I
nte
rna
t
io
na
l J
o
urna
l o
f
Adv
a
nces in Applie
d Science
s
(
I
J
AAS)
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
,
p
p
.
1
1
2
3
~
1
1
3
0
I
SS
N:
2252
-
8
8
1
4
,
DOI
:
1
0
.
1
1
5
9
1
/ijaas
.
v
1
5
.
i
3
.
pp
1
1
2
3
-
1
1
3
0
1123
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
a
s
.
ia
esco
r
e.
co
m
Interpre
ting po
ta
to disea
se cla
ss
ific
a
tion usin
g
expla
ina
ble
a
rtif
icia
l in
tellige
nce
Ra
k
esh
K
um
a
r
G
um
a
s
t
a
,
Aj
a
y
So
mk
uwa
r
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
mm
u
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
M
a
u
l
a
n
a
A
z
a
d
N
a
t
i
o
n
a
l
I
n
st
i
t
u
t
e
o
f
Te
c
h
n
o
l
o
g
y
,
B
h
o
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
Au
g
1
1
,
2
0
2
5
R
ev
is
ed
Ma
r
1
8
,
2
0
2
6
Acc
ep
ted
Au
g
1
3
,
2
0
2
6
Ti
m
e
ly
a
n
d
a
c
c
u
ra
te
c
r
o
p
d
ise
a
se
d
e
tec
ti
o
n
is
m
o
st
imp
o
r
tan
t
fo
r
e
n
su
ri
n
g
g
lo
b
a
l
fo
o
d
se
c
u
rit
y
.
F
o
r
d
e
tec
ti
n
g
d
ise
a
se
s in
c
ro
p
s,
m
a
n
y
d
iffere
n
t
m
a
c
h
in
e
lea
rn
in
g
(M
L)
m
o
d
e
ls
we
re
p
ro
p
o
se
d
.
T
h
e
se
m
o
d
e
ls
wo
r
k
a
s
a
b
lac
k
-
b
o
x
,
a
n
d
wit
h
o
u
t
p
r
o
p
e
r
e
x
p
la
n
a
ti
o
n
o
f
th
e
se
m
o
d
e
ls’
d
e
c
isio
n
s,
fa
rm
e
rs
m
a
y
fin
d
it
d
iffi
c
u
lt
t
o
tru
st
t
h
e
se
sy
ste
m
s.
F
o
r
t
h
is
,
m
a
n
y
m
o
d
e
l
e
x
p
lain
a
b
il
it
y
m
e
th
o
d
s
we
re
a
lso
p
r
o
p
o
se
d
.
All
th
e
se
m
e
th
o
d
s
h
a
v
e
b
e
e
n
e
v
a
lu
a
ted
q
u
a
li
tati
v
e
ly
,
b
u
t
q
u
a
n
t
it
a
ti
v
e
a
n
d
c
ro
ss
-
m
e
th
o
d
c
o
m
p
a
ris
o
n
s
a
re
lac
k
in
g
.
Th
is
st
u
d
y
a
d
d
re
ss
e
s
th
is
g
a
p
b
y
e
m
p
h
a
siz
in
g
q
u
a
n
ti
tativ
e
v
a
l
id
a
ti
o
n
o
f
m
o
d
e
l
s’
e
x
p
lan
a
ti
o
n
s.
Th
is
st
u
d
y
i
n
v
e
sti
g
a
tes
th
e
in
ter
p
re
tab
il
it
y
a
n
d
p
e
rfo
rm
a
n
c
e
o
f
two
a
p
p
r
o
a
c
h
e
s
fo
r
p
o
tato
lea
f
d
ise
a
se
c
las
sifi
c
a
ti
o
n
:
i)
m
a
n
u
a
l
fe
a
tu
re
e
n
g
in
e
e
rin
g
with
r
e
li
e
f
-
b
a
se
d
fe
a
tu
re
se
lec
ti
o
n
a
n
d
a
rti
ficia
l
n
e
u
ra
l
n
e
two
r
k
(AN
N)
c
las
sifica
t
io
n
wit
h
lo
c
a
l
i
n
terp
re
tab
le
m
o
d
e
l
-
a
g
n
o
st
ic
e
x
p
lan
a
ti
o
n
s
(LI
M
E)
a
n
d
i
i)
d
e
e
p
c
o
n
v
o
lu
t
io
n
a
l
n
e
u
ra
l
n
e
two
r
k
s
(CNN
s)
b
a
se
d
o
n
m
o
b
il
e
n
e
two
r
k
v
e
rsio
n
2
(
M
o
b
i
leN
e
tV2
)
with
g
ra
d
ien
t
-
we
ig
h
ted
c
las
s
a
c
ti
v
a
ti
o
n
m
a
p
p
i
n
g
(Gra
d
-
CAM).
Qu
a
n
ti
tativ
e
a
ss
e
ss
m
e
n
t
o
f
e
x
p
lan
a
ti
o
n
q
u
a
li
ty
wa
s
p
e
rfo
rm
e
d
u
si
n
g
f
id
e
li
t
y
a
n
d
r
o
b
u
stn
e
ss
m
e
tri
c
s.
Wh
il
e
LIM
E
a
c
h
ie
v
e
d
a
lo
we
r
a
v
e
ra
g
e
fid
e
li
ty
d
ro
p
1
0
.
9
0
%
a
n
d
h
ig
h
e
r
ro
b
u
stn
e
ss
8
4
.
2
7
%
c
o
m
p
a
re
d
t
o
G
ra
d
-
CAM
1
8
.
6
8
%
fi
d
e
li
ty
d
ro
p
a
n
d
7
9
.
3
8
%
ro
b
u
stn
e
ss
,
G
ra
d
-
CAM
p
ro
v
id
e
d
c
lea
re
r
v
isu
a
l
e
x
p
la
n
a
ti
o
n
s.
Th
e
se
fin
d
i
n
g
s
su
g
g
e
st
th
a
t
e
x
p
lan
a
ti
o
n
e
ffe
c
ti
v
e
n
e
ss
is
in
fl
u
e
n
c
e
d
b
y
m
o
d
e
l
d
e
sig
n
a
n
d
d
a
ta
c
h
a
ra
c
teristics
,
h
i
g
h
li
g
h
t
in
g
t
h
e
n
e
e
d
f
o
r
c
a
re
fu
l
se
lec
ti
o
n
o
f
in
terp
re
tab
il
it
y
tec
h
n
i
q
u
e
s i
n
p
re
c
isio
n
a
g
r
icu
lt
u
re
a
p
p
li
c
a
ti
o
n
s.
K
ey
w
o
r
d
s
:
E
x
p
l
a
i
n
a
b
l
e
a
r
ti
f
i
c
i
al
i
n
t
el
l
i
g
e
n
c
e
Fid
elity
Mo
d
el
in
ter
p
r
eta
b
ilit
y
Plan
t d
is
ea
s
e
clas
s
if
icatio
n
R
o
b
u
s
tn
ess
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
:
R
ak
esh
Ku
m
ar
Gu
m
asta
Dep
ar
tm
e
n
t
o
f
E
l
ec
t
r
o
n
ics
a
n
d
C
o
m
m
u
n
i
ca
ti
o
n
E
n
g
i
n
ee
r
i
n
g
,
Ma
u
la
n
a
A
za
d
Nat
io
n
a
l
I
n
s
tit
u
te
o
f
T
e
ch
n
o
l
o
g
y
B
h
o
p
al
4
6
2
0
0
3
,
I
n
d
ia
E
m
ail: r
ak
esh
g
u
m
asta4
4
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
E
ar
ly
d
etec
tio
n
o
f
cr
o
p
d
is
e
ases
is
e
s
s
en
tial
f
o
r
tim
ely
in
ter
v
en
tio
n
an
d
r
ed
u
cin
g
y
ie
ld
lo
s
s
es.
Ma
n
u
al
m
o
n
ito
r
in
g
o
f
lar
g
e
ag
r
icu
ltu
r
al
ar
ea
s
is
d
if
f
icu
lt;
th
er
ef
o
r
e,
in
tellig
en
t
an
d
au
to
m
ated
d
is
ea
s
e
d
etec
tio
n
s
y
s
tem
s
h
av
e
g
ain
e
d
in
cr
ea
s
in
g
atten
tio
n
.
Ma
ch
i
n
e
lear
n
in
g
(
ML
)
tech
n
i
q
u
es
h
av
e
b
ee
n
wid
ely
in
v
esti
g
ated
f
o
r
p
lan
t
d
is
ea
s
e
class
if
icatio
n
.
Fo
r
id
en
tify
in
g
r
elev
an
t
f
ea
tu
r
es,
r
elief
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
h
as b
ee
n
ex
p
lo
r
e
d
f
o
r
p
lan
t le
af
d
is
ea
s
e
clas
s
if
icatio
n
[
1
]
–
[
3
]
.
Dif
f
er
en
t M
L
class
if
ier
s
h
av
e
also
b
ee
n
ap
p
lied
to
d
is
tin
g
u
is
h
h
ea
lth
y
an
d
d
is
ea
s
ed
p
lan
ts
.
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
ANNs)
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVMs)
h
av
e
d
em
o
n
s
tr
ated
th
eir
ef
f
ec
tiv
en
ess
f
o
r
p
lan
t
d
is
ea
s
e
class
if
icat
io
n
[
4
]
,
[
5
]
.
Dee
p
lear
n
in
g
,
p
ar
ticu
lar
ly
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
s
(
C
NNs)
,
h
as
b
ee
n
wid
ely
ad
o
p
ted
f
o
r
au
t
o
m
atica
lly
ex
t
r
ac
tin
g
d
is
cr
im
in
ativ
e
p
atter
n
s
f
r
o
m
p
lan
t
im
ag
es
[
6
]
,
wh
ile
lig
h
t
weig
h
t
ar
ch
itectu
r
es
an
d
tr
an
s
f
er
lear
n
in
g
s
u
p
p
o
r
t
ef
f
icien
t
d
is
ea
s
e
id
en
tific
atio
n
[
7
]
.
Data
au
g
m
en
tatio
n
f
u
r
th
er
im
p
r
o
v
es
m
o
d
el
r
o
b
u
s
tn
ess
an
d
g
en
e
r
aliza
tio
n
u
n
d
er
v
ar
y
in
g
f
ield
co
n
d
itio
n
s
[
8
]
.
Alth
o
u
g
h
d
ee
p
e
r
an
d
m
o
r
e
co
m
p
le
x
d
ee
p
lear
n
in
g
m
o
d
els
ca
n
im
p
r
o
v
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
2
0
2
6
:
1
1
2
3
-
1
1
3
0
1124
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
[
9
]
,
[
1
0
]
,
th
eir
b
lack
-
b
o
x
n
atu
r
e
m
ak
es
th
e
r
ea
s
o
n
in
g
b
eh
in
d
p
r
e
d
ictio
n
s
d
if
f
icu
lt
to
u
n
d
er
s
tan
d
,
p
o
te
n
tially
lim
itin
g
tr
u
s
t a
n
d
p
r
ac
tical
ad
o
p
tio
n
.
E
x
p
lain
ab
le
ar
tific
ial
in
tellig
e
n
ce
(
XAI
)
m
eth
o
d
s
ad
d
r
ess
th
is
lim
itatio
n
b
y
id
en
tify
in
g
f
ea
tu
r
es
o
r
im
ag
e
r
e
g
io
n
s
i
n
f
lu
en
cin
g
m
o
d
el
d
ec
is
io
n
s
.
L
o
ca
l
in
te
r
p
r
etab
le
m
o
d
el
-
a
g
n
o
s
tic
e
x
p
lan
atio
n
s
(
L
I
ME
)
p
r
o
v
id
es
l
o
ca
l
m
o
d
el
-
ag
n
o
s
tic
ex
p
lan
atio
n
s
[
1
1
]
,
wh
il
e
g
r
ad
ie
n
t
-
weig
h
ted
class
ac
tiv
atio
n
m
ap
p
in
g
(
Gr
ad
-
C
AM
)
h
ig
h
lig
h
ts
class
-
d
is
cr
im
in
ativ
e
im
ag
e
r
eg
i
o
n
s
i
n
C
NNs
[
1
2
]
.
