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.
15
,
No
.
3
,
J
u
n
e
20
25
,
p
p
.
3212
~
3
2
2
5
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
3
.
pp
3
2
1
2
-
3
2
2
5
3212
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
M
o
rpho
lo
g
ica
l
features f
o
r
multi
-
mo
del rice
g
ra
in
cla
ss
ificatio
n
Su
m
a
D
.
1
,
Na
re
nd
ra
V
.
G
.
1
,
Ra
v
ira
j
a
H
o
lla
M
.
2
,
Da
rsh
a
n H
o
lla
M
.
2
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
M
a
n
i
p
a
l
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
M
a
n
i
p
a
l
A
c
a
d
e
m
y
o
f
H
i
g
h
e
r
E
d
u
c
a
t
i
o
n
,
M
a
n
i
p
a
l
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
M
a
n
i
p
a
l
I
n
st
i
t
u
t
e
o
f
Te
c
h
n
o
l
o
g
y
,
M
a
n
i
p
a
l
A
c
a
d
e
m
y
o
f
H
i
g
h
e
r
E
d
u
c
a
t
i
o
n
,
M
a
n
i
p
a
l
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
2
0
,
2
0
2
4
R
ev
is
ed
Ma
r
6
,
2
0
2
5
Acc
ep
ted
Ma
r
2
0
,
2
0
2
5
In
th
e
re
a
lm
o
f
a
g
r
icu
lt
u
re
a
n
d
f
o
o
d
p
ro
c
e
ss
in
g
,
th
e
a
u
to
m
a
ted
c
las
sifica
ti
o
n
o
f
rice
g
ra
in
s
h
o
ld
s
si
g
n
ifi
c
a
n
t
imp
o
rtan
c
e
.
T
h
e
d
iv
e
rse
v
a
rietie
s
o
f
rice
a
v
a
il
a
b
le
d
e
m
a
n
d
a
sy
ste
m
a
ti
c
a
p
p
ro
a
c
h
t
o
c
a
teg
o
riza
ti
o
n
.
T
h
is
stu
d
y
tac
k
les
th
is
c
h
a
ll
e
n
g
e
b
y
e
m
p
lo
y
in
g
d
i
v
e
rse
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
ls,
in
c
lu
d
in
g
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
i
n
e
(S
VM)
,
ra
n
d
o
m
fo
re
st
(RF
)
,
l
o
g
isti
c
re
g
re
ss
io
n
(LR)
,
d
e
c
isio
n
tree
(DT)
,
G
a
u
ss
ian
n
a
i
v
e
Ba
y
e
s
(G
NB)
,
a
n
d
k
-
n
e
a
re
st
n
e
ig
h
b
o
rs
(K
-
NN)
.
Th
e
d
a
tas
e
t,
so
u
rc
e
d
fr
o
m
Ka
g
g
le,
fe
a
tu
re
s
fi
v
e
d
isti
n
c
t
rice
ty
p
e
s:
Arb
o
ri
o
,
Ba
sm
a
ti
,
I
p
sa
la,
Ja
sm
in
e
,
a
n
d
Ka
ra
c
a
d
a
g
.
Afte
r
th
e
ima
g
e
s
u
n
d
e
rg
o
p
re
p
ro
c
e
ss
in
g
,
a
se
t
o
f
1
3
d
isti
n
c
t
m
o
r
p
h
o
l
o
g
ica
l
f
e
a
tu
re
s
is
e
x
trac
ted
.
T
h
e
se
fe
a
tu
re
s
e
n
su
r
e
a
c
o
m
p
re
h
e
n
si
v
e
re
p
re
se
n
tati
o
n
o
f
rice
g
ra
in
s
fo
r
a
c
c
u
ra
te
c
las
sifica
ti
o
n
.
Th
is
stu
d
y
a
ims
to
c
re
a
te
a
n
in
telli
g
e
n
t
sy
ste
m
fo
r
e
fficie
n
t
a
n
d
p
re
c
ise
rice
g
ra
in
c
las
sifica
ti
o
n
,
c
o
n
tri
b
u
ti
n
g
t
o
o
p
ti
m
izi
n
g
a
g
ric
u
lt
u
ra
l
a
n
d
f
o
o
d
in
d
u
str
y
p
r
o
c
e
ss
e
s.
Am
o
n
g
th
e
m
o
d
e
ls,
K
-
NN
d
e
m
o
n
stra
ted
t
h
e
h
ig
h
e
st
c
las
sifica
ti
o
n
a
c
c
u
ra
c
y
a
t
9
7
.
8
0
%
,
su
rp
a
ss
in
g
ra
n
d
o
m
fo
re
st
(9
7
.
5
1
%
),
DT
(
9
7
.
4
8
%
),
G
NB
(9
6
.
9
9
%
),
S
VM
(9
6
.
8
5
%
),
a
n
d
LR
(9
6
.
0
5
%
).
Ou
r
p
ro
p
o
se
d
K
-
NN
-
b
a
se
d
c
las
sifica
ti
o
n
m
o
d
e
l
a
c
h
iev
e
s
a
n
a
c
c
u
ra
c
y
o
f
9
7
.
8
%
,
d
e
m
o
n
stra
ti
n
g
c
o
m
p
e
ti
ti
v
e
p
e
rfo
r
m
a
n
c
e
a
n
d
o
u
tcla
ss
in
g
se
v
e
ra
l
sta
te
-
of
-
th
e
-
a
r
t
m
e
th
o
d
s
su
c
h
a
s
a
rti
ficia
l
n
e
u
ra
l
n
e
two
r
k
(AN
N)
a
n
d
m
o
d
ifi
e
d
v
is
u
a
l
g
e
o
m
e
try
g
ro
u
p
1
6
(VG
G
1
6
)
wh
il
e
m
a
in
tain
in
g
sim
p
li
c
it
y
a
n
d
c
o
m
p
u
tatio
n
a
l
e
ffi
c
ien
c
y
.
T
h
is
u
n
d
e
rsc
o
re
s
th
e
e
ffe
c
ti
v
e
n
e
ss
o
f
K
-
NN
a
n
d
RF
i
n
e
n
h
a
n
c
in
g
t
h
e
p
re
c
isio
n
o
f
r
ice
v
a
riety
c
las
sifica
ti
on
.
K
ey
w
o
r
d
s
:
Ma
ch
in
e
lear
n
in
g
Mo
r
p
h
o
l
o
g
ical
f
ea
tu
r
es
R
ice
g
r
ain
R
ice
g
r
ain
class
if
icatio
n
Seg
m
en
tatio
n
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
:
Nar
en
d
r
a
V
.
G
.
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
Ma
n
ip
al
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
,
Ma
n
i
p
al
Aca
d
em
y
o
f
Hig
h
er
E
d
u
ca
tio
n
(
MA
HE
)
Ma
n
ip
al,
Kar
n
atak
a,
5
7
6
1
0
4
,
I
n
d
ia
E
m
ail: n
ar
en
d
r
a
.
v
g
@
m
an
ip
al.
ed
u
1.
I
NT
RO
D
UCT
I
O
N
R
ice,
a
cr
u
cial
d
ietar
y
s
tap
le
f
o
r
a
s
u
b
s
tan
tial
s
eg
m
en
t
o
f
t
h
e
g
lo
b
al
p
o
p
u
latio
n
,
o
cc
u
p
ie
s
a
ce
n
tr
al
p
o
s
itio
n
in
e
n
s
u
r
in
g
wo
r
ld
w
id
e
f
o
o
d
s
ec
u
r
ity
an
d
s
ig
n
if
ican
tly
in
f
lu
e
n
ce
s
ec
o
n
o
m
ies
wh
ile
s
u
p
p
o
r
tin
g
liv
elih
o
o
d
s
.
W
ith
its
u
n
m
atch
ed
n
u
tr
itio
n
al
im
p
o
r
tan
ce
,
r
ice
s
er
v
es
as
a
p
r
im
ar
y
s
o
u
r
ce
o
f
ca
r
b
o
h
y
d
r
ates
an
d
ess
en
tial
n
u
tr
ien
ts
,
f
o
r
m
in
g
th
e
d
ietar
y
f
o
u
n
d
atio
n
f
o
r
b
illi
o
n
s
o
f
p
eo
p
le
wo
r
l
d
wid
e.
T
h
e
i
n
tr
icate
r
elatio
n
s
h
ip
b
etwe
en
th
e
r
is
in
g
g
lo
b
al
p
r
o
d
u
ctio
n
o
f
r
ice
an
d
th
e
in
cr
ea
s
i
n
g
d
em
an
d
em
p
h
asizes
th
e
n
ec
ess
ity
f
o
r
ac
cu
r
ate
class
if
icatio
n
m
eth
o
d
s
.
I
n
th
e
q
u
est
to
o
p
tim
ize
ag
r
ic
u
ltu
r
al
p
r
ac
tices
an
d
en
s
u
r
e
th
e
p
r
o
d
u
ctio
n
o
f
h
ig
h
-
q
u
ality
g
r
ain
s
,
th
is
s
tu
d
y
d
elv
es
in
to
th
e
class
if
icat
io
n
o
f
d
i
v
er
s
e
r
ice
v
ar
ieties.
W
ith
r
ice
f
ar
m
in
g
s
er
v
in
g
as
a
co
r
n
er
s
to
n
e
o
f
ec
o
n
o
m
ies,
p
a
r
ticu
lar
ly
in
Asi
a,
th
is
r
esear
ch
aim
s
to
en
h
an
ce
ag
r
icu
ltu
r
al
ef
f
icien
c
y
a
n
d
s
af
eg
u
ar
d
th
is
v
ital
g
lo
b
al
f
o
o
d
r
eso
u
r
ce
.
Ar
o
r
a
et
a
l.
[
1
]
in
tr
o
d
u
ce
an
au
t
o
m
ated
r
ice
g
r
ain
class
if
icatio
n
s
y
s
tem
em
p
lo
y
in
g
m
ac
h
in
e
le
ar
n
in
g
an
d
im
a
g
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es.
Usi
n
g
lo
g
is
tic
r
eg
r
ess
io
n
an
d
s
u
p
p
o
r
t
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
Mo
r
p
h
o
lo
g
ica
l fe
a
tu
r
es fo
r
mu
lti
-
mo
d
el
r
ice
g
r
a
in
cla
s
s
ifica
tio
n
(
S
u
ma
D.
)
3213
v
ec
to
r
m
ac
h
in
e
alg
o
r
it
h
m
s
,
t
h
ey
ac
h
iev
e
9
6
%
a
n
d
9
2
%
a
cc
u
r
ac
y
,
r
esp
ec
tiv
ely
,
in
class
if
y
in
g
v
a
r
io
u
s
r
ice
ty
p
es.
T
h
e
s
y
s
tem
s
tr
ea
m
lin
es
class
if
icatio
n
ef
f
ec
tiv
ely
b
u
t
lack
s
ad
v
an
ce
d
f
ea
tu
r
es
f
o
r
c
h
alk
in
ess
d
etec
tio
n
an
d
co
m
p
r
eh
e
n
s
iv
e
g
r
ain
q
u
al
ity
ev
alu
atio
n
,
s
u
g
g
esti
n
g
p
o
te
n
tial f
o
r
f
u
r
th
er
r
ef
in
em
e
n
t.
Nag
o
d
an
d
R
an
ath
u
n
g
a
[
2
]
p
r
esen
t
a
n
o
v
el
m
eth
o
d
f
o
r
r
ice
q
u
ality
id
e
n
tific
atio
n
u
s
in
g
im
ag
e
p
r
o
ce
s
s
in
g
an
d
m
ac
h
in
e
lea
r
n
in
g
.
