I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
9
,
No
.
2
,
J
u
n
e
2020
,
p
p
.
3
04
~
3
09
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
9
.i
2
.p
p
3
04
-
3
09
304
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Nutrien
t
d
eficie
n
cy
det
ec
tio
n in
ma
iz
e
(
Z
ea
m
a
y
s L.
)
lea
v
es
using
i
m
a
g
e p
ro
cess
ing
Nurba
it
y
Sa
bri,
Nurul S
ha
f
ek
a
h K
a
s
s
i
m
,
S
ha
f
a
f
I
bra
hi
m,
Ro
s
niza
Ro
s
la
n,
Nur
Na
bil
a
h Abu
M
a
ng
s
ho
r,
Z
a
ida
h
I
bra
hi
m
F
a
c
u
lt
y
o
f
Co
m
p
u
ter an
d
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
R
A
(UiT
M
)
M
e
lak
a
,
Ka
m
p
u
s Ja
sin
,
7
7
3
0
0
M
e
rli
m
a
u
,
M
e
lak
a
,
M
a
la
y
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
2
8
,
2
0
20
R
ev
i
s
ed
A
p
r
8
,
2
0
20
A
cc
ep
ted
A
p
r
2
5
,
2
0
20
M
a
ize
is
o
n
e
o
f
th
e
w
o
rld
'
s
le
a
d
in
g
f
o
o
d
su
p
p
li
e
s.
T
h
e
re
f
o
re
,
th
e
c
ro
p
'
s
p
ro
d
u
c
ti
o
n
m
u
st
c
o
n
ti
n
u
e
to
re
p
r
o
d
u
c
e
to
f
u
lf
il
l
th
e
m
a
r
k
e
t
d
e
m
a
n
d
.
M
a
ize
is
a
n
a
c
ti
v
e
f
e
e
d
e
r,
th
e
re
f
o
re
,
it
n
e
e
d
t
o
b
e
a
d
e
q
u
a
tely
su
p
p
li
e
d
w
it
h
n
u
tri
e
n
ts
.
T
h
e
h
e
a
lt
h
y
p
lan
ts
w
il
l
b
e
in
d
e
e
p
g
re
e
n
c
o
lo
r
to
in
d
ica
te
it
c
o
n
sist
o
f
a
d
e
q
u
a
te
n
u
tri
e
n
t.
C
u
rre
n
t
p
ra
c
ti
c
e
to
id
e
n
ti
f
y
th
e
n
u
tri
e
n
t
d
e
f
icie
n
c
y
o
n
m
a
ize
lea
f
is
th
ro
u
g
h
t
a
lab
o
ra
to
ry
tes
t.
It
is
ti
m
e
c
o
n
su
m
in
g
a
n
d
re
q
u
ired
a
g
ricu
lt
u
re
k
n
o
w
led
g
e
.
T
h
e
re
f
o
re
,
a
n
im
a
g
e
p
ro
c
e
ss
in
g
a
p
p
ro
a
c
h
h
a
s
b
e
e
n
d
o
n
e
to
im
p
ro
v
e
th
e
lab
o
ra
to
r
y
tes
t
a
n
d
e
li
m
in
a
te
a
h
u
m
a
n
e
rro
r
i
n
id
e
n
ti
f
ica
ti
o
n
p
r
o
c
e
ss
.
T
h
e
p
u
rp
o
se
o
f
th
is
r
e
se
a
r
c
h
is
to
h
e
lp
a
g
ricu
lt
u
rist,
f
a
r
m
e
rs
a
n
d
re
se
a
rc
h
e
rs
to
id
e
n
ti
fy
th
e
t
y
p
e
o
f
m
a
iz
e
n
u
tri
e
n
t
d
e
f
icie
n
c
y
to
d
e
term
in
e
a
n
a
c
ti
o
n
to
b
e
tak
e
n
.
T
h
is
re
se
a
rc
h
u
sin
g
im
a
g
e
p
ro
c
e
ss
in
g
tec
h
n
iq
u
e
s
to
d
e
term
in
e
th
e
ty
p
e
o
f
n
u
tri
e
n
t
d
e
f
icie
n
c
y
th
a
t
o
c
c
u
rs
o
n
t
h
e
p
lan
t
lea
f
.
A
c
o
m
b
in
a
ti
o
n
o
f
G
r
a
y
-
L
e
v
e
l
Co
-
Oc
c
u
rre
n
c
e
M
a
tri
x
(GL
CM
),
hu
-
h
isto
g
ra
m
a
n
d
c
o
lo
r
h
isto
g
ra
m
h
a
s
b
e
e
n
u
se
d
a
s
a
p
a
ra
m
e
ter
f
o
r
f
u
rth
e
r
c
las
si
f
ica
ti
o
n
p
ro
c
e
ss
.
Ra
n
d
o
m
f
o
re
st
tec
h
n
iq
u
e
w
a
s
u
se
d
a
s
c
las
si
f
iers
m
a
n
a
g
e
to
a
c
h
iv
e
7
8
.
3
5
%
o
f
a
c
c
u
ra
c
y
.
It
sh
o
w
s
ra
n
d
o
m
f
o
re
st
is
a
su
it
a
b
le
c
las
si
f
ier
f
o
r
n
u
tri
e
n
t
d
e
f
icie
n
c
y
d
e
tec
ti
o
n
i
n
m
a
ize
lea
v
e
s.
M
o
r
e
m
a
c
h
in
e
lea
rn
in
g
a
lg
o
rit
h
m
w
il
l
b
e
tes
ted
t
o
in
c
re
a
se
c
u
rre
n
t
a
c
c
u
ra
c
y
.
K
ey
w
o
r
d
s
:
C
o
lo
r
h
is
to
g
r
a
m
Gr
a
y
-
L
e
v
el
C
o
-
Occ
u
r
r
en
ce
Hu
-
h
i
s
to
g
r
a
m
Ma
ize
leav
es
Nu
tr
ie
n
t d
ef
icie
n
c
y
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
:
Nu
r
b
ait
y
Sab
r
i,
Facu
lt
y
o
f
C
o
m
p
u
ter
an
d
Ma
t
h
e
m
a
tical
Scie
n
ce
s
,
Un
i
v
er
s
iti T
ek
n
o
lo
g
i M
A
R
A
(
UiT
M)
Me
lak
a,
Ka
m
p
u
s
J
asin
,
7
7
3
0
0
Me
r
lim
a
u
,
Me
lak
a,
Ma
la
y
s
ia
.
E
m
ail:
n
u
r
b
ait
y
_
s
ab
r
i@
u
it
m
.
e
d
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
Z
ea
m
a
y
s
L
.
t
h
e
s
c
ien
ti
f
ic
n
a
m
e
o
f
m
aize
is
a
v
er
y
i
m
p
o
r
tan
t
s
tap
le
f
o
o
d
s
u
p
p
l
y
i
n
m
a
n
y
p
ar
ts
o
f
th
e
w
o
r
ld
f
r
o
m
m
aize
it
s
elf
to
ce
r
ea
l.
Ma
ize
ca
n
ad
ap
t
i
n
d
i
f
f
er
en
t
k
in
d
o
f
e
n
v
ir
o
n
m
e
n
t
a
n
d
g
r
o
w
i
n
g
i
n
a
w
id
er
ar
ea
th
an
o
th
er
m
aj
o
r
cr
o
p
s
u
ch
as
p
o
tato
,
w
h
ea
t
an
d
s
o
y
b
ea
n
[
1
]
.
T
h
e
m
a
ize
p
lan
t
ca
n
b
e
d
ef
in
ed
as
a
m
etab
o
lic
s
y
s
te
m
w
h
o
s
e
at
t
h
e
en
d
o
f
th
e
p
r
o
d
u
ct
w
il
l
p
r
o
d
u
ce
v
ar
iet
y
s
tar
ch
d
ep
o
s
it
ed
f
r
o
m
t
h
e
m
aize
k
er
n
el
s
s
u
c
h
as c
er
ea
ls
o
r
f
lo
u
r
.
T
h
is
p
lan
t is
th
e
m
o
s
t
p
r
o
m
i
n
en
t
ce
r
ea
l g
r
ai
n
all
o
v
er
th
e
w
o
r
ld
,
af
ter
r
ice
a
n
d
w
h
ea
t.
I
t is co
m
m
o
n
l
y
p
r
o
v
id
i
n
g
f
u
ll
-
b
o
d
ied
n
u
tr
ien
ts
f
o
r
h
u
m
an
s
an
d
a
n
i
m
als.
Ma
ize
p
lan
ts
ar
e
p
r
ec
is
el
y
i
n
t
en
s
i
v
e
f
ee
d
er
s
an
d
e
v
en
s
o
ils
n
ee
d
to
b
e
v
er
y
f
er
tile
to
f
u
ll
y
s
u
p
p
lied
n
u
tr
ie
n
t
s
as t
h
e
p
lan
t d
ev
elo
p
.
C
o
m
p
ar
ed
to
o
th
er
s
i
n
g
le
-
ce
ll
s
y
s
te
m
s
,
m
aize
h
a
s
a
lo
n
g
-
li
f
e
c
y
cle
s
i
n
ce
it e
a
s
y
to
g
r
o
w
[
1
]
.
A
s
to
in
d
icatio
n
s
o
f
n
u
tr
ie
n
t
ad
eq
u
ac
y
,
p
la
n
ts
o
u
g
h
t
to
b
e
in
d
ee
p
g
r
ee
n
co
lo
u
r
.