B
o
th
m
eth
o
d
s
h
av
e
b
ee
n
a
p
p
lied
in
p
lan
t
d
is
ea
s
e
class
if
icat
io
n
an
d
r
elate
d
ag
r
icu
ltu
r
al
ap
p
licatio
n
s
[
1
3
]
–
[
1
7
]
.
Ho
wev
e
r
,
ex
is
tin
g
s
tu
d
ies
h
av
e
p
r
e
d
o
m
in
an
tly
ev
al
u
ated
ex
p
lan
atio
n
s
q
u
alitativ
ely
,
with
lim
ited
q
u
an
titativ
e
co
m
p
ar
is
o
n
o
f
d
if
f
e
r
en
t
m
o
d
el
–
ex
p
lan
atio
n
co
m
b
i
n
atio
n
s
[
1
8
]
–
[
2
0
]
.
Alth
o
u
g
h
a
d
v
an
ce
d
e
x
p
lan
atio
n
tech
n
iq
u
es
h
av
e
also
b
ee
n
ex
p
lo
r
ed
[
2
1
]
,
[
2
2
]
,
t
h
e
r
eliab
i
lity
o
f
e
x
p
lan
atio
n
s
f
r
o
m
lig
h
t
weig
h
t
m
o
d
els
s
u
c
h
as
m
o
b
ile
n
etwo
r
k
v
er
s
io
n
2
(
Mo
b
ileNetV2
)
r
em
ain
s
in
s
u
f
f
icien
tly
q
u
an
tifie
d
[
2
3
]
.
Simp
l
er
f
ea
tu
r
e
-
b
ased
ANNs
ar
e
s
eld
o
m
b
en
c
h
m
ar
k
ed
ag
ain
s
t
d
ee
p
m
o
d
els.
R
ec
en
t
s
tu
d
ies
h
ig
h
lig
h
t
th
e
n
ee
d
f
o
r
q
u
an
titativ
e
v
alid
atio
n
o
f
m
o
d
el
e
x
p
lan
atio
n
s
[
2
4
]
,
[
2
5
]
,
a
d
ir
ec
tio
n
th
at
r
e
m
ain
s
u
n
d
er
e
x
p
lo
r
ed
in
p
lan
t
d
is
ea
s
e
r
esear
ch
.
C
o
llectiv
ely
,
ex
is
tin
g
r
esear
c
h
lack
s
a
s
y
s
tem
atic,
q
u
an
titativ
e
f
r
am
ewo
r
k
t
o
ass
ess
ex
p
lan
atio
n
f
id
elity
an
d
r
o
b
u
s
tn
ess
ac
r
o
s
s
m
o
d
el
ty
p
es.
Mo
r
eo
v
er
,
th
e
in
teg
r
atio
n
o
f
in
ter
p
r
etab
le
an
d
lig
h
tweig
h
t
m
o
d
els
f
o
r
m
o
b
ile
o
r
f
ield
-
lev
el
p
lan
t
d
is
ea
s
e
d
iag
n
o
s
is
r
em
ain
s
u
n
d
er
d
ev
elo
p
ed
,
r
ev
ea
lin
g
a
cr
itical
r
esear
ch
g
a
p
in
en
s
u
r
in
g
th
e
r
eliab
ilit
y
an
d
co
m
p
ar
ab
ilit
y
o
f
ex
p
lan
atio
n
m
eth
o
d
s
.
T
o
ad
d
r
ess
th
ese
g
ap
s
,
th
is
s
tu
d
y
p
r
o
v
id
es
a
co
m
p
ar
ativ
e
a
n
aly
s
is
o
f
two
co
m
b
in
atio
n
s
o
f
m
o
d
els
an
d
in
ter
p
r
etatio
n
m
eth
o
d
s
.
I
n
th
e
f
ir
s
t
ap
p
r
o
ac
h
,
th
is
s
tu
d
y
co
m
b
in
ed
m
a
n
u
al
f
ea
tu
r
e
e
n
g
i
n
ee
r
in
g
with
a
less
co
m
p
licated
ANN
m
o
d
el
a
n
d
th
en
ap
p
lied
t
h
e
L
I
ME
in
ter
p
r
etatio
n
m
eth
o
d
t
o
f
in
d
wh
ich
p
ar
ticu
lar
f
ea
tu
r
e
is
m
o
s
t
im
p
o
r
ta
n
t
in
m
ak
in
g
a
d
ec
is
io
n
f
o
r
a
p
ar
ticu
lar
lea
f
i
m
ag
e
.
I
n
th
e
s
ec
o
n
d
ap
p
r
o
ac
h
,
th
is
s
tu
d
y
u
s
ed
t
h
e
Mo
b
ileNetV2
ar
ch
itectu
r
e
a
n
d
ap
p
lied
th
e
Gr
ad
-
C
AM
in
ter
p
r
etatio
n
m
et
h
o
d
to
d
ir
ec
tly
v
is
u
alize
th
e
p
o
r
tio
n
o
f
th
e
im
ag
e
r
esp
o
n
s
ib
le
f
o
r
a
p
ar
ticu
lar
d
ec
is
io
n
.
E
x
ten
s
iv
e
im
ag
e
au
g
m
en
tatio
n
wa
s
u
s
ed
to
im
p
r
o
v
e
g
en
er
aliza
tio
n
to
f
ield
c
o
n
d
iti
o
n
s
.
T
h
is
co
m
p
ar
is
o
n
ev
al
u
ate
s
b
o
th
p
r
e
d
ictiv
e
p
e
r
f
o
r
m
an
ce
an
d
in
ter
p
r
etab
ilit
y
f
id
elity
,
p
r
o
v
i
d
in
g
in
s
ig
h
ts
in
to
wh
ich
X
AI
ap
p
r
o
ac
h
es
b
e
s
t
alig
n
with
d
o
m
ain
k
n
o
wled
g
e
an
d
s
u
p
p
o
r
tin
g
m
o
r
e
tr
an
s
p
a
r
en
t,
r
eliab
le
AI
f
o
r
p
r
ec
is
io
n
a
g
r
icu
ltu
r
e
.
T
h
e
o
b
jectiv
e
is
to
p
r
o
v
id
e
a
f
r
am
ewo
r
k
f
o
r
q
u
a
n
titativ
e
an
aly
s
is
o
f
L
I
ME
an
d
Gr
ad
-
C
AM
ex
p
lan
atio
n
m
eth
o
d
s
f
o
r
th
e
ag
r
icu
ltu
r
al
d
o
m
ain
.
T
h
e
ar
c
h
itectu
r
e
s
u
p
p
o
r
ts
p
r
ac
tical
m
o
b
ile
d
ep
lo
y
m
en
t
th
r
o
u
g
h
lo
w
in
f
e
r
en
ce
c
o
s
t
an
d
ef
f
icien
t
p
r
o
ce
s
s
in
g
.
T
h
e
r
elief
-
b
ased
ANN
e
n
ab
les
r
ap
id
o
n
-
d
ev
ice
co
m
p
u
tatio
n
,
wh
ile
C
NN
m
o
d
els
p
r
o
v
id
e
h
ea
tm
ap
-
b
ase
d
v
is
u
al
e
x
p
lan
atio
n
s
r
en
d
er
ed
o
n
s
m
ar
tp
h
o
n
e
s
cr
ee
n
s
.
Far
m
er
s
ca
n
ca
p
tu
r
e
leaf
im
ag
es
v
ia
m
o
b
ile
ca
m
er
as
to
o
b
tain
in
s
tan
t
d
is
ea
s
e
p
r
ed
ictio
n
s
an
d
v
is
u
alize
af
f
ec
ted
r
eg
io
n
s
,
ev
e
n
o
f
f
lin
e.
I
t
en
ab
les
f
ar
m
e
r
s
to
in
ter
p
r
et
m
o
d
el
r
ea
s
o
n
in
g
an
d
o
b
tain
ac
tio
n
ab
le
in
s
ig
h
ts
d
ir
ec
tly
i
n
th
e
f
ield
.
T
h
is
tr
an
s
p
ar
en
c
y
e
n
h
an
ce
s
u
s
er
tr
u
s
t,
s
u
p
p
o
r
ts
l
o
ca
lized
d
is
ea
s
e
d
iag
n
o
s
is
,
an
d
p
r
o
m
o
tes wid
er
a
d
o
p
tio
n
with
in
s
m
ar
t a
g
r
icu
ltu
r
e
ec
o
s
y
s
tem
s
.
T
h
e
m
ain
co
n
tr
ib
u
tio
n
s
o
f
t
h
is
wo
r
k
in
clu
d
e:
i)
d
ev
elo
p
m
e
n
t
o
f
two
c
o
m
p
le
m
en
tar
y
ex
p
lain
ab
le
m
o
d
els:
a
r
elief
-
b
ased
ANN
in
ter
p
r
et
ed
v
ia
L
I
ME
,
a
n
d
a
Mo
b
ileNetV2
m
o
d
el
an
al
y
ze
d
u
s
in
g
Gr
ad
-
C
AM
an
d
ii)
q
u
a
n
titativ
e
ass
ess
m
en
t
o
f
ex
p
lan
atio
n
q
u
ality
b
ased
o
n
f
id
elity
a
n
d
r
o
b
u
s
tn
ess
m
etr
ics.
T
h
e
r
em
ain
d
er
o
f
th
is
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
ws
.
S
ec
tio
n
2
d
escr
ib
es
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
o
lo
g
y
.
Sectio
n
3
p
r
esen
ts
th
e
r
esu
lts
an
d
d
is
cu
s
s
io
n
.
Fin
ally
,
s
ec
tio
n
4
co
n
clu
d
es
th
e
s
tu
d
y
with
in
s
ig
h
ts
o
n
f
u
tu
r
e
r
esear
ch
d
ir
ec
tio
n
s
.
2.
M
E
T
H
O
D
2
.
1
.
Da
t
a
s
et
des
cr
iptio
n
T
o
ad
d
r
ess
ex
is
tin
g
g
ap
s
in
XAI
f
o
r
p
r
ec
is
io
n
ag
r
icu
ltu
r
e,
th
is
s
tu
d
y
p
r
esen
ts
a
c
o
m
p
ar
ativ
e
an
aly
s
is
o
f
two
in
teg
r
ated
f
r
a
m
ewo
r
k
s
th
at
co
m
b
i
n
e
m
o
d
el
ar
ch
itectu
r
es
with
in
ter
p
r
etatio
n
m
eth
o
d
o
l
o
g
ies.
T
h
e
co
m
p
ar
is
o
n
aim
s
to
e
v
alu
ate
th
e
r
elatio
n
s
h
ip
b
etw
ee
n
p
r
e
d
ictiv
e
p
e
r
f
o
r
m
an
ce
an
d
in
ter
p
r
etab
ilit
y
f
id
elity
.
P
o
tato
im
ag
es
f
r
o
m
Plan
tVillag
e
d
ataset
wer
e
u
s
ed
.
T
h
e
s
elec
ted
d
ataset
co
n
tain
s
a
to
tal
o
f
2
,
1
5
2
im
ag
es
b
elo
n
g
in
g
to
t
h
r
ee
class
es
.
T
h
e
f
ir
s
t
clas
s
co
n
s
is
t
s
o
f
1
5
2
h
ea
lth
y
leav
es,
w
h
ile
th
e
s
ec
o
n
d
an
d
th
ir
d
class
es c
o
n
s
is
t o
f
1
,
0
0
0
i
m
ag
es o
f
ea
r
ly
b
lig
h
t a
n
d
1
,
0
0
0
im
ag
es o
f
late
b
lig
h
t,
r
esp
ec
tiv
ely
.
T
h
e
o
r
i
g
in
al
d
ataset
co
n
tain
s
im
ag
es
wit
h
a
r
eso
lu
tio
n
o
f
2
5
6
×
2
5
6
p
ix
el
s
,
wh
ich
wer
e
r
esized
to
2
2
4
×
2
2
4
p
i
x
els
.
