I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
rm
a
t
ics a
nd
Co
m
m
un
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
,
p
p
.
1
1
1
9
~
1
1
2
5
I
SS
N:
2252
-
8
7
7
6
,
DOI
:
1
0
.
1
1
5
9
1
/iji
ct
.
v
1
4
i
3
.
pp
1
1
1
9
-
1
1
2
5
1119
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ict.
ia
esco
r
e.
co
m
Enha
nc
i
n
g
bi
o
de
g
r
a
da
bl
e
w
a
s
te m
a
n
a
g
e
m
ent in
M
a
ur
i
t
i
u
s
throu
g
h
rea
l
-
t
i
m
e co
m
put
er
v
i
s
i
o
n
-
b
a
s
ed
s
o
rt
i
n
g
G
ee
rish
Su
dd
ul,
Av
it
a
h B
a
b
a
j
ee
,
Nund
j
ee
t
Ra
m
ba
run,
Sa
nd
hy
a
Arm
o
o
g
u
m
D
e
p
a
r
t
me
n
t
o
f
B
u
si
n
e
ss
I
n
f
o
r
mat
i
c
s
a
n
d
S
o
f
t
w
a
r
e
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
s
i
t
y
o
f
Te
c
h
n
o
l
o
g
y
M
a
u
r
i
t
i
u
s
(
U
TM
)
,
M
a
u
r
i
t
i
u
s,
Ea
st
A
f
r
i
c
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
1
5
,
2
0
2
4
R
ev
is
ed
Feb
2
1
,
2
0
2
5
Acc
ep
ted
J
u
n
9
,
2
0
2
5
M
a
u
rit
i
u
s
fa
c
e
s
a
sig
n
ifi
c
a
n
t
wa
ste
m
a
n
a
g
e
m
e
n
t
c
h
a
ll
e
n
g
e
d
u
e
to
t
h
e
in
d
isc
rimin
a
te
m
i
x
in
g
o
f
b
i
o
d
e
g
r
a
d
a
b
le
a
n
d
n
o
n
-
b
io
d
e
g
ra
d
a
b
le
wa
ste
.
Th
is
p
ra
c
ti
c
e
h
in
d
e
rs
p
ro
p
e
r
re
c
y
c
li
n
g
a
n
d
c
o
m
p
o
stin
g
e
ffo
rts
,
c
o
n
tr
ib
u
ti
n
g
to
e
n
v
iro
n
m
e
n
tal
p
o
ll
u
ti
o
n
a
n
d
re
so
u
rc
e
d
e
p
l
e
ti
o
n
.
T
h
is
re
se
a
rc
h
p
ro
p
o
se
s
a
c
o
m
p
u
ter
v
isio
n
-
b
a
se
d
sy
ste
m
fo
r
re
a
l
-
ti
m
e
c
las
sifica
ti
o
n
o
f
wa
ste
in
to
b
io
d
e
g
ra
d
a
b
le
a
n
d
n
o
n
-
b
io
d
e
g
ra
d
a
b
le
c
a
teg
o
ries
.
Tra
n
sfe
r
lea
rn
in
g
a
p
p
r
o
a
c
h
b
a
se
d
o
n
d
e
e
p
lea
rn
in
g
m
o
d
e
l
s,
sp
e
c
ifi
c
a
ll
y
De
n
se
Ne
t1
2
1
,
M
o
b
i
leN
e
t,
In
c
e
p
ti
o
n
V3
,
VG
G
1
6
a
n
d
VG
G
1
9
we
re
e
v
a
lu
a
ted
wit
h
tw
o
d
iffere
n
t
c
las
sifiers
,
th
e
K
-
n
e
a
re
st
n
e
i
g
h
b
o
rs
(KN
N)
a
n
d
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(S
VM).
Ou
r
e
x
p
e
rime
n
ts
d
e
m
o
n
s
trate
th
a
t
th
e
M
o
b
il
e
Ne
t
p
a
ired
with
S
VM
a
c
h
iev
e
s
a
c
las
sific
a
ti
o
n
a
c
c
u
ra
c
y
o
f
9
7
%
fo
r
d
e
tec
ti
o
n
i
n
re
a
lt
ime
.
Co
m
p
a
re
d
to
o
th
e
r
stu
d
ies
,
o
u
r
re
su
lt
s
d
e
m
o
n
stra
te
b
e
tt
e
r
p
e
rfo
r
m
a
n
c
e
a
n
d
re
a
lt
ime
c
las
sifica
ti
o
n
c
a
p
a
b
il
i
ti
e
s
th
ro
u
g
h
t
h
e
imp
lem
e
n
tati
o
n
o
f
a
p
ro
t
o
ty
p
e
,
fa
c
il
it
a
t
i
n
g
a
u
t
o
m
a
ti
c
s
o
rti
n
g
o
f
wa
ste
.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
R
ea
l
-
tim
e
r
ec
o
g
n
itio
n
Su
s
tain
ab
ilit
y
T
r
an
s
f
er
lear
n
i
n
g
W
aste m
an
ag
m
en
t
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
:
Gee
r
is
h
Su
d
d
u
l
Dep
ar
tm
en
t o
f
B
u
s
in
ess
I
n
f
o
r
m
atics a
n
d
So
f
twar
e
E
n
g
in
ee
r
in
g
Un
iv
er
s
ity
o
f
T
ec
h
n
o
lo
g
y
Ma
u
r
itiu
s
(
UT
M)
L
a
T
o
u
r
K
o
en
ig
,
Ma
u
r
itiu
s
,
E
ast
Af
r
ica
E
m
ail: g
.
s
u
d
d
u
l@
u
tm
.
ac
.
m
u
1.
I
NT
RO
D
UCT
I
O
N
Ma
u
r
itiu
s
,
as
a
s
m
all
is
lan
d
d
ev
elo
p
in
g
s
tate,
f
ac
es
s
ig
n
if
ican
t
waste
m
an
ag
em
en
t
ch
allen
g
es
ex
ac
er
b
ated
b
y
lim
ited
s
p
ac
e
an
d
a
h
ig
h
p
o
p
u
latio
n
d
en
s
ity
.
T
h
e
is
lan
d
g
en
er
ated
1
.
4
m
illi
o
n
to
n
n
es o
f
waste,
ex
ce
ed
in
g
th
e
g
lo
b
al
a
v
er
ag
e
in
2
0
2
0
[
1
]
.
T
o
a
d
d
r
ess
th
is
,
v
ar
io
u
s
ap
p
r
o
ac
h
es
h
av
e
b
ee
n
im
p
lem
en
te
d
,
in
clu
d
in
g
a
s
o
lid
waste
m
a
n
ag
em
en
t
s
tr
ateg
y
f
o
c
u
s
in
g
o
n
r
eso
u
r
ce
r
ec
o
v
er
y
,
en
e
r
g
y
g
en
e
r
atio
n
,
a
n
d
co
m
m
u
n
ity
en
g
a
g
em
en
t
was
im
p
lem
en
eted
[
2
]
,
[
3
]
.
Desp
ite
s
u
ch
ef
f
o
r
ts
,
lan
d
f
ill
r
e
m
ain
s
th
e
p
r
im
ar
y
d
is
p
o
s
al
s
ite,
h
an
d
lin
g
9
7
%
o
f
th
e
waste,
b
o
th
o
r
g
a
n
ic
an
d
n
o
n
-
o
r
g
an
ic
[
4
]
.
T
h
is
u
n
d
er
s
co
r
es
th
e
u
r
g
en
t
n
ee
d
f
o
r
im
p
r
o
v
e
d
waste
s
ep
ar
atio
n
an
d
m
a
n
ag
em
en
t so
lu
tio
n
s
.
T
h
e
d
iv
er
s
ity
o
f
waste
ty
p
es,
in
f
lu
en
ce
d
b
y
s
ea
s
o
n
al
ch
an
g
es,
ec
o
n
o
m
ic
f
ac
to
r
s
,
an
d
cu
ltu
r
al
p
r
ac
tices,
m
ak
es
g
en
er
al
was
te
m
an
ag
em
en
t
a
p
p
r
o
ac
h
es
i
n
ad
eq
u
ate,
r
eq
u
i
r
in
g
c
u
s
to
m
ized
s
o
lu
tio
n
s
.
T
h
e
p
r
esen
ce
o
f
b
o
th
ce
n
tr
alize
d
an
d
d
ec
en
t
r
alize
d
waste
m
an
ag
em
en
t
s
y
s
tem
s
,
in
clu
d
in
g
in
f
o
r
m
al
s
ec
to
r
in
v
o
lv
em
e
n
t,
co
n
tr
i
b
u
tes
to
in
co
n
s
is
ten
cies
an
d
d
ata
lim
itatio
n
s
[
5
]
.
T
o
o
v
e
r
co
m
e
th
ese
ch
allen
g
es,
th
is
s
tu
d
y
aim
s
to
d
ev
elo
p
a
d
ee
p
lea
r
n
in
g
p
r
o
to
ty
p
e,
u
s
in
g
co
m
p
u
ter
v
is
io
n
to
ac
cu
r
ately
c
lass
if
y
waste
in
to
b
io
d
eg
r
a
d
ab
le
(
b
io
)
an
d
n
o
n
-
b
io
d
eg
r
ad
a
b
le
(
n
o
n
-
b
io
)
ca
teg
o
r
ies.
