I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
4
,
A
ugus
t
2025
, pp.
2935
~
2944
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
4
.pp
2935
-
2944
2935
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
Im
age
an
al
ysi
s a
n
d
m
ac
h
i
n
e
l
e
ar
n
i
n
g
t
e
c
h
n
i
q
u
e
s f
or
ac
c
u
r
at
e
d
e
t
e
c
t
i
on
of
c
om
m
on
m
an
go
d
i
se
ase
s i
n
w
ar
m
c
l
i
m
at
e
s
M
d
A
b
d
u
ll
ah
A
l
R
ah
i
b
, N
az
n
in
S
u
l
t
an
a, N
ir
j
h
or
S
a
h
a, R
aj
u
M
ia
, M
on
is
h
a S
ar
k
e
r
, A
b
d
u
s
S
at
t
ar
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
S
c
i
e
nc
e
a
nd E
ngi
ne
e
r
i
ng, D
a
f
f
odi
l
I
nt
e
r
na
t
i
ona
l
U
ni
v
e
r
s
i
t
y, D
ha
ka
, B
a
ngl
a
de
s
h
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
M
a
r
3, 202
4
R
e
vi
s
e
d
F
e
b 25, 2025
A
c
c
e
pt
e
d
M
a
r
15, 2025
Mangoes
are
valuable
crops
gro
wn
in
warm
climates,
but
they
often
suffer
from
diseases
that
harm
both
the
trees
and
the
fruits.
This
paper
prop
oses
a
new
way
to
use
machine
learning
to
detect
these
diseases
early
in
mango
plants.
We
focused
on
common
issues
like
mango
fruit
disease
s,
leaf
diseases,
powdery
mildew,
anthracnose/blossom
blight,
and
dieback
,
which
are
particularly
problematic
in
places
like
Bangladesh.
Our
method
st
arts
by
improving
the
quality
of
images
of
mango
plants
and
then
ext
racting
impo
rtant
features
from
these
images.
We
use
a
technique
called
k
-
means
clusteri
ng
to
divide
the
images
into
meaningfu
l
parts
for
analysis
.
After
extractin
g
ten
key
features,
we
tested
various
ways
to
classify
the
di
seases.
The
random
forest
algorithm
stood
out,
accurately
identifying
disease
s
with
a 97.44%
success rat
e. This
research is
crucial fo
r Banglad
esh,
where
mango
farming
is
essential
for
the
economy.
By
spottin
g
diseases
early,
we
can
improve
mango
production,
quality,
and
the
livelihoods
of
farmer
s
.
This
automated
system
offers
a
practical
way
to
manage
mango
disea
ses
in
regions wit
h simil
ar climates.
K
e
y
w
o
r
d
s
:
D
is
e
a
s
e
de
te
c
ti
on
F
e
a
tu
r
e
e
xt
r
a
c
ti
on
I
m
a
ge
pr
oc
e
s
s
in
g
M
a
c
hi
ne
l
e
a
r
ni
ng
M
a
ngo f
r
ui
t
R
a
ndom f
or
e
s
t
a
lg
or
it
hm
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
M
d A
bdul
la
h A
l
R
a
hi
b
D
e
pa
r
tm
e
nt
of
C
om
put
e
r
S
c
ie
nc
e
a
nd E
ngi
ne
e
r
in
g, D
a
f
f
odi
l
I
nt
e
r
na
ti
ona
l
U
ni
ve
r
s
it
y
D
ha
ka
, B
a
ngl
a
d
e
s
h
E
m
a
il
:
a
bdul
la
h15
-
12247@
di
u.e
du.bd
1.
I
N
T
R
O
D
U
C
T
I
O
N
M
a
ngoe
s
a
r
e
m
or
e
th
a
n
ju
s
t
a
popula
r
f
r
ui
t
in
S
out
he
a
s
t
A
s
ia
;
th
e
y
a
r
e
a
vi
ta
l
c
om
pone
nt
of
th
e
a
gr
ic
ul
tu
r
a
l
e
c
onomy.
I
n
c
ount
r
ie
s
li
ke
B
a
ngl
a
de
s
h,
m
a
ngo
pr
oduc
ti
on
s
uppor
ts
m
il
li
ons
of
li
ve
li
hoods
a
nd
pl
a
ys
a
s
ig
ni
f
ic
a
nt
r
ol
e
in
dr
iv
in
g
e
c
onomi
c
gr
ow
th
.
H
ow
e
ve
r
,
th
is
c
r
uc
ia
l
s
e
c
to
r
f
a
c
e
s
s
ig
ni
f
ic
a
nt
c
ha
ll
e
nge
s
due
to
di
s
e
a
s
e
s
th
a
t
a
f
f
e
c
t
m
a
ngo
pl
a
nt
s
,
le
a
di
ng
to
s
ub
s
ta
nt
ia
l
lo
s
s
e
s
in
yi
e
ld
a
nd
qu
a
li
ty
.
A
ddr
e
s
s
in
g
th
e
s
e
c
ha
ll
e
nge
s
i
s
e
s
s
e
nt
ia
l
f
or
s
us
ta
in
in
g t
he
e
c
onomi
c
a
nd c
ul
tu
r
a
l
im
por
ta
nc
e
of
m
a
ngo pr
oduc
ti
on. Agr
ic
ul
tu
r
a
l
di
s
e
a
s
e
s
,
pa
r
ti
c
ul
a
r
ly
th
os
e
a
f
f
e
c
ti
ng
m
a
ngo
c
r
ops
,
c
a
n
ha
ve
de
va
s
ta
ti
ng
e
f
f
e
c
ts
on
bot
h
th
e
lo
c
a
l
a
nd
gl
oba
l
e
c
onomy.
F
or
in
s
ta
nc
e
,
S
a
e
e
d
e
t
al
.
[
1]
ha
ve
id
e
nt
if
ie
d
L
as
io
di
pl
odi
a
t
he
obr
om
ae
a
s
th
e
c
a
us
a
ti
ve
a
ge
nt
of
m
a
ngo
di
e
ba
c
k
di
s
e
a
s
e
in
th
e
U
ni
te
d
A
r
a
b
E
m
ir
a
te
s
,
w
it
h
s
ys
te
m
ic
f
ungi
c
id
e
s
li
ke
C
id
e
ly
®
T
op
s
how
in
g
pr
om
is
e
f
or
m
a
na
ge
m
e
nt
.
K
ha
n
e
t
al
.
[
2]
e
m
pl
oy
l
a
s
e
r
in
duc
e
d
br
e
a
kdown
s
pe
c
tr
os
c
opy
(
L
I
B
S
)
to
a
na
ly
z
e
e
le
m
e
nt
a
l
c
om
pos
it
io
n i
n m
a
ngo pulp pos
t
-
ha
r
ve
s
t,
r
e
ve
a
li
ng q
ua
li
ta
ti
ve
a
nd qua
nt
it
a
ti
ve
de
te
c
ti
on of
or
ga
ni
c
a
nd
m
in
e
r
a
l
e
le
m
e
nt
s
,
w
it
h
im
pl
ic
a
ti
ons
f
or
nut
r
it
io
n
a
nd
he
a
lt
h.
M
ia
e
t
al
.
[
3]
in
tr
oduc
e
a
nove
l
n
e
ur
a
l
ne
twor
k
e
ns
e
m
bl
e
(
N
N
E
)
f
or
m
a
ngo
le
a
f
di
s
e
a
s
e
r
e
c
ogni
ti
on
(
M
L
D
R
)
th
a
t
of
f
e
r
s
a
n
e
f
f
ic
ie
nt
a
lt
e
r
na
ti
ve
,
a
c
hi
e
vi
ng
80%
a
c
c
ur
a
c
y
a
nd
pot
e
nt
ia
ll
y
e
nha
nc
in
g
pr
oduc
ti
on.
K
um
a
r
i
e
t
a
l.
[
4
]
c
ont
r
ib
ut
e
va
lu
a
bl
e
in
s
ig
ht
s
in
to
m
a
ngo
di
s
e
a
s
e
m
a
na
ge
m
e
nt
s
tr
a
te
gi
e
s
,
ge
ne
ti
c
id
e
nt
if
i
c
a
ti
on,
a
nd
di
s
e
a
s
e
dyn
a
m
ic
s
,
a
id
in
g
in
th
e
e
nha
nc
e
m
e
nt
of
m
a
ngo
c
ul
ti
va
ti
on
pr
a
c
ti
c
e
s
a
nd
e
n
s
ur
in
g
s
us
ta
in
a
bl
e
pr
oduc
ti
on.
P
ha
m
e
t
al
.
[
5]
a
ppr
oa
c
he
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 4, A
ugus
t
2025
:
2935
-
2944
2936
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
k (
A
N
N
)
a
ppr
oa
c
h t
ha
t
s
ur
pa
s
s
e
s
c
onvol
ut
io
na
l
ne
ur
a
l
ne
twor
k
(
C
N
N
)
m
ode
ls
i
n e
a
r
ly
di
s
e
a
s
e
de
te
c
ti
on,
be
ne
f
it
ti
ng
f
a
r
m
e
r
s
w
it
h
it
s
pot
e
nt
ia
l
f
or
r
e
s
our
c
e
-
c
ons
tr
a
in
e
d
de
vi
c
e
s
.
I
s
m
a
il
e
t
al
.
[
6]
unc
ove
r
e
d
L
as
io
di
pl
odi
a
s
pe
c
ie
s
,
in
c
lu
di
ng
L
.
th
e
obr
o
m
ae
,
L
.
e
gy
pt
ia
c
ae
,
a
nd
L
.
ps
e
udot
he
obr
om
a
e
,
in
E
gypt
ia
n
m
a
ngo
pl
a
nt
a
ti
ons
,
e
m
pha
s
i
z
in
g
th
e
ne
c
e
s
s
it
y
f
or
ta
x
onomi
c
c
la
r
it
y
a
nd
pa
th
oge
ni
c
it
y
e
va
lu
a
ti
on
to
gui
de
di
s
e
a
s
e
c
ont
r
ol
m
e
a
s
ur
e
s
.
S
a
le
e
m
e
t
al
.
[
7]
ut
il
iz
e
d
C
N
N
to
a
c
hi
e
ve
96.6%
a
c
c
ur
a
c
y
in
c
la
s
s
if
yi
ng
m
a
ngo
pl
a
nt
le
a
f
phot
os
,
w
it
h
a
90.83%
de
te
c
ti
on
r
a
te
us
in
g
k
-
ne
a
r
e
s
t
ne
ig
hbor
(
KNN
)
s
e
gm
e
nt
a
ti
on.
T
he
y
f
ur
th
e
r
im
pr
ove
d
a
c
c
ur
a
c
y
to
90%
by
e
m
pl
oyi
ng
a
two
a
nd
ni
ne
c
lu
s
te
r
s
s
tr
a
te
gy
f
or
s
e
gm
e
nt
in
g
s
ic
k
c
it
r
us
le
a
f
r
e
gi
ons
w
it
h
opt
im
a
l
m
in
i
m
um
bond
pa
r
a
m
e
te
r
s
of
3%
.
Z
ha
n
e
t
al
.
[
8]
id
e
nt
if
ie
d
F
us
ar
iu
m
pr
ol
if
e
r
at
um
f
r
om
m
a
lf
or
m
e
d
m
a
ngo
s
e
e
dl
in
gs
in
C
hi
n
a
,
unde
r
s
c
or
in
g
th
e
i
m
por
ta
nc
e
of
pr
e
c
is
e
id
e
nt
if
ic
a
ti
on
f
or
di
s
e
a
s
e
unde
r
s
ta
ndi
ng. T
hi
s
s
tu
dy c
ont
r
ib
ut
e
s
i
ns
ig
ht
s
i
nt
o
F
us
a
r
iu
m
s
p
e
c
ie
s
l
in
ke
d t
o m
a
ngo ma
lf
or
m
a
ti
on.
A
not
he
r
a
ppr
oa
c
h
a
ddr
e
s
s
e
s
m
a
ngo
m
a
lf
or
m
a
ti
on
di
s
e
a
s
e
(
M
M
D
)
w
or
ld
w
id
e
a
nd
m
a
ngo
s
udde
n
de
c
li
ne
i
n O
m
a
n unde
r
s
c
or
e
s
t
he
ne
e
d f
or
f
ur
th
e
r
r
e
s
e
a
r
c
h a
nd ve
c
to
r
m
a
na
ge
m
e
nt
[
9]
, [
10]
. C
ol
in
a
e
t
al
.
[
11]
id
e
nt
if
ie
d
ni
ne
F
us
ar
iu
m
s
pe
c
ie
s
li
nke
d
to
M
M
D
in
M
e
xi
c
o,
in
c
lu
di
ng
th
e
nove
l
pa
th
oge
n
F
.
m
e
x
i
c
anum
,
c
onf
ir
m
e
d
pa
th
oge
ni
c
th
r
ough
K
oc
h'
s
pos
tu
la
te
s
.
S
hi
va
kum
a
r
e
t
al
.
[
12]
p
r
ovi
de
in
s
ig
ht
s
in
to
m
a
ngo
pos
th
a
r
ve
s
t
m
a
na
ge
m
e
nt
,
f
oc
us
in
g
on
pr
e
s
e
r
vi
ng
f
r
ui
t
qu
a
li
ty
a
nd
m
in
im
iz
in
g
lo
s
s
e
s
.
T
he
ir
r
e
vi
e
w
s
ynt
he
s
iz
e
s
r
e
s
e
a
r
c
h
on
in
nova
ti
ve
te
c
hnol
ogi
e
s
a
im
e
d
a
t
e
nh
a
nc
in
g
m
a
ngo
qua
li
ty
th
r
oughout
th
e
s
uppl
y
c
ha
in
.
