I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
, pp.
37
1
5
~
3723
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
5
.pp
3715
-
3723
3715
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
C
om
p
ar
at
i
ve
an
al
ysi
s of
c
on
vol
u
t
i
o
n
al
n
e
u
r
al
n
e
t
w
or
k
ar
c
h
i
t
e
c
t
u
r
e
s f
o
r
p
ou
l
t
r
y m
e
at
c
l
ass
i
f
i
c
at
i
on
S
e
k
h
r
a S
al
m
a
1
, M
oh
am
m
e
d
H
ab
ib
1
, A
d
il
T
an
n
ou
c
h
e
2
, Y
ou
s
s
e
f
O
u
n
e
j
j
ar
1
1
L
a
bo
r
a
t
o
r
y
o
f
S
pe
c
t
r
o
m
e
t
r
y, M
a
t
e
r
i
a
l
s
a
n
d
A
r
c
he
o
m
a
t
e
r
i
a
l
s
“
L
A
S
M
A
R
”
,
F
a
c
ul
t
y o
f
S
c
i
e
nc
e
s
,
M
ou
l
a
y
I
s
m
a
i
l
U
n
i
ve
r
s
i
t
y
, M
e
kne
s
, M
o
r
oc
c
o
2
L
a
bor
a
t
oi
r
e
de
l
'
I
ngé
ni
e
r
i
e
e
t
de
T
e
c
hnol
ogi
e
s
A
ppl
i
qué
e
s
“
L
I
T
A
”
,
E
c
ol
e
S
upé
r
i
e
ur
e
de
T
e
c
hnol
ogi
e
de
B
é
ni
M
e
l
l
a
l
,
U
ni
ve
r
s
i
t
é
S
ul
t
a
n M
oul
a
y S
l
i
m
a
ne
, B
e
ni
M
e
l
l
a
l
, M
or
oc
c
o
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
M
a
r
14
,
2024
R
e
vi
s
e
d
J
ul
6
,
2025
A
c
c
e
pt
e
d
A
ug
6
,
2025
The
increasing
demand
for
standardized
food
quality
assurance,
parti
cularly
in
regions
like
Morocco,
emphasizes
the
need
for
accurate
classificat
ion
of
poultry
meat.
This
study
evaluates
and
compares
ten
convolut
ional
neural
network
(CNN)
architectures
—
VGG19,
VGG16,
ResNet50,
GoogleNet,
MobileNetV1,
MobileNetV2,
DenseNet,
NasNet,
EfficientNet,
and
AlexNet
—
for
classifying
commonly
consumed
poultry
meat
ty
pes
in
Moroccan
markets,
including
chicken,
turkey,
fayoumi,
and
farmer’s
chicken.
A
labeled
image
datase
t
was
used
to
train
and
test
each
mode
l,
with
performance
assessed
using
metrics
such
as
accuracy,
precision,
recall,
training
time,
and
computationa
l
complexity.
Additionally,
the
study
investigates
how
dataset
size
influence
s
model
performa
nce,
addr
essing
challenges
like
limit
ed
data
availabil
ity
and
scalabil
ity.
The
results
hi
ghligh
t
DenseNet
as
the
top
-
performing
architecture,
achieving
98%
classif
ication
accuracy
while
also
demonst
rating
superior
computat
ional
efficiency.
These
findings
are
valuable
fo
r
improving
food
quality
control,
offering
data
-
driven
support
for
stakeholders
in
poultry
production,
distributio
n,
and
regulatory
bodies.
By
identifyi
ng
optimal
deep
learning
models
for
poultry
meat
classification,
the
study
contributes
to
enhancing
food
authent
ication
and
safety
in
Morocco
and
similar
regions.
It
also
encourages
the
inte
gration
of
AI
-
driven
systems
in
food
inspection
processes,
providing
sc
alable,
accurate,
and
efficient
soluti
ons
for
ensuring
standardi
zed
quality
in
the
poultry supply c
hain.