W
ith
th
e
in
teg
r
atio
n
o
f
r
e
altim
e
r
ec
o
g
n
itio
n
ca
p
ab
ilit
ies,
th
e
p
r
o
p
o
s
ed
s
y
s
tem
ca
n
ass
is
t
au
to
m
atic
s
o
r
tin
g
a
n
d
im
p
r
o
v
e
r
ec
y
clin
g
r
ates
an
d
r
eso
u
r
ce
r
ec
o
v
er
y
,
alig
n
in
g
with
Ma
u
r
itiu
s
’
co
m
m
itm
en
t
to
ac
h
iev
i
n
g
th
e
Un
ited
Natio
n
s
Su
s
tain
ab
le
Dev
elo
p
m
en
t
Go
als (
UN
SDG)
.
E
f
f
ec
tiv
e
waste
m
an
ag
em
en
t
is
cr
u
cial
f
o
r
s
u
s
ta
in
ab
le
d
ev
elo
p
m
en
t.
Sep
ar
atin
g
tr
ash
in
to
b
io
(
o
r
g
an
ic
)
an
d
n
o
n
-
b
i
o
(
in
o
r
g
an
ic)
g
r
o
u
p
s
is
n
ec
ess
ar
y
f
o
r
r
ec
y
clin
g
,
r
ed
u
cin
g
lan
d
f
ill
d
ep
en
d
en
cy
,
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
:
1
1
1
9
-
1
1
2
5
1120
d
is
p
o
s
al
s
tr
ateg
ies
[
6
]
.
Ho
wev
er
,
th
e
d
iv
er
s
ity
o
f
waste
m
ater
ial
v
ar
y
in
g
in
co
m
p
o
s
itio
n
,
s
h
ap
e,
s
ize,
tex
t
u
r
e,
an
d
c
o
lo
r
p
r
esen
ts
a
s
ig
n
if
ica
n
t
ch
allen
g
e
to
tr
ad
itio
n
al
cla
s
s
if
icatio
n
m
eth
o
d
s
,
as
p
er
M
o
h
am
m
ed
et
a
l.
[
7
]
.
Dee
p
lear
n
in
g
tec
h
n
iq
u
e
with
its
p
o
ten
tial
to
lear
n
co
m
p
le
x
p
atter
n
s
f
r
o
m
lar
g
e
d
atasets
,
in
clu
d
in
g
im
ag
es,
ca
n
o
f
f
er
a
v
iab
le
s
o
lu
tio
n
.
N
ev
er
th
eless
,
v
ar
iab
ilit
y
in
waste
ap
p
ea
r
an
ce
d
u
e
to
d
if
f
er
en
ce
s
in
co
m
p
o
s
itio
n
,
cu
ltu
r
al
p
r
ac
tices,
an
d
g
eo
g
r
a
p
h
ical
lo
ca
tio
n
s
r
eq
u
ir
es
r
o
b
u
s
t
alg
o
r
ith
m
s
th
at
ca
n
g
en
er
al
ize
ac
r
o
s
s
d
iv
er
s
e
d
atasets
[
8
]
.
T
r
ad
itio
n
al
m
et
h
o
d
s
o
f
te
n
s
tr
u
g
g
le
with
m
a
n
u
al
f
ea
tu
r
e
ex
tr
ac
tio
n
,
lead
i
n
g
to
i
n
co
n
s
is
ten
t
p
er
f
o
r
m
an
ce
ac
r
o
s
s
d
if
f
er
e
n
t
waste
ty
p
es
[
9
]
.
Nev
er
th
eless
,
d
ee
p
lear
n
in
g
m
o
d
els
ca
n
f
ac
e
th
e
r
is
k
o
f
o
v
er
f
itti
n
g
w
h
en
tr
ai
n
ed
o
n
li
m
ited
d
atasets
b
y
ca
p
tu
r
in
g
t
o
o
m
u
c
h
d
etails.
T
h
is
ca
n
r
e
s
u
lt
in
m
o
d
els
th
at
u
n
d
er
p
er
f
o
r
m
wh
e
n
class
if
y
in
g
n
ew,
u
n
s
ee
n
im
ag
es
[
1
0
]
.
Nu
m
er
o
u
s
s
tu
d
ies
h
av
e
ap
p
lied
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN)
a
lg
o
r
i
th
m
s
f
o
r
au
to
m
atic
waste
class
if
icatio
n
,
ac
h
iev
in
g
v
ar
y
in
g
lev
els
o
f
ac
cu
r
ac
y
[
1
1
]
-
[
1
3
]
.
T
r
ain
i
n
g
C
NNs
f
r
o
m
s
cr
atch
ca
n
b
e
tim
e
c
o
n
s
u
m
in
g
an
d
r
eso
u
r
ce
in
ten
s
iv
e,
p
ar
ticu
lar
ly
f
o
r
lar
g
e
d
atasets
[
1
4
]
.
T
r
an
s
f
e
r
lear
n
i
n
g
,
wh
ich
i
n
v
o
lv
es
th
e
f
in
e
-
t
u
n
in
g
o
f
p
r
e
-
tr
ain
ed
m
o
d
els
o
n
s
p
ec
if
ic
d
atasets
,
h
as
p
r
o
v
e
n
to
b
e
a
v
alu
a
b
le
a
p
p
r
o
ac
h
.
Stu
d
ies
h
av
e
s
h
o
wn
th
at
tr
an
s
f
er
lear
n
i
n
g
ca
n
s
ig
n
if
ican
tly
im
p
r
o
v
e
class
if
icatio
n
ac
cu
r
ac
y
wh
ile
r
ed
u
cin
g
tr
ai
n
in
g
tim
e.
Yan
g
an
d
T
h
u
n
g
[
1
5
]
u
s
ed
a
p
r
e
-
t
r
ain
ed
I
n
ce
p
tio
n
V3
mo
d
el
to
class
if
y
waste
im
a
g
es
f
r
o
m
th
e
T
r
ash
Net
d
ata
s
et,
ac
h
iev
in
g
a
r
em
ar
k
a
b
le
ac
cu
r
ac
y
o
f
8
3
%.
Similar
ly
,
B
ian
co
et
a
l.
[
1
6
]
em
p
lo
y
ed
th
e
R
esNet5
0
ar
ch
itectu
r
e
p
r
e
-
tr
ain
e
d
o
n
I
m
ag
eNe
t,
f
o
r
th
e
class
if
icatio
n
o
f
m
u
n
icip
al
s
o
l
id
waste
im
ag
es,
r
esu
ltin
g
in
a
n
ac
cu
r
ac
y
o
f
9
5
%.
T
h
ese
s
tu
d
ies
h
ig
h
lig
h
t
t
h
e
p
o
ten
tial
o
f
tr
an
s
f
er
lear
n
in
g
in
r
ed
u
cin
g
tr
ai
n
in
g
tim
es
an
d
im
p
r
o
v
i
n
g
m
o
d
el
p
er
f
o
r
m
a
n
ce
.
Fer
r
eir
a
et
a
l.
[
1
7
]
in
v
esti
g
ated
th
e
u
s
e
o
f
tr
an
s
f
er
lear
n
in
g
f
o
r
waste
class
if
icatio
n
b
y
f
in
e
-
tu
n
in
g
p
r
e
-
t
r
ain
ed
m
o
d
els,
s
u
ch
as
VGG1
6
,
R
e
s
Net5
0
,
an
d
I
n
ce
p
tio
n
V3
,
o
n
a
waste
class
if
ic
atio
n
d
ataset.
T
h
ey
ac
h
iev
ed
a
n
ac
cu
r
ac
y
o
f
9
2
%
u
s
in
g
th
e
R
esNet5
0
ar
ch
itectu
r
e,
d
em
o
n
s
tr
atin
g
th
e
ef
f
ec
ti
v
en
ess
o
f
tr
an
s
f
er
lear
n
in
g
in
ca
p
tu
r
in
g
in
tr
icate
waste
f
ea
tu
r
es.
R
ak
h
r
a
et
a
l.
[1
8
]
s
tu
d
y
a
p
p
lied
tr
a
n
s
f
er
l
ea
r
n
in
g
u
s
in
g
th
e
M
o
b
ileNetV2
ar
ch
itectu
r
e
f
o
r
waste
clas
s
if
icatio
n
u
s
in
g
a
cu
s
to
m
ized
I
m
ag
eNe
t
-
p
r
e
-
tr
ain
ed
Mo
b
ileNetV2
ac
h
iev
in
g
a
n
ac
cu
r
ac
y
o
f
8
9
%.
I
n
2
0
2
2
,
R
az
za
q
et
a
l.
[
1
9
]
ex
p
lo
r
ed
a
p
r
e
-
tr
ain
e
d
Den
s
eNe
t1
2
,
ac
h
iev
in
g
a
n
ac
c
u
r
ac
y
o
f
9
3
%,
th
er
e
b
y
d
em
o
n
s
tr
atin
g
th
e
m
o
d
el’
s
ab
ilit
y
to
ca
p
tu
r
e
in
tr
icate
waste
f
ea
tu
r
es.
