I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
1
5
,
No
.
2
,
A
p
r
il 2
0
2
6
,
p
p
.
1
3
8
1
~
1
3
8
9
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
2
.
p
p
1
3
8
1
-
1
3
8
9
1381
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Autom
a
ted
cla
ss
if
ica
tion o
f
a
pple
b
ruises
f
ro
m
hy
per
spectral
ima
g
es: an a
ppro
a
ch f
o
r f
ruit
qua
li
ty a
ss
ess
ment
P
eddireddy
Venk
a
t
eswa
ra
Reddy
,
Ala
g
ucha
m
y
P
a
ri
v
a
zh
a
g
a
n
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
S
c
h
o
o
l
o
f
C
o
m
p
u
t
i
n
g
,
K
a
l
a
s
a
l
i
n
g
a
m
A
c
a
d
e
my
o
f
R
e
s
e
a
r
c
h
a
n
d
E
d
u
c
a
t
i
o
n
,
K
r
i
s
h
n
a
n
k
o
i
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
6
,
2
0
2
4
R
ev
is
ed
J
an
2
4
,
2
0
2
6
Acc
ep
ted
Feb
6
,
2
0
2
6
Ap
p
le
b
ru
ise
d
e
tec
ti
o
n
p
la
y
s
a
c
ru
c
ial
ro
le
in
p
o
st
-
h
a
rv
e
st
q
u
a
li
t
y
c
o
n
tro
l
;
h
o
we
v
e
r,
c
o
n
v
e
n
ti
o
n
a
l
m
a
n
u
a
l
in
sp
e
c
ti
o
n
re
m
a
in
s
lab
o
r
-
in
ten
si
v
e
,
su
b
jec
ti
v
e
,
a
n
d
u
n
su
it
a
b
le
fo
r
lar
g
e
-
sc
a
le
in
d
u
str
ial
d
e
p
lo
y
m
e
n
t.
Th
is
st
u
d
y
p
ro
p
o
se
s
a
n
a
u
to
m
a
ted
c
las
sifica
ti
o
n
fra
m
e
wo
rk
fo
r
id
e
n
ti
fy
in
g
b
ru
ise
d
re
g
io
n
s
i
n
a
p
p
les
u
sin
g
h
y
p
e
rsp
e
c
tral
ima
g
in
g
c
o
m
b
i
n
e
d
wit
h
d
e
e
p
lea
rn
in
g
a
n
d
a
d
a
p
ti
v
e
o
p
ti
m
iza
ti
o
n
tec
h
n
i
q
u
e
s.
Th
e
p
ro
p
o
se
d
m
o
d
e
l
in
te
g
r
a
tes
a
lo
n
g
sh
o
rt
-
term
m
e
m
o
r
y
(LS
TM
)
n
e
t
wo
rk
o
p
ti
m
ize
d
u
sin
g
a
n
a
d
a
p
ti
v
e
sa
n
d
c
a
t
swa
rm
o
p
ti
m
iza
ti
o
n
(ASCS
O)
a
lg
o
rit
h
m
,
a
lo
n
g
wit
h
a
Re
sN
e
t
-
5
0
fe
a
tu
re
e
x
trac
ti
o
n
b
a
c
k
b
o
n
e
.
T
h
e
a
d
a
p
ti
v
e
b
e
h
a
v
io
r
e
m
b
e
d
d
e
d
wit
h
in
ASCS
O
d
y
n
a
m
ica
ll
y
a
d
j
u
sts
t
h
e
o
p
ti
m
iz
a
ti
o
n
p
a
ra
m
e
ters
to
e
n
h
a
n
c
e
c
o
n
v
e
rg
e
n
c
e
a
n
d
p
re
v
e
n
t
p
re
m
a
tu
re
sta
g
n
a
ti
o
n
d
u
ri
n
g
L
S
TM
h
y
p
e
rp
a
ra
m
e
ter
tu
n
i
n
g
.
Hy
p
e
rsp
e
c
tral
ima
g
e
s
we
re
p
ro
c
e
ss
e
d
to
e
x
trac
t
re
lev
a
n
t
s
p
e
c
tral
–
sp
a
ti
a
l
fe
a
tu
re
s,
wh
ich
we
re
su
b
se
q
u
e
n
tl
y
fe
d
in
to
t
h
e
o
p
ti
m
ize
d
c
las
sifier.
Ex
p
e
rime
n
tal
e
v
a
l
u
a
ti
o
n
s
d
e
m
o
n
stra
te
th
a
t
t
h
e
p
ro
p
o
se
d
h
y
b
r
id
m
o
d
e
l
sig
n
ifi
c
a
n
t
ly
o
u
tp
e
rf
o
rm
s
c
o
n
v
e
n
ti
o
n
a
l
a
n
d
b
a
se
li
n
e
d
e
e
p
lea
rn
in
g
a
p
p
ro
a
c
h
e
s,
a
c
h
iev
i
n
g
a
c
las
sifica
ti
o
n
a
c
c
u
ra
c
y
o
f
9
8
.
0
%
w
h
il
e
m
a
in
tain
i
n
g
ro
b
u
stn
e
ss
a
c
ro
ss
v
a
ry
in
g
b
ru
ise
p
a
tt
e
rn
s
a
n
d
in
ten
sity
le
v
e
ls.
Th
e
re
su
lt
s
h
ig
h
li
g
h
t
th
e
e
ffe
c
ti
v
e
n
e
ss
o
f
c
o
m
b
in
in
g
h
y
p
e
rsp
e
c
tral
ima
g
in
g
wit
h
a
d
a
p
ti
v
e
d
e
e
p
lea
rn
in
g
o
p
ti
m
iza
ti
o
n
f
o
r
h
i
g
h
-
p
re
c
isio
n
fru
i
t
q
u
a
li
ty
a
ss
e
ss
m
e
n
t.
Th
is
re
se
a
rc
h
c
o
n
t
rib
u
tes
a
re
li
a
b
le,
sc
a
lab
le
so
l
u
ti
o
n
fo
r
a
u
to
m
a
ted
b
r
u
ise
d
e
tec
ti
o
n
a
n
d
q
u
a
li
t
y
g
ra
d
in
g
in
t
h
e
fru
i
t
su
p
p
ly
c
h
a
i
n
,
o
ffe
rin
g
stro
n
g
p
o
te
n
ti
a
l
t
o
r
e
d
u
c
e
p
o
st
-
h
a
r
v
e
st
l
o
ss
e
s
a
n
d
imp
ro
v
e
o
p
e
ra
ti
o
n
a
l
e
fficie
n
c
y
i
n
t
h
e
a
g
r
o
-
fo
o
d
in
d
u
str
y
.
K
ey
w
o
r
d
s
:
Ad
ap
tiv
e
s
an
d
ca
t swar
m
o
p
tim
izatio
n
Ap
p
le
b
r
u
is
es
Fo
o
d
q
u
ality
ass
ess
m
en
t
o
p
p
o
s
itio
n
-
b
ased
lear
n
in
g
Hy
p
er
s
p
ec
tr
al
L
o
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ped
d
ir
ed
d
y
Ven
k
ateswar
a
R
ed
d
y
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
Sch
o
o
l o
f
C
o
m
p
u
tin
g
Kala
s
alin
g
am
Aca
d
em
y
o
f
R
esear
ch
an
d
E
d
u
ca
tio
n
Kr
is
h
n
an
k
o
il,
T
am
il Na
d
u
,
I
n
d
ia
E
m
ail:
v
en
k
ateswar
lu
r
ee
d
d
y
p
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Ap
p
le
b
r
u
is
in
g
is
a
m
ajo
r
p
o
s
t
-
h
ar
v
est
q
u
ality
is
s
u
e
ca
u
s
ed
b
y
m
ec
h
an
ical
s
tr
ess
es
d
u
r
in
g
h
an
d
lin
g
an
d
tr
an
s
p
o
r
tatio
n
,
lead
i
n
g
to
tis
s
u
e
d
am
ag
e,
m
icr
o
b
ial
s
u
s
ce
p
tib
ilit
y
,
r
ed
u
ce
d
s
h
elf
life
,
an
d
ec
o
n
o
m
ic
lo
s
s
[
1
]
–
[
3
]
.
E
a
r
ly
an
d
a
cc
u
r
ate
b
r
u
is
e
d
etec
tio
n
is
ess
en
tial,
y
et
co
n
v
e
n
tio
n
al
R
GB
im
ag
in
g
tech
n
i
q
u
es
ar
e
lim
ited
in
id
en
tify
in
g
s
u
b
tl
e
o
r
s
u
b
s
u
r
f
ac
e
d
a
m
ag
e
[
4
]
.
Hy
p
er
s
p
ec
tr
al
im
a
g
in
g
o
f
f
er
s
a
p
o
wer
f
u
l
non
-
d
estru
ctiv
e
alter
n
ativ
e
b
y
ca
p
tu
r
in
g
d
etailed
s
p
ec
tr
al
in
f
o
r
m
atio
n
,
en
ab
lin
g
im
p
r
o
v
ed
ea
r
ly
b
r
u
is
e
d
etec
tio
n
an
d
q
u
ality
ass
ess
m
en
t
[
5
]
–
[
7
]
.
T
h
is
s
tu
d
y
p
r
esen
ts
an
ar
tific
ial
in
tellig
en
ce
(
AI
)
-
b
ased
f
r
am
ew
o
r
k
f
o
r
ap
p
le
b
r
u
is
e
class
if
icatio
n
u
s
in
g
h
y
p
er
s
p
ec
tr
al
im
ag
es to
s
u
p
p
o
r
t e
f
f
icien
t f
r
u
it g
r
ad
in
g
an
d
q
u
ality
co
n
t
r
o
l
p
r
o
ce
s
s
es
[
8
]
–
[
1
1
]
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
em
p
lo
y
s
lin
e
-
s
ca
n
h
y
p
er
s
p
ec
tr
al
im
a
g
in
g
[
1
2
]
with
Gau
s
s
ian
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
3
8
1
-
1
3
8
9
1382
f
ilter
in
g
f
o
r
p
r
e
p
r
o
ce
s
s
in
g
[
1
3
]
,
v
is
u
al
g
eo
m
et
r
y
g
r
o
u
p
(
VG
G
)
-
1
6
f
o
r
f
ea
t
u
r
e
ex
tr
ac
tio
n
[
1
4
]
,
a
n
d
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
class
if
ier
o
p
tim
ized
u
s
in
g
ad
ap
tiv
e
s
an
d
ca
t
s
war
m
o
p
tim
izati
on
(
ASC
S
O)
[
1
5
]
–
[
1
9
]
.
T
h
e
o
p
tim
ized
m
o
d
el
ac
h
ie
v
es
h
ig
h
d
etec
tio
n
ac
cu
r
ac
y
an
d
r
ap
id
class
if
icatio
n
,
d
em
o
n
s
tr
atin
g
its
ef
f
ec
tiv
en
ess
f
o
r
r
ea
l
-
tim
e
ap
p
le
b
r
u
is
e
d
etec
tio
n
a
n
d
waste
r
ed
u
ctio
n
[
2
0
]
.