T
h
eir
ap
p
r
o
ac
h
ac
h
iev
e
s
9
6
.
0
%
s
eg
m
e
n
tatio
n
a
n
d
8
8
.
0
%
class
if
icatio
n
ac
cu
r
ac
y
ac
r
o
s
s
s
ix
r
ice
ca
te
g
o
r
ies,
s
h
o
win
g
im
p
r
o
v
e
d
s
e
ed
s
eg
m
en
tatio
n
an
d
r
ed
u
ce
d
p
r
o
ce
s
s
in
g
tim
e.
W
h
ile
ef
f
icien
t
with
l
ig
h
tweig
h
t
alg
o
r
ith
m
s
,
th
e
m
eth
o
d
r
eq
u
ir
es
en
h
an
ce
m
e
n
ts
in
s
to
n
e
id
en
tific
atio
n
ac
cu
r
ac
y
,
in
d
icatin
g
s
co
p
e
f
o
r
f
u
tu
r
e
r
esear
ch
im
p
r
o
v
em
e
n
ts
.
Kir
atir
atan
ap
r
u
k
et
a
l.
[
3
]
p
r
o
p
o
s
e
m
ac
h
in
e
v
is
io
n
f
o
r
au
to
m
ate
d
p
ad
d
y
r
i
ce
s
ee
d
class
if
icatio
n
,
o
f
f
er
in
g
a
co
s
t
-
ef
f
ec
tiv
e
alter
n
ativ
e
to
m
an
u
al
m
eth
o
d
s
.
C
o
m
p
ar
in
g
s
tatis
tical
(
lo
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
,
lin
ea
r
d
is
cr
im
in
an
t
an
aly
s
is
(
L
DA)
,
k
-
n
ea
r
est
n
eig
h
b
o
r
s
(K
-
NN)
,
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e
(
SVM)
)
an
d
d
ee
p
lear
n
i
n
g
m
o
d
els
(
v
is
u
al
g
eo
m
et
r
y
g
r
o
u
p
16
(
VGG
16
)
,
VGG1
9
,
Xce
p
tio
n
,
I
n
ce
p
tio
n
V3
,
I
n
ce
p
tio
n
R
esNetV2
)
,
SVM
ac
h
iev
es
s
u
b
g
r
o
u
p
ac
c
u
r
ac
ies
o
f
9
0
.
6
1
%,
8
2
.
7
1
%,
a
n
d
8
3
.
9
%,
wh
ile
I
n
ce
p
tio
n
R
esNetV2
attain
s
9
5
.
1
5
%.
Dee
p
lear
n
in
g
s
u
r
p
ass
es
tr
ad
itio
n
al
m
eth
o
d
s
b
y
u
p
to
1
1
.
2
4
%,
in
d
icatin
g
p
o
ten
tial
f
o
r
im
p
r
o
v
ed
s
ee
d
q
u
ality
in
s
p
ec
tio
n
in
ag
r
icu
ltu
r
e.
Au
k
k
ap
i
n
y
o
et
a
l.
[
4
]
p
r
o
p
o
s
e
an
au
t
o
m
ated
r
ice
g
r
ain
class
if
icatio
n
ap
p
r
o
ac
h
u
s
in
g
a
m
ask
r
e
g
io
n
-
b
ased
c
o
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
R
-
C
NN
)
b
ased
m
e
th
o
d
a
n
d
m
ar
k
er
-
b
ased
wate
r
s
h
ed
alg
o
r
ith
m
.
T
h
eir
m
o
d
e
l
ac
h
iev
es
a
m
ea
n
av
er
ag
e
p
r
ec
is
io
n
(
m
AP)
o
f
1
.
0
f
o
r
s
tick
y
an
d
p
ad
d
y
r
ice
g
r
a
in
s
wh
en
alig
n
ed
m
an
u
ally
,
an
d
an
av
er
ag
e
m
AP
o
f
ap
p
r
o
x
im
ately
0
.
7
5
f
o
r
cla
s
s
if
y
in
g
f
iv
e
s
u
b
ty
p
es.
I
n
tr
ig
u
in
g
ly
,
th
eir
tr
ain
ed
class
if
ier
o
u
tp
er
f
o
r
m
s
h
u
m
a
n
ex
p
er
ts
with
an
av
e
r
ag
e
m
AP
o
f
0
.
8
0
.
Kr
is
h
n
a
et
a
l.
[
5
]
h
a
v
e
d
ev
elo
p
ed
an
a
u
to
m
ated
s
y
s
tem
u
tili
zin
g
im
a
g
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
to
ca
teg
o
r
ize
r
ice
g
r
ain
s
.
T
h
eir
s
y
s
tem
,
u
tili
zin
g
MA
T
L
AB
with
n
eu
r
al
n
etwo
r
k
s
(
NN)
an
d
SVM,
ac
h
iev
es
p
r
ec
is
e
a
n
d
ef
f
icien
t
ass
ess
m
en
t
o
f
r
ic
e
q
u
ality
,
s
u
r
p
ass
in
g
tr
ad
itio
n
al
m
an
u
al
m
et
h
o
d
s
.
T
h
ey
also
s
u
g
g
est
im
p
r
o
v
em
en
ts
in
s
to
n
e
id
e
n
tific
atio
n
ac
cu
r
ac
y
t
h
r
o
u
g
h
ad
d
itio
n
al
v
al
id
atio
n
p
r
o
ce
d
u
r
es.
I
b
r
ah
im
et
a
l.
[
6
]
p
r
o
p
o
s
es
an
au
to
m
ated
m
eth
o
d
f
o
r
class
if
y
in
g
r
ice
g
r
ain
s
u
s
in
g
im
a
g
e
p
r
o
ce
s
s
in
g
m
eth
o
d
s
.
T
h
ey
ap
p
ly
f
ea
tu
r
e
e
x
tr
ac
tio
n
tech
n
iq
u
es
a
n
d
m
u
lti
-
class
SVM
class
if
icatio
n
to
d
is
tin
g
u
is
h
b
etwe
en
th
r
ee
ty
p
es
o
f
r
ice,
ac
h
iev
i
n
g
an
im
p
r
ess
iv
e
ac
cu
r
ac
y
r
ate
o
f
9
2
.
2
2
%
o
n
a
test
s
et
o
f
9
0
im
ag
es
.
R
u
s
lan
et
a
l.
[
7
]
em
p
lo
y
im
a
g
e
p
r
o
ce
s
s
in
g
tec
h
n
iq
u
es
co
m
b
in
ed
with
m
ac
h
in
e
lear
n
in
g
f
o
r
th
e
class
if
icatio
n
o
f
wee
d
y
r
ice
s
ee
d
s
.
T
h
eir
m
eth
o
d
ac
h
iev
es
an
im
p
r
ess
iv
e
8
5
.
3
%
s
en
s
itiv
ity
an
d
9
7
.
9
%
ac
cu
r
ac
y
u
s
in
g
lo
g
is
tic
r
eg
r
ess
io
n
with
R
GB
im
ag
es.
T
h
e
o
p
tim
ized
SVM
m
o
d
el
ac
h
iev
es
a
h
i
g
h
ac
cu
r
ac
y
r
ate
o
f
9
7
.
3
%.
Sin
g
h
an
d
C
h
au
d
h
u
r
y
[
8
]
p
r
esen
t
a
ca
s
ca
d
e
n
etwo
r
k
class
if
ier
d
esig
n
ed
f
o
r
th
e
cla
s
s
if
icatio
n
o
f
r
ice
g
r
ain
s
.
Utilizin
g
a
co
m
b
in
atio
n
o
f
m
o
r
p
h
o
lo
g
ical,
co
lo
r
,
tex
t
u
r
e,
an
d
wav
elet
f
ea
tu
r
es,
th
e
m
o
d
el
ac
h
iev
es
ac
cu
r
ac
y
r
at
es
o
f
9
7
.
7
5
%
with
m
o
r
p
h
o
lo
g
ical
f
ea
t
u
r
es a
n
d
9
6
.
7
5
% with
th
r
ee
s
elec
ted
f
ea
tu
r
es.
C
in
ar
an
d
Ko
k
lu
[
9
]
f
o
cu
s
o
n
class
if
y
in
g
f
iv
e
r
ice
v
ar
ieties
u
s
in
g
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
s
,
ac
h
iev
in
g
p
ea
k
ac
cu
r
ac
ies
o
f
9
7
.
9
9
%
with
r
an
d
o
m
f
o
r
est
u
s
in
g
m
o
r
p
h
o
l
o
g
ical
f
ea
tu
r
es
an
d
9
9
.
2
5
%
u
s
in
g
lo
g
is
tic
r
eg
r
ess
io
n
f
o
r
c
o
lo
r
f
ea
tu
r
es.
Far
a
h
n
ak
ian
et
a
l.
[
1
0
]
in
v
esti
g
ate
n
o
v
el
d
ee
p
-
lear
n
in
g
m
o
d
els,
ev
alu
atin
g
r
esid
u
al
n
etwo
r
k
(
R
esNet)
,
v
is
u
al
g
eo
m
etr
y
g
r
o
u
p
(
VGG)
n
etwo
r
k
,
E
f
f
icien
t
Net,
an
d
Mo
b
ileNet.
T
h
e
an
aly
s
is
s
h
o
wca
s
es
E
f
f
ic
ien
tNet
ac
h
iev
in
g
th
e
h
ig
h
est
ac
cu
r
ac
y
(
9
9
.
6
7
%),
wh
ile
Mo
b
ileNet
ex
ce
ls
in
s
p
ee
d
.
Ng
a
et
a
l.
[
1
1
]
class
if
y
1
7
Vietn
am
ese
r
ice
v
ar
ietie
s
u
s
in
g
im
ag
e
p
r
o
ce
s
s
in
g
tec
h
n
iq
u
es,
ac
h
iev
i
n
g
ac
cu
r
ac
ies
o
f
9
3
.
9
4
%
with
a
n
o
v
el
b
in
a
r
y
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
B
PS
O)
+SVM
m
eth
o
d
an
d
8
9
.
1
%
with
s
p
ar
s
e
r
ep
r
esen
tatio
n
-
b
ased
class
if
icatio
n
(
SR
C
)
.
Ku
o
et
a
l.
[
1
2
]
ac
h
iev
e
an
8
9
.
1
%
ac
cu
r
ac
y
in
id
en
tify
in
g
3
0
r
ice
g
r
ain
v
a
r
ieties
u
s
in
g
im
ag
e
an
aly
s
is
an
d
SR
C
.
C
ar
n
eir
o
et
a
l.
[
1
3
]
ef
f
e
ctiv
ely
ch
ar
ac
ter
ize
r
ice
g
r
ain
p
h
y
s
ico
ch
em
ical
co
m
p
o
s
itio
n
u
s
in
g
n
ea
r
-
in
f
r
a
r
ed
s
p
ec
tr
o
s
co
p
y
(
NI
R
)
with
m
ac
h
in
e
lear
n
in
g
m
o
d
els,
ac
h
iev
in
g
h
i
g
h
ac
cu
r
a
cy
(
9
3
.
9
%)
with
th
e
r
an
d
o
m
tr
ee
m
o
d
el
(
R
an
d
T
)
.
Ah
m
e
d
et
a
l.