R
ec
en
t
ad
v
an
ce
m
e
n
ts
in
d
ee
p
lear
n
in
g
,
s
u
ch
as
th
e
E
f
f
icien
tNet
ar
ch
i
tectu
r
e,
h
av
e
f
u
r
th
e
r
en
h
a
n
ce
d
waste
class
if
icat
io
n
p
er
f
o
r
m
an
ce
.
Vijay
ak
u
m
a
r
et
a
l.
[
2
0
]
ac
h
iev
ed
9
1
%
ac
cu
r
ac
y
o
n
th
e
W
asteNet
d
atas
et
u
s
in
g
E
f
f
icien
tNet
-
B
0
.
T
r
an
s
f
er
lear
n
in
g
,
with
p
r
etr
ain
ed
m
o
d
els
lik
e
VGG1
6
,
R
esNet5
0
,
an
d
Mo
b
ileNet,
is
p
ar
ticu
lar
ly
e
f
f
ec
tiv
e
f
o
r
wa
s
te
class
if
icatio
n
.
I
t
h
as
b
ee
n
n
o
ted
th
at
th
ese
m
o
d
els
ca
n
ac
h
iev
e
h
i
g
h
ac
cu
r
ac
y
,
o
f
ten
ex
ce
ed
in
g
9
0
%,
wh
ile
r
ed
u
cin
g
tr
ain
in
g
tim
es c
o
m
p
ar
ed
to
m
o
d
els b
u
ilt f
r
o
m
s
cr
atch
.
T
h
er
eb
y
,
u
s
in
g
tr
a
n
s
f
er
lear
n
i
n
g
f
o
r
waste
class
if
icatio
n
is
p
ar
ticu
lar
ly
r
elev
an
t
f
o
r
Ma
u
r
itiu
s
,
wh
er
e
tr
ad
itio
n
al
waste
m
an
ag
em
en
t
m
eth
o
d
s
f
ac
e
c
h
allen
g
es d
u
e
t
o
d
iv
er
s
e
an
d
v
ar
ia
b
le
waste
m
ater
ials
.
2.
M
E
T
H
O
D
2
.
1
.
Da
t
a
s
et
a
nd
pre
-
pro
ce
s
s
ing
T
h
e
"No
n
-
B
io
d
e
g
r
ad
ab
le
a
n
d
B
io
d
eg
r
ad
ab
le
W
aste
Data
s
e
t
"
d
ataset
[
2
1
]
h
as
b
ee
n
ch
o
s
en
as
th
e
im
ag
es
clo
s
ely
r
esem
b
le
th
e
ty
p
es
o
f
waste
f
o
u
n
d
in
t
h
e
Ma
u
r
itian
co
n
tex
t
.
T
h
e
d
a
tas
et
co
m
p
r
is
es
o
f
ap
p
r
o
x
im
ately
1
6
,
7
2
6
im
ag
es,
ca
teg
o
r
ized
p
r
im
ar
ily
i
n
to
tw
o
g
r
o
u
p
s
,
B
io
d
eg
r
a
d
ab
le
an
d
No
n
-
b
io
d
eg
r
ad
a
b
le
waste,
eq
u
ally
d
is
tr
ib
u
ted
.
B
i
o
d
eg
r
a
d
ab
le
waste
en
co
m
p
ass
es
o
r
g
an
ic
m
ater
ials
s
u
ch
as
f
o
o
d
s
cr
ap
s
,
p
lan
ts
,
an
d
f
r
u
its
,
wh
ich
n
at
u
r
ally
d
ec
o
m
p
o
s
e
an
d
ca
n
b
e
co
n
v
er
ted
in
to
co
m
p
o
s
t.
T
h
is
p
r
o
ce
s
s
i
s
ess
en
tia
l
f
o
r
r
ec
y
clin
g
n
u
tr
ie
n
ts
an
d
m
in
i
m
izin
g
lan
d
f
ill
u
s
e.
C
o
n
v
er
s
e
ly
,
n
o
n
-
b
io
d
eg
r
ad
ab
le
waste
in
clu
d
es
s
u
b
s
tan
ce
s
lik
e
p
last
ics an
d
m
etals th
at
d
o
n
o
t d
ec
o
m
p
o
s
e
n
atu
r
ally
.
All
th
e
im
ag
es
h
av
e
b
ee
n
r
esized
to
1
5
0
x
1
5
0
p
ix
els,
an
d
th
e
f
o
llo
win
g
d
ata
au
g
m
en
tatio
n
ap
p
r
o
ac
h
es
h
av
e
b
ee
n
a
p
p
lie
d
:
n
o
r
m
alizin
g
p
i
x
el
v
alu
es,
r
escalin
g
b
y
1
/2
5
5
,
s
h
ea
r
r
an
g
e
an
d
z
o
o
m
o
f
0
.
2
,
an
d
h
o
r
izo
n
tal
f
lip
p
in
g
.
A
v
ali
d
atio
n
s
p
lit o
f
2
0
% h
as b
ee
n
r
eser
v
ed
f
o
r
ev
alu
at
io
n
.
2
.
2.
Appro
a
ch
Ou
r
ap
p
r
o
ac
h
c
o
n
s
is
ts
o
f
u
s
in
g
tr
ad
itio
n
al
p
r
e
-
tr
ai
n
ed
C
NNs
an
d
ex
p
lo
r
e
th
e
ef
f
ec
t
o
f
d
if
f
er
en
t
class
if
ier
s
o
n
th
eir
p
er
f
o
r
m
a
n
ce
.
W
e
d
esig
n
e
d
o
u
r
ex
p
e
r
im
en
ts
to
u
s
e
VGG1
6
,
VGG1
9
,
Mo
b
ileNet,
Den
s
eNe
t1
2
1
,
an
d
I
n
ce
p
tio
n
V3
with
two
class
if
ier
s
,
th
e
K
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
.
T
h
e
p
r
o
ce
s
s
s
tar
ts
b
y
tr
ain
in
g
th
e
C
NN
m
o
d
els
o
n
th
e
well
-
estab
lis
h
ed
I
m
ag
eNe
t
d
ataset
in
o
r
d
er
to
estab
lis
h
a
b
aselin
e.
T
h
en
,
e
x
p
er
im
e
n
ts
with
KNN
an
d
SVM
ar
e
co
n
d
u
cted
as
c
lass
i
f
icatio
n
lay
er
s
.
KNN
is
k
n
o
wn
to
im
p
r
o
v
e
ac
cu
r
ac
y
b
y
clu
s
ter
in
g
s
im
il
ar
d
ata
p
o
i
n
ts
an
d
r
e
d
u
cin
g
n
o
is
e,
esp
ec
ially
in
co
m
p
lex
d
atasets
[
2
2
]
.
SVM
alg
o
r
ith
m
ex
ce
ls
in
h
ig
h
-
d
im
en
s
io
n
al
s
p
ac
es b
y
m
ax
im
izin
g
th
e
m
ar
g
in
b
etwe
en
th
e
class
es,
an
d
ca
n
th
er
ef
o
r
e
i
d
e
n
tify
th
e
m
o
s
t e
f
f
ec
tiv
e
h
y
p
er
p
lan
e
f
o
r
class
d
iv
is
io
n
[
2
3
]
.
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
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
E
n
h
a
n
ci
n
g
b
i
o
d
eg
r
a
d
a
b
le
w
a
s
te
ma
n
a
g
eme
n
t i
n
Ma
u
r
itiu
s
th
r
o
u
g
h
r
ea
l
-
time
…
(
Gee
r
is
h
S
u
d
d
u
l
)
1121
2.
2
.1
.
B
a
s
e
m
o
dels
E
ac
h
p
r
e
-
t
r
ain
ed
m
o
d
el
is
ad
a
p
ted
f
o
r
b
in
ar
y
class
if
icatio
n
b
y
im
p
o
r
tin
g
th
eir
p
r
e
-
tr
ain
e
d
weig
h
ts
o
n
th
e
I
m
ag
eNe
t
d
ataset,
ex
cl
u
d
i
n
g
th
e
to
p
class
if
icatio
n
lay
er
,
ad
d
in
g
cu
s
to
m
lay
er
s
(
Glo
b
al
Av
er
ag
ePo
o
lin
g
2
D
to
r
ed
u
ce
s
p
atial
d
im
e
n
s
io
n
s
,
a
Den
s
e
lay
er
wit
h
5
1
2
u
n
its
an
d
R
eL
U
ac
tiv
atio
n
,
an
o
p
ti
o
n
al
Dr
o
p
o
u
t
lay
er
with
a
r
ate
o
f
0
.
5
f
o
r
VGG1
6
an
d
VGG1
9
,
a
n
d
an
o
u
tp
u
t
la
y
er
with
a
s
in
g
le
u
n
it
an
d
s
ig
m
o
id
ac
tiv
atio
n
)
.
All
lay
er
s
o
f
th
e
b
ase
m
o
d
el
ar
e
th
en
f
r
o
ze
n
.
T
h
e
m
o
d
els
ar
e
co
m
p
iled
u
s
in
g
th
e
Ad
am
o
p
tim
ize
r
with
a
lear
n
in
g
r
ate
o
f
0
.