K
e
y
w
o
r
d
s
:
C
onvolut
io
na
l
ne
ur
a
l
ne
twor
ks
D
e
e
p l
e
a
r
ni
ng
F
ood qua
li
ty
a
s
s
e
s
s
m
e
nt
I
m
a
ge
a
na
ly
s
is
M
e
a
t
a
ut
he
nt
ic
a
ti
on
P
oul
tr
y m
e
a
t
c
la
s
s
if
ic
a
ti
on
T
r
a
ns
f
e
r
l
e
a
r
ni
ng
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
S
e
khr
a
S
a
lm
a
L
a
bor
a
to
r
y of
S
pe
c
tr
om
e
t
r
y, M
a
te
r
ia
ls
a
nd A
r
c
he
om
a
te
r
ia
ls
“
L
A
S
M
A
R
”
, F
a
c
ul
ty
of
S
c
ie
nc
e
s
M
oul
a
y I
s
m
a
il
U
ni
ve
r
s
it
y
M
e
kne
s
50000,
M
or
oc
c
o
E
m
a
il
:
s
e
khr
a
s
a
lm
a
3@
gm
a
il
.c
om
1.
I
N
T
R
O
D
U
C
T
I
O
N
P
oul
tr
y m
e
a
t
c
ons
ti
tu
te
s
a
c
or
ne
r
s
to
ne
of
gl
oba
l
f
ood s
e
c
ur
it
y,
pr
ovi
di
ng a
r
ic
h
s
our
c
e
of
pr
ot
e
in
a
nd
e
s
s
e
nt
ia
l
nut
r
ie
nt
s
.
I
n
M
or
oc
c
o,
poul
tr
y
pr
oduc
ti
on
is
pa
r
ti
c
ul
a
r
ly
vi
ta
l,
c
ont
r
ib
ut
in
g
s
ig
ni
f
ic
a
nt
ly
to
bot
h
na
ti
ona
l
di
e
ta
r
y
ne
e
ds
a
nd
th
e
a
gr
ic
ul
tu
r
a
l
e
c
onomy.
A
s
c
ons
u
m
e
r
de
m
a
nd
f
or
hi
gh
-
qua
li
ty
poul
tr
y
pr
oduc
ts
in
te
ns
if
ie
s
,
a
c
c
ur
a
te
a
nd
e
f
f
ic
ie
nt
c
la
s
s
if
ic
a
ti
on
of
poul
tr
y
m
e
a
t
be
c
om
e
s
pa
r
a
m
ount
f
or
m
a
in
ta
in
in
g
c
ons
um
e
r
tr
us
t,
e
ns
ur
in
g
f
ood
s
a
f
e
ty
,
a
nd
s
uppor
ti
ng
f
a
ir
m
a
r
k
e
t
pr
a
c
ti
c
e
s
.
T
hi
s
ne
c
e
s
s
it
a
te
s
th
e
a
dopt
io
n
of
a
dva
nc
e
d
te
c
hnol
ogi
e
s
c
a
pa
bl
e
of
a
ut
om
a
ti
ng
a
nd
s
ta
nda
r
di
z
in
g
th
e
c
la
s
s
if
ic
a
ti
on
pr
oc
e
s
s
.
C
onvolut
io
na
l
ne
ur
a
l
ne
twor
ks
(
C
N
N
s
)
,
a
hi
ghl
y
e
f
f
e
c
ti
ve
c
a
te
gor
y
of
de
e
p
le
a
r
ni
ng
m
ode
ls
s
pe
c
ia
li
z
e
d
in
im
a
ge
pr
oc
e
s
s
in
g,
h
a
ve
pr
ove
n
to
be
a
c
om
pe
ll
in
g
a
ppr
oa
c
h
to
a
ddr
e
s
s
in
g
th
i
s
c
h
a
ll
e
nge
.