I
n
th
is
s
tu
d
y
,
8
0
% o
f
th
e
im
a
g
es we
r
e
u
s
ed
f
o
r
tr
ain
in
g
,
wh
ile
th
e
r
em
ain
in
g
2
0
% we
r
e
u
s
ed
f
o
r
t
esti
n
g
.
2
.
2
.
Da
t
a
a
ug
m
ent
a
t
i
o
n
T
h
is
s
tu
d
y
em
p
lo
y
ed
im
a
g
e
au
g
m
en
tatio
n
tech
n
iq
u
es
to
i
m
p
r
o
v
e
th
e
g
en
er
aliza
tio
n
a
b
ilit
y
an
d
r
o
b
u
s
tn
ess
o
f
th
e
C
NN
m
o
d
el
in
th
e
f
ield
[
2
6
]
.
Du
r
in
g
i
m
ag
e
au
g
m
en
tatio
n
,
th
e
d
ata
s
et
is
s
y
n
th
etica
lly
ex
p
an
d
e
d
b
y
ap
p
l
y
in
g
r
an
d
o
m
tr
an
s
f
o
r
m
atio
n
s
.
Fo
r
th
is
p
u
r
p
o
s
e,
a
r
o
tatio
n
r
an
g
e
o
f
±
3
0
d
eg
r
ee
s
,
r
an
d
o
m
h
o
r
izo
n
tal
f
lip
p
in
g
,
a
n
d
b
r
ig
h
tn
ess
ad
ju
s
tm
en
t
in
t
h
e
r
a
n
g
e
o
f
0
.
7
t
o
1
.
3
,
a
zo
o
m
v
ar
iatio
n
o
f
u
p
to
2
0
%
,
an
d
Gau
s
s
ian
n
o
is
e
with
a
s
m
all
v
ar
ian
ce
wer
e
ap
p
lied
.
T
h
ese
a
u
g
m
en
tatio
n
s
r
ep
licate
n
atu
r
al
v
ar
iab
ilit
y
in
f
ield
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
I
n
terp
r
etin
g
p
o
ta
to
d
is
ea
s
e
cla
s
s
ifica
tio
n
u
s
in
g
ex
p
la
in
a
b
le
a
r
tifi
cia
l in
tellig
en
ce
(
R
a
ke
s
h
K
u
ma
r
Gu
ma
s
ta
)
1125
co
n
d
itio
n
s
s
u
c
h
as
ch
an
g
es
in
lig
h
tin
g
,
ca
m
e
r
a
a
n
g
les,
a
n
d
leaf
ap
p
ea
r
an
ce
.
Gau
s
s
ian
n
o
is
e
s
im
u
lates
s
en
s
o
r
im
p
er
f
ec
tio
n
s
an
d
m
in
o
r
e
n
v
ir
o
n
m
e
n
tal
ar
tifa
cts,
en
co
u
r
ag
in
g
th
e
m
o
d
el
to
lear
n
m
o
r
e
r
o
b
u
s
t
an
d
n
o
is
e
-
to
ler
an
t r
ep
r
esen
tatio
n
s
.
T
h
e
au
g
m
en
ted
d
ataset
was s
u
b
s
eq
u
en
tly
u
s
ed
to
tr
ain
th
e
Mo
b
ileNetV2
-
b
ased
C
NN
m
o
d
el.
2
.
3
.
Appro
a
ch
A
:
m
a
nu
a
l f
e
a
t
ure
eng
ineering
+
re
lief
+
a
rt
if
icia
l neura
l net
wo
rk
T
h
e
f
ir
s
t
ap
p
r
o
ac
h
co
n
s
is
ts
o
f
f
o
u
r
s
tag
es
.
T
h
e
f
ir
s
t
s
tag
e
is
f
ea
tu
r
e
ex
tr
ac
tio
n
,
th
e
s
ec
o
n
d
is
f
ea
tu
r
e
s
elec
tio
n
,
th
e
th
ir
d
is
tr
ain
in
g
an
d
test
in
g
o
f
th
e
class
if
icatio
n
m
o
d
el,
an
d
th
e
f
o
u
r
th
is
m
o
d
el
in
ter
p
r
etatio
n
.
I
n
f
ea
tu
r
e
ex
tr
a
ctio
n
s
tag
e
,
th
is
s
tu
d
y
ex
tr
ac
t
ed
a
co
m
p
r
e
h
en
s
iv
e
s
et
o
f
h
an
d
cr
af
te
d
f
ea
tu
r
es
f
r
o
m
ea
ch
leaf
im
ag
e
to
ca
p
tu
r
e
co
lo
r
,
tex
t
u
r
e,
an
d
s
tr
u
ctu
r
al
in
f
o
r
m
ati
o
n
.
C
o
lo
r
ch
ar
ac
ter
is
tics
ar
e
q
u
an
tifie
d
u
s
in
g
h
is
to
g
r
am
s
o
f
th
e
r
ed
,
g
r
ee
n
,
an
d
b
lu
e
ch
an
n
els,
ea
ch
co
m
p
u
ted
with
3
2
b
in
s
to
ca
p
tu
r
e
th
e
d
is
tr
ib
u
tio
n
o
f
in
ten
s
ities
.
T
ex
tu
r
e
f
ea
tu
r
es
,
i
n
clu
d
in
g
co
n
tr
ast,
h
o
m
o
g
e
n
eity
,
d
is
s
im
ilar
ity
,
an
g
u
lar
s
ec
o
n
d
m
o
m
en
t
(
ASM)
,
an
d
c
o
r
r
elatio
n
,
ar
e
ex
t
r
ac
ted
,
wh
ich
d
escr
ib
e
s
p
atial
r
elatio
n
s
h
ip
s
o
f
p
i
x
el
in
ten
s
ities
.
I
n
s
tatis
tical
d
escr
ip
to
r
s
,
th
e
m
ea
n
a
n
d
s
t
an
d
ar
d
d
ev
iatio
n
o
f
ea
ch
c
o
lo
r
ch
an
n
el
ar
e
ca
lcu
lated
t
o
s
u
m
m
ar
ize
co
lo
r
v
ar
iatio
n
s
.
T
h
e
en
t
r
o
p
y
o
f
t
h
e
g
r
ay
s
ca
le
im
ag
e
is
esti
m
ated
to
m
ea
s
u
r
e
th
e
co
m
p
lex
ity
o
f
in
ten
s
ity
p
atter
n
s
.
His
to
g
r
am
o
f
o
r
ien
te
d
g
r
ad
ie
n
ts
(
HOG)
f
ea
tu
r
es
ar
e
ex
tr
ac
ted
to
ch
ar
ac
ter
ize
s
h
ap
e
an
d
lo
ca
l
g
r
ad
ien
ts
,
u
s
in
g
eig
h
t
o
r
ien
tatio
n
s
an
d
3
2
×
32
-
p
ix
el
ce
lls
,
p
r
o
v
id
in
g
a
c
o
m
p
ac
t
r
ep
r
esen
tatio
n
o
f
ed
g
e
in
f
o
r
m
atio
n
.
T
h
ese
d
iv
er
s
e
f
ea
tu
r
es
a
r
e
c
o
n
ca
ten
ated
in
to
a
s
in
g
le
v
ec
to
r
t
h
at
s
er
v
es
as
in
p
u
t
f
o
r
s
u
b
s
eq
u
en
t
ML
class
if
ier
s
.
I
n
th
i
s
way
,
a
to
tal
o
f
5
0
1
f
ea
tu
r
es
wer
e
ex
tr
ac
ted
f
r
o
m
ea
ch
im
ag
e
.
T
h
e
n
am
es
o
f
th
ese
o
r
ig
in
al
f
ea
tu
r
es
ar
e
s
to
r
ed
in
a
lis
t
ca
lled
“f
ea
tu
r
e_
n
am
es”.
I
n
f
ea
tu
r
e
s
elec
tio
n
s
tag
e,
th
e
r
elief
alg
o
r
ith
m
was
ap
p
lied
to
id
en
tify
th
e
1
2
8
m
o
s
t
r
elev
an
t
f
ea
tu
r
es
f
o
r
cla
s
s
if
icatio
n
.
Her
e
,
th
e
p
u
r
p
o
s
e
o
f
u
s
in
g
R
elief
is
to
id
en
tify
f
ea
tu
r
es
th
at
g
lo
b
ally
h
el
p
d
is
tin
g
u
is
h
b
etw
ee
n
class
es
an
d
im
p
r
o
v
e
m
o
d
e
l
class
if
icatio
n
ac
cu
r
ac
y
.
I
t
is
also
u
s
ed
to
r
ed
u
ce
th
e
d
im
en
s
io
n
ality
o
r
co
m
p
le
x
ity
o
f
th
e
m
o
d
el
.
T
h
e
in
d
ic
es
o
f
th
ese
t
op
-
r
an
k
ed
f
ea
tu
r
es,
b
ased
o
n
th
eir
im
p
o
r
tan
ce
s
co
r
es,
ar
e
s
to
r
ed
in
th
e
ar
r
ay
“selecte
d
_
in
d
ices”.
T
h
ese
in
d
ices
in
d
icate
t
h
e
p
o
s
itio
n
s
o
f
th
e
s
elec
ted
f
ea
tu
r
es in
th
e
o
r
i
g
in
a
l f
ea
tu
r
e
s
p
ac
e.
Fo
r
th
e
class
if
icatio
n
m
o
d
el,
an
ANN
is
d
esig
n
ed
u
s
in
g
a
s
eq
u
en
ti
al
ar
ch
itectu
r
e
co
m
p
r
is
in
g
f
o
u
r
h
id
d
en
lay
er
s
.
T
h
e
h
i
d
d
en
la
y
er
s
in
clu
d
ed
2
5
6
,
1
2
8
,
6
4
,
an
d
3
2
n
e
u
r
o
n
s
ea
ch
with
r
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U
)
ac
tiv
atio
n
,
f
o
llo
wed
b
y
b
atch
n
o
r
m
aliza
tio
n
an
d
a
d
r
o
p
o
u
t
lay
e
r
with
d
if
f
e
r
en
t
r
ate
to
r
ed
u
ce
o
v
er
f
itti
n
g
.
T
h
e
o
u
t
p
u
t
lay
er
em
p
lo
y
ed
a
s
o
f
tm
ax
ac
tiv
atio
n
f
u
n
ctio
n
with
a
n
u
m
b
er
o
f
n
eu
r
o
n
s
eq
u
al
to
th
e
n
u
m
b
er
o
f
tar
g
et
class
es
to
p
r
o
d
u
ce
class
p
r
o
b
ab
ilit
ies.
A
b
atch
s
ize
o
f
3
2
is
s
elec
ted
,
an
d
th
e
Ad
am
o
p
tim
izer
is
u
s
ed
to
tr
ain
th
e
ANN
m
o
d
el
with
a
lear
n
in
g
r
ate
o
f
0
.
001
.
T
o
en
h
a
n
ce
in
ter
p
r
etab
ilit
y
,
th
is
s
tu
d
y
u
s
e
d
L
I
ME
f
r
am
ewo
r
k
in
tab
u
lar
m
o
d
e
to
in
te
r
p
r
et
th
e
AN
N
m
o
d
el
p
r
ed
ictio
n
s
.
L
I
ME
h
ig
h
lig
h
ts
th
e
to
p
2
0
m
o
s
t
in
f
lu
en
tial
f
ea
tu
r
es
in
th
e
m
o
d
el’
s
d
ec
is
io
n
.
Ho
wev
er
,
L
I
ME
p
r
o
v
id
es
th
e
ex
p
la
n
atio
n
in
ter
m
s
o
f
co
lu
m
n
in
d
ices
(
lim
e_
co
lu
m
n
_
in
d
ex
)
c
o
r
r
esp
o
n
d
in
g
to
th
e
r
ed
u
ce
d
f
ea
tu
r
e
s
et
o
f
r
elief
.
