I
n
t
e
r
n
at
ion
al
Jou
r
n
a
l
of
I
n
f
o
r
m
at
ics
an
d
Com
m
u
n
icat
ion
T
e
c
h
n
ol
ogy
(
I
J
-
I
CT
)
Vo
l
.
14
,
N
o
.
1
,
A
pr
i
l
20
25
,
pp.
240
~
249
I
S
S
N:
2252
-
8776
,
DO
I
:
10
.
11591/i
ji
c
t
.
v
14
i
1
.
pp
24
0
-
249
240
Jou
r
n
al
h
o
m
e
page
:
ht
tp:
//
ij
ict
.
iaes
c
or
e
.
c
om
A
h
y
b
r
id
ap
p
r
oac
h
of
p
at
t
e
r
n
r
e
c
ogn
ition
t
o d
e
t
e
c
t
m
ar
in
e
an
i
m
al
s
Vij
ayal
ak
s
h
m
i
B
al
ac
h
an
d
r
an
1
,
T
h
an
ga
Ram
ya
S
h
an
m
u
gave
l
2
,
Ram
ar
K
ad
ar
k
a
r
ayan
d
i
3
,
Vij
ayal
ak
s
h
m
i
K
an
d
h
as
am
y
1
1
D
e
pa
r
tm
e
nt
of
C
o
mpu
te
r
S
c
ie
n
c
e
a
nd E
ngi
n
e
e
r
in
g, R
a
m
c
o
I
ns
ti
tu
te
of
T
e
c
hn
ol
o
g
y
, R
a
ja
pa
la
y
a
m, I
ndi
a
2
D
e
pa
r
tm
e
nt
of
C
o
mpu
te
r
S
c
ie
n
c
e
a
nd D
e
s
ig
n, R
M
K
E
ngi
ne
e
r
i
ng C
o
ll
e
ge
, C
h
e
nna
i,
I
ndi
a
3
D
e
pa
r
tm
e
nt
of
C
o
mpu
te
r
S
c
ie
n
c
e
a
nd E
ngi
n
e
e
r
in
g, R
M
K
C
o
ll
e
g
e
of
E
ngi
n
e
e
r
in
g a
nd
T
e
c
hn
o
l
oy
, C
h
e
nna
i,
I
ndi
a
Ar
t
ic
l
e
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
i
ve
d
J
u
l
11,
2024
R
e
vi
s
e
d
Oc
t
5,
2024
A
c
c
e
pt
e
d
N
o
v
19,
2024
A
c
q
u
i
ri
n
g
u
p
-
to
-
d
at
e
k
n
o
w
l
ed
g
e
ab
o
u
t
v
ar
i
o
u
s
an
i
m
al
s
w
i
l
l
h
a
v
e
a
s
i
g
n
i
fi
c
an
t
i
m
p
ac
t
o
n
e
ff
ec
t
i
v
el
y
m
an
a
g
i
n
g
s
p
ec
i
e
s
w
i
t
h
i
n
t
h
e
eco
s
y
s
t
em
.
Man
u
al
l
y
i
d
e
n
t
i
f
y
i
n
g
a
n
i
m
a
l
s
an
d
t
h
e
i
r
t
rai
t
s
co
n
t
i
n
u
es
t
o
b
e
a
c
o
s
t
l
y
a
n
d
t
i
me
-
c
o
n
s
u
m
i
n
g
p
ro
ce
s
s
.
T
h
e
d
e
v
el
o
p
me
n
t
o
f
a
s
y
s
t
em
u
s
i
n
g
t
h
e
m
o
s
t
r
ece
n
t
d
ev
e
l
o
p
men
t
s
i
n
co
m
p
u
t
e
r
v
i
s
i
o
n
m
a
c
h
i
n
e
l
e
arn
i
n
g
w
as
n
ece
s
s
ar
y
t
o
a
d
d
r
e
s
s
t
h
e
i
s
s
u
e
s
o
f
d
e
t
ec
t
i
n
g
s
h
ark
s
an
d
aq
u
at
i
c
s
p
ec
i
e
s
i
n
ar
eas
fi
l
l
ed
w
i
t
h
s
u
rf
e
rs
,
ro
ck
s
,
a
n
d
v
ari
o
u
s
o
t
h
e
r
p
o
t
e
n
t
i
a
l
fa
l
s
e
p
o
s
i
t
i
v
e
s
.
In
t
h
e
o
ce
a
n
mo
s
t
o
f
t
h
e
s
p
e
c
i
e
s
ar
e
co
l
d
-
b
l
o
o
d
ed
a
n
i
m
a
l
s
h
e
n
ce
t
h
ey
c
a
n
n
o
t
b
e
t
rac
k
e
d
w
i
t
h
t
h
e
r
m
al
c
a
me
ras
.
O
ce
an
’
s
d
y
n
a
m
i
c
e
n
v
i
r
o
n
me
n
t
aff
ec
t
s
s
i
m
p
l
e
t
ec
h
n
i
q
u
e
s
l
i
k
e
c
o
l
o
r
s
e
p
arat
i
o
n
,
i
n
t
en
s
i
t
y
h
i
s
t
o
g
ra
m
s
,
an
d
o
p
t
i
c
a
l
f
l
o
w
.
H
en
ce
a
h
y
b
ri
d
ap
p
ro
a
c
h
u
s
i
n
g
co
n
v
o
l
u
t
i
o
n
a
l
n
e
u
ra
l
n
e
t
w
o
r
k
-
s
u
p
p
o
rt
v
ec
t
o
r
mac
h
i
n
e
(CN
N
-
SV
M)
c
l
as
s
i
fi
e
r
i
s
p
ro
p
o
s
e
d
t
o
p
e
rf
o
r
m
t
h
e
p
at
t
e
rn
reco
g
n
i
t
i
o
n
.
A
C
N
N
i
s
em
p
l
o
y
ed
f
o
r
f
e
at
u
r
e
e
x
t
rac
t
i
o
n
b
y
u
s
i
n
g
t
h
e
h
i
s
t
o
g
ra
m
o
f
g
ra
d
i
en
t
s
v
al
u
e
.
Su
b
s
e
q
u
en
t
l
y
,
a
S
V
M
c
l
as
s
i
fi
e
r
i
s
em
p
l
o
y
ed
t
o
i
d
en
t
i
f
y
a
n
d
c
at
e
g
o
ri
s
e
m
ari
n
e
s
p
ec
i
e
s
i
n
t
h
e
v
i
c
i
n
i
t
y
o
f
t
h
e
s
e
a
co
as
t
.
T
h
i
s
s
e
rv
e
s
t
o
n
o
t
i
f
y
i
n
d
i
v
i
d
u
al
s
w
h
o
en
g
a
g
e
i
n
s
w
i
mm
i
n
g
a
c
t
i
v
i
t
i
e
s
i
n
t
h
e
o
ce
an
.
T
h
e
s
u
g
g
e
s
t
ed
mo
d
e
l
i
s
e
v
a
l
u
at
e
d
a
g
ai
n
s
t
al
t
e
rn
at
i
v
e
m
a
c
h
i
n
e
l
e
arn
i
n
g
ap
p
ro
ach
e
s
,
a
n
d
i
t
ac
h
i
e
v
e
s
a
s
u
p
e
ri
o
r
a
ccu
racy
o
f
9
5
%
c
o
m
p
ar
e
d
t
o
t
h
e
o
t
h
e
rs
.
K
e
y
w
o
r
d
s
:
C
l
a
s
s
if
i
e
r
C
o
n
v
o
l
ut
i
o
n
n
e
ur
a
l
n
e
t
wo
r
k
H
i
s
t
o
gr
a
m
o
f
gr
a
d
i
e
n
t
s
M
a
r
i
n
e
a
nim
a
l
s
P
a
tt
e
r
n
r
e
c
o
gni
t
i
o
n
S
uppor
t
v
e
c
to
r
m
a
c
hi
ne
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
cen
s
e.
C
or
r
e
s
pon
din
g
A
u
th
or
:
V
i
j
a
y
a
l
a
k
s
hmi
B
a
l
a
c
h
a
n
dr
a
n
De
pa
r
t
m
e
n
t
o
f
C
o
m
put
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
ne
e
r
i
n
g
,
R
a
m
c
o
I
n
s
t
i
t
ut
e
o
f
T
e
c
hn
o
l
o
g
y
R
a
j
a
pa
l
a
y
a
m
,
T
a
mi
l
na
du,
I
n
d
i
a
E
m
a
i
l
:
bvij
i
.
a
k
il
a
@
g
m
a
il
.
c
o
m
1.
I
NT
RODU
C
T
I
ON
S
i
n
c
e
t
h
e
m
a
j
o
r
i
t
y
o
f
m
a
r
i
ne
or
ga
ni
s
m
s
a
r
e
c
o
l
d
-
bl
o
o
de
d,
t
h
e
r
m
a
l
c
a
m
e
r
a
s
c
a
nn
ot
tr
a
c
k
t
h
e
m
.
S
h
a
r
ks
a
r
e
c
r
uc
i
a
l
t
o
t
h
e
e
c
o
l
o
g
y
b
e
c
a
u
s
e
t
h
e
y
ke
e
p
t
h
e
s
pe
c
i
e
s
l
o
we
r
do
wn
i
n
t
h
e
f
o
o
d
c
h
a
i
n
a
li
ve
a
n
d
a
c
t
a
s
a
ga
uge
o
f
t
h
e
h
e
a
l
t
h
o
f
t
h
e
o
c
e
a
n
.
I
n
t
h
e
o
c
e
a
n
s
of
t
h
e
wo
r
l
d,
t
h
e
r
e
a
r
e
a
b
o
u
t
500
di
f
f
e
r
e
n
t
s
pe
c
i
e
s
o
f
s
ha
r
k.
I
t
c
o
m
e
s
i
n
a
r
a
n
ge
o
f
c
o
l
o
ur
s
a
n
d
s
i
z
e
s
,
r
a
n
g
i
ng
f
r
o
m
39
f
e
e
t
(
12
m
e
t
r
e
s
)
to
l
e
s
s
t
h
a
n
o
n
e
m
e
t
r
e
(
3
f
e
e
t
)
.
T
h
e
s
a
l
t
wa
t
e
r
c
r
o
c
o
di
l
e
s
h
a
ve
t
h
e
s
t
r
o
n
ge
s
t
p
r
o
pe
ns
i
t
i
e
s
to
vi
e
w
pe
o
pl
e
a
s
pr
e
y
a
m
o
n
g
a
l
l
c
r
o
c
o
di
l
i
a
n
s
.
T
h
e
t
w
o
m
a
j
o
r
m
a
r
i
n
e
a
nim
a
l
s
t
h
a
t
c
o
ul
d
po
s
e
a
t
h
r
e
a
t
to
s
wi
mm
e
r
s
a
r
e
s
a
l
t
wa
t
e
r
c
r
o
c
o
di
l
e
s
a
n
d
gr
e
a
t
whi
t
e
s
ha
r
ks
.
W
o
r
l
dw
i
de
,
t
h
e
r
e
h
a
v
e
b
e
e
n
nu
m
e
r
o
us
i
n
c
i
de
n
t
s
o
f
s
h
a
r
k
a
n
d
s
a
l
t
wa
t
e
r
c
r
o
c
o
di
le
a
tt
a
c
ks
,
m
a
ny
o
f
w
hi
c
h
too
k
pl
a
c
e
i
n
s
h
a
ll
o
w
s
e
a
s
.
T
h
e
ma
j
o
r
i
t
y
o
f
a
s
s
a
u
l
t
s
t
a
ke
pl
a
c
e
i
n
s
e
a
s
c
l
o
s
e
to
s
h
o
r
e
,
us
ua
l
l
y
c
l
o
s
e
t
o
o
r
b
e
t
we
e
n
s
a
n
d
b
a
r
s
w
h
e
r
e
s
h
a
r
ks
mi
g
h
t
ge
t
c
a
ugh
t
a
t
l
o
w
t
i
de
w
hil
e
f
e
e
d
i
ng.
S
l
o
pe
d
a
r
e
a
s
a
r
e
a
l
s
o
po
s
s
i
bl
e
p
l
a
c
e
s
f
o
r
a
tt
a
c
ks
.
S
wi
m
m
e
r
s
a
n
d
s
ur
f
e
r
s
a
r
e
t
h
e
us
ua
l
t
a
r
ge
t
s
o
f
t
h
e
a
t
t
a
c
ks
,
whi
c
h
m
a
inl
y
t
a
ke
p
l
a
c
e
i
n
t
h
e
s
ur
f
z
o
ne
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
hy
br
id
appr
oac
h
o
f
pat
ter
n
r
e
c
ognit
ion
to
de
tec
t
mar
ine
animals
(
V
ij
ay
alaks
hmi
B
alachandr
an
)
241
P
e
o
pl
e
w
h
o
s
w
i
m
i
n
t
h
e
s
e
a
c
l
o
s
e
to
t
h
e
c
o
a
s
t
a
r
e
s
e
r
i
o
us
ly
t
h
r
e
a
t
e
n
e
d
by
b
ot
h
s
ha
r
ks
a
n
d
s
a
l
t
wa
t
e
r
c
r
o
c
o
di
l
e
s
.
I
n
or
de
r
to
i
de
n
t
i
f
y
t
h
e
s
e
m
a
r
i
ne
s
pe
c
i
e
s
us
i
n
g
hi
g
h
r
e
s
o
l
ut
i
o
n
c
a
m
e
r
a
s
a
n
d
wa
r
n
s
w
im
m
e
r
s
a
s
t
h
e
y
a
ppr
o
a
c
h
t
h
e
c
o
a
s
t
,
a
pa
tt
e
r
n
r
e
c
o
gn
i
t
i
o
n
s
y
s
t
e
m
m
u
s
t
b
e
de
v
e
l
o
pe
d.
T
he
f
o
l
l
o
w
i
ng
t
a
s
ks
m
us
t
b
e
c
o
m
p
l
e
t
e
d
i
n
o
r
de
r
to
s
p
ot
s
h
a
r
ks
a
n
d
c
r
o
c
o
di
l
e
s
.
F
i
nd
i
ng
t
h
e
s
us
p
i
c
i
o
us
s
ha
pe
ne
a
r
t
h
e
c
o
a
s
t
a
n
d
de
t
e
r
m
i
ni
n
g
wh
e
t
h
e
r
o
r
n
ot
i
t
i
s
a
h
a
r
m
f
u
l
s
e
a
s
pe
c
i
e
s
.
T
o
s
a
v
e
t
h
e
s
w
i
mm
e
r
s
’
l
i
ve
s
,
i
t
i
s
i
m
po
r
t
a
n
t
to
wa
r
n
t
h
e
m
i
f
t
h
e
r
e
a
r
e
s
h
a
r
ks
o
r
c
r
o
c
o
di
l
e
s
pr
e
s
e
n
t
.
T
h
e
f
o
l
l
o
w
i
ng
f
a
c
t
o
r
s
m
a
k
e
i
t
h
a
r
de
r
to
di
s
c
e
r
n
pa
tt
e
r
n
s
a
n
d
f
i
nd
m
a
r
i
ne
o
r
ga
ni
s
m
s
i
n
t
h
e
m
a
r
i
t
i
m
e
e
nvi
r
o
nm
e
n
t
:
Us
e
a
hi
g
h
-
r
e
s
o
l
ut
i
o
n
c
a
m
e
r
a
i
n
a
dr
o
n
e
t
o
i
de
n
t
i
f
y
s
h
a
r
ks
a
n
d
c
r
o
c
o
di
l
e
s
f
r
o
m
a
m
i
n
im
u
m
o
f
60
m
e
t
r
e
s
a
b
o
v
e
t
h
e
gr
o
un
d.
T
h
e
o
c
e
a
n
’
s
i
nc
r
e
d
i
bly
d
y
na
mi
c
e
nvi
r
o
nm
e
n
t
c
a
us
e
d
by
t
h
e
wa
v
e
s
,
g
l
a
r
e
s
,
a
n
d
s
h
a
do
ws
.
A
ut
o
m
a
t
i
c
m
a
r
i
ne
a
nim
a
l
i
de
n
t
i
f
i
c
a
t
i
o
n
m
a
y
b
e
c
a
r
r
i
e
d
o
u
t
wi
t
h
e
a
s
e
t
h
a
n
ks
to
t
h
e
c
a
pa
bi
li
t
i
e
s
o
f
m
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hni
que
s
a
n
d
i
n
c
r
e
a
s
e
d
pr
o
c
e
s
s
i
ng
po
we
r
.
