
f
o
r
m
e
d
b
etter
w
h
en
tr
ain
ed
b
y
au
g
m
e
n
ted
im
a
g
es.
T
h
e
s
u
m
m
ar
y
o
f
th
e
class
if
icatio
n
r
esu
lts
f
o
r
ea
ch
d
i
s
ea
s
e,
n
am
ely
u
p
war
d
c
u
r
li
n
g
(
UC
)
,
m
o
s
aic/m
o
ttli
n
g
(
M)
an
d
b
ac
ter
ial
s
p
o
t
(
B
S)
d
is
ea
s
es,
ar
e
s
h
o
wn
in
th
e
co
n
f
u
s
io
n
m
atr
i
x
in
T
a
b
le
2
.
I
t
is
d
em
o
n
s
tr
ated
th
at
I
n
ce
p
tio
n
V3
h
as
p
r
o
d
u
ce
d
th
e
h
ig
h
est
tr
u
e
p
o
s
itiv
ev
alu
e
wh
e
n
class
if
y
in
g
b
ac
ter
ial
s
p
o
ts
f
o
r
b
o
th
ca
s
es
b
u
t
th
e
m
o
s
aic/m
o
ttli
n
g
,
th
e
I
n
ce
p
tio
n
V
3
m
o
d
el,
h
as
th
e
h
ig
h
est
tr
u
e
p
o
s
itiv
e
v
alu
e
wh
en
tr
ain
ed
b
y
a
u
g
m
en
te
d
im
ag
es.
Nev
er
th
eless
,
th
e
tr
u
e
p
o
s
itiv
e
v
alu
e
o
f
th
e
I
n
ce
p
tio
n
V3
m
o
d
el
is
lo
wer
th
a
n
th
e
tr
u
e
p
o
s
itiv
e
v
alu
e
o
f
E
f
f
i
cien
tNetB
0
wh
en
tr
ain
e
d
b
y
o
r
ig
in
al
im
ag
es
.
All
th
r
ee
m
eth
o
d
s
h
av
e
eq
u
iv
alen
t
p
er
f
o
r
m
an
ce
wh
en
class
if
y
in
g
th
e
u
p
war
d
c
u
r
lin
g
d
is
ea
s
e.
I
t
is
also
s
h
o
wn
t
h
at
th
e
m
o
d
els
h
av
e
d
if
f
icu
lty
class
if
y
in
g
th
e
m
o
s
aic/m
o
ttli
n
g
d
is
ea
s
e,
as
th
e
p
r
o
d
u
ce
d
tr
u
e
p
o
s
itiv
e
v
alu
e
is
th
e
lo
west
co
m
p
ar
ed
to
o
t
h
er
d
is
ea
s
es.
T
h
e
co
m
p
lex
f
ea
tu
r
es
o
f
th
e
m
o
s
aic/m
o
ttli
n
g
d
i
s
ea
s
e
ar
e
ap
p
r
o
x
im
ately
s
im
ilar
to
th
e
u
p
war
d
cu
r
lin
g
d
is
ea
s
e
th
at
ca
u
s
ed
th
e
m
eth
o
d
s
to
m
is
class
if
y
.
T
ab
le
1
.
T
h
e
VGG1
6
,
I
n
ce
p
tio
n
V3
an
d
E
f
f
icien
tNetB
0
test
in
g
p
er
f
o
r
m
an
ce
Tr
a
i
n
i
n
g
S
e
t
M
o
d
e
l
s
A
c
c
u
r
a
c
y
(
%)
P
r
e
c
i
s
i
o
n
(
%)
R
e
c
a
l
l
(
%)
F
1
S
c
o
r
e
F
i
r
st
s
e
t
(
o
r
i
g
i
n
a
l
i
ma
g
e
s)
V
G
G
1
6
9
5
.
1
7
9
5
.
0
0
9
5
.
0
0
0
.
9
5
I
n
c
e
p
t
i
o
n
V
3
9
7
.
5
0
9
7
.
0
0
9
7
.
0
0
0
.
9
7
E
f
f
i
c
i
e
n
t
N
e
t
B
0
9
7
.
6
7
9
8
.
0
0
9
8
.
0
0
0
.
9
8
S
e
c
o
n
d
s
e
t
(
a
u
g
me
n
t
e
d
i
ma
g
e
s)
V
G
G
1
6
9
5
.
8
3
9
6
.
0
0
9
6
.
0
0
0
.
9
6
In
c
e
p
t
i
o
n
V
3
9
8
.
8
3
9
9
.
0
0
9
9
.
0
0
0
.
9
9
Ef
f
i
c
i
e
n
t
N
e
t
B
0
9
6
.