So
m
e
o
f
m
aize
p
lan
t
f
ac
in
g
n
u
tr
ie
n
t
d
e
f
icie
n
c
y
as
a
r
es
u
lt
o
f
s
e
v
er
al
a
g
r
o
n
o
m
ic
a
n
d
en
v
ir
o
n
m
en
t
f
ac
to
r
s
.
T
r
a
d
i
tio
n
al
n
u
tr
ien
t
d
ef
icien
c
y
d
iag
n
o
s
tic
m
et
h
o
d
s
r
eq
u
ir
e
co
m
p
r
eh
e
n
s
i
v
e
s
o
il
o
r
p
lan
t
tis
s
u
e
lab
o
r
ato
r
y
test
in
g
o
r
m
a
n
u
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
N
u
tr
ien
t d
eficien
cy
d
etec
tio
n
in
ma
iz
e
(
Zea
ma
ys L.)
lea
ve
s
u
s
in
g
ima
g
e
p
r
o
ce
s
s
in
g
(
N
u
r
b
a
ity
S
a
b
r
i
)
305
ex
a
m
in
at
io
n
b
y
f
ar
m
er
s
[
2
]
.
T
h
is
is
a
ted
io
u
s
an
d
lab
o
r
io
u
s
w
o
r
k
,
th
o
u
g
h
,
a
n
d
ca
n
b
e
ac
h
i
ev
ed
o
n
l
y
w
it
h
v
er
y
s
m
al
l sa
m
p
le
s
[
3
]
.
T
h
er
ef
o
r
e,
m
o
s
t o
f
t
h
e
m
eth
o
d
s
o
f
p
r
o
d
u
c
tio
n
ar
e
ap
p
r
o
x
im
a
te.
Nu
tr
ie
n
ts
d
e
f
icie
n
c
y
d
etec
tio
n
i
m
p
o
r
tan
t
to
e
n
s
u
r
e
all
m
aize
g
r
o
w
n
in
ac
co
r
d
an
ce
w
i
th
w
h
at
i
s
r
eq
u
ir
es
s
u
c
h
as
Nitr
o
g
e
n
,
I
r
o
n
an
d
m
o
r
e
.
T
h
is
n
u
tr
ie
n
t
d
ef
i
cien
c
y
i
n
m
aize
v
is
u
all
y
ca
n
b
e
s
ee
n
th
r
o
u
g
h
th
e
l
ea
v
es
o
f
af
f
ec
ted
p
lan
t
s
u
s
i
n
g
i
m
a
g
e
p
r
o
ce
s
s
in
g
.
Mo
r
eo
v
er
,
im
a
g
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
e
ca
n
ass
u
r
e
th
at
t
h
e
m
et
h
o
d
o
lo
g
y
m
a
y
b
e
r
eliab
l
e
an
d
ab
le
to
b
e
d
etec
tio
n
to
o
ls
in
ag
r
ic
u
lt
u
r
al
f
ield
[
4
]
.
T
h
e
u
s
e
o
f
i
m
a
g
e
p
r
o
ce
s
s
in
g
is
m
o
r
e
ac
c
u
r
ate
s
in
ce
i
t
ca
n
ca
p
t
u
r
e
th
e
d
i
f
f
er
en
ce
s
o
f
p
atter
n
,
co
lo
u
r
,
an
d
th
e
s
u
r
f
ac
e
t
h
at
af
f
ec
ted
.
Un
d
er
s
ta
n
d
in
g
t
h
es
e
s
ig
n
s
w
ill
h
elp
d
eter
m
i
n
e
co
r
r
ec
tiv
e
ac
tio
n
to
n
o
r
m
a
lize
th
e
p
la
n
t
[
2
]
.
C
las
s
i
f
icatio
n
p
r
o
ce
s
s
i
n
i
m
a
g
e
p
r
o
ce
s
s
in
g
co
n
s
is
t
s
o
f
f
ea
tu
r
es
an
d
cla
s
s
i
f
ier
.
I
t
is
e
s
s
e
n
tia
l
to
co
n
ce
n
tr
ate
o
n
th
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
s
ta
g
e
as
it
h
as
a
n
o
b
s
er
v
ab
l
y
a
f
f
ec
ts
o
n
t
h
e
co
m
p
ete
n
ce
o
f
t
h
e
r
ec
o
g
n
itio
n
s
y
s
te
m
[
5
]
.
T
h
er
e
ar
e
d
if
f
er
en
t te
ch
n
iq
u
e
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
s
u
ch
as te
x
tu
r
e,
co
lo
r
an
d
s
h
ap
e
[
6
]
.
Gab
o
r
f
ilter
h
as
b
ee
n
u
s
ed
b
y
r
esear
ch
er
f
o
r
v
ar
io
u
s
te
x
t
u
r
e
an
al
y
s
is
ap
p
licatio
n
s
[
7
]
.
Gab
o
r
f
ilter
i
s
in
te
n
d
ed
to
test
t
h
e
w
h
o
le
r
ec
u
r
r
en
ce
d
o
m
ai
n
o
f
a
n
i
m
a
g
e
b
y
p
o
r
tr
ay
in
g
t
h
e
ce
n
ter
f
r
eq
u
e
n
c
y
a
n
d
o
r
ien
tat
io
n
p
ar
am
eter
s
[
8
]
.
B
y
d
ep
lo
y
i
n
g
o
f
th
e
Gab
o
r
f
ilter
,
i
m
ag
e
s
te
g
an
al
y
s
is
i
m
p
r
o
v
ed
s
c
h
e
m
in
g
t
o
o
l
in
co
n
s
eq
u
e
n
ce
u
s
e
o
f
t
h
e
s
te
g
an
al
y
s
is
[
9
]
.
Ho
w
e
v
er
,
Gab
o
r
f
ilter
s
h
av
e
p
o
o
r
ex
ec
u
tio
n
w
h
en
th
e
i
m
ag
e
is
f
r
a
g
m
en
ted
n
u
m
er
o
u
s
s
m
aller
tex
tu
r
e,
i
n
th
is
m
a
n
n
er
i
n
f
lu
e
n
ci
n
g
t
h
e
p
r
ec
is
io
n
o
f
i
m
a
g
e
s
e
g
m
e
n
tat
io
n
[
1
0
]
.
GL
C
M
o
n
e
o
f
th
e
w
ell
k
n
o
w
tex
t
u
r
e
f
ea
tu
r
es
co
n
s
is
t
o
f
co
-
o
cc
u
r
r
en
c
e
m
atr
ices
r
esu
lts
ar
e
b
etter
th
an
o
th
er
te
x
t
u
r
e
d
is
tin
ct
m
et
h
o
d
s
[
1
1
]
.
I
t
is
a
co
n
v
e
n
tio
n
al
m
e
th
o
d
o
f
te
x
t
u
r
e
f
ea
tu
r
e
e
x
tr
ac
tio
n
t
h
at
ca
n
b
e
u
s
e
f
u
l
f
o
r
i
m
a
g
e
class
i
f
icatio
n
,
s
e
g
m
e
n
tatio
n
,
r
ec
o
g
n
itio
n
an
d
m
o
r
e
[
1
1
]
.
C
o
lo
r
m
o
m
e
n
t
(
C
M)
co
n
s
id
er
as
an
e
f
f
ec
tiv
e
a
n
d
s
i
m
p
le
m
et
h
o
d
f
o
r
co
lo
r
f
ea
t
u
r
e
[
8
]
.
T
h
er
e
ar
e
o
v
er
all
th
r
ee
s
tep
in
co
lo
r
m
o
m
e
n
t
w
h
ic
h
ar
e
m
ea
n
,
Var
ia
n
ce
an
d
s
k
e
w
n
es
s
[
1
2
]
.
A
cc
o
r
d
in
g
to
p
r
ev
io
u
s
ar
ticle
also
m
e
n
tio
n
t
h
at
C
M
h
a
s
a
co
m
p
ac
t
f
ea
tu
r
e
as
it
is
o
n
l
y
r
eq
u
ir
ed
th
r
ee
co
lo
r
co
m
p
o
n
e
n
ts
[
1
2
]
.
Ho
w
e
v
er
,
C
M
is
lo
w
in
d
is
cr
i
m
i
n
atio
n
p
o
w
er
ac
co
r
d
in
g
to
p
r
ev
io
u
s
ar
ticle.
C
o
lo
r
h
is
to
g
r
a
m
v
er
y
ef
f
ec
tiv
e
a
n
d
g
i
v
e
ac
t
u
al
p
r
esen
tat
io
n
v
is
u
aliza
tio
n
[
1
2
]
.
I
n
ad
d
itio
n
,
co
lo
r
h
is
to
g
r
a
m
i
s
u
s
ef
u
l
to
r
ec
o
g
n
ize
i
m
a
g
e.
T
h
e
i
m
p
l
e
m
e
n
ted
co
lo
r
h
is
to
g
r
a
m
ap
p
r
o
ac
h
h
a
s
p
r
o
v
en
to
b
e
v
er
y
ea
s
y
a
n
d
e
f
f
icien
t
to
e
n
f
o
r
ce
[
1
3
]
.