[
1
4
]
ca
teg
o
r
ize
im
ag
e
-
b
ased
r
ice
g
r
ain
s
u
s
in
g
g
eo
m
et
r
ic,
d
ee
p
lear
n
in
g
,
s
u
p
er
v
is
ed
,
u
n
s
u
p
er
v
is
ed
,
an
d
s
ta
tis
tical
ap
p
r
o
ac
h
es,
h
ig
h
lig
h
tin
g
th
e
ef
f
icac
y
o
f
d
ee
p
lear
n
in
g
tech
n
iq
u
es.
Srim
u
ly
an
i
an
d
Mu
s
d
h
o
lifa
h
[
1
5
]
en
h
an
ce
r
ice
v
ar
iety
id
en
tific
atio
n
in
I
n
d
o
n
esia
u
s
in
g
NN
,
ac
h
iev
in
g
im
p
r
o
v
ed
ac
cu
r
ac
y
th
r
o
u
g
h
g
e
o
m
etr
y
f
ea
tu
r
es.
Sin
g
h
an
d
C
h
au
d
h
u
r
y
[
1
6
]
class
if
y
f
o
u
r
v
ar
ieties
o
f
b
u
lk
r
ice
g
r
ai
n
im
ag
es
u
s
in
g
b
ac
k
-
p
r
o
p
a
g
atio
n
n
eu
r
al
n
etwo
r
k
(
B
PNN)
,
ac
h
iev
in
g
an
av
e
r
a
g
e
class
if
icatio
n
ac
cu
r
ac
y
ex
ce
ed
in
g
9
6
%
ac
r
o
s
s
all
f
ea
tu
r
es
an
d
d
atasets
.
Azn
an
et
a
l.
[
1
7
]
em
p
lo
y
e
d
c
o
m
p
u
ter
v
is
io
n
a
n
d
m
ac
h
in
e
lear
n
in
g
m
et
h
o
d
s
to
ca
teg
o
r
ize
co
m
m
er
cial
r
ice
s
am
p
les
b
ased
o
n
d
im
en
s
io
n
less
m
o
r
p
h
o
m
etr
ic
an
d
co
l
o
r
p
ar
am
eter
s
ex
tr
ac
ted
f
r
o
m
s
m
ar
tp
h
o
n
e
p
h
o
t
o
s
.
T
h
eir
ar
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN)
m
o
d
el,
u
s
in
g
B
ay
esian
r
eg
u
lar
izatio
n
(
B
R
)
tech
n
iq
u
e,
ac
h
iev
e
d
th
e
h
ig
h
est
class
if
icatio
n
ac
cu
r
ac
y
o
f
9
3
.
9
%
am
o
n
g
1
5
r
ice
v
a
r
ieties.
Ham
za
h
an
d
Mo
h
am
e
d
[
1
8
]
d
is
cu
s
s
ed
th
e
s
ig
n
if
ican
ce
o
f
em
p
l
o
y
in
g
tec
h
n
o
lo
g
y
to
class
if
y
wh
ite
r
ice
g
r
ain
q
u
ality
,
ac
h
iev
i
n
g
a
h
ig
h
ac
cu
r
ac
y
o
f
9
6
%
u
s
in
g
B
PNN.
T
o
s
u
cc
ess
f
u
ll
y
b
r
ee
d
r
ice
an
d
s
atis
f
y
cu
s
to
m
er
d
esire
s
,
it
is
ess
en
tial
to
d
eter
m
in
e
th
e
ch
ar
ac
ter
is
tics
o
f
r
ice
g
r
ain
q
u
ality
(
R
GQ)
,
wh
ich
in
clu
d
e
m
illi
n
g
,
s
to
r
ag
e,
co
o
k
i
n
g
,
n
u
tr
itio
n
al
v
alu
e,
an
d
m
ar
k
et
q
u
alities
.
R
eg
io
n
al
p
r
ef
er
en
ce
s
d
if
f
er
;
f
o
r
ex
am
p
le,
Mid
d
le
E
aster
n
cu
s
to
m
er
s
p
r
ef
er
f
r
ag
r
a
n
t,
well
-
m
illed
lo
n
g
-
g
r
ain
r
ice
,
wh
er
e
as
E
u
r
o
p
ea
n
s
c
h
o
o
s
e
lo
n
g
-
g
r
ain
,
n
o
n
ar
o
m
atic
r
ice.
Glo
b
al
d
em
an
d
f
o
r
h
ig
h
-
q
u
ality
r
ice
is
g
r
o
win
g
.
I
t
ca
n
b
e
v
er
y
h
el
p
f
u
l
to
g
e
n
er
at
e
n
ew
r
ice
v
ar
ieties
with
im
p
r
o
v
e
d
R
GQ
if
th
e
g
en
etic
m
ec
h
a
n
is
m
s
u
n
d
er
l
y
in
g
g
r
ai
n
q
u
ality
q
u
an
titativ
e
tr
ait
lo
ci
(
QT
L
s
)
an
d
th
e
ir
co
n
s
tr
ain
ts
ar
e
u
n
d
er
s
to
o
d
[
1
9
]
.
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.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
1
2
-
3
2
2
5
3214
Ah
ad
et
a
l.
[
2
0
]
co
m
p
a
r
e
C
NN
-
b
ased
d
ee
p
lear
n
in
g
ar
ch
ite
ctu
r
es
f
o
r
d
etec
tin
g
an
d
lo
ca
li
zin
g
n
in
e
ep
id
em
ic
r
ice
d
is
ea
s
es
in
B
an
g
lad
esh
,
ac
h
iev
in
g
a
n
ac
cu
r
ac
y
o
f
9
8
%
with
an
en
s
em
b
le
m
o
d
el.
T
r
an
-
T
h
i
-
Kim
et
a
l.
[
2
1
]
class
if
y
1
7
r
ice
g
r
ain
v
ar
ieties
u
s
in
g
C
NN
m
o
d
els,
ac
h
ie
v
in
g
a
cc
u
r
ac
ies
o
f
9
2
.
8
2
%
with
ANN,
9
6
.
4
1
%
with
m
o
d
if
ied
VGG1
6
,
an
d
9
7
.
8
8
%
with
m
o
d
if
ied
R
esNet5
0
.
Patel
an
d
Sh
ar
af
f
[
2
2
]
ca
teg
o
r
ize
ten
p
ad
d
y
r
ice
v
ar
ieties
u
s
in
g
im
ag
e
p
r
o
ce
s
s
in
g
,
ac
h
iev
in
g
a
d
v
an
tag
es
i
n
s
p
ee
d
an
d
co
s
t
-
ef
f
ec
tiv
en
ess
.
Ku
r
ad
e
et
a
l.
[
2
3
]
in
t
r
o
d
u
ce
a
c
o
s
t
-
ef
f
ec
tiv
e
r
ice
q
u
ality
ass
ess
m
en
t
s
y
s
t
em
u
s
in
g
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
,
ac
h
iev
i
n
g
a
h
ig
h
ac
cu
r
ac
y
o
f
7
7
%
w
ith
r
an
d
o
m
f
o
r
est
class
if
ier
.
Dee
p
ik
a
et
a
l.
[
2
4
]
ev
alu
ate
g
r
ain
q
u
ality
tr
aits
in
2
1
r
ice
h
y
b
r
i
d
s
u
s
in
g
d
ig
it
al
im
ag
in
g
,
ac
cu
r
ately
class
if
y
in
g
g
r
ai
n
s
ize
an
d
ty
p
e,
an
d
id
en
tify
i
n
g
p
o
ten
tial
r
eso
u
r
ce
s
f
o
r
ar
o
m
a
-
ty
p
e
r
ice
b
r
ee
d
in
g
p
r
o
g
r
am
s
.
T
h
is
s
tu
d
y
ad
d
r
ess
es
th
e
ch
allen
g
es
o
f
m
an
u
al
r
ice
g
r
ai
n
class
if
icatio
n
,
wh
ich
is
o
f
ten
lab
o
r
-
in
ten
s
iv
e
an
d
p
r
o
n
e
to
s
u
b
je
ctiv
e
in
co
n
s
is
ten
cies.
B
y
ap
p
ly
in
g
m
ac
h
in
e
lear
n
in
g
tec
h
n
iq
u
es,
we
aim
to
r
ev
o
lu
tio
n
ize
t
h
e
class
if
icatio
n
p
r
o
ce
s
s
,
en
s
u
r
in
g
p
r
ec
is
io
n
a
n
d
co
n
s
is
ten
cy
.
Key
co
n
tr
i
b
u
ti
o
n
s
in
clu
d
e:
a.
Au
to
m
atio
n
o
f
r
ice
g
r
ain
class
if
icatio
n
:
tr
an
s
itio
n
in
g
f
r
o
m
tr
ad
itio
n
al
h
u
m
a
n
-
d
ep
e
n
d
en
t m
eth
o
d
s
to
a
f
u
lly
au
to
m
ated
s
y
s
tem
u
s
in
g
ad
v
an
ce
d
m
ac
h
i
n
e
lear
n
in
g
al
g
o
r
ith
m
s
,
in
clu
d
i
n
g
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
Gau
s
s
ian
n
aiv
e
B
ay
e
s
(
GNB)
,
lo
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
,
d
ec
is
io
n
tr
ee
(
DT
)
,
k
-
n
e
ar
est
n
eig
h
b
o
r
s
(K
-
NN)
,
an
d
r
an
d
o
m
f
o
r
est (
R
F)
.
b.
E
n
h
an
cin
g
class
if
icatio
n
ac
cu
r
ac
y
:
lev
er
a
g
in
g
d
ata
-
d
r
iv
e
n
alg
o
r
ith
m
s
to
im
p
r
o
v
e
t
h
e
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
o
f
r
ice
g
r
ain
class
if
icatio
n
,
m
ee
tin
g
t
h
e
d
em
a
n
d
s
o
f
p
r
ec
is
io
n
ag
r
icu
ltu
r
e.
c.
Featu
r
e
ex
tr
ac
tio
n
f
o
r
g
r
ain
r
ep
r
esen
tatio
n
:
id
en
tify
in
g
an
d
ex
tr
ac
tin
g
cr
itical
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
to
co
m
p
r
eh
e
n
s
iv
ely
r
ep
r
esen
t r
ic
e
g
r
ain
s
,
en
a
b
lin
g
r
o
b
u
s
t c
lass
if
icatio
n
.
d.
Per
f
o
r
m
an
ce
e
v
alu
atio
n
o
f
alg
o
r
ith
m
s
:
co
n
d
u
ctin
g
co
m
p
ar
ativ
e
an
aly
s
is
o
f
v
ar
io
u
s
m
a
ch
in
e
lear
n
in
g
m
o
d
els to
id
en
tif
y
th
e
m
o
s
t e
f
f
ec
tiv
e
ap
p
r
o
ac
h
f
o
r
r
ice
g
r
ain
class
if
icatio
n
.
e.
R
ig
o
r
o
u
s
v
alid
atio
n
:
e
n
s
u
r
in
g
r
eliab
ilit
y
th
r
o
u
g
h
ex
ten
s
iv
e
test
in
g
an
d
an
aly
s
is
to
v
alid
a
te
th
e
s
y
s
tem
'
s
p
er
f
o
r
m
an
ce
ac
r
o
s
s
d
iv
er
s
e
r
ic
e
g
r
ain
v
a
r
ieties.
T
h
is
s
tu
d
y
s
ig
n
if
ican
tl
y
co
n
tr
ib
u
tes
to
p
r
ec
is
io
n
a
g
r
ic
u
ltu
r
e
b
y
ad
v
an
cin
g
th
e
s
y
s
tem
atic
ca
teg
o
r
izatio
n
o
f
r
ice
v
ar
ietie
s
,
p
av
in
g
th
e
way
f
o
r
m
o
r
e
c
o
n
s
is
ten
t
an
d
ef
f
icien
t
ag
r
icu
l
tu
r
al
p
r
ac
tices.