8
3
9
7
.
0
0
9
7
.
0
0
0
.
9
7
T
ab
le
2
.
C
o
n
f
u
s
io
n
Ma
tr
ix
o
f
t
h
e
m
eth
o
d
s
th
at
wer
e
tr
ain
ed
u
s
in
g
o
r
ig
in
al
an
d
a
u
g
m
e
n
ted
im
ag
es
F
i
r
st
Tr
a
i
n
i
n
g
S
e
t
(
O
r
i
g
i
n
a
l
I
mag
e
s)
S
e
c
o
n
d
Tr
a
i
n
i
n
g
S
e
t
(
A
u
g
me
n
t
e
d
I
mag
e
s)
V
G
G
1
6
I
n
c
e
p
t
i
o
n
V
3
Ef
f
i
c
i
e
n
t
N
e
t
B
0
V
G
G
1
6
I
n
c
e
p
t
i
o
n
V
3
Ef
f
i
c
i
e
n
t
N
e
t
B
0
Tr
u
e
La
b
e
l
BS
0
.
9
7
0
.
0
2
0
.
0
1
0
.
9
9
0
.
0
0
.
1
0
.
9
8
0
.
0
0
.
0
2
0
.
9
8
0
.
0
2
0
.
0
1
.
0
0
.
0
0
.
0
0
.
9
8
0
.
0
0
.
0
2
M
0
.
0
1
0
.
9
1
0
.
0
8
0
.
0
2
0
.
9
6
0
.
0
2
0
.
0
0
.
9
8
0
.
0
2
0
.
0
2
0
.
9
0
0
.
0
8
0
.
0
0
.
9
8
0
.
0
2
0
.
0
0
.
9
4
0
.
0
6
UC
0
.
0
0
.
0
2
0
.
9
8
0
.
0
0
.
0
3
0
.
9
7
0
.
0
0
.
0
2
0
.
9
8
0
.
0
0
.
0
2
0
.
9
8
0
.
0
0
.
0
2
0
.
9
8
0
.
0
0
.
0
2
0
.
9
8
BS
M
UC
BS
M
UC
BS
M
UC
BS
M
UC
BS
M
UC
BS
M
UC
P
r
e
d
i
c
t
e
d
La
b
e
l
5.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
S
T
h
e
ef
f
icac
y
o
f
d
ee
p
lear
n
i
n
g
alg
o
r
ith
m
s
,
n
am
ely
VGG1
6
,
I
n
ce
p
tio
n
V3
a
n
d
E
f
f
icie
n
tNet
B
0
,
m
eth
o
d
s
in
class
if
y
in
g
th
r
ee
ty
p
es
o
f
ch
i
l
li
p
lan
t
d
is
ea
s
e
s
,
n
am
ely
u
p
war
d
cu
r
lin
g
,
m
o
s
aic/m
o
ttli
n
g
an
d
b
ac
ter
ial
s
p
o
t
f
r
o
m
a
d
ataset
th
at
co
n
s
is
ts
o
f
3
,
0
0
0
im
ag
es
in
an
u
n
co
n
tr
o
lled
co
n
d
itio
n
an
d
u
n
d
er
v
ar
io
u
s
im
ag
in
g
co
n
d
itio
n
s
is
d
em
o
n
s
tr
ated
.
T
h
e
ex
p
er
im
en
t
r
esu
lts
s
h
o
wed
th
at
I
n
ce
p
tio
n
V3
h
as
b
etter
p
er
f
o
r
m
an
ce
th
an
th
e
VGG1
6
an
d
E
f
f
i
cien
t
NetB0
in
clas
s
if
y
in
g
b
ac
ter
ial
s
p
o
t
im
ag
es
.
Sti
ll,
all
th
e
m
o
d
els
h
av
e
d
if
f
icu
lty
class
if
y
in
g
th
e
m
o
s
aic/m
o
ttli
n
g
d
is
ea
s
e
d
u
e
to
th
e
c
o
m
p
lex
f
ea
tu
r
es
o
f
th
e
m
o
s
aic/m
o
ttli
n
g
d
is
ea
s
e
th
at
ar
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
25
,
No
.
3
,
Ma
r
ch
20
22
:
1
4
4
2
-
1
4
4
9
1448
ap
p
r
o
x
im
ately
s
im
ilar
to
th
e
u
p
war
d
c
u
r
lin
g
d
is
ea
s
e.
All
th
r
ee
m
eth
o
d
s
h
av
e
eq
u
i
v
alen
t
p
er
f
o
r
m
an
ce
w
h
en
class
if
y
in
g
th
e
u
p
war
d
cu
r
lin
g
d
is
ea
s
e.