His
to
g
r
a
m
Or
ien
tat
io
n
G
r
ad
ien
t
(
HOG)
d
etec
ts
ed
g
e
o
r
g
r
ad
ien
t
f
o
r
m
t
h
at
is
v
er
y
d
escr
ip
tiv
e
o
f
lo
ca
l
s
h
ap
e
an
d
d
o
es
s
o
i
n
a
lo
ca
l
r
ep
r
esen
tat
io
n
w
it
h
o
n
l
y
an
ea
s
il
y
m
an
a
g
ed
d
e
g
r
ee
o
f
in
-
v
ar
ia
n
ce
o
f
lo
ca
l
g
eo
m
etr
y
[
1
4
]
.
Ho
w
e
v
er
,
it
ta
k
es
a
lo
n
g
co
m
p
u
tatio
n
ti
m
e
[
1
5
]
.
A
f
ast
an
d
v
er
y
s
i
m
p
le
s
h
ap
e
f
ea
t
u
r
es
w
h
ic
h
i
s
h
u
m
o
m
e
n
t
h
a
s
b
ee
n
i
n
tr
o
d
u
ce
[
1
6
]
.
I
t
is
th
e
b
est
m
et
h
o
d
i
n
i
m
g
e
p
r
o
ce
s
s
i
n
g
s
tr
ateg
ie
s
[
1
7
]
.
T
h
er
ef
o
r
e,
th
i
s
r
esear
ch
i
m
p
le
m
e
n
t
co
m
b
in
atio
n
o
f
G
L
C
M,
co
lo
r
h
is
to
g
r
a
m
a
n
d
h
u
m
o
m
en
t
to
as a
f
ea
tu
r
es to
b
e
u
s
ed
in
cl
ass
i
f
ier
.
Su
p
p
o
r
t
Vec
to
r
Ma
ch
i
n
e
h
as
b
ee
n
u
s
ed
to
d
etec
t
a
leaf
d
is
ea
s
e
o
n
g
r
a
p
e
lea
f
[
1
8
]
.
Ho
w
ev
er
,
it
i
s
d
if
f
ic
u
lt
to
s
p
ec
i
f
y
t
h
e
b
est
p
ar
a
m
eter
to
u
s
e
if
d
ata
is
n
o
t
s
ep
ar
ate
lin
ea
r
l
y
[
1
9
]
.
C
NN
is
o
n
e
o
f
ex
ce
lle
n
t
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
ap
p
r
o
ac
h
i
n
A
r
ti
f
icia
l
I
n
telli
g
en
t
t
h
at
i
m
p
le
m
e
n
t
g
e
n
er
al
a
n
d
d
etail
tas
k
.
Ma
n
y
C
NN
ar
ch
itect
u
r
es
h
a
s
b
ee
n
u
s
ed
f
o
r
i
m
ag
e
cla
s
s
i
f
icatio
n
an
d
r
ec
o
g
n
itio
n
s
u
ch
as
A
le
x
N
et
an
d
L
eNe
t
[
2
0
]
.
E
n
h
a
n
ce
m
en
ts
in
co
n
v
o
lu
t
io
n
al
n
e
u
r
al
n
et
w
o
r
k
s
(
C
NNs)
r
ec
en
tl
y
h
av
e
m
ad
e
t
h
e
m
t
h
e
b
est
i
n
c
lass
a
m
o
n
g
m
ac
h
in
e
le
ar
n
i
n
g
ap
p
r
o
ac
h
es
f
o
r
ad
d
r
ess
in
g
co
m
p
u
ter
v
is
io
n
is
s
u
es,
esp
ec
ial
l
y
in
i
m
ag
e
class
i
f
icatio
n
[
2
1
]
.
T
h
is
ar
ch
itectu
r
e
h
a
s
also
b
ee
n
i
m
p
le
m
en
ts
o
n
Ma
ize
leaf
t
o
id
en
tify
t
h
e
d
is
ea
s
e
o
n
m
aiz
e
p
lan
ts
an
d
a
h
ig
h
ac
cu
r
ac
y
ac
h
iv
e
[
2
2
]
.
Ho
w
e
v
e
r
,
o
n
e
m
ain
d
is
ad
v
an
ta
g
e
o
f
C
NN
b
ased
m
et
h
o
d
s
is
t
h
at
th
e
y
u
s
u
all
y
n
ee
d
lar
g
e
d
atasets
to
tr
ain
a
f
ea
s
ib
le
m
o
d
el
[
2
3
]
.
R
an
d
o
m
Fo
r
est
s
(
R
F)
is
am
o
n
g
t
h
e
m
o
s
t
e
f
f
ec
tiv
e
an
d
ef
f
icie
n
t
m
ac
h
in
e
lear
n
i
n
g
ap
p
r
o
ac
h
es
in
to
d
a
y
's
al
g
o
r
ith
m
s
[
2
4
]
.
B
ec
au
s
e
o
f
t
h
eir
h
i
g
h
p
r
ed
ictiv
e
p
r
ec
is
io
n
,
r
a
n
d
o
m
f
o
r
ests
s
i
n
ce
th
e
n
h
a
v
e
p
r
o
v
en
to
b
e
s
u
cc
ess
f
u
l
i
n
s
o
m
a
n
y
f
ield
s
[
2
5
]
.
Du
e
to
a
g
r
ea
t
s
u
cc
e
s
s
o
f
r
an
d
o
m
f
o
r
est,
i
m
p
le
m
e
n
tatio
n
o
f
th
is
alg
o
r
ith
m
w
il
l b
e
d
o
n
e
to
class
if
y
n
u
tr
ie
n
t d
ef
ic
ien
c
y
o
f
m
ai
ze
leaf
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
I
m
ag
e
p
r
o
ce
s
s
i
n
g
co
n
s
i
s
t
o
f
s
tep
b
y
s
tep
p
r
o
ce
s
s
o
n
class
if
y
th
e
n
u
tr
ie
n
t
i
n
to
th
r
ee
cla
s
s
es
n
a
m
e
Nitr
o
g
en
,
P
o
tass
i
u
m
a
n
d
Ma
g
n
e
s
i
u
m
d
ef
icien
c
y
.
B
elo
w
s
h
o
w
s
th
e
f
lo
w
o
f
clas
s
icatio
n
s
tar
tin
g
f
r
o
m
i
n
p
u
t
i
m
a
g
e,
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
t
u
r
e
ex
tr
ac
tio
n
,
a
n
d
class
i
f
ica
ti
o
n
.
Fig
u
r
e
1
ill
u
s
tr
ates
th
e
leaf
d
etec
tio
n
a
n
d
class
i
f
icatio
n
d
iag
r
a
m
f
o
r
n
u
tr
ien
t c
las
s
i
f
icatio
n
o
n
m
aize
le
af
.
Fig
u
r
e
1
.
L
ea
f
d
etec
tio
n
an
d
cl
ass
i
f
icatio
n
In
p
u
t
Ima
g
e
Pre
-
p
ro
c
e
s
s
i
n
g
F
e
atu
re
E
x
tr
ac
ti
o
n
Cl
as
s
i
ficatio
n
O
u
tp
u
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
20
20
:
3
04
–
3
09
306
2
.
1
.
I
np
ut
i
m
a
g
e
I
n
th
is
s
tu
d
y
,
t
h
r
ee
t
y
p
e
o
f
n
u
tr
ien
t
d
ef
icie
n
c
y
w
h
ic
h
ar
e
Nitr
o
g
en
,
Ma
g
n
esi
u
m
a
n
d
P
o
tass
i
u
m
ar
e
b
ein
g
co
llected
.
Up
to
3
0
d
is
tin
ct
i
m
ag
e
s
o
f
clas
s
es
o
f
n
u
t
r
ien
t
d
e
f
icie
n
c
y
m
e
n
tio
n
ed
will
b
e
p
r
o
ce
s
s
es
to
d
etec
t
ty
p
e
o
f
n
u
tr
ie
n
t
d
ef
ic
i
en
c
y
o
f
t
h
e
m
aize
leaf
.
T
h
e
d
ataset
o
f
m
aize
lea
f
w
ill
b
e
d
iv
id
ed
in
to
t
w
o
p
ar
titi
o
n
w
h
ic
h
ar
e
tr
ain
i
n
g
a
n
d
test
in
g
.
Fi
g
u
r
e
2
s
h
o
w
s
ex
a
m
p
le
o
f
i
m
ag
e
o
f
n
u
tr
ie
n
t d
ef
i
cien
c
y
.
(
a)
(
b
)
(
c)
Fig
u
r
e
2
.
I
m
a
g
e
o
f
(
a)
Nitr
o
g
e
n
,
(
b
)
Po
tass
iu
m
a
n
d
(
c)
Ma
g
n
esiu
m
d
ef
ic
ien
c
y
[
2
6
]
2
.
2
.
P
re
pro
ce
s
s
ing
An
i
n
p
u
t
i
m
a
g
e
co
n
ta
in
s
o
f
u
n
w
a
n
ted
n
o
is
e,
p
r
ep
r
o
ce
s
s
i
n
g
p
h
ase
i
s
to
r
e
m
o
v
e
n
o
is
e
a
n
d
en
h
a
n
ce
m
en
t
o
f
i
m
a
g
e.
I
m
ag
e
P
r
ep
r
o
ce
s
s
in
g
i
s
to
ex
p
el
th
e
u
n
d
esira
b
le
n
o
is
e
i
n
i
m
ag
e
p
u
r
s
u
ed
b
y
s
ec
tio
n
an
d
s
m
o
o
t
h
i
n
g
o
f
t
h
e
i
m
ag
e
a
n
d
co
m
p
leted
to
i
m
p
r
o
v
e
t
h
e
q
u
alit
y
o
f
t
h
e
i
m
a
g
e
[
2
7
]
.