T
he
ir
a
bi
li
ty
to
di
s
c
e
r
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3715
-
3723
3716
in
tr
ic
a
te
pa
tt
e
r
ns
a
nd
f
e
a
tu
r
e
s
f
r
om
i
m
a
ge
da
ta
w
it
h
m
in
im
a
l
pr
e
-
pr
oc
e
s
s
in
g
m
a
ke
s
th
e
m
id
e
a
ll
y
s
ui
te
d
f
or
c
om
pl
e
x
c
la
s
s
if
ic
a
ti
on
ta
s
ks
in
th
e
f
ood
in
dus
tr
y.
T
he
a
ppl
ic
a
ti
on
of
C
N
N
s
in
f
ood
c
la
s
s
if
ic
a
ti
on
ha
s
ga
in
e
d
c
ons
id
e
r
a
bl
e
tr
a
c
ti
on
in
r
e
c
e
nt
y
e
a
r
s
,
w
it
h
s
uc
c
e
s
s
f
ul
im
pl
e
m
e
nt
a
ti
ons
in
a
r
e
a
s
s
uc
h
a
s
f
r
ui
t
a
nd
ve
ge
t
a
bl
e
s
or
ti
ng,
m
e
a
t
qua
li
ty
a
s
s
e
s
s
m
e
nt
,
a
nd
s
pe
c
ie
s
id
e
nt
if
ic
a
ti
on.
S
e
ve
r
a
l
s
tu
di
e
s
ha
v
e
e
xpl
or
e
d
th
e
pot
e
nt
ia
l
of
C
N
N
s
f
or
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on
s
pe
c
if
ic
a
ll
y.
F
or
e
xa
m
pl
e
,
our
pr
e
vi
ous
w
o
r
k
[
1
]
de
m
ons
tr
a
te
d
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
th
e
M
obi
le
N
e
tV2
pr
e
-
tr
a
in
e
d
m
ode
l
f
or
th
is
ta
s
k.
O
th
e
r
r
e
s
e
a
r
c
h
in
[
2]
ha
s
e
m
pl
oy
e
d
C
N
N
s
to
c
la
s
s
if
y
c
hi
c
ke
n
br
e
a
s
ts
ba
s
e
d
on
vi
s
ua
l
c
ha
r
a
c
te
r
is
ti
c
s
li
ke
c
ol
or
,
te
xt
ur
e
,
a
nd
f
a
t
di
s
tr
ib
ut
io
n,
w
hi
le
f
ur
th
e
r
s
tu
di
e
s
in
[
3]
ha
v
e
in
ve
s
ti
ga
te
d
th
e
ir
u
s
e
in
a
ut
he
nt
ic
a
ti
ng
poul
tr
y
m
e
a
t
pr
oduc
ts
to
de
t
e
c
t
a
dul
te
r
a
ti
on
a
nd
e
ns
ur
e
pr
oduc
t
in
te
gr
it
y.
D
e
s
pi
te
th
e
s
e
a
dva
nc
e
m
e
nt
s
,
a
c
om
pr
e
he
ns
iv
e
c
om
pa
r
a
ti
ve
a
na
ly
s
is
of
va
r
io
us
C
N
N
a
r
c
hi
te
c
tu
r
e
s
ta
il
or
e
d
to
th
e
s
pe
c
if
ic
c
ont
e
xt
of
M
or
oc
c
a
n
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on
r
e
m
a
in
s
la
c
ki
ng.
S
uc
h
a
n
a
na
ly
s
is
is
c
r
uc
ia
l
f
or
unde
r
s
ta
ndi
ng
th
e
s
tr
e
ngt
hs
a
nd
li
m
it
a
ti
ons
of
di
f
f
e
r
e
nt
m
ode
ls
a
nd
f
o
r
de
ve
lo
pi
ng
r
obus
t
c
la
s
s
if
ic
a
ti
on
s
ys
te
m
s
th
a
t
a
c
c
om
m
oda
te
r
e
gi
ona
l
va
r
ia
ti
ons
in
poul
tr
y
br
e
e
d
s
,
pr
oc
e
s
s
in
g
m
e
th
ods
, a
nd c
ons
um
e
r
pr
e
f
e
r
e
nc
e
s
.