T
o
f
in
d
o
u
t
wh
ic
h
o
r
ig
in
al
f
ea
t
u
r
es
th
ese
in
d
ice
s
r
ef
er
to
,
ea
ch
lim
e
_
co
lu
m
n
_
in
d
ex
is
r
em
ap
p
e
d
to
its
o
r
i
g
in
al
f
ea
tu
r
e
_
in
d
e
x
u
s
in
g
s
elec
ted
_
in
d
ices
as
s
h
o
wn
in
(
1
)
.
T
h
e
ac
tu
al
f
ea
tu
r
e
n
am
e
is
th
en
r
etr
iev
ed
u
s
in
g
th
is
in
d
ex
f
r
o
m
th
e
f
ea
tu
r
e_
n
a
m
es lis
t
as sh
o
wn
in
(
2
)
.
_
=
[
_
_
]
(
1
)
_
=
_
[
_
]
(
2
)
T
h
is
m
ap
p
in
g
e
n
s
u
r
es
th
at
th
e
ex
p
lan
atio
n
s
p
r
o
d
u
ce
d
b
y
L
I
ME
ar
e
in
ter
p
r
etab
le
in
ter
m
s
o
f
th
e
o
r
ig
i
n
al
im
ag
e
-
d
er
iv
e
d
f
ea
tu
r
es,
ev
en
t
h
o
u
g
h
th
e
m
o
d
el
o
p
er
ates o
n
a
r
ed
u
ce
d
f
ea
tu
r
e
s
u
b
s
et
af
ter
r
elief
s
elec
tio
n
.
2
.
4
.
Appro
a
ch
B
:
co
nv
o
lutio
na
l neura
l net
wo
rk
+
g
ra
dient
-
weig
hte
d c
la
s
s
a
ct
iv
a
t
io
n
m
a
pp
ing
I
n
th
e
s
ec
o
n
d
ap
p
r
o
ac
h
,
two
s
tag
es
ar
e
u
s
ed
:
th
e
f
ir
s
t
is
C
NN
m
o
d
el
tr
ain
in
g
,
f
o
llo
wed
b
y
th
e
Gr
ad
-
C
AM
in
ter
p
r
etatio
n
m
e
th
o
d
.
T
r
ain
in
g
d
ee
p
lear
n
in
g
m
o
d
els
g
en
er
ally
r
eq
u
ir
es
a
lar
g
e
a
m
o
u
n
t
o
f
tr
ain
in
g
d
ata
.
Sin
ce
o
u
r
d
ataset
d
o
es
n
o
t
h
a
v
e
lar
g
e
n
u
m
b
er
o
f
tr
ain
in
g
im
a
g
es,
a
tr
a
n
s
f
er
l
ea
r
n
in
g
ap
p
r
o
ac
h
is
em
p
lo
y
ed
,
lev
er
a
g
in
g
t
h
e
M
o
b
ileNetV2
ar
ch
itectu
r
e
as
th
e
b
ase
m
o
d
el
.
T
h
e
Mo
b
ile
NetV2
m
o
d
el
was
in
itialized
with
weig
h
ts
lear
n
e
d
o
n
I
m
ag
eNe
t
a
n
d
co
n
f
ig
u
r
e
d
with
i
n
clu
d
e_
t
o
p
=
Fals
e
to
r
em
o
v
e
its
o
r
ig
i
n
al
class
if
icatio
n
h
ea
d
.
Du
r
in
g
th
e
f
ir
s
t
s
tag
e
o
f
tr
ain
i
n
g
,
o
n
ly
t
h
e
to
p
lay
er
s
wer
e
r
etr
ain
e
d
,
wh
ile
th
e
p
r
etr
ain
ed
Mo
b
ileNetV2
la
y
er
s
r
em
ain
e
d
f
r
o
ze
n
.
T
h
is
s
tep
en
ab
led
th
e
m
o
d
el
t
o
ad
a
p
t
its
h
ig
h
-
le
v
e
l
r
ep
r
esen
tatio
n
s
to
th
e
tar
g
et
d
ataset
with
o
u
t
d
is
r
u
p
tin
g
t
h
e
p
r
et
r
ain
ed
c
o
n
v
o
lu
tio
n
al
f
ilter
s
.
T
h
e
m
o
d
el
was
co
m
p
iled
u
s
in
g
th
e
Ad
am
o
p
tim
izer
an
d
th
e
s
p
a
r
s
e
ca
teg
o
r
ical
cr
o
s
s
-
en
tr
o
p
y
lo
s
s
f
u
n
ctio
n
,
an
d
tr
ain
ed
t
h
e
m
o
d
el
f
o
r
eig
h
t
ep
o
ch
s
u
s
in
g
b
o
th
tr
ain
in
g
a
n
d
v
alid
atio
n
s
ets.
Fo
r
f
i
n
e
-
tu
n
in
g
,
th
e
b
ase
m
o
d
el
Mo
b
ileNetV2
was
u
n
f
r
o
ze
n
b
y
s
ettin
g
b
ase_
m
o
d
el.
T
r
ain
a
b
le
=
tr
u
e
,
en
a
b
lin
g
all
c
o
n
v
o
lu
tio
n
al
lay
er
s
to
b
e
u
p
d
ate
d
d
u
r
in
g
tr
ain
i
n
g
.
T
h
e
m
o
d
el
was
th
en
r
etr
ain
ed
f
o
r
an
ad
d
itio
n
al
s
ev
en
ep
o
ch
s
,
s
tar
tin
g
f
r
o
m
th
e
last
co
m
p
leted
ep
o
ch
o
f
t
h
e
in
itial
tr
ain
in
g
.
T
h
is
p
r
o
ce
s
s
allo
ws t
h
e
f
ea
tu
r
e
ex
tr
ac
to
r
to
b
e
r
ef
in
ed
jo
in
tly
with
th
e
class
if
ier
,
l
ea
d
in
g
to
im
p
r
o
v
e
d
d
is
cr
im
in
ativ
e
p
er
f
o
r
m
an
ce
.
Ov
er
all,
th
is
two
-
s
tag
e
tr
ain
i
n
g
s
tr
ateg
y
,
co
n
s
is
tin
g
o
f
in
itial
tr
an
s
f
er
lear
n
in
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
2
0
2
6
:
1
1
2
3
-
1
1
3
0
1126
with
f
r
o
ze
n
weig
h
ts
,
f
o
llo
wed
b
y
en
d
-
to
-
en
d
f
i
n
e
-
tu
n
i
n
g
,
b
a
lan
ce
s
th
e
b
en
ef
its
o
f
lev
er
ag
i
n
g
p
r
io
r
k
n
o
wled
g
e
f
r
o
m
lar
g
e
-
s
ca
le
d
atasets
wi
th
th
e
ab
ilit
y
to
s
p
ec
ialize
t
h
e
m
o
d
el
to
th
e
ch
ar
ac
ter
is
tics
o
f
th
e
tar
g
et
d
o
m
ain
.
Gr
ad
-
C
AM
is
ap
p
lied
to
v
is
u
a
lize
th
e
p
o
r
tio
n
o
f
t
h
e
im
ag
e
r
esp
o
n
s
ib
le
f
o
r
a
p
ar
ticu
lar
d
ec
is
io
n
.
2
.
5
.
E
v
a
lua
t
i
o
n m
et
rics
B
o
th
ANN
an
d
Mo
b
ileNetV2
m
o
d
els
wer
e
ev
alu
ate
d
u
s
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
as d
ef
in
ed
in
(
3
)
t
o
(
6
)
.
=
+
+
+
+
(
3
)
=
+
(
4
)
=
+
(
5
)
1
−
=
2
×
×
+
(
6
)
W
h
er
e
is
tr
u
e
p
o
s
itiv
es,
is
f
alse p
o
s
itiv
es,
is
tr
u
e
n
eg
ativ
es,
an
d
is
f
alse n
eg
ativ
es.
Acc
u
r
ac
y
r
ef
lects
o
v
er
all
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
an
d
m
e
asu
r
es
h
o
w
well
t
h
e
m
o
d
el
d
is
tin
g
u
is
h
es
b
etwe
en
h
ea
lth
y
an
d
d
is
ea
s
ed
leav
es
o
r
id
en
tifie
s
s
p
ec
if
ic
d
is
ea
s
es.
Pre
ci
s
io
n
m
ea
s
u
r
es
th
e
co
r
r
ec
tn
ess
o
f
p
o
s
itiv
e
p
r
ed
ictio
n
s
,
r
ec
all
m
ea
s
u
r
es
th
e
d
etec
tio
n
o
f
d
is
ea
s
ed
s
am
p
les,
an
d
F1
-
s
co
r
e
b
alan
ce
s
b
o
th
[
2
7
]
.
E
x
p
lan
atio
n
m
eth
o
d
s
wer
e
s
y
s
tem
atica
lly
ev
alu
ated
u
s
in
g
f
id
elity
an
d
r
o
b
u
s
tn
ess
.
Fid
elity
m
ea
s
u
r
es
h
o
w
well
an
ex
p
lan
atio
n
r
ef
lects
th
e
m
o
d
el'
s
d
ec
is
io
n
-
m
ak
in
g
p
r
o
ce
s
s
,
wh
ile
th
e
f
id
elity
d
r
o
p
q
u
an
tifie
s
th
e
d
ec
r
ea
s
e
in
m
o
d
el
co
n
f
id
en
ce
o
r
ac
cu
r
ac
y
wh
en
im
p
o
r
tan
t f
e
atu
r
es id
en
tifie
d
b
y
th
e
ex
p
lan
atio
n
ar
e
r
em
o
v
ed
.
R
o
b
u
s
tn
ess
as
s
es
s
es
th
e
co
n
s
is
ten
cy
an
d
r
eliab
ilit
y
o
f
ex
p
lan
atio
n
s
u
n
d
er
v
ar
y
in
g
i
n
p
u
t
co
n
d
itio
n
s
.
I
n
ag
r
icu
ltu
r
al
d
is
ea
s
e
d
etec
tio
n
,
r
o
b
u
s
t
m
eth
o
d
s
s
h
o
u
ld
m
ai
n
tain
m
ea
n
in
g
f
u
l
ex
p
la
n
atio
n
s
ev
en
f
o
r
n
o
is
y
o
r
d
eg
r
ad
e
d
im
ag
es,
h
el
p
in
g
u
s
er
s
id
en
tify
d
is
ea
s
e
-
af
f
ec
ted
r
e
g
i
o
n
s
.
Hy
p
er
p
ar
a
m
eter
s
ettin
g
s
f
o
r
c
alcu
latin
g
f
id
elity
an
d
r
o
b
u
s
tn
ess
o
f
b
o
th
m
eth
o
d
s
ar
e
as
f
o
l
lo
ws.
Fo
r
L
I
ME
(
o
n
ANN
m
o
d
el)
:
n
u
m
b
er
o
f
p
er
tu
r
b
ed
s
am
p
les
=
5
,
0
0
0
,
Ker
n
el
wid
t
h
=
0
.
2
5
,
n
u
m
b
er
o
f
t
o
p
f
ea
tu
r
es
r
em
o
v
ed
=
5
,
d
is
tan
ce
m
etr
ic
=
E
u
clid
ea
n
,
p
er
tu
r
b
atio
n
ty
p
e
=
Gau
s
s
ian
n
o
is
e
with
σ
=
0
.
0
2
.
Fo
r
Gr
a
d
-
C
AM
(
o
n
C
NN
m
o
d
el)
:
tar
g
et
lay
e
r
=
last
co
n
v
o
lu
tio
n
al
b
lo
c
k
o
f
Mo
b
ileNetV2
,
g
r
ad
ien
t
n
o
r
m
aliza
tio
n
=
tr
u
e,
h
ea
tm
ap
r
eso
lu
tio
n
=
2
2
4
×
2
2
4
p
ix
els,
an
d
p
er
tu
r
b
at
io
n
s
tep
f
o
r
th
e
r
o
b
u
s
tn
ess
test
=
5
% p
ix
el
n
o
is
e
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
er
f
o
r
m
an
ce
an
d
i
n
ter
p
r
e
tab
ilit
y
o
f
th
e
p
r
o
p
o
s
ed
p
o
tat
o
leaf
d
is
ea
s
e
class
if
icatio
n
f
r
am
ewo
r
k
wer
e
ev
alu
ated
u
s
in
g
b
o
t
h
tr
a
d
itio
n
al
ANN
an
d
C
NN
ar
ch
i
tectu
r
es.