T
r
a
ng
e
t
al
.
[
13]
in
tr
oduc
e
a
n
im
a
ge
-
ba
s
e
d
di
s
e
a
s
e
id
e
n
ti
f
ic
a
ti
on
m
e
th
od
us
in
g
de
e
p
ne
ur
a
l
ne
twor
ks
,
a
c
hi
e
vi
ng
a
n
88.46%
a
c
c
ur
a
c
y
in
id
e
nt
if
yi
ng
c
om
m
on
m
a
ngo
di
s
e
a
s
e
s
.
T
hi
s
a
ppr
oa
c
h
s
ur
pa
s
s
e
s
ot
h
e
r
pr
e
-
tr
a
in
e
d
m
ode
ls
,
of
f
e
r
in
g
pr
om
is
e
f
or
e
f
f
ic
ie
nt
pl
a
nt
di
s
e
a
s
e
de
te
c
ti
on.
Z
a
in
ur
i
e
t
al
.
[
14]
in
ve
s
ti
ga
te
d
pot
a
s
s
iu
m
phos
phona
t
e
a
nd
s
a
li
c
yl
ic
a
c
id
tr
e
a
tm
e
nt
s
f
or
a
nt
hr
a
c
nos
e
c
ont
r
ol
in
m
a
ngo
f
r
ui
t.
W
hi
le
no
e
f
f
e
c
ts
w
e
r
e
obs
e
r
ve
d
in
it
ia
ll
y,
s
a
li
c
yl
ic
a
c
id
tr
e
a
tm
e
nt
s
s
how
e
d
pr
om
is
e
in
r
e
duc
in
g
di
s
e
a
s
e
s
e
v
e
r
it
y
a
nd
s
lo
w
in
g
f
r
ui
t
r
ip
e
ni
ng
in
s
ubs
e
que
nt
s
e
a
s
on
s
G
in
in
g
e
t
al
.
[
15]
a
ddr
e
s
s
in
g
li
m
it
a
ti
ons
in
m
a
ngo
f
a
r
m
in
g
te
c
hni
que
s
,
a
di
s
e
a
s
e
r
e
c
ogni
ti
on
s
y
s
te
m
us
in
g
im
a
ge
pr
oc
e
s
s
in
g
of
f
e
r
s
pr
a
c
ti
c
a
l
be
ne
f
it
s
.
R
a
ha
m
a
n
e
t
al
.
[
16]
ut
i
li
z
e
d
m
a
c
hi
n
e
l
e
a
r
ni
n
g
te
c
hni
que
s
on
m
a
n
go
f
r
ui
t
a
nd
le
a
f
phot
o
s
,
a
c
hi
e
vi
ng
97
.81%
a
c
c
ur
a
c
y
w
it
h
D
e
ns
e
N
e
t1
69
.
T
he
ir
A
ndr
oi
d
a
pp
a
i
ds
in
di
s
e
a
s
e
id
e
nt
if
ic
a
ti
on
a
nd
p
e
s
ti
c
id
e
r
e
c
om
m
e
nd
a
ti
on.
S
in
gh
e
t
al
.
[
1
7]
in
tr
odu
c
ti
on
of
a
m
ul
ti
la
y
e
r
c
onvo
lu
ti
on
a
l
ne
ur
a
l
ne
tw
or
k
(
M
C
N
N
)
f
or
d
ia
gno
s
in
g
A
nt
h
r
ac
no
s
e
pr
ovi
d
e
s
a
pr
om
is
i
ng
s
ol
ut
io
n
f
or
di
s
e
a
s
e
m
a
n
a
g
e
m
e
nt
.
R
a
jb
o
ngs
hi
e
t
al
.
[
18]
a
ddr
e
s
s
th
e
c
r
uc
ia
l
r
ol
e
of
di
s
e
a
s
e
a
c
k
now
l
e
dgm
e
nt
in
e
nha
n
c
in
g
h
a
r
ve
s
t
yi
e
ld
by
e
m
pl
oyi
n
g
C
N
N
s
,
p
a
r
ti
c
ul
a
r
ly
D
e
n
s
e
N
e
t2
01
,
w
hi
c
h
a
c
hi
e
v
e
s
h
ig
h
a
c
c
ur
a
c
y
(
98.00%
)
in
m
a
n
go
le
a
f
di
s
e
a
s
e
c
la
s
s
if
ic
a
ti
on,
a
nd
s
tr
e
a
m
li
ni
ng
de
te
c
ti
on
m
e
th
od
s
.
T
hi
s
r
e
s
e
a
r
c
h
c
om
pi
la
t
io
n
by
P
la
nc
a
r
te
e
t
al
.
[
19]
pr
ovi
de
s
c
om
pr
e
h
e
ns
i
ve
in
s
ig
ht
s
i
nt
o
v
a
r
io
us
a
s
p
e
c
t
s
of
m
a
ngo
di
s
e
a
s
e
m
a
na
ge
m
e
nt
a
nd
unde
r
s
ta
ndi
ng.
I
t
e
xpl
or
e
s
th
e
pr
e
va
l
e
n
c
e
a
nd
ge
ne
ti
c
id
e
n
ti
f
ic
a
ti
o
n
of
m
a
lf
or
m
a
ti
on
di
s
e
a
s
e
in
M
e
xi
c
a
n
m
a
ngo
nur
s
e
r
ie
s
,
hi
ghl
i
ght
in
g
th
e
c
r
it
i
c
a
l
ne
e
d
f
or
pa
th
og
e
n
-
f
r
e
e
pl
a
nt
in
g
m
a
te
r
ia
l
to
c
ur
b
di
s
e
a
s
e
s
pr
e
a
d
dur
in
g
or
c
h
a
r
d
e
s
ta
bl
is
hm
e
nt
.
S
il
im
e
la
a
nd
K
or
s
te
n
[
20]
e
va
lu
a
te
d
th
e
B
ac
il
lu
s
li
c
he
ni
fo
r
m
is
f
or
c
ont
r
ol
li
ng
m
a
ngo
f
r
ui
t
di
s
e
a
s
e
s
unde
r
s
c
or
in
g
it
s
pot
e
nt
ia
l
in
di
s
e
a
s
e
m
a
n
a
ge
m
e
nt
,
w
hi
le
A
ti
ns
ky
e
t
al
.
[
21]
r
e
s
e
a
r
c
h
on
F
us
ar
iu
m
m
angi
fe
r
ae
in
f
e
c
ti
on
dyna
m
ic
s
s
he
d
li
ght
on
di
s
e
a
s
e
c
yc
le
s
a
nd
opt
im
a
l
c
ondi
ti
ons
f
o
r
f
unga
l
gr
ow
th
.
W
ongs
il
a
e
t
al
.
[
22]
a
im
to
de
s
ig
n
a
n
a
lg
or
it
hm
f
or
de
te
c
ti
ng
a
nt
hr
a
c
nos
e
-
in
f
e
c
te
d
m
a
ngoe
s
s
how
c
a
s
in
g
pr
om
is
in
g
a
dva
nc
e
m
e
nt
s
in
di
s
e
a
s
e
de
te
c
ti
on
te
c
hnol
ogy.
T
he
pr
opos
a
l
of
a
n
a
dva
nc
e
d
a
le
r
t
s
ys
te
m
f
or
di
s
e
a
s
e
out
br
e
a
k
f
or
e
c
a
s
ti
ng
by
J
a
w
a
de
e
t
al
.
[
23]
of
f
e
r
s
in
nova
ti
ve
s
ol
ut
io
ns
f
o
r
ti
m
e
ly
di
s
e
a
s
e
m
a
na
ge
m
e
nt
.
F
ur
th
e
r
m
or
e
,
in
ve
s
ti
ga
ti
ons
in
to
c
hi
to
s
a
n
a
nd
s
pe
r
m
id
in
e
f
r
ui
t
c
oa
ti
ngs
by
J
ongs
r
i
e
t
al
.
[
24
]
pr
e
s
e
nt
pr
a
c
ti
c
a
l
a
ppr
oa
c
he
s
f
or
e
nha
nc
in
g
po
s
t
-
ha
r
ve
s
t
qua
li
ty
.
A
s
ur
ve
y
on
nur
s
e
r
y
di
s
e
a
s
e
s
in
B
a
ngl
a
de
s
h
by
S
a
r
ke
r
e
t
al
.
[
25]
pr
ov
id
e
s
in
s
ig
ht
s
in
to
r
e
gi
ona
l
di
s
e
a
s
e
pr
e
va
le
nc
e
a
nd
e
f
f
ic
a
c
y
of
c
ont
r
ol
m
e
a
s
ur
e
s
.
F
in
a
ll
y, a
c
om
pr
e
he
ns
iv
e
r
e
vi
e
w
of
m
a
ngo a
nt
hr
a
c
nos
e
c
ont
r
ib
ut
e
s
t
o i
m
pr
ove
d di
s
e
a
s
e
c
ont
r
ol
s
tr
a
te
gi
e
s
.
T
hi
s
s
tu
dy
a
im
s
to
br
id
ge
th
is
ga
p
by
de
ve
lo
pi
ng
a
n
a
dv
a
nc
e
d
m
a
c
hi
ne
le
a
r
ni
ng
-
ba
s
e
d
s
y
s
te
m
f
or
th
e
e
a
r
ly
de
te
c
ti
on
a
nd
di
a
gno
s
is
of
m
a
ngo
di
s
e
a
s
e
s
in
B
a
ngl
a
de
s
h.
B
y
le
ve
r
a
gi
ng
C
N
N
s
a
nd
ot
he
r
s
ta
te
-
of
-
th
e
-
a
r
t
de
e
p
le
a
r
ni
ng
te
c
hni
que
s
,
th
e
r
e
s
e
a
r
c
h
s
e
e
ks
to
a
c
c
ur
a
te
ly
id
e
nt
if
y
c
om
m
on
m
a
ngo
di
s
e
a
s
e
s
f
r
om
im
a
ge
s
of
le
a
ve
s
a
nd
f
r
ui
ts
.
T
hi
s
a
ppr
oa
c
h
not
onl
y
pr
om
is
e
s
to
e
nha
nc
e
di
s
e
a
s
e
s
ur
ve
il
la
nc
e
c
a
pa
bi
li
ti
e
s
but
a
l
s
o
of
f
e
r
s
pr
a
c
ti
c
a
l
s
ol
ut
io
n
s
to
im
pr
ove
m
a
ngo
pr
oduc
ti
on
pr
a
c
ti
c
e
s
.
T
hi
s
r
e
s
e
a
r
c
h
w
il
l
in
vol
ve
s
e
ve
r
a
l
ke
y s
te
p
s
. I
ni
ti
a
ll
y, hi
gh
-
qua
li
ty
i
m
a
ge
s
of
m
a
ngo pla
nt
s
w
il
l
be
c
ol
le
c
te
d a
nd pr
e
pr
oc
e
s
s
e
d t
o
e
nha
nc
e
f
e
a
tu
r
e
e
xt
r
a
c
ti
on.
T
he
s
e
im
a
ge
s
w
il
l
th
e
n
be
a
na
ly
z
e
d
us
in
g
C
N
N
s
to
id
e
nt
if
y
di
s
e
a
s
e
pa
tt
e
r
ns
a
nd
c
la
s
s
if
y
th
e
m
a
c
c
ur
a
te
ly
.
T
he
s
ys
te
m
w
il
l
be
te
s
te
d
a
nd
va
li
da
te
d
a
ga
in
s
t
e
xi
s
ti
ng
da
ta
s
e
t
s
to
e
ns
ur
e
it
s
r
e
li
a
bi
li
ty
a
nd
e
f
f
e
c
ti
ve
ne
s
s
.
T
h
e
ul
ti
m
a
te
goa
l
is
to
pr
ovi
de
a
s
c
a
la
bl
e
a
nd
e
f
f
ic
ie
nt
to
ol
th
a
t
c
a
n
be
us
e
d
by
f
a
r
m
e
r
s
a
nd a
gr
ic
ul
tu
r
a
l
pr
of
e
s
s
io
na
ls
i
n B
a
ngl
a
de
s
h t
o m
it
ig
a
te
t
he
i
m
pa
c
t
of
m
a
ngo dis
e
a
s
e
s
.
2.
R
E
S
E
A
R
C
H
M
E
T
H
O
D
O
L
O
G
Y
E
xpl
a
in
in
g
th
e
s
ig
n
s
of
m
a
ny
il
ln
e
s
s
e
s
c
a
n
be
e
xt
r
e
m
e
l
y s
im
il
a
r
,
it
c
a
n
be
ve
r
y
di
f
f
ic
ul
t
to
di
s
ti
ngui
s
h
a
he
a
l
th
y m
a
ngo le
a
f
or
f
r
ui
t
f
r
om
a
di
s
e
a
s
e
d one
j
us
t
by l
ooki
n
g a
t
it
. A
s
ui
ta
b
le
c
o
ur
s
e
of
t
r
e
a
tm
e
nt
c
a
nn
ot
be
s
ta
r
t
e
d
in
s
uc
h
c
a
s
e
s
w
it
hou
t
a
pr
e
c
i
s
e
d
ia
gn
os
i
s
.