0
0
0
1
,
b
in
ar
y
cr
o
s
s
-
en
tr
o
p
y
l
o
s
s
f
u
n
ctio
n
,
a
n
d
ac
c
u
r
ac
y
as
th
e
ev
alu
atio
n
m
etr
ic
,
an
d
tr
ain
e
d
f
o
r
u
p
t
o
5
0
e
p
o
ch
s
with
ca
llb
ac
k
s
f
o
r
s
av
in
g
th
e
b
est
-
p
e
r
f
o
r
m
in
g
m
o
d
el
an
d
h
altin
g
tr
ain
in
g
to
p
r
ev
en
t
o
v
er
f
itti
n
g
.
2.
2
.2
.
Cus
t
o
m
ized
m
o
dels
I
n
s
tead
o
f
u
s
in
g
th
e
m
o
d
els
f
o
r
d
ir
ec
t
class
if
icatio
n
,
we
f
o
cu
s
o
n
th
ese
two
cu
s
to
m
f
u
n
ctio
n
s
to
r
esh
ap
e
th
e
f
ea
tu
r
e
m
ap
s
g
e
n
e
r
ated
b
y
t
h
e
m
o
d
els:
−
T
r
an
s
f
er
L
ea
r
n
i
n
g
with
KNN
:
KNN
co
m
p
ar
es
a
n
ew
im
ag
e
to
its
m
o
s
t
s
im
ilar
n
eig
h
b
o
r
s
f
r
o
m
th
e
tr
ain
in
g
d
ata
(
u
s
in
g
5
n
eig
h
b
o
r
s
f
o
r
b
alan
ce
d
ac
cu
r
ac
y
)
to
d
eter
m
in
e
its
class
.
T
h
is
s
im
p
le
y
et
p
o
wer
f
u
l
ap
p
r
o
ac
h
is
wid
ely
u
s
ed
in
tas
k
s
lik
e
f
ac
ial
r
ec
o
g
n
itio
n
[
2
4
]
.
−
T
r
an
s
f
er
L
ea
r
n
in
g
wit
h
SVM
:
T
h
is
tech
n
iq
u
e
lev
er
ag
es
f
ea
tu
r
e
m
ap
s
to
class
if
y
im
ag
es
u
s
in
g
an
SVM
with
a
lin
ea
r
k
er
n
el.
Oth
er
k
er
n
els
ca
n
b
e
ex
p
l
o
r
ed
f
o
r
f
u
r
th
er
o
p
tim
izatio
n
.
SVMs
ex
ce
l
in
v
ar
io
u
s
im
ag
e
an
aly
s
is
task
s
lik
e
o
b
ject
d
etec
tio
n
[
2
5
]
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
E
x
perim
ent
s
R
esu
lts
f
r
o
m
th
e
ex
p
e
r
im
en
ts
ar
e
p
r
esen
ted
in
T
ab
le
1
.
A
co
m
p
ar
is
o
n
o
f
th
e
d
if
f
e
r
en
t
a
p
p
r
o
ac
h
es
s
h
o
ws
s
ig
n
if
ican
t
im
p
r
o
v
em
e
n
t
in
p
e
r
f
o
r
m
an
ce
.
W
h
ile
De
n
s
eNe
t1
2
1
an
d
I
n
ce
p
tio
n
V3
ac
h
iev
e
th
e
h
ig
h
est
tr
ain
in
g
ac
cu
r
ac
ies
at
9
9
.
7
9
%
an
d
9
9
.
0
3
%
r
esp
ec
t
iv
el
y
,
th
eir
test
ac
cu
r
ac
ies
ar
e
o
n
l
y
5
1
.
0
%,
in
d
icatin
g
p
o
ten
tial
o
v
er
f
itti
n
g
.
T
h
is
d
is
cr
ep
an
cy
b
etwe
en
tr
ai
n
in
g
an
d
test
in
g
ac
cu
r
ac
y
is
o
b
s
er
v
e
d
ac
r
o
s
s
all
m
o
d
els,
s
u
g
g
esti
n
g
th
e
n
ee
d
f
o
r
f
u
r
th
er
tu
n
in
g
a
n
d
r
e
g
u
lar
izatio
n
tech
n
iq
u
es to
im
p
r
o
v
e
g
en
er
aliza
t
i
o
n
.
T
ab
le
1
.
E
x
p
er
im
en
tal
r
esu
lts
M
o
d
e
l
Ep
o
c
h
Tr
a
i
n
a
c
c
u
r
a
c
y
%
Te
st
a
c
c
u
r
a
c
y
%
Te
st
a
c
c
u
r
a
c
y
%
Te
st
a
c
c
u
r
a
c
y
%
W
i
t
h
o
u
t
C
l
a
s
si
f
i
e
r
W
i
t
h
o
u
t
C
l
a
s
si
f
i
e
r
W
i
t
h
K
N
N
C
l
a
s
si
f
i
e
r
W
i
t
h
S
V
M
C
l
a
s
si
f
i
e
r
M
o
b
i
l
e
N
e
t
14
9
8
.
8
9
5
0
.
0
9
5
.
0
9
7
.
0
V
G
G
1
6
12
9
5
.
9
4
9
.
0
8
9
.
0
9
5
.
0
V
G
G
1
9
20
9
5
.
8
9
5
0
.
0
9
0
.
0
9
0
.
0
I
n
c
e
p
t
i
o
n
V
3
16
9
9
.
0
3
5
1
.
0
9
4
.
0
9
6
.
0
D
e
n
seN
e
t
1
2
1
20
9
9
.
7
9
5
1
.
0
9
6
.
0
9
4
.
0
W
h
en
p
air
ed
with
class
if
ier
s
,
th
e
m
o
d
els
ex
h
i
b
it
v
ar
y
i
n
g
p
e
r
f
o
r
m
a
n
ce
.
Mo
b
ileNet
ac
h
ie
v
es
th
e
b
est
test
ac
cu
r
ac
y
o
f
9
7
%
with
th
e
SVM
wh
ile
I
n
ce
p
tio
n
V3
p
er
f
o
r
m
s
q
u
ite
well
with
SVM
,
r
e
ac
h
in
g
9
6
.
0
%
test
ac
cu
r
ac
y
.
T
h
e
VGG
m
o
d
els
d
em
o
n
s
tr
ate
lo
wer
tr
ain
in
g
an
d
test
in
g
ac
cu
r
ac
ies
co
m
p
ar
e
d
t
o
th
e
o
th
e
r
m
o
d
els.
T
h
e
Den
s
eNe
t1
2
1
d
em
o
n
s
tr
ates
a
p
er
f
o
r
m
an
ce
o
f
9
6
%
with
th
e
KNN
class
if
ier
,
as
c
o
m
p
ar
ed
t
o
o
n
ly
5
1
%
with
o
u
t
an
y
class
if
ier
.
T
h
e
ch
o
ice
o
f
class
if
ier
is
s
ig
n
if
ica
n
t,
with
SVM
g
en
er
ally
o
u
tp
er
f
o
r
m
in
g
KNN
f
o
r
mo
s
t
m
o
d
els.
Ho
wev
er
,
th
e
tr
ain
in
g
ac
cu
r
ac
y
o
f
th
e
Den
s
eNe
t1
2
1
m
o
d
el
at
9
9
.
7
9
%
an
d
I
n
ce
p
tio
n
V3
at
9
9
.
0
3
% c
lear
l
y
ex
h
i
b
it si
g
n
s
f
o
r
o
v
e
r
f
itti
n
g
.
T
h
e
ex
p
er
im
e
n
ts
em
p
lo
y
ed
e
ar
ly
s
to
p
p
in
g
to
p
r
ev
en
t
o
v
e
r
f
itti
n
g
an
d
en
s
u
r
e
m
o
d
el
p
e
r
f
o
r
m
a
n
ce
d
u
r
in
g
v
alid
atio
n
.
T
h
e
b
atc
h
s
ize,
wh
ich
d
eter
m
in
es
th
e
n
u
m
b
er
o
f
im
ag
es
g
e
n
er
ated
f
o
r
ea
ch
s
tep
o
f
an
ep
o
ch
,
in
c
r
ea
s
ed
th
r
o
u
g
h
o
u
t
th
e
ex
p
er
im
e
n
ts
.
T
h
e
n
u
m
b
er
o
f
ep
o
c
h
s
an
d
s
tep
s
p
er
ep
o
c
h
v
ar
ied
,
with
f
ix
e
d
v
alu
es
cr
ea
tin
g
tr
ad
e
-
o
f
f
s
b
et
wee
n
ea
r
ly
s
to
p
p
in
g
a
n
d
o
v
er
f
itti
n
g
r
is
k
s
.
T
h
e
r
esu
lts
s
u
g
g
est
th
at
lar
g
er
b
atch
s
izes
en
h
an
ce
f
ea
tu
r
e
i
d
en
tif
icatio
n
ca
p
ab
ilit
ies
b
y
ex
p
o
s
in
g
m
o
d
els
to
a
g
r
ea
ter
v
ar
i
ety
o
f
im
a
g
es
an
d
au
g
m
en
tatio
n
s
d
u
r
in
g
tr
ain
i
n
g
,
b
u
t
th
is
is
co
n
s
tr
ain
ed
b
y
m
em
o
r
y
lim
itatio
n
s
,
r
estrictin
g
th
e
b
atch
s
ize
to
a
m
ax
im
u
m
o
f
2
5
6
.