T
h
e
C
NN
m
o
d
el
cle
ar
ly
o
u
tp
er
f
o
r
m
e
d
th
e
ANN
ac
r
o
s
s
all
ev
alu
atio
n
m
ea
s
u
r
es.
Sp
ec
i
f
ically
,
t
h
e
C
NN
m
o
d
el
b
ased
o
n
Mo
b
ileNetV2
with
f
in
e
-
tu
n
in
g
o
n
th
e
tar
g
et
d
ataset
ac
h
iev
ed
a
test
ac
cu
r
ac
y
o
f
9
9
.
7
6
%,
alo
n
g
with
a
p
r
ec
is
io
n
o
f
9
9
.
8
2
%,
r
ec
all
o
f
9
9
.
8
2
%,
a
n
d
an
F1
-
s
co
r
e
o
f
9
9
.
8
2
%.
I
n
co
m
p
ar
is
o
n
,
th
e
ANN
m
o
d
el
r
ea
ch
e
d
a
lo
w
er
test
ac
cu
r
ac
y
o
f
9
9
.
0
7
%,
with
co
r
r
esp
o
n
d
in
g
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
v
alu
es
o
f
9
9
.
0
8
%,
9
9
.
0
7
%,
an
d
9
9
.
0
7
%,
r
esp
ec
tiv
ely
.
T
h
ese
r
esu
lts
clea
r
ly
d
em
o
n
s
tr
ate
th
at
th
e
C
NN’
s
ab
ilit
y
to
lear
n
h
ier
ar
ch
ical
f
ea
t
u
r
e
r
ep
r
esen
tatio
n
s
d
ir
ec
tly
f
r
o
m
im
ag
es
r
esu
lted
in
h
ig
h
er
d
is
cr
im
in
ativ
e
p
er
f
o
r
m
a
n
ce
an
d
b
etter
o
v
e
r
all
g
en
er
aliza
tio
n
.
W
h
ile
th
e
AN
N
co
m
b
in
e
d
with
r
elief
-
s
elec
ted
f
ea
tu
r
es
o
f
f
e
r
ed
r
ea
s
o
n
ab
ly
g
o
o
d
ac
c
u
r
ac
y
,
its
r
elian
ce
o
n
m
a
n
u
ally
e
n
g
in
ee
r
ed
in
p
u
ts
lim
ited
its
ef
f
ec
tiv
en
ess
r
elativ
e
to
th
e
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
.
Fo
r
in
ter
p
r
etab
ilit
y
,
L
I
ME
w
as
ap
p
lied
to
th
e
ANN
m
o
d
el,
wh
ile
Gr
ad
-
C
AM
was
ap
p
lied
to
th
e
C
NN
m
o
d
el
.
L
I
ME
p
r
o
v
id
ed
f
ea
tu
r
e
-
lev
el
e
x
p
lan
atio
n
s
b
y
h
ig
h
lig
h
tin
g
th
e
t
o
p
2
0
c
o
n
tr
i
b
u
tin
g
h
an
d
c
r
af
ted
f
ea
tu
r
es
r
esp
o
n
s
ib
le
f
o
r
ea
ch
p
r
e
d
ictio
n
.
I
n
an
ex
a
m
p
le
in
s
tan
ce
,
L
I
ME
id
e
n
tifie
d
f
ea
tu
r
es
lik
e
C
H_
B
_
b
in
_
2
2
,
C
H_
B
_
b
in
_
2
1
,
C
H_
R
_
b
in
_
3
2
,
an
d
HOG_
5
9
as
m
o
s
t
in
f
lu
en
tial
in
co
r
r
ec
tly
p
r
ed
ictin
g
th
e
class
“
ea
r
ly
b
lig
h
t”.
T
o
q
u
an
ti
tativ
ely
ass
es
s
th
e
q
u
ality
o
f
th
ese
ex
p
lan
atio
n
s
,
f
id
elity
an
d
r
o
b
u
s
tn
ess
s
co
r
es
wer
e
co
m
p
u
te
d
.
T
h
e
a
v
er
a
g
e
f
id
elity
d
r
o
p
f
o
r
L
I
ME
o
n
3
0
t
est
im
ag
es
was
1
0
.
9
0
%,
in
d
ic
atin
g
th
at
r
em
o
v
in
g
th
e
to
p
5
f
ea
tu
r
es
f
r
o
m
th
e
in
p
u
t
r
ed
u
ce
d
th
e
m
o
d
el'
s
p
r
ed
ictio
n
co
n
f
id
e
n
ce
,
th
u
s
v
alid
atin
g
th
eir
im
p
o
r
tan
ce
.
T
h
e
r
o
b
u
s
tn
ess
o
f
L
I
ME
ex
p
lan
atio
n
s
,
co
m
p
u
te
d
b
ased
o
n
th
e
c
o
n
s
is
ten
cy
o
f
to
p
5
f
ea
t
u
r
es
ac
r
o
s
s
s
im
ilar
im
ag
es
,
was 8
4
.
2
7
%,
r
e
f
lectin
g
s
tab
le
ex
p
lan
atio
n
s
.
Fig
u
r
e
1
s
h
o
ws
th
e
i
n
te
r
p
r
eta
b
ilit
y
co
m
p
a
r
is
o
n
o
f
C
NN
an
d
ANN
m
o
d
els
u
s
in
g
Gr
a
d
-
C
AM
an
d
L
I
ME
ex
p
la
n
atio
n
s
f
o
r
p
o
tato
leaf
d
is
ea
s
e
class
if
icatio
n
.
Fig
u
r
e
1
(
a
)
s
h
o
ws
th
e
o
r
ig
in
al
t
est
im
ag
es
with
tr
u
e
class
lab
el
s
(
ea
r
ly
b
lig
h
t,
late
b
lig
h
t,
an
d
h
ea
lth
y
)
.
Fig
u
r
e
1
(
b
)
p
r
esen
ts
Gr
ad
-
C
AM
v
is
u
aliza
tio
n
s
h
ig
h
lig
h
tin
g
s
p
atial
r
eg
io
n
s
t
h
at
m
o
s
t
in
f
lu
en
ce
d
th
e
C
NN’
s
p
r
ed
ictio
n
s
.
Fig
u
r
e
1
(
c)
d
e
p
icts
L
I
ME
f
ea
tu
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
I
n
terp
r
etin
g
p
o
ta
to
d
is
ea
s
e
cla
s
s
ifica
tio
n
u
s
in
g
ex
p
la
in
a
b
le
a
r
tifi
cia
l in
tellig
en
ce
(
R
a
ke
s
h
K
u
ma
r
Gu
ma
s
ta
)
1127
co
n
tr
ib
u
tio
n
p
l
o
ts
f
o
r
th
e
AN
N
m
o
d
el
,
s
h
o
win
g
t
h
e
r
elativ
e
im
p
o
r
ta
n
ce
an
d
p
o
lar
ity
o
f
f
e
atu
r
es
d
r
iv
in
g
ea
ch
class
if
icatio
n
d
ec
is
io
n
.
T
o
g
e
th
er
,
th
ese
r
esu
lts
illu
s
tr
ate
co
m
p
lem
en
tar
y
in
s
ig
h
ts
:
Gr
ad
-
C
AM
p
r
o
v
id
es
r
eg
io
n
-
le
v
el
v
is
u
al
f
o
c
u
s
,
wh
ile
L
I
ME
r
ev
ea
ls
f
ea
tu
r
e
-
le
v
el
attr
ib
u
tio
n
co
n
tr
i
b
u
tin
g
to
d
is
ea
s
e
class
id
en
tific
atio
n
.
Gr
ad
-
C
AM
p
r
o
v
id
es
s
p
atial
ex
p
lan
atio
n
s
d
ir
ec
tly
o
n
th
e
in
p
u
t
im
ag
e
b
y
g
en
er
atin
g
h
ea
tm
a
p
s
th
at
lo
ca
lized
th
e
m
o
s
t
d
is
cr
im
in
ativ
e
r
eg
io
n
s
.
T
h
e
h
ig
h
lig
h
ted
ar
ea
(
Fig
u
r
e
1
)
c
o
r
r
esp
o
n
d
s
clo
s
ely
to
t
h
e
d
is
ea
s
e
-
af
f
ec
ted
r
eg
i
o
n
o
f
t
h
e
leaf
,
af
f
ir
m
in
g
t
h
e
r
ele
v
an
ce
o
f
th
e
m
o
d
el'
s
atten
tio
n
.
T
h
e
av
er
ag
e
f
id
elity
d
r
o
p
f
o
r
Gr
a
d
-
C
AM
was
o
b
s
er
v
ed
to
b
e
1
8
.
6
8
%,
wh
ich
,
alth
o
u
g
h
h
ig
h
er
th
a
n
L
I
ME
’
s
,
r
ef
lect
s
th
e
m
o
d
el'
s
h
ea
v
y
r
elian
ce
o
n
s
p
atial
p
atter
n
s
.
Gr
ad
-
CA
M
r
o
b
u
s
tn
ess
was
m
ea
s
u
r
ed
at
7
9
.
3
8
%,
in
d
icatin
g
g
en
er
ally
co
n
s
is
ten
t
atten
tio
n
ac
r
o
s
s
s
im
ilar
in
p
u
ts
b
u
t sli
g
h
tly
lo
wer
s
tab
ilit
y
co
m
p
ar
ed
to
L
I
ME
.
(
a)
(
b
)
(
c)
Fig
u
r
e
1
.
I
n
ter
p
r
etab
ilit
y
co
m
p
ar
is
o
n
o
f
C
NN
an
d
ANN
m
o
d
els u
s
in
g
Gr
ad
-
C
AM
an
d
L
I
ME
ex
p
lan
atio
n
s
f
o
r
p
o
tato
leaf
d
is
ea
s
e
class
if
icati
o
n
: (
a)
o
r
ig
in
al
test
im
ag
e
,
(
b
)
Gr
ad
-
C
AM
in
ter
p
r
etatio
n
,
an
d
(
c)
LIME
ex
p
lan
atio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
2
0
2
6
:
1
1
2
3
-
1
1
3
0
1128
T
h
e
ANN
co
m
b
in
ed
with
L
I
ME
p
r
o
d
u
ce
d
e
x
p
lan
atio
n
s
in
th
e
f
o
r
m
o
f
n
u
m
er
ica
l
f
ea
tu
r
e
im
p
o
r
tan
ce
s
,
p
r
o
v
id
in
g
a
r
a
n
k
ed
lis
t
o
f
th
e
m
o
s
t
in
f
lu
e
n
tial
f
ea
tu
r
es
f
o
r
ea
ch
p
r
ed
i
ctio
n
.
I
t
ex
p
licitly
q
u
an
tifie
s
th
e
co
n
tr
ib
u
tio
n
o
f
ea
ch
in
p
u
t
f
ea
tu
r
e
.
I
n
co
n
tr
ast,
th
e
C
NN
with
Gr
ad
-
C
A
M
g
en
er
ates
v
is
u
a
l
h
ea
tm
ap
s
h
ig
h
lig
h
tin
g
s
p
atial
r
eg
io
n
s
wi
th
in
th
e
in
p
u
t
im
ag
es
th
at
m
o
s
t
s
tr
o
n
g
ly
in
f
l
u
en
ce
th
e
m
o
d
el’
s
d
ec
is
io
n
.
I
t
h
elp
s
u
s
to
v
is
u
ally
in
s
p
ec
t
th
e
d
is
ea
s
ed
ar
ea
s
o
f
th
e
leaf
.