T
h
e
s
u
b
s
eq
u
en
t
s
ec
tio
n
s
o
f
th
is
p
ap
er
u
n
f
o
ld
s
ea
m
less
ly
,
with
s
ec
tio
n
2
d
etailin
g
m
ater
ials
an
d
m
eth
o
d
s
f
o
r
r
ice
g
r
ain
class
if
icatio
n
u
tili
zin
g
d
iv
er
s
e
m
ac
h
i
n
e
lear
n
i
n
g
a
lg
o
r
ith
m
s
,
s
ec
tio
n
3
o
f
f
er
i
n
g
a
co
m
p
r
eh
e
n
s
iv
e
an
aly
s
is
o
f
r
esu
lts
,
an
d
s
ec
tio
n
4
en
ca
p
s
u
latin
g
o
u
r
f
in
d
in
g
s
wh
ile
p
r
o
p
o
s
in
g
av
e
n
u
es
f
o
r
f
u
tu
r
e
r
esear
ch
in
th
e
d
o
m
ain
o
f
r
ice
g
r
ain
class
if
icatio
n
.
2.
M
AT
E
R
I
AL
S AN
D
M
E
T
H
O
DS
T
h
is
s
ec
tio
n
o
u
tlin
es
th
e
s
eq
u
en
tial
p
r
o
ce
s
s
im
p
lem
e
n
ted
to
ac
h
iev
e
th
e
o
b
jectiv
es
o
f
r
ice
g
r
ain
class
if
icatio
n
.
T
h
e
f
lo
wch
ar
t
s
h
o
wn
in
Fig
u
r
e
1
illu
s
tr
ates
th
e
o
v
e
r
all
p
r
o
ce
s
s
o
f
r
ice
g
r
ain
class
if
icatio
n
m
eth
o
d
o
l
o
g
y
.
T
h
e
s
u
b
s
ec
tio
n
s
s
p
an
n
in
g
f
r
o
m
2
.
1
to
2
.
6
will p
r
o
v
id
e
a
b
r
ief
o
v
er
v
iew
o
f
ea
ch
s
tep
o
u
tlin
ed
in
th
e
f
lo
wch
ar
t.
Fig
u
r
e
1
.
I
n
f
o
g
r
ap
h
ic
f
lo
wch
a
r
t o
f
o
v
er
all
p
r
o
ce
s
s
2
.
1
.
Da
t
a
c
o
llect
io
n
T
h
e
s
tu
d
y
b
eg
in
s
with
th
e
co
llectio
n
o
f
d
ata
f
r
o
m
a
p
u
b
lic
ly
av
ailab
le
r
ep
o
s
ito
r
y
o
n
Kag
g
le
[
2
5
]
.
T
h
e
d
ataset
is
s
y
s
tem
atica
lly
o
r
g
an
ized
in
t
o
a
m
ain
d
ir
ec
to
r
y
n
am
e
d
'R
ice_
Data
s
et,
'
wh
ich
co
n
tain
s
f
iv
e
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
Mo
r
p
h
o
lo
g
ica
l fe
a
tu
r
es fo
r
mu
lti
-
mo
d
el
r
ice
g
r
a
in
cla
s
s
ifica
tio
n
(
S
u
ma
D.
)
3215
s
u
b
f
o
ld
er
s
.
E
ac
h
s
u
b
f
o
ld
er
r
ep
r
esen
ts
a
s
p
ec
if
ic
class
lab
el
co
r
r
esp
o
n
d
in
g
to
d
if
f
er
e
n
t
v
ar
ieties
o
f
r
ice:
Ar
b
o
r
io
,
B
asm
ati,
Sala,
J
as
m
in
e,
an
d
Kar
ac
a
d
ag
.
T
h
ese
s
u
b
f
o
ld
er
s
co
llectiv
ely
h
o
u
s
e
a
to
tal
o
f
7
5
,
0
0
0
im
ag
es,
with
ea
ch
class
f
o
ld
er
co
n
tain
in
g
p
r
ec
is
ely
1
5
,
0
0
0
im
ag
es.
T
h
is
well
-
s
tr
u
ctu
r
ed
d
ataset
s
er
v
es
as
a
r
o
b
u
s
t
f
o
u
n
d
atio
n
f
o
r
th
e
cla
s
s
if
icatio
n
task
,
en
s
u
r
in
g
a
b
alan
ce
d
an
d
co
m
p
r
eh
e
n
s
iv
e
r
ep
r
esen
tatio
n
o
f
th
e
f
iv
e
r
ice
v
ar
ieties.
Su
ch
m
eticu
lo
u
s
o
r
g
an
izatio
n
aid
s
in
ef
f
ec
tiv
e
p
r
ep
r
o
ce
s
s
in
g
,
tr
ain
in
g
,
an
d
ev
alu
atio
n
o
f
th
e
class
if
icatio
n
m
o
d
els d
ev
e
lo
p
ed
in
t
h
e
s
tu
d
y
.
2
.
2
.
Da
t
a
pr
epa
ra
t
io
n
Up
o
n
ac
q
u
ir
in
g
th
e
im
a
g
es
f
r
o
m
th
e
Ka
g
g
le
r
e
p
o
s
ito
r
y
,
a
Pan
d
as
Data
Fra
m
e,
d
e
p
icted
in
T
ab
le
1
,
is
g
en
er
ated
to
s
y
s
tem
atica
lly
ar
r
an
g
e
th
e
d
ata.
T
h
e
Data
Fra
m
e
co
m
p
r
is
es
two
p
r
im
ar
y
co
lu
m
n
s
:
o
n
e
f
o
r
s
to
r
in
g
th
e
f
ile
p
ath
s
d
ir
ec
tin
g
t
o
th
e
g
ath
er
ed
im
ag
es
a
n
d
an
o
th
e
r
f
o
r
th
e
co
r
r
esp
o
n
d
in
g
class
lab
els.
T
h
is
o
r
g
an
ized
f
o
r
m
at
f
a
cilitates
ef
f
icien
t
m
an
ip
u
latio
n
a
n
d
a
n
aly
s
is
o
f
th
e
d
ataset.
I
n
th
is
p
h
ase,
im
a
g
e
s
wer
e
tr
an
s
f
o
r
m
e
d
in
to
g
r
ay
s
ca
les
an
d
s
u
b
jecte
d
to
Ots
u
’
s
th
r
esh
o
ld
in
g
f
o
r
b
in
ar
y
s
eg
m
en
tatio
n
.
C
o
n
n
ec
ted
co
m
p
o
n
en
t
lab
elin
g
id
en
tifie
d
d
is
tin
ct
r
eg
io
n
s
,
f
ac
ilit
atin
g
ess
en
tial
p
r
o
p
er
ty
ex
tr
ac
tio
n
.
T
h
ese
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
aim
to
n
o
r
m
alize
an
d
s
tr
ea
m
lin
e
d
at
a
f
o
r
ef
f
icien
t
f
ea
tu
r
e
ex
t
r
ac
tio
n
.
T
h
e
Fig
u
r
es
2
(
a)
,
2
(
b
)
,
an
d
2
(
c)
d
ep
ict
th
e
o
r
ig
in
al,
g
r
ay
s
ca
le,
an
d
s
eg
m
e
n
ted
r
ice
g
r
ain
im
ag
es,
r
esp
ec
t
iv
ely
.
T
ab
le
1
.
Data
Fra
m
e
s
am
p
les c
o
m
p
r
is
in
g
im
a
g
e
p
ath
s
an
d
co
r
r
esp
o
n
d
i
n
g
lab
els
S
l
.
N
o
.
F
i
l
e
p
a
t
h
La
b
e
l
1
R
i
c
e
_
D
a
t
a
se
t
/
A
r
b
o
r
i
o
/
A
r
b
o
r
i
o
(
9
0
9
9
)
.
j
p
g
A
r
b
o
r
i
o
2
R
i
c
e
_
D
a
t
a
se
t
/
A
r
b
o
r
i
o
/
A
r
b
o
r
i
o
(
9
1
)
.
j
p
g
A
r
b
o
r
i
o
…
…
…
7
4
9
9
9
R
i
c
e
_
D
a
t
a
se
t
/
K
a
r
a
c
a
d
a
g
/
K
a
r
a
c
a
d
a
g
(
1
0
8
9
7
)
.
j
p
g
K
a
r
a
c
a
d
a
g
7
5
0
0
0
R
i
c
e
_
D
a
t
a
se
t
/
K
a
r
a
c
a
d
a
g
/
K
a
r
a
c
a
d
a
g
(
1
0
8
9
8
)
.
j
p
g
K
a
r
a
c
a
d
a
g
(
a)
(
b
)
(
c)
Fig
u
r
e
2
.
Pre
p
r
o
ce
s
s
in
g
s
tep
s
(
a)
o
r
ig
i
n
al
im
ag
e
,
(
b
)
g
r
ay
s
ca
le
im
ag
e
,
an
d
(
c)
s
eg
m
e
n
ted
i
m
ag
e
T
h
e
r
atio
n
ale
b
eh
i
n
d
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
em
p
lo
y
ed
lies
in
th
eir
ab
ilit
y
to
s
tan
d
ar
d
ize
an
d
en
h
an
ce
t
h
e
q
u
ality
o
f
th
e
in
p
u
t
d
ata
f
o
r
s
u
b
s
eq
u
e
n
t
an
al
y
s
is
.
C
o
n
v
er
tin
g
im
a
g
es
to
g
r
ay
s
ca
le
s
im
p
lifie
s
co
m
p
u
tatio
n
al
co
m
p
le
x
ity
wh
ile
p
r
eser
v
in
g
ess
en
tial
v
is
u
al
in
f
o
r
m
atio
n
n
ec
ess
ar
y
f
o
r
class
if
icatio
n
.
Ots
u
's
th
r
esh
o
ld
in
g
f
o
r
b
in
a
r
y
s
eg
m
e
n
tatio
n
en
a
b
les
th
e
s
ep
ar
atio
n
o
f
r
ice
g
r
ain
s
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
,
f
ac
ilit
atin
g
m
o
r
e
p
r
ec
is
e
f
ea
tu
r
e
e
x
tr
ac
tio
n
.
Ad
d
itio
n
ally
,
co
n
n
ec
ted
c
o
m
p
o
n
e
n
t
lab
elin
g
aid
s
in
id
en
t
if
y
in
g
an
d
is
o
latin
g
d
is
tin
ct
r
eg
io
n
s
with
in
th
e
s
e
g
m
en
ted
im
a
g
es,
th
er
e
b
y
e
n
h
an
cin
g
th
e
ac
cu
r
ac
y
o
f
m
o
r
p
h
o
lo
g
ical
p
r
o
p
er
ty
ex
tr
ac
tio
n
.
O
v
er
all,
th
ese
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
s
aim
to
s
tr
ea
m
lin
e
th
e
d
ata
an
d
o
p
tim
ize
it
f
o
r
ef
f
icien
t
f
ea
tu
r
e
ex
tr
ac
tio
n
,
u
ltima
tely
im
p
r
o
v
in
g
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
clas
s
if
icatio
n
m
o
d
el.
2
.
3
.
F
e
a
t
ur
e
ex
t
ra
c
t
io
n
T
h
e
f
ea
tu
r
e
ex
tr
ac
tio
n
p
h
ase
i
n
v
o
lv
es
tr
an
s
f
o
r
m
in
g
r
aw
d
at
a
in
to
a
f
o
r
m
at
s
u
itab
le
f
o
r
m
o
d
elin
g
.