T
h
e
class
if
icatio
n
r
ep
o
r
t
f
o
r
th
e
b
est
-
p
er
f
o
r
m
in
g
,
Mo
b
ileNet
m
o
d
el
with
SVM
C
las
s
if
ier
d
em
o
n
s
tr
ates
a
h
ig
h
p
r
ec
is
io
n
(
0
.
9
6
f
o
r
B
io
,
0
.
9
9
f
o
r
No
n
-
b
io
)
a
n
d
r
ec
all
(
0
.
9
9
f
o
r
B
io
,
0
.
9
6
f
o
r
No
n
-
b
io
)
,
lead
in
g
to
an
F1
-
s
co
r
e
o
f
0
.
9
7
f
o
r
b
o
th
class
es.
B
o
th
ca
teg
o
r
ies
s
h
o
wed
eq
u
al
s
u
p
p
o
r
t
w
ith
1
6
7
2
i
n
s
tan
ce
s
ea
ch
,
co
n
tr
i
b
u
tin
g
to
an
o
v
er
a
ll
ac
cu
r
ac
y
o
f
9
7
%.
T
h
e
c
o
n
s
is
ten
t
m
ac
r
o
an
d
weig
h
ted
av
e
r
ag
es,
b
o
th
at
0
.
9
7
,
u
n
d
er
s
co
r
e
th
e
b
alan
ce
d
an
d
e
f
f
ec
tiv
e
p
er
f
o
r
m
a
n
ce
o
f
th
is
m
o
d
el
co
m
b
in
atio
n
ac
r
o
s
s
th
e
ca
teg
o
r
ies.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
:
1
1
1
9
-
1
1
2
5
1122
T
h
e
co
n
f
u
s
io
n
m
atr
ix
f
r
o
m
Fig
u
r
e
1
,
in
d
icate
s
a
s
tr
o
n
g
p
er
f
o
r
m
an
ce
in
b
o
th
class
es
with
h
ig
h
n
u
m
b
er
s
o
f
tr
u
e
p
o
s
itiv
es.
T
h
e
m
o
d
el
h
as
a
h
ig
h
e
r
r
ate
o
f
f
alse
n
eg
ativ
es
f
o
r
'
B
io
d
eg
r
ad
ab
le'
,
wh
er
e
it
m
is
class
if
ied
7
5
in
s
tan
ce
s
as
'
No
n
-
b
io
d
eg
r
ad
a
b
le'
,
co
m
p
ar
e
d
to
1
7
in
s
tan
ce
s
o
f
'
B
io
d
eg
r
ad
ab
le'
m
is
clas
s
if
ied
as
'
No
n
-
b
io
d
eg
r
ad
a
b
le'
.
T
h
is
s
u
g
g
ests
th
e
m
o
d
el
m
ig
h
t
b
e
s
lig
h
tly
m
o
r
e
co
n
s
er
v
ativ
e
in
p
r
ed
ictin
g
item
s
as
'
No
n
-
b
io
d
eg
r
ad
a
b
le'
,
lead
in
g
t
o
a
f
ew
m
o
r
e
er
r
o
r
s
in
class
if
y
in
g
'
B
io
d
eg
r
ad
a
b
le'
i
tem
s
in
co
r
r
ec
tly
.
Ov
er
all,
th
e
m
o
d
el
ac
h
iev
es h
ig
h
a
cc
u
r
ac
y
in
d
is
tin
g
u
is
h
in
g
b
etwe
en
th
e
two
class
es.
Fig
u
r
e
1.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
Mo
b
ileNet
with
SVM
3
.
2
.
Rea
l
-
t
im
e
a
pp
lica
t
io
n
T
h
e
Mo
b
ileNet
m
o
d
el
with
SVM
class
if
ier
m
o
d
el
was
in
te
g
r
ated
in
t
o
a
we
b
ap
p
licatio
n
f
o
r
test
in
g
in
r
ea
l
tim
e
s
ce
n
ar
io
s
.
T
h
e
s
y
s
tem
was
d
esig
n
ed
to
wo
r
k
wit
h
b
ac
k
g
r
o
u
n
d
im
ag
es,
a
n
d
o
u
t
o
f
s
ev
e
r
al
p
o
s
s
ib
le
waste
o
b
jects,
ac
cu
r
ate
r
esu
lt
s
wer
e
n
o
ted
.
Fig
u
r
e
2
s
h
o
ws
s
o
m
e
o
f
th
e
test
ca
s
es
an
d
t
h
e
r
esp
o
n
s
es
o
f
th
e
s
y
s
tem
.
Fig
u
r
e
2
.
R
ea
l tim
e
test
ca
s
es
3
.
3
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
I
n
co
n
tr
ast,
R
ad
et
a
l.
[
2
6
]
en
h
an
ce
d
p
er
f
o
r
m
a
n
ce
b
y
in
co
r
p
o
r
atin
g
b
atch
n
o
r
m
aliza
tio
n
a
n
d
d
r
o
p
o
u
t
lay
er
s
,
ac
h
iev
in
g
a
n
8
7
%
ac
c
u
r
ac
y
ac
r
o
s
s
s
ix
waste
cla
s
s
e
s
.
W
h
ile
th
ese
s
tu
d
ies
u
n
d
er
s
co
r
e
C
NNs
'
p
o
ten
tial,
th
ey
also
r
ev
ea
l
lim
itatio
n
s
in
tr
ain
in
g
f
r
o
m
s
cr
atch
,
s
u
c
h
as
tim
e
co
n
s
u
m
p
ti
o
n
an
d
r
eso
u
r
ce
d
em
a
n
d
s
,
esp
ec
ially
f
o
r
v
ast
d
atasets
[
1
4
]
.
Yan
g
an
d
T
h
u
n
g
[
1
5
]
ac
h
iev
ed
8
3
%
ac
cu
r
ac
y
with
th
e
I
n
ce
p
tio
n
V3
m
o
d
el,
wh
ile
B
ian
co
et
a
l.
[
1
6
]
r
ep
o
r
ted
9
5
%
ac
cu
r
ac
y
u
s
in
g
th
e
R
esNet5
0
ar
ch
itectu
r
e.
T
h
e
s
e
r
esu
lts
illu
s
tr
ate
tr
an
s
f
er
lear
n
i
n
g
'
s
e
f
f
icien
cy
in
ca
p
t
u
r
in
g
in
tr
icate
waste
f
ea
tu
r
es.
R
ec
en
t
s
tu
d
ies
h
a
v
e
ex
ten
d
ed
t
h
ese
f
in
d
in
g
s
.
Fer
r
eir
a
et
a
l.
[
1
7
]
d
em
o
n
s
tr
ated
s
ig
n
i
f
ican
t
ac
cu
r
ac
y
im
p
r
o
v
e
m
en
ts
th
r
o
u
g
h
tr
an
s
f
er
lear
n
in
g
,
ac
h
iev
in
g
9
2
%
with
R
esNet5
0
.
R
ak
h
r
a
et
a
l.
[
1
8
]
f
u
r
th
e
r
v
alid
ated
M
o
b
ileNetV2
'
s
ef
f
ec
tiv
en
ess
in
r
ea
l
-
tim
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
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
E
n
h
a
n
ci
n
g
b
i
o
d
eg
r
a
d
a
b
le
w
a
s
te
ma
n
a
g
eme
n
t i
n
Ma
u
r
itiu
s
th
r
o
u
g
h
r
ea
l
-
time
…
(
Gee
r
is
h
S
u
d
d
u
l
)
1123
ap
p
licatio
n
s
,
ac
h
iev
in
g
8
9
%
ac
cu
r
ac
y
,
an
d
R
az
za
q
et
a
l.
[
1
9
]
h
ig
h
lig
h
ted
Den
s
eNe
t1
2
1
'
s
ca
p
ab
ilit
ies
with
9
3
% a
cc
u
r
ac
y
.
B
y
i
n
t
e
g
r
a
ti
n
g
K
N
N
a
n
d
S
VM
w
i
t
h
t
r
a
n
s
f
e
r
l
ea
r
n
i
n
g
m
o
d
e
l
s
,
w
e
h
a
v
e
a
c
h
i
e
v
e
d
h
i
g
h
e
r
a
c
c
u
r
a
c
y
l
e
v
e
l
s
t
h
a
n
t
h
e
m
e
n
ti
o
n
e
d
p
r
e
v
i
o
u
s
w
o
r
k
s
,
s
i
g
n
i
f
i
c
a
n
tl
y
s
u
r
p
a
s
s
i
n
g
t
r
a
d
i
t
i
o
n
a
l
C
N
N
b
e
n
c
h
m
a
r
k
s
.
A
d
d
it
i
o
n
a
l
l
y
,
t
h
e
i
n
c
l
u
s
i
o
n
o
f
r
ea
l
-
t
i
m
e
r
e
c
o
g
n
i
ti
o
n
c
a
p
a
b
i
l
it
i
es
d
is
t
i
n
g
u
is
h
e
s
o
u
r
m
e
t
h
o
d
f
r
o
m
m
a
n
y
e
x
i
s
tin
g
m
o
d
e
l
s
,
e
n
a
b
li
n
g
i
m
m
e
d
i
a
t
e
w
a
s
t
e
s
o
r
t
i
n
g
a
n
d
t
h
e
r
e
b
y
i
m
p
r
o
v
i
n
g
r
e
c
y
c
l
i
n
g
p
r
o
c
e
s
s
e
s
a
n
d
r
es
o
u
r
c
e
r
e
c
o
v
e
r
y
r
a
t
e
s
.