A
n
i
m
a
g
e
o
f
m
aize
lea
f
ca
p
tu
r
ed
th
en
r
esize
to
5
0
0
X
5
0
0
p
ix
els
to
r
ed
u
ce
p
r
o
ce
s
s
in
g
ti
m
e.
T
h
e
i
m
ag
e
s
n
ee
d
ed
to
b
e
en
h
an
ce
s
b
ef
o
r
e
g
o
th
r
o
u
g
h
to
th
e
n
e
x
t
p
r
o
ce
s
s
.
I
m
a
g
es
ar
e
f
ilter
u
s
i
n
g
m
ed
ian
f
il
ter
.
Me
d
ian
f
ilter
is
a
n
o
n
-
li
n
ea
r
p
r
o
ce
s
s
to
r
e
m
o
v
e
u
n
w
a
n
ted
n
o
is
e
a
n
d
o
u
tlier
s
f
r
o
m
t
h
e
in
p
u
t i
m
ag
e
s
.
̂
(
,
)
=
me
dia
n
(
,
)
∈
{
(
,
)
}
(
1
)
T
h
e
f
o
r
m
u
la
ab
o
v
e
is
t
h
e
o
u
tp
u
t o
f
m
ed
ian
f
ilter
w
h
er
e,
f
(
x
,
y
)
is
th
e
o
r
ig
in
al
i
m
a
g
e
an
d
g
(
s
,
t)
is
th
e
o
u
tp
u
t
i
m
a
g
e.
S is
a
t
w
o
-
d
i
m
e
n
s
io
n
al
m
a
s
k
,
w
h
er
e
t
h
e
m
a
s
k
s
ize
is
m
X
m
(
w
h
er
e
m
i
s
u
s
u
a
ll
y
o
d
d
)
s
u
ch
as 3
x
3
,
5
x
5
,
etc.
2
.
3
.
F
e
a
t
ure
e
x
t
ra
ct
io
n
On
ce
i
m
a
g
e
p
r
ep
r
o
ce
s
s
in
g
is
f
in
is
h
ed
,
it i
s
i
m
p
o
r
ta
n
t to
g
et
t
h
e
m
o
s
t i
m
p
o
r
tan
t q
u
al
ities
o
f
th
e
lea
v
es
f
o
r
s
ep
ar
atin
g
t
h
e
m
w
it
h
r
esp
e
ct
o
f
ea
ch
d
ef
icie
n
c
y
[
2
8
]
.
P
r
o
ce
s
s
o
f
ex
tr
ac
ti
n
g
r
elate
d
i
n
f
o
r
m
atio
n
f
r
o
m
i
n
p
u
t
i
m
a
g
e
is
ca
lled
f
ea
t
u
r
e
ex
tr
ac
t
io
n
.
I
t
is
also
to
tr
an
s
f
o
r
m
i
n
g
i
n
p
u
t
i
m
a
g
e
i
n
to
a
s
et
o
f
f
ea
t
u
r
es.
I
n
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
o
lo
g
y
,
a
co
m
p
r
eh
e
n
s
i
v
e
ex
p
er
i
m
e
n
tatio
n
is
ca
r
r
ied
o
u
t
co
n
s
id
er
in
g
te
x
t
u
r
e
f
ea
tu
r
es
f
r
o
m
th
e
m
a
ize
leaf
.
2
.
4
.
T
ex
t
ure
f
ea
t
ure
Gr
a
y
L
e
v
el
C
o
-
o
cc
u
r
r
en
ce
M
atr
ix
(
GL
C
M)
ap
p
r
o
ac
h
d
ef
in
es
th
e
s
h
ap
e
o
f
d
is
tr
ib
u
t
io
n
o
f
v
ar
io
u
s
to
n
es
i
n
te
n
s
i
ties
in
th
e
i
m
ag
e
,
w
h
ic
h
ar
e
d
eter
m
in
ed
b
y
ac
q
u
ir
in
g
t
h
e
co
o
cc
u
r
r
en
ce
m
a
t
r
ices
o
f
th
e
i
m
a
g
e.
GL
C
M
i
s
cr
ea
ted
f
r
o
m
a
g
r
a
y
-
s
ca
le
i
m
a
g
e.
A
co
-
o
cc
u
r
r
en
ce
m
atr
i
x
p
o
r
tr
a
y
s
t
h
e
f
r
eq
u
e
n
c
y
at
w
h
ich
a
s
p
ec
i
f
i
c
g
r
a
y
lev
e
l
is
s
h
o
w
n
in
a
p
ar
ticu
lar
s
p
atial
r
elatio
n
s
h
ip
,
in
r
elatio
n
to
an
o
th
er
g
r
a
y
le
v
el
i
n
an
i
m
a
g
e.
I
n
th
i
s
w
a
y
,
t
h
e
co
-
o
cc
u
r
r
en
ce
m
atr
i
x
is
a
n
o
u
t
lin
e
o
n
h
o
w
t
h
e
p
i
x
el
s
v
al
u
e
s
ar
e
ex
h
ib
ited
alo
n
g
s
i
d
e
to
an
o
th
er
v
al
u
e
in
a
s
m
all
w
i
n
d
o
w
.
B
ec
a
u
s
e
o
f
t
h
eir
e
x
ten
s
i
v
e
d
i
m
e
n
s
io
n
a
lit
y
,
t
h
e
G
L
C
M
's
ar
e
s
e
n
s
iti
v
e
to
th
e
s
ize
o
f
t
h
e
tex
t
u
r
e
s
a
m
p
les
o
n
w
h
ich
t
h
e
y
ar
e
ass
e
s
s
ed
.
C
o
n
s
eq
u
en
t
l
y
,
th
e
q
u
a
n
tit
y
o
f
d
ar
k
d
i
m
e
n
s
io
n
s
i
s
f
r
eq
u
en
t
l
y
d
i
m
in
is
h
ed
.
T
h
e
t
w
o
-
d
i
m
e
n
s
i
o
n
al
ar
r
a
y
d
en
o
ted
b
y
P
o
f
b
o
th
r
o
w
s
a
n
d
co
lu
m
n
s
i
s
s
i
g
n
i
f
y
i
n
g
to
a
s
et
o
f
p
o
s
s
ib
le
g
r
a
y
le
v
el
s
in
i
m
a
g
e
v
alu
e
s
.
P
(
i,
j
|
d
x
,
d
y
)
is
t
h
e
r
e
lativ
e
f
r
eq
u
e
n
c
y
w
i
th
w
h
ic
h
t
w
o
p
ix
el
s
,
s
ep
ar
ated
b
y
a
d
is
ta
n
ce
d
(
d
x
,
d
y
)
.
2
.
4
.
1
.
Co
ntr
a
s
t
I
t
also
ca
lled
as
in
er
tia
o
r
s
u
m
o
f
s
q
u
ar
e
v
ar
ian
ce
.
T
h
e
f
u
n
ctio
n
i
s
to
c
alcu
late
t
h
e
in
te
n
s
it
y
o
f
t
h
e
co
n
tr
ast.
T
h
e
co
n
tr
ast
s
tatis
ti
ca
l
f
ea
tu
r
e
u
s
ed
to
m
ea
s
u
r
e
t
h
e
lo
ca
l
v
ar
iatio
n
s
i
n
th
e
G
L
C
M
an
d
r
esto
r
in
g
a
ca
lcu
latio
n
co
n
tr
ast
o
f
t
h
e
in
t
en
s
it
y
b
et
w
ee
n
t
h
e
p
ix
el
s
an
d
th
eir
n
ei
g
h
b
o
r
s
o
v
er
th
e
w
h
o
le
im
a
g
e.
N
is
an
u
n
k
n
o
w
n
v
al
u
e.
C
o
n
tr
as
t is 0
i
f
th
e
i
m
a
g
e
is
co
n
s
ta
n
t (
d
i
m
e
n
s
io
n
o
f
i
n
p
u
t i
m
a
g
e
s
ize)
.
Eq
u
atio
n
(
2
)
s
h
o
w
s
t
h
e
eq
u
atio
n
f
o
r
co
n
tr
ast.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
N
u
tr
ien
t d
eficien
cy
d
etec
tio
n
in
ma
iz
e
(
Zea
ma
ys L.)
lea
ve
s
u
s
in
g
ima
g
e
p
r
o
ce
s
s
in
g
(
N
u
r
b
a
ity
S
a
b
r
i
)
307
∑
,
−
1
,
=
0
(
−
)
2
(
2
)
2
.
4
.
2
.
Co
rr
ela
t
io
n
C
o
r
r
elatio
n
tex
t
u
r
e
is
a
r
etu
r
n
m
ea
s
u
r
e
o
f
li
n
ea
r
d
ep
en
d
en
cy
g
r
a
y
le
v
el
b
et
w
ee
n
th
e
p
r
ed
eter
m
i
n
ed
p
ix
els
at
t
h
e
s
p
ec
i
f
ied
p
o
s
itio
n
s
r
elat
iv
e
to
th
o
s
e
n
e
ig
h
b
o
r
s
o
v
er
th
e
en
t
ir
e
i
m
ag
e.
Mo
r
e
o
v
er
,
it
is
p
r
o
v
id
ed
s
a
m
e
m
ea
s
u
r
e
m
en
t to
a
u
to
co
r
r
elatio
n
m
e
th
o
d
.