E
ar
ly
ap
p
le
b
r
u
is
e
d
etec
tio
n
m
eth
o
d
s
r
elied
o
n
R
GB
im
ag
in
g
d
u
e
to
lo
w
co
s
t,
b
u
t
th
ese
tech
n
iq
u
es
ar
e
in
ef
f
ec
tiv
e
f
o
r
d
etec
tin
g
s
u
b
s
u
r
f
ac
e
b
r
u
is
es
an
d
ar
e
h
ig
h
ly
s
en
s
itiv
e
to
lig
h
tin
g
v
ar
iatio
n
s
,
lim
itin
g
in
d
u
s
tr
ial
u
s
e.
T
h
er
m
al
im
ag
in
g
im
p
r
o
v
ed
in
ter
n
al
d
am
ag
e
d
etec
tio
n
b
u
t
s
u
f
f
e
r
ed
f
r
o
m
in
s
tab
ilit
y
u
n
d
er
v
ar
y
in
g
am
b
ien
t
tem
p
er
atu
r
e
s
.
T
o
o
v
er
co
m
e
th
ese
is
s
u
es,
n
ea
r
-
in
f
r
ar
e
d
(
NI
R
)
an
d
h
y
p
er
s
p
ec
tr
al
im
ag
in
g
tech
n
iq
u
es
wer
e
in
tr
o
d
u
ce
d
,
o
f
f
er
in
g
e
n
h
an
ce
d
s
p
ec
tr
al
s
en
s
itiv
ity
f
o
r
n
o
n
-
d
estru
ctiv
e
b
r
u
is
e
d
etec
tio
n
.
T
h
e
s
tu
d
ies
[
2
1
]
,
[
2
2
]
d
em
o
n
s
tr
a
ted
h
ig
h
d
etec
tio
n
ac
cu
r
ac
y
u
s
in
g
h
y
p
er
s
p
ec
tr
al
d
ata
c
o
m
b
in
ed
with
d
ee
p
lear
n
in
g
,
th
o
u
g
h
at
th
e
ex
p
en
s
e
o
f
in
cr
ea
s
ed
co
m
p
u
t
atio
n
al
co
m
p
lex
ity
a
n
d
lim
ited
g
en
er
aliza
tio
n
.
Oth
er
s
tu
d
ies
u
s
in
g
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN
)
an
d
tr
a
d
itio
n
al
class
if
ier
s
ac
h
iev
ed
p
r
o
m
is
in
g
r
esu
lts
b
u
t
r
em
ai
n
ed
d
ataset
-
d
ep
en
d
e
n
t
an
d
s
en
s
itiv
e
to
n
o
is
e
[
2
3
]
,
[
2
4
]
.
A
p
o
r
tab
le
h
y
p
er
s
p
ec
tr
al
s
en
s
in
g
d
ev
ice
ac
h
iev
ed
h
i
g
h
ac
cu
r
ac
y
in
f
r
u
it f
ir
m
n
ess
ass
es
s
m
en
t,
b
u
t its
ef
f
ec
tiv
en
ess
is
lim
ited
f
o
r
co
m
p
le
x
b
r
u
is
e
class
if
icatio
n
in
v
o
lv
in
g
m
u
lti
p
le
ca
teg
o
r
ies
a
n
d
v
ar
y
in
g
s
ev
er
ity
lev
els
[
2
5
]
.
Op
tim
iza
tio
n
-
b
ased
m
eth
o
d
s
s
u
ch
as
s
an
d
ca
t
s
war
m
o
p
tim
izatio
n
(
SC
SO)
im
p
r
o
v
ed
lear
n
in
g
ef
f
icien
cy
b
u
t
s
tr
u
g
g
led
with
h
ig
h
-
d
im
e
n
s
io
n
al
d
ee
p
m
o
d
els
[
2
6
]
.
Ov
er
all,
ex
is
tin
g
ap
p
r
o
ac
h
es
lack
ef
f
ec
tiv
e
s
p
ec
tr
al
d
e
p
en
d
e
n
c
y
m
o
d
elin
g
an
d
r
o
b
u
s
t
o
p
tim
iz
atio
n
.
T
h
ese
lim
itatio
n
s
m
o
ti
v
ate
th
e
p
r
o
p
o
s
ed
ASC
SO
-
L
STM
f
r
am
ewo
r
k
,
wh
ich
ex
p
lo
its
s
eq
u
en
tial
s
p
ec
tr
al
in
f
o
r
m
atio
n
w
h
ile
en
h
an
cin
g
r
o
b
u
s
tn
ess
an
d
co
m
p
u
tat
io
n
al
ef
f
icien
cy
f
o
r
ap
p
le
b
r
u
is
e
class
if
icatio
n
.
2.
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
e
d
s
t
u
d
y
f
o
cu
s
es
o
n
ap
p
le
b
r
u
i
s
e
c
la
s
s
i
f
i
c
at
i
o
n
u
s
i
n
g
h
y
p
er
s
p
e
c
tr
a
l
i
m
a
g
i
n
g
,
a
s
i
l
l
u
s
t
r
a
t
ed
in
F
i
g
u
r
e
1
.
Ap
p
l
e
i
m
a
g
e
s
a
r
e
a
c
q
u
i
r
e
d
u
s
i
n
g
a
h
y
p
er
s
p
e
c
tr
a
l
ca
m
er
a
to
ca
p
tu
r
e
d
e
ta
i
l
e
d
s
p
e
c
tr
a
l
i
n
f
o
r
m
a
t
io
n
a
cr
o
s
s
m
u
l
t
ip
l
e
w
a
v
e
l
en
g
th
s
.
T
h
e
a
cq
u
ir
e
d
im
a
g
e
s
a
r
e
p
r
ep
r
o
c
e
s
s
e
d
u
s
i
n
g
a
G
au
s
s
i
a
n
f
i
l
t
e
r
t
o
r
e
d
u
c
e
n
o
i
s
e
a
n
d
en
h
an
c
e
i
m
a
g
e
q
u
a
l
i
ty
.
D
e
e
p
f
ea
t
u
r
e
s
a
r
e
t
h
en
ex
t
r
a
ct
e
d
u
s
i
n
g
th
e
VG
G
-
1
6
C
N
N
[
2
7
]
,
w
h
i
c
h
e
f
f
e
c
t
i
v
el
y
id
e
n
t
if
i
e
s
d
i
s
c
r
i
m
in
a
t
i
v
e
ch
a
r
a
c
te
r
i
s
t
i
c
s
o
f
h
e
a
l
th
y
a
n
d
b
r
u
i
s
e
d
ap
p
l
e
t
i
s
s
u
e
s
.
T
h
e
s
e
f
e
a
t
u
r
e
s
a
r
e
c
l
a
s
s
i
f
i
ed
u
s
in
g
a
L
S
T
M
n
e
t
w
o
r
k
,
w
h
i
ch
i
s
w
e
l
l
s
u
i
t
e
d
f
o
r
m
o
d
e
l
i
n
g
t
h
e
s
e
q
u
en
t
i
a
l
s
p
e
c
t
r
a
l
i
n
f
o
r
m
a
t
io
n
i
n
h
e
r
en
t
in
h
y
p
e
r
s
p
ec
t
r
a
l
d
a
t
a.
T
o
i
m
p
r
o
v
e
c
l
a
s
s
i
f
i
ca
t
i
o
n
a
c
cu
r
ac
y
a
n
d
c
o
m
p
u
ta
t
i
o
n
a
l
e
f
f
i
c
i
en
cy
,
o
p
t
im
i
z
a
t
io
n
te
c
h
n
i
q
u
e
s
a
r
e
e
m
p
l
o
y
ed
to
au
t
o
m
a
t
i
c
a
l
l
y
t
u
n
e
th
e
L
S
T
M
h
y
p
e
r
p
a
r
am
e
t
e
r
s
.
T
h
i
s
o
p
t
i
m
iz
e
d
f
r
a
m
e
wo
r
k
e
n
ab
l
es
a
c
c
u
r
a
t
e
an
d
e
f
f
i
c
i
en
t
a
p
p
le
b
r
u
i
s
e
d
e
t
e
c
t
io
n
,
s
u
p
p
o
r
t
in
g
im
p
r
o
v
e
d
q
u
a
l
i
ty
c
o
n
tr
o
l
in
ap
p
le
p
r
o
d
u
c
t
i
o
n
an
d
d
i
s
t
r
i
b
u
t
io
n
.
Fig
u
r
e
1
.
Ap
p
le
b
r
u
is
e
class
if
i
ca
tio
n
f
r
o
m
h
y
p
er
s
p
ec
tr
al
im
a
g
es u
s
in
g
an
o
p
tim
ally
co
n
f
ig
u
r
ed
L
STM
m
o
d
el
2
.
1
.
Da
t
a
s
et
des
cr
iptio
n
R
ed
f
u
ji
ap
p
les
wer
e
co
llecte
d
f
r
o
m
a
lo
ca
l
m
ar
k
et
in
Kash
m
ir
,
I
n
d
ia.
Sam
p
les
(
6
–
8
cm
d
iam
eter
)
with
o
u
t
v
is
ib
le
d
e
f
ec
ts
wer
e
c
lean
ed
an
d
s
to
r
ed
at
2
5
°C
an
d
7
5
%
r
elativ
e
h
u
m
id
ity
.
C
o
n
t
r
o
lled
b
r
u
is
es
wer
e
in
d
u
ce
d
b
y
d
r
o
p
p
in
g
a
b
all
f
r
o
m
h
eig
h
ts
o
f
2
0
0
–
5
0
0
m
m
at
eq
u
ato
r
ial
p
o
s
itio
n
s
.
A
to
tal
o
f
1
,
3
4
5
h
y
p
er
s
p
ec
tr
al
s
am
p
les
wer
e
ca
teg
o
r
ized
in
to
f
iv
e
b
r
u
is
e
class
es:
s
m
all
(
2
9
8
)
,
m
ed
iu
m
(
2
8
5
)
,
lar
g
e
(
2
4
4
)
,
s
tem
(
2
7
0
)
,
a
n
d
ca
ly
x
(
2
4
8
)
.
T
h
e
d
ataset
was
d
i
v
id
ed
in
to
tr
ain
in
g
,
v
alid
atio
n
,
an
d
test
in
g
s
u
b
s
ets,
as
d
etailed
in
T
ab
le
1
.
T
h
e
tr
ain
in
g
s
et
was
u
s
ed
f
o
r
m
o
d
el
lear
n
in
g
,
th
e
v
alid
atio
n
s
et
f
o
r
h
y
p
er
p
ar
am
eter
tu
n
in
g
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
u
to
ma
ted
cla
s
s
ifica
tio
n
o
f a
p
p
le
b
r
u
is
es fr
o
m
h
yp
ers
p
ec
tr
a
l
ima
g
es
…
(
P
ed
d
ir
ed
d
y
V
en
ka
tesw
a
r
a
R
ed
d
y
)
1383
ea
r
ly
s
to
p
p
in
g
,
a
n
d
th
e
test
s
et
f
o
r
u
n
b
iased
p
er
f
o
r
m
a
n
ce
ev
alu
atio
n
.
Stra
tifie
d
s
am
p
lin
g
was
ap
p
lied
to
p
r
eser
v
e
class
b
alan
ce
ac
r
o
s
s
all
s
u
b
s
ets.
T
ab
le
1
.
Data
s
et
p
ar
titi
o
n
in
g
f
o
r
tr
ain
in
g
,
v
alid
atio
n
,
an
d
test
in
g
p
h
ases
D
a
t
a
s
e
t
p
a
r
t
i
t
i
o
n
P
e
r
c
e
n
t
a
g
e
(
%)
N
u
mb
e
r
o
f
s
a
m
p
l
e
s
Tr
a
i
n
i
n
g
se
t
70
9
4
2
V
a
l
i
d
a
t
i
o
n
se
t
15
2
0
2
Te
st
i
n
g
s
e
t
15
2
0
1
To
t
a
l
1
0
0
1
,
3
4
5
2
.