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
d
em
o
n
s
tr
ates
th
e
e
f
f
ec
tiv
en
ess
o
f
o
u
r
class
if
icatio
n
ap
p
r
o
ac
h
f
o
r
waste
m
an
ag
em
en
t.
T
h
e
p
r
e
-
tr
ain
ed
Mo
b
ileNet
co
m
b
i
n
ed
with
th
e
SVM
clas
s
if
ier
,
ac
h
iev
ed
a
n
ac
cu
r
ac
y
o
f
9
7
%
in
class
if
y
in
g
b
io
d
eg
r
a
d
ab
le
a
n
d
n
o
n
-
b
io
d
e
g
r
ad
ab
le
waste.
T
h
is
ap
p
r
o
ac
h
o
f
f
e
r
s
a
b
etter
s
o
l
u
tio
n
to
t
r
ad
itio
n
al
m
eth
o
d
s
,
en
h
an
cin
g
r
ec
y
clin
g
an
d
r
eso
u
r
ce
r
ec
o
v
er
y
.
W
h
ile
s
till
a
p
r
o
o
f
o
f
co
n
ce
p
t,
th
is
ap
p
r
o
ac
h
ca
n
b
e
in
teg
r
ated
in
a
s
y
s
tem
o
f
au
to
m
ati
c
waste
class
if
icatio
n
.
At
th
is
s
tag
e,
t
h
e
lim
itatio
n
r
em
ain
s
th
e
r
ela
tiv
ely
h
ig
h
tr
ain
in
g
ac
cu
r
ac
y
,
wh
ich
ca
n
b
e
r
eso
l
v
ed
with
m
o
r
e
im
a
g
es,
an
d
t
h
e
ch
allen
g
e
o
f
ca
p
tu
r
i
n
g
s
ev
er
al
waste
p
r
o
d
u
cts
f
r
o
m
a
v
id
e
o
s
tr
ea
m
o
f
im
a
g
es.
Ov
er
all,
th
e
r
esu
lts
h
ig
h
li
g
h
t
th
e
p
o
ten
tial
o
f
m
ac
h
in
e
lear
n
in
g
t
o
ad
d
r
ess
waste
m
an
ag
em
en
t
ch
allen
g
es
wh
ich
ca
n
b
e
ad
ap
te
d
f
o
r
Ma
u
r
itiu
s
an
d
o
t
h
er
d
e
v
elo
p
in
g
r
e
g
io
n
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Gee
r
is
h
Su
d
d
u
l
✓
✓
✓
✓
✓
✓
✓
✓
✓
Av
itah
B
ab
ajee
✓
✓
✓
✓
✓
✓
✓
✓
✓
Nu
n
d
jeet
R
am
b
ar
u
n
✓
✓
✓
✓
✓
✓
✓
✓
✓
San
d
h
y
a
Ar
m
o
o
g
u
m
✓
✓
✓
✓
✓
✓
✓
✓
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
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
u
s
ed
f
o
r
th
is
s
tu
d
y
is
av
ailab
le
f
r
o
m
R
ay
h
a
n
Z
a
m
za
m
y
,
o
n
Ka
g
g
le:
“No
n
-
an
d
-
b
io
d
eg
r
ad
a
b
le
waste
d
ataset”.
So
u
r
ce
:
h
t
tp
s
://www.
k
ag
g
le.
co
m
/d
atasets
/r
ay
h
an
za
m
za
m
y
/
n
o
n
-
a
n
d
-
b
i
o
d
eg
r
a
d
ab
le
-
waste
-
d
ataset
RE
F
E
R
E
NC
E
S
[
1
]
A
n
A
n
g
e
l
,
“
W
a
st
e
g
e
n
e
r
a
t
i
o
n
:
M
a
u
r
i
t
i
u
s
w
a
s
t
e
st
a
t
i
st
i
c
s,
”
A
n
An
g
e
l
N
e
w
sr
o
o
m
,
2
0
2
3
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
h
q
.
a
n
a
n
g
e
l
.
e
a
r
t
h
/
n
e
w
s
.
[
A
c
c
e
s
se
d
:
A
u
g
.
5
,
2
0
2
4
]
.
[
2
]
M
i
n
i
s
t
r
y
o
f
En
v
i
r
o
n
me
n
t
,
“
S
o
l
i
d
w
a
st
e
ma
n
a
g
e
m
e
n
t
st
r
a
t
e
g
y
,
”
G
o
v
e
r
n
m
e
n
t
o
f
M
a
u
ri
t
i
u
s
,
2
0
2
3
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
e
n
v
i
r
o
n
me
n
t
.
g
o
v
m
u
.
o
r
g
.
[
A
c
c
e
sse
d
:
A
u
g
.
5
,
2
0
2
4
]
.
[
3
]
A
g
e
n
c
e
F
r
a
n
ç
a
i
se
d
e
D
é
v
e
l
o
p
p
e
me
n
t
(
A
F
D
)
,
“
S
u
p
p
o
r
t
i
n
g
s
o
l
i
d
w
a
s
t
e
ma
n
a
g
e
m
e
n
t
i
n
M
a
u
r
i
t
i
u
s
,
”
AFD
,
2
0
2
3
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
a
f
d
.
f
r
/
e
n
.
[
A
c
c
e
sse
d
:
A
u
g
.
5
,
2
0
2
4
]
.
[
4
]
S
t
a
t
i
st
i
c
s
M
a
u
r
i
t
i
u
s
,
“
En
v
i
r
o
n
me
n
t
st
a
t
i
s
t
i
c
s
–
Y
e
a
r
2
0
2
0
,
”
G
o
v
e
rn
m
e
n
t
o
f
M
a
u
r
i
t
i
u
s
,
2
0
2
1
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
s
t
a
t
sma
u
r
i
t
i
u
s.
g
o
v
m
u
.
o
r
g
/
P
a
g
e
s
/
S
t
a
t
i
s
t
i
c
s/
ESI
/
En
v
i
r
o
n
me
n
t
/
E
n
v
_
Y
r
2
0
.
a
sp
x
.
[
A
c
c
e
sse
d
:
A
u
g
.
5
,
2
0
2
4
]
.
[
5
]
S
u
s
t
a
i
n
a
b
i
l
i
t
y
,
“
S
o
l
i
d
w
a
st
e
g
e
n
e
r
a
t
i
o
n
a
n
d
d
i
sp
o
sa
l
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s:
A
s
u
r
v
e
y
o
f
s
o
l
u
t
i
o
n
s
a
n
d
c
h
a
l
l
e
n
g
e
s
,
”
S
u
s
t
a
i
n
a
b
i
l
i
t
y
,
2
0
2
2
.
[
6
]
P
.
K
o
w
l
e
ss
e
r
,
“
A
n
O
v
e
r
v
i
e
w
o
f
C
i
r
c
u
l
a
r
Ec
o
n
o
m
y
i
n
M
a
u
r
i
t
i
u
s,
”
i
n
C
i
rc
u
l
a
r
Ec
o
n
o
m
y
:
G
l
o
b
a
l
P
e
rsp
e
c
t
i
v
e
,
S
.
G
h
o
s
h
,
Ed
.
S
i
n
g
a
p
o
r
e
:
S
p
r
i
n
g
e
r
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
15
-
1
0
5
2
-
6
_
1
4
[
7
]
M
.
A
.
M
o
h
a
mm
e
d
e
t
a
l
,
“
A
u
t
o
m
a
t
e
d
w
a
st
e
-
s
o
r
t
i
n
g
a
n
d
r
e
c
y
c
l
i
n
g
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
a
n
d
f
e
a
t
u
r
e
s
f
u
s
i
o
n
:
A
d
i
g
i
t
a
l
-
e
n
a
b
l
e
d
c
i
r
c
u
l
a
r
e
c
o
n
o
m
y
v
i
si
o
n
f
o
r
smar
t
c
i
t
i
e
s
,
”
S
p
ri
n
g
e
r
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
0
4
2
-
0
2
1
-
1
1
5
3
7
-
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
:
1
1
1
9
-
1
1
2
5
1124
[
8
]
A
.
G
.
M
u
k
h
e
r
j
e
e
,
U
.
R
.
W
a
n
j
a
r
i
,
R
.
C
h
a
k
r
a
b
o
r
t
y
,
K
.
R
e
n
u
,
B
.
V
e
l
l
i
n
g
i
r
i
,
A
.
G
e
o
r
g
e
,
C
.
R
.
S
.
R
a
j
a
n
,
a
n
d
A
.
V
.
G
o
p
a
l
a
k
r
i
s
h
n
a
n
,
“
A
r
e
v
i
e
w
o
n
m
o
d
e
r
n
a
n
d
sm
a
r
t
t
e
c
h
n
o
l
o
g
i
e
s
f
o
r
e
f
f
i
c
i
e
n
t
w
a
s
t
e
d
i
s
p
o
s
a
l
a
n
d
ma
n
a
g
e
me
n
t
,
”
J
o
u
r
n
a
l
o
f
E
n
v
i
ro
n
m
e
n
t
a
l
Ma
n
a
g
e
m
e
n
t
,
v
o
l
.
2
9
7
,
2
0
2
1
,
A
r
t
.
n
o
.