T
hi
s
c
a
n
o
pe
n
up
a
v
a
s
t
n
e
w
a
r
r
a
y
o
f
o
pp
o
r
t
un
i
t
i
e
s
f
o
r
de
v
e
l
o
p
i
ng
s
p
e
c
i
a
li
z
e
d
wa
r
ni
ng
s
y
s
t
e
m
a
n
d
da
t
a
c
o
l
l
e
c
t
i
o
n
s
c
e
n
a
r
i
o
s
.
B
y
do
i
n
g
t
hi
s
,
t
h
e
m
a
n
ua
l
wo
r
k
a
n
d
a
s
s
o
c
i
a
t
e
d
e
x
pe
n
s
e
s
a
r
e
de
c
r
e
a
s
e
d,
a
n
d
t
h
e
m
a
r
i
ne
e
c
o
s
y
s
t
e
m
s
i
s
pr
e
s
e
r
ve
d.
T
h
e
us
e
o
f
c
o
m
put
e
r
vi
s
i
o
n
too
l
s
to
r
e
s
e
a
r
c
h
a
n
d
t
r
a
c
k
t
h
e
m
a
r
i
n
e
e
c
o
s
y
s
t
e
m
ha
s
gr
o
wn
i
n
r
e
c
e
n
t
y
e
a
r
s
.
T
h
e
m
a
i
n
us
e
s
i
nv
o
l
ve
o
bj
e
c
t
de
t
e
c
t
i
o
n
,
whi
c
h
i
s
h
e
l
p
f
u
l
f
o
r
tr
a
c
k
i
n
g
f
i
s
h
po
pu
l
a
t
i
o
n
s
[
1]
,
[
2
]
,
de
t
e
c
t
i
n
g
s
e
a
l
s
[
3]
,
l
o
c
a
t
i
n
g
wh
a
l
e
[
4]
h
o
t
s
p
ot
s
,
a
n
d
c
o
r
a
l
r
e
e
f
f
i
s
h
po
pu
l
a
t
i
o
ns
[
5]
.
T
h
e
un
pr
o
v
o
ke
d
s
h
a
r
k
a
tt
a
c
ks
c
a
n
b
e
a
v
o
i
de
d
by
us
i
n
g
t
h
e
po
we
r
o
f
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
li
ge
nc
e
(
AI
)
a
n
d
m
a
c
hi
ne
l
e
a
r
ni
ng
(
ML
)
a
l
go
r
i
t
hm
s
t
o
de
t
e
c
t
t
h
e
o
c
c
ur
r
e
n
c
e
o
f
s
h
a
r
ks
i
n
t
h
e
s
e
a
s
h
o
r
e
.
T
h
e
s
h
a
r
ks
pott
e
r
[
6]
c
l
a
s
s
if
i
e
s
t
h
e
de
t
e
c
t
e
d
o
bj
e
c
t
i
n
t
o
o
n
e
o
f
t
h
e
16
c
a
t
e
g
o
r
i
e
s
,
i
f
i
t
i
d
e
n
t
i
f
i
e
s
a
s
h
a
r
k
t
h
e
a
l
e
r
t
ge
n
e
r
a
t
e
d
a
n
d
t
h
e
s
w
i
mm
e
r
s
c
a
n
b
e
wa
r
ne
d.
P
r
o
p
o
s
e
d
a
m
e
t
h
o
d
f
o
r
de
t
e
c
t
i
n
g
t
h
e
do
l
p
hi
ns
a
n
d
s
h
a
r
k
us
i
n
g
t
h
e
s
h
a
pe
f
e
a
t
ur
e
pr
o
pe
r
t
i
e
s
f
r
o
m
t
h
e
a
e
r
i
a
l
im
a
g
e
s
.
T
h
e
m
e
t
h
o
do
l
o
g
y
b
a
s
e
d
o
n
t
h
e
t
w
o
-
di
m
e
n
s
i
o
n
a
l
de
f
o
r
m
a
bl
e
m
o
de
l
[
7]
i
n
whi
c
h
t
h
e
r
e
f
e
r
e
n
c
e
v
a
r
i
a
bl
e
wa
s
o
pt
i
m
i
z
e
d
to
p
r
o
duc
e
l
o
w
e
r
r
o
r
r
a
t
e
a
n
d
hi
g
h
a
c
c
ur
a
c
y
.
A
de
e
p
l
e
a
r
ni
ng
a
l
go
r
i
t
hm
wa
s
pr
e
s
e
n
t
e
d
to
i
de
n
t
i
f
y
W
hi
t
e
s
h
a
r
ks
in
un
d
e
r
wa
t
e
r
e
nvi
r
o
nm
e
n
t
s
,
pr
o
vi
d
i
ng
a
s
s
i
s
t
a
nc
e
to
di
v
e
r
s
a
n
d
o
t
h
e
r
a
f
i
c
i
o
n
a
do
s
o
f
u
n
de
r
wa
t
e
r
s
po
r
t
s
.
T
h
e
s
t
ud
y
us
e
d
a
YO
L
O
v
3
a
l
go
r
i
t
hm
[
8]
whi
c
h
ut
i
li
s
e
s
c
o
n
v
o
l
ut
i
o
n
a
l
n
e
ur
a
l
ne
t
wor
ks
(
C
NN
s
)
to
r
e
c
o
gni
s
e
o
bj
e
c
t
s
,
m
a
k
e
pr
e
d
i
c
t
i
o
ns
a
t
m
a
ny
s
c
a
l
e
s
,
a
n
d
f
o
r
e
c
a
s
t
b
o
un
d
i
ng
b
o
xe
s
us
i
ng
l
o
g
i
s
t
i
c
r
e
gr
e
s
s
i
o
n
.
T
h
e
y
t
a
c
k
l
e
d
t
h
e
c
o
n
c
e
r
n
s
r
e
l
a
t
e
d
to
t
h
e
un
de
r
s
e
a
e
nvir
o
nm
e
n
t
wh
e
n
i
de
n
t
i
f
yi
ng
t
h
e
s
pe
c
i
e
s
.
De
e
p
C
NN
b
a
s
e
d
o
bj
e
c
t
de
t
e
c
t
i
o
n
m
o
de
l
[
9]
wa
s
d
e
s
i
g
n
e
d
to
de
t
e
c
t
s
h
a
r
k
a
n
d
ot
h
e
r
m
a
r
i
ne
a
nim
a
l
s
.
I
t
a
l
s
o
pe
r
f
o
r
m
s
t
h
e
r
e
g
i
o
n
s
e
g
m
e
n
t
a
t
i
o
n
t
a
s
k
f
o
r
s
ha
r
k
de
t
e
c
t
i
o
n
whi
c
h
wa
s
do
n
e
by
f
a
s
t
R
C
NN
a
l
o
n
g
wi
t
h
VG
G16
a
r
c
hi
t
e
c
t
ur
e
t
h
a
t
i
m
pr
o
v
e
s
t
h
e
a
c
c
ur
a
c
y
r
a
t
i
o
.
T
r
a
n
s
f
e
r
l
e
a
r
ni
ng
a
n
d
C
NN
ba
s
e
d
m
o
de
l
[
10]
de
s
i
g
n
e
d
f
o
r
c
l
a
s
s
if
i
c
a
t
i
o
n
o
f
s
h
a
r
ks
.
A
s
ha
r
k
a
l
e
r
t
i
n
g
s
y
s
t
e
m
[
11]
i
n
w
hi
c
h
t
h
e
s
h
a
r
k
de
t
e
c
t
i
o
n
wa
s
do
n
e
us
i
n
g
t
h
e
de
e
p
ne
ur
a
l
n
e
t
wo
r
k
-
b
a
s
e
d
YO
L
O
a
l
go
r
i
t
hm
w
a
s
us
e
d.
I
t
c
a
n
a
l
s
o
de
t
e
c
t
ot
h
e
r
s
e
v
e
r
a
l
d
i
s
t
i
nc
t
o
bj
e
c
t
s
(
s
h
a
r
ks
,
r
a
y
s
,
s
ur
f
e
r
s
,
pa
dd
l
e
b
o
a
r
de
r
s
)
.
T
hi
s
a
l
e
r
t
i
n
g
s
y
s
t
e
m
i
s
t
r
a
i
n
e
d
b
a
s
e
d
o
n
a
s
i
n
g
l
e
b
e
a
c
h
l
o
c
a
t
i
o
n
whi
c
h
m
a
y
pe
r
f
o
r
m
we
ll
i
n
t
h
a
t
pa
r
t
i
c
u
l
a
r
l
o
c
a
t
i
o
n
a
n
d
n
e
e
d
t
o
i
m
pr
o
v
e
t
h
e
m
o
de
l
w
i
t
h
by
t
r
a
i
ni
ng
w
i
t
h
ot
h
e
r
l
o
c
a
t
i
o
n
da
t
a
.
P
r
o
p
o
s
e
d
t
w
o
a
l
go
r
i
t
hm
s
f
o
r
s
a
l
t
wa
t
e
r
c
r
o
c
o
di
l
e
de
t
e
c
t
i
o
n
dur
i
n
g
da
y
t
i
m
e
us
i
ng
m
u
l
t
i
-
f
e
a
t
ur
e
j
o
i
n
t
de
s
c
r
i
pt
o
r
[
12]
.
T
h
e
f
i
r
s
t
a
l
go
r
i
t
hm
us
e
s
f
e
a
t
ur
e
s
c
o
l
o
r
a
n
d
Ho
G
.
A
n
o
t
h
e
r
a
l
go
r
i
t
hm
m
a
ke
s
us
e
o
f
c
o
l
o
r
a
n
d
s
c
a
l
e
-
i
nva
r
i
a
n
t
f
e
a
t
ur
e
t
r
a
n
s
f
o
r
m
f
e
a
t
ur
e
de
s
c
r
i
pt
or
s
to
i
de
n
t
i
f
y
t
h
e
t
a
r
ge
t.
A
n
a
ppr
o
a
c
h
[
13]
f
o
r
de
t
e
c
t
i
n
g
t
h
e
c
a
pt
u
r
e
d
i
m
a
ge
s
o
f
dugo
n
gs
us
i
ng
a
pa
tt
e
r
n
r
e
c
o
gni
t
i
o
n
a
l
go
r
i
t
hm
t
h
a
t
m
a
ke
us
e
t
h
e
f
e
a
t
ur
e
s
f
o
r
m
t
h
e
c
a
pt
ur
e
d
i
m
a
ge
.
T
h
e
i
n
i
t
i
a
l
m
e
t
h
o
d
f
o
r
bl
o
b
d
e
t
e
c
t
i
o
n
c
o
m
bi
ne
s
m
o
r
ph
o
l
o
g
i
c
a
l
pr
o
c
e
dur
e
s
w
i
t
h
c
o
l
o
r
a
n
a
ly
s
i
s
.
T
h
e
s
e
c
o
n
d
m
e
t
h
o
d
e
m
p
l
o
y
s
a
s
h
a
pe
pr
of
i
li
ng
m
e
t
h
o
do
l
o
g
y
a
l
o
n
g
w
i
t
h
t
h
e
s
a
t
ur
a
t
i
o
n
c
h
a
nn
e
l
f
r
o
m
t
h
e
HSV
c
o
l
o
r
s
pa
c
e
to
i
de
n
t
i
f
y
m
a
mm
a
l
.
T
h
e
r
e
i
s
a
n
e
e
d
to
de
a
l
w
i
t
h
t
h
e
wa
t
e
r
t
ur
bi
d
i
t
y
,
w
i
n
d,
wa
v
e
s
pe
e
d
a
n
d
pe
r
i
o
d,
a
n
d
s
unl
i
g
h
t
whi
l
e
do
i
n
g
t
h
e
a
u
to
m
a
t
i
c
de
t
e
c
t
i
o
n
.
A
m
u
l
t
i
s
pe
c
t
r
a
l
i
m
a
g
i
ng
-
b
a
s
e
d
m
e
th
o
d
[
14
]
wa
s
pr
o
p
o
s
e
d
to
f
i
nd
o
u
t
t
h
e
gr
e
a
t
whi
t
e
s
h
a
r
ks
i
n
t
h
e
P
a
c
i
f
i
c
Oc
e
a
n
o
f
f
c
o
a
s
t
o
f
S
a
n
D
i
e
go
.
T
h
e
we
a
k
l
y
s
upe
r
vi
s
e
d
m
e
t
h
o
d
[
15]
to
de
t
e
c
t
t
h
e
m
a
r
i
ne
a
nim
a
l
s
f
r
o
m
t
h
e
a
e
r
i
a
l
im
a
g
e
s
pr
o
p
o
s
e
d
whi
c
h
do
e
s
n
ot
r
e
qui
r
e
to
s
pe
n
d
m
u
c
h
t
i
m
e
o
n
t
h
e
a
n
n
o
t
a
t
i
o
n
.
A
m
u
l
t
i
s
t
e
p
p
i
pe
li
ne
wa
s
pr
o
p
o
s
e
d
to
m
a
p
t
h
e
a
n
o
m
a
ly
m
a
ps
t
o
t
h
e
r
e
l
e
va
n
t
r
e
gi
o
n
pr
o
p
o
s
a
l
f
r
o
m
wh
e
r
e
t
h
e
o
bj
e
c
t
s
c
a
n
b
e
de
t
e
c
t
e
d
us
i
n
g
t
h
e
pa
t
c
h
d
i
s
t
r
i
b
ut
i
o
n
m
o
de
l
l
i
ng
m
e
t
h
o
d.
T
h
e
C
NN
n
e
t
w
o
r
k
m
o
de
l
[
16]
i
s
pr
o
p
o
s
e
d
f
o
r
t
h
e
a
uto
m
a
t
e
d
c
o
un
t
i
n
g
a
n
d
i
de
n
t
i
f
i
c
a
t
i
o
n
o
f
f
i
s
h
s
pe
c
i
e
s
.
P
r
e
s
e
n
t
e
d
a
n
o
v
e
l
d
e
e
p
n
e
ur
a
l
n
e
t
wo
r
k
m
o
de
l
[
17]
f
o
r
t
h
e
de
t
e
c
t
i
o
n
o
f
f
i
s
h
i
n
un
de
r
wa
t
e
r
e
n
vi
r
o
nm
e
n
t
s
a
n
d
c
o
n
duc
t
e
d
a
c
o
m
pa
r
a
t
i
v
e
a
n
a
ly
s
i
s
us
i
n
g
t
h
e
s
uppo
r
t
v
e
c
to
r
m
a
c
hi
ne
s
(
S
VM
)
c
l
a
s
s
if
i
e
r
.
A
m
e
t
h
o
d
f
o
r
qui
c
k
ly
a
n
d
a
c
c
ur
a
t
e
l
y
de
t
e
c
t
i
n
g
a
n
d
r
e
c
o
gni
s
i
ng
f
i
s
h
i
n
u
n
de
r
wa
t
e
r
ph
oto
gr
a
ph
s
us
i
n
g
t
h
e
f
a
s
t
R
-
C
NN
a
l
go
r
i
t
hm
wa
s
pr
o
p
o
s
e
d
i
n
[
18]
.
De
e
p
l
e
a
r
ni
ng
’
s
e
x
p
l
o
s
i
ve
gr
o
w
t
h
h
a
s
r
e
c
e
n
t
l
y
e
n
c
o
ur
a
ge
d
s
i
g
ni
f
i
c
a
n
t
t
h
e
or
e
t
i
c
a
l
a
d
v
a
n
c
e
m
e
n
t
s
a
n
d
r
e
a
l
-
wo
r
l
d
a
pp
li
c
a
t
i
o
n
s
o
f
c
o
m
put
e
r
-
vi
s
i
o
n
-
ba
s
e
d
un
de
r
wa
t
e
r
o
bj
e
c
t
r
e
c
o
gn
i
t
i
o
n
m
e
t
h
o
ds
.
I
n
t
hi
s
t
h
e
R
-
C
NN
a
n
d
m
a
s
k
R
C
NN
a
ppr
o
a
c
h
[
19]
,
de
e
p
n
e
ur
a
l
n
e
t
w
o
r
k
m
o
de
l
M
o
bil
e
Ne
t
V1
[
20]
wa
s
us
e
d
to
de
t
e
c
t
t
h
e
un
de
r
wa
t
e
r
o
bj
e
c
t
s
a
n
d
m
a
r
i
n
e
c
r
e
a
t
ur
e
s
.