1
.
1
.
Cro
s
s
-
v
a
lid
a
t
io
n str
a
t
e
g
y
T
o
en
h
an
ce
m
o
d
el
r
eliab
ilit
y
an
d
g
e
n
er
aliza
b
ilit
y
,
a
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
was
ap
p
lied
i
n
ad
d
itio
n
to
th
e
f
ix
ed
d
ataset
s
p
lit.
T
h
e
tr
ain
in
g
s
et
(
7
0
%)
was
p
ar
titi
o
n
ed
i
n
to
f
iv
e
f
o
ld
s
;
in
ea
ch
i
ter
atio
n
,
f
o
u
r
f
o
ld
s
wer
e
u
s
ed
f
o
r
tr
ain
in
g
a
n
d
o
n
e
f
o
r
v
alid
atio
n
.
T
h
is
p
r
o
ce
s
s
was
r
ep
ea
ted
f
iv
e
tim
es,
an
d
t
h
e
f
in
al
r
esu
lts
wer
e
o
b
tain
ed
b
y
av
er
a
g
in
g
p
er
f
o
r
m
an
ce
ac
r
o
s
s
all
f
o
ld
s
.
T
h
is
s
tr
ateg
y
im
p
r
o
v
es
r
o
b
u
s
tn
ess
,
r
ed
u
ce
s
o
v
er
f
itti
n
g
,
an
d
p
r
o
v
id
es a
r
eliab
le
esti
m
at
e
o
f
p
e
r
f
o
r
m
an
ce
o
n
u
n
s
ee
n
d
ata.
2
.
2
.
H
y
persp
ec
t
ra
l im
a
g
ing
s
y
s
t
em
T
h
e
g
r
ap
h
ic
s
h
o
ws
a
h
y
p
er
s
p
ec
tr
al
im
ag
in
g
s
y
s
tem
with
s
lit
w
id
th
s
o
f
3
0
μ
m
an
d
a
s
p
ec
tr
al
r
eso
lu
tio
n
o
f
ab
o
u
t
2
.
8
n
m
.
An
im
ag
in
g
s
p
ec
tr
o
g
r
a
p
h
(
I
m
Sp
ec
to
r
VNI
R
-
V1
0
E
-
E
MC
C
D)
with
a
s
tan
d
ar
d
2
3
m
m
C
-
m
o
u
n
t
zo
o
m
len
s
,
a
150
-
W
att
h
alo
g
en
lam
p
with
two
-
lin
e
lig
h
tin
g
f
ib
er
s
to
en
s
u
r
e
u
n
if
o
r
m
v
is
ib
le
n
ea
r
-
in
f
r
ar
ed
(
Vis
-
NI
R
)
illu
m
in
atio
n
,
an
elec
tr
o
n
-
m
u
ltip
ly
in
g
ch
ar
g
e
-
co
u
p
led
d
ev
ice
(
E
MCC
D
)
ca
m
er
a
(
An
d
o
r
L
u
ca
E
MCC
D
DL
-
6
0
4
M)
,
a
s
am
p
le
d
is
p
lace
m
en
t
p
latf
o
r
m
,
an
d
a
c
o
m
p
u
ter
(
Dell
E
6
5
2
0
,
I
n
tel
C
o
r
e
i5
-
2
5
2
0
M@2
.
5
GH
z,
8
GB
R
AM
)
m
ak
e
u
p
th
is
s
etu
p
.
T
h
e
r
eso
lu
tio
n
o
f
th
e
im
ag
es
th
at
wer
e
s
h
o
t
was
1
0
0
4
b
y
1
0
0
2
.
T
o
en
h
a
n
ce
im
ag
e
q
u
ality
,
o
n
ly
d
ata
in
th
e
wav
ele
n
g
th
r
an
g
e
o
f
4
5
0
to
1
,
0
0
0
n
m
wer
e
r
etain
ed
;
d
ata
o
u
ts
id
e
o
f
th
is
r
an
g
e
wer
e
r
em
o
v
ed
d
u
e
to
th
e
lim
its
o
f
th
e
ch
a
r
g
e
-
co
u
p
led
d
ev
ice
(
CCD
)
d
etec
to
r
.
2
.
3
.
P
re
-
pro
ce
s
s
ing
Hy
p
er
s
p
ec
tr
al
d
ata
wer
e
p
r
ep
r
o
ce
s
s
ed
u
s
in
g
s
tan
d
ar
d
n
o
r
m
al
v
ar
iate
(
SNV)
tr
an
s
f
o
r
m
atio
n
to
r
ed
u
ce
illu
m
in
atio
n
an
d
s
ca
tter
in
g
e
f
f
ec
ts
.
T
h
is
was
f
o
llo
wed
b
y
v
ec
to
r
an
d
m
in
–
m
a
x
n
o
r
m
aliza
tio
n
to
en
s
u
r
e
u
n
if
o
r
m
f
ea
tu
r
e
s
ca
lin
g
.
T
h
is
p
r
ep
r
o
ce
s
s
in
g
en
h
an
ce
d
s
p
ec
tr
al
d
is
cr
im
in
atio
n
,
im
p
r
o
v
ed
t
r
ain
in
g
s
tab
ilit
y
,
an
d
s
ig
n
if
ican
tly
b
o
o
s
ted
th
e
class
if
icatio
n
p
er
f
o
r
m
an
ce
an
d
g
en
er
aliza
tio
n
o
f
th
e
ASC
SO
-
L
S
T
M
m
o
d
el.
2
.
4
.
F
e
a
t
ure
ex
t
r
a
ct
io
n
T
h
e
f
ea
tu
r
e
ex
tr
ac
tio
n
p
r
o
c
ess
in
ap
p
le
b
r
u
is
e
cla
s
s
if
i
ca
tio
n
in
v
o
lv
es
u
tili
zin
g
th
e
VGG
-
16
ar
ch
itectu
r
e,
a
C
NN
d
ev
elo
p
e
d
b
y
th
e
VGG
[
2
8
]
.
VGG
-
1
6
is
well
ac
claim
ed
f
o
r
its
v
er
s
a
tili
ty
in
a
v
ar
iety
o
f
co
m
p
u
ter
v
is
io
n
task
s
,
in
clu
d
in
g
f
ea
t
u
r
e
ex
tr
a
ctio
n
an
d
im
ag
e
ca
teg
o
r
izatio
n
.
Fu
n
d
a
m
en
tally
,
VGG
-
1
6
co
m
p
r
is
es
1
6
weig
h
t
lay
er
s
,
p
r
im
ar
ily
co
n
s
is
tin
g
o
f
1
3
co
n
v
o
lu
tio
n
al
lay
er
s
task
ed
with
ex
tr
ac
tin
g
d
etailed
p
atter
n
s
an
d
d
ata
f
r
o
m
in
p
u
t
im
ag
es.
B
y
u
s
in
g
3
×
3
f
ilter
s
with
a
s
tr
id
e
o
f
1
a
n
d
"sa
m
e"
p
ad
d
i
n
g
,
th
ese
co
n
v
o
l
u
tio
n
al
lay
er
s
p
r
eser
v
e
im
p
o
r
tan
t
s
p
atial
in
f
o
r
m
at
io
n
ac
r
o
s
s
th
e
n
etwo
r
k
.
Ad
d
itio
n
ally
,
VGG
-
16
in
co
r
p
o
r
ates
2
×
2
m
ax
-
p
o
o
lin
g
lay
er
s
af
ter
ea
ch
b
lo
c
k
o
f
c
o
n
v
o
lu
ti
o
n
al
lay
er
s
at
a
s
tr
id
e
o
f
2
.
T
h
ese
lay
er
s
d
o
wn
s
am
p
le
th
e
f
ea
tu
r
e
m
ap
s
,
g
r
ad
u
ally
r
e
d
u
cin
g
th
eir
s
p
atial
s
ize
wh
ile
au
g
m
en
tin
g
th
eir
d
ep
th
.
T
h
is
ar
ch
itectu
r
al
d
esig
n
allo
ws
t
h
e
n
etwo
r
k
to
ef
f
ec
tiv
ely
ca
p
tu
r
e
h
ier
a
r
ch
ical
f
ea
tu
r
es,
e
s
s
en
tial
f
o
r
ac
cu
r
ate
ap
p
le
b
r
u
is
e
class
if
icatio
n
.
2
.
5
.
L
o
ng
s
ho
rt
-
t
er
m
m
emo
ry
L
STM
n
etwo
r
k
s
ar
e
em
p
l
o
y
e
d
to
class
if
y
ap
p
le
b
r
u
is
es
in
to
s
m
all,
m
ed
iu
m
,
lar
g
e,
s
tem
,
an
d
ca
ly
x
ca
teg
o
r
ies
b
y
m
o
d
elin
g
s
eq
u
en
tial
d
ep
en
d
en
cies
in
h
y
p
er
s
p
ec
tr
al
f
ea
tu
r
es
e
x
tr
ac
ted
u
s
in
g
VGG
-
1
6
.
T
h
e
L
STM
n
etwo
r
k
wo
u
ld
th
e
n
lear
n
to
an
aly
ze
th
ese
f
ea
tu
r
es
o
v
er
tim
e,
co
n
s
id
er
in
g
th
e
s
eq
u
en
tial
n
atu
r
e
o
f
th
e
d
ata
[
2
9
]
.
B
y
lear
n
in
g
tem
p
o
r
al
p
atter
n
s
f
r
o
m
la
b
eled
d
at
a,
th
e
L
STM
ef
f
ec
tiv
ely
ca
p
t
u
r
es
b
r
u
is
e
-
r
elate
d
v
ar
iatio
n
s
,
en
a
b
lin
g
a
cc
u
r
ate
an
d
r
o
b
u
s
t
class
if
icatio
n
f
o
r
ap
p
le
q
u
ality
ass
ess
m
en
t.
T
h
e
m
ath
em
atica
l
r
ep
r
esen
tatio
n
o
f
th
e
L
STM
n
etwo
r
k
is
as (
1
)
to
(
5
)
.
=
(
+
ℎ
ℎ
−
1
+
)
(
1
)
=
(
+
ℎ
ℎ
−
1
+
)
(
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
3
8
1
-
1
3
8
9
1384
̃
=
ℎ
(
+
ℎ
ℎ
−
1
+
)
(
3
)
=
(
+
ℎ
ℎ
−
1
+
)
(
4
)
ℎ
=
⋅
ℎ
(
)
(
5
)
T
h
e
in
p
u
t
v
ec
to
r
at
tim
e
is
r
ep
r
esen
ted
b
y
in
th
is
eq
u
atio
n
,
th
e
h
id
d
e
n
s
tate
at
tim
e
t
is
d
en
o
ted
by
ℎ
,
th
e
in
p
u
t,
f
o
r
g
et,
an
d
o
u
t
p
u
t
g
ates
a
r
e
r
ep
r
esen
ted
b
y
,
,
an
d
,
r
esp
ec
tiv
ely
,
an
d
th
e
m
em
o
r
y
ce
ll
is
in
d
icate
d
b
y
̃
.
T
h
e
m
o
d
el'
s
tr
ain
ab
le
p
ar
am
eter
s
co
n
s
is
t
o
f
weig
h
ts
an
d
b
iases
b
.