T
h
e
f
a
r
m
e
r
s
c
a
n
lo
s
e
a
lo
t
of
m
on
e
y
if
it
b
e
c
om
e
s
in
f
e
s
te
d
w
it
h
p
e
s
t
s
. W
e
ha
v
e
d
e
c
i
de
d t
o
f
o
c
us
on
t
hi
s
to
pi
c
f
or
o
ur
s
t
udy
a
s
a
r
e
s
ul
t
of
t
he
s
e
f
a
c
to
r
s
. A
n
im
a
g
e
-
f
il
te
r
in
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
age
analy
s
is
and mac
hi
ne
l
e
ar
ni
ng t
e
c
hni
que
s
f
or
ac
c
ur
at
e
d
e
te
c
ti
on of
…
(
M
d A
bdul
la
h A
l
R
ahi
b
)
2937
m
ode
l
th
a
t
c
onv
e
r
ts
c
or
r
e
l
a
ti
on
s
be
tw
e
e
n
qua
li
t
y
r
a
ti
ng
s
a
nd
ir
r
e
le
v
a
nt
pi
c
t
ur
e
s
our
c
e
s
in
to
c
or
r
e
l
a
ti
on
s
w
it
h
a
r
a
ndom
f
or
e
s
t
a
lg
or
it
hm
m
a
y
b
e
c
r
e
a
t
e
d.
T
he
y
m
u
s
t
b
e
a
dd
r
e
s
s
e
d
be
f
or
e
c
om
p
a
r
in
g
t
he
m
to
a
ddi
ti
ona
l
m
a
c
hi
n
e
le
a
r
ni
ng
m
o
de
l
te
c
hni
que
s
to
pr
odu
c
e
m
or
e
m
e
th
o
di
c
a
ll
y
a
nd
r
e
s
ul
t
in
a
m
or
e
e
f
f
e
c
ti
v
e
de
pl
oym
e
n
t.
T
o
a
c
hi
e
ve
our
ul
t
im
a
t
e
go
a
l
of
r
e
duc
i
ng
m
a
ngo
di
s
e
a
s
e
,
w
e
ha
ve
e
xp
e
r
im
e
n
te
d
w
it
h
a
v
a
r
ie
ty
of
di
f
f
e
r
e
nt
ty
pe
s
of
di
s
e
a
s
e
f
or
m
a
ts
.
T
he
m
a
ng
o di
s
e
a
s
e
de
te
c
ti
on pr
o
c
e
s
s
i
s
s
ho
w
n i
n F
i
gur
e
1.
F
ig
ur
e
1. M
a
ngo dis
e
a
s
e
de
te
c
ti
on pr
oc
e
s
s
2.1. Dat
a c
ol
le
c
t
io
n
F
lo
w
e
r
di
s
e
a
s
e
,
a
nt
hr
a
c
nos
e
/b
lo
s
s
om
bl
ig
ht
,
m
a
ngo
de
f
or
m
it
y,
ba
c
te
r
ia
l
c
a
nke
r
,
le
a
f
s
pot
,
s
te
m
e
nd
r
ot
,
di
e
ba
c
k,
twi
g
bl
ig
ht
,
gum
m
os
is
,
ba
r
k
s
pl
it
ti
ng,
ba
r
k
s
c
a
li
ng,
a
nd
w
il
ti
ng
m
a
ngo
s
udd
e
n
de
a
th
s
yndr
om
e
(
M
S
D
S
)
w
e
r
e
a
m
ong
th
e
m
a
ngo
di
s
e
a
s
e
s
w
e
s
tu
di
e
d. T
he
s
e
s
t
a
ti
s
ti
c
s
w
e
r
e
ta
ke
n
be
twe
e
n
M
a
r
c
h
a
nd
J
ul
y
of
2023. I
nt
e
r
ne
t
da
ta
i
s
r
a
r
e
ly
ga
th
e
r
e
d. R
e
a
l
-
ti
m
e
da
ta
w
e
r
e
c
ol
le
c
te
d i
n R
a
js
ha
hi
,
B
a
ngl
a
de
s
h.
2.2. De
s
c
r
ip
t
io
n
of
m
an
go d
is
e
as
e
M
a
ngo
f
r
ui
t
in
c
lu
de
s
a
va
r
ie
ty
of
phyt
oc
he
m
ic
a
l
s
,
in
c
lu
di
ng
pol
yphe
nol
s
,
a
s
c
or
bi
c
a
c
id
,
a
nd
c
a
r
ot
e
noi
ds
,
w
hi
c
h
ha
ve
he
a
lt
h
-
pr
om
ot
in
g
qua
li
ti
e
s
du
e
to
th
e
ir
a
nt
io
xi
da
nt
c
ha
r
a
c
te
r
is
ti
c
s
[
12]
.
F
r
ui
t
is
th
e
m
a
in
us
a
ge
of
it
.
H
ow
e
ve
r
,
s
e
ve
r
a
l
di
s
e
a
s
e
s
ha
ve
a
n
im
pa
c
t
o
n
it
s
pr
oduc
ti
on.
S
o,
w
e
f
ound
di
s
e
a
s
e
s
th
e
r
e
.
T
he
de
s
c
r
ip
ti
on
a
nd
f
ig
ur
e
of
di
f
f
e
r
e
nt
ty
pe
s
of
m
a
ngo
di
s
e
a
s
e
s
a
r
e
s
how
n
in
T
a
bl
e
s
1
to
4
a
nd
th
e
he
a
lt
hy
m
a
ngo
f
ig
ur
e
is
s
how
n
in
F
ig
ur
e
2
.
T
he
e
di
bl
e
he
a
d
of
a
he
a
lt
hy
m
a
ngo
s
houl
d
be
w
hi
te
a
nd
lo
ok
c
om
pa
c
t
,
w
it
h
c
r
is
p,
th
e
c
ol
or
of
he
a
lt
hy
a
m
a
ngo
a
nd
th
e
le
a
ve
s
s
houl
d
be
gr
e
e
n.
A
m
a
ngo
th
a
t
is
in
good
he
a
lt
h
ha
s
no w
r
in
kl
e
s
, bl
a
c
k s
ta
in
s
, or
i
m
pe
r
f
e
c
ti
ons
.
F
ig
ur
e
2. H
e
a
lt
hy ma
ngo
T
a
bl
e
1.
M
a
ngo
f
lo
w
e
r
di
s
e
a
s
e
N
a
m
e
D
i
s
e
a
s
e
de
s
c
r
i
pt
i
on
F
i
gur
e
P
ow
de
r
y
m
i
l
de
w
T
he
s
ym
pt
om
s
c
a
n
be
not
i
c
e
d
on
t
he
i
nf
l
or
e
s
c
e
nc
e
,
t
he
s
t
a
l
k
of
t
he
i
nf
l
or
e
s
c
e
n
c
e
,
t
he
l
e
a
ve
s
,
a
nd
t
he
young f
r
ui
t
s
.
O
ne
of
t
he
di
s
e
a
s
e
'
s
di
s
t
i
ngui
s
hi
ng
f
e
a
t
ur
e
s
i
s
t
he
s
upe
r
f
i
c
i
a
l
,
pow
de
r
y,
w
hi
t
e
f
unga
l
gr
ow
t
h
i
n
t
he
s
e
a
r
e
a
s
.
A
c
c
or
di
ng
t
o
r
e
por
t
s
,
t
he
i
l
l
ne
s
s
c
a
us
e
s
20
t
o
80%
c
r
op
l
o
s
s
.
C
oni
di
a
=9
-
32
°
C
a
nd
>25%
R
.
H
.
D
i
s
e
a
s
e
=10
-
31
°
C
w
i
t
h a
r
e
l
a
t
i
ve
hum
i
di
t
y of
60
-
90%
.
A
nt
hr
a
c
nos
e
/
B
l
os
s
om
bl
i
ght
A
nt
hr
a
c
nos
e
i
s
f
ound
i
n
s
e
ve
r
a
l
pa
r
t
s
of
t
he
m
a
ngo
t
r
e
e
.
T
he
f
i
r
s
t
s
i
gn
s
of
t
he
i
l
l
ne
s
s
a
r
e
bl
a
c
ki
s
h
-
br
ow
n
pa
t
c
he
s
on
f
l
ow
e
r
s
a
nd
pe
dun
c
l
e
s
.
S
m
a
l
l
bl
a
c
k
s
p
ot
s
a
ppe
a
r
on
pa
ni
c
l
e
s
a
nd
ope
n
f
l
ow
e
r
s
,
a
nd
a
s
t
he
y
gr
ow
l
a
r
ge
r
,
t
he
y
d
a
m
a
ge
t
he
bl
os
s
om
s
.
D
i
s
e
a
s
e
d
bl
oom
s
br
e
a
k
of
f
,
l
e
a
vi
ng
m
or
e
p
e
r
s
i
s
t
e
nt
s
pi
ke
s
on
t
he
pe
dunc
l
e
s
,
r
e
s
ul
t
i
ng
i
n
s
e
ve
r
e
c
r
op
l
o
s
s
(
10
-
90%
)
.25
-
30
°
C
a
nd
95
-
97%
R
.H
.
O
ve
r
l
y
n
i
t
r
oge
nous
.
M
a
ngo
m
a
l
f
or
m
a
t
i
on
D
e
s
pi
t
e
be
i
ng
i
ni
t
i
a
l
l
y
do
c
um
e
nt
e
d
i
n
I
ndi
a
ove
r
a
c
e
nt
ur
y
a
go
(
10,
14,
34)
,
t
hi
s
di
s
e
a
s
e
w
a
s
not
di
s
c
ove
r
e
d
i
n
M
e
xi
c
o
unt
i
l
1958
[
11]
.
R
e
c
e
nt
r
e
s
e
a
r
c
h
s
ugge
s
t
s
t
ha
t
t
he
s
i
c
kne
s
s
i
s
c
a
u
s
e
d
by
a
f
ungus
.
F
l
or
a
l
m
a
l
f
or
m
a
t
i
on
(
M
F
)
a
nd
ve
ge
t
a
t
i
ve
m
a
l
f
or
m
a
t
i
on
(
M
V
)
a
r
e
t
w
o
di
f
f
e
r
e
nt
c
a
t
e
gor
i
e
s
of
s
ym
pt
om
s
t
ha
t
t
he
w
or
k
e
r
s
ha
ve
doc
um
e
nt
e
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 4, A
ugus
t
2025
:
2935
-
2944
2938
T
a
bl
e
2.
M
a
ngo le
a
f
di
s
e
a
s
e
N
a
m
e
D
i
s
e
a
s
e
de
s
c
r
i
pt
i
on
F
i
gur
e
A
nt
hr
a
c
nos
e
A
s
s
ym
pt
om
s
,
r
ound
or
i
r
r
e
gul
a
r
da
r
k
t
o
de
e
p
br
ow
n
pa
t
c
he
s
of
va
r
yi
ng
s
i
z
e
s
c
a
n
be
f
ound
on
t
he
l
e
a
f
s
ur
f
a
c
e
.
I
n
hum
i
d
s
e
t
t
i
ngs
,
t
he
f
ungus
m
ul
t
i
pl
i
e
s
r
a
pi
d
l
y.
Y
oung
l
e
a
ve
s
a
t
t
r
a
c
t
m
or
e
i
ns
e
c
t
s
t
ha
n
ol
de
r
one
s
do.
A
t
t
a
c
ks
by
i
ns
e
c
t
s
m
a
y
m
a
ke
i
t
e
a
s
i
e
r
f
or
pa
t
hoge
ns
t
o e
nt
e
r
, l
e
a
di
ng t
o a
hi
gh pr
e
va
l
e
nc
e
of
di
s
e
a
s
e
.
A
l
t
e
r
na
r
i
a
l
e
a
f
s
pot
T
he
c
ondi
t
i
on
be
gi
ns
w
i
t
h
l
i
t
t
l
e
,
br
ow
ni
s
h
c
i
r
c
ul
a
r
s
pot
s
on
t
he
s
ur
f
a
c
e
o
f
l
e
a
v
e
s
.
L
a
t
e
r
,
t
he
l
e
a
f
l
a
m
i
na
a
c
qui
r
e
s
a
de
ns
e
pa
t
t
e
r
n
of
br
ow
n
a
nd
bl
a
c
k
s
pe
c
ks
.
T
he
l
ow
e
r
s
i
de
of
t
he
l
e
a
ve
s
e
xhi
bi
t
s
m
or
e
pr
onounc
e
d
s
ym
pt
om
s
. I
t
i
s
di
s
c
ove
r
e
d t
ha
t
younge
r
l
e
a
ve
s
a
r
e
m
or
e
pr
one
t
ha
n ol
de
r
one
s
.
B
a
c
t
e
r
i
a
l
C
a
nke
r
O
n
l
e
a
ve
s
,
t
he
a
pe
x
i
s
ge
n
e
r
a
l
l
y
de
ns
e
l
y
pa
c
k
e
d
w
i
t
h
s
m
a
l
l
,
w
a
t
e
r
-
s
oa
k
e
d,
i
r
r
e
gul
a
r
t
o
a
ngul
a
r
r
a
i
s
e
d
l
e
s
i
ons
.
W
hi
l
e
young
e
r
l
e
a
ve
s
ha
ve
m
or
e
obvi
ous
a
nd
br
oa
de
r
ha
l
os
t
ha
n
ol
de
r
l
e
a
ve
s
,
w
hi
c
h
c
a
n
onl
y
be
s
e
e
n
i
n
br
i
ght
s
unl
i
ght
,
e
l
de
r
l
e
a
ve
s
ha
ve
na
r
r
ow
e
r
ha
l
os
.
W
he
n
a
l
e
a
f
i
s
s
e
r
i
ous
l
y
a
f
f
e
c
t
e
d,
i
t
be
c
om
e
s
ye
l
l
ow
a
nd f
a
l
l
s
of
f
. 25
-
30
°
C
a
nd
>90R
.H
.
T
a
bl
e
3.
M
a
ngo
p
os
t
-
ha
r
ve
s
t
di
s
e
a
s
e
N
a
m
e
D
i
s
e
a
s
e
d
e
s
c
r
i
pt
i
on
F
i
gur
e
F
r
ui
t
a
nt
hr
a
c
nos
e
T
he
pr
e
-
ha
r
ve
s
t
i
nf
e
c
t
i
on
c
a
us
e
s
po
s
t
-
ha
r
ve
s
t
r
ot
s
.