I
n
co
n
clu
s
io
n
,
d
ee
p
l
ea
r
n
in
g
alg
o
r
ith
m
s
h
av
e
s
h
o
w
n
a
g
r
ea
t
p
o
ten
tial
f
o
r
class
if
y
in
g
c
h
i
l
li
p
lan
t
d
is
ea
s
es.
T
h
e
p
er
f
o
r
m
an
ce
alg
o
r
i
th
m
s
ca
n
b
e
f
u
r
t
h
er
im
p
r
o
v
e
d
b
y
e
x
p
o
s
in
g
th
e
m
to
h
ig
h
-
c
o
m
p
lex
ity
im
a
g
es
an
d
s
ev
er
al
o
th
e
r
ty
p
es
o
f
d
is
ea
s
es,
wh
ich
will
cr
ea
te
m
o
r
e
o
p
p
o
r
tu
n
ities
f
o
r
d
ev
elo
p
in
g
m
o
r
e
ad
v
a
n
ce
d
cla
s
s
if
ier
s
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
au
th
o
r
s
wis
h
to
ac
k
n
o
wl
ed
g
e
th
e
f
in
a
n
cial
s
u
p
p
o
r
t
r
e
ce
iv
ed
f
r
o
m
th
e
Fu
n
d
am
en
tal
R
esear
ch
Gr
an
t
Sch
em
e
(
FR
GS)
,
Min
is
tr
y
o
f
Hig
h
er
E
d
u
ca
tio
n
,
with
g
r
an
t
n
u
m
b
er
5
5
4
0
0
7
8
(
I
D:
1
5
9
4
8
a
n
d
r
ef
er
en
ce
co
d
e:
FR
GS/1
/2
0
1
8
/W
AB
0
1
/UPM/0
2
/3
0
)
.
T
h
e
au
t
h
o
r
s
wo
u
ld
also
wan
t
to
e
x
ten
d
ap
p
r
e
ciatio
n
to
Mr
s
.
No
o
r
Azlin
a
Ab
d
u
l A
ziz
an
d
C
u
i H
ai
ru
f
o
r
th
eir
s
u
p
p
o
r
t in
th
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
S
.
Lo
b
,
M
.
N
.
M
.
A
r
i
s
,
S
.
N
.
M
.
S
i
d
i
q
u
e
,
N
.
F
.
I
b
r
a
h
i
m,
a
n
d
X
.
Ji
n
,
“
G
r
o
w
t
h
d
e
v
e
l
o
p
me
n
t
a
n
d
n
a
t
u
r
a
l
i
n
f
e
c
t
i
o
n
i
n
c
i
d
e
n
c
e
o
f
t
o
b
a
c
c
o
m
o
sa
i
c
v
i
r
u
s
(
T
M
V
)
o
n
si
l
i
c
o
n
-
t
r
e
a
t
e
d
c
h
i
l
l
i
(
c
a
p
si
c
u
m
a
n
n
u
u
m
L.
)
c
u
l
t
i
v
a
t
e
d
i
n
c
o
mm
e
r
c
i
a
l
s
o
i
l
,
”
M
a
l
a
y
s
i
a
n
Ap
p
l
i
e
d
Bi
o
l
o
g
y
,
v
o
l
.
4
6
,
n
o
.
3
,
p
p
.
2
2
1
–
2
2
6
,
2
0
1
7
.
[
2
]
D
e
p
a
r
t
me
n
t
o
f
S
t
a
t
i
s
t
i
c
s
M
a
l
a
y
s
i
a
,
“
P
r
e
ss
R
e
l
e
a
se
S
u
p
p
l
y
a
n
d
U
t
i
l
i
z
a
t
i
o
n
A
c
c
o
u
n
t
s
S
e
l
e
c
t
e
d
A
g
r
i
c
u
l
t
u
r
a
l
C
o
mm
o
d
i
t
i
e
s,”
D
e
p
a
rt
m
e
n
t
o
f
S
t
a
t
i
s
t
i
c
s
,
n
o
.
D
e
c
e
m
b
e
r
,
p
p
.
2
0
1
3
–
2
0
1
5
,
2
0
2
1
.
[
3
]
A
.
N
o
r
f
a
d
z
i
l
a
h
,
“
S
c
r
e
e
n
i
n
g
o
f
S
e
l
e
c
t
e
d
C
u
c
u
m
b
e
r
M
o
s
a
i
c
V
i
r
u
s
R
e
s
i
st
a
n
c
e
C
h
i
l
l
i
G
e
n
o
t
y
p
e
s
O
b
t
a
i
n
e
d
f
r
o
m
W
o
r
l
d
V
e
g
e
t
a
b
l
e
C
e
n
t
e
r
f
o
r
H
i
g
h
Y
i
e
l
d
i
n
M
a
l
a
y
si
a
n
C
o
n
d
i
t
i
o
n
,
”
2
0
1
8
.