T
h
e
L
STM
ce
ll
(
alo
n
g
with
th
e
e
q
u
atio
n
s
)
with
a
f
o
r
g
et
g
ate
ar
e
s
h
o
wn
in
Fig
u
r
e
2
.
Hy
p
er
p
a
r
am
eter
o
p
tim
izatio
n
p
lay
s
a
cr
u
cial
r
o
le
in
im
p
r
o
v
in
g
L
S
T
M
-
b
ased
ap
p
le
b
r
u
is
e
class
if
icatio
n
b
y
tu
n
in
g
k
ey
p
a
r
a
m
eter
s
s
u
ch
as
t
h
e
n
u
m
b
er
o
f
la
y
er
s
,
h
id
d
en
u
n
its
,
d
r
o
p
o
u
t,
an
d
lear
n
in
g
r
at
e.
Au
to
m
ated
o
p
tim
izatio
n
m
eth
o
d
s
r
ed
u
ce
th
e
co
m
p
lex
ity
a
n
d
tim
e
ass
o
ciate
d
with
m
an
u
al
tu
n
in
g
,
p
r
ev
e
n
t
o
v
er
f
itti
n
g
o
r
u
n
d
er
f
itti
n
g
,
an
d
en
a
b
le
ef
f
icien
t
u
s
e
o
f
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
wh
ile
en
h
an
cin
g
m
o
d
el
g
en
e
r
aliza
tio
n
an
d
p
er
f
o
r
m
an
ce
.
Fig
u
r
e
2
.
L
STM
f
o
r
ap
p
le
b
r
u
i
s
es c
las
s
if
icatio
n
2
.
6
.
Sa
nd
ca
t
s
wa
rm
o
pti
m
iza
t
io
n
T
h
e
b
eh
av
io
r
o
f
s
an
d
ca
ts
in
th
eir
n
atu
r
al
h
ab
itat
s
er
v
ed
as
th
e
m
o
d
el
f
o
r
th
e
SC
SO
a
lg
o
r
ith
m
as
s
h
o
wn
in
Alg
o
r
ith
m
1
.
T
h
e
t
wo
m
ain
b
eh
av
io
r
s
ex
h
i
b
ited
b
y
th
ese
ca
ts
ar
e
ac
tiv
ely
s
ea
r
ch
in
g
f
o
r
p
r
ey
an
d
s
u
cc
ess
f
u
lly
o
b
tain
in
g
it.
T
h
e
p
r
o
g
r
a
m
m
ed
em
u
lates
th
e
ex
c
ep
tio
n
al
lo
w
-
f
r
eq
u
en
cy
s
o
u
n
d
d
etec
tio
n
ca
p
ac
ity
o
f
th
e
s
an
d
ca
t,
a
u
n
iq
u
e
ch
ar
ac
ter
is
tic
th
at
m
ak
es
ef
f
icien
t
p
r
ey
lo
ca
tio
n
p
o
s
s
ib
le,
b
o
th
a
b
o
v
e
an
d
u
n
d
e
r
th
e
s
u
r
f
ac
e.
T
h
is
ex
ce
p
tio
n
al
s
k
ill em
p
o
wer
s
s
an
d
ca
ts
to
s
wif
tly
p
in
p
o
i
n
t a
n
d
s
eize
th
eir
p
r
ey
[
3
0
]
.
Alg
o
r
ith
m
1
.
S
an
d
ca
t swar
m
o
p
tim
izatio
n
alg
o
r
ith
m
1
:
I
n
itialize
p
o
p
u
latio
n
u
s
in
g
o
p
p
o
s
itio
n
-
b
ased
lea
r
n
in
g
.
2
:
E
v
alu
ate
f
itn
ess
o
f
ea
c
h
s
o
l
u
tio
n
.
3
:
I
d
en
tif
y
th
e
b
est s
o
lu
tio
n
.
4
:
Fo
r
t =
1
to
m
ax
im
u
m
iter
atio
n
s
d
o
a.
Fo
r
ea
ch
s
an
d
ca
t
d
o
i.
Select
m
o
v
em
en
t d
ir
ec
tio
n
u
s
i
n
g
r
o
u
lette
wh
ee
l selec
tio
n
.
ii.
Up
d
ate
p
o
s
itio
n
u
s
in
g
ex
p
l
o
r
a
tio
n
o
r
e
x
p
lo
itatio
n
s
tr
ateg
y
.
iii.
Ap
p
ly
c
au
c
h
y
m
u
tatio
n
to
e
n
h
an
ce
d
iv
er
s
ity
.
iv
.
E
n
f
o
r
ce
b
o
u
n
d
ar
y
co
n
s
tr
ain
ts
.
b.
E
n
d
f
o
r
c.
Up
d
ate
th
e
g
lo
b
al
b
est s
o
lu
tio
n
.
5
:
E
n
d
f
o
r
6
:
R
etu
r
n
th
e
b
est s
o
lu
tio
n
.
2
.
6
.
1
.
Ca
uchy
m
uta
t
io
n
A
wid
er
m
u
tatio
n
s
ca
le
is
in
tr
o
d
u
ce
d
u
tili
zin
g
t
h
e
c
au
c
h
y
d
is
tr
ib
u
tio
n
.
T
h
e
g
e
n
er
al
f
o
r
m
u
la
f
o
r
its
p
r
o
b
a
b
ilit
y
d
en
s
ity
f
u
n
ctio
n
is
g
iv
en
b
y
(
6
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
u
to
ma
ted
cla
s
s
ifica
tio
n
o
f a
p
p
le
b
r
u
is
es fr
o
m
h
yp
ers
p
ec
tr
a
l
ima
g
es
…
(
P
ed
d
ir
ed
d
y
V
en
ka
tesw
a
r
a
R
ed
d
y
)
1385
(
)
=
1
(
1
+
(
(
−
)
⁄
)
2
)
(
6
)
T
h
e
r
an
d
o
m
v
ar
ia
b
le
=
(
)
h
as
a
u
n
if
o
r
m
d
is
tr
ib
u
tio
n
in
t
h
e
r
a
n
g
e
[
0
,
1
]
wh
en
a
r
an
d
o
m
v
ar
iab
le
h
as
a
d
is
tr
ib
u
tio
n
f
u
n
ctio
n
F
.
T
h
is
is
th
e
ca
lcu
latio
n
p
r
o
ce
s
s
f
o
r
a
c
au
ch
y
r
a
n
d
o
m
v
ar
ia
b
le.
I
n
th
e
ev
en
t
wh
en
F
is
r
ev
er
s
ed
,
th
e
r
an
d
o
m
v
ar
iab
le
ca
n
th
u
s
em
p
lo
y
a
u
n
if
o
r
m
d
en
s
ity
to
r
esem
b
le
r
an
d
o
m
v
ar
iab
le
b
ec
au
s
e
=
−
1
(
)
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
is
ev
alu
ated
u
s
in
g
co
m
p
r
eh
en
s
iv
e
p
e
r
f
o
r
m
a
n
ce
m
etr
ics,
in
clu
d
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
e
ca
ll,
F1
-
s
co
r
e,
f
al
s
e
d
is
co
v
er
y
r
ate
(
FDR
)
,
f
als
e
p
o
s
itiv
e
r
ate
(
FPR
)
,
f
alse
n
eg
ativ
e
r
ate
(
FNR
)
,
Ma
tth
ews
co
r
r
elatio
n
co
ef
f
i
cien
t
(
MCC
)
,
n
eg
ativ
e
p
r
ed
ictiv
e
v
alu
e
(
NPV
)
,
an
d
s
p
ec
if
icity
.
R
ec
eiv
er
o
p
er
atin
g
c
h
ar
ac
ter
is
tic
(
R
OC
)
cu
r
v
es,
co
n
f
u
s
io
n
m
atr
ices,
an
d
co
n
v
er
g
e
n
ce
an
aly
s
is
ar
e
also
u
s
ed
to
ass
es
s
o
p
tim
izatio
n
e
f
f
ec
tiv
en
ess
.
C
o
m
p
ar
ativ
e
r
esu
lts
ac
r
o
s
s
L
STM
v
ar
ian
ts
an
d
C
NN
m
o
d
els
(
VGG
-
1
6
an
d
R
esNet
-
5
0
)
ar
e
r
ep
o
r
ted
u
s
in
g
tr
u
e
p
o
s
itiv
e,
tr
u
e
n
eg
ativ
e
,
f
alse
p
o
s
itiv
e,
an
d
f
alse
n
e
g
ativ
e
m
ea
s
u
r
es
to
q
u
an
tify
class
if
icatio
n
p
er
f
o
r
m
an
ce
in
ap
p
le
b
r
u
is
e
d
etec
tio
n
.
3
.
1
.
P
er
f
o
r
m
a
nce
ev
a
lua
t
io
n m
a
t
rice
s
Per
f
o
r
m
an
ce
ev
al
u
atio
n
m
etr
i
cs
q
u
an
titativ
ely
ass
es
s
th
e
ef
f
ec
tiv
en
ess
o
f
ap
p
le
b
r
u
is
e
class
if
icatio
n
f
r
o
m
h
y
p
er
s
p
ec
tr
al
im
a
g
es
.
T
o
g
eth
er
with
th
e
co
n
f
u
s
io
n
m
atr
ix
,
h
elp
id
e
n
tify
m
o
d
el
s
tr
e
n
g
th
s
,
wea
k
n
ess
es,
an
d
co
m
p
ar
ativ
e
p
er
f
o
r
m
an
ce
,
en
s
u
r
in
g
ac
cu
r
ate
a
n
d
r
eliab
le
d
etec
tio
n
.
T
h
e
p
e
r
f
o
r
m
an
ce
r
esu
lts
o
f
s
u
g
g
este
d
ASC
S
O
m
eth
o
d
f
o
r
d
esig
n
in
g
L
STM
f
o
r
th
e
ca
teg
o
r
izatio
n
o
f
ap
p
le
b
r
u
is
in
g
f
r
o
m
h
y
p
er
s
p
ec
tr
al
im
ag
es
s
h
o
w
s
tr
o
n
g
p
r
e
d
ictiv
e
ca
p
ab
il
ity
,
ac
h
iev
in
g
a
n
ac
cu
r
ac
y
o
f
9
7
.
5
%,
FDR
o
f
2
.
9
%,
F1
-
s
co
r
e
o
f
9
1
.
4
%,
FNR
o
f
2
%,
FP
R
o
f
3
%,
MCC
o
f
9
5
%,
NPV
o
f
9
8
%,
p
r
ec
is
io
n
o
f
9
7
%,
r
ec
all
o
f
9
8
%,
a
n
d
s
p
ec
if
ic
ity
o
f
9
7
%
.
Fig
u
r
e
3
s
h
o
ws
th
e
p
er
f
o
r
m
an
ce
co
m
p
ar
is
o
n
o
f
th
e
em
p
lo
y
e
d
m
o
d
els
f
o
r
ap
p
le
b
r
u
is
e
clas
s
if
icatio
n
.
Fig
u
r
e
3
(
a)
p
r
esen
ts
th
e
ac
cu
r
ac
y
co
m
p
ar
is
o
n
am
o
n
g
t
h
e
m
o
d
els.
Fig
u
r
e
3
(
b
)
s
h
o
ws
th
e
p
r
ec
is
io
n
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
els,
wh
ile
Fig
u
r
e
3
(
c)
illu
s
tr
ate
s
th
e
r
ec
all
r
esu
lts
.