1
1
3
3
4
7
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
e
n
v
m
a
n
.
2
0
2
1
.
1
1
3
3
4
7
.
[
9
]
M
.
A
.
R
a
d
,
M
.
A
b
d
e
r
r
a
h
i
m,
a
n
d
C
.
K
a
n
e
l
l
a
k
i
s,
“
S
m
a
r
t
w
a
st
e
m
a
n
a
g
e
me
n
t
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
:
C
h
a
l
l
e
n
g
e
s
a
n
d
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
s
,
”
S
u
st
a
i
n
a
b
i
l
i
t
y
,
v
o
l
.
1
2
,
n
o
.
1
5
,
p
.
6
1
4
2
,
2
0
2
0
.
[
1
0
]
Y
.
Ze
n
g
,
M
.
Q
i
a
o
,
M
.
Li
,
a
n
d
S
.
W
a
n
g
,
“
D
e
e
p
l
e
a
r
n
i
n
g
i
n
w
a
st
e
c
l
a
ssi
f
i
c
a
t
i
o
n
:
A
c
o
m
p
r
e
h
e
n
si
v
e
s
u
r
v
e
y
,
”
J
o
u
rn
a
l
o
f
En
v
i
r
o
n
m
e
n
t
a
l
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
2
5
6
,
2
0
2
0
,
A
r
t
.
n
o
.
1
0
9
9
5
3
.
[
1
1
]
B
.
Ji
a
n
g
,
J
.
W
a
n
g
,
J.
F
a
n
,
J.
L
u
,
a
n
d
Y
.
L
u
o
,
“
A
u
t
o
m
a
t
i
c
w
a
s
t
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
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
,
”
Wa
s
t
e
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
1
0
8
,
p
p
.
1
3
6
–
1
4
4
,
2
0
2
0
.
[
1
2
]
V
.
P
.
B
r
i
n
t
h
a
,
R
.
R
e
k
h
a
,
J.
N
a
n
d
h
i
n
i
,
N
.
S
r
e
e
k
a
a
r
t
h
i
c
k
,
B
.
I
sh
w
a
r
y
a
a
,
a
n
d
R
.
R
a
h
u
l
,
“
A
u
t
o
ma
t
i
c
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
so
l
i
d
w
a
st
e
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
S
p
r
i
n
g
e
r
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
030
-
2
4
0
5
1
-
6
_
8
3
.
[
1
3
]
A
.
A
i
s
h
w
a
r
y
a
,
P
.
W
a
d
h
w
a
,
O
.
O
w
a
i
s,
a
n
d
V
.
V
a
sh
i
s
h
t
,
“
Ef
f
i
c
i
e
n
t
d
i
f
f
e
r
e
n
t
i
a
t
i
o
n
o
f
b
i
o
d
e
g
r
a
d
a
b
l
e
a
n
d
n
o
n
-
b
i
o
d
e
g
r
a
d
a
b
l
e
mu
n
i
c
i
p
a
l
w
a
st
e
u
si
n
g
a
n
o
v
e
l
M
o
b
i
l
e
Y
O
LO
a
l
g
o
r
i
t
h
m,”
I
n
t
e
r
n
a
t
i
o
n
a
l
I
n
f
o
rm
a
t
i
o
n
a
n
d
E
n
g
i
n
e
e
ri
n
g
T
e
c
h
n
o
l
o
g
y
Ass
o
c
i
a
t
i
o
n
,
2
0
2
1
.
[
1
4
]
W
.
R
a
w
a
t
a
n
d
Z
.
W
a
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
f
o
r
i
m
a
g
e
c
l
a
ssi
f
i
c
a
t
i
o
n
:
A
c
o
m
p
r
e
h
e
n
s
i
v
e
r
e
v
i
e
w
,
”
N
e
u
ra
l
C
o
m
p
u
t
a
t
i
o
n
,
v
o
l
.
2
9
,
n
o
.
9
,
p
p
.
2
3
5
2
–
2
4
4
9
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
6
2
/
n
e
c
o
_
a
_
0
0
9
9
0
.
[
1
5
]
M
.
Y
a
n
g
a
n
d
G
.
T
h
u
n
g
,
“
Tr
a
n
sf
e
r
l
e
a
r
n
i
n
g
f
o
r
w
a
st
e
c
l
a
s
si
f
i
c
a
t
i
o
n
u
s
i
n
g
t
h
e
Tr
a
s
h
N
e
t
d
a
t
a
se
t
,
”
S
t
a
n
f
o
r
d
U
n
i
v
e
r
si
t
y
,
2
0
1
6
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
c
s
2
2
9
.
s
t
a
n
f
o
r
d
.
e
d
u
/
p
r
o
j
2
0
1
6
/
r
e
p
o
r
t
/
Th
u
n
g
Y
a
n
g
-
C
l
a
ss
i
f
i
c
a
t
i
o
n
O
f
Tr
a
s
h
F
o
r
R
e
c
y
c
l
a
b
i
l
i
t
y
S
t
a
t
u
s
-
r
e
p
o
r
t
.
p
d
f
.
[
A
c
c
e
sse
d
:
A
u
g
.
5
,
2
0
2
4
]
.
[
1
6
]
S
.
B
i
a
n
c
o
,
R
.
C
a
d
e
n
e
,
L
.
C
e
l
o
n
a
,
a
n
d
P
.
N
a
p
o
l
e
t
a
n
o
,
“
B
e
n
c
h
m
a
r
k
a
n
a
l
y
s
i
s
o
f
r
e
p
r
e
s
e
n
t
a
t
i
v
e
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
a
r
c
h
i
t
e
c
t
u
r
e
s
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
6
,
p
p
.
6
4
2
7
0
–
6
4
2
7
7
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
8
.
2
8
7
7
8
9
0
.
[
1
7
]
C
.
F
e
r
r
e
i
r
a
,
J.
M
.
d
e
A
l
me
i
d
a
,
S
.
C
.
O
l
i
v
e
i
r
a
,
a
n
d
T
.
R
e
d
d
y
,
“
W
a
st
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
w
i
t
h
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
:
A
n
e
v
a
l
u
a
t
i
o
n
o
f
p
re
-
t
r
a
i
n
e
d
m
o
d
e
l
s,”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
7
7
,
p
p
.
3
2
4
–
3
3
1
,
2
0
2
0
.
[
1
8
]
M
.
R
a
k
h
r
a
,
A
.
C
h
a
u
h
a
n
,
N
.
K
u
ma
r
,
a
n
d
S
.
S
h
r
i
v
a
s
t
a
v
a
,
“
Tr
a
n
sf
e
r
l
e
a
r
n
i
n
g
f
o
r
w
a
s
t
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
M
o
b
i
l
e
N
e
t
V
2
,
”
Ma
t
e
r
i
a
l
s T
o
d
a
y
:
Pr
o
c
e
e
d
i
n
g
s
,
v
o
l
.
4
5
,
p
p
.
4
2
5
8
–
4
2
6
3
,
2
0
2
1
.
[
1
9
]
M
.
R
a
z
z
a
q
,
M
.
Y
.
J
a
v
e
d
,
a
n
d
M
.
F
.
M
u
sh
t
a
q
,
“
C
l
a
ssi
f
y
i
n
g
w
a
s
t
e
u
si
n
g
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
w
i
t
h
D
e
n
s
e
N
e
t
a
r
c
h
i
t
e
c
t
u
r
e
,
”
p
r
e
se
n
t
e
d
a
t
I
EE
E
C
o
n
f
e
re
n
c
e
(
s
p
e
c
i
f
i
c
e
v
e
n
t
n
o
t
st
a
t
e
d
)
,
2
0
2
2
.
[
2
0
]
V
.
V
i
j
a
y
a
k
u
mar,
R
.
T
h
i
r
u
n
a
v
u
k
a
r
a
su
,
a
n
d
S
.
B
a
l
a
,
“
Ef
f
i
c
i
e
n
t
N
e
t
f
o
r
w
a
st
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
si
n
g
t
h
e
W
a
st
e
N
e
t
d
a
t
a
se
t
,
”
J
o
u
r
n
a
l
o
f
C
l
e
a
n
e
r P
r
o
d
u
c
t
i
o
n
,
v
o
l
.
3
4
0
,
A
r
t
.
n
o
.
1
3
0
7
4
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
c
l
e
p
r
o
.
2
0
2
2
.
1
3
0
7
4
9
.
[
2
1
]
R
a
y
h
a
n
Za
m
z
a
m
y
,
“
N
o
n
-
a
n
d
-
b
i
o
d
e
g
r
a
d
a
b
l
e
w
a
st
e
d
a
t
a
s
e
t
,
”
K
a
g
g
l
e
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
k
a
g
g
l
e
.
c
o
m
/
d
a
t
a
se
t
s/
r
a
y
h
a
n
z
a
mz
a
m
y
/
n
o
n
-
a
n
d
-
b
i
o
d
e
g
r
a
d
a
b
l
e
-
w
a
st
e
-
d
a
t
a
se
t
.
[
A
c
c
e
s
se
d
:
A
u
g
.
5
,
2
0
2
4
]
.
[
2
2
]
L.
P
e
t
e
r
s
o
n
,
“
K
-
n
e
a
r
e
st
n
e
i
g
h
b
o
r
,
”
S
c
h
o
l
a
rp
e
d
i
a
,
v
o
l
.