I
n
th
is
p
h
ase,
ess
en
tial
1
3
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
wer
e
d
er
iv
ed
f
r
o
m
th
e
s
eg
m
en
ted
r
ice
g
r
ain
im
ag
es.
T
h
is
in
clu
d
es
ch
ar
ac
ter
is
tics
s
u
ch
as
m
ajo
r
an
d
m
i
n
o
r
ax
is
len
g
th
s
,
p
er
im
eter
,
ec
ce
n
tr
icity
,
ar
ea
,
co
n
v
ex
ar
ea
,
ex
ten
t,
s
o
lid
ity
,
o
r
ien
tatio
n
,
e
q
u
iv
alen
t
d
iam
eter
,
co
m
p
ac
tn
ess
,
asp
ec
t
r
atio
,
an
d
r
o
u
n
d
n
ess
.
T
h
e
m
ajo
r
an
d
m
in
o
r
ax
is
len
g
th
s
r
ep
r
esen
t t
h
e
lo
n
g
est an
d
s
h
o
r
test
d
iam
et
er
s
o
f
th
e
r
ice
g
r
ain
,
r
esp
ec
tiv
ely
.
Per
im
eter
r
ef
er
s
to
th
e
len
g
th
o
f
th
e
r
ice
g
r
ai
n
's
b
o
u
n
d
ar
y
,
wh
ile
ec
ce
n
tr
icit
y
d
escr
ib
es
th
e
s
h
ap
e
o
f
th
e
r
ice
g
r
ain
's
ellip
s
e
.
Ar
ea
an
d
co
n
v
ex
ar
ea
q
u
a
n
tify
th
e
s
ize
o
f
th
e
r
ice
g
r
ain
a
n
d
its
co
n
v
e
x
h
u
ll,
r
esp
ec
tiv
ely
.
E
x
ten
t
m
ea
s
u
r
es
t
h
e
r
atio
o
f
th
e
ar
ea
o
f
t
h
e
r
ice
g
r
ain
to
th
e
ar
ea
o
f
its
b
o
u
n
d
in
g
b
o
x
,
p
r
o
v
id
i
n
g
i
n
s
ig
h
ts
in
to
th
e
ex
ten
t
o
f
co
v
er
ag
e.
So
lid
ity
in
d
icate
s
th
e
co
m
p
ac
tn
ess
o
f
t
h
e
r
ice
g
r
ain
,
ca
lcu
lated
as
t
h
e
r
atio
o
f
th
e
ar
ea
o
f
th
e
r
ice
g
r
ain
to
its
co
n
v
ex
a
r
ea
.
Or
ie
n
tatio
n
d
escr
ib
es
th
e
a
n
g
le
o
f
th
e
m
ajo
r
a
x
is
o
f
th
e
r
ice
g
r
ai
n
's
b
o
u
n
d
in
g
ellip
s
e
with
th
e
h
o
r
izo
n
tal
ax
is
.
E
q
u
iv
alen
t
d
iam
eter
r
ep
r
esen
ts
th
e
d
iam
eter
o
f
a
ci
r
cle
with
th
e
s
am
e
ar
ea
as
t
h
e
r
ic
e
g
r
ain
.
C
o
m
p
ac
tn
ess
q
u
an
tifie
s
h
o
w
clo
s
ely
th
e
r
ice
g
r
ain
's
s
h
ap
e
ap
p
r
o
ac
h
es
th
at
o
f
a
cir
cle.
Asp
ec
t
r
atio
i
s
th
e
r
atio
o
f
th
e
m
ajo
r
ax
is
len
g
th
to
th
e
m
in
o
r
ax
is
len
g
t
h
,
in
d
icatin
g
elo
n
g
atio
n
o
r
c
o
m
p
ac
tn
ess
.
L
astl
y
,
r
o
u
n
d
n
ess
m
ea
s
u
r
es
h
o
w
clo
s
e
th
e
r
ice
g
r
ain
's
s
h
ap
e
is
to
a
p
er
f
ec
t
ci
r
cle,
with
a
v
al
u
e
o
f
1
in
d
icatin
g
a
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.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
1
2
-
3
2
2
5
3216
p
er
f
ec
t
cir
cle.
T
h
ese
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
wer
e
ca
r
ef
u
lly
s
elec
ted
f
o
r
t
h
eir
r
elev
an
ce
to
r
ice
g
r
ai
n
ch
ar
ac
ter
is
tics
an
d
th
eir
p
o
ten
t
ial
to
co
n
tr
ib
u
te
t
o
ac
cu
r
ate
cl
ass
if
icatio
n
.
T
h
ese
f
ea
tu
r
es
s
er
v
e
as
q
u
an
t
itativ
e
d
escr
ip
to
r
s
,
ca
p
tu
r
in
g
cr
u
cial
in
f
o
r
m
atio
n
ab
o
u
t
th
e
s
h
ap
e
an
d
s
tr
u
ctu
r
e
o
f
in
d
iv
id
u
al
r
ice
g
r
ain
s
.
T
h
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
f
ac
ilit
ates
th
e
cr
ea
tio
n
o
f
a
f
ea
tu
r
e
m
atr
ix
,
f
o
r
m
in
g
th
e
b
asis
f
o
r
s
u
b
s
eq
u
en
t
m
ac
h
in
e
lear
n
in
g
m
o
d
el
tr
ain
in
g
an
d
an
aly
s
is
.
Fu
r
th
er
m
o
r
e,
th
ese
ex
tr
ac
ted
f
ea
tu
r
es
wer
e
wr
itten
to
a
C
S
V
f
ile
f
o
r
co
m
p
r
eh
e
n
s
iv
e
d
ata
s
to
r
ag
e
a
n
d
f
u
r
t
h
er
ex
p
l
o
r
atio
n
.
T
ab
le
2
v
is
u
ally
p
r
esen
ts
th
e
Fil
e
p
ath
,
L
a
b
el
an
d
th
e
1
3
ex
tr
ac
te
d
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
f
o
r
th
r
ee
r
ice
g
r
ai
n
i
m
ag
es,
p
r
o
v
id
in
g
a
v
is
u
al
r
ep
r
esen
tatio
n
o
f
th
e
q
u
an
titativ
e
ch
ar
ac
ter
is
tics
d
er
iv
ed
f
r
o
m
th
e
im
ag
e
an
aly
s
is
p
r
o
ce
s
s
.
T
ab
le
2
.
Fil
e
p
ath
,
lab
el,
an
d
t
h
e
1
3
e
x
tr
ac
ted
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es o
f
r
ice
s
am
p
les
F
i
l
e
p
a
t
h
La
b
e
l
A
r
e
a
P
e
r
i
me
t
e
r
M
a
j
o
r
a
x
i
s
M
i
n
o
r
a
x
i
s
Ex
t
e
n
t
R
i
c
e
_
D
a
t
a
se
t
/
A
r
b
o
r
i
o
/
A
r
b
o
r
i
o
(
9
0
9
9
)
.
j
p
g
A
r
b
o
r
i
o
7
8
3
5
3
6
6
.
1
3
2
1
3
8
.
3
1
6
7
3
.
3
1
9
0
.
6
2
5
R
i
c
e
_
D
a
t
a
se
t
/
A
r
b
o
r
i
o
/
A
r
b
o
r
i
o
(
9
1
)
.
j
p
g
A
r
b
o
r
i
o
7
6
2
5
3
6
1
.
9
8
9
1
3
6
.
5
8
4
7
2
.
4
2
8
0
.
6
3
5
R
i
c
e
_
D
a
t
a
se
t
/
A
r
b
o
r
i
o
/
A
r
b
o
r
i
o
(
9
1
0
)
.
j
p
g
A
r
b
o
r
i
o
6
8
5
9
3
4
2
.
9
3
6
1
3
5
.
9
2
6
6
4
.
7
5
0
0
.
7
6
4
…
…
…
…
…
…
…
E
c
c
e
n
t
r
i
c
i
t
y
C
o
n
v
e
x
a
rea
S
o
l
i
d
i
t
y
Ori
e
n
t
a
t
i
o
n
E
q
u
i
v
a
l
e
n
t
d
i
a
me
t
e
r
C
o
mp
a
c
t
n
e
ss
A
sp
e
c
t
r
a
t
i
o
R
o
u
n
d
n
e
ss
0
.
8
4
7
8
0
7
6
0
.
9
7
0
0
.
7
2
8
9
9
.
8
7
9
1
.
3
6
1
1
.
8
8
6
0
.
5
2
1
0
.
8
4
7
7
8
7
2
0
.
9
6
8
-
0
.
6
2
2
9
8
.
5
3
1
1
.
3
6
7
1
.
8
8
5
0
.
5
2
0
0
.
8
7
9
6
9
8
6
0
.
9
8
1
-
0
.
1
1
3
9
3
.
4
5
1
1
.
3
6
4
2
.
0
9
9
0
.
4
7
3
…
…
…
…
…
…
…
…
2
.
4
.
F
e
a
t
ur
e
s
elec
t
io
n
I
n
th
e
f
ea
tu
r
e
s
elec
tio
n
s
tag
e,
a
m
eticu
lo
u
s
p
r
o
ce
s
s
is
u
n
d
er
tak
en
to
id
e
n
tify
an
d
r
etain
th
e
m
o
s
t
in
f
o
r
m
ativ
e
s
u
b
s
et
o
f
f
ea
tu
r
es
am
o
n
g
th
e
ex
tr
ac
ted
m
o
r
p
h
o
lo
g
ical
d
escr
ip
to
r
s
.
R
ec
u
r
s
iv
e
f
ea
tu
r
e
elim
in
atio
n
(
R
FE)
is
em
p
lo
y
e
d
t
o
s
y
s
tem
atica
lly
elim
in
ate
less
s
ig
n
if
ican
t
f
ea
tu
r
es,
en
s
u
r
in
g
th
at
th
e
r
ef
in
ed
f
ea
tu
r
e
s
et
m
ain
tain
s
th
e
h
ig
h
est
r
elev
an
ce
f
o
r
s
u
b
s
eq
u
en
t
m
ac
h
in
e
lear
n
in
g
m
o
d
el
t
r
ain
in
g
.
T
h
is
s
tr
ateg
ic
s
elec
tio
n
en
h
an
ce
s
m
o
d
el
in
ter
p
r
etab
ili
ty
,
m
itig
ates
o
v
er
f
itti
n
g
,
an
d
o
p
tim
izes
th
e
p
r
ed
ictiv
e
ca
p
a
b
ilit
y
o
f
th
e
ch
o
s
en
f
ea
tu
r
es.
2
.
5
.
Appl
y
ma
chine le
a
rning
a
lg
o
rit
hm
s
T
h
e
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
u
s
ed
in
th
is
s
tu
d
y
ar
e
s
u
m
m
ar
ized
in
T
ab
le
3
,
p
r
o
v
id
in
g
a
co
m
p
r
eh
e
n
s
iv
e
o
v
er
v
iew
o
f
th
e
class
if
icat
io
n
m
eth
o
d
s
em
p
l
o
y
ed
.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
s
ar
e
em
p
lo
y
ed
f
o
r
th
eir
ef
f
ec
tiv
en
ess
in
h
an
d
lin
g
co
m
p
lex
d
ec
is
io
n
b
o
u
n
d
a
r
ies
an
d
h
ig
h
-
d
im
e
n
s
io
n
al
f
ea
tu
r
e
s
p
ac
es.