T
h
e
L
STM
–
ASC
SO
m
o
d
el
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
s
all
o
th
er
s
,
ac
h
iev
in
g
th
e
h
ig
h
est
ac
cu
r
ac
y
(
0
.
9
8
)
,
p
r
ec
is
io
n
(
0
.
9
7
)
,
r
ec
all
(
0
.
9
8
)
,
s
p
ec
if
icity
(
0
.
9
7
)
,
MCC
(
0
.
9
5
)
,
an
d
NPV
(
0
.
9
8
)
,
wh
ile
r
e
co
r
d
in
g
th
e
lo
west
FDR
(
0
.
0
3
)
,
FP
R
(
0
.
0
3
)
,
an
d
FNR
(
0
.
0
2
)
.
T
h
ese
r
esu
lts
d
em
o
n
s
tr
ate
its
s
u
p
er
io
r
r
eliab
ilit
y
in
m
in
im
izin
g
f
alse
d
etec
tio
n
s
an
d
m
is
s
ed
b
r
u
is
es.
Ov
er
all,
th
e
ad
ap
ti
v
e
ASC
SO
-
b
ased
o
p
tim
izati
o
n
s
ig
n
if
ican
tly
en
h
an
ce
s
L
STM
p
er
f
o
r
m
a
n
ce
co
m
p
ar
ed
to
L
STM
-
SC
SO,
L
STM
-
C
SO,
b
aselin
e
L
STM
,
a
n
d
C
NN
-
b
ased
m
o
d
els
(
VGG
-
1
6
an
d
R
esNet
-
5
0
)
.
T
ab
le
2
s
h
o
ws
th
at
t
h
e
p
r
o
p
o
s
ed
ASC
SO
–
L
STM
o
u
tp
er
f
o
r
m
s
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
r
an
d
o
m
f
o
r
est
(
R
F),
VGG
-
1
6
,
R
esNet
-
5
0
,
a
n
d
E
f
f
icien
tNet
-
B
0
,
ac
h
iev
in
g
t
h
e
h
ig
h
est
ac
cu
r
ac
y
(
0
.
9
8
)
,
p
r
ec
is
io
n
(
0
.
9
7
)
,
an
d
r
ec
all
(
0
.
9
8
)
,
co
n
f
ir
m
in
g
its
s
u
p
er
io
r
s
p
ec
tr
al
–
tem
p
o
r
al
f
e
atu
r
e
lear
n
i
n
g
a
n
d
r
o
b
u
s
tn
ess
f
o
r
h
y
p
e
r
s
p
ec
tr
al
ap
p
le
b
r
u
is
e
class
if
icatio
n
.
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
E
v
alu
atin
g
t
h
e
p
er
f
o
r
m
an
ce
o
f
em
p
lo
y
ed
m
o
d
els f
o
r
class
if
y
in
g
ap
p
le
b
r
u
is
es
of
(
a)
ac
cu
r
ac
y
,
(
b
)
p
r
ec
is
io
n
,
an
d
(
c)
r
ec
all
T
ab
le
2.
Per
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
o
f
b
aselin
e
m
o
d
els an
d
p
r
o
p
o
s
ed
ASC
SO
–
L
STM
M
o
d
e
l
LSTM
–
A
S
C
S
O
LSTM
–
S
C
S
O
LSTM
–
C
S
O
LSTM
S
V
M
RF
VGG
-
16
Ef
f
i
c
i
e
n
t
N
e
t
-
B0
R
e
sN
e
t
-
50
A
c
c
u
r
a
c
y
0
.
9
8
0
.
9
5
0
.
9
4
0
.
9
1
0
.
8
9
0
.
9
0
0
.
9
0
0
.
9
2
0
.
8
8
P
r
e
c
i
s
i
o
n
0
.
9
7
0
.
9
5
0
.
9
4
0
.
9
0
0
.
8
8
0
.
8
9
0
.
8
9
0
.
9
1
0
.
8
8
R
e
c
a
l
l
0
.
9
8
0
.
9
4
0
.
9
4
0
.
9
2
0
.
8
7
0
.
8
8
0
.
9
0
0
.
9
1
0
.
8
8
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
3
8
1
-
1
3
8
9
1386
3
.
2
.
Co
nfusi
o
n m
a
t
rix
T
h
e
co
n
f
u
s
io
n
m
atr
ix
an
aly
s
is
is
p
r
esen
ted
in
Fig
u
r
e
4
,
s
h
o
win
g
th
e
p
er
f
o
r
m
an
ce
o
f
d
if
f
er
en
t
tech
n
iq
u
es.
T
h
e
p
r
o
p
o
s
ed
L
S
T
M
–
ASC
SO
m
o
d
el
ac
h
iev
es
th
e
b
est
class
-
wis
e
d
is
cr
im
in
atio
n
with
m
in
im
al
m
is
class
if
icatio
n
ac
r
o
s
s
all
ap
p
le
b
r
u
is
e
ca
teg
o
r
ies
(
Fig
u
r
e
4
(
a)
.
I
t
o
u
tp
e
r
f
o
r
m
s
th
e
o
p
tim
ized
L
STM
v
ar
ian
ts
,
in
clu
d
in
g
L
STM
–
SC
SO a
n
d
L
STM
–
C
SO (
F
ig
u
r
es 4
(
b
)
an
d
4
(
c)
)
,
as we
ll a
s
th
e
b
aselin
e
L
STM
(
Fig
u
r
e
4
(
d
)
)
an
d
C
NN
-
b
ased
m
o
d
els,
VGG
-
1
6
(
Fig
u
r
e
4
(
e)
)
a
n
d
R
esNet
-
5
0
(
Fig
u
r
e
4
(
f
)
)
,
wh
ich
ex
h
ib
i
t
h
ig
h
er
c
r
o
s
s
-
class
co
n
f
u
s
io
n
d
u
e
to
wea
k
er
s
p
ec
t
r
al
–
tem
p
o
r
al
m
o
d
elin
g
.
(
a)
(
b
)
(
c)
(
d
)
(
e)
(f)
Fig
u
r
e
4
.
C
o
n
f
u
s
io
n
m
atr
i
x
ac
r
o
s
s
d
if
f
er
en
t te
ch
n
iq
u
es
of
(
a)
L
STM
-
ASC
S
O,
(
b
)
L
STM
-
SC
SO,
(
c)
L
STM
-
C
SO,
(
d
)
L
STM
,
(
e
)
VGG
-
1
6
,
an
d
(
f
)
R
esNet
-
50
3
.
3
.
Rec
eiv
er
o
pera
t
ing
cha
r
a
ct
er
is
t
ic
a
nd
c
o
nv
er
g
ence
g
ra
ph
Fig
u
r
e
5
p
r
esen
ts
th
e
R
OC
cu
r
v
e,
illu
s
tr
atin
g
th
e
m
o
d
el
’
s
ab
ilit
y
to
d
is
tin
g
u
is
h
b
etwe
en
b
r
u
is
ed
an
d
non
-
b
r
u
is
ed
ap
p
les
b
y
p
lo
ttin
g
th
e
tr
u
e
p
o
s
itiv
e
r
ate
ag
ain
s
t
t
h
e
FPR
ac
r
o
s
s
d
if
f
er
en
t
th
r
esh
o
ld
s
.
T
h
is
an
aly
s
is
r
ef
lects
th
e
class
if
icatio
n
ef
f
ec
tiv
en
ess
an
d
th
e
tr
ad
e
-
o
f
f
b
e
twee
n
s
en
s
itiv
ity
an
d
s
p
ec
if
icity
.
Fig
u
r
e
6
s
h
o
ws
th
e
co
n
v
er
g
e
n
ce
b
eh
av
i
o
r
o
f
th
e
o
p
tim
izatio
n
alg
o
r
ith
m
d
u
r
in
g
L
STM
t
r
ain
in
g
.
T
h
e
cu
r
v
e
illu
s
tr
ates
ch
an
g
es
in
p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
iter
atio
n
s
,
in
d
icatin
g
co
n
v
er
g
e
n
ce
s
p
ee
d
an
d
tr
ain
in
g
s
tab
ilit
y
.
A
s
m
o
o
th
,
s
tab
le
tr
en
d
f
o
r
th
e
p
r
o
p
o
s
ed
m
eth
o
d
c
o
n
f
i
r
m
s
ef
f
ec
tiv
e
o
p
tim
izatio
n
,
wh
ile
f
lu
ctu
atio
n
s
m
ay
s
u
g
g
est co
n
v
er
g
en
ce
is
s
u
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
u
to
ma
ted
cla
s
s
ifica
tio
n
o
f a
p
p
le
b
r
u
is
es fr
o
m
h
yp
ers
p
ec
tr
a
l
ima
g
es
…
(
P
ed
d
ir
ed
d
y
V
en
ka
tesw
a
r
a
R
ed
d
y
)
1387
Fig
u
r
e
5
.
R
OC
cu
r
v
e
Fig
u
r
e
6
.
C
o
n
v
er
g
e
n
ce
g
r
a
p
h
4.
CO
NCLU
SI
O
N
T
h
is
r
esear
ch
in
v
esti
g
ated
t
h
e
au
to
m
ated
class
if
icatio
n
o
f
a
p
p
le
b
r
u
is
es
u
s
in
g
h
y
p
er
s
p
ec
tr
al
im
ag
in
g
co
m
b
in
ed
with
a
n
ad
ap
tiv
el
y
o
p
tim
ized
L
STM
n
etwo
r
k
.
A
n
o
v
el
o
p
tim
izatio
n
s
tr
ateg
y
,
ASC
SO,
was
in
tr
o
d
u
ce
d
to
tu
n
e
th
e
L
STM
h
y
p
er
p
ar
am
eter
s
ef
f
ec
tiv
ely
.
E
x
ten
s
iv
e
ex
p
er
im
en
tal
ev
alu
atio
n
u
s
in
g
s
tatis
t
ical
an
d
class
if
icatio
n
m
etr
ics
s
u
ch
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
al
l,
MCC
,
s
p
ec
if
icity
,
an
d
er
r
o
r
r
ates
d
em
o
n
s
tr
ated
th
at
th
e
p
r
o
p
o
s
ed
L
STM
-
ASC
S
O
f
r
am
ewo
r
k
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
e
d
tr
ad
itio
n
al
L
STM
m
o
d
els,
co
n
v
en
tio
n
al
o
p
tim
izatio
n
ap
p
r
o
ac
h
es,
an
d
C
NN
-
b
ased
ar
ch
itectu
r
es
in
clu
d
in
g
VGG
-
1
6
a
n
d
R
esNet
-
5
0
.
T
h
e
in
co
r
p
o
r
atio
n
o
f
a
d
ap
tiv
e
m
ec
h
an
is
m
s
,
n
am
ely
o
p
p
o
s
itio
n
-
b
ased
lear
n
in
g
an
d
C
au
ch
y
d
is
tr
ib
u
tio
n
,
en
h
a
n
ce
d
th
e
ex
p
lo
r
ati
o
n
-
ex
p
lo
itatio
n
b
alan
ce
o
f
th
e
o
p
tim
izatio
n
p
r
o
ce
s
s
,
lead
in
g
to
im
p
r
o
v
e
d
co
n
v
er
g
e
n
ce
b
eh
a
v
io
r
an
d
b
etter
m
o
d
el
g
en
e
r
aliza
tio
n
.