[
4
]
N
.
Y
u
s
u
f
,
S
.
M
.
F
a
z
i
,
N
.
A
.
A
l
i
,
a
n
d
N
.
I
.
F
a
u
z
i
,
“
Ef
f
e
c
t
s
o
f
C
o
l
l
e
t
o
t
r
i
c
h
u
m
c
a
p
si
c
i
i
n
f
e
c
t
i
o
n
o
n
t
h
e
g
r
o
w
t
h
a
n
d
a
n
t
i
o
x
i
d
a
t
i
v
e
r
e
sp
o
n
s
e
o
n
d
e
f
e
n
se
mec
h
a
n
i
sms
o
f
C
a
p
s
i
c
u
m
a
n
n
u
u
m
,
”
2
0
1
6
.
[
5
]
M
.
P
a
r
i
s
i
,
D
.
A
l
i
o
t
o
,
a
n
d
P
.
Tr
i
p
o
d
i
,
“
O
v
e
r
v
i
e
w
o
f
b
i
o
t
i
c
s
t
r
e
sses
i
n
p
e
p
p
e
r
(
C
a
p
s
i
c
u
m
s
p
p
.
)
:
S
o
u
r
c
e
s
o
f
g
e
n
e
t
i
c
r
e
s
i
st
a
n
c
e
,
mo
l
e
c
u
l
a
r
b
r
e
e
d
i
n
g
a
n
d
g
e
n
o
m
i
c
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Mo
l
e
c
u
l
a
r
S
c
i
e
n
c
e
s
,
v
o
l
.
2
1
,
n
o
.
7
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
i
j
ms
2
1
0
7
2
5
8
7
.
[
6
]
K
.
P
u
s
h
p
a
n
a
t
h
a
n
,
M
.
H
a
n
a
f
i
,
S
.
M
a
s
h
o
h
o
r
,
a
n
d
W
.
F
.
F
a
z
l
i
l
I
l
a
h
i
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
m
e
d
i
c
i
n
a
l
p
l
a
n
t
s
r
e
c
o
g
n
i
t
i
o
n
:
a
r
e
v
i
e
w
,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
Re
v
i
e
w
,
v
o
l
.
5
4
,
n
o
.
1
,
p
p
.
305
–
3
2
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
6
2
-
0
2
0
-
0
9
8
4
7
-
0.
[
7
]
A
.
P
i
c
o
n
,
M
.
S
e
i
t
z
,
A
.
A
l
v
a
r
e
z
-
G
i
l
a
,
P
.
M
o
h
n
k
e
,
A
.
O
r
t
i
z
-
B
a
r
r
e
d
o
,
a
n
d
J
.
E
c
h
a
z
a
r
r
a
,
“
C
r
o
p
c
o
n
d
i
t
i
o
n
a
l
C
o
n
v
o
l
u
t
i
o
n
a
l
N
e
u
r
a
l
N
e
t
w
o
r
k
s
f
o
r
mass
i
v
e
m
u
l
t
i
-
c
r
o
p
p
l
a
n
t
d
i
sea
se
c
l
a
ssi
f
i
c
a
t
i
o
n
o
v
e
r
c
e
l
l
p
h
o
n
e
a
c
q
u
i
r
e
d
i
ma
g
e
s
t
a
k
e
n
o
n
r
e
a
l
f
i
e
l
d
c
o
n
d
i
t
i
o
n
s,
”
C
o
m
p
u
t
e
rs
a
n
d
E
l
e
c
t
r
o
n
i
c
s
i
n
A
g
r
i
c
u
l
t
u
r
e
,
v
o
l
.
1
6
7
,
n
o
.
S
e
p
t
e
m
b
e
r
,
p
.
1
0
5
0
9
3
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
a
g
.
2
0
1
9
.
1
0
5
0
9
3
.
[
8
]
R
.
I
.
H
a
sa
n
,
S
.
M
.
Y
u
s
u
f
,
a
n
d
L.
A
l
z
u
b
a
i
d
i
,
“
R
e
v
i
e
w
o
f
t
h
e
s
t
a
t
e
o
f
t
h
e
a
r
t
o
f
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
p
l
a
n
t
d
i
s
e
a
s
e
s:
A
b
r
o
a
d
a
n
a
l
y
s
i
s
a
n
d
d
i
s
c
u
ss
i
o
n
,
”
Pl
a
n
t
s
,
v
o
l
.
9
,
n
o
.