A
co
r
r
elatio
n
is
ca
lcu
la
ted
u
s
i
n
g
th
i
s
f
o
r
m
u
l
a:
∑
,
−
1
,
=
0
[
(
−
)
(
−
√
(
2
)
(
2
)
]
(
3
)
N
is
an
u
n
k
n
o
w
n
v
a
lu
e,
µ
is
m
ea
n
b
ased
o
n
th
e
r
ef
er
en
ce
to
th
e
p
ix
els
an
d
σ
is
a
s
tan
d
ar
d
d
ev
iatio
n
.
C
o
r
r
elatio
n
r
an
g
e
-
1
i
s
f
o
r
n
e
g
ati
v
e
co
r
r
elativ
e
i
m
ag
e
an
d
1
is
f
o
r
p
o
s
itiv
e
co
r
r
elativ
e
i
m
ag
e.
On
t
h
e
o
th
e
r
h
an
d
,
f
o
r
a
co
n
s
ta
n
t i
m
a
g
e
(
d
im
en
s
io
n
o
f
in
p
u
t i
m
ag
e
s
ize)
,
th
e
co
r
r
elatio
n
is
NaN
.
2
.
4
.
3
.
H
o
m
o
g
eneit
y
T
h
e
m
ea
s
u
r
es
o
f
t
h
e
s
m
o
o
th
n
es
s
h
o
m
o
g
e
n
eit
y
d
is
tr
ib
u
tio
n
o
f
th
e
g
r
a
y
le
v
el
o
f
t
h
e
i
m
ag
e
it
i
s
ap
p
r
o
x
im
a
tel
y
i
n
v
er
s
e
l
y
co
r
r
elate
d
w
it
h
co
n
tr
a
s
t.
I
f
co
n
tr
ast
i
s
s
m
al
l,
f
o
r
t
h
e
m
o
s
t
p
ar
t
h
o
m
o
g
en
e
it
y
is
s
u
b
s
ta
n
tial.
I
f
w
eig
h
t
s
d
ec
li
n
e
a
w
a
y
f
r
o
m
t
h
e
d
iag
o
n
al,
t
h
e
ca
lcu
lated
tex
t
u
r
e
m
ea
s
u
r
e
w
il
l
b
e
b
ig
g
er
f
o
r
w
i
n
d
o
w
s
w
i
th
litt
le
d
if
f
er
en
c
e
[
2
9
]
.
R
etu
r
n
h
o
m
o
g
en
ei
t
y
w
ei
g
h
ts
v
al
u
es
b
y
t
h
e
i
n
v
er
s
e
o
f
t
h
e
C
o
n
tr
ast
w
ei
g
h
t,
w
it
h
w
e
ig
h
t
s
d
i
m
in
i
s
h
in
g
ex
p
o
n
e
n
tia
ll
y
f
ar
f
r
o
m
G
L
C
M
d
iag
o
n
al.
N
is
a
n
u
n
k
n
o
wn
v
alu
e.
R
a
n
g
e
o
f
h
o
m
o
g
en
eit
y
i
s
1
if
it is
a
d
ia
g
o
n
al
GL
C
M
.
Ho
m
o
g
en
e
it
y
f
ea
tu
r
e
f
o
r
m
u
la
as
:
∑
,
1
+
(
−
)
2
−
1
,
=
0
(
4
)
2
.
4
.
4
.
E
ntr
o
py
E
n
tr
o
p
y
is
a
s
tati
s
tical
m
ea
s
u
r
e
o
f
r
an
d
o
m
n
ess
th
at
ca
n
b
e
u
tili
ze
d
to
d
escr
ib
e
t
h
e
tex
t
u
r
e
o
f
t
h
e
m
aize
lea
f
i
m
a
g
e
a
n
d
g
e
n
er
a
ll
y
cla
s
s
i
f
ied
a
s
a
f
ir
s
t
-
d
eg
r
e
e
m
ea
s
u
r
e.
T
h
e
i
m
ag
e
s
w
it
h
a
lar
g
er
n
u
m
b
er
o
f
d
is
tr
ib
u
ted
g
r
a
y
le
v
els
h
a
v
e
b
ig
g
er
e
n
tr
o
p
y
.
N
is
a
n
u
n
k
n
o
w
n
v
al
u
e
an
d
l
n
is
t
h
i
s
lo
g
ar
it
h
m
an
d
u
s
e
s
a
b
ase
clo
s
e
to
2
.
7
1
8
s
am
e
w
i
th
lo
g
w
it
h
a
b
ase
o
f
1
0
.
E
n
tr
o
p
y
f
ea
tu
r
e
as f
o
llo
w
i
n
g
f
o
r
m
u
la
:
∑
,
(
−
ln
,
)
−
1
,
=
0
(
5
)
2
.
5
.
Co
l
o
r
f
ea
t
ure
A
h
i
s
to
g
r
a
m
co
u
n
t
s
t
h
e
p
i
x
el
s
n
u
m
b
er
i
n
ea
c
h
f
o
r
m
a
n
d
c
an
b
e
ea
s
il
y
g
e
n
er
ated
b
y
r
ea
d
in
g
ea
c
h
p
ix
el
o
f
ea
ch
i
m
a
g
e
o
n
ce
a
n
d
in
cr
ea
s
i
n
g
th
e
h
is
to
g
r
a
m
's
co
r
r
ec
t
b
in
[
1
3
]
.
T
h
e
h
is
to
g
r
am
f
ea
t
u
r
es
r
ep
r
esen
t
s
tatis
t
ical
-
b
ased
f
ea
tu
r
e
s
,
w
h
er
eb
y
th
e
h
is
to
g
r
a
m
i
s
o
f
t
en
u
s
ed
as
a
r
ep
r
esen
tatio
n
o
f
th
e
d
is
tr
ib
u
tio
n
o
f
lik
eli
h
o
o
d
o
f
th
e
i
m
a
g
e
in
te
n
s
it
y
lev
e
ls
[
3
0
]
.
T
h
e
co
l
o
r
h
is
to
g
r
a
m
f
o
r
an
i
m
a
g
e
is
cr
ea
ted
b
y
m
ap
p
in
g
i
n
p
u
t
v
alu
e
s
i
n
a
s
m
a
ller
s
et
o
f
co
lo
r
s
w
it
h
i
n
t
h
e
i
m
a
g
e
f
r
o
m
a
lar
g
e
s
et
o
f
o
u
tp
u
t
v
alu
e
s
an
d
co
u
n
t
in
g
t
h
e
n
u
m
b
e
r
o
f
p
ix
el
s
o
f
ea
ch
co
lo
r
[
3
1
]
.
S
ca
n
n
i
n
g
t
h
e
i
m
a
g
e,
s
et
tin
g
co
l
o
r
v
alu
e
s
to
t
h
e
h
is
to
g
r
a
m
s
ca
l
e,
an
d
co
n
s
tr
u
cti
n
g
th
e
h
is
to
g
r
a
m
u
s
i
n
g
co
lo
r
attr
ib
u
tes
a
s
in
d
icato
r
s
ar
e
ea
s
y
p
r
o
ce
s
s
es
f
o
r
cr
ea
tin
g
co
lo
r
h
is
to
g
r
a
m
ch
ar
ac
ter
is
tic
s
.
Fo
r
th
is
s
t
u
d
y
,
to
co
m
p
u
te
th
e
co
lo
r
h
is
to
g
r
a
m
is
u
s
in
g
th
e
f
o
llo
w
in
g
p
ar
a
m
eter
s
s
u
ch
a
s
i
m
a
g
es,
ch
a
n
n
els,
m
as
k
,
h
i
s
to
g
r
a
m
s
ize
an
d
r
an
g
es.
2
.
6
.
Sh
a
pe
f
ea
t
ure
Sh
ap
e
i
s
t
h
e
p
r
i
m
ar
y
s
o
u
r
ce
o
f
i
n
f
o
r
m
atio
n
u
s
ed
to
r
ec
o
g
n
i
ze
o
b
j
ec
ts
.
No
v
is
u
al
co
n
te
n
t
o
b
j
ec
t
ca
n
b
e
p
r
o
p
er
ly
r
ec
o
g
n
ized
w
it
h
o
u
t
s
h
ap
e.
Mo
m
e
n
t
i
n
v
ar
ia
n
t
s
a
r
e
ess
en
t
iall
y
th
e
r
e
g
io
n
d
escr
ip
to
r
s
th
at
ar
e
m
o
s
t
p
o
p
u
lar
an
d
w
id
el
y
[
3
2
]
.
2
.
7
.
Ra
nd
o
m
f
o
re
s
t
T
h
e
r
an
d
o
m
f
o
r
est
cla
s
s
i
f
ier
i
s
clo
s
e
to
th
e
to
p
o
f
th
e
class
if
ier
r
an
k
in
g
s
.
R
an
d
o
m
f
o
r
est
co
u
ld
b
e
u
s
ed
f
o
r
cla
s
s
i
f
icat
io
n
as
w
el
l
as
f
o
r
r
eg
r
ess
io
n
.
A
r
an
d
o
m
f
o
r
est
's
cu
m
u
lati
v
e
p
r
ed
ictio
n
er
r
o
r
is
tig
h
tl
y
co
r
r
elate
d
w
i
th
i
n
d
iv
id
u
al
tr
e
es
'
i
n
te
n
s
i
t
y
a
n
d
d
en
s
it
y
i
n
t
h
e
f
o
r
est.