T
o
g
et
h
er
,
th
ese
m
eth
o
d
s
p
r
o
v
id
e
a
co
m
p
lem
en
tar
y
u
n
d
er
s
tan
d
i
n
g
o
f
m
o
d
el
d
ec
is
io
n
-
m
ak
in
g
,
f
r
o
m
in
te
r
p
r
etab
le
n
u
m
er
ical
f
e
atu
r
es
to
in
t
u
itiv
e
v
is
u
al
ev
id
en
ce
,
e
n
s
u
r
in
g
t
r
an
s
p
ar
en
cy
an
d
tr
u
s
two
r
th
in
ess
i
n
au
to
m
ated
p
lan
t d
is
ea
s
e
d
ete
ctio
n
s
y
s
tem
s
.
T
h
e
r
esu
lts
o
f
th
is
s
tu
d
y
d
em
o
n
s
tr
ate
th
at
b
o
th
ap
p
r
o
ac
h
es
ac
h
iev
ed
h
ig
h
class
if
icatio
n
p
er
f
o
r
m
an
ce
,
with
th
e
co
m
p
u
tatio
n
ally
ef
f
ic
ien
t
Mo
b
ileNetV2
C
NN
m
o
d
el
r
ea
ch
in
g
an
o
v
e
r
all
ac
cu
r
ac
y
o
f
9
9
.
7
6
%,
co
n
f
ir
m
in
g
th
e
ef
f
ec
tiv
en
ess
o
f
d
ee
p
co
n
v
o
lu
tio
n
al
ar
c
h
itectu
r
es
f
o
r
p
lan
t
d
is
ea
s
e
d
etec
tio
n
[
1
7
]
,
[
2
6
]
.
T
h
is
is
co
n
s
is
ten
t
with
p
r
ev
i
o
u
s
r
e
p
o
r
ts
th
at
f
in
e
-
t
u
n
ed
C
NNs
ca
n
ac
h
iev
e
n
ea
r
-
p
er
f
ec
t
ac
cu
r
ac
y
u
n
d
er
co
n
tr
o
lled
co
n
d
itio
n
s
[
2
2
]
,
[
2
3
]
.
T
h
e
s
tu
d
y
in
te
g
r
ates
a
co
m
p
r
eh
en
s
iv
e
au
g
m
en
tatio
n
p
ip
elin
e
in
clu
d
in
g
f
lip
s
,
r
o
tatio
n
s
,
b
r
i
g
h
tn
ess
an
d
c
o
n
tr
ast
ad
ju
s
tm
en
ts
,
an
d
Gau
s
s
ian
n
o
is
e.
T
h
is
m
u
lti
-
f
ac
eted
s
tr
ateg
y
im
p
r
o
v
es
r
o
b
u
s
tn
ess
to
v
ar
iab
le
f
ield
i
m
ag
in
g
co
n
d
itio
n
s
.
E
ar
lier
a
g
r
icu
ltu
r
al
s
tu
d
ies
f
o
cu
s
ed
s
o
le
ly
o
n
class
if
icatio
n
ac
cu
r
ac
y
with
o
u
t
in
ter
p
r
etab
ilit
y
ass
es
s
m
en
ts
[
5
]
,
[
7
]
,
[
8
]
.
I
n
co
n
tr
ast,
th
e
a
p
p
r
o
ac
h
p
r
o
d
u
ce
s
class
-
d
is
cr
im
in
ativ
e
h
ea
tm
ap
s
.
Ho
wev
er
,
in
ter
p
r
etab
ilit
y
m
etr
ics
r
ev
ea
led
n
u
an
ce
d
d
if
f
er
en
ce
s
b
etwe
en
m
eth
o
d
s
.
L
I
ME
ex
p
lan
atio
n
s
ap
p
lied
t
o
th
e
r
el
ief
-
s
elec
ted
ANN
m
o
d
el
ac
h
iev
ed
a
lo
wer
av
er
a
g
e
f
id
elity
d
r
o
p
o
f
1
0
.
9
0
%
an
d
h
ig
h
er
r
o
b
u
s
tn
ess
o
f
8
4
.
2
7
%,
wh
ile
Gr
ad
-
C
AM
ex
p
lan
atio
n
s
f
o
r
th
e
C
NN
m
o
d
el
e
x
h
ib
ite
d
a
h
ig
h
er
f
id
elity
d
r
o
p
o
f
1
8
.
6
8
%
a
n
d
lo
wer
r
o
b
u
s
tn
ess
o
f
7
9
.
3
8
%.
T
h
ese
f
in
d
in
g
s
p
ar
tially
co
n
tr
ast
with
ea
r
lier
s
tu
d
ies
th
at
h
av
e
r
ep
o
r
ted
g
r
ea
ter
s
tab
il
ity
an
d
f
id
elity
f
o
r
g
r
ad
ie
n
t
-
b
ased
v
is
u
al
ex
p
lan
atio
n
s
in
p
l
an
t
d
is
ea
s
e
class
if
icatio
n
.
I
n
th
e
ex
p
er
i
m
en
ts
,
th
e
s
im
p
ler
f
ea
tu
r
e
r
ep
r
esen
tatio
n
s
d
er
iv
e
d
f
r
o
m
m
an
u
al
f
ea
t
u
r
e
en
g
in
ee
r
in
g
co
m
b
in
e
d
with
r
elief
s
elec
tio
n
m
ay
h
av
e
co
n
t
r
ib
u
ted
to
m
o
r
e
r
o
b
u
s
t
lo
ca
l
ex
p
lan
atio
n
s
wh
en
p
er
tu
r
b
e
d
b
y
d
ata
au
g
m
en
tatio
n
s
s
u
ch
as r
o
tatio
n
,
b
r
ig
h
t
n
ess
v
ar
iatio
n
,
an
d
Gau
s
s
ian
n
o
is
e.
Mo
r
eo
v
er
,
wh
ile
Gr
ad
-
C
AM
p
r
o
v
id
e
d
v
is
u
ally
i
n
tu
itiv
e
h
e
atm
ap
s
h
ig
h
lig
h
tin
g
d
is
ea
s
e
r
eg
io
n
s
,
its
q
u
an
titativ
e
r
o
b
u
s
tn
ess
was
s
o
m
ewh
at
lo
wer
,
s
u
g
g
esti
n
g
t
h
at
ex
p
lan
atio
n
s
tab
ilit
y
ca
n
d
ep
en
d
s
tr
o
n
g
ly
o
n
mod
el
co
m
p
lex
ity
a
n
d
th
e
h
o
m
o
g
en
eity
o
f
im
a
g
e
b
ac
k
g
r
o
u
n
d
s
.
T
h
is
d
iv
e
r
g
en
ce
h
ig
h
lig
h
t
s
th
e
im
p
o
r
tan
ce
o
f
ev
alu
atin
g
in
ter
p
r
etab
ilit
y
m
eth
o
d
s
n
o
t
o
n
ly
q
u
alitativ
ely
b
u
t
also
u
s
in
g
o
b
jectiv
e
m
etr
ics
tailo
r
ed
to
ag
r
icu
ltu
r
al
d
atasets
.
Ov
er
all,
th
e
r
esu
lts
in
d
icat
e
th
at
w
h
ile
C
NNs
o
f
f
er
s
u
p
er
i
o
r
class
if
icatio
n
p
e
r
f
o
r
m
an
ce
,
s
im
p
ler
m
o
d
els
with
f
ea
tu
r
e
s
elec
tio
n
ca
n
y
ield
e
x
p
lan
atio
n
s
with
h
ig
h
e
r
f
id
elity
an
d
s
tab
ilit
y
u
n
d
er
ce
r
tain
co
n
d
itio
n
s
,
s
u
p
p
o
r
tin
g
th
e
ca
s
e
f
o
r
h
y
b
r
id
ap
p
r
o
ac
h
es
th
at
b
alan
ce
ac
cu
r
ac
y
an
d
in
ter
p
r
e
tab
ilit
y
in
p
r
ec
is
io
n
ag
r
icu
ltu
r
e
a
p
p
licatio
n
s
.
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
co
m
p
ar
ed
two
ap
p
r
o
ac
h
es
f
o
r
p
lan
t
d
is
ea
s
e
cla
s
s
if
icatio
n
an
d
in
ter
p
r
etab
ilit
y
:
an
ANN
lev
er
ag
in
g
m
an
u
ally
en
g
in
ee
r
ed
f
ea
tu
r
es
s
elec
ted
b
y
r
elief
an
d
in
ter
p
r
eted
u
s
in
g
L
I
ME
,
a
n
d
a
Mo
b
ileNetV2
-
b
ased
C
NN
with
Gr
ad
-
C
AM
v
is
u
al
ex
p
lan
atio
n
s
.
T
h
e
Mo
b
ileNetV2
m
o
d
el
ac
h
iev
e
d
a
h
ig
h
class
if
icatio
n
ac
cu
r
ac
y
o
f
9
9
.
7
6
%,
c
o
n
f
ir
m
in
g
th
e
ef
f
ec
tiv
e
n
ess
o
f
d
ee
p
lear
n
in
g
in
p
lan
t
d
is
ea
s
e
d
e
tectio
n
.
Ho
wev
er
,
L
I
ME
ex
p
lan
atio
n
s
p
r
o
v
id
e
d
a
lo
wer
f
id
elity
d
r
o
p
(
1
0
.
9
0
%)
an
d
h
ig
h
er
r
o
b
u
s
tn
ess
(
8
4
.
2
7
%)
co
m
p
a
r
ed
to
Gr
ad
-
C
AM
(
1
8
.
6
8
%
f
id
elity
d
r
o
p
an
d
7
9
.
3
8
%
r
o
b
u
s
tn
ess
)
,
in
d
icatin
g
th
at
s
im
p
ler
m
o
d
els
with
ex
p
licit
f
ea
tu
r
e
s
elec
tio
n
ca
n
y
ield
m
o
r
e
s
tab
le
an
d
f
aith
f
u
l
ex
p
lan
atio
n
s
u
n
d
er
d
ata
au
g
m
e
n
tatio
n
s
s
im
u
latin
g
f
ield
co
n
d
itio
n
s
.
T
h
ese
f
in
d
i
n
g
s
h
ig
h
lig
h
t
th
e
im
p
o
r
tan
ce
o
f
ev
alu
atin
g
in
ter
p
r
etab
ilit
y
m
eth
o
d
s
q
u
an
titativ
ely
an
d
co
n
tex
tu
ally
to
en
s
u
r
e
tr
u
s
two
r
th
y
d
ep
lo
y
m
en
t
in
p
r
ec
is
io
n
ag
r
icu
ltu
r
e.
Fo
r
f
u
tu
r
e
wo
r
k
,
we
p
lan
t
o
ex
ten
d
th
is
co
m
p
ar
ativ
e
a
n
al
y
s
is
to
m
o
r
e
co
m
p
lex
a
n
d
d
iv
er
s
e
d
atasets
with
r
ea
l
-
wo
r
ld
f
ield
im
ag
es
co
n
tain
in
g
b
ac
k
g
r
o
u
n
d
v
a
r
iab
ilit
y
an
d
o
c
clu
s
io
n
s
.
I
n
co
r
p
o
r
atin
g
ad
d
iti
o
n
al
in
ter
p
r
etab
ilit
y
tec
h
n
iq
u
e
s
,
s
u
ch
as
SHAP
o
r
in
teg
r
ated
g
r
ad
ien
ts
,
co
u
ld
p
r
o
v
id
e
f
u
r
th
er
in
s
ig
h
ts
in
to
th
e
s
tr
en
g
th
s
an
d
lim
itatio
n
s
o
f
d
if
f
er
en
t
ex
p
lan
atio
n
m
eth
o
d
s
.
C
o
llab
o
r
atio
n
with
ag
r
icu
ltu
r
al
e
x
p
er
ts
ca
n
f
u
r
th
er
tailo
r
in
ter
p
r
etab
ilit
y
o
u
tp
u
ts
to
s
u
p
p
o
r
t
in
f
o
r
m
e
d
d
ec
is
io
n
-
m
ak
in
g
in
th
e
f
ield
.