M
o
de
l
s
in
[
21]
,
[
22
]
,
de
m
o
ns
t
r
a
t
e
d
a
hy
b
r
i
d
a
ppr
o
a
c
h
f
o
r
r
e
a
l
t
i
m
e
f
i
s
h
s
p
e
c
i
e
s
a
n
d
do
l
p
hi
n
s
pe
c
i
e
s
i
de
n
t
i
f
i
c
a
t
i
o
n
.
S
VM
b
a
s
e
d
a
ppr
o
a
c
h
e
s
w
e
r
e
us
e
d
i
n
v
a
r
i
o
us
pr
e
d
ict
i
o
n
t
a
s
k
[
23]
.
A
n
o
v
e
r
vi
e
w
o
f
s
e
ve
r
a
l
s
wa
r
m
i
n
t
e
ll
i
ge
n
c
e
m
e
t
h
o
ds
a
n
d
t
h
e
i
r
a
pp
l
i
c
a
t
i
o
ns
i
n
im
a
g
e
pr
o
c
e
s
s
i
n
g
wa
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
14
,
N
o.
1
,
A
pr
i
l
20
25
:
240
-
249
242
pr
e
s
e
n
t
e
d
i
n
[
24]
.
T
hi
s
pr
o
vi
de
d
a
n
o
ppo
r
t
un
i
t
y
t
o
i
de
n
t
i
f
y
a
n
d
c
h
o
o
s
e
t
h
e
e
s
s
e
n
t
i
a
l
o
pt
i
m
i
z
a
t
i
o
n
a
l
go
r
i
t
hm
f
o
r
t
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
t
a
s
k.
I
n
s
u
m
m
a
r
y
,
de
s
p
i
t
e
t
h
e
r
e
c
e
n
t
r
i
s
e
i
n
s
h
a
r
k
a
tt
a
c
ks
wo
r
l
dw
i
de
,
r
e
l
a
t
i
ve
l
y
f
e
w
e
f
f
o
r
t
s
h
a
v
e
be
e
n
m
a
de
t
o
a
u
to
m
a
t
e
t
h
e
de
t
e
c
t
i
o
n
pr
o
c
e
dur
e
gi
v
e
n
t
h
e
s
e
r
i
o
us
n
e
s
s
o
f
t
h
e
c
o
ns
e
que
n
c
e
.
T
h
e
m
a
jo
r
i
t
y
o
f
publi
c
a
t
i
o
n
s
i
n
t
h
e
l
i
t
e
r
a
t
ur
e
e
m
p
l
o
y
e
d
a
s
t
a
n
da
r
d
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
o
a
c
h
,
we
r
e
i
n
e
f
f
i
c
i
e
n
t
,
a
n
d
h
a
d
l
o
w
pr
e
c
i
s
i
o
n
.
I
n
t
hi
s
pa
pe
r
,
a
hy
b
r
i
d
a
ppr
o
a
c
h
f
o
r
pa
tt
e
r
n
r
e
c
o
gni
t
i
o
n
a
n
d
de
t
e
c
t
i
o
n
us
i
n
g
hi
s
t
o
gr
a
m
o
f
gr
a
d
i
e
n
t
s
(
H
o
G
)
,
C
NN
,
S
VM
a
l
go
r
i
t
hm
s
a
r
e
pr
o
po
s
e
d
to
f
i
nd
t
h
e
pa
t
t
e
r
n
i
n
t
h
e
m
a
r
i
n
e
a
ni
m
a
l
s
s
u
c
h
a
s
s
ha
r
ks
a
n
d
c
r
o
c
o
di
l
e
s
a
n
d
i
de
n
t
i
f
y
t
h
e
a
nim
a
l
s
r
e
a
c
hi
ng
t
h
e
s
e
a
s
h
o
r
e
a
n
d
wa
r
n
t
h
e
s
w
im
m
e
r
s
a
n
d
s
ur
f
e
r
s
n
e
a
r
t
h
e
c
o
a
s
t
a
l
r
e
g
i
o
n
.
2.
RE
S
E
AR
CH
M
E
T
HO
D
I
n
t
h
i
s
wo
r
k
a
hy
b
r
i
d
m
o
de
l
to
i
de
n
t
i
f
y
a
n
d
c
l
a
s
s
if
y
t
h
e
m
a
r
i
ne
a
nim
a
l
s
u
c
h
a
s
s
h
a
r
k
a
n
d
c
r
o
c
o
di
l
e
i
s
s
h
o
w
n
i
n
F
i
g
ur
e
1.
T
h
e
pr
o
p
o
s
e
d
m
o
de
l
m
a
ke
us
e
o
f
t
h
e
b
e
s
t
o
f
t
h
e
f
u
n
c
t
i
o
n
a
li
t
i
e
s
f
r
o
m
t
h
e
C
NN
a
n
d
S
VM
.
T
h
e
Ho
G
f
e
a
t
ur
e
e
x
t
r
a
c
to
r
e
x
tr
a
c
t
s
t
h
e
hi
s
to
gr
a
m
gr
a
di
e
n
t
s
f
r
o
m
t
h
e
i
m
a
ge
a
n
d
C
NN
s
uppo
r
t
s
t
h
e
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
a
n
d
t
h
e
o
p
t
i
m
i
z
e
d
S
VM
pe
r
f
o
r
m
s
t
h
e
c
las
s
if
i
c
a
t
i
o
n
t
a
s
k
.
F
i
gur
e
1
.
P
a
tt
e
r
n
r
e
c
o
gni
t
i
o
n
m
o
de
l
2
.
1.
Dat
a
s
e
t
T
h
e
da
t
a
’
s
r
e
l
e
v
a
n
t
to
t
h
e
t
r
a
i
ni
ng
a
n
d
t
h
e
t
e
s
t
i
n
g
o
f
t
h
e
m
o
de
l
wa
s
c
o
l
l
e
c
t
e
d
f
r
o
m
va
r
i
o
us
s
o
ur
c
e
s
[
25
]
-
[
27
]
.
T
h
e
i
m
a
g
e
da
t
a
s
e
t
c
o
n
s
i
s
t
s
o
f
s
a
l
t
wa
t
e
r
c
r
o
c
o
di
l
e
a
n
d
s
h
a
r
k
i
m
a
ge
s
e
a
c
h
o
f
700
i
m
a
g
e
s
.
Da
t
a
a
ugm
e
n
t
a
t
i
o
n
t
e
c
h
ni
que
s
a
r
e
e
m
p
l
o
y
e
d
to
e
x
pa
n
d
t
h
e
da
t
a
s
e
t,
r
e
s
u
l
t
i
n
g
i
n
a
tot
a
l
o
f
1
,
200
p
h
o
to
s
f
o
r
e
a
c
h
c
a
t
e
gor
y
.
T
h
e
da
t
a
s
e
t
i
s
pa
r
t
i
t
i
o
n
e
d
i
n
t
o
t
h
r
e
e
s
ubs
e
t
s
:
t
r
a
i
n
,
t
e
s
t,
a
n
d
v
a
l
i
d
a
t
i
o
n
.
T
h
e
d
i
s
t
r
i
b
ut
i
o
n
i
s
a
s
f
o
l
l
o
ws
:
70%
o
f
t
h
e
da
t
a
i
s
a
l
l
o
c
a
t
e
d
f
o
r
tr
a
i
ni
ng
t
h
e
m
o
de
l
,
10%
i
s
a
l
l
o
c
a
t
e
d
f
o
r
m
o
de
l
v
a
li
da
t
i
o
n
,
a
n
d
t
h
e
r
e
m
a
i
n
i
ng
20%
i
s
a
l
l
o
c
a
t
e
d
f
o
r
m
o
de
l
t
e
s
t
i
n
g.
2
.
2
.
P
r
e
p
r
oc
e
s
s
in
g
T
h
e
s
t
a
n
da
r
d
da
t
a
a
ug
m
e
n
t
a
t
i
o
n
t
e
c
h
ni
que
s
s
uc
h
a
s
f
li
pp
i
ng,
r
ot
a
t
i
o
n
,
s
h
e
a
r
a
n
d
b
r
i
g
h
t
n
e
s
s
c
h
a
n
ge
s
a
r
e
a
pp
l
i
e
d
t
o
t
h
e
a
c
t
ua
l
da
t
a
s
e
t
to
i
n
c
r
e
a
s
e
t
h
e
s
i
z
e
o
f
t
h
e
da
t
a
s
e
t.
T
h
e
i
m
a
ge
s
a
r
e
r
e
c
e
i
ve
d
f
r
o
m
t
h
e
v
a
r
i
o
us
s
o
ur
c
e
s
h
e
nc
e
to
pr
o
vi
de
unif
o
r
m
i
t
y
a
m
o
n
g
t
h
e
da
t
a
,
a
l
l
t
h
e
i
m
a
ge
s
a
r
e
r
e
s
i
z
e
d
to
t
h
e
r
a
t
i
o
o
f
1:2
(
w
i
dt
h
:he
i
g
h
t
)
.
T
h
e
i
m
a
ge
s
a
r
e
c
o
nv
e
r
t
e
d
i
n
t
o
g
r
e
y
s
c
a
l
e
i
m
a
ge
s
b
e
f
o
r
e
e
x
t
r
a
c
t
i
n
g
t
h
e
f
e
a
t
ur
e
s
f
r
o
m
t
h
e
im
a
ge
s
.
2
.
3
.
F
e
at
u
r
e
e
x
t
r
ac
t
ion
2
.
3.
1.
Hi
s
t
ogr
am
of
g
r
ad
ient
s
f
e
at
u
r
e
Us
i
n
g
t
h
e
Ho
G
f
e
a
t
ur
e
de
s
c
r
i
pt
or
,
o
n
e
c
a
n
e
x
t
r
a
c
t
f
e
a
t
ur
e
s
f
r
o
m
a
n
i
m
a
ge
to
f
a
c
i
li
t
a
t
e
o
bj
e
c
t
de
t
e
c
t
i
o
n
.
I
t
e
m
p
h
a
s
i
s
e
s
t
h
e
f
o
r
m
a
n
d
c
o
m
po
s
i
t
i
o
n
o
f
a
n
o
bj
e
c
t
.
I
t
e
x
t
r
a
c
t
s
t
h
e
gr
a
di
e
n
t
a
n
d
o
r
i
e
n
t
a
t
i
o
n
o
f
t
h
e
e
dge
s
a
n
d
t
h
e
s
e
a
r
e
c
a
l
c
u
l
a
t
e
d
i
n
l
o
c
a
li
z
e
d
po
r
t
i
o
n
s
.
F
o
r
e
a
c
h
o
f
t
h
e
s
e
r
e
g
i
o
ns
a
hi
s
t
o
gr
a
m
wa
s
g
e
n
e
r
a
t
e
d
whi
c
h
pr
o
vi
de
t
h
e
i
m
po
r
t
a
n
t
f
e
a
t
ur
e
s
i
n
t
h
e
i
m
a
ge
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
hy
br
id
appr
oac
h
o
f
pat
ter
n
r
e
c
ognit
ion
to
de
tec
t
mar
ine
animals
(
V
ij
ay
alaks
hmi
B
alachandr
an
)
243
C
a
l
c
u
l
a
t
i
n
g
t
h
e
gr
a
d
i
e
n
t
s
o
f
a
pi
c
t
ur
e
e
x
po
s
e
s
t
h
e
s
pe
c
i
f
i
c
a
r
e
a
s
wh
e
r
e
t
h
e
i
n
t
e
n
s
i
t
y
o
f
p
i
x
e
l
gr
a
d
i
e
n
t
s
un
de
r
go
e
s
a
l
t
e
r
a
t
i
o
n
s
.
T
h
e
ke
r
n
e
l
s
ut
i
l
i
s
e
d
f
o
r
c
o
m
put
i
n
g
t
h
e
gr
a
di
e
n
t
s
c
o
n
s
i
s
t
o
f
a
v
e
r
t
i
c
a
l
gr
a
d
i
e
n
t
ke
r
n
e
l
:
[
-
1,
0,
1]
a
n
d
a
h
o
r
i
z
o
n
t
a
l
gr
a
d
i
e
n
t
ke
r
n
e
l
:
[
-
1]
.
T
h
e
gr
a
di
e
n
t
o
f
t
h
e
pi
c
t
ur
e
i
s
de
t
e
r
m
i
ne
d
by
c
o
m
bi
n
i
ng
t
h
e
ve
r
t
i
c
a
l
gr
a
d
i
e
n
t
(
Gx
)
a
n
d
t
h
e
ho
r
i
z
o
n
t
a
l
gr
a
d
i
e
n
t
(
G
y
)
,
us
i
n
g
t
h
e
m
a
g
ni
t
ude
a
n
d
a
n
g
l
e
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
i
m
a
ge
,
a
s
s
pe
c
if
i
e
d
i
n
(
1)
a
n
d
(
2)
.
=
√
2
+
2
(
1)
(
)
=
|
−
1
(
)
|
(
2)
Af
t
e
r
t
h
e
gr
a
di
e
n
t
i
s
c
a
l
c
u
l
a
t
e
d
a
hi
s
t
o
gr
a
m
o
f
9
bi
n
s
wa
s
f
o
r
m
e
d
to
bi
n
t
h
e
gr
a
di
e
n
t
m
a
g
ni
t
ud
e
v
a
l
ue
s
w.
r
.
t
gr
a
di
e
n
t
di
r
e
c
t
i
o
n
.
Gr
a
di
e
n
t
l
e
v
e
l
s
m
igh
t
v
a
r
y
de
pe
n
d
i
ng
o
n
l
i
g
h
t
i
n
g
a
n
d
f
o
r
e
gr
o
un
d/b
a
c
kgr
o
un
d
c
o
n
t
r
a
s
t
h
e
n
c
e
bl
o
c
k
n
o
r
m
a
li
z
a
t
i
o
n
i
s
a
pp
li
e
d.
F
i
n
a
l
ly
,
t
h
e
hi
s
t
o
gr
a
m
i
s
n
o
r
m
a
l
i
z
e
d
to
f
o
r
m
t
h
e
f
e
a
t
ur
e
v
e
c
t
o
r
whi
c
h
r
e
pr
e
s
e
n
t
t
h
e
c
o
n
c
i
s
e
a
n
d
s
uc
c
i
nc
t
r
e
pr
e
s
e
n
t
a
t
i
o
n
o
f
pa
r
t
i
c
u
l
a
r
pa
t
c
h
e
s
o
f
t
h
e
i
m
a
ge
s
.
T
h
e
pi
c
t
ur
e
s
t
h
a
t
w
i
ll
b
e
l
e
a
r
n
e
d
a
r
e
c
o
nv
e
r
t
e
d
i
n
t
o
a
n
a
r
r
a
y
o
f
Ho
G
f
e
a
t
ur
e
s
.
T
h
e
c
h
o
s
e
n
f
e
a
t
ur
e
s
m
a
y
e
n
c
o
m
pa
s
s
t
h
e
c
r
uc
i
a
l
i
n
f
o
r
m
a
t
i
o
n
f
r
o
m
t
h
e
i
nput
da
t
a
,
s
o
e
n
a
bli
ng
t
h
e
a
c
c
o
m
p
li
s
hm
e
n
t
o
f
t
h
e
i
n
t
e
n
de
d
j
o
b
ut
i
li
s
i
ng
a
s
m
a
ll
e
r
qua
n
t
i
t
y
o
f
da
t
a
i
ns
t
e
a
d
o
f
t
h
e
o
r
i
g
i
na
l
u
nm
o
d
if
i
e
d
da
t
a
.
2
.
4
.
Convol
u
t
ion
al
n
e
u
r
al
n
e
t
wo
r
k
T
h
e
C
NN
i
s
e
m
p
l
o
y
e
d
i
n
t
a
s
ks
i
nv
o
l
vi
n
g
pi
c
t
ur
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
a
n
d
de
tec
t
i
o
n
.
T
h
i
s
i
s
b
e
c
a
us
e
t
h
e
c
o
n
v
o
l
ut
i
o
n
l
a
y
e
r
o
f
t
h
e
C
NN
e
f
f
e
c
t
i
v
e
l
y
r
e
duc
e
s
t
h
e
hi
gh
d
i
m
e
n
s
i
o
n
a
li
t
y
o
f
t
h
e
da
t
a
.