Ad
d
itio
n
ally
,
t
h
e
an
aly
s
is
o
f
R
OC
ch
ar
ac
ter
is
tic
s
an
d
co
n
v
er
g
en
ce
b
eh
av
io
r
p
r
o
v
id
e
d
f
u
r
th
er
ev
id
e
n
ce
o
f
t
h
e
s
tab
ilit
y
an
d
lear
n
in
g
e
f
f
ici
en
cy
o
f
th
e
p
r
o
p
o
s
ed
f
r
am
ew
o
r
k
.
T
h
e
o
u
tco
m
es
o
f
th
is
s
tu
d
y
d
em
o
n
s
tr
ate
th
e
p
o
ten
tial
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
to
s
u
p
p
o
r
t
r
eliab
le
an
d
a
cc
u
r
ate
f
r
u
it
q
u
ality
ass
es
s
m
en
t
in
p
o
s
t
-
h
ar
v
est
p
r
o
ce
s
s
in
g
en
v
ir
o
n
m
en
ts
.
Fu
tu
r
e
wo
r
k
will
f
o
cu
s
o
n
ex
ten
d
in
g
th
e
m
o
d
el
to
o
th
er
f
r
u
it
v
a
r
ieties,
r
ed
u
cin
g
c
o
m
p
u
tatio
n
al
c
o
m
p
lex
ity
f
o
r
r
ea
l
-
tim
e
in
d
u
s
tr
ial
d
ep
lo
y
m
en
t,
an
d
ex
p
lo
r
in
g
ad
v
an
ce
d
f
ea
tu
r
e
s
elec
tio
n
tec
h
n
iq
u
es to
f
u
r
th
e
r
im
p
r
o
v
e
d
et
ec
tio
n
ac
cu
r
ac
y
an
d
s
y
s
tem
s
c
alab
ilit
y
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
d
ec
lar
e
th
at
n
o
f
in
an
cial
s
u
p
p
o
r
t
was
r
ec
eiv
ed
f
o
r
th
e
r
esear
ch
,
au
t
h
o
r
s
h
ip
,
an
d
/o
r
p
u
b
licatio
n
o
f
th
is
ar
ticle.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
3
8
1
-
1
3
8
9
1388
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
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Ped
d
ir
ed
d
y
Ven
k
ateswar
a
R
ed
d
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
A
l
a
g
u
c
h
a
m
y
P
a
r
i
v
a
z
h
a
g
a
n
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
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
th
o
r
d
ec
lar
es th
at
th
e
r
e
is
n
o
co
n
f
lict o
f
in
ter
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
i
n
g
s
o
f
th
is
s
tu
d
y
ar
e
n
o
t
p
u
b
licly
av
ailab
le
b
u
t
ca
n
b
e
p
r
o
v
i
d
ed
b
y
t
h
e
co
r
r
esp
o
n
d
in
g
au
th
o
r
,
[
PVR
]
,
u
p
o
n
r
ea
s
o
n
a
b
le
r
eq
u
est.
T
h
e
d
ataset
co
n
s
is
ts
o
f
h
y
p
e
r
s
p
e
ctr
al
im
ag
es
o
f
r
ed
f
u
ji
ap
p
les
co
llected
f
r
o
m
a
l
o
ca
l
m
ar
k
et
in
Kash
m
ir
,
I
n
d
ia,
an
d
in
clu
d
es
1
,
3
4
5
s
am
p
les
ca
teg
o
r
ized
in
t
o
f
iv
e
b
r
u
is
e
class
es
(
s
m
all,
m
ed
iu
m
,
lar
g
e,
s
tem
,
an
d
ca
ly
x
)
.
T
h
e
d
ataset
was
d
iv
id
ed
in
to
tr
ain
in
g
,
v
alid
atio
n
,
a
n
d
test
in
g
s
u
b
s
ets f
o
r
m
o
d
el
d
ev
e
lo
p
m
en
t a
n
d
ev
alu
atio
n
.
RE
F
E
R
E
NC
E
S
[
1
]
S
.
D
.
G
o
l
o
mb
e
k
a
n
d
M
.
M
.
B
l
a
n
k
e
,
“
O
r
c
h
a
r
d
ma
n
a
g
e
me
n
t
st
r
a
t
e
g
i
e
s
t
o
r
e
d
u
c
e
b
r
u
i
ses
o
n
a
p
p
l
e
s
i
n
I
n
d
i
a
:
a
r
e
v
i
e
w
,
”
Ve
g
e
t
o
s
,
v
o
l
.
3
5
,
n
o
.
1
,
p
p
.
1
–
8
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s4
2
5
3
5
-
021
-
0
0
3
2
6
-
7.
[
2
]
T.
X
u
,
X
.
Z
h
a
n
g
,
Y
.
Zh
u
,
X
.
X
u
,
a
n
d
X
.
R
a
o
,
“
Ev
o
l
u
t
i
o
n
p
a
t
t
e
r
n
i
n
b
r
u
i
se
d
t
i
ssu
e
o
f
‘
r
e
d
d
e
l
i
c
i
o
u
s’
a
p
p
l
e
,
”
F
o
o
d
s
,
v
o
l
.
1
3
,
n
o
.
4
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
f
o
o
d
s1
3
0
4
0
6
0
2
.
[
3
]
E.
E.
O
k
e
r
e
,
A
.
A
m
b
a
w
,
W
.
J
.
P
e
r
o
l
d
,
a
n
d
U
.
L.
O
p
a
r
a
,
“
V
i
s
-
N
I
R
a
n
d
S
W
I
R
h
y
p
e
r
s
p
e
c
t
r
a
l
i
m
a
g
i
n
g
me
t
h
o
d
t
o
d
e
t
e
c
t
b
r
u
i
ses
i
n
p
o
m
e
g
r
a
n
a
t
e
f
r
u
i
t
,
”
Fro
n
t
i
e
rs
i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
4
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s
.
2
0
2
3
.
1
1
5
1
6
9
7
.
[
4
]
G
.
Ö
z
d
o
ğ
a
n
,
X
.
L
i
n
,
a
n
d
D
.
-
W
.
S
u
n
,
“
R
a
p
i
d
a
n
d
n
o
n
i
n
v
a
si
v
e
se
n
s
o
r
y
a
n
a
l
y
ses
o
f
f
o
o
d
p
r
o
d
u
c
t
s
b
y
h
y
p
e
r
s
p
e
c
t
r
a
l
i
m
a
g
i
n
g
:
r
e
c
e
n
t
a
p
p
l
i
c
a
t
i
o
n
d
e
v
e
l
o
p
m
e
n
t
s,
”
T
re
n
d
s i
n
Fo
o
d
S
c
i
e
n
c
e
& Te
c
h
n
o
l
o
g
y
,
v
o
l
.
1
1
1
,
p
p
.
1
5
1
–
1
6
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
i
f
s.2
0
2
1
.
0
2
.
0
4
4
.
[
5
]
M
.
A
l
-
D
a
i
r
i
,
P
.
B
.
P
a
t
h
a
r
e
,
a
n
d
R
.
A
l
-
Y
a
h
y
a
i
,
“
B
r
u
i
se
d
a
ma
g
e
s
u
sc
e
p
t
i
b
i
l
i
t
y
o
f
t
o
ma
t
o
,
”
i
n
M
e
c
h
a
n
i
c
a
l
D
a
m
a
g
e
i
n
Fre
s
h
H
o
rt
i
c
u
l
t
u
r
a
l
Pr
o
d
u
c
e
,
S
i
n
g
a
p
o
r
e
:
S
p
r
i
n
g
e
r
,
2
0
2
3
,
p
p
.
1
7
3
–
1
8
6
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
981
-
99
-
7
0
9
6
-
4
_
9
.
[
6
]
H
.
Y
i
n
,
B
.
Li
,
Y
.
L
i
u
,
F
.
Z
h
a
n
g
,
C
.
S
u
,
a
n
d
A
.
O
.
-
y
a
n
g
,
“
D
e
t
e
c
t
i
o
n
o
f
e
a
r
l
y
b
r
u
i
s
e
s
o
n
l
o
q
u
a
t
u
si
n
g
h
y
p
e
r
s
p
e
c
t
r
a
l
i
m
a
g
i
n
g
t
e
c
h
n
o
l
o
g
y
c
o
u
p
l
e
d
w
i
t
h
b
a
n
d
r
a
t
i
o
a
n
d
i
m
p
r
o
v
e
d
O
t
s
u
me
t
h
o
d
,
”
S
p
e
c
t
r
o
c
h
i
m
i
c
a
Ac
t
a
P
a
rt
A:
M
o
l
e
c
u
l
a
r
a
n
d
B
i
o
m
o
l
e
c
u
l
a
r
S
p
e
c
t
ro
s
c
o
p
y
,
v
o
l
.
2
8
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
sa
a
.
2
0
2
2
.
1
2
1
7
7
5
.
[
7
]
Y
.
T
a
n
g
e
t
a
l
.
,
“
A
p
p
l
e
b
r
u
i
s
e
g
r
a
d
i
n
g
u
si
n
g
p
i
e
c
e
w
i
se
n
o
n
l
i
n
e
a
r
c
u
r
v
e
f
i
t
t
i
n
g
f
o
r
h
y
p
e
r
s
p
e
c
t
r
a
l
i
ma
g
i
n
g
d
a
t
a
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
8
,
p
p
.
1
4
7
4
9
4
–
1
4
7
5
0
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
3
0
1
5
8
0
8
.
[
8
]
I
.
B
a
e
k
e
t
a
l
.
,
“
D
e
t
e
r
m
i
n
a
t
i
o
n
o
f
sp
e
c
t
r
a
l
r
e
so
l
u
t
i
o
n
s
f
o
r
m
u
l
t
i
s
p
e
c
t
r
a
l
d
e
t
e
c
t
i
o
n
o
f
a
p
p
l
e
b
r
u
i
s
e
s
u
si
n
g
v
i
s
i
b
l
e
/
n
e
a
r
-
i
n
f
r
a
r
e
d
h
y
p
e
r
s
p
e
c
t
r
a
l
r
e
f
l
e
c
t
a
n
c
e
i
ma
g
i
n
g
,
”
F
ro
n
t
i
e
rs
i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
3
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s.
2
0
2
2
.
9
6
3
5
9
1
.
[
9
]
J.
F
.
I
.
N
t
u
r
a
m
b
i
r
w
e
,
E.
A
.
H
u
s
sei
n
,
M
.
V
a
c
c
a
r
i
,
C
.
Th
r
o
n
,
W
.
J.
P
e
r
o
l
d
,
a
n
d
U
.
L
.
O
p
a
r
a
,
“
F
e
a
t
u
r
e
r
e
d
u
c
t
i
o
n
f
o
r
t
h
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
b
r
u
i
se
d
a
ma
g
e
t
o
a
p
p
l
e
f
r
u
i
t
u
si
n
g
a
c
o
n
t
a
c
t
l
e
ss
F
T
-
N
I
R
sp
e
c
t
r
o
sc
o
p
y
w
i
t
h
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
F
o
o
d
s
,
v
o
l
.
1
2
,
n
o
.
1
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
f
o
o
d
s
1
2
0
1
0
2
1
0
.
[
1
0
]
X
.
W
a
n
g
,
J.
H
a
n
,
C
.
L
i
u
,
a
n
d
T
.