A
d
d
in
g
s
ig
n
i
f
ica
n
t
r
an
d
o
m
n
e
s
s
i
n
t
h
e
b
ase
m
o
d
els,
tr
ee
s
a
n
d
cr
ea
ti
n
g
s
u
b
s
ets
o
f
th
e
p
r
ed
icto
r
s
y
s
te
m
ca
n
r
e
f
i
n
es
b
a
g
g
in
g
to
s
ep
ar
ate
th
e
a
tr
ee
n
o
d
es
o
f
r
an
d
o
m
f
o
r
est
[
3
3
]
.
I
n
th
e
tr
ain
i
n
g
p
h
a
s
e,
it
a
u
to
m
atica
ll
y
ca
lc
u
late
s
t
h
e
ap
p
r
o
p
r
iate
s
co
r
e
f
o
r
ea
c
h
ele
m
e
n
t.
Af
ter
t
h
at
it
s
ca
les
d
o
w
n
th
e
s
i
g
n
if
ican
ce
f
o
r
t
h
e
t
o
tal
o
f
all
s
co
r
es
to
b
e
1
.
T
h
e
o
v
er
all
ex
p
la
n
atio
n
ca
n
b
e
d
ef
in
ed
b
y
u
s
i
n
g
t
h
e
Gi
n
i in
d
e
x
.
=
1
−
∑
(
)
2
=
1
(
6
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
2
,
J
u
n
e
20
20
:
3
04
–
3
09
308
I
n
ad
d
itio
n
,
ab
o
v
e
eq
u
at
io
n
i
n
clu
d
e
s
t
h
e
clas
s
a
n
d
li
k
eli
h
o
o
d
to
d
ec
id
e
w
h
ic
h
b
r
an
c
h
Gi
n
i
i
s
m
o
s
t
li
k
el
y
t
o
o
cc
u
r
o
n
a
n
o
d
e.
T
h
u
s
,
p
i
is
t
h
e
ab
s
o
l
u
te
f
r
eq
u
e
n
c
y
o
f
t
h
e
class
th
at
w
il
l
b
e
f
i
n
d
s
in
t
h
e
d
ataset,
an
d
c
i
s
t
h
e
n
u
m
b
er
o
f
clas
s
es.
T
h
e
f
u
n
cti
o
n
o
f
clas
s
i
f
icatio
n
p
r
o
ce
s
s
i
s
to
class
i
f
y
i
m
a
g
e
ac
co
r
d
in
g
to
th
e
t
y
p
e
o
f
n
u
tr
ie
n
t
d
ef
icien
c
y
a
n
d
its
ac
cu
r
ac
y
p
e
r
ce
n
tag
e.
I
n
t
h
i
s
r
esear
ch
3.
RE
SU
L
T
S
A
ND
D
I
SCU
SS
I
O
N
T
h
e
test
in
g
r
es
u
lt
o
f
t
h
e
n
u
tr
ien
t
d
ef
icien
c
y
o
f
m
aize
lea
f
is
d
iv
id
ed
in
to
t
h
r
ee
t
y
p
e
s
w
h
ic
h
ar
e
Ma
g
n
e
s
i
u
m
,
Nitr
o
g
e
n
,
P
o
tass
iu
m
an
d
h
ea
lt
h
y
s
h
o
w
s
i
n
T
ab
le
1
.
T
h
e
class
if
ier
s
h
o
w
s
t
h
e
m
a
x
i
m
u
m
p
r
o
b
a
b
ilit
y
f
o
r
t
h
e
clas
s
it
p
r
ed
icts
.
C
o
n
f
u
s
io
n
m
atr
i
x
is
u
s
e
d
to
ca
lcu
late
ac
cu
r
ac
y
p
er
ce
n
tag
e
f
o
r
t
h
e
o
v
er
al
l
s
y
s
te
m
.
T
h
e
r
es
u
lt
o
f
p
o
t
ass
i
u
m
d
etec
t
io
n
e
f
f
ec
ted
b
y
i
ts
s
i
m
ilar
it
y
w
it
h
m
a
g
n
e
s
i
u
m
f
ea
t
u
r
es.
T
h
u
s
,
p
o
tass
iu
m
h
as
t
h
e
lo
w
e
s
t
n
u
m
b
er
o
f
co
r
r
ec
t
d
etec
tio
n
as
it
m
o
s
tl
y
d
etec
t
s
it
as
a
m
ag
n
e
s
i
u
m
a
s
it
h
a
s
th
e
clo
s
er
f
ea
t
u
r
es
as
m
a
g
n
e
s
iu
m
.
T
h
e
ac
cu
r
ac
y
ac
h
ie
v
e
7
8
.
3
5
p
er
ce
n
t
as
co
n
s
eq
u
e
n
ce
o
f
lo
w
d
etec
tio
n
o
n
p
o
tass
iu
m
.
T
ab
le
1
.
R
esu
lt o
f
t
h
e
id
en
ti
f
ic
atio
n
Ty
p
e
N
u
mb
e
r
o
f
T
e
st
e
d
I
mag
e
N
u
mb
e
r
o
f
I
mag
e
I
d
e
n
t
i
f
i
e
d
C
o
r
r
e
c
t
l
y
N
u
mb
e
r
o
f
I
mag
e
I
d
e
n
t
i
f
i
e
d
I
n
c
o
r
r
e
c
t
l
y
H
e
a
l
t
h
y
31
31
0
M
a
g
n
e
si
u
m
35
35
0
N
i
t
r
o
g
e
n
36
26
10
P
o
t
a
ssi
u
m
32
11
21
%
=
100
(
7
)
4.
CO
NCLU
SI
O
N
T
h
is
r
esear
ch
class
i
f
ie
s
f
o
u
r
ty
p
e
o
f
class
w
h
ic
h
is
h
ea
lt
y
le
af
,
n
u
tr
i
en
t,
m
a
g
n
e
s
i
u
a
m
a
n
d
p
o
tass
iu
m
.
So
m
e
o
f
t
h
e
n
u
tr
ien
t
d
ef
ic
ie
n
c
y
is
li
k
el
y
to
h
av
e
t
h
e
s
a
m
e
tr
ait,
th
er
e
f
o
r
e
it
is
d
if
f
ic
u
lt
to
clas
s
i
f
y
th
e
s
e
n
u
tr
ie
n
t.
Mo
r
e
d
ata
w
i
ll
b
e
c
o
llected
in
t
h
e
f
u
t
u
r
e
b
y
ad
d
i
n
g
m
o
r
e
tr
ai
n
i
n
g
d
ata
f
o
r
class
i
f
icatio
n
p
r
o
ce
s
s
.
B
esid
es,
th
e
s
a
m
e
d
ataset
will
b
e
test
ed
w
ith
o
th
er
av
a
i
lab
le
m
ac
h
i
n
e
lear
n
i
n
g
to
in
cr
ea
s
e
th
e
cu
r
r
en
t
ac
cu
r
ac
y
ac
h
iev
e
b
y
r
a
n
d
o
m
f
o
r
est cla
s
s
i
f
ier
.
ACK
NO
WL
E
D
G
E
M
E
NT
T
h
e
au
th
o
r
s
w
o
u
ld
li
k
e
to
th
an
k
th
e
M
in
i
s
tr
y
o
f
Hig
h
er
E
d
u
ca
tio
n
,
Ma
la
y
s
ia
an
d
Un
i
v
er
s
it
i
T
ek
n
o
lo
g
i M
A
R
A
f
o
r
th
e
r
ese
ar
ch
f
u
n
d
i
n
g
a
n
d
s
u
p
p
o
r
t v
ia
g
r
an
t n
u
m
b
er
6
0
0
-
I
R
MI
/F
R
GS
5
/3
(
2
1
5
/2
0
1
9
)
.
RE
F
E
R
E
NC
E
S
[1
]
S
.
Ha
k
e
&
J.
Ro
ss
-
Ib
a
rra
,
“
G
e
n
e
ti
c
Ev
o
lu
ti
o
n
a
ry
a
n
d
P
la
n
t
Bre
e
d
in
g
In
sig
h
ts
f
ro
m
T
h
e
Do
m
e
stica
ti
o
n
o
f
M
a
ize
,
“
EL
if
e
,
4
,
p
p
.
1
–
8
,
2
0
1
5
.
[2
]
L
.
Na
ir
&
K.
K.
S
a
ju
,
”
Clas
si
f
ic
a
ti
o
n
o
f
M
a
c
ro
n
u
tri
e
n
t
De
f
icie
n
c
ies
in
M
a
ize
P
lan
t
Us
in
g
M
a
c
h
in
e
L
e
a
rn
in
g
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
),
8
(
6
),
p
p
.
4
1
9
7
–
4
2
0
3
,
2
0
1
8
.
[3
]
S
.
B.
Ja
d
h
a
v
e
t
a
l.
,
“
Co
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s
f
o
r
le
a
f
i
m
a
g
e
-
b
a
se
d
p
lan
t,
”
IAE
S
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Arti
fi
c
ia
l
I
n
telli
g
e
n
c
e
(
IJ
-
AI)
,
p
p
.
3
2
8
-
3
4
1
,
2
0
1
8
.
[4
]
S
.
S
ri
d
a
ra
n
e
,
&
A
.
S
.