T
he
r
e
a
r
e
bl
a
c
k
pa
t
c
he
s
i
n
s
t
or
a
ge
.
I
ni
t
i
a
l
l
y
c
i
r
c
ul
a
r
,
t
he
s
pot
s
m
or
ph
i
nt
o
l
a
r
ge
,
une
ve
n
bl
ot
c
he
s
t
ha
t
c
ove
r
t
he
e
nt
i
r
e
f
r
ui
t
.
T
he
f
ungus
r
ot
s
t
he
f
r
ui
t
de
e
pl
y
a
nd
c
a
us
e
s
m
a
s
s
i
ve
,
d
e
e
p
f
r
a
c
t
ur
e
s
i
n
t
he
a
r
e
a
s
.
A
s
a
r
e
s
ul
t
,
e
f
f
e
c
t
i
ve
,
s
a
f
e
,
a
nd a
f
f
or
da
bl
e
pl
a
nt
pr
ot
e
c
t
i
on w
a
ys
a
r
e
i
nt
e
gr
a
l
[
14]
.
S
t
e
m
e
nd r
ot
A
s
t
he
f
r
ui
t
r
i
pe
ns
,
t
he
s
t
e
m
e
nd
t
ur
ns
da
r
k
or
bl
a
c
k.
W
i
t
hi
n
t
w
o
t
o
t
h
r
e
e
da
ys
,
t
he
e
nt
i
r
e
f
r
ui
t
t
ur
ns
bl
a
c
k,
a
nd
t
he
di
s
e
a
s
e
s
pr
e
a
ds
dow
nw
a
r
d,
da
m
a
gi
ng ha
l
f
of
t
he
f
r
ui
t
'
s
s
ur
f
a
c
e
. A
l
t
hough t
he
e
nt
i
r
e
f
r
ui
t
t
ypi
c
a
l
l
y ha
s
a
bl
us
h,
w
r
i
nkl
e
s
a
r
e
a
l
s
o
not
i
c
e
a
bl
e
.
T
he
a
f
f
l
i
c
t
e
d
s
ki
n
r
e
m
a
i
ns
s
ol
i
d,
but
r
ot
de
ve
l
ops
i
n t
he
pul
p be
ne
a
t
h.
T
a
bl
e
4.
M
a
ngo
de
c
li
ne
s
di
s
or
de
r
s
di
s
e
a
s
e
N
a
m
e
D
i
s
e
a
s
e
d
e
s
c
r
i
pt
i
on
F
i
gur
e
D
i
e
ba
c
k
A
l
t
hough
t
he
di
s
e
a
s
e
i
s
vi
s
i
bl
e
t
hr
oughout
t
he
ye
a
r
,
i
t
i
s
di
s
t
i
ngui
s
he
d
by t
he
t
op
-
to
-
bot
t
om
dr
y
i
ng
of
t
w
i
gs
,
pa
r
t
i
c
ul
a
r
l
y
i
n
e
l
de
r
t
r
e
e
s
,
f
ol
l
ow
e
d
by
l
e
a
f
dr
yi
ng,
gi
vi
ng
t
he
i
m
a
ge
of
f
i
r
e
s
c
or
c
h.
T
he
uppe
r
l
e
a
ve
s
dr
y
out
a
nd
l
os
e
t
he
i
r
c
ol
or
w
i
t
h
t
i
m
e
.
T
he
bor
de
r
r
ol
l
s
upw
a
r
d
a
s
t
he
e
nt
i
r
e
l
e
a
f
dr
i
e
s
.
W
h
e
n
s
uc
h l
e
a
ve
s
s
hr
i
nk a
nd f
a
l
l
of
f
w
i
t
hi
n a
m
ont
h.
T
w
i
g bl
i
ght
T
he
t
w
i
gs
de
ve
l
op
e
l
onga
t
e
d,
da
r
k,
ne
c
r
ot
i
c
pa
t
c
he
s
due
t
o
di
s
e
a
s
e
.
T
he
l
e
a
ve
s
gr
a
dua
l
l
y
dr
oop
t
hr
oughout
t
he
ups
w
i
ng
be
f
or
e
f
a
l
l
i
ng.
O
f
f
T
he
ve
r
y
young
br
a
nc
he
s
be
gi
n
t
o
dr
y
f
r
om
t
he
t
i
p
dow
n.
I
nj
ur
i
e
s
,
i
ns
e
c
t
a
t
t
a
c
ks
,
hot
pl
a
nt
s
w
i
t
h w
e
a
k r
oot
s
, w
a
t
e
r
s
t
r
e
s
s
, f
r
os
t
, a
nd phys
i
c
a
l
ha
r
m
.
G
um
m
os
i
s
G
um
m
os
i
s
a
f
f
e
c
t
s
30
-
40%
of
young
m
a
ngo
t
r
e
e
s
,
pa
r
t
i
c
ul
a
r
l
y
t
hos
e
pl
a
nt
e
d
i
n s
a
ndy s
oi
l
, but
i
t
ha
s
b
e
e
n obs
e
r
ve
d i
n ot
he
r
m
a
ngo
-
gr
ow
i
ng c
ondi
t
i
ons
a
s
w
e
l
l
.
T
he
pr
e
s
e
nc
e
of
e
nor
m
ous
a
m
ount
s
of
gum
f
l
ow
i
ng
f
r
om
t
he
s
ur
f
a
c
e
of
t
he
da
m
a
ge
d
w
ood,
t
he
b
a
r
k
of
t
he
t
r
unk,
a
nd
on
l
a
r
ge
r
br
a
nc
he
s
di
s
t
i
ngui
s
he
s
t
he
i
l
l
ne
s
s
.
B
a
r
ks
c
r
a
c
ki
ng
T
he
e
m
e
r
ge
nc
e
of
l
a
r
ge
,
de
e
p
l
ongi
t
udi
na
l
f
i
s
s
ur
e
s
i
s
a
s
i
gn
of
ba
r
k
c
r
a
c
ki
ng.
A
l
t
hough
t
he
unde
r
l
yi
ng
w
ood
i
s
di
s
c
ove
r
e
d
t
o
be
s
i
gni
f
i
c
a
nt
l
y
pi
t
t
e
d,
r
oot
i
ng
i
s
not
c
onne
c
t
e
d
t
o
t
he
f
i
s
s
ur
e
s
.
A
l
ong
w
i
t
h
t
he
f
i
s
s
ur
e
s
,
gum
poc
ke
t
s
a
r
e
a
l
s
o
di
s
c
e
r
ni
bl
e
.
L
a
t
e
r
,
a
s
t
h
e
ba
r
k
dr
i
e
s
a
nd
i
s
t
a
k
e
n
of
f
,
i
t
c
a
us
e
s
t
he
c
ons
e
qu
e
nc
e
s
of
gi
r
dl
i
ng, ye
l
l
ow
i
ng, a
nd l
e
a
f
l
os
s
.
R
oot
r
ot
W
a
t
e
r
-
s
oa
ke
d
pa
t
c
h
e
s
t
ha
t
a
r
e
c
i
r
c
ul
a
r
t
o
a
s
ym
m
e
t
r
i
c
a
l
be
c
om
e
i
nf
e
c
t
e
d
a
t
or
be
l
ow
gr
ound
l
e
ve
l
.
T
he
s
e
pa
t
c
he
s
gr
ow
i
n
s
i
z
e
,
e
ve
nt
ua
l
l
y
e
nc
i
r
c
l
i
ng
t
he
e
nt
i
r
e
s
t
e
m
ba
s
e
. I
n l
i
ght
of
t
hi
s
, t
he
di
s
e
a
s
e
d
t
i
s
s
u
e
s
be
gi
n
t
o de
ge
ne
r
a
t
e
a
nd
be
c
om
e
m
us
hy, da
r
k br
ow
n, or
bl
a
c
k.
M
a
ngo
s
udde
n
de
a
t
h
s
yndr
om
e
S
i
gni
f
i
c
a
nt
pr
e
di
s
pos
i
ng
f
a
c
t
or
s
f
or
t
h
i
s
i
l
l
ne
s
s
ha
ve
be
e
n
i
de
nt
i
f
i
e
d
a
s
i
m
pr
ope
r
w
a
t
e
r
i
ng
a
nd
r
oot
da
m
a
ge
.
M
a
ngo
t
r
e
e
s
w
i
t
h
t
he
c
ondi
t
i
on
w
i
l
t
.
C
a
nke
r
s
m
a
y
gr
ow
ove
r
va
s
c
ul
a
r
di
s
c
ol
or
a
t
i
on
a
nd
di
s
c
ha
r
ge
gum
f
r
om
t
he
s
t
e
m
.
W
i
l
t
e
d
l
e
a
ve
s
t
ypi
c
a
l
l
y
dr
y
out
a
nd
c
ur
l
,
a
l
t
hough
t
he
y
r
e
m
a
i
n
a
t
t
a
c
he
d t
o t
he
t
r
e
e
f
or
a
f
e
w
w
e
e
ks
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
age
analy
s
is
and mac
hi
ne
l
e
ar
ni
ng t
e
c
hni
que
s
f
or
ac
c
ur
at
e
d
e
te
c
ti
on of
…
(
M
d A
bdul
la
h A
l
R
ahi
b
)
2939
2
.3. I
m
age
a
c
q
u
is
it
io
n
T
he
f
ir
s
t
pha
s
e
of
our
a
ppr
oa
c
h
is
pi
c
tu
r
e
a
c
qui
s
it
io
n,
in
w
hi
c
h
w
e
obt
a
in
e
xa
m
pl
e
phot
os
f
o
r
tr
a
in
in
g
a
nd
te
s
ti
ng.
I
m
a
ge
s
f
r
om
a
phone
a
nd
a
s
e
le
c
t
h
a
ndf
ul
f
r
om
th
e
in
te
r
ne
t
w
e
r
e
us
e
d
in
th
is
r
e
s
e
a
r
c
h.
T
he
pl
a
nt
s
in
th
e
e
xa
m
pl
e
phot
ogr
a
phs
a
r
e
bot
h
una
f
f
e
c
te
d
a
nd
s
ic
k.
T
he
m
a
ngo
im
a
ge
is
t
a
ke
n
w
it
h
a
phone
c
a
m
e
r
a
a
nd s
a
v
e
d i
n di
gi
ta
l
m
e
di
a
i
n a
c
om
m
on digi
ta
l
f
or
m
a
t.
T
he
R
G
B
f
or
m
a
t
of
t
he
s
e
pi
c
tu
r
e
s
.
2.4. I
m
age
p
r
e
p
r
oc
e
s
s
in
g
I
m
a
ge
s
ga
th
e
r
e
d
f
r
om
m
ul
ti
pl
e
s
our
c
e
s
a
r
e
r
e
f
e
r
r
e
d
to
a
s
r
a
w
p
ic
tu
r
e
s
a
nd
c
a
nnot
b
e
us
e
d
di
r
e
c
tl
y
in
th
e
f
ol
lo
w
in
g
pha
s
e
.
A
s
pe
c
ts
of
im
a
ge
pr
oc
e
s
s
in
g
in
c
lu
de
im
pr
ovi
ng
a
nd
f
il
te
r
in
g
th
e
im
a
ge
,
r
e
m
ovi
ng
noi
s
e
a
nd
unde
s
ir
e
d
obj
e
c
ts
,
a
nd
c
r
oppi
ng
th
e
im
a
ge
to
th
e
de
s
ir
e
d
s
iz
e
.
F
ir
s
tl
y,
w
e
a
ppl
y
i
m
a
ge
r
e
s
i
z
in
g
w
hi
c
h
r
e
s
iz
in
g
th
e
in
put
im
a
ge
is
c
r
it
ic
a
l
f
or
c
a
te
gor
iz
a
ti
on.
I
m
a
ge
s
a
r
e
r
e
duc
e
d
in
s
iz
e
f
or
a
ddi
ti
ona
l
pr
oc
e
s
s
in
g.
T
he
n
w
e
a
ppl
y
im
a
ge
f
il
te
r
in
g
w
hi
c
h
is
a
s
m
oot
hi
ng
f
il
te
r
,
f
or
e
xa
m
pl
e
,
c
a
n
r
e
m
ove
noi
s
e
f
r
om
phot
ogr
a
phs
.
F
in
a
ll
y,
c
ont
r
a
s
t
e
nha
nc
e
m
e
nt
is
th
e
pr
oc
e
s
s
w
he
r
e
hi
s
to
gr
a
m
e
qua
li
z
a
ti
on
is
th
e
m
e
th
od
th
a
t
is
us
e
d
to
im
pr
ove
c
ont
r
a
s
t.
I
m
a
ge
s
a
r
e
i
m
pr
ove
d t
o pe
r
f
or
m
be
tt
e
r
.
2.5. Color
c
on
ve
r
s
io
n
G
r
e
e
n
r
e
pr
e
s
e
nt
s
m
os
tl
y
th
e
he
a
lt
hy
c
om
pone
nt
of
th
e
pl
a
nt
;
th
us
,
w
e
boos
t
it
s
va
lu
e
by
59%
.
R
e
d
a
nd
bl
ue
,
on
th
e
ot
he
r
ha
nd,
m
us
t
de
c
r
e
a
s
e
in
va
lu
e
by
30
a
nd
11%
,
r
e
s
pe
c
ti
ve
ly
.
T
hus
,
th
e
c
onve
r
s
io
n'
s
e
qua
ti
on
is
a
s
f
ol
lo
w
s
:
n
e
w
gr
a
y
s
c
a
le
im
a
ge
=
(
0.3
×
R
)
+
(
0.59
×
G
)
+
(
0.11
×
B
)
.