1
0
,
p
p
.
1
–
2
5
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
p
l
a
n
t
s
9
1
0
1
3
0
2
.
[
9
]
J.
L
i
u
a
n
d
X
.
W
a
n
g
,
“
P
l
a
n
t
d
i
s
e
a
se
s
a
n
d
p
e
s
t
s
d
e
t
e
c
t
i
o
n
b
a
se
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
:
a
r
e
v
i
e
w
,
”
Pl
a
n
t
M
e
t
h
o
d
s
,
v
o
l
.
1
7
,
n
o
.
1
,
p
p
.
1
–
1
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
3
0
0
7
-
0
2
1
-
0
0
7
2
2
-
9.
[
1
0
]
Ü
.
A
t
i
l
a
,
M
.
U
ç
a
r
,
K
.
A
k
y
o
l
,
a
n
d
E
.
U
ç
a
r
,
“
P
l
a
n
t
l
e
a
f
d
i
sea
se
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
Ef
f
i
c
i
e
n
t
N
e
t
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
,
”
E
c
o
l
o
g
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
6
1
,
n
o
.
O
c
t
o
b
e
r
2
0
2
0
,
p
.
1
0
1
1
8
2
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
c
o
i
n
f
.
2
0
2
0
.
1
0
1
1
8
2
.
[
1
1
]
J.
C
h
e
n
,
J
.
C
h
e
n
,
D
.
Zh
a
n
g
,
Y
.
S
u
n
,
a
n
d
Y
.
A
.
N
a
n
e
h
k
a
r
a
n
,
“
U
si
n
g
d
e
e
p
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
f
o
r
i
ma
g
e
-
b
a
sed
p
l
a
n
t
d
i
s
e
a
s
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
C
o
m
p
u
t
e
rs
a
n
d
El
e
c
t
ro
n
i
c
s
i
n
A
g
ri
c
u
l
t
u
r
e
,
v
o
l
.
1
7
3
,
n
o
.
A
p
r
i
l
,
p
.
1
0
5
3
9
3
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
a
g
.
2
0
2
0
.
1
0
5
3
9
3
.
[
1
2
]
A
.
R
a
m
d
a
n
,
A
.
H
e
r
y
a
n
a
,
A
.
A
r
i
sal
,
R
.
B
.
S
.
K
u
su
m
o
,
a
n
d
H
.
F
.
P
a
r
d
e
d
e
,
“
T
r
a
n
sf
e
r
Le
a
r
n
i
n
g
a
n
d
F
i
n
e
-
T
u
n
i
n
g
f
o
r
D
e
e
p
L
e
a
r
n
i
n
g
-
B
a
se
d
Te
a
D
i
se
a
ses
D
e
t
e
c
t
i
o
n
o
n
S
m
a
l
l
D
a
t
a
se
t
s,”
Pr
o
c
e
e
d
i
n
g
-
2
0
2
0
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Ra
d
a
r,
A
n
t
e
n
n
a
,
Mi
c
r
o
w
a
v
e
,
El
e
c
t
r
o
n
i
c
s
a
n
d
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
,
I
C
RAM
ET
2
0
2
0
,
p
p
.
2
0
6
–
2
1
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
R
A
M
ET
5
1
0
8
0
.
2
0
2
0
.
9
2
9
8
5
7
5
.
[
1
3
]
A
.
R
.
Lu
a
i
b
i
,
T.
M
.
S
a
l
ma
n
,
a
n
d
A
.
H
.
M
i
r
y
,
“
D
e
t
e
c
t
i
o
n
o
f
c
i
t
r
u
s
l
e
a
f
d
i
s
e
a
ses
u
si
n
g
a
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
1
,
n
o
.
2
,
p
p
.
1
7
1
9
–
1
7
2
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
1
i
2
.
p
p
1
7
1
9
-
1
7
2
7
.
[
1
4
]
W
.
Z
h
u
,
B
.
B
r
a
u
n
,
L.
H
.
C
h
i
a
n
g
,
a
n
d
J.
A
.
R
o
m
a
g
n
o
l
i
,
“
I
n
v
e
st
i
g
a
t
i
o
n
o
f
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
f
o
r
i
ma
g
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
n
d
i
m
p
a
c
t
on
t
r
a
i
n
i
n
g
s
a
m
p
l
e
s
i
z
e
,
”
C
h
e
m
o
m
e
t
r
i
c
s
a
n
d
I
n
t
e
l
l
i
g
e
n
t
L
a
b
o
r
a
t
o
r
y
S
y
st
e
m
s
,
v
o
l
.
2
1
1
,
n
o
.
F
e
b
r
u
a
r
y
,
p
.