Mo
r
e
o
v
er
,
ex
p
l
o
r
i
n
g
h
y
b
r
id
m
o
d
els
th
at
in
teg
r
ate
h
an
d
cr
af
ted
a
n
d
lear
n
ed
f
ea
tu
r
es m
ay
h
el
p
b
ala
n
ce
in
ter
p
r
eta
b
ilit
y
with
p
r
ed
i
ctiv
e
p
er
f
o
r
m
an
ce
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
I
n
terp
r
etin
g
p
o
ta
to
d
is
ea
s
e
cla
s
s
ifica
tio
n
u
s
in
g
ex
p
la
in
a
b
le
a
r
tifi
cia
l in
tellig
en
ce
(
R
a
ke
s
h
K
u
ma
r
Gu
ma
s
ta
)
1129
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
R
a
k
e
s
h
Ku
m
ar
G
u
m
a
s
t
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ajay
So
m
k
u
war
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
t
h
o
r
s
d
ec
lar
e
t
h
at
th
e
r
e
ar
e
n
o
k
n
o
wn
co
n
f
licts
o
f
in
ter
est
ass
o
ciate
d
with
th
is
p
u
b
licatio
n
.
T
h
er
e
ar
e
n
o
f
in
a
n
cial
o
r
p
e
r
s
o
n
al
r
elatio
n
s
h
ip
s
th
at
co
u
ld
in
ap
p
r
o
p
r
iately
in
f
l
u
en
ce
o
r
b
ias
th
e
co
n
ten
t
o
f
th
is
wo
r
k
.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
av
ailab
ilit
y
is
n
o
t
ap
p
l
icab
le
to
th
is
p
ap
er
as
n
o
n
ew
d
ata
wer
e
cr
ea
ted
o
r
a
n
aly
ze
d
i
n
th
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
R
o
m
a
n
,
M
.
M
.
R
a
h
ma
n
,
S
.
A
.
H
a
i
d
e
r
,
T.
A
k
r
a
m,
a
n
d
S
.
R
.
N
a
q
v
i
,
“
I
n
t
e
g
r
a
t
i
n
g
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
:
a
h
y
b
r
i
d
a
p
p
r
o
a
c
h
f
o
r
smar
t
a
g
r
i
c
u
l
t
u
r
e
a
p
p
l
i
c
a
t
i
o
n
s
,
”
Al
g
o
ri
t
h
m
s
,
v
o
l
.
1
8
,
n
o
.
4
,
A
p
r
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
a
1
8
0
4
0
2
2
2
.
[
2
]
K
.
S
y
e
d
,
S
.
S
.
A
.
B
e
g
u
m
,
A
.
R
.
P
a
l
a
k
a
y
a
l
a
,
G
.
V
.
V
.
La
k
s
h
mi
,
a
n
d
S
.
G
o
r
i
k
a
p
u
d
i
,
“
L
e
a
f
d
i
s
e
a
s
e
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
f
o
o
d
c
r
o
p
s
w
i
t
h
e
f
f
i
c
i
e
n
t
f
e
a
t
u
r
e
d
i
m
e
n
s
i
o
n
a
l
i
t
y
r
e
d
u
c
t
i
o
n
,
”
PLO
S
O
n
e
,
v
o
l
.
2
0
,
n
o
.
8
,
A
u
g
.
2
0
2
5
,
d
o
i
:
1
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
o
n
e
.
0
3
2
8
3
4
9
.
[
3
]
K
.
K
i
r
a
a
n
d
L.
A
.
R
e
n
d
e
l
l
,
“
A
p
r
a
c
t
i
c
a
l
a
p
p
r
o
a
c
h
t
o
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
,
”
i
n
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
Pr
o
c
e
e
d
i
n
g
s
1
9
9
2
,
B
u
r
l
i
n
g
t
o
n
,
U
n
i
t
e
d
S
t
a
t
e
s:
M
o
r
g
a
n
K
a
u
f
ma
n
n
,
1
9
9
2
,
p
p
.
2
4
9
–
2
5
6
,
d
o
i
:
1
0
.
1
0
1
6
/
B
9
7
8
-
1
-
5
5
8
6
0
-
2
4
7
-
2
.
5
0
0
3
7
-
1.
[
4
]
S
.
C
.
G
i
r
o
n
e
s,
S
.
M
u
n
e
r
a
,
M
.
M
.
S
o
b
e
r
,
J.
B
l
a
sc
o
,
S
.
C
u
b
e
r
o
,
a
n
d
J
.
G
.
S
a
n
c
h
i
s
,
“
A
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s i
n
a
g
r
i
c
u
l
t
u
r
e
,
t
h
e
c
o
r
e
o
f
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
:
w
h
a
t
,
w
h
e
n
,
a
n
d
w
h
y
,
”
C
o
m
p
u
t
e
rs
a
n
d
El
e
c
t
r
o
n
i
c
s
i
n
A
g
ri
c
u
l
t
u
re
,
v
o
l
.
2
3
0
,
M
a
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
p
a
g
.
2
0
2
5
.
1
0
9
9
3
8
.
[
5
]
P
.
S
r
i
d
h
a
r
a
n
d
P
.
A
n
g
a
mu
t
h
u
,
“
E
n
h
a
n
c
i
n
g
i
ma
g
e
b
a
s
e
d
c
l
a
ssi
f
i
c
a
t
i
o
n
f
o
r
c
r
o
p
d
i
sea
s
e
d
e
t
e
c
t
i
o
n
u
s
i
n
g
a
mu
l
t
i
c
l
a
s
s
S
V
M
a
p
p
r
o
a
c
h
w
i
t
h
k
e
r
n
e
l
c
o
mp
a
r
i
s
o
n
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
rt
s
,
v
o
l
.
1
5
,
n
o
.
1
,
N
o
v
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
5
-
2
3
5
6
8
-
w.
[
6
]
T.
N
y
a
w
o
se,
R
.
C
.
M
a
sw
a
n
g
a
n
y
i
,
a
n
d
P
.
K
h
u
m
a
l
o
,
“
A
r
e
v
i
e
w
o
n
t
h
e
d
e
t
e
c
t
i
o
n
o
f
p
l
a
n
t
d
i
s
e
a
s
e
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s,
”
J
o
u
r
n
a
l
o
f
I
m
a
g
i
n
g
,
v
o
l
.
1
1
,
n
o
.
1
0
,
S
e
p
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
j
i
ma
g
i
n
g
1
1
1
0
0
3
2
6
.
[
7
]
B
.
S
a
m
b
a
n
a
e
t
a
l
.
,
“
A
n
e
f
f
i
c
i
e
n
t
p
l
a
n
t
d
i
se
a
se
d
e
t
e
c
t
i
o
n
u
s
i
n
g
t
r
a
n
s
f
e
r
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
rt
s
,
v
o
l
.
1
5
,
n
o
.
1
,
M
a
y
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
5
-
0
2
2
7
1
-
w.
[
8
]
M
.
B
.
D
i
a
l
l
o
e
t
a
l
.
,
“
E
n
h
a
n
c
e
d
-
R
I
C
A
P
:
a
n
o
v
e
l
d
a
t
a
a
u
g
me
n
t
a
t
i
o
n
s
t
r
a
t
e
g
y
f
o
r
i
m
p
r
o
v
e
d
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
p
l
a
n
t
d
i
s
e
a
se
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
d
m
o
b
i
l
e
d
i
a
g
n
o
si
s
,
”
Fro
n
t
i
e
rs
i
n
P
l
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
6
,
S
e
p
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s.
2
0
2
5
.
1
6
4
6
6
1
1
.
[
9
]
E.
-
M
.
N
i
k
o
l
a
d
o
s
,
A
.
W
o
n
g
p
r
o
mm
o
o
n
,
O
.
M
.
A
o
d
h
a
,
G
.
C
a
m
b
r
a
y
,
a
n
d
D
.
A
.
O
y
a
r
z
ú
n
,
“
A
c
c
u
r
a
c
y
a
n
d
d
a
t
a
e
f
f
i
c
i
e
n
c
y
i
n
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
o
f
p
r
o
t
e
i
n
e
x
p
r
e
ssi
o
n
,
”
N
a
t
u
r
e
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
1
3
,
n
o
.
1
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
4
6
7
-
0
2
2
-
3
4
9
0
2
-
5.
[
1
0
]
Y
.
B
a
h
r
i
,
E
.
D
y
e
r
,
J
.
K
a
p
l
a
n
,
J
.
L
e
e
,
a
n
d
U
.
S
h
a
r
ma,
“
E
x
p
l
a
i
n
i
n
g
n
e
u
r
a
l
s
c
a
l
i
n
g
l
a
w
s
,
”
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
N
a
t
i
o
n
a
l
Ac
a
d
e
m
y
o
f
S
c
i
e
n
c
e
s
,
v
o
l
.
1
2
1
,
n
o
.
2
7
,
J
u
l
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
7
3
/
p
n
a
s
.
2
3
1
1
8
7
8
1
2
1
.
[
1
1
]
M
.
T
.
R
i
b
e
i
r
o
,
S
.
S
i
n
g
h
,
a
n
d
C
.
G
u
e
s
t
r
i
n
,
“
‘
W
h
y
sh
o
u
l
d
I
t
r
u
st
y
o
u
?
’
:
e
x
p
l
a
i
n
i
n
g
t
h
e
p
r
e
d
i
c
t
i
o
n
s
o
f
a
n
y
c
l
a
ssi
f
i
e
r
,
”
i
n
2
2
n
d
AC
M
S
I
G
K
D
D
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
K
n
o
w
l
e
d
g
e
D
i
sc
o
v
e
ry
a
n
d
D
a
t
a
M
i
n
i
n
g
,
A
u
g
.
2
0
1
6
,
p
p
.
1
1
3
5
–
1
1
4
4
,
do
i
:
1
0
.
1
1
4
5
/
2
9
3
9
6
7
2
.
2
9
3
9
7
7
8
.
[
1
2
]
R
.
R
.
S
e
l
v
a
r
a
j
u
,
M
.
C
o
g
sw
e
l
l
,
A
.
D
a
s,
R
.
V
e
d
a
n
t
a
m,
D
.
P
a
r
i
k
h
,
a
n
d
D
.
B
a
t
r
a
,
“
G
r
a
d
-
C
A
M
:
v
i
s
u
a
l
e
x
p
l
a
n
a
t
i
o
n
s
f
r
o
m
d
e
e
p
n
e
t
w
o
r
k
s
v
i
a
g
r
a
d
i
e
n
t
-
b
a
se
d
l
o
c
a
l
i
z
a
t
i
o
n
,
”
i
n
2
0
1
7
I
EEE
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
s
i
o
n
,
O
c
t
.
2
0
1
7
,
p
p
.
6
1
8
–
626
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
V
.
2
0
1
7
.
7
4
.
[
1
3
]
R
.
P
.
P
o
r
f
í
r
i
o
,
R
.
N
.
M
a
d
e
i
r
a
,
a
n
d
P
.
A
.
S
a
n
t
o
s,
“
Ex
p
l
o
r
i
n
g
e
x
p
l
a
i
n
a
b
l
e
A
I
t
e
c
h
n
i
q
u
e
s
f
o
r
p
l
a
n
t
d
i
se
a
se
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
d
i
g
i
t
a
l
a
g
r
i
c
u
l
t
u
r
e
,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
6
5
,
p
p
.
1
7
5
–
1
8
2
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
5
.
0
7
.
1
7
0
.
[
1
4
]
R
.
K
u
s
u
ma
a
n
d
R
.
R
a
j
k
u
m
a
r
,
“
P
l
a
n
t
l
e
a
f
d
i
se
a
se
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
t
e
c
h
n
i
q
u
e
s:
a
r
e
v
i
e
w
,
”
I
n
d
o
n
e
si
a
n
J
o
u
r
n
a
l
o
f
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
8
,
n
o
.
2
,
pp
.
1
3
0
8
-
1
3
2
3
,
M
a
y
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
3
8
.
i
2
.
p
p
1
3
0
8
-
1
3
2
3
.
[
1
5
]
V
.