V
ij
e
n
d
ra
n
,
“
A
n
a
l
y
sis
o
f
M
a
ize
Cro
p
L
e
a
f
u
sin
g
M
u
l
ti
v
a
riate
Im
a
g
e
A
n
a
l
y
sis
f
o
r
Id
e
n
ti
fy
in
g
S
o
il
De
f
icie
n
c
y
A
n
a
l
y
sis
o
f
M
a
iz
e
C
ro
p
L
e
a
f
u
sin
g
M
u
lt
iv
a
riate
I
m
a
g
e
A
n
a
l
y
sis
f
o
r
Id
e
n
ti
fy
in
g
S
o
il
De
f
icie
n
c
y
,
”
Res
e
a
rc
h
J
o
u
rn
a
l
o
f
A
p
p
li
e
d
S
c
ie
n
c
e
s,
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
8
(1
9
),
2
0
1
4
.
[5
]
B.
Ch
it
ra
d
e
v
i
&
P
.
S
rim
a
th
i,
”
A
n
Ov
e
rv
i
e
w
o
n
Im
a
g
e
P
ro
c
e
ss
in
g
T
e
c
h
n
iq
u
e
s,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
n
o
v
a
ti
v
e
Res
e
a
rc
h
in
C
o
mp
u
ter
a
n
d
C
o
mm
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
2
(
1
1
N
o
v
2
0
1
4
),
p
p
.
6
4
6
6
–
6
4
7
2
,
2
0
1
4
.
[6
]
S
.
Ba
rh
m
i
&
F
a
tn
i
El
,
O.,
”
Ho
u
rly
W
in
d
S
p
e
e
d
F
o
re
c
a
stin
g
Ba
se
d
o
n
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
,
”
IAE
S
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Arti
fi
c
i
a
l
I
n
telli
g
e
n
c
e
(
IJ
-
AI)
,
p
p
.
2
8
6
-
2
9
1
,
2
0
1
9
.
[7
]
A
.
Ch
a
u
d
h
a
ry
&
S
.
S
.
S
in
g
h
,
”
L
u
n
g
Ca
n
c
e
r
D
e
tec
ti
o
n
u
sin
g
Dig
it
a
l
I
m
a
g
e
P
ro
c
e
ss
in
g
,
”
Res
e
a
rc
h
J
o
u
rn
a
l
o
f
En
g
i
n
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
3
8
(
2
),
1
3
5
1
–
1
3
5
9
,
2
0
1
2
.
[8
]
D.
T
ian
,
“
A
Re
v
ie
w
o
n
Im
a
g
e
F
e
a
tu
re
Ex
trac
ti
o
n
a
n
d
Re
p
re
se
n
tatio
n
T
e
c
h
n
iq
u
e
s
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
M
u
lt
ime
d
ia
a
n
d
U
b
iq
u
it
o
u
s E
n
g
i
n
e
e
rin
g
,
8
(4
)
,
2
0
1
3
.
[9
]
M
.
S
h
a
rm
a
&
B.
S
in
g
h
,
“
F
e
a
tu
re
Ex
trac
ti
o
n
a
n
d
A
n
a
l
y
sis
u
sin
g
G
a
b
o
r
F
il
ter
a
n
d
Hig
h
e
r
Ord
e
r
S
tatisti
c
s
f
o
r
th
e
JP
EG
S
teg
a
n
o
g
ra
p
h
y
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
p
p
li
e
d
E
n
g
in
e
e
rin
g
Res
e
a
rc
h
,
1
3
(5
),
p
p
.
2
9
4
5
–
2
9
5
4
,
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
N
u
tr
ien
t d
eficien
cy
d
etec
tio
n
in
ma
iz
e
(
Zea
ma
ys L.)
lea
ve
s
u
s
in
g
ima
g
e
p
r
o
ce
s
s
in
g
(
N
u
r
b
a
ity
S
a
b
r
i
)
309
[1
0
]
Y.
W
ica
k
so
n
o
&
V.
S
u
h
a
rt
o
n
o
,
“
Co
lo
r
a
n
d
T
e
x
tu
re
F
e
a
tu
re
Ex
trac
ti
o
n
Us
i
n
g
G
a
b
o
r
F
il
ter
-
L
o
c
a
l
Bin
a
ry
P
a
tt
e
rn
s
f
o
r
I
m
a
g
e
S
e
g
m
e
n
tatio
n
w
it
h
F
u
z
z
y
C
-
M
e
a
n
s,”
J
o
u
rn
a
l
o
f
In
telli
g
e
n
t
S
y
ste
ms
,
1
(
1
),
p
p
.
1
5
–
2
1
,
2
0
1
5
.
[1
1
]
X
.
Zh
a
n
g
e
t
a
l.
,
“
A
S
tu
d
y
f
o
r
T
e
x
tu
re
F
e
a
tu
re
Ex
tr
a
c
ti
o
n
Of
Hig
h
-
Re
so
lu
ti
o
n
S
a
telli
te
Im
a
g
e
s
Ba
se
d
o
n
A
Dire
c
ti
o
n
M
e
a
su
re
a
n
d
G
ra
y
L
e
v
e
l
Co
-
Oc
c
u
rre
n
c
e
M
a
tri
x
F
u
si
o
n
A
l
g
o
rit
h
m
,
”
S
e
n
so
rs
,
1
7
(7
)
,
2
0
1
7
.
[1
2
]
A
.
M
a
so
o
d
&
M
.
S
h
a
rif
,
“
Co
n
te
n
t
-
Ba
se
d
Im
a
g
e
R
e
tri
e
v
a
l
F
e
a
tu
re
s :
A
S
u
rv
e
y
,
”
In
t.
J
.
Ad
v
a
n
c
e
d
Ne
two
rk
in
g
a
n
d
Ap
p
li
c
a
ti
o
n
s
,
1
0
(
1
),
p
p
.
3
7
4
1
–
3
7
5
7
,
2
0
1
8
.
[1
3
]
F
.
A
la
m
d
a
r
&
M
.
R.
Ke
y
v
a
n
p
o
u
r,
“
A
Ne
w Co
lo
r
F
e
a
tu
re
Ex
trac
ti
o
n
M
e
th
o
d
Ba
se
d
On
Qu
a
d
Histo
g
ra
m
,
”
Pro
c
e
d
ia
En
v
iro
n
me
n
ta
l
S
c
ien
c
e
s,
1
0
(
PA
RT
A),
p
p
.
7
7
7
–
7
8
3
,
2
0
1
1
.
[1
4
]
T
.
S
u
ra
sa
k
,
e
t
a
l.
,
”
Histo
g
ra
m
O
f
Orie
n
ted
G
ra
d
ien
ts
f
o
r
Hu
m
a
n
De
tec
ti
o
n
in
V
i
d
e
o
,
”
Pro
c
e
e
d
in
g
s
o
f
2
0
1
8
5
th
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
B
u
sin
e
ss
a
n
d
I
n
d
u
stri
a
l
Res
e
a
rc
h
:
S
ma
rt
T
e
c
h
n
o
lo
g
y
fo
r
Ne
x
t
Ge
n
e
ra
ti
o
n
o
f
In
fo
rm
a
t
io
n
,
E
n
g
i
n
e
e
rin
g
,
B
u
sin
e
ss
a
n
d
S
o
c
ia
l
S
c
ien
c
e
,
ICBIR
2
0
1
8
,
p
p
.
1
7
2
–
1
7
6
,
2
0
1
8
.
[1
5
]
M
.
Ja
m
sh
e
d
e
t,
a
l.
,
”
S
ig
n
if
ica
n
t
HO
G
-
Histo
g
ra
m
o
f
Ori
e
n
ted
G
r
a
d
ien
t
F
e
a
tu
re
S
e
lec
ti
o
n
f
o
r
Hu
m
a
n
De
tec
ti
o
n
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
Ap
p
l
ica
ti
o
n
s
,
1
3
2
(1
7
),
p
p
.
2
0
–
2
4
,
2
0
1
5
.
[1
6
]
S
.
Zek
o
v
ich
&
M
.
T
u
b
a
,
“
Hu
M
o
m
e
n
ts
Ba
s
e
d
Ha
n
d
w
rit
ten
Dig
it
s
Re
c
o
g
n
it
io
n
A
lg
o
rit
h
m
,
”
Re
c
e
n
t
A
d
v
a
n
c
e
s
in
Kn
o
w
led
g
e
En
g
in
e
e
rin
g
a
n
d
S
y
ste
m
s S
c
ien
c
e
,
p
p
.
9
8
–
1
0
3
,
2
0
1
3
.
[1
7
]
M
.
F
e
rn
a
n
d
o
,
”
No
v
e
l
A
p
p
ro
a
c
h
to
Us
e
HU
M
o
m
e
n
ts
w
it
h
I
m
a
g
e
P
ro
c
e
ss
in
g
T
e
c
h
n
iq
u
e
s
f
o
r
Re
a
l
T
i
m
e
S
ig
n
L
a
n
g
u
a
g
e
Co
m
m
u
n
ica
ti
o
n
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Ima
g
e
Pro
c
e
s
sin
g
(
IJ
IP)
,
9
(
6
),
p
p
.
3
3
5
–
3
4
5
,
2
0
1
5
.
[1
8
]
S
.
S
.
Ko
t
h
a
w
a
le,
e
t
a
l.
,
“
G
ra
p
e
Lea
f
Dise
a
se
D
e
tec
ti
o
n
Us
in
g
S
VM
Clas
sif
ier,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
n
o
v
a
t
ive
Res
e
a
rc
h
in
C
o
mp
u
ter
a
n
d
C
o
mm
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
2
0
1
8
.
[1
9
]
D.
A
n
g
u
it
a
e
t
a
l.