Giv
en
th
e
in
tr
icate
m
o
r
p
h
o
lo
g
ical
ch
ar
ac
ter
is
tics
o
f
r
ice
g
r
ain
s
,
S
VM
's
ab
ilit
y
to
cr
ea
te
o
p
tim
al
h
y
p
e
r
p
lan
es
f
o
r
class
if
icatio
n
is
ad
v
an
tag
eo
u
s
.
T
h
e
lin
ea
r
k
er
n
el
is
ch
o
s
en
f
o
r
its
s
im
p
licity
an
d
s
u
itab
ilit
y
f
o
r
th
e
d
ataset.
R
an
d
o
m
Fo
r
ests
ar
e
c
h
o
s
en
f
o
r
th
eir
r
o
b
u
s
tn
ess
an
d
a
b
ilit
y
to
h
an
d
le
m
an
y
f
ea
tu
r
es.
I
n
th
e
co
n
tex
t
o
f
r
ice
g
r
ain
class
if
icatio
n
,
wh
er
e
n
u
m
er
o
u
s
m
o
r
p
h
o
lo
g
ical
f
ea
t
u
r
e
s
co
n
tr
ib
u
te
to
th
e
d
if
f
e
r
en
tiatio
n
o
f
class
es,
R
F
p
r
o
v
id
es
an
en
s
em
b
le
o
f
d
ec
i
s
io
n
tr
ee
s
f
o
r
im
p
r
o
v
ed
ac
cu
r
ac
y
.
Settin
g
th
e
tr
ee
co
u
n
t
in
th
e
f
o
r
est
at
1
0
0
ac
h
iev
es
a
b
alan
ce
b
etwe
en
c
o
m
p
u
tatio
n
al
ef
f
icien
c
y
an
d
t
h
e
ef
f
ec
tiv
en
ess
o
f
th
e
m
o
d
el.
L
o
g
is
tic
r
eg
r
ess
io
n
is
s
elec
ted
f
o
r
its
s
im
p
licity
an
d
ef
f
icien
cy
in
b
in
ar
y
an
d
m
u
lticlas
s
clas
s
if
icatio
n
t
ask
s
.
I
n
r
ice
g
r
ain
class
if
icatio
n
,
wh
er
e
in
ter
p
r
et
ab
ilit
y
is
v
alu
ab
le
,
L
R
p
r
o
v
i
d
es
a
s
tr
aig
h
tf
o
r
war
d
p
r
o
b
ab
ili
s
tic
f
r
am
ewo
r
k
.
T
h
e
'sag
a'
s
o
lv
er
is
em
p
lo
y
ed
f
o
r
o
p
tim
izatio
n
.
T
h
e
iter
atio
n
lim
i
t is estab
lis
h
ed
at
1
4
,
0
0
0
t
o
en
s
u
r
e
co
n
v
er
g
e
n
ce
.
T
ab
le
3
.
Ma
ch
in
e
lear
n
in
g
alg
o
r
ith
m
s
u
s
ed
f
o
r
class
if
icatio
n
A
l
g
o
r
i
t
h
m
D
e
scri
p
t
i
o
n
S
u
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
S
u
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
c
l
a
ssi
f
i
c
a
t
i
o
n
t
a
s
k
s
R
a
n
d
o
m
f
o
r
e
s
t
En
se
mb
l
e
l
e
a
r
n
i
n
g
m
e
t
h
o
d
f
o
r
c
l
a
ssi
f
i
c
a
t
i
o
n
t
a
s
k
s
Lo
g
i
s
t
i
c
r
e
g
r
e
ssi
o
n
R
e
g
r
e
ssi
o
n
a
n
a
l
y
si
s
f
o
r
b
i
n
a
r
y
c
l
a
ssi
f
i
c
a
t
i
o
n
t
a
s
k
s
D
e
c
i
s
i
o
n
t
r
e
e
T
r
e
e
-
l
i
k
e
mo
d
e
l
f
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
n
d
r
e
g
r
e
ss
i
o
n
t
a
s
k
s
G
a
u
ss
i
a
n
n
a
i
v
e
B
a
y
e
s
P
r
o
b
a
b
i
l
i
st
i
c
c
l
a
ssi
f
i
e
r
b
a
s
e
d
o
n
B
a
y
e
s' t
h
e
o
r
e
m
K
-
n
e
a
r
e
s
t
n
e
i
g
h
b
o
r
s
I
n
st
a
n
c
e
-
b
a
se
d
l
e
a
r
n
i
n
g
m
e
t
h
o
d
f
o
r
c
l
a
ssi
f
i
c
a
t
i
o
n
Dec
is
io
n
tr
ee
s
ar
e
s
elec
ted
f
o
r
th
eir
in
ter
p
r
etab
ilit
y
an
d
s
u
ita
b
ilit
y
f
o
r
b
o
th
n
u
m
er
ical
an
d
ca
teg
o
r
ical
d
ata.
I
n
r
ice
g
r
ain
class
if
icatio
n
,
wh
er
e
u
n
d
er
s
tan
d
in
g
th
e
d
ec
is
io
n
p
ath
is
cr
u
cial,
d
ec
is
io
n
tr
ee
s
p
r
o
v
id
e
a
clea
r
s
tr
u
ctu
r
e
b
ased
o
n
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es.
R
ed
u
ce
d
er
r
o
r
p
r
u
n
in
g
(
R
E
P)
is
ap
p
lied
to
o
p
tim
ize
th
e
tr
ee
s
tr
u
ctu
r
e
b
y
r
em
o
v
in
g
u
n
n
ec
e
s
s
ar
y
b
r
an
c
h
es
an
d
p
r
ev
en
t
o
v
er
f
itti
n
g
.
Gau
s
s
ian
n
aiv
e
B
ay
es
is
ch
o
s
en
f
o
r
its
s
im
p
licity
an
d
ef
f
ec
tiv
en
ess
in
h
an
d
lin
g
co
n
tin
u
o
u
s
d
ata.
I
n
th
e
co
n
te
x
t
o
f
r
ice
g
r
ain
cl
ass
if
icatio
n
,
wh
er
e
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
Mo
r
p
h
o
lo
g
ica
l fe
a
tu
r
es fo
r
mu
lti
-
mo
d
el
r
ice
g
r
a
in
cla
s
s
ifica
tio
n
(
S
u
ma
D.
)
3217
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
ca
n
b
e
co
n
s
id
er
ed
as
co
n
tin
u
o
u
s
v
ar
iab
les,
GNB's
as
s
u
m
p
tio
n
o
f
f
ea
tu
r
e
in
d
ep
en
d
en
ce
s
im
p
lifie
s
th
e
m
o
d
elin
g
p
r
o
ce
s
s
.
T
h
e
u
tili
za
tio
n
o
f
K
-
NN
is
d
o
n
e
d
u
e
t
o
its
s
im
p
licity
an
d
ca
p
ac
ity
to
ca
p
tu
r
e
lo
ca
l
p
a
tter
n
s
with
in
th
e
d
ata.
I
n
r
i
ce
g
r
ain
class
if
icatio
n
,
wh
er
e
th
e
s
im
ilar
ity
o
f
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
is
s
ig
n
if
ican
t,
K
-
NNs
ap
p
r
o
ac
h
o
f
ass
ig
n
in
g
lab
els
b
ased
o
n
n
ei
g
h
b
o
r
in
g
in
s
tan
ce
s
p
r
o
v
es e
f
f
ec
tiv
e.
T
h
e
n
u
m
b
er
o
f
n
eig
h
b
o
r
s
(
k
)
is
s
et
to
5
f
o
r
a
b
alan
ce
d
tr
a
d
e
-
o
f
f
b
etwe
en
b
ias an
d
v
ar
ia
n
ce
.
T
h
e
r
atio
n
ale
b
eh
in
d
th
e
s
elec
tio
n
o
f
s
p
ec
if
ic
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
s
tem
s
f
r
o
m
th
e
ir
p
r
o
v
e
n
ef
f
ec
tiv
en
ess
in
class
if
icat
io
n
task
s
,
p
ar
ticu
lar
ly
in
co
n
tex
ts
lik
e
r
ice
g
r
ain
class
if
icati
o
n
.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
is
ch
o
s
en
f
o
r
its
a
b
ilit
y
to
h
an
d
le
h
ig
h
-
d
im
en
s
io
n
al
d
ata
an
d
n
o
n
-
lin
ea
r
d
ec
is
io
n
b
o
u
n
d
a
r
ies
ef
f
ec
tiv
ely
.
R
an
d
o
m
f
o
r
est
is
s
elec
ted
d
u
e
to
its
r
o
b
u
s
tn
ess
to
o
v
er
f
itti
n
g
an
d
its
ca
p
ab
il
ity
to
h
an
d
le
lar
g
e
d
atasets
with
h
ig
h
d
im
en
s
io
n
ality
.
L
o
g
is
tic
r
eg
r
ess
io
n
is
in
clu
d
ed
f
o
r
its
s
im
p
licity
,
in
ter
p
r
etab
ilit
y
,
an
d
s
u
itab
ilit
y
f
o
r
b
in
ar
y
class
if
icatio
n
task
s
.
Dec
i
s
io
n
tr
ee
is
c
h
o
s
en
f
o
r
its
in
tu
itiv
e
r
ep
r
esen
tatio
n
o
f
d
ec
is
io
n
r
u
les
an
d
ea
s
e
o
f
u
n
d
e
r
s
tan
d
in
g
.
Ga
u
s
s
ian
n
aiv
e
B
ay
es
is
in
clu
d
ed
f
o
r
its
s
im
p
licity
,
s
ca
lab
ilit
y
,
an
d
ef
f
icien
cy
,
p
ar
ticu
lar
ly
in
ca
s
es
o
f
s
m
all
tr
ain
in
g
d
atasets
.
L
astl
y
,
k
-
n
ea
r
est
n
eig
h
b
o
r
s
ar
e
s
elec
ted
f
o
r
th
eir
s
im
p
licity
an
d
f
lex
ib
ilit
y
in
h
an
d
lin
g
m
u
lti
-
class
class
if
icat
io
n
p
r
o
b
lem
s
,
r
ely
in
g
o
n
lo
ca
l
in
f
o
r
m
atio
n
r
ath
e
r
th
an
ass
u
m
in
g
a
s
p
ec
if
ic
d
ata
d
is
tr
ib
u
tio
n
.
C
o
llectiv
ely
,
th
ese
alg
o
r
ith
m
s
o
f
f
er
a
d
iv
er
s
e
r
a
n
g
e
o
f
m
eth
o
d
o
l
o
g
ies
th
at
c
o
m
p
le
m
en
t
ea
ch
o
th
er
,
e
n
s
u
r
in
g
co
m
p
r
eh
en
s
iv
e
ex
p
lo
r
atio
n
o
f
th
e
r
ice
g
r
ai
n
class
if
icatio
n
p
r
o
b
lem
.
Fu
r
th
e
r
m
o
r
e,
th
e
o
p
tim
izatio
n
o
f
h
y
p
er
p
ar
am
eter
s
in
v
o
lv
ed
tech
n
iq
u
es
s
u
ch
as
g
r
i
d
s
ea
r
ch
,
wh
ich
s
y
s
tem
atica
lly
ex
p
lo
r
es
d
iv
e
r
s
e
p
ar
am
ete
r
co
m
b
i
n
atio
n
s
to
id
en
tif
y
t
h
e
o
p
tim
al
m
o
d
el.