4
,
p
.
1
8
8
3
,
2
0
0
9
,
d
o
i
:
1
0
.
4
2
4
9
/
s
c
h
o
l
a
r
p
e
d
i
a
.
1
8
8
3
.
[
2
3
]
C
.
C
o
r
t
e
s
a
n
d
V
.
V
a
p
n
i
k
,
“
S
u
p
p
o
r
t
-
v
e
c
t
o
r
n
e
t
w
o
r
k
s,
”
Ma
c
h
i
n
e
L
e
a
rn
i
n
g
,
v
o
l
.
2
0
,
n
o
.
3
,
p
p
.
2
7
3
–
2
9
7
,
1
9
9
5
,
d
o
i
:
1
0
.
1
0
0
7
/
B
F
0
0
9
9
4
0
1
8
.
[
2
4
]
S
.
A
g
g
a
r
w
a
l
,
M
.
K
.
S
h
a
r
ma,
H
.
O
.
K
.
J
h
a
,
A
.
P
.
S
i
n
g
h
,
R
.
S
h
a
r
m
a
,
a
n
d
G
.
M
i
t
r
a
,
“
C
o
m
p
a
r
i
n
g
p
r
e
-
t
r
a
i
n
e
d
a
l
g
o
r
i
t
h
ms
f
o
r
f
a
c
i
a
l
e
mo
t
i
o
n
r
e
c
o
g
n
i
t
i
o
n
:
A
n
a
n
a
l
y
si
s
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
5
,
n
o
.
1
,
p
p
.
5
4
–
6
0
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
5
4
5
/
2
6
6
3
3
5
8
2
.
2
0
2
3
.
v
5
.
i
1
a
.
9
0
.
[
2
5
]
F
.
M
.
P
i
n
t
o
,
G
.
A
.
M
e
l
o
,
L.
M
.
H
o
n
ó
r
i
o
,
A
.
L.
M
.
M
a
r
c
a
t
o
,
A
.
G
.
S
.
C
o
n
c
e
i
ç
ã
o
,
a
n
d
A
.
O
.
Ti
mo
t
h
e
o
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
p
p
l
i
e
d
t
o
v
e
g
e
t
a
t
i
o
n
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
n
d
r
e
m
o
v
a
l
u
si
n
g
mu
l
t
i
d
i
me
n
s
i
o
n
a
l
a
e
r
i
a
l
d
a
t
a
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
0
,
n
o
.
2
1
,
p
.
6
1
8
7
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
0
2
1
6
1
8
7
.
[
2
6
]
M
.
A
.
R
a
d
,
M
.
A
b
d
e
r
r
a
h
i
m,
a
n
d
C
.
K
a
n
e
l
l
a
k
i
s
,
“
C
N
N
-
b
a
se
d
w
a
s
t
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
w
i
t
h
b
a
t
c
h
n
o
r
ma
l
i
z
a
t
i
o
n
a
n
d
d
r
o
p
o
u
t
,
”
p
r
e
se
n
t
e
d
a
t
I
EE
E
C
o
n
f
e
re
n
c
e
(
s
p
e
c
i
f
i
c
e
v
e
n
t
n
o
t
st
a
t
e
d
)
,
2
0
1
7
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
G
e
e
r
ish
S
u
d
d
u
l
re
c
e
iv
e
d
h
is P
h
.
D.
fr
o
m
t
h
e
Un
i
v
e
rsity
o
f
Tec
h
n
o
lo
g
y
,
M
a
u
ri
ti
u
s
(UTM
).
He
is
c
u
rre
n
tl
y
a
S
e
n
i
o
r
Lec
tu
re
r
a
t
th
e
UTM
,
in
th
e
De
p
a
rtme
n
t
o
f
Bu
sin
e
ss
In
fo
rm
a
ti
c
s
a
n
d
S
o
ftwa
re
En
g
i
n
e
e
rin
g
u
n
d
e
r
t
h
e
S
c
h
o
o
l
o
f
I
n
n
o
v
a
t
iv
e
Tec
h
n
o
lo
g
ies
a
n
d
En
g
i
n
e
e
rin
g
.
He
h
a
s
b
e
e
n
a
c
ti
v
e
l
y
in
v
o
l
v
e
d
i
n
re
se
a
r
c
h
a
n
d
tea
c
h
i
n
g
si
n
c
e
2
0
0
5
,
a
n
d
c
u
rre
n
tl
y
h
is
re
se
a
rc
h
wo
rk
fo
c
u
se
s
o
n
d
if
fe
re
n
t
a
sp
e
c
ts
o
f
m
a
c
h
in
e
lea
rn
in
g
su
c
h
a
s
c
o
m
p
u
ter
v
isi
o
n
a
n
d
n
a
t
u
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
g
.
su
d
d
u
l@
u
tm.a
c
.
m
u
.
Av
ita
h
Ba
b
a
je
e
h
a
s
c
o
m
p
let
e
d
a
M
a
ste
r'
s
d
e
g
re
e
in
Artifi
c
ial
In
telli
g
e
n
c
e
wit
h
M
a
c
h
in
e
Lea
rn
in
g
fro
m
S
c
h
o
o
l
o
f
In
n
o
v
a
ti
v
e
Tec
h
n
o
lo
g
ies
a
n
d
En
g
i
n
e
e
rin
g
,
Un
i
v
e
rsity
o
f
Tec
h
n
o
l
o
g
y
La
T
o
u
r
Ko
e
n
i
g
,
P
o
in
te
-
a
u
x
-
S
a
b
les
,
Re
p
u
b
li
c
o
f
M
a
u
rit
iu
s
.
S
h
e
h
a
s
wo
r
k
e
d
o
n
se
v
e
ra
l
m
a
c
h
in
e
le
a
rn
i
n
g
p
ro
jec
t
s,
with
a
fo
c
u
s
o
n
d
e
e
p
lea
rn
i
n
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
v
it
a
h
.
b
a
b
a
jee
@u
m
a
il
.
u
tm.a
c
.
m
u
.
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
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
E
n
h
a
n
ci
n
g
b
i
o
d
eg
r
a
d
a
b
le
w
a
s
te
ma
n
a
g
eme
n
t i
n
Ma
u
r
itiu
s
th
r
o
u
g
h
r
ea
l
-
time
…
(
Gee
r
is
h
S
u
d
d
u
l
)
1125
Nun
d
je
e
t
Ra
m
b
a
r
u
n
h
a
s
c
o
m
p
lete
d
a
M
a
ste
r'
s
d
e
g
re
e
in
Art
ifi
c
ial
In
telli
g
e
n
c
e
with
M
a
c
h
in
e
Lea
rn
i
n
g
fr
o
m
S
c
h
o
o
l
o
f
In
n
o
v
a
ti
v
e
Tec
h
n
o
lo
g
ies
a
n
d
E
n
g
i
n
e
e
rin
g
,
Un
i
v
e
rsity
o
f
Tec
h
n
o
lo
g
y
La
To
u
r
Ko
e
n
i
g
,
P
o
in
te
-
a
u
x
-
S
a
b
les
,
Re
p
u
b
li
c
o
f
M
a
u
rit
i
u
s
.
He
h
a
s
wo
rk
e
d
o
n
se
v
e
ra
l
m
a
c
h
in
e
lea
rn
i
n
g
p
ro
jec
ts
,
wit
h
a
fo
c
u
s o
n
d
e
e
p
lea
rn
i
n
g
.
H
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
u
n
d
jee
t.
ra
m
b
a
ru
n
@u
m
a
il
.
u
tm.a
c
.
m
u
.
S
a
n
d
h
y
a
Ar
m
o
o
g
u
m
re
c
e
iv
e
d
h
e
r
P
h
.
D.
fr
o
m
th
e
Un
i
v
e
rsit
y
o
f
Tec
h
n
o
l
o
g
y
,
M
a
u
rit
i
u
s
(UTM
)
.
S
h
e
is
c
u
rre
n
t
ly
a
n
As
so
c
iate
P
r
o
fe
ss
o
r
a
t
t
h
e
UTM
,
i
n
th
e
De
p
a
rtme
n
t
o
f
In
d
u
strial
S
y
ste
m
s
E
n
g
i
n
e
e
rin
g
u
n
d
e
r
th
e
S
c
h
o
o
l
o
f
I
n
n
o
v
a
ti
v
e
Te
c
h
n
o
l
o
g
ies
a
n
d
E
n
g
i
n
e
e
rin
g
.
S
h
e
h
a
s
b
e
e
n
a
c
ti
v
e
ly
in
v
o
l
v
e
d
in
re
se
a
rc
h
a
n
d
tea
c
h
in
g
sin
c
e
2
0
0
3
,
a
n
d
c
u
rre
n
tl
y
h
e
r
re
se
a
rc
h
wo
rk
f
o
c
u
se
s
o
n
d
iffere
n
t
a
s
p
e
c
ts
o
f
m
a
c
h
in
e
lea
rn
in
g
,
c
y
b
e
rse
c
u
rit
y
a
n
d
b
i
g
d
a
ta
a
n
a
ly
ti
c
s.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sa
n
d
h
y
a
.
a
rm
o
o
g
u
m
@u
tm.a
c
.
m
u
.
Evaluation Warning : The document was created with Spire.PDF for Python.