C
NN
f
un
c
t
i
o
n
s
a
s
a
f
e
a
t
ur
e
e
x
tr
a
c
tor
t
h
r
ough
o
u
t
t
h
e
tr
a
i
ni
n
g
ph
a
s
e
.
T
h
e
f
e
a
t
u
r
e
e
x
t
r
a
c
tor
e
m
p
l
o
y
e
d
by
C
NN
c
o
n
s
i
s
t
s
of
d
i
s
t
i
n
c
t
n
e
ur
a
l
n
e
t
w
o
r
k
a
r
c
hi
t
e
c
t
u
r
e
s
,
wi
t
h
t
h
e
we
i
gh
t
s
b
e
i
n
g
e
s
t
a
b
l
i
s
h
e
d
by
t
h
e
tr
a
i
ni
n
g
pr
o
c
e
s
s
.
De
s
pi
t
e
t
h
e
a
dde
d
c
o
m
p
l
e
xi
t
y
o
f
t
h
e
l
e
a
r
ni
ng
a
ppr
o
a
c
h
,
C
NN
pr
o
v
i
de
s
e
n
h
a
n
c
e
d
pi
c
t
ur
e
r
e
c
ogni
t
i
o
n
by
us
i
n
g
a
de
e
pe
r
n
e
ur
a
l
n
e
t
w
or
k
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
wi
t
h
m
o
r
e
l
a
y
e
r
s
.
C
N
N
i
s
a
c
o
m
put
a
t
i
o
n
a
l
m
o
de
l
de
s
i
g
n
e
d
to
e
x
t
r
a
c
t
r
e
l
e
v
a
n
t
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
f
r
o
m
i
n
put
i
m
a
ge
s
.
T
h
e
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
n
e
t
w
or
k
u
t
i
l
i
s
e
s
t
h
e
i
n
put
pi
c
t
ur
e
a
s
i
t
s
f
i
r
s
t
r
e
f
e
r
e
n
c
e
.
T
h
e
C
NN
m
o
de
l
’
s
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
n
e
t
w
or
k
i
s
c
o
m
po
s
e
d
o
f
a
s
e
r
i
e
s
o
f
c
o
n
v
o
l
ut
i
o
n
a
l
l
a
y
e
r
s
a
n
d
p
oo
l
i
n
g
l
a
y
e
r
s
s
t
a
c
ke
d
toge
t
h
e
r
.
T
h
e
c
o
n
v
o
l
ut
i
o
n
l
a
y
e
r
pe
r
f
o
r
m
s
t
h
e
c
o
n
v
o
l
ut
i
o
n
o
pe
r
a
t
i
o
n
us
i
n
g
f
i
lt
e
r
s
a
nd
tr
a
n
s
f
o
r
m
s
t
h
e
i
n
put
i
m
a
ge
.
C
o
n
v
o
l
ut
i
o
n
i
s
a
pr
oc
e
s
s
i
n
whi
c
h
a
s
ke
r
n
e
l
o
r
f
i
l
t
e
r
i
s
m
o
v
e
d
o
v
e
r
t
h
e
i
m
a
ge
a
n
d
tr
a
n
s
f
o
r
m
a
t
i
o
n
t
a
ke
s
pl
a
c
e
b
a
s
e
d
o
n
t
h
e
v
a
l
ue
s
i
n
t
h
e
f
i
l
t
e
r
.
F
e
a
t
ur
e
m
a
ps
a
r
e
c
a
l
c
u
l
a
t
e
d
b
a
s
e
d
o
n
(
3)
:
[
,
]
=
(
∗
ℎ
)
[
,
]
=
∑
∑
ℎ
[
,
]
[
−
,
−
]
(
3)
w
h
e
r
e
f
de
n
otes
t
h
e
i
n
put
i
m
a
ge
,
h
i
n
d
i
c
a
t
e
s
t
h
e
k
e
r
n
e
l
,
m
a
n
d
n
i
n
d
i
c
a
t
e
s
t
h
e
r
o
ws
a
n
d
c
o
l
u
m
n
s
o
f
th
e
r
e
s
ul
t
m
a
t
r
i
x
.
T
h
e
p
oo
l
i
n
g
l
a
y
e
r
o
f
t
h
e
n
e
ur
a
l
n
e
t
w
or
k
r
e
tai
n
s
t
h
e
f
i
n
e
de
t
a
i
l
s
i
n
t
h
e
i
n
put
a
n
d
e
n
h
a
n
c
e
s
t
h
e
pr
oc
e
s
s
o
f
r
e
duc
i
n
g
d
i
m
e
n
s
i
o
ns
by
c
a
l
c
u
l
a
t
i
n
g
t
h
e
a
v
e
r
a
ge
or
l
o
we
s
t
v
a
l
ue
,
de
pe
n
di
n
g
o
n
t
h
e
s
pe
c
i
f
i
c
r
e
qui
r
e
m
e
n
t
s
o
f
t
h
e
a
ppl
i
c
a
t
i
o
n
.
Ul
t
i
m
a
t
e
l
y
,
a
f
ul
ly
c
o
n
n
e
c
t
e
d
l
a
y
e
r
i
s
a
dde
d
to
t
h
e
f
i
n
a
l
l
a
y
e
r
i
n
o
r
d
e
r
to
a
c
qui
r
e
kn
o
wl
e
dge
f
r
o
m
t
h
e
o
u
tpu
t
o
f
t
h
e
p
r
e
c
e
di
n
g
l
a
y
e
r
a
n
d
c
a
r
r
y
o
u
t
t
h
e
d
e
s
i
r
e
d
ope
r
a
t
i
o
n
.
T
h
e
c
o
n
v
o
l
ut
i
o
n
a
n
d
p
oo
l
i
ng
l
a
y
e
r
s
i
n
a
C
NN
i
n
h
e
r
e
n
t
l
y
o
pe
r
a
te
o
n
a
tw
o
-
di
m
e
n
s
i
o
na
l
pl
a
n
e
due
to
t
h
e
pr
i
m
a
r
y
e
m
p
h
a
s
i
s
o
f
C
NN
s
on
i
m
a
ge
pr
o
c
e
s
s
i
n
g.
2
.
5
.
S
u
p
p
or
t
ve
c
t
or
m
ac
h
in
e
s
S
VM
c
a
n
e
f
f
e
c
t
i
v
e
ly
h
a
n
d
l
e
b
o
t
h
c
o
n
t
i
n
uo
us
a
n
d
c
a
t
e
g
o
r
i
c
a
l
da
t
a
.
S
VM
c
o
n
s
t
r
uc
t
s
a
hy
pe
r
p
l
a
ne
i
n
a
hi
g
h
-
d
i
m
e
ns
i
o
n
a
l
s
pa
c
e
to
s
e
pa
r
a
t
e
di
f
f
e
r
e
n
t
c
l
a
s
s
e
s
.
T
h
e
S
VM
c
o
n
s
t
r
uc
t
s
a
n
o
p
t
i
m
a
l
hy
pe
r
p
l
a
ne
t
h
r
o
ugh
a
n
i
t
e
r
a
t
i
v
e
pr
o
c
e
s
s
i
n
o
r
de
r
to
m
i
n
im
i
s
e
e
r
r
or
.
T
h
e
pr
i
m
a
r
y
o
bj
e
c
t
i
v
e
o
f
S
VM
i
s
to
de
t
e
r
m
i
ne
t
h
e
m
a
xim
u
m
m
a
r
g
i
na
l
hy
pe
r
p
l
a
ne
(
M
M
H)
t
h
a
t
e
f
f
e
c
t
i
v
e
ly
d
i
v
i
d
e
s
t
h
e
da
t
a
s
e
t
i
n
t
o
s
e
v
e
r
a
l
gr
o
ups
.
S
VM
us
e
d
a
hype
r
p
l
a
n
e
a
s
a
de
c
i
s
i
o
n
b
o
un
da
r
y
t
o
di
f
f
e
r
e
n
t
i
a
t
e
o
r
c
l
a
s
s
i
f
y
b
e
t
we
e
n
t
w
o
g
r
o
ups
.
S
VM
i
s
c
o
m
m
o
nly
e
m
p
l
o
ye
d
i
n
t
h
e
t
a
s
ks
o
f
i
m
a
ge
c
l
a
s
s
i
f
i
c
a
t
i
o
n
a
n
d
o
bj
e
c
t
de
t
e
c
t
i
o
n
.
T
h
e
da
t
a
p
o
i
n
t
s
t
h
a
t
a
r
e
i
n
c
l
o
s
e
s
t
pr
o
xi
m
i
t
y
t
o
th
e
hy
pe
r
p
l
a
n
e
a
r
e
r
e
f
e
r
r
e
d
to
a
s
s
uppor
t
v
e
c
to
r
s
.
B
y
c
a
l
c
u
l
a
t
i
n
g
t
h
e
m
a
r
g
i
ns
,
t
h
e
s
e
po
i
n
t
s
w
i
l
l
m
o
r
e
a
c
c
ur
a
t
e
l
y
d
e
l
i
ne
a
t
e
t
h
e
di
vi
d
i
ng
l
i
ne
.
T
h
e
s
e
n
o
t
i
o
n
s
a
r
e
o
f
gr
e
a
t
e
r
i
m
po
r
t
a
n
c
e
i
n
t
h
e
c
o
n
s
t
r
uc
t
i
o
n
o
f
t
h
e
c
l
a
s
s
if
i
e
r
.
A
hy
p
e
r
p
l
a
n
e
i
s
a
p
l
a
n
e
us
e
d
to
m
a
k
e
de
c
i
s
i
o
ns
by
s
e
pa
r
a
t
i
n
g
a
gr
o
up
o
f
o
bj
e
c
t
s
w
i
t
h
d
i
s
t
i
nc
t
c
l
a
s
s
m
e
m
be
r
s
hi
p
s
.
I
t
i
s
de
f
i
ne
d
by
t
he
s
e
t
o
f
po
i
n
t
s
x
t
h
a
t
s
a
t
i
s
f
i
e
s
(
4)
.
.
+
=
0
(
4)
Any
po
i
n
t
x
t
h
a
t
d
oe
s
n
ot
s
a
t
i
s
f
y
t
h
e
(
)
e
q
ua
t
i
o
n
m
us
t
b
e
l
o
c
a
t
e
d
o
n
o
n
e
o
f
t
h
e
t
w
o
s
i
de
s
o
f
t
h
e
hy
pe
r
p
l
a
n
e
.
C
h
oo
s
e
tw
o
a
d
di
t
i
o
n
a
l
hy
pe
r
p
l
a
n
e
s
(
H₁,
H₂)
t
h
a
t
a
r
e
e
qua
l
l
y
s
pa
c
e
d
f
r
o
m
H₀,
m
a
ki
n
g
s
ur
e
t
h
a
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
14
,
N
o.
1
,
A
pr
i
l
20
25
:
240
-
249
244
e
a
c
h
da
t
a
p
o
i
n
t
i
s
o
n
t
h
e
c
or
r
e
c
t
s
i
de
wi
t
h
n
o
da
t
a
i
n
b
e
t
we
e
n
.
T
h
e
r
e
f
o
r
e
,
i
t
i
s
p
o
s
s
i
bl
e
to
de
f
i
ne
t
h
e
hy
pe
r
p
l
a
n
e
s
(
H₁,
H₂)
us
i
n
g
(
5)
a
n
d
(
6
)
.
(
1
)
.
+
≥
1
=
1
(
5)
(
2
)
.
+
≤
1
=
−
1
(
6)
T
h
e
s
e
t
w
o
h
y
pe
r
p
l
a
n
e
s
c
a
n
b
e
c
o
m
bi
n
e
d
i
n
to
a
s
i
n
g
l
e
e
qua
t
i
o
n
f
o
r
a
s
e
pa
r
a
t
i
n
g
hy
pe
r
pl
a
n
e
us
i
n
g
(
7
)
.
(
.
+
)
≥
1
∀
∈
{
1
,
2
,
…
.
,
}
(
7)
T
h
e
o
p
t
i
m
a
l
s
e
pa
r
a
t
i
n
g
hy
pe
r
pl
a
n
e
i
s
t
h
e
s
o
l
ut
i
o
n
th
a
t
i
s
f
a
r
t
h
e
s
t
a
wa
y
f
r
o
m
t
h
e
c
l
o
s
e
s
t
da
t
a
p
o
i
n
t
or
i
n
ot
h
e
r
te
r
m
s
m
a
xim
i
z
e
s
t
h
e
m
a
r
g
i
n
.
I
t
c
a
n
a
l
s
o
b
e
i
m
a
g
i
n
e
d
a
s
a
pa
r
a
l
l
e
l
l
i
n
e
c
utt
i
n
g
t
h
e
m
a
r
g
i
n
i
n
t
w
o
h
a
l
v
e
s
.
T
h
e
m
a
i
n
o
bj
e
c
t
i
v
e
f
u
n
c
t
i
o
n
i
s
de
f
i
ne
d
by
(
8)
.
(
,
)
=
1
2
‖
‖
2
+
1
∑
m
a
x
(
0
,
1
−
=
1
(
.
+
)
)
(
8)
T
h
e
f
i
r
s
t
t
e
r
m
i
s
b
a
s
i
c
a
l
ly
r
e
s
po
n
s
i
bl
e
f
o
r
m
a
xi
m
i
z
i
n
g
t
h
e
m
a
r
gi
n
,
e
x
pr
e
s
s
e
d
a
s
a
m
i
ni
mi
z
a
t
i
o
n
pr
o
b
l
e
m
w
i
t
h
a
n
a
dde
d
r
e
gul
a
r
i
z
a
t
i
o
n
pa
r
a
m
e
t
e
r
λ
.
T
h
e
s
e
c
o
n
d
ter
m
i
s
t
h
e
de
f
i
ni
t
i
o
n
o
f
a
s
e
pa
r
a
t
i
n
g
hy
pe
r
p
l
a
n
e
i
s
a
l
o
s
s
f
un
c
t
i
o
n
c
a
l
l
e
d
t
h
e
Hi
n
ge
l
o
s
s
.
T
hi
s
ter
m
i
s
r
e
s
p
o
n
s
i
bl
e
f
o
r
e
n
s
ur
i
n
g
t
h
a
t
t
h
e
m
o
de
l
pr
e
di
c
t
t
h
e
c
or
r
e
c
t
c
l
a
s
s
l
a
b
e
l
w
i
t
h
e
n
o
ugh
m
a
r
g
i
n
.
F
o
r
e
x
a
m
p
l
e
,
i
f
y
ᵢ
=
1
a
n
d
xᵢ
i
s
c
o
r
r
e
c
t
l
y
c
l
a
s
s
i
f
i
e
d,
c
a
l
c
u
l
a
t
i
n
g
t
h
e
h
i
n
ge
l
o
s
s
wi
ll
r
e
s
u
l
t
i
n
z
e
r
o
s
i
n
c
e
max
(
0
,
1
–
1)
=
0
.
H
o
we
v
e
r
,
i
f
t
h
e
c
l
a
s
s
l
a
b
e
l
i
s
f
a
l
s
e
l
y
pr
e
d
i
c
t
e
d
t
h
e
h
i
n
ge
l
o
s
s
wi
l
l
r
e
s
u
l
t
i
n
a
v
a
l
ue
g
r
e
a
t
e
r
t
h
a
n
z
e
r
o.
T
h
e
l
o
s
s
f
un
c
t
i
o
n
c
a
n
b
e
o
p
t
i
m
i
z
e
d
w
i
t
h
g
r
a
di
e
n
t
de
s
c
e
n
t
by
m
a
k
i
n
g
s
m
a
l
l
s
t
e
ps
i
n
t
h
e
opp
o
s
i
t
e
di
r
e
c
t
i
o
n
to
m
i
nim
i
z
e
t
h
e
l
o
s
s
.
S
VM
i
s
a
m
a
xim
a
l
m
a
r
g
i
n
c
l
a
s
s
i
f
i
e
r
w
hi
c
h
i
s
b
a
s
ica
l
ly
p
e
r
f
o
r
m
i
ng
b
e
t
t
e
r
o
n
l
i
ne
a
r
ly
s
e
pa
r
a
bl
e
da
t
a
.