Ad
d
itio
n
ally
,
cr
o
s
s
-
v
alid
atio
n
was
u
tili
ze
d
to
en
h
a
n
ce
th
is
p
r
o
ce
s
s
b
y
e
v
alu
atin
g
th
e
m
o
d
el's
g
en
er
aliza
tio
n
ab
ilit
y
ac
r
o
s
s
v
ar
io
u
s
d
ata
s
u
b
s
ets.
T
h
r
o
u
g
h
th
ese
m
eth
o
d
o
lo
g
ies,
o
u
r
m
o
d
els
wer
e
f
in
e
-
tu
n
ed
to
attain
o
p
tim
al
p
er
f
o
r
m
an
ce
o
n
t
h
e
d
a
taset wh
ile
m
itig
atin
g
th
e
r
is
k
o
f
o
v
e
r
f
itti
n
g
.
2
.
6
.
M
o
del
ev
a
lua
t
i
o
n a
nd
da
t
a
v
is
ua
liza
t
io
n
I
n
th
is
p
h
ase,
a
co
m
p
r
eh
e
n
s
iv
e
s
et
o
f
m
etr
ics
an
d
tec
h
n
i
q
u
es
h
as
b
ee
n
em
p
lo
y
ed
t
o
ass
es
s
an
d
p
o
r
tr
ay
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
class
if
icatio
n
m
o
d
els.
T
h
e
c
o
n
f
u
s
io
n
m
atr
i
x
p
r
esen
ts
a
d
etailed
b
r
ea
k
d
o
w
n
o
f
p
r
ed
icted
an
d
ac
t
u
al
class
lab
e
ls
,
s
h
ed
d
in
g
lig
h
t o
n
th
e
m
o
d
e
l's
clas
s
if
icatio
n
p
er
f
o
r
m
an
ce
.
Fig
u
r
e
3
illu
s
tr
ates
th
e
co
n
f
u
s
io
n
m
atr
i
x
f
o
r
m
u
lticlas
s
cla
s
s
if
ica
tio
n
,
aid
in
g
in
th
e
in
ter
p
r
etatio
n
o
f
class
if
icatio
n
r
esu
lts
.
Fro
m
th
e
co
n
f
u
s
io
n
m
atr
ix
,
s
ev
e
r
al
m
etr
ics
ar
e
ca
lcu
lated
to
h
o
li
s
tically
ass
e
s
s
th
e
p
er
f
o
r
m
an
c
e
o
f
t
h
e
r
ice
g
r
ain
class
if
icatio
n
m
o
d
el.
E
q
u
atio
n
s
(
1
)
-
(
4
)
p
r
o
v
id
e
f
o
r
m
u
las
f
o
r
av
er
ag
e
d
p
r
ec
is
io
n
,
av
er
a
g
ed
r
ec
all,
F1
-
s
co
r
e,
an
d
av
er
a
g
ed
ac
c
u
r
ac
y
ca
lc
u
latio
n
s
.
Fig
u
r
e
3
.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
m
u
lticlas
s
class
if
icatio
n
A
v
e
r
a
ge
pr
e
c
ision
=
∑
+
=
1
×
100
(
1
)
A
v
e
r
a
ge
r
e
c
a
l
l
=
∑
+
=
1
×
100
(
2
)
F1
−
s
c
or
e
=
2
×
∑
+
=
1
×
∑
+
=
1
∑
+
=
1
+
∑
+
=
1
×
100
(
3
)
A
v
e
r
a
ge
a
c
c
ur
a
c
y
=
∑
+
+
+
+
=
1
×
100
(
4
)
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.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
1
2
-
3
2
2
5
3218
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
er
f
o
r
m
a
n
ce
ev
alu
atio
n
o
f
o
u
r
r
ice
g
r
ain
class
if
icatio
n
m
o
d
els
was
ca
r
r
ied
o
u
t
o
n
a
s
y
s
tem
eq
u
ip
p
e
d
with
a
d
u
al
-
co
r
e
I
n
t
el®
C
o
r
e
™
i3
-
2
3
7
0
M
p
r
o
ce
s
s
o
r
an
d
4
GB
o
f
R
AM
.
T
o
ass
e
s
s
th
e
ef
f
ec
tiv
en
es
s
o
f
th
e
m
o
d
els
in
d
is
tin
g
u
is
h
in
g
b
etwe
en
d
if
f
er
en
t
v
ar
ieties
o
f
r
ice,
we
em
p
lo
y
ed
v
a
r
io
u
s
m
etr
ics
as
o
u
tlin
ed
in
s
ec
tio
n
3
,
in
clu
d
i
n
g
3
.
1
co
n
f
u
s
io
n
m
atr
ix
,
3
.
2
p
er
f
o
r
m
a
n
ce
m
etr
ics
,
3
.
3
r
ec
ei
v
er
o
p
e
r
atin
g
ch
ar
ac
ter
is
tic
(
R
OC
)
cu
r
v
e
,
an
d
3
.
4
p
er
f
o
r
m
an
ce
co
m
p
ar
is
o
n
an
d
ju
s
tific
at
io
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
.
T
h
ese
m
etr
ics,
alo
n
g
with
th
e
s
u
b
s
eq
u
en
t d
is
cu
s
s
io
n
,
p
r
o
v
id
e
a
c
o
m
p
r
eh
en
s
iv
e
ev
alu
atio
n
o
f
t
h
e
class
if
icatio
n
m
o
d
els.
3
.
1
.
Co
nfusi
o
n
ma
t
rix
I
n
T
ab
les
4
to
9
,
th
e
c
o
n
f
u
s
io
n
m
atr
ices
d
ep
ict
th
e
p
e
r
f
o
r
m
an
ce
o
f
v
a
r
io
u
s
m
o
d
els
-
SVM,
R
F
,
L
R
,
DT
,
GNB,
an
d
K
-
NN
r
esp
ec
ti
v
ely
.
E
ac
h
r
o
w
s
ig
n
if
ies
th
e
tr
u
e
class
,
an
d
ea
c
h
co
lu
m
n
d
e
n
o
tes
th
e
p
r
ed
icted
class
.
Diag
o
n
al
elem
en
ts
(
to
p
-
lef
t
to
b
o
tto
m
-
r
i
g
h
t)
s
ig
n
if
y
ac
cu
r
ate
p
r
ed
ictio
n
s
,
wh
ile
o
f
f
-
d
iag
o
n
al
elem
en
ts
in
d
icate
m
is
class
if
icatio
n
s
.
T
ab
le
4
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
SVM
P
r
e
d
i
c
t
e
d
La
b
e
l
s
T
r
u
e
L
a
b
e
l
s
A
r
b
o
r
i
o
B
a
sm
a
t
i
I
p
sal
a
Jasmi
n
e
K
a
r
a
c
a
d
a
g
A
r
b
o
r
i
o
3
6
1
8
0
1
10
1
3
4
B
a
sm
a
t
i
0
3
5
3
9
0
1
9
0
0
I
p
sal
a
10
0
3
8
6
2
10
0
Jasmi
n
e
8
1
3
1
6
3
5
4
8
1
K
a
r
a
c
a
d
a
g
90
0
0
0
3
5
9
2
T
ab
le
5
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
R
F
P
r
e
d
i
c
t
e
d
La
b
e
l
s
T
r
u
e
L
a
b
e
l
s
A
r
b
o
r
i
o
B
a
sm
a
t
i
I
p
sal
a
Jasmi
n
e
K
a
r
a
c
a
d
a
g
A
r
b
o
r
i
o
3
6
0
1
0
7
8
1
4
7
B
a
sm
a
t
i
0
3
5
6
4
0
75
0
I
p
sal
a
7
0
3
8
6
2
13
0
Jasmi
n
e
7
97
7
3
5
8
2
1
K
a
r
a
c
a
d
a
g
98
0
0
0
3
5
8
4
T
ab
le
6
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
L
R
P
r
e
d
i
c
t
e
d
La
b
e
l
s
T
r
u
e
L
a
b
e
l
s
A
r
b
o
r
i
o
B
a
sm
a
t
i
I
p
sal
a
Jasmi
n
e
K
a
r
a
c
a
d
a
g
A
r
b
o
r
i
o
3
5
2
8
0
4
25
2
0
6
B
a
sm
a
t
i
0
3
5
3
1
0
1
9
8
0
I
p
sal
a
43
0
3
8
3
4
5
0
Jasmi
n
e
19
1
3
5
11
3
5
2
5
4
K
a
r
a
c
a
d
a
g
92
0
0
0
3
5
9
0
T
ab
le
7
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
DT
P
r
e
d
i
c
t
e
d
La
b
e
l
s
T
r
u
e
L
a
b
e
l
s
A
r
b
o
r
i
o
B
a
sm
a
t
i
I
p
sal
a
Jasmi
n
e
K
a
r
a
c
a
d
a
g
A
r
b
o
r
i
o
3
5
8
0
0
3
11
1
6
9
B
a
sm
a
t
i
1
3
6
6
1
0
67
0
I
p
sal
a
9
0
3
8
5
8
14
1
Jasmi
n
e
4
1
0
5
11
3
5
7
3
1
K
a
r
a
c
a
d
a
g
75
0
0
0
3
6
0
7
T
ab
le
8
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
GNB
P
r
e
d
i
c
t
e
d
La
b
e
l
s
T
r
u
e
L
a
b
e
l
s
A
r
b
o
r
i
o
B
a
sm
a
t
i
I
p
sal
a
Jasmi
n
e
K
a
r
a
c
a
d
a
g
A
r
b
o
r
i
o
3
6
0
3
0
0
18
1
4
2
B
a
sm
a
t
i
0
3
5
6
0
0
1
6
9
0
I
p
sal
a
`
1
7
0
3
4
8
0
25
0
Jasmi
n
e
0
79
1
3
6
1
3
1
K
a
r
a
c
a
d
a
g
1
1
2
0
0
0
3
5
7
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Mo
r
p
h
o
lo
g
ica
l fe
a
tu
r
es fo
r
mu
lti
-
mo
d
el
r
ice
g
r
a
in
cla
s
s
ifica
tio
n
(
S
u
ma
D.
)
3219
3
.
2
.
P
er
f
o
rma
nce
met
rics
Fig
u
r
es
4
to
7
co
m
p
ar
e
th
e
p
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e,
an
d
ac
cu
r
ac
y
o
f
s
ix
m
ac
h
in
e
lear
n
in
g
m
o
d
els
in
class
if
y
in
g
f
iv
e
d
if
f
er
en
t
ty
p
es
o
f
r
ice
g
r
ain
s
.
I
n
ter
m
s
o
f
p
r
ec
is
io
n
in
Fig
u
r
e
4
,
K
-
NN
ac
h
iev
es
th
e
h
ig
h
est
av
er
ag
e
p
r
ec
is
io
n
(
9
7
.
6
0
%),
f
o
llo
wed
b
y
R
F
(
9
7
.
4
0
%),
DT
(
9
7
.
4
0
%),
GNB
(
9
7
.
0
0
%),
SVM
(
9
6
.
6
0
%),
an
d
L
R
(
9
6
.
0
0
%).
Fo
r
r
ec
all
in
Fig
u
r
e
5
,
K
-
NN
ag
ain
lea
d
s
with
th
e
h
ig
h
est av
er
ag
e
r
ec
all
(
9
7
.
8
0
%),
with
R
F a
n
d
DT
(
b
o
th
9
7
.
4
0
%),
GNB
(
9
7
.
0
0
%),
SVM
(
9
6
.
8
0
%),
a
n
d
L
R
(
9
6
.
2
0
%)
f
o
llo
win
g
.