,
“
M
o
d
e
l
S
e
lec
ti
o
n
f
o
r
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
s:
A
d
v
a
n
tag
e
s
a
n
d
Disa
d
v
a
n
tag
e
s
o
f
th
e
M
a
c
h
in
e
L
e
a
rn
in
g
T
h
e
o
r
y
,
”
T
h
e
2
0
1
0
i
n
ter
n
a
ti
o
n
a
l
j
o
i
n
t
c
o
n
fer
e
n
c
e
o
n
n
e
u
r
a
l
n
e
tw
o
rk
s (
IJ
CNN)
,
p
p
.
1
-
8
,
2
0
1
0
.
[2
0
]
S
.
W
a
ll
e
li
g
n
,
”
S
o
y
b
e
a
n
P
lan
t
Dis
e
a
se
Id
e
n
ti
f
ica
ti
o
n
Us
in
g
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
t
w
o
r
k
,
”
Arti
fi
c
i
a
l
In
telli
g
e
n
c
e
Res
e
a
rc
h
S
o
c
iety
Co
n
fer
e
n
c
e
(
FL
AIR
S
-
3
1
)
S
o
y
b
e
a
n
,
p
p
.
1
4
6
–
1
5
1
,
2
0
1
7
.
[2
1
]
R.
J.
Ne
lso
n
,
e
t
a
l.
“
A
u
to
m
a
ted
Id
e
n
ti
f
ica
ti
o
n
o
f
No
rt
h
e
rn
L
e
a
f
B
li
g
h
t
-
In
f
e
c
ted
M
a
ize
P
lan
ts
f
ro
m
F
ield
Im
a
g
e
r
y
Us
in
g
De
e
p
L
e
a
rn
in
g
,
”
Ph
y
to
p
a
t
h
o
l
o
g
y
,
1
0
7
(1
1
),
p
p
.
1
4
2
6
–
1
4
3
2
,
2
0
1
7
.
[2
2
]
C.
De
Ch
a
n
t,
e
t
a
l.
,
"
A
u
to
m
a
ted
Id
e
n
ti
f
ica
ti
o
n
Of
No
rth
e
rn
L
e
a
f
B
li
g
h
t
-
In
f
e
c
ted
M
a
ize
P
lan
ts
F
ro
m
F
ield
Im
a
g
e
r
y
Us
in
g
De
e
p
L
e
a
rn
in
g
,
”
p
p
.
1
4
2
6
-
1
4
3
2
,
2
0
1
7
.
[2
3
]
H.
W
a
n
g
,
e
t
a
l.
,
“
A
h
y
b
rid
CNN
F
e
a
tu
re
M
o
d
e
l
f
o
r
P
u
lm
o
n
a
ry
No
d
u
le
M
a
li
g
n
a
n
c
y
Risk
D
iffere
n
ti
a
ti
o
n
,
”
Ima
g
i
n
g
fo
r P
a
ti
e
n
t
-
C
u
sto
mize
d
S
imu
l
a
ti
o
n
s a
n
d
S
y
ste
ms
fo
r P
o
in
t
-
of
-
Ca
re
Ultra
so
u
n
d
,
p
p
.
1
9
-
2
6
,
2
0
1
7
.
[2
4
]
D.
Na
y
a
k
,
e
t
a
l.
,
“
Bra
in
M
R
I
m
a
g
e
Clas
sif
i
c
a
ti
o
n
Us
in
g
T
w
o
-
Dim
e
n
sio
n
a
l
Disc
re
te
W
a
v
e
l
e
t
T
ra
n
sf
o
r
m
A
n
d
A
d
a
b
o
o
st W
it
h
Ra
n
d
o
m
F
o
re
sts,“
Ne
u
ro
c
o
mp
u
ti
n
g
,
p
p
.
1
8
8
–
1
9
7
,
2
0
1
6
.
[2
5
]
D.
De
n
isk
o
&
M
.
M
.
Ho
f
fm
a
n
,
“
Clas
si
f
ica
ti
o
n
a
n
d
In
tera
c
ti
o
n
In
Ra
n
d
o
m
F
o
re
sts,”
Pro
c
e
e
d
in
g
s
o
f
th
e
Na
ti
o
n
a
l
Aca
d
e
my
o
f
S
c
ien
c
e
s o
f
t
h
e
Un
i
ted
S
t
a
tes
o
f
Ame
ric
a
,
1
1
5
(8
),
p
p
.
1
6
9
0
–
1
6
9
2
,
”
2
0
1
8
.
[2
6
]
M
.
A
n
it
h
a
&
K.
Ka
a
rth
ik
,
“
A
n
a
ly
sis
o
f
Nu
tri
e
n
t
Re
q
u
irem
e
n
t
o
f
Cro
p
s
Us
in
g
Its
L
e
a
f
,
”
J
o
u
rn
a
l
o
f
Ch
e
mic
a
l
a
n
d
Ph
a
rm
a
c
e
u
t
ica
l
S
c
ien
c
e
s
,
(8
),
p
p
.
9
9
–
1
0
3
,
2
0
1
6
.
[2
7
]
N.
G
.
Dh
a
k
a
d
e
t
a
l.
,
“
L
e
a
f
Dis
e
a
se
De
tec
ti
o
n
Us
in
g
I
m
a
g
e
P
ro
c
e
ss
in
g
f
o
r
P
e
sticid
e
s
S
p
ra
y
in
g
,
“
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
E
n
g
i
n
e
e
rin
g
a
n
d
Res
e
a
rc
h
De
v
e
lo
p
me
n
t,
4
(
0
4
),
p
p
.
6
8
7
–
6
8
9
,
2
0
1
8
.
[2
8
]
V
.
M
a
rc
e
lo
,
e
t
a
l.
,
“
A
u
to
m
a
ti
c
D
e
tec
ti
o
n
o
f
Nu
tri
ti
o
n
a
l
De
f
icie
n
c
ies
In
Co
f
f
e
e
T
re
e
L
e
a
v
e
s
T
h
ro
u
g
h
S
h
a
p
e
A
n
d
T
e
x
tu
re
De
sc
rip
to
rs,”
J
o
u
rn
a
l
o
f
Dig
it
a
l
In
f
o
rm
a
ti
o
n
M
a
n
a
g
e
me
n
t,
1
5
(
1
),
2
0
1
7
.
[2
9
]
G
.
K
a
u
sh
a
l
&
R.
Ba
la,“
GL
CM
a
n
d
KN
N
b
a
se
d
A
l
g
o
rit
h
m
f
o
r
P
la
n
t
Dise
a
se
D
e
tec
ti
o
n
,
“
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
Res
e
a
rc
h
in
El
e
c
trica
l
El
e
c
tro
n
ics
a
n
d
I
n
stru
me
n
t
a
ti
o
n
En
g
i
n
e
e
rin
g
,
6
(
7
),
p
p
.
5
8
4
5
–
5
8
5
2
.
[3
0
]
S
.
S
e
rg
y
á
n
,
“
Co
lo
r
Histo
g
ra
m
F
e
a
tu
re
s
Ba
s
e
d
I
m
a
g
e
Cla
ss
i
f
ica
t
i
o
n
In
Co
n
ten
t
-
Ba
se
d
Im
a
g
e
R
e
tri
e
v
a
l
S
y
ste
m
s,”
S
AM
I
2
0
0
8
6
th
I
n
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m
o
n
A
p
p
li
e
d
M
a
c
h
i
n
e
In
telli
g
e
n
c
e
a
n
d
I
n
fo
rm
a
ti
c
s
-
Pro
c
e
e
d
in
g
s
,
(F
e
b
ru
a
ry
2
0
0
8
),
p
p
.
2
2
1
–
2
2
4
,
2
0
1
7
.
[3
1
]
K.
Ro
y
&
J.
M
u
k
h
e
rjee
,
”
I
m
a
g
e
S
im
il
a
rit
y
M
e
a
su
re
u
sin
g
Co
lo
r
Histo
g
ra
m
,
C
o
lo
r
Co
h
e
re
n
c
e
V
e
c
to
r,
a
n
d
S
o
b
e
l
M
e
th
o
d
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
S
c
ien
c
e
a
n
d
Res
e
a
rc
h
,
2
(
1
),
p
p
.
2
3
1
9
–
7
0
6
4
,
2
0
1
3
.
[3
2
]
A
.
T
h
o
m
a
s
&
K.
S
re
e
k
u
m
a
r,
“
A
S
u
rv
e
y
o
n
Im
a
g
e
F
e
a
tu
re
De
s
c
r
ip
to
rs
-
C
o
lo
r
,
S
h
a
p
e
a
n
d
T
e
x
tu
re
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
In
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
ies
,
5
(
6
),
p
p
.
7
8
4
7
–
7
8
5
0
,
2
0
1
4
.
[3
3
]
D.
M
o
h
a
m
m
e
d
,
e
t
a
l.
,
”
Ov
e
rl
a
p
p
e
d
M
u
sic
S
e
g
m
e
n
tatio
n
Us
in
g
A
Ne
w
Eff
e
c
ti
v
e
F
e
a
tu
re
a
n
d
R
a
n
d
o
m
F
o
re
sts,”
IAE
S
I
n
ter
n
a
t
io
n
a
l
J
o
u
r
n
a
l
o
f
Art
if
icia
l
I
n
telli
g
e
n
c
e
(
IJ
-
AI)
,
p
p
.
1
8
1
-
1
8
9
,
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.