1
0
4
2
6
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
h
e
m
o
l
a
b
.
2
0
2
1
.
1
0
4
2
6
9
.
[
1
5
]
S
.
N
i
u
,
Y
.
L
i
u
,
J.
W
a
n
g
,
a
n
d
H
.
S
o
n
g
,
“
A
D
e
c
a
d
e
S
u
r
v
e
y
o
f
Tr
a
n
sf
e
r
Le
a
r
n
i
n
g
(
2
0
1
0
–
2
0
2
0
)
,
”
I
EE
E
T
ra
n
s
a
c
t
i
o
n
s
o
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
,
n
o
.
2
,
p
p
.
1
5
1
–
1
6
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
t
a
i
.
2
0
2
1
.
3
0
5
4
6
0
9
.
[
1
6
]
Y
.
W
u
,
X
.
Q
i
n
,
Y
.
P
a
n
,
a
n
d
C
.
Y
u
a
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
b
a
s
e
d
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
f
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
f
l
o
w
e
r
s,”
2
0
1
8
I
EEE
3
rd
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
i
g
n
a
l
a
n
d
I
m
a
g
e
Pro
c
e
ssi
n
g
,
I
C
S
I
P
2
0
1
8
,
p
p
.
5
6
2
–
5
6
6
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
S
I
P
R
O
C
ESS
.
2
0
1
8
.
8
6
0
0
5
3
6
.
[
1
7
]
C
.
T
a
n
,
F
.
S
u
n
,
T.
K
o
n
g
,
W
.
Zh
a
n
g
,
C
.
Y
a
n
g
,
a
n
d
C
.
L
i
u
,
“
A
s
u
r
v
e
y
o
n
d
e
e
p
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
,
”
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
ser
i
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
1
1
4
1
LN
C
S
,
p
p
.
2
7
0
–
2
7
9
,
2
0
1
8
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
0
1
4
2
4
-
7
_
2
7
.
[
1
8
]
T.
P
u
r
w
a
n
i
n
g
s
i
h
,
I
.
A
.
A
n
j
a
n
i
,
a
n
d
P
.
B
.
U
t
a
m
i
,
“
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
mp
l
e
me
n
t
a
t
i
o
n
f
o
r
C
h
i
l
i
C
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
Pro
c
e
e
d
i
n
g
-
2
0
1
8
I
n
t
e
r
n
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
o
n
A
d
v
a
n
c
e
d
I
n
t
e
l
l
i
g
e
n
t
I
n
f
o
r
m
a
t
i
c
s:
R
e
v
o
l
u
t
i
o
n
i
ze
I
n
t
e
l
l
i
g
e
n
t
I
n
f
o
r
m
a
t
i
c
s
S
p
e
c
t
ru
m
f
o
r
H
u
m
a
n
i
t
y
,
S
AI
N
2
0
1
8
,
p
p
.
1
9
0
–
1
9
4
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
S
A
I
N
.
2
0
1
8
.
8
6
7
3
3
7
3
.
[
1
9
]
A
.
R
.
B
a
h
t
i
a
r
,
P
r
a
n
o
w
o
,
A
.
J
.
S
a
n
t
o
s
o
,
a
n
d
J.
Ju
h
a
r
i
a
h
,
“
D
e
e
p
L
e
a
r
n
i
n
g
D
e
t
e
c
t
e
d
N
u
t
r
i
e
n
t
D
e
f
i
c
i
e
n
c
y
i
n
C
h
i
l
i
P
l
a
n
t
,
”
2
0
2
0
8
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
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
y
,
I
C
o
I
C
T
2
0
2
0
,
p
p
.
5
–
8
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
o
I
C
T4
9
3
4
5
.
2
0
2
0
.
9
1
6
6
2
2
4
.
[
2
0
]
M
.
T
a
n
a
n
d
Q
.
V
.
L
e
,
“
Ef
f
i
c
i
e
n
t
N
e
t
:
R
e
t
h
i
n
k
i
n
g
mo
d
e
l
sc
a
l
i
n
g
f
o
r
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,
”
3
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
,
I
C
ML
2
0
1
9
,
v
o
l
.
2
0
1
9
-
Ju
n
e
,
p
p
.
1
0
6
9
1
–
1
0
7
0
0
,
2
0
1
9
.
[
2
1
]
F
.
L
i
u
,
Y
.
W
a
n
g
,
F
.
C
.
W
a
n
g
,
Y
.
Z
.
Zh
a
n
g
,
a
n
d
J.