T
hi
s
s
tu
dy
a
ddr
e
s
s
e
s
th
is
g
a
p
in
our
r
e
s
e
a
r
c
h
te
a
m
[
4]
–
[
7]
by
pr
e
s
e
nt
in
g
a
c
om
pa
r
a
ti
ve
a
n
a
ly
s
is
of
te
n
di
s
ti
nc
t
C
N
N
a
r
c
hi
te
c
tu
r
e
s
f
or
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
o
n
in
M
or
oc
c
o.
U
ti
li
z
in
g
a
di
ve
r
s
e
da
ta
s
e
t
of
poul
tr
y
m
e
a
t
im
a
ge
s
r
e
pr
e
s
e
nt
a
ti
ve
of
th
e
M
or
oc
c
a
n
m
a
r
ke
t,
w
e
e
va
lu
a
te
th
e
pe
r
f
or
m
a
nc
e
of
th
e
s
e
a
r
c
hi
te
c
tu
r
e
s
in
a
c
c
ur
a
te
ly
di
s
ti
ngui
s
hi
ng
b
e
twe
e
n
di
f
f
e
r
e
nt
poul
tr
y
ty
pe
s
a
nd
c
ut
s
.
T
hr
ough
r
ig
or
ous
te
s
ti
ng
a
nd
c
om
pa
r
is
on
of
ke
y
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
,
w
e
a
im
to
id
e
nt
if
y
th
e
opt
im
a
l
C
N
N
a
r
c
hi
te
c
tu
r
e
s
f
or
m
a
xi
m
iz
in
g
c
la
s
s
if
ic
a
ti
on
a
c
c
ur
a
c
y
a
nd
e
f
f
ic
ie
nc
y
in
th
e
M
or
oc
c
a
n
poul
tr
y
in
dus
tr
y.
T
he
f
in
di
ngs
o
f
th
is
r
e
s
e
a
r
c
h
a
id
in
th
e
pr
ogr
e
s
s
io
n
of
a
dva
nc
e
d
to
ol
s
f
or
qua
li
ty
c
ont
r
ol
,
tr
a
c
e
a
bi
li
ty
,
a
nd
c
ons
um
e
r
pr
ot
e
c
ti
on
w
it
hi
n t
he
M
or
oc
c
a
n poult
r
y s
e
c
to
r
, w
it
h br
oa
de
r
i
m
pl
ic
a
ti
ons
f
or
f
ood s
e
c
ur
it
y a
nd e
c
onomi
c
de
ve
lo
pm
e
nt
.
2.
R
E
L
A
T
E
D
WORK
I
n
th
e
r
e
a
lm
of
f
ood
c
la
s
s
if
ic
a
ti
on,
pa
r
ti
c
ul
a
r
ly
c
onc
e
r
ni
ng
th
e
c
a
te
gor
iz
a
ti
on
of
poul
tr
y
m
e
a
t,
a
w
e
a
lt
h
of
pr
io
r
in
ve
s
ti
ga
ti
ons
ha
s
la
id
th
e
g
r
oundwor
k
f
or
th
e
c
om
pa
r
a
ti
ve
a
na
ly
s
is
e
xpounde
d
w
it
hi
n
th
is
s
tu
dy.
O
ve
r
ti
m
e
,
r
e
s
e
a
r
c
he
r
s
ha
ve
d
e
lv
e
d
in
to
a
n
a
r
r
a
y
of
m
e
th
odol
ogi
c
a
l
a
ppr
oa
c
he
s
,
s
p
a
nni
ng
f
r
om
c
onve
nt
io
na
l
m
a
c
hi
ne
le
a
r
ni
ng
te
c
hni
que
s
to
th
e
m
or
e
in
tr
ic
a
t
e
r
e
a
lm
s
of
de
e
p
le
a
r
ni
ng
m
e
th
odol
ogi
e
s
.