B
o
p
p
a
n
a
,
“
P
l
a
n
t
d
i
se
a
se
d
e
t
e
c
t
i
o
n
u
si
n
g
h
y
b
r
i
d
M
o
b
i
l
e
N
e
t
V
2
-
c
o
m
p
a
c
t
C
N
N
a
r
c
h
i
t
e
c
t
u
r
e
w
i
t
h
LI
M
E
i
n
t
e
g
r
a
t
i
o
n
,
”
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s E
n
g
i
n
e
e
ri
n
g
a
n
d
Ma
n
a
g
e
m
e
n
t
,
v
o
l
.
1
0
,
n
o
.
1
3
s,
p
p
.
5
5
4
–
5
6
8
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
5
2
7
8
3
/
j
i
se
m.v
1
0
i
1
3
s
.
2
1
1
1
.
[
1
6
]
M
.
N
a
h
i
d
u
z
z
a
ma
n
e
t
a
l
.
,
“
E
x
p
l
a
i
n
a
b
l
e
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
a
u
t
o
m
a
t
i
c
mu
l
b
e
r
r
y
l
e
a
f
d
i
s
e
a
s
e
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
Fro
n
t
i
e
rs
i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
4
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s
.
2
0
2
3
.
1
1
7
5
5
1
5
.
[
1
7
]
V
.
M
.
G
u
t
i
é
r
r
e
z
e
t
a
l
.
,
“
A
I
-
d
r
i
v
e
n
p
l
a
n
t
h
e
a
l
t
h
a
ss
e
ssm
e
n
t
:
a
c
o
mp
a
r
a
t
i
v
e
a
n
a
l
y
si
s
o
f
I
n
c
e
p
t
i
o
n
V
3
,
R
e
sN
e
t
-
5
0
a
n
d
V
i
T
w
i
t
h
S
H
A
P
f
o
r
a
c
c
u
r
a
t
e
d
i
se
a
se
i
d
e
n
t
i
f
i
c
a
t
i
o
n
i
n
t
a
r
o
,
”
Ag
ro
n
o
m
y
,
v
o
l
.
1
5
,
n
o
.
1
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
o
n
o
m
y
1
5
0
1
0
0
7
7
.
[
1
8
]
S
.
M
.
A
l
h
a
mm
a
d
,
D
.
S
.
K
h
a
f
a
g
a
,
W
.
M
.
El
-
h
a
d
y
,
F
.
M
.
S
a
m
y
,
a
n
d
K
.
M
.
H
o
sn
y
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
n
d
e
x
p
l
a
i
n
a
b
l
e
A
I
f
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
p
o
t
a
t
o
l
e
a
f
d
i
sea
ses,
”
Fro
n
t
i
e
rs
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
7
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
r
a
i
.
2
0
2
4
.
1
4
4
9
3
2
9
.
[
1
9
]
K
.
G
o
p
a
l
a
n
,
S
.
S
r
i
n
i
v
a
s
a
n
,
P
r
a
g
y
a
,
M
.
S
i
n
g
h
,
S
.
K
.
M
a
t
h
i
v
a
n
a
n
,
a
n
d
U
.
M
o
o
r
t
h
y
,
“
C
o
r
n
l
e
a
f
d
i
s
e
a
s
e
d
i
a
g
n
o
s
i
s
:
e
n
h
a
n
c
i
n
g
a
c
c
u
r
a
c
y
w
i
t
h
r
e
s
n
e
t
1
5
2
a
n
d
g
r
a
d
-
c
a
m
f
o
r
e
x
p
l
a
i
n
a
b
l
e
A
I
,
”
B
MC
Pl
a
n
t
B
i
o
l
o
g
y
,
v
o
l
.
2
5
,
n
o
.
1
,
A
p
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
2
8
7
0
-
0
2
5
-
0
6
3
8
6
-
0.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
2
0
2
6
:
1
1
2
3
-
1
1
3
0
1130
[
2
0
]
M
.
J.
K
a
r
i
m,
M
.
O
.
F
.
G
o
n
i
,
M
.
N
a
h
i
d
u
z
z
a
ma
n
,
M
.
A
h
sa
n
,
J.
H
a
i
d
e
r
,
a
n
d
M
.
K
o
w
a
l
s
k
i
,
“
E
n
h
a
n
c
i
n
g
a
g
r
i
c
u
l
t
u
r
e
t
h
r
o
u
g
h
r
e
a
l
-
t
i
m
e
g
r
a
p
e
l
e
a
f
d
i
se
a
se
c
l
a
ss
i
f
i
c
a
t
i
o
n
v
i
a
a
n
e
d
g
e
d
e
v
i
c
e
w
i
t
h
a
l
i
g
h
t
w
e
i
g
h
t
C
N
N
a
r
c
h
i
t
e
c
t
u
r
e
a
n
d
G
r
a
d
-
C
A
M
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
1
4
,
n
o
.
1
,
Ju
l
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
4
-
6
6
9
8
9
-
9.
[
2
1
]
S
.
M
o
s
t
a
f
a
,
D
.
M
o
n
d
a
l
,
K
.
P
a
n
j
v
a
n
i
,
L.
K
o
c
h
i
a
n
,
a
n
d
I
.
S
t
a
v
n
e
ss,
“
E
x
p
l
a
i
n
a
b
l
e
d
e
e
p
l
e
a
r
n
i
n
g
i
n
p
l
a
n
t
p
h
e
n
o
t
y
p
i
n
g
,
”
Fr
o
n
t
i
e
rs
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
6
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
8
9
/
f
r
a
i
.
2
0
2
3
.
1
2
0
3
5
4
6
.
[
2
2
]
J.
A
mar
a
,
B
.
K
.
R
i
e
s,
a
n
d
S
.
S
a
mu
e
l
,
“
C
o
n
c
e
p
t
e
x
p
l
a
i
n
a
b
i
l
i
t
y
f
o
r
p
l
a
n
t
d
i
se
a
ses
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
i
n
1
8
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
i
n
t
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
si
o
n
,
I
m
a
g
i
n
g
a
n
d
C
o
m
p
u
t
e
r
G
ra
p
h
i
c
s
T
h
e
o
ry
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
2
0
2
3
,
p
p
.
2
4
6
–
253
,
d
o
i
:
1
0
.
5
2
2
0
/
0
0
1
1
6
6
7
9
0
0
0
0
3
4
1
7
.
[
2
3
]
J.
Lu
,
X
.
Li
u
,
X
.
M
a
,
J.
To
n
g
,
a
n
d
J.
P
e
n
g
,
“
I
mp
r
o
v
e
d
M
o
b
i
l
e
N
e
t
V
2
c
r
o
p
d
i
s
e
a
se
i
d
e
n
t
i
f
i
c
a
t
i
o
n
m
o
d
e
l
f
o
r
i
n
t
e
l
l
i
g
e
n
t
a
g
r
i
c
u
l
t
u
r
e
,
”
Pe
e
r
J
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
9
,
2
0
2
3
,
d
o
i
:
1
0
.
7
7
1
7
/
p
e
e
r
j
-
c
s.
1
5
9
5
.
[
2
4
]
H
.
K
.
K
o
n
d
a
v
e
e
t
i
a
n
d
C
.
G
.
S
i
m
h
a
d
r
i
,
“
E
v
a
l
u
a
t
i
o
n
o
f
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
u
si
n
g
e
x
p
l
a
i
n
a
b
l
e
A
I
w
i
t
h
q
u
a
l
i
t
a
t
i
v
e
a
n
d
q
u
a
n
t
i
t
a
t
i
v
e
a
n
a
l
y
si
s
f
o
r
r
i
c
e
l
e
a
f
d
i
s
e
a
se
d
e
t
e
c
t
i
o
n
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
5
,
n
o
.
1
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
025
-
1
4
3
0
6
-
3.
[
2
5
]
A
.
B
h
u
j
e
l
,
N
.
-
E
.
K
i
m
,
E.
A
r
u
l
mo
z
h
i
,
J.
K
.
B
a
s
a
k
,
a
n
d
H
.
-
T.
K
i
m,
“
A
l
i
g
h
t
w
e
i
g
h
t
a
t
t
e
n
t
i
o
n
-
b
a
se
d
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
f
o
r
t
o
ma
t
o
l
e
a
f
d
i
s
e
a
s
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
Ag
r
i
c
u
l
t
u
re
,
v
o
l
.
1
2
,
n
o
.
2
,
F
e
b
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
i
c
u
l
t
u
r
e
1
2
0
2
0
2
2
8
.
[
2
6
]
A
.
M
u
mu
n
i
a
n
d
F
.
M
u
mu
n
i
,
“
D
a
t
a
a
u
g
m
e
n
t
a
t
i
o
n
:
a
c
o
mp
r
e
h
e
n
s
i
v
e
su
r
v
e
y
o
f
m
o
d
e
r
n
a
p
p
r
o
a
c
h
e
s
,
”
Arr
a
y
,
v
o
l
.
1
6
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
r
r
a
y
.
2
0
2
2
.
1
0
0
2
5
8
.
[
2
7
]
A
.
R
.
A
l
-
S
h
a
mas
n
e
h
a
n
d
R
.
W
.
I
b
r
a
h
i
m
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
o
f
t
o
ma
t
o
l
e
a
f
i
mag
e
s
f
o
r
d
e
t
e
c
t
i
o
n
o
f
p
l
a
n
t
d
i
s
e
a
s
e
u
si
n
g
c
o
n
f
o
r
ma
b
l
e
p
o
l
y
n
o
mi
a
l
s
i
m
a
g
e
f
e
a
t
u
r
e
s,
”
Me
t
h
o
d
sX
,
v
o
l
.
1
3
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
mex
.
2
0
2
4
.
1
0
2
8
4
4
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ra
k
e
sh
K
u
m
a
r
G
u
m
a
sta
is
p
re
se
n
tl
y
p
u
rsu
i
n
g
a
P
h
.
D
.
i
n
t
h
e
D
e
p
a
rtme
n
t
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
a
t
M
a
u
lan
a
A
z
a
d
Na
ti
o
n
a
l
I
n
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
B
h
o
p
a
l,
In
d
ia.
H
e
re
c
e
iv
e
d
a
n
M
.
E.
d
e
g
re
e
i
n
Dig
it
a
l
C
o
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
fro
m
UIT
-
RG
P
V
Bh
o
p
a
l
.
He
h
a
s
m
o
re
th
a
n
1
2
y
e
a
r
s
o
f
e
x
p
e
rien
c
e
i
n
re
se
a
rc
h
,
d
e
v
e
lo
p
m
e
n
t,
a
n
d
tea
c
h
i
n
g
.
He
h
a
s
c
o
n
d
u
c
ted
trai
n
in
g
f
o
r
i
n
d
u
stri
e
s.
His
a
re
a
s
o
f
e
x
p
e
rti
se
a
re
M
ATLAB,
VLS
I
,
a
n
d
AI.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ra
k
e
sh
g
u
m
a
sta
4
4
@g
m
a
il
.
c
o
m
.
Dr
.
Aja
y
S
o
m
k
u
wa
r
is
p
re
s
e
n
tl
y
wo
rk
in
g
a
s
a
p
ro
fe
s
so
r
in
t
h
e
D
e
p
a
rtme
n
t
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
a
t
M
a
u
lan
a
Az
a
d
Na
ti
o
n
a
l
I
n
stit
u
te
o
f
Tec
h
n
o
lo
g
y
.
He
re
c
e
iv
e
d
h
is
P
h
.
D
.
fr
o
m
In
d
ian
I
n
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
Ne
w
De
lh
i,
I
n
d
ia
,
in
2
0
0
3
.
He
h
a
s
p
u
b
li
sh
e
d
m
o
re
t
h
a
n
1
0
0
p
a
p
e
r
s
o
f
n
a
ti
o
n
a
l
a
n
d
i
n
tern
a
ti
o
n
a
l
re
p
u
te.
His
re
se
a
rc
h
a
re
a
in
c
lu
d
e
s
si
g
n
a
l
p
r
o
c
e
ss
in
g
,
sp
e
e
c
h
sig
n
a
l
p
r
o
c
e
ss
in
g
,
a
n
d
b
i
o
sig
n
a
l
e
n
g
i
n
e
e
rin
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
As
o
m
k
u
wa
r@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.