Li
n
,
“
I
n
t
e
l
l
i
g
e
n
t
a
n
d
s
e
c
u
r
e
c
o
n
t
e
n
t
-
b
a
s
e
d
i
ma
g
e
r
e
t
r
i
e
v
a
l
f
o
r
mo
b
i
l
e
u
s
e
r
s,”
I
EEE
A
c
c
e
ss
,
v
o
l
.
7
,
p
p
.
1
1
9
2
0
9
–
1
1
9
2
2
2
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
3
5
2
2
2
.
[
2
2
]
C
.
S
z
e
g
e
d
y
,
V
.
V
a
n
h
o
u
c
k
e
,
S
.
I
o
f
f
e
,
J.
S
h
l
e
n
s,
a
n
d
Z
.
W
o
j
n
a
,
“
R
e
t
h
i
n
k
i
n
g
t
h
e
I
n
c
e
p
t
i
o
n
A
r
c
h
i
t
e
c
t
u
r
e
f
o
r
C
o
mp
u
t
e
r
V
i
si
o
n
,
”
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
I
EEE
C
o
m
p
u
t
e
r
S
o
c
i
e
t
y
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
si
o
n
a
n
d
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
,
v
o
l
.
2
0
1
6
-
D
e
c
e
m
b
e
r
,
p
p
.
2
8
1
8
–
2
8
2
6
,
2
0
1
6
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
6
.
3
0
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
E
ffica
cy
o
f c
h
ili p
la
n
t d
is
ea
s
es c
la
s
s
ifica
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
:
a
p
r
elimin
a
r
y
s
tu
d
y
(
S
u
h
a
n
a
R
o
z
la
n
)
1449
[
2
3
]
Y
.
B
a
z
i
,
M
.
M
.
A
.
R
a
h
h
a
l
,
H
.
A
l
h
i
c
h
r
i
,
a
n
d
N
.
A
l
a
j
l
a
n
,
“
S
i
m
p
l
e
y
e
t
e
f
f
e
c
t
i
v
e
f
i
n
e
-
t
u
n
i
n
g
o
f
d
e
e
p
c
n
n
s
u
s
i
n
g
a
n
a
u
x
i
l
i
a
r
y
c
l
a
ss
i
f
i
c
a
t
i
o
n
l
o
ss f
o
r
r
e
m
o
t
e
se
n
s
i
n
g
sce
n
e
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
1
,
n
o
.
2
4
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s1
1
2
4
2
9
0
8
.
[
2
4
]
A
.
K
r
i
z
h
e
v
s
k
y
,
I
.
S
u
t
s
k
e
v
e
r
,
a
n
d
G
.
E.
H
i
n
t
o
n
,
“
I
mag
e
N
e
t
c
l
a
ss
i
f
i
c
a
t
i
o
n
w
i
t
h
d
e
e
p
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,
”
C
o
m
m
u
n
i
c
a
t
i
o
n
s
o
f
t
h
e
A
C
M
,
v
o
l
.
6
0
,
n
o
.
6
,
p
p
.
8
4
–
9
0
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
4
5
/
3
0
6
5
3
8
6
.
[
2
5
]
D
.
M
a
s
t
e
r
s
a
n
d
C
.
L
u
sc
h
i
,
“
R
e
v
i
s
i
t
i
n
g
S
ma
l
l
B
a
t
c
h
Tr
a
i
n
i
n
g
f
o
r
D
e
e
p
N
e
u
r
a
l
N
e
t
w
o
r
k
s
,
”
p
p
.
1
–
1
8
,
2
0
1
8
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
a
r
x
i
v
.
o
r
g
/
a
b
s/
1
8
0
4
.
0
7
6
1
2
.
[
2
6
]
V
.
M
a
e
d
a
-
G
u
t
i
é
r
r
e
z
e
t
a
l
.
,
“
C
o
m
p
a
r
i
so
n
o
f
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
r
c
h
i
t
e
c
t
u
r
e
s
f
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
t
o
ma
t
o
p
l
a
n
t
d
i
s
e
a
se
s,”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s (
S
w
i
t
z
e
rl
a
n
d
)
,
v
o
l
.
1
0
,
n
o
.
4
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
0
0
4
1
2
4
5
.
[
2
7
]
S
.
P
.
M
o
h
a
n
t
y
,
D
.
P
.
H
u
g
h
e
s,
a
n
d
M
.
S
a
l
a
t
h
é
,
“
U
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
i
m
a
g
e
-
b
a
s
e
d
p
l
a
n
t
d
i
s
e
a
se
d
e
t
e
c
t
i
o
n
,
”
F
ro
n
t
i
e
rs
i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
7
,
n
o
.
S
e
p
t
e
m
b
e
r
,
2
0
1
6
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s
.
2
0
1
6
.