T
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
m
a
r
i
ne
a
ni
m
a
l
s
uc
h
a
s
s
h
a
r
k
a
n
d
s
a
l
t
wa
t
e
r
c
r
o
c
o
di
l
e
s
c
a
n
b
e
pe
r
f
o
r
m
e
d
us
in
g
S
VM
s
i
nc
e
t
h
e
da
t
a
s
e
t
i
s
li
ne
a
r
ly
s
e
p
a
r
a
bl
e
t
ha
t
m
a
xim
i
z
e
s
t
h
e
de
c
i
s
i
o
n
b
o
un
d
a
r
y
b
e
t
we
e
n
t
h
e
t
wo
c
l
a
s
s
e
s
b
y
f
i
nd
i
n
g
t
h
e
o
pt
i
m
a
l
s
e
pa
r
a
t
i
n
g
hy
p
e
r
p
l
a
ne
.
T
h
e
S
VM
h
a
v
e
b
e
e
n
t
r
a
i
n
e
d
a
n
d
t
e
s
t
e
d
wi
t
h
t
h
e
Ho
G
f
e
a
t
ur
e
e
x
t
r
a
c
t
e
d
f
r
o
m
t
h
e
im
a
ge
d
a
t
a
s
e
t
s
a
n
d
s
uppo
r
t
v
e
c
to
r
c
l
a
s
s
i
f
i
e
r
(
S
VC
)
i
s
us
e
d
i
n
t
h
e
S
V
M
m
o
de
l
.
He
r
e
t
h
e
c
l
a
s
s
if
i
c
a
t
i
o
n
wo
r
ks
b
a
s
e
d
o
n
t
h
e
l
i
ne
a
r
da
t
a
a
n
d
i
t
b
a
s
e
d
o
n
t
h
e
m
a
r
g
i
n
c
l
a
s
s
i
f
i
e
r
.
2
.
6
.
P
ar
t
icle
s
wa
r
m
op
t
im
i
z
at
ion
T
h
e
P
a
r
t
i
c
l
e
s
wa
r
m
o
pt
i
m
i
z
a
t
i
o
n
(
PSO
)
a
l
go
r
i
t
hm
i
s
a
c
o
m
put
e
r
m
e
t
h
o
d
t
h
a
t
dr
a
ws
i
t
s
i
n
s
p
i
r
a
t
i
o
n
f
r
o
m
t
h
e
c
o
o
pe
r
a
t
i
v
e
m
o
v
e
m
e
n
t
o
f
na
t
ur
a
l
s
pe
c
i
e
s
,
i
n
c
l
ud
i
ng
f
i
s
h
a
n
d
bi
r
d
s
,
to
a
c
c
o
m
p
li
s
h
a
c
o
m
m
o
n
o
bj
e
c
t
i
v
e
.
I
n
Al
go
r
i
t
hm
1
e
x
p
l
a
i
ns
t
h
e
P
S
O
i
nv
o
l
ve
s
s
e
a
r
c
hi
ng
a
c
r
o
s
s
t
h
e
s
o
l
ut
i
o
n
s
pa
c
e
o
f
a
pr
o
bl
e
m
w
i
t
h
a
gr
o
up
o
f
pa
r
t
i
c
l
e
s
(
r
e
pr
e
s
e
n
t
i
n
g
pot
e
n
t
i
a
l
s
o
l
ut
i
o
n
s
)
.
B
a
s
e
d
o
n
i
t
s
o
wn
b
e
s
t
-
kn
o
wn
a
ns
we
r
(
pe
r
s
o
n
a
l
b
e
s
t
)
a
n
d
t
h
e
b
e
s
t
s
o
l
ut
i
o
n
f
o
un
d
by
t
h
e
e
n
t
i
r
e
gr
o
up
(
gl
o
b
a
l
b
e
s
t
)
,
e
a
c
h
pa
r
t
i
c
l
e
m
o
di
f
i
e
s
i
t
s
p
o
s
i
t
i
o
n
.
P
a
r
t
i
c
l
e
s
c
a
n
c
o
n
v
e
r
ge
o
v
e
r
i
t
e
r
a
t
i
o
n
s
t
o
wa
r
ds
i
d
e
a
l
s
o
l
ut
i
o
ns
t
h
a
n
k
s
to
t
h
i
s
c
oo
pe
r
a
t
i
v
e
m
o
v
e
m
e
n
t
.
Al
go
r
i
t
hm
1
.
T
h
e
P
S
O
Initialize random number of particles
For each particle
Do
Initialize particle position x
i
Initialize the initial position x
i
as the best known position pbest
i
Update the swarms best position g
best
= pbest
i
, if fitness(pbest
i
) <fitness(g
best
)
Initialize the particles velocity as v
i
Repeat until a termination criterion is met:
Update the particle velocity
Update the particles position
Calculate
Update p
articles best known position as pbest
i
=x
i
, if
fitness(x
i
)<fitness(pbest
i
)
Update the swarms best known position as g
best
=pbest
i
, if
fitness(pbest
i
)<fgbest)
Return the best particle of swarm
w
h
e
r
e
x
i
a
n
d
v
i
de
n
o
t
e
s
t
h
e
po
s
i
t
i
o
n
a
n
d
v
e
l
o
c
i
t
y
o
f
pa
r
t
i
c
l
e
i
.
F
i
t
ne
s
s
(
x
i
)
de
n
o
t
e
s
t
h
e
f
i
t
n
e
s
s
f
u
n
c
t
i
o
n
t
o
f
i
n
d
t
h
e
pa
r
t
i
c
l
e
m
a
xi
m
u
m
f
i
t
n
e
s
s
va
l
u
e
.
T
h
e
pa
r
a
m
e
t
e
r
s
o
f
t
h
e
S
VM
m
o
de
l
i
s
o
p
t
i
mi
z
e
d
f
o
r
th
e
i
m
a
g
e
c
l
a
s
s
if
i
c
a
t
i
o
n
t
a
s
k.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
hy
br
id
appr
oac
h
o
f
pat
ter
n
r
e
c
ognit
ion
to
de
tec
t
mar
ine
animals
(
V
ij
ay
alaks
hmi
B
alachandr
an
)
24
5
2
.
6.
1.
P
S
O
-
S
V
M
o
p
t
im
iz
at
ion
T
a
bl
e
1
l
i
s
t
s
t
h
e
pa
r
a
m
e
t
e
r
s
p
i
c
ke
d
f
o
r
o
p
t
i
mi
z
a
t
i
o
n
to
ge
t
h
e
r
w
i
t
h
t
h
e
v
a
l
u
e
s
o
f
i
t
s
s
e
a
r
c
h
s
pa
c
e
.
T
h
e
pa
r
a
m
e
t
e
r
s
us
e
d
to
c
o
n
s
t
r
uc
t
t
h
e
pa
r
t
i
c
l
e
s
f
o
r
o
p
t
i
mi
z
a
t
i
o
n
,
e
a
c
h
o
f
whi
c
h
r
e
pr
e
s
e
n
t
s
a
pot
e
n
t
i
a
l
s
o
l
ut
i
o
n
,
a
r
e
us
e
d
h
e
r
e
t
o
i
ni
t
i
a
li
z
e
t
h
e
a
l
go
r
i
t
hm
.
I
n
t
hi
s
c
a
s
e
,
a
pa
r
t
i
c
l
e
i
s
s
pe
c
i
f
i
e
d
w
i
t
h
t
h
r
e
e
po
s
i
t
i
o
ns
,
e
a
c
h
o
f
whi
c
h
c
o
r
r
e
s
po
n
ds
to
a
n
o
p
t
i
mi
z
a
t
i
o
n
t
a
r
ge
t
pa
r
a
m
e
t
e
r
.
T
h
e
f
o
l
l
o
w
i
ng
de
s
c
r
i
be
s
h
o
w
S
VM
i
s
o
p
t
i
mi
z
e
d
us
i
ng
P
S
O.
a)
P
r
o
vi
d
e
t
h
e
da
t
a
to
t
h
e
S
VM
a
s
t
r
a
i
ni
ng
i
n
put
.
b)
C
r
e
a
t
e
t
h
e
po
pul
a
t
i
o
n
a
n
d
s
e
t
t
h
e
P
S
O
pa
r
a
m
e
t
e
r
s
to
t
h
e
i
r
i
ni
t
i
a
l
v
a
l
ue
s
.
c)
B
a
s
e
d
o
n
t
h
e
P
S
O
a
l
go
r
i
t
hm
’
s
r
e
s
u
l
t
s
,
de
f
i
ne
t
h
e
S
VM
m
o
de
l
’
s
pa
r
a
m
e
t
e
r
s
.
d)
T
h
e
S
VM
m
o
de
l
s
h
o
u
l
d
b
e
t
r
a
i
n
e
d
a
n
d
e
v
a
l
ua
t
e
d
to
de
t
e
r
m
i
ne
a
c
c
ur
a
c
y
,
whi
c
h
i
s
t
h
e
P
S
O
a
l
go
r
i
t
hm
’
s
go
a
l
.
e)
T
h
e
a
c
c
ur
a
c
y
v
a
l
u
e
,
whi
c
h
i
s
t
h
e
go
a
l
f
u
n
c
t
i
o
n
f
o
r
t
h
e
P
S
O
a
l
go
r
i
t
hm
,
i
s
o
b
t
a
i
ne
d
a
f
t
e
r
t
h
e
S
VM
m
o
de
l
h
a
s
b
e
e
n
t
r
a
i
n
e
d
a
n
d
e
v
a
l
u
a
t
e
d.
f)
P
S
O
a
s
s
e
s
s
e
s
t
h
e
o
bj
e
c
t
i
v
e
f
u
n
c
t
i
o
n
to
de
t
e
r
m
i
ne
t
h
e
o
pt
i
m
u
m
o
pt
i
m
a
l
.
g)
Upda
t
e
t
h
e
P
S
O
pa
r
a
m
e
t
e
r
s
a
n
d
c
o
n
t
i
n
ue
do
i
n
g
s
o
un
t
i
l
t
h
e
m
a
xim
u
m
i
t
e
r
a
t
i
o
n
i
s
r
e
a
c
h
e
d.
h)
P
r
o
vi
d
e
t
h
e
pa
r
t
i
c
l
e
w
i
t
h
t
h
e
Gb
e
s
t
v
a
l
ue
,
w
hi
c
h
i
s
t
h
e
S
VM
m
o
de
l
’
s
o
p
t
i
m
a
l
v
a
l
u
e
.
T
a
bl
e
1.
S
e
a
r
c
h
s
pa
c
e
f
o
r
P
S
O
P
a
r
ti
c
l
e
P
a
r
a
me
t
e
r
S
e
a
r
c
h
s
pa
c
e
X
1
K
e
r
na
l
[
“
li
ne
a
r
”
, “
r
b
f
”
,
“
p
o
l
y
”
]
X
2
K
e
r
n
e
l
c
o
e
f
f
i
c
i
e
nt
-
ga
mm
a
[
‘
s
c
a
l
e
’
,
‘
a
ut
o
’
]
X
3
P
e
na
lt
y
pa
r
a
m
e
t
e
r
C
[
0.1,100]
T
h
e
S
VM
m
o
d
e
l
op
t
i
m
i
z
e
d
wi
t
h
P
S
O
a
l
g
or
i
t
hm
f
o
un
d
t
h
e
b
e
s
t
p
a
r
a
m
e
ter
s
f
or
c
l
a
s
s
i
f
i
c
a
t
i
o
n
.
A
n
R
B
F
f
un
c
t
i
o
n
i
s
us
e
d
a
s
a
ke
r
n
e
l
i
n
th
e
S
VM
m
o
d
e
l
.
T
h
e
p
a
r
a
m
e
te
r
s
ga
m
m
a
a
n
d
C
a
r
e
c
h
o
s
e
n
a
s
‘
s
c
a
l
e
’
a
n
d
1
0
.
T
hi
s
m
ode
l
us
e
s
th
e
f
e
a
t
u
r
e
v
e
c
tor
s
f
r
o
m
t
h
e
C
N
N
m
o
d
e
l
a
s
i
n
pu
t
a
n
d
pe
r
f
or
m
t
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
t
h
e
i
m
a
ge
.
2
.
7
.
E
val
u
at
ion
m
e
t
r
ics
T
h
e
m
a
r
i
ne
a
nim
a
l
c
l
a
s
s
i
f
i
c
a
t
i
o
n
a
l
go
r
i
t
hm
pe
r
f
o
r
m
a
nc
e
c
a
n
b
e
m
e
a
s
ur
e
d
w
i
t
h
t
h
e
f
o
l
l
o
w
i
ng
p
e
r
f
o
r
m
a
n
c
e
m
e
t
r
i
c
s
g
i
ve
n
i
n
(
9)
-
(
11)
.
T
h
e
pe
r
c
e
n
t
a
ge
o
f
a
c
c
ur
a
t
e
f
o
r
e
c
a
s
t
s
i
s
c
a
ll
e
d
a
c
c
ur
a
c
y
.
R
e
c
a
l
l
,
a
l
s
o
kn
o
wn
a
s
s
e
ns
i
t
i
vi
t
y
,
r
e
f
e
r
s
to
t
h
e
pr
o
p
o
r
t
i
o
n
o
f
t
r
ue
po
s
i
t
i
v
e
s
,
w
h
e
r
e
a
s
pr
e
c
i
s
i
o
n
i
s
t
h
e
pe
r
c
e
n
t
a
ge
o
f
a
c
t
ua
l
po
s
i
t
i
v
e
s
a
m
o
n
g
a
ll
t
h
e
v
a
l
ue
s
pr
o
j
e
c
t
e
d
a
s
po
s
i
t
i
ve
.
T
h
e
F
1
-
S
c
o
r
e
i
s
a
m
e
t
r
i
c
t
h
a
t
qua
n
t
i
f
i
e
s
t
h
e
pe
r
f
o
r
m
a
n
c
e
o
f
a
c
l
a
s
s
if
i
e
r
by
t
a
k
i
n
g
t
h
e
ha
r
m
o
ni
c
m
e
a
n
o
f
i
t
s
a
c
c
ur
a
c
y
a
n
d
r
e
c
a
ll
a
n
d
c
a
l
c
u
l
a
t
e
d
us
i
n
g
(
12)
.
=
(
+
)
(
+
+
+
)
(
9)
=
(
)
(
+
)
(
10)
=
(
)
(
+
)
(
11)
1
−
=
(
2
∗
∗
)
(
+
)
(
12)
wh
e
r
e
T
P
,
T
N,
F
P
,
a
n
d
F
N
i
nd
i
c
a
t
e
s
t
h
e
t
r
ue
p
o
s
i
t
i
v
e
(
T
P
)
,
t
r
ue
n
e
ga
t
i
v
e
(
T
N)
,
f
a
l
s
e
po
s
i
t
i
v
e
(
F
P
)
,
a
n
d
f
a
l
s
e
n
e
ga
t
i
v
e
(
F
N)
r
e
s
pe
c
t
i
v
e
ly
.
3.
RE
S
UL
T
S
AN
D
DI
S
CU
S
S
I
ON
3.
1.
P
r
e
p
r
oc
e
s
s
in
g
T
h
e
a
c
t
ua
l
o
r
t
h
e
or
i
g
i
na
l
s
i
z
e
i
m
a
ge
i
s
s
h
o
w
n
i
n
F
i
gur
e
2.
T
h
e
a
c
t
ua
l
d
i
m
e
n
s
i
o
ns
o
f
t
h
e
i
m
a
ge
m
a
y
d
i
f
f
e
r
.
T
h
e
r
e
i
s
a
n
e
e
d
to
r
e
s
i
z
e
t
h
e
i
m
a
ge
i
n
t
o
t
h
e
r
e
qu
i
r
e
d
c
r
i
t
e
r
i
a
i
n
w
hi
c
h
t
h
e
wi
dt
h
i
s
to
h
e
i
g
h
t
to
b
e
i
n
t
h
e
f
o
r
m
a
t
o
f
1:2
whi
c
h
r
e
s
e
m
b
l
e
s
t
h
a
t
t
h
e
s
i
z
e
o
f
t
he
i
m
a
ge
s
c
a
n
be
i
n
d
im
e
n
s
i
o
ns
o
f
(
64
×
128)
o
r
(
1
28
×
256)
.