F
e
n
g
,
“
N
o
n
-
d
e
s
t
r
u
c
t
i
v
e
a
ss
e
ssm
e
n
t
o
f
a
p
p
l
e
i
n
t
e
r
n
a
l
q
u
a
l
i
t
y
u
si
n
g
r
o
t
a
t
i
o
n
a
l
h
y
p
e
r
sp
e
c
t
r
a
l
i
ma
g
i
n
g
,
”
Fr
o
n
t
i
e
rs i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
5
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s
.
2
0
2
4
.
1
4
3
2
1
2
0
.
[
1
1
]
M
.
P
a
l
u
mb
o
e
t
a
l
.
,
“
Emer
g
i
n
g
p
o
s
t
h
a
r
v
e
s
t
t
e
c
h
n
o
l
o
g
i
e
s
t
o
e
n
h
a
n
c
e
t
h
e
s
h
e
l
f
-
l
i
f
e
o
f
f
r
u
i
t
a
n
d
v
e
g
e
t
a
b
l
e
s:
a
n
o
v
e
r
v
i
e
w
,
”
F
o
o
d
s
,
v
o
l
.
1
1
,
n
o
.
2
3
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
f
o
o
d
s
1
1
2
3
3
9
2
5
.
[
1
2
]
H
.
Z
h
a
i
,
P
.
X
i
e
,
X
.
X
i
e
,
a
n
d
S
.
S
.
S
h
a
,
“
D
e
e
p
l
e
a
r
n
i
n
g
-
e
n
a
b
l
e
d
h
y
p
e
r
sp
e
c
t
r
a
l
i
ma
g
i
n
g
f
o
r
h
i
g
h
-
a
c
c
u
r
a
c
y
n
o
n
-
d
e
st
r
u
c
t
i
v
e
q
u
a
n
t
i
f
i
c
a
t
i
o
n
o
f
n
u
t
r
i
t
i
o
n
a
l
c
o
m
p
o
n
e
n
t
s
i
n
m
u
l
t
i
-
v
a
r
i
e
t
y
a
p
p
l
e
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
,
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s
.
2
0
2
5
.
1
6
3
4
7
8
5
.
[
1
3
]
L.
Zh
a
n
g
,
C
.
Li
u
,
J.
H
a
n
,
K
.
S
u
n
,
a
n
d
Y
.
F
e
n
g
,
“
S
E
-
e
n
h
a
n
c
e
d
1
-
D
C
N
N
w
i
t
h
f
u
l
l
-
b
a
n
d
h
y
p
e
r
s
p
e
c
t
r
a
l
i
m
a
g
i
n
g
f
o
r
r
a
p
i
d
a
n
d
a
c
c
u
r
a
t
e
m
a
i
z
e
see
d
v
a
r
i
e
t
y
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
Fr
o
n
t
i
e
rs
i
n
P
l
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
6
,
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s.
2
0
2
5
.
1
5
8
7
8
4
5
.
[
1
4
]
G
.
S
.
B
.
Ja
h
a
n
g
e
e
r
a
n
d
T
.
D
.
R
a
j
k
u
mar,
“
Ea
r
l
y
d
e
t
e
c
t
i
o
n
o
f
b
r
e
a
st
c
a
n
c
e
r
u
si
n
g
h
y
b
r
i
d
o
f
s
e
r
i
e
s
n
e
t
w
o
r
k
a
n
d
V
G
G
-
1
6
,
”
Mu
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
0
,
n
o
.
5
,
p
p
.
7
8
5
3
–
7
8
8
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
0
-
0
9
9
1
4
-
2.
[
1
5
]
M
.
S
.
M
.
A
l
i
,
M
.
S
.
B
.
I
sl
a
m,
M
.
E.
M
a
j
i
d
,
S
.
B
.
A
.
K
a
s
h
e
m
,
A
.
K
h
a
n
d
a
k
a
r
,
a
n
d
M
.
E.
H
.
C
h
o
w
d
h
u
r
y
,
“
A
h
y
b
r
i
d
3
D
C
N
N
-
LST
M
mo
d
e
l
w
i
t
h
s
o
f
t
s
p
a
t
i
a
l
a
t
t
e
n
t
i
o
n
m
e
c
h
a
n
i
sm
f
o
r
a
c
c
u
r
a
t
e
h
y
p
e
r
s
p
e
c
t
r
a
l
i
ma
g
e
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
Re
m
o
t
e
S
e
n
s
i
n
g
A
p
p
l
i
c
a
t
i
o
n
s:
S
o
c
i
e
t
y
a
n
d
E
n
v
i
r
o
n
m
e
n
t
,
v
o
l
.
4
0
,
p
.
1
0
1
7
7
9
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
sase
.
2
0
2
5
.
1
0
1
7
7
9
.
[
1
6
]
M
.
S
u
p
r
i
y
a
,
V
.
S
u
b
i
t
h
a
,
R
.
S
u
m
a
t
h
i
,
a
n
d
P
.
A
.
A
.
R
a
t
h
y
,
“
Th
e
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
c
r
o
p
s
e
l
e
c
t
i
o
n
a
n
d
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
u
si
n
g
B
i
-
LST
M
i
n
a
g
r
i
c
u
l
t
u
r
e
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
1
6
8
–
1
7
4
,
2
0
2
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
u
to
ma
ted
cla
s
s
ifica
tio
n
o
f a
p
p
le
b
r
u
is
es fr
o
m
h
yp
ers
p
ec
tr
a
l
ima
g
es
…
(
P
ed
d
ir
ed
d
y
V
en
ka
tesw
a
r
a
R
ed
d
y
)
1389
[
1
7
]
F
.
K
i
a
n
i
,
F
.
A
.
A
n
k
a
,
a
n
d
F
.
Er
e
n
e
l
,
“
P
S
C
S
O
:
e
n
h
a
n
c
e
d
s
a
n
d
c
a
t
sw
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
i
n
sp
i
r
e
d
b
y
t
h
e
p
o
l
i
t
i
c
a
l
sy
s
t
e
m
t
o
so
l
v
e
c
o
m
p
l
e
x
p
r
o
b
l
e
ms,”
A
d
v
a
n
c
e
s
i
n
E
n
g
i
n
e
e
ri
n
g
S
o
f
t
w
a
re
,
v
o
l
.
1
7
8
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
d
v
e
n
g
s
o
f
t
.
2
0
2
3
.
1
0
3
4
2
3
.
[
1
8
]
D
.
W
u
,
H
.
R
a
o
,
C
.
W
e
n
,
H
.
J
i
a
,
Q
.
L
i
u
,
a
n
d
L
.
A
b
u
a
l
i
g
a
h
,
“
M
o
d
i
f
i
e
d
sa
n
d
c
a
t
sw
a
r
m
o
p
t
i
mi
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
so
l
v
i
n
g
c
o
n
st
r
a
i
n
e
d
e
n
g
i
n
e
e
r
i
n
g
o
p
t
i
m
i
z
a
t
i
o
n
p
r
o
b
l
e
ms,
”
M
a
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
0
,
n
o
.
2
2
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
ma
t
h
1
0
2
2
4
3
5
0
.
[
1
9
]
E.
H
a
ssa
n
,
S
.
A
.
G
h
a
z
a
l
a
h
,
N
.
E
l
-
R
a
sh
i
d
y
,
T.
A
.
El
-
H
a
f
e
e
z
,
a
n
d
M
.
Y
.
S
h
a
ms
,
“
S
u
s
t
a
i
n
a
b
l
e
d
e
e
p
v
i
si
o
n
s
y
st
e
ms
f
o
r
d
a
t
e
f
r
u
i
t
q
u
a
l
i
t
y
a
s
sessm
e
n
t
u
si
n
g
a
t
t
e
n
t
i
o
n
-
e
n
h
a
n
c
e
d
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
,
”
Fr
o
n
t
i
e
rs
i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
6
,
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s
.
2
0
2
5
.
1
5
2
1
5
0
8
.
[
2
0
]
Y
.
Y
u
a
n
,
Z.
Y
a
n
g
,
H
.
Li
u
,
H
.
W
a
n
g
,
J.
Li
,
a
n
d
L.
Zh
a
o
,
“
D
e
t
e
c
t
i
o
n
o
f
e
a
r
l
y
b
r
u
i
se
i
n
a
p
p
l
e
u
s
i
n
g
n
e
a
r
-
i
n
f
r
a
r
e
d
c
a
mer
a
i
m
a
g
i
n
g
t
e
c
h
n
o
l
o
g
y
c
o
m
b
i
n
e
d
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
n
f
r
a
r
e
d
Ph
y
si
c
s &
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
2
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
f
r
a
r
e
d
.
2
0
2
2
.
1
0
4
4
4
2
.
[
2
1
]
B
.
L
i
,
T.
M
a
,
T.
I
n
a
g
a
k
i
,
a
n
d
S
.
Tsu
c
h
i
k
a
w
a
,
“
En
h
a
n
c
e
d
d
e
t
e
c
t
i
o
n
o
f
e
a
r
l
y
b
r
u
i
ses
i
n
a
p
p
l
e
s
u
s
i
n
g
n
e
a
r
-
i
n
f
r
a
r
e
d
h
y
p
e
r
s
p
e
c
t
r
a
l
i
ma
g
i
n
g
w
i
t
h
g
e
o
m
e
t
r
i
c
a
l
i
n
f
l
u
e
n
c
e
c
o
r
r
e
c
t
i
o
n
f
o
r
u
n
i
v
e
r
s
a
l
si
z
e
a
d
a
p
t
a
t
i
o
n
,
”
P
o
st
h
a
rve
s
t
Bi
o
l
o
g
y
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
1
9
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
o
st
h
a
r
v
b
i
o
.
2
0
2
4
.
1
1
3
2
8
2
.
[
2
2
]
Q
.
P
a
n
g
,
W
.
H
u
a
n
g
,
S
.
F
a
n
,
Q
.
Z
h
o
u
,
Z.
W
a
n
g
,
a
n
d
X
.
T
i
a
n
,
“
D
e
t
e
c
t
i
o
n
o
f
e
a
r
l
y
b
r
u
i
ses
o
n
a
p
p
l
e
s
u
si
n
g
h
y
p
e
r
sp
e
c
t
r
a
l
i
m
a
g
i
n
g
c
o
m
b
i
n
i
n
g
w
i
t
h
Y
O
LO
v
3
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m,”
J
o
u
r
n
a
l
o
f
F
o
o
d
Pr
o
c
e
ss
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
4
5
,
n
o
.
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
1
1
/
j
f
p
e
.
1
3
9
5
2
.
[
2
3
]
Z.
Ü
n
a
l
,
T.
K
ı
z
ı
l
d
e
n
i
z
,
M
.
Ö
z
d
e
n
,
H
.
A
k
t
a
ş,
a
n
d
Ö
.
K
a
r
a
g
ö
z
,
“
D
e
t
e
c
t
i
o
n
o
f
b
r
u
i
s
e
s
o
n
r
e
d
a
p
p
l
e
s
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s,”
S
c
i
e
n
t
i
a
H
o
r
t
i
c
u
l
t
u
ra
e
,
v
o
l
.
3
2
9
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
sc
i
e
n
t
a
.
2
0
2
4
.
1
1
3
0
2
1
.
[
2
4
]
D
.
H
u
,
D
.
Q
i
u
,
S
.
Y
u
,
T.
J
i
a
,
T.
Zh
o
u
,
a
n
d
X
.