0
1
4
1
9
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
S
u
h
a
n
a
Ro
z
l
a
n
re
c
e
iv
e
d
th
e
Ba
c
h
e
lo
r
o
f
El
e
c
tro
n
ic
E
n
g
i
n
e
e
rin
g
fr
o
m
Un
iv
e
rsiti
Tu
n
Hu
ss
e
in
O
n
n
M
a
lay
sia
(
UTHM)
,
i
n
2
0
0
3
.
S
h
e
is
c
u
rr
e
n
tl
y
p
u
rs
u
in
g
t
h
e
M
a
ste
r
o
f
Co
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
a
tUn
iv
e
rsiti
P
u
tra
M
a
lay
sia
(UP
M
).
He
r
re
se
a
rc
h
in
tere
st
in
c
l
u
d
e
s
c
o
m
p
u
ter
v
isio
n
a
p
p
li
c
a
ti
o
n
s
i
n
a
g
ricu
lt
u
re
,
u
si
n
g
d
e
e
p
lea
rn
in
g
m
e
th
o
d
s
to
stu
d
y
p
la
n
t
lea
f
d
ise
a
se
d
e
tec
ti
o
n
in
c
o
m
p
l
e
x
b
a
c
k
g
ro
u
n
d
s.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
g
s5
5
7
5
2
@s
tu
d
e
n
t.
u
p
m
.
e
d
u
.
m
y
.
Ma
r
sy
ita
H
a
n
a
fi
is
c
u
rre
n
tl
y
a
S
e
n
io
r
Lec
tu
re
r
a
t
th
e
De
p
a
rtm
e
n
t
o
f
C
o
m
p
u
ter
a
n
d
Co
m
m
u
n
ica
ti
o
n
S
y
ste
m
s,
F
a
c
u
l
ty
o
f
E
n
g
i
n
e
e
rin
g
,
Un
iv
e
rsiti
P
u
tra
M
a
lay
sia
(UPM
).
S
h
e
re
c
e
iv
e
d
h
e
r
P
h
.
D.
i
n
Im
a
g
e
P
ro
c
e
ss
in
g
a
n
d
Artifi
c
ial
I
n
telli
g
e
n
c
e
fro
m
Im
p
e
rial
Co
ll
e
g
e
Lo
n
d
o
n
,
UK
,
in
2
0
1
2
.
He
r
re
se
a
rc
h
in
tere
sts
a
re
ima
g
e
p
r
o
c
e
ss
in
g
a
n
d
a
rti
f
icia
l
in
tell
ig
e
n
c
e
t
h
a
t
in
c
lu
d
e
a
u
t
o
n
o
m
o
u
s
v
e
h
icle
s,
p
re
c
isio
n
a
g
ricu
lt
u
re
,
b
i
o
m
e
tri
c
,
a
n
d
Io
T
-
b
a
se
d
in
telli
g
e
n
t
m
o
n
it
o
r
in
g
sy
ste
m
s.
S
h
e
is
a
c
ti
v
e
ly
in
v
o
l
v
e
d
wit
h
m
u
lt
i
-
d
isc
ip
l
in
e
re
se
a
rc
h
stu
d
y
in
g
AI
a
n
d
ima
g
e
p
ro
c
e
ss
in
g
a
p
p
li
c
a
ti
o
n
s
wi
th
th
e
F
a
c
u
lt
y
o
f
Ag
ricu
l
tu
re
,
F
a
c
u
lt
y
o
f
M
e
d
icin
e
,
a
n
d
F
a
c
u
lt
y
o
f
E
d
u
c
a
ti
o
n
.
S
h
e
h
a
s
b
e
e
n
in
v
o
lv
e
d
wi
th
IEE
E
a
s
a
m
e
m
b
e
r
si
n
c
e
2
0
1
2
.
S
h
e
is
th
e
p
ri
n
c
ip
a
l
in
v
e
stig
a
t
o
r
a
n
d
c
o
ll
a
b
o
ra
to
r
f
o
r
p
ro
jec
ts
u
n
d
e
r
th
e
l
o
c
a
l
fu
n
d
in
g
b
o
d
ies
;
n
a
m
e
ly
th
e
M
a
lay
sia
n
M
in
istr
y
o
f
Hi
g
h
e
r
E
d
u
c
a
ti
o
n
(
M
o
HE),
a
n
d
Re
se
a
rc
h
Un
iv
e
rsit
y
G
ra
n
t
S
c
h
e
m
e
(RUG
S
)
(n
o
w
k
n
o
w
n
a
s P
u
tra In
i
ti
a
ti
v
e
G
ra
n
t)
UPM
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
a
rsy
it
a
@u
p
m
.
e
d
u
.
m
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.