F
or
in
s
ta
nc
e
, nume
r
ous
s
tu
di
e
s
ha
ve
de
lv
e
d i
nt
o t
he
e
f
f
ic
a
c
y of
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
s
(
S
V
M
s
)
,
r
a
ndom f
or
e
s
ts
,
a
nd
k
-
ne
a
r
e
s
t
ne
ig
hbor
s
(k
-
N
N
)
a
lg
or
it
hm
s
in
di
s
c
e
r
ni
ng
a
nd
c
la
s
s
if
yi
ng
poul
tr
y
m
e
a
t
ba
s
e
d
on
a
di
v
e
r
s
e
a
r
r
a
y
of
vi
s
ua
l
a
tt
r
ib
ut
e
s
a
nd
f
e
a
tu
r
e
s
[
8
]
,
[
9]
.
M
or
e
ove
r
,
th
e
a
dve
nt
a
nd
s
ubs
e
que
nt
m
a
tu
r
a
ti
on
of
de
e
p
le
a
r
ni
ng
m
e
th
odol
ogi
e
s
,
not
a
bl
y
C
N
N
s
,
ha
v
e
s
pa
r
ke
d
c
on
s
id
e
r
a
bl
e
in
te
r
e
s
t
w
it
hi
n
th
e
r
e
s
e
a
r
c
h
c
om
m
uni
ty
.
C
N
N
a
r
c
hi
te
c
tu
r
e
s
s
uc
h
a
s
V
G
G
,
R
e
s
N
e
t,
a
nd
I
nc
e
pt
io
n
ha
ve
e
m
e
r
ge
d
a
s
s
ta
lwa
r
ts
in
th
e
c
la
s
s
if
ic
a
ti
on
of
va
r
io
us
f
ood
it
e
m
s
,
in
c
lu
di
ng
poul
tr
y
m
e
a
t,
s
how
c
a
s
in
g
r
e
m
a
r
ka
bl
e
pe
r
f
or
m
a
nc
e
a
nd
a
c
c
ur
a
c
y
in
di
s
c
e
r
ni
ng
in
tr
ic
a
te
pa
tt
e
r
ns
a
nd
f
e
a
tu
r
e
s
w
it
hi
n
vi
s
ua
l
da
ta
[
10]
,
[
11]
.
I
t
i
s
w
it
hi
n
th
is
c
ont
e
xt
th
a
t
our
s
tu
dy
e
nde
a
vor
s
to
c
ont
r
ib
ut
e
s
ig
ni
f
ic
a
nt
ly
.
B
y
s
ys
te
m
a
ti
c
a
ll
y
c
om
pa
r
in
g
te
n
w
id
e
ly
e
m
pl
oye
d
C
N
N
a
r
c
hi
te
c
tu
r
e
s
on
a
s
ta
nda
r
di
z
e
d
a
nd
r
ig
or
ous
ly
c
ur
a
te
d
poul
tr
y
m
e
a
t
da
ta
s
e
t,
our
a
na
ly
s
is
a
im
s
to
s
he
d
li
ght
on
th
e
ir
r
e
s
pe
c
ti
ve
s
tr
e
ngt
hs
,
w
e
a
kn
e
s
s
e
s
,
a
nd
a
ppl
ic
a
bi
li
ty
in
r
e
a
l
-
w
or
ld
s
c
e
na
r
io
s
.
T
hr
ough
th
is
e
nde
a
vor
,
w
e
a
im
not
onl
y
to
a
dva
nc
e
th
e
und
e
r
s
ta
ndi
ng
of
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on
but
a
ls
o
to
pr
ovi
de
pr
a
c
ti
ti
one
r
s
a
nd
r
e
s
e
a
r
c
he
r
s
a
li
ke
w
it
h va
lu
a
bl
e
i
ns
ig
ht
s
i
nt
o t
he
opt
im
a
l
s
e
le
c
ti
on a
nd de
pl
o
ym
e
nt
of
C
N
N
m
ode
ls
f
or
s
im
il
a
r
t
a
s
ks
.