F
ig
ur
e
3
s
how
s
th
e
c
ol
or
c
onve
r
s
io
n
in
im
a
ge
pr
e
pr
oc
e
s
s
in
g.
T
he
pi
c
tu
r
e
of
th
e
be
f
or
e
c
ol
or
c
onve
r
s
io
n
a
nd
th
e
a
f
te
r
pi
c
tu
r
e
of
a
f
te
r
c
ol
or
c
onve
r
s
io
n
is
s
how
n i
n F
ig
ur
e
s
3(
a
)
a
nd 3(
b)
.
(
a
)
(
b)
F
ig
ur
e
3. C
ol
or
c
onve
r
s
io
n i
n i
m
a
ge
pr
e
pr
oc
e
s
s
in
g
of
(
a
)
be
f
or
e
a
nd (
b)
a
f
te
r
2.6. I
m
age
s
e
g
m
e
n
t
at
io
n
I
n
th
is
r
e
s
e
a
r
c
h,
w
e
us
e
d
th
e
k
-
m
e
a
n
s
c
lu
s
te
r
in
g
te
c
hni
que
t
o
s
e
pa
r
a
te
im
a
ge
s
in
to
th
r
e
e
gr
oups
.
U
s
in
g
a
s
e
t
of
c
r
it
e
r
ia
,
a
na
lo
gou
s
pi
xe
ls
a
r
e
c
om
bi
ne
d
to
c
a
te
gor
iz
e
im
a
ge
s
in
to
k
gr
oups
.
I
ts
pur
pos
e
is
to
r
e
duc
e
t
he
t
ot
a
l
s
qua
r
e
d di
s
ta
nc
e
be
twe
e
n t
he
a
s
s
oc
ia
te
d c
lu
s
te
r
a
nd t
he
t
r
a
in
in
g
i
m
a
ge
s
. F
ir
s
t,
a
n
R
G
B
i
m
a
ge
ha
d
to
be
c
onve
r
te
d
to
L
*a
*b*,
w
he
r
e
L
s
ta
nds
f
or
th
e
lu
m
in
os
it
y
la
ye
r
(
"
L
*"
)
a
nd
a
*b*
f
o
r
th
e
c
hr
o
m
a
ti
c
it
y
la
ye
r
.
T
he
E
uc
li
de
a
n
di
s
ta
n
c
e
m
e
tr
ic
,
w
hi
c
h
is
s
how
n
a
s
f
ol
lo
w
s
,
is
us
e
d
to
de
te
r
m
in
e
th
e
di
s
ta
n
c
e
onc
e
th
e
im
a
ge
ha
s
b
e
e
n
di
vi
de
d
in
to
th
r
e
e
di
s
ti
nc
t
gr
oups
.
W
h
e
r
e
th
e
r
e
a
r
e
two
-
pi
xe
l
c
oor
di
na
te
s
(
X
1,
Y
1)
a
nd
(
X
2, Y
2)
, t
he
E
uc
li
de
a
n
di
s
ta
nc
e
(
d)
i
s
i
de
nt
ic
a
l.
2.7.
F
e
at
u
r
e
e
xt
r
ac
t
io
n
F
e
a
tu
r
e
e
xt
r
a
c
ti
on
is
a
n
im
por
ta
nt
pa
r
t
of
im
a
ge
a
na
ly
s
is
s
in
c
e
it
e
xt
r
a
c
ts
s
ig
ni
f
ic
a
nt
in
f
or
m
a
ti
on
f
r
om
im
a
ge
s
.
E
ve
r
y
di
s
e
a
s
e
,
a
s
w
e
know,
h
a
s
di
s
ti
ngui
s
hi
n
g
c
ha
r
a
c
te
r
is
ti
c
s
th
a
t
he
lp
us
c
om
pr
e
he
nd
it
.
T
e
xt
ur
e
,
f
or
m
,
a
nd
c
ol
or
a
r
e
e
xa
m
pl
e
s
of
th
e
s
e
pr
ope
r
ti
e
s
.
T
h
e
gr
a
y
-
le
ve
l
c
o
-
oc
c
ur
r
e
nc
e
m
a
tr
ix
(
G
L
C
M
)
w
a
s
us
e
d
in
our
s
tu
dy
to
e
xa
m
in
e
f
iv
e
di
f
f
e
r
e
nt
f
or
m
s
of
te
xt
u
r
e
:
c
ont
r
a
s
t,
e
ne
r
gy,
hom
oge
ne
it
y,
c
or
r
e
la
ti
on,
a
nd
e
nt
r
opy. T
he
i
nput
i
m
a
ge
i
s
a
l
s
o ut
il
iz
e
d t
o c
a
lc
ul
a
te
t
he
m
e
a
n,
s
ta
nda
r
d de
vi
a
ti
on, va
r
ia
nc
e
, a
nd kur
to
s
i
s
.
2.8.
C
la
s
s
if
i
c
at
io
n
U
s
in
g
th
e
ir
c
ol
le
c
te
d
f
e
a
tu
r
e
s
,
th
e
e
xt
r
a
c
te
d
c
la
s
s
e
s
of
th
e
i
m
a
ge
s
a
r
e
c
la
s
s
if
ie
d
us
in
g
th
e
im
a
ge
c
la
s
s
if
ic
a
ti
on
te
c
hni
qu
e
.
O
ur
s
ugge
s
te
d
c
la
s
s
if
ic
a
ti
on
s
y
s
te
m
h
a
s
f
iv
e
c
la
s
s
e
s
:
f
lo
w
e
r
di
s
e
a
s
e
s
,
le
a
f
di
s
e
a
s
e
s
,
f
r
ui
t
di
s
e
a
s
e
s
/p
o
s
t
-
ha
r
ve
s
t
di
s
e
a
s
e
s
,
d
e
c
li
ne
di
s
or
de
r
s
,
a
nd
h
e
a
lt
hy
.
T
o
c
l
a
s
s
if
y
our
s
ys
te
m
,
w
e
,
th
e
r
e
f
or
e
,
a
dopt
e
d
a
m
ul
ti
c
la
s
s
a
ppr
oa
c
h.
I
n
th
is
to
pi
c
,
th
e
r
e
a
r
e
num
e
r
ou
s
c
la
s
s
if
ie
r
s
a
va
il
a
bl
e
.
W
e
c
ont
r
a
s
t
a
va
r
ie
ty
of
c
la
s
s
if
ie
r
s
,
in
c
lu
di
ng
r
a
ndom
f
or
e
s
t
,
I
B
K
,
K
-
S
ta
r
,
s
uppor
t
ve
c
t
or
m
a
c
hi
ne
,
de
c
is
io
n
tr
e
e
,
a
nd
r
a
ndom
f
or
e
s
t
.
T
he
a
c
c
ur
a
c
y of
di
s
e
a
s
e
r
e
c
ogni
ti
on i
s
be
tt
e
r
(
97.44%
)
w
it
h
r
a
ndom f
or
e
s
t
, t
hough.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 4, A
ugus
t
2025
:
2935
-
2944
2940
2.9.
M
od
e
l
s
t
u
d
y
T
he
s
tu
dy
f
oc
us
e
s
on
c
l
a
s
s
if
yi
ng
a
lg
or
it
hm
s
on
th
e
ba
s
is
of
a
c
c
ur
a
c
y.
E
a
c
h
m
od
e
l
di
s
c
r
im
in
a
te
s
by
s
om
e
m
e
th
od
-
in
s
ta
nc
e
-
ba
s
e
d
le
a
r
ni
ng,
e
ns
e
m
bl
e
le
a
r
ni
ng,
or
ne
ur
a
l
ne
twor
k.
T
he
goa
l
is
to
e
s
ta
bl
i
s
h
w
hi
c
h
m
ode
l
is
m
os
t
r
e
li
a
bl
e
f
or
c
la
s
s
if
ic
a
ti
on pur
pos
e
s
.
‒
K
-
S
ta
r
:
a
s
lo
w
, s
tr
a
ig
ht
f
or
w
a
r
d,
a
nd i
ns
ta
nc
e
-
ba
s
e
d c
la
s
s
if
ie
r
i
s
t
he
K
-
S
ta
r
m
ode
l.
E
nt
r
opy
m
e
a
s
ur
e
m
e
nt
is
th
e
pr
in
c
ip
a
l
us
e
of
th
e
K
-
S
ta
r
a
lg
or
it
hm
.
T
he
r
e
a
r
e
a
lo
t
of
m
e
r
it
s
w
he
n
ut
il
iz
in
g
e
nt
r
opy
f
or
ga
ugi
ng
di
s
ta
nc
e
. A
lt
hough we
a
c
hi
e
ve
a
good
c
la
s
s
if
ic
a
ti
on a
c
c
ur
a
c
y o
f
86.08%
w
it
h
K
-
S
ta
r
.
‒
R
a
ndom
c
om
m
it
te
e
:
it
i
s
a
s
upe
r
vi
s
e
d
a
lg
or
it
hm
f
r
om
th
e
M
e
ta
gr
oup.
T
he
ov
e
r
a
ll
m
e
a
n
of
th
e
pr
e
di
c
ti
ons
ge
ne
r
a
te
d
by
e
ve
r
y
ba
s
e
c
la
s
s
if
ie
r
is
th
e
n
a
ppl
i
e
d
to
m
a
ke
th
e
ul
ti
m
a
te
c
la
s
s
if
ic
a
ti
on
de
c
is
io
n. T
h
e
s
upe
r
vi
s
e
d l
e
a
r
ni
ng me
th
od i
s
m
uc
h s
im
pl
e
r
, w
it
h a
n a
c
c
ur
a
c
y
s
c
or
e
of
80.40%
.
‒
I
ns
ta
nc
e
-
ba
s
e
d
a
lg
or
it
hm
(
I
BK
)
:
th
e
I
B
K
is
a
not
he
r
na
m
e
f
or
K
N
N
.
E
uc
li
de
a
n
di
s
ta
nc
e
m
a
tr
ix
w
a
s
e
m
pl
oye
d.
H
e
r
e
,
th
e
K
va
lu
e
f
or
e
c
a
s
ts
th
e
num
be
r
of
c
lo
s
e
s
t
ne
ig
hbor
s
ba
s
e
d
on
th
e
di
s
ta
nc
e
be
tw
e
e
n
th
e
m
. T
he
I
B
K
pr
ovi
de
s
84.88%
a
c
c
ur
a
c
y.
‒
B
a
ggi
ng:
th
e
ba
ggi
ng
c
la
s
s
if
ie
r
is
a
m
e
ta
-
a
lg
or
it
hm
te
c
hni
que
th
a
t
th
e
m
a
c
hi
ne
le
a
r
ni
ng
e
ns
e
m
bl
e
us
e
s
to
ge
ne
r
a
te
c
la
s
s
if
ie
r
s
f
or
e
a
c
h
s
a
m
pl
e
of
th
e
tr
a
in
in
g
da
ta
s
e
t.
I
t
is
us
e
d
in
m
a
c
hi
ne
le
a
r
ni
ng
to
c
he
c
k
s
ta
bi
li
ty
,
in
c
r
e
a
s
e
a
c
c
ur
a
c
y,
r
e
duc
e
va
r
ia
ti
on,
a
nd
pr
e
ve
nt
ove
r
f
it
ti
ng.
T
he
c
la
s
s
if
ie
r
pr
ovi
de
s
84.88%
a
c
c
ur
a
c
y.
‒
M
ul
ti
la
ye
r
pe
r
c
e
pt
r
on
(
M
L
P
)
:
th
e
M
L
P
s
is
c
om
bi
ne
hi
dde
n
la
ye
r
s
a
nd
th
e
ba
c
kpr
opa
ga
ti
on
m
e
th
od.
M
L
P
ha
s
be
e
n
u
s
e
d
by
m
a
ny
a
c
a
de
m
ic
s
s
in
c
e
it
f
unc
ti
on
s
be
tt
e
r
w
it
h
num
e
r
ic
a
l
va
lu
e
s
.
W
it
h
th
is
a
lg
or
it
hm
, t
he
a
c
c
ur
a
c
y s
c
or
e
i
s
88.40%
.
‒
R
a
ndomi
z
a
bl
e
f
il
te
r
e
d
c
la
s
s
if
ie
r
:
a
ba
s
ic
f
il
te
r
e
d
c
la
s
s
if
ie
r
a
dj
us
tm
e
nt
s
how
in
g
th
e
m
ode
l
w
it
h
a
r
a
ndomi
z
a
bl
e
f
il
te
r
a
c
ti
ng
a
s
th
e
m
a
in
c
la
s
s
if
ie
r
.
E
a
c
h
ba
s
e
c
la
s
s
if
ie
r
is
bui
lt
us
in
g
a
uni
que
r
a
ndom
num
be
r
s
e
e
d
ut
il
iz
in
g
a
r
a
ndomi
z
a
bl
e
f
il
te
r
c
la
s
s
if
ie
r
s
in
c
e
it
is
a
n
e
n
s
e
m
bl
e
ba
s
e
c
la
s
s
if
ie
r
.
T
h
e
f
in
a
l
out
c
om
e
is
e
s
t
a
bl
is
he
d
by
a
ve
r
a
gi
ng
th
e
pr
e
di
c
ti
ons
of
e
a
c
h
ba
s
e
c
l
a
s
s
if
ie
r
.
T
h
e
a
c
c
ur
a
c
y
of
th
is
c
la
s
s
if
ie
r
i
s
81.52%
.
‒
M
ul
ti
c
la
s
s
c
la
s
s
if
ie
r
:
w
he
n
us
in
g
2
-
c
la
s
s
c
la
s
s
if
ie
r
s
to
a
na
l
yz
e
m
ul
ti
c
la
s
s
da
t
a
s
e
t
s
,
th
e
m
ul
ti
c
la
s
s
c
la
s
s
if
ie
r
is
a
M
e
ta
te
c
hni
que
.