T
h
e
r
e
s
i
z
e
d
a
n
d
gr
a
y
s
c
a
l
e
c
o
n
ve
r
t
e
d
i
m
a
ge
i
s
s
h
o
wn
i
n
F
i
gur
e
3.
T
h
e
gr
a
d
i
e
n
t
a
n
d
a
n
g
l
e
o
f
t
h
e
a
c
t
ua
l
i
m
a
ge
a
r
e
s
h
o
wn
i
n
F
i
gur
e
s
4
a
n
d
5.
F
i
gur
e
6
d
i
s
p
l
a
y
s
t
h
e
Ho
G
pi
c
t
ur
e
o
f
t
h
e
pr
o
vi
de
d
s
c
a
l
e
d
i
m
a
ge
.
T
h
e
Ho
G
f
e
a
t
ur
e
s
t
h
a
t
h
a
ve
b
e
e
n
r
e
t
r
i
e
ve
d
a
r
e
ut
i
l
i
s
e
d
a
s
i
n
p
ut
f
o
r
t
h
e
C
NN
m
o
de
l
to
d
o
a
ddi
t
i
o
n
a
l
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
14
,
N
o.
1
,
A
pr
i
l
20
25
:
240
-
249
246
F
i
gur
e
2.
A
c
t
ua
l
im
a
ge
F
i
gur
e
3.
R
e
s
i
z
e
d
gr
a
y
s
c
a
l
e
i
m
a
ge
F
i
gur
e
4.
Gr
a
di
e
n
t
F
i
gur
e
5.
A
n
g
l
e
F
i
gur
e
6.
H
o
G
3.
2.
CNN
m
od
e
l
p
ar
a
m
e
t
e
r
T
h
e
C
NN
m
o
de
l
i
s
de
s
i
g
n
e
d
w
i
t
h
4
c
o
n
v
o
l
ut
i
o
n
l
a
y
e
r
s
w
i
t
h
t
h
e
n
u
m
be
r
o
f
n
e
ur
o
n
s
o
f
{32,
64
,
128,
32}
,
t
h
e
ke
r
n
e
l
s
i
z
e
o
f
3*3
a
n
d
t
h
e
a
c
t
i
v
a
t
i
o
n
f
u
n
c
t
i
o
n
o
f
r
e
l
u
.
Ne
x
t
,
t
h
e
m
a
x
poo
l
i
ng
l
a
y
e
r
a
nd
t
h
e
f
l
a
t
t
e
n
l
a
y
e
r
a
r
e
a
dde
d.
F
o
l
l
o
we
d
by
t
h
e
de
ns
e
l
a
y
e
r
w
i
t
h
256,
512
n
e
ur
o
n
s
a
dd
e
d
to
t
h
e
m
o
de
l
.
T
h
e
m
o
de
l
i
s
c
o
m
p
li
e
d
w
i
t
h
t
h
e
a
da
m
o
pt
i
mi
z
e
r
a
n
d
t
h
e
l
o
s
s
f
u
n
c
t
i
o
n
i
s
o
f
bi
na
r
y
_c
r
o
s
s
e
n
t
r
y
.
T
h
e
o
u
t
c
o
m
e
o
f
th
e
C
NN
m
o
de
l
i
s
t
h
e
f
e
a
t
ur
e
v
e
c
to
r
.
3.
3.
Re
s
u
l
t
s
T
h
e
m
o
de
l
i
s
e
v
a
l
ua
t
e
d
a
n
d
i
t
s
pe
r
f
o
r
m
a
nc
e
c
a
n
be
i
m
pr
o
v
e
d
by
u
s
i
ng
t
h
e
K
-
f
o
l
d
c
r
o
s
s
v
a
li
da
t
i
o
n
t
e
c
h
ni
que
s
.
T
hi
s
va
l
i
d
a
t
i
o
n
t
e
c
hni
que
i
s
a
pp
li
e
d
t
o
a
v
o
i
d
t
h
e
o
v
e
r
f
i
t
t
i
n
g
o
f
t
h
e
m
o
de
l
by
r
e
duc
i
ng
t
h
e
v
a
r
i
a
n
c
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
hy
br
id
appr
oac
h
o
f
pat
ter
n
r
e
c
ognit
ion
to
de
tec
t
mar
ine
animals
(
V
ij
ay
alaks
hmi
B
alachandr
an
)
247
o
f
t
h
e
pe
r
f
o
r
m
a
n
c
e
e
s
t
i
m
a
t
e
a
n
d
i
t
pr
o
vi
de
s
a
n
o
ppo
r
t
un
i
t
y
t
o
t
h
e
m
o
de
l
to
l
e
a
r
n
w
i
t
h
m
o
r
e
a
m
o
un
t
o
f
da
t
a
.
T
h
e
c
o
m
pa
r
i
s
o
n
o
f
t
h
e
pr
o
p
o
s
e
d
Ho
G
b
a
s
e
d
C
NN
_S
VM
m
o
de
l
w
i
t
h
v
a
r
i
o
us
m
a
c
hi
ne
l
e
a
r
ni
ng
m
o
de
l
s
w
i
t
h
o
ut
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
i
s
g
i
ve
n
i
n
T
a
bl
e
2.
T
a
bl
e
3
pr
o
vi
de
s
a
d
e
t
a
i
l
e
d
c
o
m
pa
r
i
s
o
n
o
f
m
o
de
l
pe
r
f
o
r
m
a
n
c
e
us
i
ng
d
i
f
f
e
r
e
n
t
c
o
n
f
i
gur
a
t
i
o
n
s
:
Ho
G+
S
VM
,
H
o
G+
C
NN
,
a
n
d
t
h
e
pr
o
p
o
s
e
d
H
oG+
C
NN
+
S
VM
m
o
de
l
.
T
h
e
pe
r
f
o
r
m
a
n
c
e
m
e
t
r
i
c
s
i
nc
l
ude
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
a
n
d
F1
-
s
c
o
r
e
.
T
h
e
pr
o
p
o
s
e
d
m
o
de
l
,
H
o
G+
C
NN
+
S
VM
,
s
h
o
ws
s
upe
r
i
o
r
pe
r
f
o
r
m
a
n
c
e
a
c
r
o
s
s
a
l
l
m
e
t
r
i
c
s
,
w
i
t
h
a
n
a
c
c
ur
a
c
y
o
f
95.
7%
,
pr
e
c
i
s
i
o
n
o
f
0.
93,
r
e
c
a
l
l
o
f
0.
94,
a
n
d
F1
-
s
c
o
r
e
o
f
0.
94.
T
hi
s
s
ugge
s
t
s
t
h
a
t
t
h
e
i
n
t
e
gr
a
t
i
o
n
o
f
C
NN
a
n
d
S
VM
wi
t
h
Ho
G
f
e
a
t
ur
e
s
s
i
g
nif
i
c
a
n
t
l
y
e
nh
a
n
c
e
s
t
h
e
m
o
de
l
’
s
e
f
f
e
c
t
i
v
e
n
e
s
s
.
I
n
c
o
n
t
r
a
s
t
,
t
h
e
Ho
G+
C
NN
m
o
de
l
a
c
hi
e
ve
s
a
n
a
c
c
ur
a
c
y
o
f
91%
,
a
n
d
t
he
Ho
G+
S
VM
m
o
de
l
h
a
s
t
h
e
l
o
we
s
t
a
c
c
ur
a
c
y
a
t
89.
85
%
.
T
h
e
s
e
r
e
s
u
l
t
s
un
de
r
s
c
o
r
e
t
h
e
i
m
pr
o
v
e
d
pe
r
f
o
r
m
a
n
c
e
a
n
d
r
ob
us
t
ne
s
s
o
f
t
h
e
pr
o
p
o
s
e
d
Ho
G+
C
NN
+
S
VM
m
o
de
l
f
o
r
t
h
e
g
i
ve
n
t
a
s
k.
T
a
bl
e
2.
C
o
m
pa
r
i
s
o
n
o
f
t
h
e
m
o
de
l
w
i
t
h
o
ut
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
M
o
de
l
A
c
c
u
r
a
c
y
P
r
e
c
is
i
o
n
R
e
c
a
ll
F1
-
s
c
o
r
e
DT
72.2
0.62
0.72
0.69
S
V
M
76.80
0.66
0.75
0.70
K
N
N
79.72
0.69
0.79
0.72
C
N
N
88.80
0.74
0.84
0.77
H
oG
+
C
N
N
+
S
V
M
(
pr
o
p
o
s
e
d m
o
d
e
l
)
95.7
0.93
0.94
0.94
T
a
bl
e
3.
C
o
m
pa
r
i
s
o
n
o
f
m
o
de
l
pe
r
f
o
r
m
a
n
c
e
w
i
t
h
ot
h
e
r
m
o
de
l
s
M
o
de
l
A
c
c
u
r
a
c
y
P
r
e
c
is
i
o
n
R
e
c
a
ll
F1
-
s
c
o
r
e
H
oG
+
S
V
M
89.85
0.88
0.89
0.88
H
oG
+
C
N
N
91
0.89
0.90
0.89
H
oG
+
C
N
N
+
S
V
M
(
pr
o
p
o
s
e
d m
o
d
e
l
)
95.7
0.93
0.94
0.94
F
i
gur
e
7
c
o
m
pa
r
e
s
t
h
r
e
e
m
o
de
l
s
’
pe
r
f
o
r
m
a
n
c
e
–
Ho
G+
S
VM
,
H
o
G+
C
NN
,
a
n
d
Ho
G+
C
NN
+
S
VM
–
o
n
f
o
ur
e
v
a
l
ua
t
i
o
n
m
e
t
r
i
c
s
:
a
c
c
ur
a
c
y
,
pr
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
a
n
d
F1
-
s
c
o
r
e
.
a)
Ho
G+
S
VM
:
i
n
t
hi
s
m
o
de
l
,
Ho
G
i
s
us
e
d
f
o
r
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
whil
e
S
VM
i
s
e
m
p
l
o
y
e
d
f
o
r
c
l
a
s
s
if
i
c
a
t
i
o
n
I
t
s
h
o
ws
m
o
de
r
a
t
e
l
y
hi
g
h
pe
r
f
o
r
m
a
n
c
e
a
c
r
o
s
s
a
l
l
m
e
t
r
i
c
s
,
w
i
t
h
v
a
l
ue
s
ge
n
e
r
a
ll
y
r
a
n
g
i
ng
a
r
o
un
d
t
h
e
mi
d
-
80%
m
a
r
k.
b)
Ho
G+
C
NN
:
t
hi
s
m
o
de
l
e
m
p
l
o
y
s
C
N
N
a
l
o
n
g
w
i
t
h
Ho
G
f
e
a
t
ur
e
s
.
I
t
pe
r
f
o
r
m
s
a
li
t
t
l
e
b
e
t
t
e
r
wi
t
h
a
ll
m
e
a
s
ur
e
s
t
h
a
n
Ho
G+
S
VM
m
o
de
l
w
h
o
s
e
pe
r
f
o
r
m
a
n
c
e
s
a
r
e
c
o
n
s
t
a
n
t
l
y
j
us
t
o
v
e
r
85%
i
n
a
ll
t
he
s
e
i
nd
i
c
a
t
o
r
s
.
c)
Ho
G+
C
NN
+
S
VM
(
pr
o
p
o
s
e
d
m
o
de
l
)
:
t
hi
s
pr
o
p
o
s
e
d
m
o
de
l
i
n
t
e
gr
a
t
e
s
t
h
e
f
e
a
t
ur
e
s
o
f
Ho
G,
C
NN
,
a
n
d
S
VM
whi
c
h
h
e
l
p
t
o
o
u
t
pe
r
f
o
r
m
t
h
e
ot
h
e
r
t
w
o
m
o
de
l
s
a
c
r
o
s
s
a
l
l
m
e
t
r
i
c
s
,
wi
t
h
pe
r
f
o
r
m
a
n
c
e
pe
r
c
e
n
t
a
ge
s
n
e
a
r
i
ng
o
r
r
e
a
c
hi
ng
90%
.
T
hi
s
s
h
o
ws
t
h
a
t
t
h
e
pr
o
p
o
s
e
d
Ho
G_C
NN
_S
VM
m
o
de
l
pe
r
f
o
r
m
s
b
e
t
t
e
r
wi
t
h
a
n
a
c
c
ur
a
c
y
o
f
95.
7%
f
o
r
t
h
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
m
a
r
i
ne
a
nim
a
l
s
.
T
hi
s
s
ho
ws
t
h
a
t
f
e
a
t
ur
e
e
x
tr
a
c
t
i
o
n
i
s
a
c
r
uc
i
a
l
c
o
m
po
ne
n
t
o
f
t
h
e
pa
tt
e
r
n
r
e
c
o
gni
t
i
o
n
pr
o
c
e
s
s
f
o
r
un
de
r
wa
t
e
r
ph
oto
s
.
F
i
gur
e
7.
P
e
r
f
o
r
m
a
n
c
e
c
o
m
pa
r
i
s
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8776
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
,
Vo
l
.
14
,
N
o.
1
,
A
pr
i
l
20
25
:
240
-
249
248
4.
CONC
L
USI
ON
A
s
t
r
e
a
m
li
ne
d,
a
uto
m
a
t
e
d,
a
n
d
i
n
s
t
a
n
t
a
ne
o
us
s
ur
v
e
il
l
a
n
c
e
m
e
t
h
o
d
i
s
vi
t
a
l
f
o
r
i
de
n
t
i
f
yi
ng
d
i
ve
r
s
e
e
n
t
i
t
i
e
s
(
s
uc
h
a
s
h
u
m
a
n
a
c
t
i
vi
t
i
e
s
,
s
i
z
a
bl
e
f
i
s
h
,
s
ha
r
ks
,
wh
a
l
e
s
,
a
n
d
s
ur
f
e
r
s
)
o
n
b
e
a
c
h
e
s
,
i
n
o
r
de
r
to
pr
e
v
e
n
t
unf
o
r
e
s
e
e
n
de
a
t
h
s
a
n
d
m
i
s
ha
p
s
.
T
hi
s
s
t
ud
y
i
n
t
r
o
d
u
c
e
s
a
f
e
a
t
ur
e
e
x
t
r
a
c
to
r
b
a
s
e
d
o
n
de
e
p
l
e
a
r
ni
ng,
whi
c
h
is
c
o
m
bi
ne
d
w
i
t
h
a
m
a
c
hi
ne
l
e
a
r
ni
ng
c
l
a
s
s
i
f
i
e
r
.
T
h
e
pur
po
s
e
i
s
t
o
a
u
to
m
a
t
i
c
a
ll
y
i
de
n
t
i
f
y
pa
t
t
e
r
n
s
a
n
d
c
a
t
e
g
o
r
i
s
e
m
a
r
i
ne
s
pe
c
i
e
s
.
T
hi
s
a
ppr
o
a
c
h
a
i
m
s
t
o
m
i
n
i
mi
s
e
th
e
ne
e
d
f
o
r
h
u
m
a
n
i
n
t
e
r
v
e
n
t
i
o
n
a
n
d
l
o
w
e
r
a
s
s
o
c
i
a
t
e
d
c
o
s
t
s
.
T
h
e
C
NN
m
o
de
l
e
x
t
r
a
c
t
s
t
h
e
e
s
s
e
n
t
i
a
l
a
n
d
s
i
g
nif
i
c
a
n
t
H
o
G
i
nf
o
r
m
a
t
i
o
n
f
r
o
m
t
h
e
i
m
a
ge
.
T
h
e
s
e
c
ha
r
a
c
t
e
r
i
s
t
i
c
s
we
r
e
pr
o
v
i
d
e
d
a
s
i
nput
to
t
h
e
S
VM
c
l
a
s
s
if
i
e
r
i
n
or
de
r
to
c
a
t
e
g
o
r
i
s
e
t
h
e
m
a
r
i
ne
s
pe
c
i
e
s
f
o
un
d
a
lo
n
g
t
h
e
s
h
o
r
e
l
i
ne
,
w
hi
c
h
m
i
g
h
t
pot
e
n
t
i
a
ll
y
im
pa
c
t
t
h
o
s
e
s
w
i
mm
i
ng
i
n
t
he
wa
t
e
r
.