Similar
ly
,
f
o
r
th
e
F1
-
s
co
r
e
in
Fig
u
r
e
6
,
K
-
NN
m
ain
tain
s
th
e
h
ig
h
est
av
e
r
ag
e
F1
-
s
co
r
e
(
9
7
.
8
0
%),
f
o
llo
wed
b
y
R
F
an
d
DT
(
b
o
th
9
7
.
6
0
%)
,
SVM
an
d
GNB (
b
o
th
9
7
.
0
0
%)
,
an
d
L
R
(
9
6
.
0
0
%).
Fin
ally
,
in
ter
m
s
o
f
ac
cu
r
ac
y
i
n
Fig
u
r
e
7
,
K
-
NN
em
er
g
es a
s
th
e
h
ig
h
est
-
p
er
f
o
r
m
i
n
g
class
if
ier
with
9
7
.
8
0
%
ac
cu
r
ac
y
,
f
o
llo
wed
b
y
R
F
(
9
7
.
5
1
%),
D
T
(
9
7
.
4
8
%),
GNB
(
9
6
.
9
9
%),
an
d
b
o
t
h
SVM
an
d
L
R
(
9
6
.
8
5
%).
T
ab
le
9
.
C
o
n
f
u
s
io
n
m
atr
ix
f
o
r
K
-
NN
P
r
e
d
i
c
t
e
d
La
b
e
l
s
T
r
u
e
L
a
b
e
l
s
A
r
b
o
r
i
o
B
a
sm
a
t
i
I
p
sal
a
Jasmi
n
e
K
a
r
a
c
a
d
a
g
A
r
b
o
r
i
o
3
6
1
0
0
1
8
1
4
4
B
a
sm
a
t
i
2
3
5
6
4
0
73
0
I
p
sal
a
`
1
1
0
3
8
5
6
15
0
Jasmi
n
e
6
1
7
6
6
3
6
0
5
1
K
a
r
a
c
a
d
a
g
78
0
0
0
3
6
0
4
Fig
u
r
e
4
.
C
o
m
p
a
r
is
o
n
o
f
p
r
ec
i
s
io
n
o
f
v
a
r
io
u
s
m
ac
h
i
n
e
lear
n
i
n
g
m
o
d
els f
o
r
r
ice
g
r
ain
class
if
icatio
n
Fig
u
r
e
5
.
C
o
m
p
a
r
is
o
n
o
f
r
ec
al
l o
f
v
ar
i
o
u
s
m
ac
h
in
e
lea
r
n
in
g
m
o
d
els f
o
r
r
ice
g
r
ain
class
if
icatio
n
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.
15
,
No
.
3
,
J
u
n
e
20
25
:
3
2
1
2
-
3
2
2
5
3220
Fig
u
r
e
6
.
C
o
m
p
a
r
is
o
n
o
f
F
1
-
s
co
r
e
o
f
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
m
o
d
els f
o
r
r
ice
g
r
ain
class
if
icatio
n
Fig
u
r
e
7
.
C
o
m
p
a
r
is
o
n
o
f
ac
cu
r
ac
y
o
f
v
ar
io
u
s
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els f
o
r
r
ice
g
r
ain
class
if
icatio
n
3
.
3
.
Rec
eiv
er
o
pera
t
ing
cha
r
a
ct
er
is
t
ic
(
RO
C)
curv
e
I
n
m
u
lti
-
class
class
if
icatio
n
,
th
e
R
OC
cu
r
v
e
s
ig
n
i
f
ies
th
e
o
v
er
all
d
is
cr
im
in
ato
r
y
p
o
wer
o
f
th
e
m
o
d
el
ac
r
o
s
s
all
class
es.
T
h
e
ar
ea
u
n
d
er
th
e
R
OC
c
u
r
v
e
(
AUC)
p
r
o
v
id
es
a
m
ea
s
u
r
e
o
f
th
e
m
o
d
el’
s
ab
ilit
y
to
d
is
tin
g
u
is
h
b
etwe
en
d
if
f
er
en
t
class
es,
with
h
ig
h
er
AUC
in
d
icatin
g
b
etter
p
er
f
o
r
m
a
n
ce
.
Ov
er
all,
th
e
R
OC
cu
r
v
e
f
o
r
m
u
lti
-
class
class
if
ic
atio
n
p
r
o
v
id
es
v
alu
ab
le
i
n
s
ig
h
ts
in
to
th
e
m
o
d
el’
s
class
if
icatio
n
ac
cu
r
ac
y
ac
r
o
s
s
m
u
ltip
le
class
es
an
d
th
e
ef
f
ec
t
iv
en
ess
o
f
th
e
ch
o
s
en
d
is
cr
im
in
atio
n
th
r
esh
o
ld
.
T
o
en
h
an
ce
t
h
e
in
ter
p
r
etab
ilit
y
o
f
th
e
r
esu
lts
,
R
OC
cu
r
v
es
h
av
e
b
ee
n
u
tili
ze
d
.
Ad
d
itio
n
ally
,
th
e
AUC
i
s
ca
lcu
lated
,
p
r
o
v
i
d
in
g
a
s
u
m
m
ar
ize
d
m
etr
ic
f
o
r
th
e
m
o
d
el’
s
d
is
cr
im
in
atio
n
ab
ilit
y
.
T
h
e
r
o
b
u
s
t
elu
cid
atio
n
o
f
th
e
m
o
d
els’
ef
f
ec
tiv
en
ess
in
th
e
in
tr
icate
task
o
f
r
ice
g
r
ain
cla
s
s
if
icatio
n
is
ac
h
iev
ed
th
r
o
u
g
h
th
e
co
n
ce
r
ted
ap
p
licatio
n
a
n
d
in
teg
r
atio
n
o
f
a
d
i
v
e
r
s
e
a
r
r
a
y
o
f
m
o
d
e
l
e
v
a
l
u
a
tio
n
t
e
c
h
n
i
q
u
e
s
a
n
d
s
o
p
h
is
t
ic
a
t
ed
d
a
t
a
v
is
u
a
li
z
a
t
i
o
n
m
e
t
h
o
d
s
.
W
e
p
e
r
f
o
r
m
e
d
a
o
n
e
-
w
a
y
a
n
a
l
y
s
is
o
f
v
a
r
ia
n
c
e
(
ANO
V
A
)
s
t
a
t
is
ti
c
a
l
t
es
t
t
o
as
s
e
s
s
if
t
h
e
r
e
we
r
e
a
n
y
s
t
at
is
t
i
ca
l
l
y
s
ig
n
i
f
i
c
a
n
t
v
a
r
i
at
i
o
n
s
i
n
a
c
c
u
r
a
c
y
a
c
r
o
s
s
t
h
e
m
ea
n
s
o
f
t
h
e
s
i
x
d
i
f
f
e
r
e
n
t
m
o
d
e
ls
u
t
il
i
z
e
d
i
n
t
h
e
a
n
a
l
y
s
i
s
.
T
h
e
o
b
ta
i
n
e
d
h
i
g
h
p
-
v
a
l
u
e
i
n
d
i
c
a
t
es
t
h
a
t
o
u
r
m
o
d
e
ls
r
e
v
e
al
n
o
n
o
t
a
b
l
e
d
i
s
p
a
r
it
i
es
i
n
a
c
c
u
r
a
c
y
a
m
o
n
g
t
h
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
s
.
Fig
u
r
e
s
8
(
a)
t
o
8
(
f
)
illu
s
tr
ate
s
th
e
R
OC
cu
r
v
es
f
o
r
s
ix
cl
ass
if
icatio
n
m
o
d
els
-
SVM,
R
F
,
L
R
,
DT
,
GNB,
an
d
K
-
NN
r
esp
ec
tiv
ely
.
T
h
ese
cu
r
v
es
v
is
u
ally
r
ep
r
e
s
en
t
ea
ch
m
o
d
el’
s
ab
ilit
y
to
d
is
tin
g
u
is
h
b
etwe
en
tr
u
e
p
o
s
itiv
e
an
d
f
alse
p
o
s
itiv
e
r
ates
ac
r
o
s
s
d
if
f
er
en
t
class
if
icatio
n
th
r
esh
o
l
d
s
.
T
h
e
R
OC
an
aly
s
is
f
ac
ilit
ates
a
co
m
p
r
eh
e
n
s
iv
e
ass
ess
m
en
t
o
f
ea
ch
m
o
d
el’
s
ef
f
icac
y
in
b
in
ar
y
class
if
icatio
n
task
s
,
aid
in
g
in
th
e
s
elec
tio
n
o
f
o
p
tim
al
m
o
d
els
b
ased
o
n
th
ei
r
d
is
cr
im
in
ato
r
y
p
o
wer
an
d
o
v
er
all
ac
cu
r
ac
y
.
T
h
e
co
r
r
esp
o
n
d
in
g
AUC
v
alu
es,
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
Mo
r
p
h
o
lo
g
ica
l fe
a
tu
r
es fo
r
mu
lti
-
mo
d
el
r
ice
g
r
a
in
cla
s
s
ifica
tio
n
(
S
u
ma
D.
)
3221
s
u
m
m
ar
ized
in
T
ab
le
1
0
,
p
r
o
v
id
e
a
co
n
cise
m
ea
s
u
r
e
o
f
d
is
cr
im
in
ativ
e
p
er
f
o
r
m
an
ce
,
with
h
ig
h
er
AUC
v
alu
es
in
d
icatin
g
s
u
p
er
io
r
d
is
cr
im
in
a
to
r
y
ab
ilit
ies.
Dif
f
er
en
t m
o
d
els p
er
f
o
r
m
d
if
f
er
en
tly
f
o
r
ea
ch
r
ice
g
r
ain
class
.
Fo
r
ex
am
p
le,
I
n
T
ab
le
1
0
,
in
t
h
e
Ar
b
o
r
io
class
,
th
e
GNB
m
o
d
el
ac
h
iev
es
th
e
h
ig
h
est
AUC
v
alu
e
o
f
0
.
9
9
7
6
,
in
d
icatin
g
s
tr
o
n
g
p
er
f
o
r
m
an
ce
in
d
is
tin
g
u
is
h
in
g
Ar
b
o
r
i
o
r
ic
e
g
r
ain
s
.
Similar
ly
,
in
th
e
J
as
m
in
e
class
,
th
e
R
F
m
o
d
el
ac
h
ie
v
es
th
e
h
ig
h
est
AUC
v
alu
e
o
f
0
.
9
9
8
1
.
W
h
ile
R
F
g
en
er
ally
p
er
f
o
r
m
s
wel
l
ac
r
o
s
s
all
class
es,
ce
r
tain
m
o
d
els
m
ay
ex
ce
l
in
s
p
ec
if
ic
class
es.
Fo
r
ex
am
p
le,
th
e
L
R
m
o
d
el
ac
h
iev
es
a
p
ar
ticu
lar
ly
h
ig
h
AUC
v
alu
e
o
f
0
.
9
9
9
7
f
o
r
t
h
e
I
p
s
ala
class
,
in
d
icatin
g
its
ef
f
ec
tiv
en
ess
in
d
is
tin
g
u
is
h
in
g
I
p
s
ala
r
ic
e
g
r
ain
s
.
(
a)
(
b
)
(
c)
(
d
)
(
e)
(f)
Fig
u
r
e
8
.
R
OC
f
o
r
d
if
f
er
e
n
t c
l
ass
if
icatio
n
m
o
d
els (
a)
SVM,
(
b
)
R
F
,
(
c
)
L
R
,
(
d
)
DT
,
(
e)
GNB,
an
d
(
f
)
K
-
NN
Evaluation Warning : The document was created with Spire.PDF for Python.