T
hi
s
c
la
s
s
if
ie
r
c
a
n
a
l
s
o
a
ppl
y
e
r
r
or
-
c
or
r
e
c
ti
ng
out
put
c
ode
s
.
W
e
ge
t
83.76%
a
c
c
ur
a
c
y us
in
g t
he
m
ul
ti
c
la
s
s
C
la
s
s
if
ie
r
a
lg
or
it
hm
.
‒
S
uppor
t
ve
c
to
r
m
a
c
hi
ne
:
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
a
r
e
s
upe
r
vi
s
e
d
le
a
r
ni
ng
m
ode
ls
w
it
h
r
e
la
te
d
l
e
a
r
ni
ng
a
lg
or
it
hm
s
th
a
t
a
r
e
us
e
d
in
m
a
c
hi
ne
le
a
r
ni
ng
to
a
na
ly
z
e
d
a
t
a
f
or
r
e
gr
e
s
s
io
n
a
nd
c
l
a
s
s
if
ic
a
ti
on.
T
he
non
-
li
ne
a
r
pr
obl
e
m
o
f
c
la
s
s
if
ic
a
ti
on
c
oul
d
a
ls
o
be
r
e
s
ol
ve
d
w
it
h
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
w
it
h
th
e
us
e
of
ke
r
ne
l
f
unc
ti
ons
w
hi
c
h
tr
a
ns
f
or
m
th
e
in
f
o
r
m
a
ti
on
th
a
t
a
r
r
iv
e
s
in
to
a
s
pa
c
e
w
it
h
m
or
e
di
m
e
ns
io
ns
.
T
he
a
c
c
ur
a
c
y s
c
or
e
us
in
g t
hi
s
a
lg
or
it
hm
i
s
94.80%
.
‒
R
a
ndom
f
or
e
s
t:
tr
a
di
ti
ona
l
m
a
c
hi
ne
le
a
r
ni
ng
us
e
s
th
e
s
upe
r
vi
s
e
d
r
a
ndom
f
or
e
s
t
a
lg
or
it
hm
.
I
t
us
e
s
le
a
r
ni
ng
ba
s
e
d
on
de
c
i
s
io
n
tr
e
e
s
.
B
e
c
a
us
e
it
c
a
n
ha
ndl
e
bot
h
c
a
te
gor
ic
a
l
a
nd
num
e
r
ic
a
l
d
a
ta
,
it
is
a
s
tr
ong
c
la
s
s
if
ie
r
.
T
he
r
a
ndom
f
or
e
s
t
m
e
th
od
is
s
im
pl
e
e
nough
f
or
hum
a
ns
to
unde
r
s
ta
nd.
T
h
e
ul
ti
m
a
te
r
e
s
ul
t
f
or
i
ll
ne
s
s
i
de
nt
if
ic
a
ti
on i
s
de
te
r
m
in
e
d by the
hi
ghe
s
t
nu
m
be
r
of
vot
e
s
r
e
c
e
iv
e
d f
r
om
e
a
c
h de
c
is
io
n
tr
e
e
node
. C
om
pa
r
e
d t
o ot
he
r
c
la
s
s
if
ie
r
s
, t
hi
s
one
ha
s
a
hi
ghe
r
a
c
c
ur
a
c
y s
c
or
e
of
97.44%
.
‒
P
A
R
T
:
a
r
ul
e
s
-
ba
s
e
d
c
la
s
s
if
ie
r
is
a
P
A
R
T
.
I
n
th
is
r
ul
e
s
-
ba
s
e
d
c
la
s
s
if
ie
r
,
c
la
s
s
pr
e
di
c
ti
on
is
done
us
in
g
a
s
s
oc
ia
ti
on
r
ul
e
s
a
m
ong
a
ll
th
e
a
tt
r
ib
ut
e
s
.
T
h
e
s
e
pr
e
c
is
e
f
or
e
c
a
s
ts
a
r
e
r
e
ga
r
de
d
a
s
c
ove
r
a
ge
.
T
h
e
y
ha
v
e
th
e
c
a
pa
c
it
y t
o f
or
e
c
a
s
t
m
or
e
t
ha
n one
c
onc
lu
s
io
n.
T
he
a
c
c
ur
a
c
y of
t
hi
s
c
la
s
s
if
ie
r
i
s
78.76%
.
3.
R
E
S
U
L
T
S
A
N
D
A
N
A
L
Y
S
I
S
500 photos
of
va
r
io
us
m
a
ngo
i
ll
ne
s
s
e
s
a
r
e
c
ol
le
c
te
d f
r
om
R
a
js
ha
hi
, B
a
ngl
a
de
s
h f
or
our
e
xpe
r
im
e
nt
a
l
in
ve
s
ti
ga
ti
on,
a
nd
a
ddi
ti
ona
l
im
a
ge
s
a
r
e
c
ol
le
c
te
d
onl
in
e
.
F
o
r
e
a
c
h
of
th
e
f
ol
lo
w
in
g
c
a
te
gor
ie
s
:
pow
de
r
y
m
il
de
w
,
a
nt
hr
a
c
nos
e
/b
lo
s
s
om
bl
ig
ht
,
m
a
ngo
de
f
or
m
it
y,
le
a
f
d
is
e
a
s
e
s
,
f
r
ui
t
di
s
e
a
s
e
s
,
de
c
li
ne
di
s
or
de
r
s
,
a
nd
he
a
lt
hy
im
a
ge
s
,
th
e
pi
c
tu
r
e
d
a
ta
ba
s
e
in
c
lu
de
s
100
im
a
g
e
s
.
I
m
a
ge
s
th
a
t
ha
ve
b
e
e
n
a
c
qui
r
e
d
a
r
e
pr
e
-
pr
oc
e
s
s
e
d
us
in
g
te
c
hni
que
s
in
c
lu
di
ng
noi
s
e
r
e
duc
ti
on,
c
ont
r
a
s
t
e
nh
a
nc
e
m
e
nt
,
a
nd
f
il
te
r
in
g.
E
uc
li
de
a
n
di
s
ta
n
c
e
m
e
tr
ic
s
a
nd
k
-
m
e
a
ns
c
lu
s
te
r
in
g
a
r
e
us
e
d
to
s
e
gm
e
nt
im
a
ge
s
.
10
f
e
a
tu
r
e
s
a
r
e
th
e
n
ta
ke
n
f
r
om
th
e
im
a
ge
a
nd
s
a
ve
d
f
or
la
te
r
a
na
ly
s
is
.
T
he
r
e
a
r
e
no
lo
nge
r
a
ny
unne
c
e
s
s
a
r
y
f
e
a
tu
r
e
s
.
T
hi
s
a
ll
ow
s
th
e
di
s
e
a
s
e
s
to
be
id
e
nt
if
ie
d
a
nd
c
a
te
gor
iz
e
d
us
in
g
r
a
ndom
f
or
e
s
t
.
T
he
s
te
p
-
by
-
s
te
p
m
a
ngo
di
s
e
a
s
e
de
te
c
ti
on
pr
oc
e
s
s
a
na
ly
s
i
s
is
s
how
n
in
T
a
bl
e
5.
T
h
e
r
a
ndom
f
or
e
s
t
m
e
th
od
pr
oduc
e
d
th
e
hi
gh
e
s
t
a
c
c
ur
a
c
y
(
97.44%
)
in
th
is
a
na
ly
s
i
s
a
nd
pe
r
f
or
m
a
nc
e
of
r
a
ndom
f
o
r
e
s
t
c
la
s
s
if
ie
r
is
s
how
n
in
T
a
bl
e
6.
T
he
c
om
p
a
r
is
o
n
of
th
e
ot
he
r
c
la
s
s
if
ie
r
s
is
s
how
n
in
F
ig
ur
e
4
.
I
n
e
a
r
li
e
r
s
tu
di
e
s
,
r
e
s
e
a
r
c
he
r
s
a
tt
e
m
pt
e
d
to
c
r
e
a
te
a
uni
que
m
od
e
l
to
id
e
nt
if
y
pa
r
ti
c
ul
a
r
d
is
e
a
s
e
s
li
ke
onl
y
le
a
f
di
s
e
a
s
e
or
onl
y
f
r
ui
t
di
s
e
a
s
e
s
[
3]
,
[
5]
,
[
16
]
–
[
19]
.
B
ut
in
th
is
r
e
s
e
a
r
c
h,
w
e
tr
ie
d
to
de
te
c
t
a
hi
ghe
r
num
be
r
of
m
a
ngo dis
e
a
s
e
s
(
li
ke
le
a
f
di
s
e
a
s
e
s
, f
r
ui
t
di
s
e
a
s
e
s
, a
nd othe
r
s
)
us
in
g m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd ge
t
good a
c
c
ur
a
c
y.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
age
analy
s
is
and mac
hi
ne
l
e
ar
ni
ng t
e
c
hni
que
s
f
or
ac
c
ur
at
e
d
e
te
c
ti
on of
…
(
M
d A
bdul
la
h A
l
R
ahi
b
)
2941
T
a
bl
e
5. S
te
p by s
te
p m
a
ngo dis
e
a
s
e
de
te
c
ti
on pr
oc
e
s
s
a
na
ly
s
i
s
O
r
i
gi
na
l
i
m
a
ge
R
e
s
i
z
e
d i
m
a
ge
C
ont
r
a
s
t
-
e
nha
nc
e
d i
m
a
ge
S
e
gm
e
nt
e
d i
m
a
ge
T
ype
D
e
c
l
i
ne
d
i
s
e
a
s
e
s
P
os
t
-
ha
r
ve
s
t
di
s
e
a
s
e
L
e
a
f
di
s
e
a
s
e
s
F
l
ow
e
r
di
s
e
a
s
e
H
e
a
l
t
hy
m
a
ngo
T
a
bl
e
6.
P
e
r
f
or
m
a
nc
e
of
r
a
ndom f
or
e
s
t
c
la
s
s
if
ie
r
T
ype
P
r
e
c
i
s
i
on (
%
)
S
pe
c
i
f
i
c
i
t
y (
%
)
S
e
ns
i
t
i
vi
t
y (
%
)
F
N
R
(
%
)
F
P
R
(
%
)
A
c
c
ur
a
c
y (
%
)
A
vg. a
c
c
ur
a
c
y (
%
)
F
l
ow
e
r
di
s
e
a
s
e
s
93.00
98.25
93.00
7.00
1.75
97.20
L
e
a
f
di
s
e
a
s
e
s
93.81
98.50
91.00
9.00
1.50
97.00
P
os
t
ha
r
ve
s
t
di
s
e
a
s
e
s
92.16
98.00
94.00
6.00
2.00
97.20
97.44
D
e
c
l
i
ne
di
s
or
de
r
s
95.15
98.75
98.00
2.00
1.25
98.60
H
e
a
l
t
hy m
a
ngo
93.88
98.50
92.00
8.00
1.50
97.20
F
ig
ur
e
4
. C
om
pa
r
is
on of
t
he
ot
he
r
c
la
s
s
if
ie
r
s
4.
C
O
N
C
L
U
S
I
O
N
B
a
ngl
a
de
s
h
l
e
a
ds
th
e
w
a
y
in
pi
one
e
r
in
g
a
ut
om
a
ti
c
m
a
ngo
di
s
e
a
s
e
r
e
c
ogni
ti
on,
c
oupl
in
g
it
w
it
h
da
ta
m
in
in
g
te
c
hni
que
s
to
f
or
e
c
a
s
t
c
r
op
yi
e
ld
s
ba
s
e
d
on
c
li
m
a
te
in
p
ut
s
.
T
hi
s
in
nova
ti
ve
s
ys
te
m
a
im
s
to
a
c
c
ur
a
te
ly
id
e
nt
if
y
va
r
io
us
m
a
ngo
a
il
m
e
nt
s
,
in
c
lu
di
ng
de
c
li
ne
di
s
or
de
r
s
,
f
r
ui
t,
pos
t
-
ha
r
ve
s
t,
le
a
f
,
a
nd
f
lo
w
e
r
di
s
e
a
s
e
s
,
a
s
w
e
ll
a
s
he
a
lt
hy
m
a
ngo
f
r
ui
t.
C
om
m
e
nc
in
g
pos
t
-
phot
o
c
a
pt
ur
e
,
th
e
pr
oc
e
s
s
in
vol
ve
s
e
s
s
e
nt
ia
l
pr
e
pr
oc
e
s
s
in
g
s
te
ps
,
s
u
c
h
a
s
im
a
ge
r
e
s
iz
in
g,
c
ont
r
a
s
t
e
nha
nc
e
m
e
nt
,
a
n
d
R
G
B
c
ol
or
s
c
he
m
e
a
dj
us
tm
e
nt
.
P
ic
tu
r
e
s
e
gm
e
nt
a
ti
on,
f
a
c
il
it
a
te
d
by
E
uc
li
de
a
n
di
s
ta
nc
e
a
nd
k
-
m
e
a
ns
c
lu
s
te
r
in
g,
pr
e
c
e
de
s
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
ut
il
iz
in
g
two
s
e
ts
of
G
L
C
M
f
e
a
tu
r
e
s
,
to
ta
li
ng
10
a
tt
r
ib
ut
e
s
.
T
he
s
e
e
xt
r
a
c
te
d
f
e
a
tu
r
e
s
pow
e
r
da
ta
c
la
s
s
if
ic
a
ti
on,
r
e
s
ul
ti
ng i
n a
n i
m
pr
e
s
s
iv
e
a
ve
r
a
ge
a
c
c
ur
a
c
y r
a
te
of
97.44%
f
or
il
ln
e
s
s
i
de
nt
if
ic
a
ti
on. T
hi
s
i
nt
e
gr
a
te
d a
ppr
oa
c
h
s
ig
ni
f
ie
s
a
s
ig
ni
f
ic
a
nt
le
a
p
f
or
w
a
r
d
in
m
a
ngo
di
s
e
a
s
e
m
a
n
a
ge
m
e
nt
,
pr
om
is
in
g
im
pr
ove
d
c
r
op
he
a
lt
h
a
nd
yi
e
ld
pr
e
di
c
ti
ons
.