T
h
e
r
e
s
u
l
t
s
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
s
ugge
s
t
e
d
m
e
t
h
o
d
de
m
o
ns
t
r
a
t
e
e
n
ha
n
c
e
d
a
c
c
ur
a
c
y
o
f
95%
i
n
c
o
m
pa
r
i
s
o
n
to
t
h
e
a
l
t
e
r
n
a
t
i
v
e
m
a
c
hi
ne
l
e
a
r
ni
ng
m
e
t
h
o
do
l
o
g
y
t
h
a
t
d
o
e
s
n
ot
i
nv
o
l
v
e
f
e
a
t
ur
e
e
x
t
r
a
c
ti
o
n
.
T
h
e
r
e
f
o
r
e
,
i
t
i
s
e
vi
de
n
t
t
h
a
t
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
o
n
pl
a
y
s
a
c
r
uc
i
a
l
r
o
l
e
i
n
pr
e
d
i
c
t
i
n
g
m
a
r
i
ne
c
r
e
a
t
ur
e
s
.
RE
F
E
R
E
NC
E
S
[
1]
E
.
M
.
D
it
r
ia
,
M
.
S
ie
ve
r
s
,
S
.
L
ope
z
-
M
a
r
c
a
n
o
,
E
.
L
.
J
in
ks
,
a
nd
R
.
M
.
C
o
nn
o
ll
y
,
“
D
e
e
p
l
e
a
r
ni
ng
f
o
r
a
ut
o
ma
te
d
a
na
l
y
s
is
of
f
is
h
a
bunda
nc
e
:
th
e
b
e
n
e
f
it
s
of
t
r
a
in
in
g
a
c
r
o
s
s
mul
ti
pl
e
ha
bi
ta
ts
,
”
E
nv
ir
onm
e
nt
al
M
oni
to
r
in
g
and
A
s
s
e
s
s
m
e
nt
,
v
ol
.
192,
n
o
.
11,
O
c
t.
2020,
d
o
i:
10.1007/s
10661
-
020
-
08653
-
z.
[
2]
A
.
J
a
la
l,
A
.
S
a
lm
a
n,
A
.
M
ia
n,
M
.
S
ho
r
ti
s
,
a
nd
F
.
S
ha
f
a
it
,
“
F
is
h
de
t
e
c
ti
o
n
a
nd
s
pe
c
i
e
s
c
la
s
s
if
i
c
a
ti
o
n
in
und
e
r
w
a
t
e
r
e
n
v
ir
o
n
m
e
n
ts
us
in
g
de
e
p
l
e
a
r
ni
ng
w
it
h
t
e
mp
or
a
l
in
f
or
ma
ti
o
n,”
E
c
ol
ogi
c
al
I
nf
or
m
at
ic
s
,
v
o
l.
57,
p.
101088
,
M
a
y
2
020,
do
i:
10.1016/j
.
e
c
o
in
f
.2020.101088.
[
3]
A
.
B
.
S
a
lb
e
r
g,
“
D
e
t
e
c
ti
o
n
of
s
e
a
ls
in
r
e
m
o
t
e
s
e
ns
in
g
im
a
ge
s
us
in
g
f
e
a
tu
r
e
s
e
x
t
r
a
c
t
e
d
f
r
o
m
de
e
p
c
o
n
vo
lu
t
i
o
na
l
ne
ur
a
l
n
e
tw
o
r
ks
,
”
in
I
nt
e
r
nat
io
nal
G
e
os
c
ie
nc
e
and
R
e
m
ot
e
Se
ns
in
g
Sy
m
pos
iu
m
(
I
G
A
R
SS)
,
J
ul
.
2015
,
vol
.
2015
-
N
ove
mb
e
r
,
pp.
1893
–
18
96,
do
i:
10.1109/
I
G
A
R
S
S
.2015.7326163.
[
4]
E
.
G
ui
r
a
d
o
,
S
.
T
a
bi
k,
M
.
L
.
R
iv
a
s
,
D
.
A
lc
a
r
a
z
-
S
e
gur
a
,
a
nd
F
.
H
e
r
r
e
r
a
,
“
W
ha
le
c
o
un
ti
ng
in
s
a
te
ll
it
e
a
nd
a
e
r
ia
l
im
a
ge
s
w
it
h
de
e
p
le
a
r
ni
ng,”
Sc
ie
nt
if
ic
R
e
por
ts
, v
o
l.
9, n
o
. 1, O
c
t.
20
19, d
o
i:
10.10
38/
s
41598
-
019
-
50795
-
9.
[
5]
M
.
D
ua
r
te
,
J
.
A
guz
z
i,
a
nd
E
.
F
a
n
e
ll
i,
“
A
ut
o
ma
t
e
d
v
id
e
o
m
o
n
it
or
in
g
of
c
o
r
a
l
r
e
e
f
f
is
h
p
o
pul
a
ti
o
ns
,”
M
ar
in
e
E
c
ol
ogy
P
r
ogr
e
s
s
Se
r
ie
s
, v
o
l.
681, pp. 105
–
119, 2022.
[
6]
N
.
S
ha
r
ma
,
M
.
S
a
qi
b,
P
.
S
c
ul
l
y
-
P
o
w
e
r
,
a
nd
M
.
B
lu
m
e
n
s
te
in
,
“
S
ha
r
kS
po
t
te
r
:
s
ha
r
k
de
t
e
c
ti
o
n
w
it
h
d
r
o
n
e
s
f
o
r
huma
n
s
a
f
e
t
y
a
nd
e
n
v
ir
o
nm
e
nt
a
l
pr
o
t
e
c
ti
o
n,”
i
n
H
um
ani
ty
D
r
iv
e
n A
I
, S
pr
in
g
e
r
I
nt
e
r
na
ti
o
na
l
P
ubl
is
hi
ng, 2022, pp. 223
–
237.
[
7]
S
.
G
ur
u
r
a
ts
a
kul
,
D
.
G
ib
bi
ns
,
a
nd
D
.
K
e
a
r
ne
y
,
“
A
s
im
pl
e
d
e
f
o
r
ma
bl
e
m
o
d
e
l
f
o
r
s
ha
r
k
r
e
c
o
gni
t
i
o
n,”
in
P
r
oc
e
e
di
ngs
-
2011
C
anadian C
onf
e
r
e
n
c
e
on C
om
put
e
r
and R
obot
V
is
io
n, C
R
V
2011
, M
a
y
2011, pp. 234
–
241, d
o
i
:
10.1109/C
R
V
.2011.38.
[
8]
N
.
E
.
M
e
r
e
n
c
il
la
,
A
.
S
a
r
r
a
ga
A
l
o
n,
G
.
J
.
O
.
F
e
r
na
ndo
,
E
.
M
.
C
e
pe
,
a
nd
D
.
C
.
M
a
lu
na
o
,
“
S
ha
r
k
-
E
Y
E
:
a
d
e
e
p
in
f
e
r
e
n
c
e
c
o
n
vo
lu
ti
o
na
l
ne
u
r
a
l
ne
tw
o
r
k
of
s
ha
r
k
de
te
c
ti
o
n
f
or
unde
r
w
a
t
e
r
di
v
in
g
s
ur
ve
il
la
n
c
e
,”
in
P
r
oc
e
e
di
ngs
o
f
2nd
I
E
E
E
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on
C
om
put
at
io
nal
I
nt
e
ll
ig
e
nc
e
and
K
now
le
dge
E
c
onomy
,
I
C
C
I
K
E
2021
,
M
a
r
.
2021,
pp.
384
–
3
88,
do
i:
10.1109/
I
C
C
I
K
E
51210.2021.9410715.
[
9]
N
.
S
ha
r
ma
,
P
.
S
c
ul
l
y
-
P
o
w
e
r
,
a
nd
M
.
B
lu
m
e
ns
te
in
,
“
S
ha
r
k
d
e
t
e
c
t
i
o
n
f
r
o
m
a
e
r
ia
l
im
a
ge
r
y
us
in
g
r
e
gi
o
n
-
ba
s
e
d
C
N
N
,
a
s
tu
d
y
,”
in
L
e
c
tu
r
e
N
ot
e
s
in
C
o
m
put
e
r
Sc
ie
n
c
e
(
in
c
lu
di
ng
s
ubs
e
r
ie
s
L
e
c
tu
r
e
N
ot
e
s
in
A
r
ti
f
ic
ia
l
I
nt
e
ll
ig
e
nc
e
and
L
e
c
tu
r
e
N
ot
e
s
in
B
io
in
f
or
m
at
ic
s
)
, v
o
l.
11320
L
N
A
I
, S
pr
in
g
e
r
I
nt
e
r
na
ti
o
na
l
P
ubl
i
s
hi
ng, 2018, pp. 224
–
236.
[
10]
J
. J
e
nr
e
tt
e
,
Z
. Y
. C
.
L
iu
, P
. C
hi
m
o
t
e
,
T
.
H
a
s
ti
e
,
E
. F
ox
, a
nd F
. F
e
r
r
e
tt
i,
“
S
ha
r
k d
e
te
c
ti
o
n a
nd
c
la
s
s
i
f
ic
a
ti
o
n w
i
th
ma
c
hi
n
e
l
e
a
r
ni
n
g,”
E
c
ol
ogi
c
al
I
nf
or
m
at
ic
s
, v
o
l.
69, p. 101673, J
ul
. 2022, d
o
i:
10.1
016/
j.
e
c
o
in
f
.2022.101673.
[
11]
R
.
G
o
r
ki
n
e
t
al
.
,
“
S
ha
r
k
e
y
e
:
R
e
a
l
-
ti
m
e
a
ut
o
n
o
m
o
us
p
e
r
s
o
na
l
s
ha
r
k
a
le
r
ti
ng
v
ia
a
e
r
ia
l
s
ur
ve
il
la
nc
e
,”
D
r
one
s
,
vo
l.
4,
n
o
.
2,
pp. 1
–
17, M
a
y
2020, d
o
i
:
10.3390/dr
o
n
e
s
4020018.
[
12]
M
.
H
e
ma
la
th
a
,
M
.
A
.
M
ut
hi
a
h,
a
nd
B
.
V
e
nka
ta
la
ks
mi
,
“
M
ul
ti
-
f
e
a
tu
r
e
j
o
in
t
d
e
s
c
r
ip
t
o
r
ba
s
e
d
im
a
g
e
d
e
t
e
c
t
i
o
n
a
lg
or
it
hm
f
or
c
r
oc
o
di
l
e
d
e
t
e
c
ti
o
n,”
in
P
r
oc
e
e
di
ngs
of
2016
I
nt
e
r
nat
io
nal
C
onf
e
r
e
nc
e
on
A
dv
anc
e
d
C
om
m
uni
c
at
io
n
C
ont
r
ol
and
C
om
put
in
g
T
e
c
hnol
ogi
e
s
, I
C
A
C
C
C
T
2016
, M
a
y
2017, pp. 527
–
530, d
o
i
:
1
0.1109/I
C
A
C
C
C
T
.2016.7831696.
[
13]
L
.
M
e
ji
a
s
,
G
.
D
uc
l
o
s
,
A
.
H
o
dgs
o
n,
a
nd
F
.
M
a
ir
e
,
A
ut
om
at
e
d
m
ar
in
e
m
am
m
al
d
e
te
c
ti
on
f
r
om
ae
r
ia
l
image
r
y
.
T
o
a
ppe
a
r
in
M
T
S
/
I
E
E
E
O
C
E
A
N
S
, S
a
n D
i
e
g
o
, U
S
A
, 2013.
[
14]
J
. L
o
p
e
z
, J
. S
c
h
oo
n
ma
ke
r
,
a
nd S
. S
a
gg
e
s
e
, “
A
ut
o
ma
te
d
de
t
e
c
ti
o
n
of
ma
r
in
e
a
ni
ma
ls
us
in
g mul
ti
s
pe
c
tr
a
l
im
a
gi
ng,”
i
n
2014 Oc
e
ans
-
St
. J
ohn
’
s
, O
C
E
A
N
S 2014
, S
e
p. 2015, pp. 1
–
6, d
o
i:
10.1109
/
O
C
E
A
N
S
.2014.7003132.
[
15]
P
.
B
e
r
g,
D
.
S
.
M
a
ia
,
M
.
T
.
P
ha
m,
a
nd
S
.
L
e
f
è
v
r
e
,
“
W
e
a
kl
y
s
upe
r
v
is
e
d
d
e
t
e
c
ti
o
n
of
ma
r
in
e
a
ni
ma
ls
in
hi
gh
r
e
s
o
lu
t
i
o
n
a
e
r
ia
l
im
a
ge
s
,”
R
e
m
ot
e
Se
ns
in
g
, v
ol
. 14, n
o
. 2, p. 339, J
a
n. 2022, do
i:
10.3390/r
s
14020339.
[
16]
A
.
S
a
lm
a
n
e
t
al
.
,
“
F
is
h
s
pe
c
ie
s
c
la
s
s
if
i
c
a
ti
o
n
in
unc
o
ns
tr
a
in
e
d
unde
r
w
a
t
e
r
e
n
vi
r
o
nm
e
nt
s
ba
s
e
d
o
n
d
e
e
p
l
e
a
r
ni
ng,”
L
imnology
and
O
c
e
anogr
aphy
:
M
e
th
ods
, v
o
l.
14, n
o
. 9, pp. 570
–
585, M
a
y
201
6, do
i:
10.1002/l
om3.10113.
[
17]
S
.
V
il
lo
n,
M
.
C
ha
umo
nt
,
G
.
S
ubs
o
l,
S
.
V
il
lé
ge
r
,
T
.
C
la
ve
r
i
e
,
a
nd
D
.
M
o
ui
ll
o
t,
“
C
or
a
l
r
e
e
f
f
is
h
de
t
e
c
ti
o
n
a
nd
r
e
c
o
gni
ti
o
n
in
u
nde
r
w
a
t
e
r
v
id
e
o
s
b
y
s
upe
r
v
is
e
d
ma
c
hi
ne
le
a
r
n
in
g:
C
o
mpa
r
is
o
n
b
e
tw
e
e
n
de
e
p
l
e
a
r
ni
ng
a
nd
H
O
G
+
S
V
M
me
th
o
ds
,”
in
L
e
c
t
ur
e
N
ot
e
s
in
C
om
put
e
r
Sc
ie
nc
e
(
in
c
lu
di
ng
s
ubs
e
r
ie
s
L
e
c
tu
r
e
N
ot
e
s
in
A
r
ti
f
ic
ia
l
I
nt
e
ll
ig
e
nc
e
and
L
e
c
tu
r
e
N
ot
e
s
in
B
io
in
f
or
m
at
i
c
s
)
,
vo
l.
10016
L
N
C
S
,
S
pr
in
g
e
r
I
n
te
r
na
ti
o
na
l
P
ubl
is
hi
ng, 2016, pp.
160
–
171.
[
18]
X
.
L
i,
M
.
S
ha
ng,
H
.
Q
in
,
a
nd
L
.
C
he
n,
“
F
a
s
t
a
c
c
u
r
a
te
f
is
h
de
te
c
ti
o
n
a
nd
r
e
c
o
gni
ti
o
n
of
und
e
r
w
a
t
e
r
im
a
ge
s
w
it
h
F
a
s
t
R
-
C
N
N
,”
O
c
t.
2016, d
o
i:
10.23919/
oc
e
a
ns
.2015.7404464.
[
19]
M
.
J
o
o
E
r
,
J
.
C
he
n,
a
nd
Y
.
Z
ha
ng,
“
M
a
r
in
e
r
o
b
o
ti
c
s
4.0:
pr
e
s
e
nt
a
nd
f
ut
ur
e
of
r
e
a
l
-
ti
m
e
d
e
te
c
ti
o
n
t
e
c
hni
qu
e
s
f
or
unde
r
w
a
te
r
o
bj
e
c
ts
,”
i
n
I
ndus
tr
y
4.0
-
P
e
r
s
pe
c
ti
v
e
s
and A
ppl
ic
at
io
ns
, I
nt
e
c
h
O
pe
n, 2023.
[
20]
O
.
I
tt
oo
a
nd
S
.
P
uda
r
ut
h,
“
A
ut
o
ma
ti
c
r
e
c
o
gni
ti
o
n
of
ma
r
in
e
c
r
e
a
tu
r
e
s
u
s
in
g
d
e
e
p
l
e
a
r
ni
ng,”
I
nt
e
r
nat
io
nal
J
our
nal
of
A
dv
an
c
e
d
C
om
put
e
r
Sc
ie
n
c
e
and A
ppl
ic
at
io
ns
, v
o
l.