Y
a
n
,
“
I
n
t
e
g
r
a
t
i
o
n
o
f
o
p
t
i
c
a
l
p
r
o
p
e
r
t
y
ma
p
p
i
n
g
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
r
e
a
l
-
t
i
m
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
e
a
r
l
y
b
r
u
i
ses
o
f
a
p
p
l
e
s,”
F
o
o
d
a
n
d
B
i
o
p
r
o
c
e
ss
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
7
,
n
o
.
9
,
p
p
.
2
7
4
5
–
2
7
5
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
9
4
7
-
023
-
0
3
2
6
0
-
5.
[
2
5
]
S
.
W
e
n
g
,
S
.
Y
u
,
B
.
G
u
o
,
P
.
T
a
n
g
,
a
n
d
D
.
Li
a
n
g
,
“
N
o
n
-
d
e
s
t
r
u
c
t
i
v
e
d
e
t
e
c
t
i
o
n
o
f
s
t
r
a
w
b
e
r
r
y
q
u
a
l
i
t
y
u
si
n
g
mu
l
t
i
-
f
e
a
t
u
r
e
s
o
f
h
y
p
e
r
s
p
e
c
t
r
a
l
i
ma
g
i
n
g
a
n
d
m
u
l
t
i
v
a
r
i
a
t
e
me
t
h
o
d
s,
”
S
e
n
so
rs
,
v
o
l
.
2
0
,
n
o
.
1
1
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
0
1
1
3
0
7
4
.
[
2
6
]
A
.
S
e
y
y
e
d
a
b
b
a
si
a
n
d
F
.
K
i
a
n
i
,
“
S
a
n
d
c
a
t
sw
a
r
m o
p
t
i
mi
z
a
t
i
o
n
:
a
n
a
t
u
r
e
-
i
n
sp
i
r
e
d
a
l
g
o
r
i
t
h
m
t
o
so
l
v
e
g
l
o
b
a
l
o
p
t
i
m
i
z
a
t
i
o
n
p
r
o
b
l
e
ms,
”
En
g
i
n
e
e
ri
n
g
w
i
t
h
C
o
m
p
u
t
e
rs
,
v
o
l
.
3
9
,
n
o
.
4
,
p
p
.
2
6
2
7
–
2
6
5
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s0
0
3
6
6
-
022
-
0
1
6
0
4
-
x.
[
2
7
]
A
.
G
.
D
e
v
i
,
A
.
Th
o
t
a
,
G
.
N
i
t
h
y
a
,
S
.
M
a
j
j
i
,
A
.
G
o
p
a
t
o
t
i
,
a
n
d
L
.
D
h
a
v
a
ma
n
i
,
“
A
d
v
a
n
c
e
m
e
n
t
o
f
d
i
g
i
t
a
l
i
ma
g
e
s
t
e
g
a
n
o
g
r
a
p
h
y
u
si
n
g
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
,
”
i
n
2
0
2
2
I
n
t
e
r
n
a
t
i
o
n
a
l
I
n
t
e
r
d
i
s
c
i
p
l
i
n
a
ry
H
u
m
a
n
i
t
a
ri
a
n
C
o
n
f
e
r
e
n
c
e
f
o
r
S
u
s
t
a
i
n
a
b
i
l
i
t
y
,
2
0
2
2
,
p
p
.
2
5
0
–
2
5
4
.
d
o
i
:
1
0
.
1
1
0
9
/
I
I
H
C
5
5
9
4
9
.
2
0
2
2
.
1
0
0
6
0
2
3
0
.
[
2
8
]
Y
.
N
a
g
a
r
a
j
u
,
V
e
n
k
a
t
e
sh
,
S
.
S
w
e
t
h
a
,
a
n
d
S
.
S
t
a
l
i
n
,
“
A
p
p
l
e
a
n
d
g
r
a
p
e
l
e
a
f
d
i
s
e
a
se
s
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
v
i
a
f
i
n
e
-
t
u
n
e
d
c
l
a
ss
i
f
i
e
r
,
”
i
n
2
0
2
0
I
EE
E
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
A
p
p
l
i
e
d
N
e
t
w
o
rk
T
e
c
h
n
o
l
o
g
i
e
s
,
2
0
2
0
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
M
LA
N
T
5
0
9
6
3
.
2
0
2
0
.
9
3
5
5
9
9
1
.
[
2
9
]
X
.
S
h
i
,
X
.
C
h
a
i
,
C
.
Y
a
n
g
,
X
.
X
i
a
,
a
n
d
T
.
S
u
n
,
“
V
i
s
i
o
n
-
b
a
se
d
a
p
p
l
e
q
u
a
l
i
t
y
g
r
a
d
i
n
g
w
i
t
h
mu
l
t
i
-
v
i
e
w
s
p
a
t
i
a
l
n
e
t
w
o
r
k
,
”
C
o
m
p
u
t
e
rs
a
n
d
El
e
c
t
r
o
n
i
c
s i
n
Ag
r
i
c
u
l
t
u
re
,
v
o
l
.
1
9
5
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
a
g
.
2
0
2
2
.
1
0
6
7
9
3
.
[
3
0
]
Y
.
Li
a
n
d
G
.
W
a
n
g
,
“
S
a
n
d
c
a
t
sw
a
r
m
o
p
t
i
mi
z
a
t
i
o
n
b
a
se
d
o
n
s
t
o
c
h
a
st
i
c
v
a
r
i
a
t
i
o
n
w
i
t
h
e
l
i
t
e
c
o
l
l
a
b
o
r
a
t
i
o
n
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
8
9
9
8
9
–
9
0
0
0
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
2
0
1
1
4
7
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Pedd
ire
d
d
y
Ve
n
k
a
te
sw
a
r
a
Re
d
d
y
is
c
u
rre
n
tl
y
a
re
se
a
rc
h
sc
h
o
lar
i
n
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
a
t
Ka
las
a
li
n
g
a
m
Ac
a
d
e
m
y
o
f
Re
se
a
rc
h
a
n
d
E
d
u
c
a
ti
o
n
,
In
d
ia.
He
h
a
s
m
o
re
th
a
n
1
7
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
i
n
th
e
field
o
f
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
e
n
g
in
e
e
r
in
g
.
His
re
se
a
rc
h
in
tere
sts
i
n
c
l
u
d
e
a
rti
f
icia
l
i
n
telli
g
e
n
c
e
,
m
a
c
h
in
e
lea
rn
in
g
,
d
a
ta
m
in
in
g
,
c
o
m
p
u
ter
v
isio
n
,
a
n
d
d
e
e
p
le
a
rn
in
g
.
He
h
a
s
sig
n
ifi
c
a
n
t
e
x
p
e
rien
c
e
i
n
tea
c
h
i
n
g
c
o
re
c
o
m
p
u
ter
sc
ien
c
e
s
u
b
jec
ts
a
n
d
g
u
id
i
n
g
u
n
d
e
rg
ra
d
u
a
te
a
n
d
p
o
stg
ra
d
u
a
te
stu
d
e
n
t
p
r
o
jec
ts.
He
is
a
c
ti
v
e
ly
e
n
g
a
g
e
d
i
n
a
c
a
d
e
m
ic
re
se
a
rc
h
a
n
d
h
a
s
p
u
b
li
s
h
e
d
se
v
e
ra
l
p
a
p
e
rs
i
n
re
p
u
ted
jo
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
His
c
u
rre
n
t
re
se
a
rc
h
fo
c
u
se
s o
n
d
e
v
e
lo
p
in
g
in
telli
g
e
n
t
c
o
m
p
u
tat
io
n
a
l
m
o
d
e
ls
a
n
d
a
d
v
a
n
c
e
d
a
l
g
o
ri
th
m
s
t
o
a
d
d
re
ss
re
a
l
-
wo
rld
p
ro
b
lem
s
in
ima
g
e
a
n
a
ly
sis
a
n
d
d
a
ta
-
d
ri
v
e
n
a
p
p
l
ica
ti
o
n
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
v
e
n
k
a
tes
wa
rlu
re
e
d
d
y
p
@
g
m
a
il
.
c
o
m
.
Ala
g
u
c
h
a
m
y
Pa
r
i
v
a
z
h
a
g
a
n
is
p
re
se
n
tl
y
wo
r
k
in
g
a
s
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
,
in
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
,
S
c
h
o
o
l
o
f
Co
m
p
u
ti
n
g
,
Ka
las
a
li
n
g
a
m
Ac
a
d
e
m
y
o
f
Re
se
a
rc
h
a
n
d
Ed
u
c
a
ti
o
n
(KA
RE
-
De
e
m
e
d
to
b
e
Un
i
v
e
rsity
),
Krish
n
a
n
k
o
il
,
I
n
d
ia
.
He
h
a
s
p
u
rsu
e
d
h
is
f
u
ll
-
ti
m
e
P
h
.
D
.
fro
m
S
c
h
o
o
l
o
f
El
e
c
tro
n
ics
En
g
in
e
e
rin
g
at
Ve
ll
o
re
I
n
stit
u
te
o
f
Tec
h
n
o
lo
g
y
(VIT
Ch
e
n
n
a
i
Ca
m
p
u
s),
In
d
ia.
He
re
c
e
iv
e
d
h
is
M
.
E
.
(Co
n
tro
l
a
n
d
In
stru
m
e
n
tati
o
n
)
i
n
2
0
1
2
fr
o
m
Ve
lam
m
a
l
En
g
in
e
e
rin
g
Co
ll
e
g
e
,
A
n
n
a
Un
i
v
e
rsity
,
Ch
e
n
n
a
i
a
n
d
B
.
E
.
(El
e
c
tro
n
ics
a
n
d
C
o
m
m
u
n
ic
a
ti
o
n
E
n
g
i
n
e
e
rin
g
)
in
2
0
1
0
fr
o
m
M
a
d
h
a
En
g
i
n
e
e
rin
g
C
o
ll
e
g
e
,
An
n
a
Un
i
v
e
rsity
,
Ch
e
n
n
a
i.
His
re
se
a
rc
h
in
tere
sts
a
re
d
ig
it
a
l
im
a
g
e
p
ro
c
e
ss
in
g
,
b
io
m
e
tri
c
s,
b
io
m
e
d
ica
l
ima
g
e
p
ro
c
e
ss
in
g
,
fa
c
e
re
c
o
g
n
it
i
o
n
,
e
a
r
re
c
o
g
n
it
i
o
n
,
a
r
ti
ficia
l
in
telli
g
e
n
c
e
,
p
a
tt
e
rn
re
c
o
g
n
it
i
o
n
,
c
o
m
p
u
ter
v
isio
n
,
a
n
d
o
b
jec
t
re
c
o
g
n
it
i
o
n
.
He
h
a
s
p
u
b
l
i
sh
e
d
m
o
re
th
a
n
1
2
re
se
a
rc
h
a
rti
c
les
/p
a
p
e
rs
a
b
o
u
t
e
a
r
re
c
o
g
n
it
i
o
n
,
fa
c
e
re
c
o
g
n
it
i
o
n
,
fa
c
e
d
e
tec
ti
o
n
,
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
in
v
a
rio
u
s
in
ter
n
a
ti
o
n
a
l
jo
u
rn
a
ls.
He
h
a
s
m
o
re
th
a
n
1
1
y
e
a
rs
o
f
tea
c
h
i
n
g
a
n
d
re
se
a
rc
h
e
x
p
e
rien
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
.
p
a
riv
a
z
h
a
g
a
n
@k
lu
.
a
c
.
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.