A
C
K
N
O
WL
E
D
G
M
E
N
T
S
W
e
w
oul
d
li
ke
to
th
a
nk
D
I
U
,
N
L
P
,
a
nd
M
L
R
e
s
e
a
r
c
h
L
a
b
a
t
D
a
f
f
odi
l
I
nt
e
r
na
ti
ona
l
U
ni
ve
r
s
it
y.
W
e
w
oul
d l
ik
e
t
o e
xpr
e
s
s
our
s
in
c
e
r
e
gr
a
ti
tu
de
a
nd t
h
a
nks
t
o t
he
uni
ve
r
s
it
y a
ut
hor
it
y.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 4, A
ugus
t
2025
:
2935
-
2944
2942
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
he
a
ut
hor
s
de
c
l
a
r
e
th
a
t
th
i
s
r
e
s
e
a
r
c
h
ha
s
not
be
e
n
f
in
a
nc
ia
ll
y
s
uppor
te
d.
I
t
w
a
s
c
a
r
r
ie
d
out
in
de
pe
nde
nt
ly
a
nd no f
in
a
nc
ia
l
a
s
s
is
ta
n
c
e
f
r
om
a
ny f
undi
ng a
ge
nc
y, i
ns
ti
tu
ti
on, or
or
ga
ni
z
a
ti
on w
a
s
a
c
c
e
pt
e
d.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
M
d A
bdul
la
h A
l
R
a
hi
b
✓
✓
✓
✓
✓
✓
✓
✓
N
a
z
ni
n S
ul
ta
na
✓
✓
✓
✓
✓
✓
✓
N
ir
jh
or
S
a
ha
✓
✓
✓
✓
✓
✓
R
a
ju
M
ia
✓
✓
✓
✓
✓
✓
M
oni
s
ha
S
a
r
ke
r
✓
✓
✓
✓
✓
✓
A
bdus
S
a
tt
a
r
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ll
a
ut
hor
he
r
e
by
s
ta
te
s
th
a
t
no
f
in
a
nc
ia
l,
pe
r
s
ona
l,
p
r
of
e
s
s
io
na
l,
pol
it
ic
a
l,
id
e
ol
ogi
c
a
l,
or
r
e
li
gi
ous
in
te
r
e
s
ts
e
xi
s
ts
th
a
t
c
oul
d
ha
ve
in
f
lu
e
nc
e
d
th
e
r
e
s
e
a
r
c
h
a
na
ly
s
is
,
or
in
te
r
pr
e
ta
ti
on
of
th
e
d
a
ta
pr
e
s
e
nt
e
d
h
e
r
e
in
.
C
om
pe
ti
ng i
nt
e
r
e
s
ts
i
f
a
ny ha
ve
be
e
n di
s
c
lo
s
e
d by a
ll
a
ut
hor
s
, a
nd none
i
m
pe
de
t
he
i
m
pa
r
ti
a
ll
y of
t
hi
s
w
or
k.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
t
a
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
a
va
il
a
bl
e
f
r
om
th
e
c
or
r
e
s
ponding
a
ut
hor
,
[
M
A
A
R
]
,
upon
r
e
a
s
ona
bl
e
r
e
que
s
t.
T
h
e
da
ta
a
r
e
not
publ
ic
ly
a
va
il
a
bl
e
due
to
pr
iv
a
c
y
c
onc
e
r
ns
a
nd
in
s
ti
tu
ti
ona
l
pol
ic
y r
e
s
tr
ic
ti
ons
r
e
ga
r
di
ng t
he
ha
ndl
in
g of
s
e
ns
it
i
ve
or
pe
r
s
ona
ll
y i
de
nt
if
ia
bl
e
i
nf
or
m
a
ti
on.
R
E
F
E
R
E
N
C
E
S
[
1]
E
.
E
.
S
a
e
e
d
e
t
al
.
,
“
D
e
t
e
c
t
i
on
a
nd
m
a
na
g
e
m
e
nt
of
m
a
ngo
di
e
ba
c
k
di
s
e
a
s
e
i
n
t
he
U
ni
t
e
d
A
r
a
b
E
m
i
r
a
t
e
s
,”
I
nt
e
r
nat
i
onal
J
ou
r
nal
o
f
M
ol
e
c
ul
ar
Sc
i
e
nc
e
s
, vol
. 18, no. 10, 2017, doi
:
10.3390/
i
j
m
s
18102086.
[
2]
S
.
K
ha
n,
R
.
U
l
l
a
h,
H
.
A
l
i
,
A
.
W
a
he
e
d,
a
nd
Q
.
A
bba
s
,
“
E
l
e
m
e
nt
a
l
a
na
l
ys
i
s
of
m
a
ngo
r
i
pe
ne
d
by
di
f
f
e
r
e
nt
pos
t
ha
r
ve
s
t
t
r
e
a
t
m
e
nt
s
us
i
ng l
a
s
e
r
i
nduc
e
d br
e
a
kdow
n
s
pe
c
t
r
os
c
opi
c
,
”
O
pt
i
k
, vol
. 246, 2021, doi
:
10.1016/
j
.i
j
l
e
o.2021.167770.
[
3]
M
.
R
.
M
i
a
,
S
.
R
oy,
S
.
K
.
D
a
s
,
a
nd
M
.
A
.
R
a
hm
a
n,
“
M
a
ngo
l
e
a
f
di
s
e
a
s
e
r
e
c
ogni
t
i
on
us
i
ng
ne
ur
a
l
ne
t
w
or
k
a
nd
s
uppor
t
ve
c
t
or
m
a
c
hi
ne
,”
I
r
an J
our
nal
of
C
o
m
put
e
r
Sc
i
e
n
c
e
, vol
. 3, no. 3, pp. 185
–
193, 2020,
doi
:
10.1007/
s
42044
-
020
-
00057
-
z.
[
4]
P
. K
um
a
r
i
, P
. R
a
ke
s
h, a
nd
R
. S
i
ngh, “
A
nt
hr
a
c
no
s
e
of
m
a
ngo i
nc
i
t
e
d by
c
ol
l
e
t
ot
r
i
c
hum
gl
oe
os
por
i
oi
de
s
:
a
c
om
pr
e
he
ns
i
v
e
r
e
vi
e
w
,
”
I
nt
e
r
nat
i
onal
J
our
nal
of
P
ur
e
&
A
ppl
i
e
d B
i
os
c
i
e
nc
e
, vol
. 5, no. 1, pp. 48
–
56, 2017, doi
:
10.18782/
2320
-
7051.2478.
[
5]
T
.
N
.
P
ha
m
,
L
.
V
.
T
r
a
n,
a
nd
S
.
V
.
T
.
D
a
o,
“
E
a
r
l
y
di
s
e
a
s
e
c
l
a
s
s
i
f
i
c
a
t
i
on
of
m
a
ngo
l
e
a
ve
s
us
i
ng
f
e
e
d
-
f
or
w
a
r
d
ne
ur
a
l
ne
t
w
or
k
a
nd
hybr
i
d m
e
t
a
he
ur
i
s
t
i
c
f
e
a
t
ur
e
s
e
l
e
c
t
i
on,”
I
E
E
E
A
c
c
e
s
s
, vol
. 8, pp. 189960
–
189973, 2020, doi
:
10.1109/
A
C
C
E
S
S
.2020.3031914.
[
6]
A
.
M
.
I
s
m
a
i
l
,
G
.
C
i
r
v
i
l
l
e
r
i
,
G
.
P
ol
i
z
z
i
,
P
.
W
.
C
r
ous
,
J
.
Z
.
G
r
oe
ne
w
a
l
d,
a
nd
L
.
L
om
ba
r
d,
“
L
as
i
odi
pl
odi
a
s
pe
c
i
e
s
a
s
s
oc
i
a
t
e
d
w
i
t
h
di
e
ba
c
k
di
s
e
a
s
e
of
m
a
ngo
(
m
angi
f
e
r
a
i
ndi
c
a
)
i
n
E
gypt
,”
A
us
t
r
al
as
i
an
P
l
an
t
P
at
hol
ogy
,
vol
.
41,
no.
6,
pp.
649
–
660,
2012,
doi
:
10.1007/
s
13313
-
012
-
0163
-
1.
[
7]
R
.
S
a
l
e
e
m
,
J
.
H
.
S
ha
h,
M
.
S
ha
r
i
f
,
a
nd
G
.
J
.
A
ns
a
r
i
,
“
M
a
ngo
l
e
a
f
di
s
e
a
s
e
i
de
nt
i
f
i
c
a
t
i
on
us
i
ng
f
ul
l
y
r
e
s
ol
ut
i
on
c
onvol
ut
i
ona
l
ne
t
w
or
k,”
C
om
put
e
r
s
, M
at
e
r
i
al
s
and C
ont
i
nua
, vol
. 69, no. 3, pp. 3581
–
3601,
2021, doi
:
10.32604/
c
m
c
.2021.017700.
[
8]
R
.
L
.
Z
ha
n
e
t
al
.
,
“
M
a
ngo
m
a
l
f
or
m
a
t
i
on
di
s
e
a
s
e
i
n
S
out
h
C
hi
na
c
a
u
s
e
d
by
f
u
s
ar
i
um
pr
ol
i
f
e
r
at
um
,”
J
our
nal
of
P
hy
t
opat
hol
ogy
,
vol
. 158, no. 11
–
12, pp. 721
–
725, 2010, doi
:
10.1111/
j
.1439
-
0434.2010.01688.
x.
[
9]
W
.
F
.
O
.
M
a
r
a
s
a
s
,
R
.
C
.
P
l
oe
t
z
,
M
.
J
.
W
i
ngf
i
e
l
d,
B
.
D
.
W
i
ngf
i
e
l
d,
a
nd
E
.
T
.
S
t
e
e
nka
m
p,
“
M
a
ngo
m
a
l
f
or
m
a
t
i
on
di
s
e
a
s
e
a
nd
t
h
e
a
s
s
oc
i
a
t
e
d
F
us
ar
i
um
s
p
e
c
i
e
s
,
”
P
hy
t
opat
hol
ogy
, vol
. 96, no. 6, pp. 667
–
672, 2006, doi
:
10.1094/
P
H
Y
T
O
-
96
-
0667.
[
10]
A
.
O
.
A
l
A
da
w
i
e
t
al
.
,
“
A
e
t
i
ol
ogy
a
nd
c
a
us
a
l
a
ge
nt
s
of
m
a
ngo
s
udde
n
de
c
l
i
ne
di
s
e
a
s
e
i
n
t
he
S
ul
t
a
na
t
e
of
O
m
a
n,”
E
ur
ope
an
J
our
nal
of
P
l
ant
P
at
hol
ogy
, vol
. 116, no. 4, pp. 247
–
254, 2006, doi
:
10.1007/
s
1
0658
-
006
-
9056
-
x.
[
11]
G
.
O
.
-
C
ol
i
na
e
t
al
.
,
“
I
de
nt
i
f
i
c
a
t
i
on
a
nd
c
h
a
r
a
c
t
e
r
i
z
a
t
i
on
of
a
nov
e
l
e
t
i
ol
ogi
c
a
l
a
ge
nt
of
m
a
ngo
m
a
l
f
or
m
a
t
i
on
di
s
e
a
s
e
i
n
M
e
xi
c
o,
f
us
ar
i
um
m
e
x
i
c
anu
m
s
p. nov
.,”
P
hy
t
opat
hol
ogy
, vol
. 100, no. 11, pp. 1176
–
118
4, 2010, doi
:
10.1094/
P
H
Y
T
O
-
01
-
10
-
0029.
[
12]
D
.
S
i
va
kum
a
r
,
Y
.
J
i
a
ng,
a
nd
E
.
M
.
Y
a
hi
a
,
“
M
a
i
nt
a
i
ni
ng
m
a
ngo
(
m
angi
f
e
r
a
i
ndi
c
a
L
.)
f
r
ui
t
qua
l
i
t
y
dur
i
ng
t
he
e
xpor
t
c
ha
i
n,”
F
ood
R
e
s
e
ar
c
h I
nt
e
r
nat
i
onal
, vol
. 44, no. 5, pp. 1254
–
1263, 2011, doi
:
10.1016/
j
.f
oodr
e
s
.2010.11.022.
[
13]
K
.
T
r
a
ng,
L
.
T
onT
ha
t
,
N
.
G
.
M
.
T
ha
o,
a
nd
N
.
T
.
T
.
T
hi
,
“
M
a
ngo
di
s
e
a
s
e
s
i
de
n
t
i
f
i
c
a
t
i
on
by
a
de
e
p
r
e
s
i
dua
l
n
e
t
w
or
k
w
i
t
h
c
ont
r
a
s
t
e
nha
nc
e
m
e
nt
a
nd
t
r
a
ns
f
e
r
l
e
a
r
ni
ng,”
i
n
2019
I
E
E
E
C
onf
e
r
e
nc
e
on
Sus
t
ai
nabl
e
U
t
i
l
i
z
at
i
on
and
D
e
v
e
l
opm
e
nt
i
n
E
ngi
ne
e
r
i
ng
and
T
e
c
hnol
ogi
e
s
(
C
SU
D
E
T
)
, 2019, pp. 138
–
142
,
doi
:
10.1109/
C
S
U
D
E
T
47057.20
19.9214620.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
I
m
age
analy
s
is
and mac
hi
ne
l
e
ar
ni
ng t
e
c
hni
que
s
f
or
ac
c
ur
at
e
d
e
te
c
ti
on of
…
(
M
d A
bdul
la
h A
l
R
ahi
b
)
2943
[
14]
Z
a
i
nur
i
,
D
.