15, n
o
. 1, pp. 47
–
64,
2024, do
i:
10.14569/
I
J
A
C
S
A
.2024.0150106.
[
21]
N
.
S
ha
hna
z
,
M
.
M
.
I
s
la
m,
M
.
A
.
S
a
r
k
e
r
,
a
nd
M
.
R
a
hma
n,
“
H
y
br
id
d
e
e
p
l
e
a
r
ni
ng
a
ppr
o
a
c
h
f
or
ma
r
i
n
e
ma
mm
a
l
d
e
t
e
c
ti
o
n
u
s
in
g
a
c
o
us
ti
c
da
ta
,”
J
our
nal
of
M
ar
in
e
S
c
ie
nc
e
and E
ngi
ne
e
r
in
g
, vo
l
. 10, no
. 2, p. 123, 2022.
[
22]
X
.
Z
ha
ng,
J
.
W
.
Y
.
L
i,
a
nd
H
.
X
u,
“
A
h
y
br
id
d
e
e
p
l
e
a
r
ni
ng
mo
d
e
l
f
or
r
e
a
l
-
ti
me
f
is
h
s
pe
c
i
e
s
r
e
c
o
gni
t
i
o
n
in
und
e
r
w
a
t
e
r
v
id
e
o
s
,”
I
E
E
E
T
r
ans
ac
ti
ons
on Ge
os
c
i
e
nc
e
and R
e
m
ot
e
Se
n
s
in
g
, v
o
l.
61
, pp. 4512
–
4524, 2023.
[
23]
R
.
V
e
nka
te
s
h,
S
.
A
na
nt
ha
r
a
ja
n,
a
nd
S
.
G
una
s
e
ka
r
a
n,
“
M
ul
ti
-
g
r
a
di
e
nt
b
oo
s
t
e
d
a
da
pt
i
ve
S
V
M
-
ba
s
e
d
pr
e
di
c
ti
o
n
of
h
e
a
r
t
di
s
e
a
s
e
,”
I
nt
e
r
nat
io
nal
J
our
nal
of
C
om
put
e
r
s
, C
om
m
uni
c
at
io
n
s
and
C
ont
r
ol
, vo
l.
18, n
o
. 5, Aug. 2023, d
o
i:
10.15837/i
j
c
c
c
.2023.5.4994.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
I
n
f
&
C
o
m
m
u
n
T
e
c
hn
o
l
I
S
S
N:
2252
-
8776
A
hy
br
id
appr
oac
h
o
f
pat
ter
n
r
e
c
ognit
ion
to
de
tec
t
mar
ine
animals
(
V
ij
ay
alaks
hmi
B
alachandr
an
)
249
[
24]
M
.
X
u,
L
.
C
a
o
,
D
.
L
u,
Z
.
H
u,
a
nd
Y
.
Y
ue
,
“
A
ppl
ic
a
ti
o
n
of
s
w
a
r
m
in
te
ll
ig
e
n
c
e
o
pt
im
i
z
a
ti
o
n
a
lg
o
r
it
hms
in
im
a
ge
pr
oc
e
s
s
in
g:
a
c
o
mpr
e
h
e
ns
i
v
e
r
e
v
i
e
w
of
a
na
l
y
s
is
,
s
y
nt
he
s
is
,
a
nd
o
pt
im
iz
a
t
i
o
n,”
B
io
m
ime
ti
c
s
,
vol
.
8,
n
o
.
2,
p.
235,
J
un.
2
023,
do
i:
10.3390/bi
omi
m
e
ti
c
s
8020235.
[
25]
“
S
ha
r
k S
pe
c
i
e
s
,”
K
aggl
e
. ht
tp
s
:/
/ww
w
.ka
ggl
e
.c
o
m/
da
ta
s
e
ts
/l
a
r
u
s
s
o
94/
s
ha
r
k
-
s
pe
c
i
e
s
(
a
c
c
e
s
s
e
d F
e
b. 14, 2024)
.
[
26]
P
.
L
.
F
a
lk
in
gha
m
a
nd
J
.
M
il
a
n,
“
C
r
oc
o
di
l
e
t
r
a
c
ks
a
nd
t
r
a
c
kw
a
y
s
:
P
h
o
t
o
g
r
a
mm
e
tr
i
c
m
o
d
e
ls
a
nd
ph
o
t
o
da
ta
s
e
ts
,”
Z
e
nodo,
D
at
as
e
t
,
2015. htt
ps
:/
/d
o
i.
or
g/
10.5281/
z
e
n
o
d
o
.31711.
[
27]
“
C
r
oc
o
di
le
,”
I
m
age
s
.c
v
. ht
tp
s
:/
/i
ma
ge
s
.c
v/
c
a
t
e
g
o
r
y
/
c
r
oc
o
di
l
e
(
a
c
c
e
s
s
e
d M
a
r
. 01, 2024)
.
B
I
OG
RA
P
HI
E
S
OF
AU
T
HO
RS
V
i
ja
y
a
l
a
k
s
hm
i
Ba
l
a
ch
a
n
dra
n
i
s
w
o
r
k
i
n
g
as
an
a
s
s
i
s
t
an
t
p
ro
fe
s
s
o
r
(SG
)
i
n
t
h
e
D
e
p
art
men
t
o
f
C
o
m
p
u
t
e
r
S
ci
en
ce
an
d
E
n
g
i
n
ee
ri
n
g
,
Ra
mc
o
In
s
t
i
t
u
t
e
o
f
T
ec
h
n
o
l
o
g
y
,
Raj
ap
al
a
y
a
m
,
I
n
d
i
a.
Sh
e
h
as
co
m
p
l
e
t
ed
h
e
r
B.
E
.
a
n
d
M.
E
.
D
e
g
r
ee
i
n
C
o
m
p
u
t
e
r
S
c
i
en
ce
a
n
d
E
n
g
i
n
ee
r
i
n
g
fro
m
PSR
E
n
g
i
n
ee
ri
n
g
C
o
l
l
e
g
e,
A
n
n
a
U
n
i
v
e
rs
i
t
y
.
H
e
r
ar
e
as
o
f
i
n
t
e
r
e
s
t
are
i
m
ag
e
p
ro
ce
s
s
i
n
g
,
d
at
a
s
c
i
en
ce
an
d
d
ee
p
l
e
ar
n
i
n
g
.
Sh
e
h
as
p
u
b
l
i
s
h
ed
1
3
p
a
p
e
rs
i
n
j
o
u
rn
a
l
an
d
i
n
t
e
r
n
at
i
o
n
al
c
o
n
f
e
ren
ce
s
.
Sh
e
i
s
a
l
i
f
e
m
em
b
e
r
o
f
IST
E
.
Sh
e
c
a
n
b
e
co
n
t
a
c
t
ed
at
em
a
i
l
:
b
v
i
j
i
.
ak
i
l
a
@
g
m
a
i
l
.
co
m
.
Th
a
n
g
a
R
a
m
y
a
Sh
a
nm
u
g
a
v
el
B.
E
,
M.
S
.
(b
y
R
es
),
Ph
.
D
.
,
i
s
w
o
rk
i
n
g
as
a
n
a
s
s
o
c
i
at
e
p
ro
f
e
s
s
o
r
i
n
t
h
e
D
e
p
art
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
en
ce
a
n
d
E
n
g
i
n
ee
r
i
n
g
,
R
.
M.
K
.
E
n
g
i
n
ee
r
i
n
g
Co
l
l
eg
e
,
Ch
en
n
ai
.
S
h
e
o
b
t
ai
n
ed
h
e
r
B
.
E
.
(CSE
)
fr
o
m
D
r.
Si
v
an
t
h
i
A
d
i
t
an
ar
Co
l
l
eg
e
o
f
E
n
g
i
n
ee
r
i
n
g
an
d
M.
S
.
b
y
R
e
s
e
ar
ch
(
I
CE
)
fr
o
m
A
n
n
a
U
n
i
v
e
rs
i
t
y
,
C
h
e
n
n
ai
.
Sh
e
h
as
o
b
t
ai
n
ed
h
e
r
Ph
.
D
.
i
n
I
n
f
o
r
m
at
i
o
n
an
d
C
o
mm
u
n
i
c
at
i
o
n
E
n
g
i
n
ee
r
i
n
g
fro
m
A
n
n
a
U
n
i
v
e
rs
i
t
y
,
Ch
en
n
a
i
,
i
n
2
0
1
7
.
Sh
e
h
as
b
ee
n
i
n
t
h
e
t
e
ach
i
n
g
p
ro
fe
s
s
i
o
n
f
o
r
t
h
e
p
as
t
2
0
y
e
ars
an
d
p
u
b
l
i
s
h
e
d
mo
re
t
h
an
3
0
p
ap
e
rs
i
n
v
ari
o
u
s
In
t
e
rn
at
i
o
n
a
l
J
o
u
rn
a
l
s
an
d
Co
n
fe
r
en
ce
.
H
e
r
ar
e
as
o
f
i
n
t
e
r
e
s
t
i
n
c
l
u
d
e
p
ro
g
ra
mmi
n
g
l
a
n
g
u
ag
e
s
,
d
at
ab
as
e
m
a
n
ag
eme
n
t
,
c
l
o
u
d
c
o
m
p
u
t
i
n
g
an
d
d
at
a
m
i
n
i
n
g
.
Sh
e
h
as
g
o
t
o
ra
c
l
e
j
a
v
a
i
n
t
e
rn
at
i
o
n
a
l
ce
rt
i
f
i
c
at
i
o
n
,
I
BM
RA
D
,
T
IV
O
L
I
,
a
n
d
D
B2
c
e
rt
i
f
i
c
at
i
o
n
an
d
o
b
t
ai
n
e
d
h
e
r
PRP
ce
rt
i
f
i
c
at
i
o
n
fro
m
W
i
p
ro
.
S
h
e
i
s
t
h
e
l
i
f
e
mem
b
e
r
o
f
I
ST
E
.
S
h
e
c
an
b
e
c
o
n
t
ac
t
e
d
at
em
a
i
l
:
t
h
a
n
g
ara
my
a
@
g
m
a
i
l
.
c
o
m
.
R
a
m
a
r
Ka
da
rk
a
ra
y
a
n
di
c
u
rr
e
n
t
l
y
w
o
rk
i
n
g
as
a
p
ri
n
ci
p
al
i
n
,
R.
M.
K
.
Co
l
l
eg
e
o
f
E
n
g
i
n
ee
r
i
n
g
an
d
T
ec
h
n
o
l
o
g
y
,
C
h
e
n
n
ai
.
H
e
o
b
t
ai
n
e
d
h
i
s
b
ac
h
el
o
r’s
d
eg
r
ee
i
n
E
l
ec
t
ro
n
i
c
s
an
d
Co
mm
u
n
i
c
at
i
o
n
E
n
g
i
n
ee
r
i
n
g
fro
m
G
o
v
e
rn
men
t
Co
l
l
e
g
e
o
f
E
n
g
i
n
ee
ri
n
g
,
T
i
ru
n
el
v
el
i
a
n
d
P
o
s
t
G
rad
u
at
i
o
n
i
n
A
p
p
l
i
e
d
E
l
ec
t
ro
n
i
c
s
fr
o
m
PSG
Co
l
l
eg
e
o
f
T
ech
n
o
l
o
g
y
,
C
o
i
m
b
at
o
r
e
.
H
e
o
b
t
ai
n
ed
h
i
s
D
o
c
t
o
ra
l
D
e
g
r
ee
i
n
C
o
m
p
u
t
e
r
S
ci
e
n
ce
E
n
g
i
n
ee
ri
n
g
fro
m
Man
o
n
m
a
n
i
a
u
m
Su
n
d
aran
ar
U
n
i
v
e
rs
i
t
y
,
T
i
ru
n
e
l
v
el
i
i
n
2
0
0
1
.
H
e
h
as
3
4
y
e
ars
o
f
t
e
a
ch
i
n
g
e
x
p
e
r
i
e
n
ce
i
n
c
l
u
d
i
n
g
2
9
y
e
ars
o
f
R
e
s
e
ar
c
h
e
x
p
e
r
i
e
n
ce.
H
i
s
e
x
t
e
n
s
i
v
e
r
e
s
e
ar
c
h
e
x
p
e
ri
en
ce
h
as
p
ro
d
u
ced
4
6
Ph
.
D
.
s
o
far
a
n
d
5
S
ch
o
l
ars
ar
e
p
u
rs
i
n
g
P
h
.
D
.
H
e
h
as
p
u
b
l
i
s
h
e
d
1
3
2
re
s
e
ar
c
h
art
i
c
l
e
s
i
n
v
ari
o
u
s
i
n
t
e
r
n
at
i
o
n
al
j
o
u
r
n
al
s
an
d
9
1
o
f
t
h
em
ar
e
S
c
o
p
u
s
In
d
e
x
ed
.
H
e
c
an
b
e
co
n
t
a
c
t
ed
at
em
a
i
l
:
k
ram
ar
.
e
i
n
s
t
e
i
n
@
g
m
ai
l
.
c
o
m
.
V
i
ja
y
a
l
a
k
s
hm
i
Ka
n
d
h
a
s
a
m
y
recei
v
ed
h
e
r
b
a
c
h
el
o
r’s
d
e
g
r
ee
an
d
a
m
as
t
e
r’s
i
n
c
o
m
p
u
t
e
r
s
ci
e
n
ce
a
n
d
en
g
i
n
ee
r
i
n
g
fro
m
Ma
d
u
ra
i
K
a
maraj
U
n
i
v
e
rs
i
t
y
,
Mad
u
rai
,
T
a
m
i
l
n
a
d
u
,
I
n
d
i
a
.
S
h
e
o
b
t
ai
n
e
d
P
h
.
D
.
d
e
g
r
ee
i
n
C
o
m
p
u
t
e
r
S
ci
en
ce
a
n
d
E
n
g
i
n
ee
r
i
n
g
fr
o
m
A
n
n
a
U
n
i
v
e
rs
i
t
y
,
C
h
e
n
n
ai
,
T
a
m
i
l
n
a
d
u
,
I
n
d
i
a,
i
n
2
0
0
8
.
H
er
re
s
e
ar
ch
i
n
t
e
r
e
s
t
s
i
n
c
l
u
d
e
art
i
f
ci
al
i
n
t
el
l
i
g
en
ce
,
n
e
t
w
o
rk
s
,
b
i
g
d
at
a
a
n
al
y
t
i
c
s
,
a
n
d
o
p
t
i
m
i
z
at
i
o
n
.
Sh
e
i
s
c
u
rr
en
t
l
y
w
o
rk
i
n
g
as
a
p
ro
fe
s
s
o
r
an
d
h
e
ad
in
t
h
e
D
e
p
art
me
n
t
o
f
C
o
m
p
u
t
e
r
Sc
i
en
ce
an
d
E
n
g
i
n
ee
ri
n
g
at
Ramco
I
n
s
t
i
t
u
t
e
o
f
T
ec
h
n
o
l
o
g
y
,
Raj
ap
al
a
y
a
m
,
T
a
m
i
l
N
a
d
u
,
I
n
d
i
a.
S
h
e
h
as
p
u
b
l
i
s
h
e
d
3
0
art
i
c
l
e
s
i
n
p
e
e
r
r
e
v
i
ew
ed
j
o
u
r
n
al
s
an
d
p
r
e
s
en
t
ed
p
ap
e
rs
i
n
m
o
r
e
t
h
a
n
7
0
n
at
i
o
n
al
a
n
d
i
n
t
e
rn
at
i
o
n
a
l
c
o
n
f
e
r
e
n
ce
s
.
T
o
t
al
l
y
e
i
g
h
t
s
ch
o
l
ars
c
o
m
p
l
e
t
e
d
t
h
ei
r
Ph
.
D
.
u
n
d
e
r
h
e
r
s
u
p
e
r
v
i
s
i
o
n
i
n
v
ari
o
u
s
d
o
m
ai
n
s
o
f
Co
m
p
u
t
e
r
S
ci
e
n
ce
an
d
E
n
g
i
n
ee
ri
n
g
.
Sh
e
c
a
n
b
e
co
n
t
a
c
t
ed
at
em
a
i
l
:
v
i
j
a
y
al
a
k
s
h
m
i
k
@
r
i
t
rj
p
m.
a
c
.
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.