C
.
J
oyc
e
,
A
.
H
.
W
e
a
r
i
ng,
L
.
C
oa
t
e
s
,
a
nd
L
.
T
e
r
r
y,
“
E
f
f
e
c
t
s
of
phos
phona
t
e
a
nd
s
a
l
i
c
yl
i
c
a
c
i
d
t
r
e
a
t
m
e
nt
s
o
n
a
nt
hr
a
c
nos
e
di
s
e
a
s
e
de
ve
l
opm
e
nt
a
nd
r
i
pe
ni
ng
of
‘
ke
ns
i
ngt
on
pr
i
de
’
m
a
ngo
f
r
ui
t
,”
A
us
t
r
al
i
an
J
our
nal
of
E
x
pe
r
i
m
e
nt
al
A
gr
i
c
ul
t
ur
e
, vol
. 41, no. 6, pp. 805
–
813, 2001, doi
:
10.1071/
E
A
99104.
[
15]
R
.
A
.
J
.
M
.
G
i
n
i
ng
e
t
a
l
.
,
“
H
a
r
u
m
a
ni
s
m
a
n
go l
e
a
f
di
s
e
a
s
e
r
e
c
ogn
i
t
i
o
n
s
ys
t
e
m
us
i
ng i
m
a
ge
pr
oc
e
s
s
i
ng
t
e
c
hn
i
q
ue
,”
I
nd
one
s
i
an
J
o
ur
na
l
of
E
l
e
c
t
r
i
c
al
E
ng
i
ne
e
r
i
n
g
and
C
o
m
pu
t
e
r
Sc
i
e
nc
e
,
vo
l
.
23
, n
o.
1,
p
p.
37
8
–
3
86
, 2
02
1,
do
i
:
10
.11
59
1/
i
j
e
e
c
s
.v
23
.i
1.p
p3
78
-
3
86.
[
16]
N
.
R
a
ha
m
a
n
e
t
al
.
,
“
A
d
e
e
p
l
e
a
r
ni
ng
ba
s
e
d
s
m
a
r
t
phone
a
ppl
i
c
a
t
i
on
f
or
de
t
e
c
t
i
ng
m
a
ngo
di
s
e
a
s
e
s
a
nd
pe
s
t
i
c
i
de
s
ugg
e
s
t
i
ons
,
”
I
nt
e
r
nat
i
onal
J
our
nal
of
C
om
put
i
ng and D
i
gi
t
al
Sy
s
t
e
m
s
, vol
. 13, no. 1, pp. 1273
–
1286, 2023, doi
:
10.12785/
i
j
c
ds
/
1301104.
[
17]
U
.
P
.
S
i
ngh,
S
.
S
.
C
houha
n,
S
.
J
a
i
n,
a
nd
S
.
J
a
i
n,
“
M
ul
t
i
l
a
ye
r
c
onvol
ut
i
on
ne
ur
a
l
ne
t
w
or
k
f
o
r
t
he
c
l
a
s
s
i
f
i
c
a
t
i
on
of
m
a
ngo
l
e
a
v
e
s
i
nf
e
c
t
e
d by a
nt
hr
a
c
nos
e
di
s
e
a
s
e
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 7, pp. 43721
–
43729, 2019,
doi
:
10.1109/
A
C
C
E
S
S
.2019.2907383.
[
18]
A
.
R
a
j
bongs
hi
,
T
.
K
ha
n,
M
.
M
.
R
a
hm
a
n,
A
.
P
r
a
m
a
ni
k,
S
.
M
.
T
.
S
i
ddi
que
e
,
a
nd
N
.
R
.
C
ha
kr
a
bor
t
y,
“
R
e
c
ogni
t
i
on
of
m
a
ngo
l
e
a
f
di
s
e
a
s
e
us
i
ng
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
k
m
ode
l
s
:
a
t
r
a
ns
f
e
r
l
e
a
r
ni
ng
a
ppr
oa
c
h,”
I
ndone
s
i
an
J
our
nal
of
E
l
e
c
t
r
i
c
al
E
ngi
ne
e
r
i
ng
and C
om
put
e
r
Sc
i
e
n
c
e
, vol
. 23, no. 3, pp. 1681
–
1688, 2021, doi
:
10.11591/
i
j
e
e
c
s
.v23.i
3.pp1681
-
1688.
[
19]
A
.
S
.
-
P
l
a
nc
a
r
t
e
e
t
a
l
.
,
“
M
a
ngo
nur
s
e
r
i
e
s
a
s
s
our
c
e
s
of
F
us
ar
i
um
m
e
x
i
c
anum
,
c
a
us
e
of
m
a
ngo
m
a
l
f
or
m
a
t
i
on
di
s
e
a
s
e
i
n
c
e
nt
r
a
l
w
e
s
t
e
r
n M
e
xi
c
o,”
P
hy
t
opa
r
as
i
t
i
c
a
, vol
. 43, no. 4, pp. 427
–
435, 2015, doi
:
10.1
007/
s
12600
-
015
-
0471
-
4.
[
20]
M
.
S
i
l
i
m
e
l
a
a
nd
L
.
K
or
s
t
e
n,
“
E
va
l
ua
t
i
on
of
pr
e
-
ha
r
ve
s
t
bac
i
l
l
us
l
i
c
he
ni
f
or
m
i
s
s
pr
a
ys
t
o
c
ont
r
ol
m
a
ngo
f
r
ui
t
di
s
e
a
s
e
s
,”
C
r
o
p
P
r
ot
e
c
t
i
on
, vol
. 26,
no. 10, pp. 1474
–
1481, 2007, doi
:
10.1016/
j
.c
r
opr
o.2006.12
.011.
[
21]
E
.
G
.
-
A
t
i
ns
ky,
A
.
S
z
t
e
j
nbe
r
g,
M
.
M
a
ym
on,
H
.
V
i
nt
a
l
,
D
.
S
ht
i
e
nbe
r
g,
a
n
d
S
.
F
r
e
e
m
a
n,
“
I
nf
e
c
t
i
on
dyna
m
i
c
s
of
f
u
s
ar
i
u
m
m
angi
f
e
r
ae
,
c
a
us
a
l
a
ge
nt
of
m
a
ngo
m
al
f
or
m
at
i
on
di
s
e
as
e
,”
P
hy
t
opat
hol
ogy
,
vol
.
99,
no.
6,
pp.
775
–
781,
2009,
doi
:
10.1094/
P
H
Y
T
O
-
99
-
6
-
0775.
[
22]
S
.
W
ongs
i
l
a
,
P
.
C
ha
nt
r
a
s
r
i
,
a
nd
P
.
S
ur
e
e
phong,
“
M
a
c
hi
n
e
l
e
a
r
ni
ng
a
l
gor
i
t
hm
de
ve
l
opm
e
nt
f
or
de
t
e
c
t
i
on
of
m
a
ngo
i
nf
e
c
t
e
d
by
ant
hr
ac
nos
e
di
s
e
as
e
,
”
i
n
2021 J
oi
nt
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on D
i
gi
t
al
A
r
t
s
,
M
e
di
a and T
e
c
hnol
ogy
w
i
t
h E
C
T
I
N
or
t
he
r
n S
e
c
t
i
on
C
onf
e
r
e
nc
e
on
E
l
e
c
t
r
i
c
al
,
E
l
e
c
t
r
oni
c
s
,
C
om
put
e
r
and
T
e
l
e
c
om
m
u
ni
c
at
i
on
E
ngi
ne
e
r
i
ng
,
2021,
pp.
249
–
252
,
doi
:
10.1109/
E
C
T
I
D
A
M
T
N
C
O
N
51128.2021.9425737.
[
23]
P
.
B
.
J
a
w
a
de
,
D
.
C
ha
ugul
e
,
D
.
P
a
t
i
l
,
a
nd
H
.
S
hi
nde
,
“
D
i
s
e
a
s
e
pr
e
di
c
t
i
on
o
f
m
a
ngo
c
r
op
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
I
oT
,”
L
e
ar
ni
ng and A
nal
y
t
i
c
s
i
n I
nt
e
l
l
i
ge
nt
Sy
s
t
e
m
s
, vol
. 3, pp. 254
–
260, 2020, doi
:
1
0.1007/
978
-
3
-
030
-
24322
-
7_33.
[
24]
P
. J
ongs
r
i
, P
. R
oj
s
i
t
t
hi
s
a
k, T
. W
a
ngs
om
boond
e
e
, a
nd K
.
S
e
r
a
yphe
a
p, “
I
nf
l
ue
nc
e
of
c
hi
t
os
a
n c
o
a
t
i
ng c
om
bi
ne
d w
i
t
h s
p
e
r
m
i
di
ne
on
a
nt
hr
a
c
nos
e
di
s
e
a
s
e
a
nd
qua
l
i
t
i
e
s
of
‘
N
a
m
D
ok
M
a
i
’
m
a
ngo
a
f
t
e
r
ha
r
ve
s
t
,”
Sc
i
e
nt
i
a
H
or
t
i
c
ul
t
ur
ae
,
vol
.
224,
pp.
180
–
187,
2017
,
doi
:
10.1016/
j
.s
c
i
e
nt
a
.2017.06.011.
[
25]
S
.
R
.
S
a
r
ke
r
,
M
.
R
.
I
s
l
a
m
,
a
nd
I
.
H
o
s
s
a
i
n,
“
P
r
e
va
l
e
nc
e
a
nd
e
c
o
-
f
r
i
e
ndl
y
m
a
na
ge
m
e
nt
of
s
om
e
i
m
por
t
a
nt
nur
s
e
r
y
di
s
e
a
s
e
s
of
m
a
ngo i
n B
a
ngl
a
de
s
h,”
J
our
nal
of
A
gr
i
c
ul
t
ur
al
Sc
i
e
n
c
e
, vol
. 8, no. 1, pp. 205
–
220, 2015, doi
:
10.5539/
j
a
s
.v8n1p205.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Md
Abdullah
Al
Rahib
received
a
bachelor
degree
in
comput
er
science
an
d
engineerin
g
from
Daffodil
Internati
onal
Universi
ty,
Dhaka,
Banglades
h
in
2023.
His
research
interests
are
machine
learning,
deep
learning,
data
mining,
data
sc
ience,
n
atural
language
processing
,
and
artificial
intelligence.
He
can
be
contacted
at
email:
abdullah15
-
12247@
diu.edu.bd.
Naznin
Sultana
is
an
Associate
Professor
in
the
Department
of
Computer
Scienc
e
and
Engine
ering,
Daff
odil
Inter
nationa
l
Univer
sity
since
2
015.
She
comple
ted
her
b
achelor
degree
from
Jahangirn
agar
Universi
ty
in
electroni
cs
and
computer
science
and
master’s
degree
from
the
same
institution
in
computer
science
and
e
ngineerin
g
.
Her
research
interest
includes
natural
language
processing
,
neural
network,
mach
ine
learning
,
and
image
processing.
She
is
an
a
ssociate
member
of
Bangladesh
Computer
Society,
a
leading
professional
and
learned
society
in
the
field
of
computers
and
information
systems
in
Banglade
sh. She c
an be
contacted
at email
: naznin
.cse@
diu.edu.
bd.
Nirjhor
Saha
received
a
bachelor
degree
in
computer
scien
ce
an
d
engineering
from
Daffodil
International
University,
Dhaka,
Bangladesh
in
2023.
His
research
interests
include
machine
learning,
data
science,
and
artificial
intell
igence.
He
can
be
contacted
at
e
mail: nirjhor15
-
12207@
diu.edu.bd.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 4, A
ugus
t
2025
:
2935
-
2944
2944
Raju
Mia
is
a
software
engineer
at
Nimusoft
Technologies
Ltd.
He
received
a
bachelor
degree
in
computer
science
and
engineering
from
Daffodil
I
nternational
University,
Dhaka,
Bangladesh
in
2023.
His
research
interests
in
clude
machine
le
arning,
data
science,
and
artificial
intell
igence
. He can be contac
ted at email: raju15
-
11995@
diu.edu.bd.
Monish
a
Sarker
completed
her
M.Sc.
in
computer
science
and
engineerin
g
at
Daffodil
International
University,
Bangladesh.
Her
research
intere
sts
include
in
machine
learning
.
She
can be cont
acted at em
ail:
sarker22205103
002@
diu.edu.bd.
Abdus S
attar
was born in Com
illa,
Bangladesh, in
1983. Currently
he is
workin
g
as
Assistant
Professor
at
the
Department
of
Computer
Science
and
Engineering
at
Daffodil
International
University
,
Faculty
of
Science
and
Information
Technol
ogy.
Previousl
y,
he
wa
s
employed
as
Assist
ant
Professo
r
in
the
Department
of
Comput
er
Sci
ence
and
Engineeri
ng
at
Britannia
Universi
ty,
Comilla.
H
e
is
a
Ph
.
D.
Student
at
Bangladesh
University
Professionals,
Banglade
sh.
He
recei
ved
Bachelo
r
of
Science
in
Computer
Science
and
Enginee
ring
from
Ahsanullah
University
of
Science
and
Technology,
Bangladesh
an
d
Master’s
Program
of
Interactive
Systems
Engineering
from
KTH
-
Royal
Institute
of
Te
chnology,
Sweden.
His
research
interests
include
machine
learning
,
IoT,
human
compute
r
interaction
,
IoT,
and
interactio
n design
. He can be contac
ted at email:
abdus.cse@
diu.edu.
bd.
Evaluation Warning : The document was created with Spire.PDF for Python.