3.
M
A
T
E
R
I
A
L
S
A
N
D
M
E
T
H
O
D
S
T
hi
s
m
e
th
od
f
oc
us
e
s
on
c
l
a
s
s
if
yi
ng
f
our
s
pe
c
if
ic
poul
tr
y
c
a
te
gor
ie
s
-
c
hi
c
ke
n,
tu
r
ke
y,
F
a
youmi
,
a
nd
c
hi
c
ke
n
f
a
r
m
e
r
-
us
in
g
va
r
io
us
pr
e
-
tr
a
in
e
d
C
N
N
a
r
c
hi
te
c
tu
r
e
s
.
I
n
th
is
s
tu
dy,
w
e
e
xpl
oi
t
th
e
c
a
pa
bi
li
ti
e
s
of
te
n
di
ve
r
s
e
C
N
N
a
r
c
hi
te
c
tu
r
e
s
,
in
c
lu
di
ng
V
G
G
19,
V
G
G
16,
R
e
s
N
e
t5
0,
G
oogl
e
N
e
t,
M
obi
le
N
e
tV1,
M
obi
le
N
e
tV2,
D
e
ns
e
N
e
t,
N
a
s
N
e
t,
E
f
f
ic
ie
nt
N
e
t
,
a
nd
A
le
xN
e
t.
U
nl
ik
e
c
onve
nt
io
na
l
m
e
th
ods
th
a
t
of
te
n
r
e
qui
r
e
tr
a
in
in
g
la
r
ge
m
ode
ls
f
r
om
s
c
r
a
tc
h
on
e
xt
e
ns
iv
e
da
ta
s
e
ts
,
by
ut
il
iz
in
g
p
r
e
-
tr
a
in
e
d
f
e
a
tu
r
e
m
a
ps
,
w
e
c
ir
c
um
ve
nt
th
e
ne
e
d
to
s
ta
r
t
tr
a
in
in
g
f
r
om
th
e
gr
ound
up,
th
e
r
e
by
s
a
vi
ng
c
om
put
a
ti
on
a
l
r
e
s
our
c
e
s
a
nd
ti
m
e
.
T
hr
ough
th
is
a
ppr
oa
c
h,
th
e
r
e
s
ul
ti
ng
m
ode
ls
d
e
m
ons
tr
a
te
a
r
e
m
a
r
ka
bl
e
a
bi
li
ty
to
vi
s
ua
ll
y
di
f
f
e
r
e
nt
ia
te
be
twe
e
n
c
hi
c
ke
n,
tu
r
ke
y,
F
a
youmi
, a
nd c
hi
c
ke
n f
a
r
m
e
r
s
w
it
h a
hi
gh l
e
ve
l
of
a
c
c
ur
a
c
y
a
c
r
os
s
t
he
va
r
io
us
C
N
N
a
r
c
hi
te
c
tu
r
e
s
t
e
s
te
d.
4.
I
M
A
G
E
S
A
C
Q
U
I
S
I
T
I
O
N
T
hi
s
s
tu
dy
on
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on
c
onc
e
nt
r
a
te
d
on
c
h
ic
ke
n,
tu
r
ke
y,
a
nd
c
hi
c
ke
n
f
a
r
m
e
r
s
,
a
c
knowle
dgi
ng
th
a
t
M
or
oc
c
a
ns
a
r
e
a
m
ong
th
e
hi
ghe
s
t
c
on
s
um
e
r
s
of
m
e
a
t
gl
oba
ll
y,
a
ve
r
a
gi
ng
30
ki
lo
g
r
a
m
s
c
ons
um
e
d
pe
r
pe
r
s
on
e
a
c
h
ye
a
r
.
I
n
M
or
oc
c
o,
a
s
in
m
a
ny
ot
he
r
c
ount
r
ie
s
,
m
e
a
t
hol
ds
s
ig
ni
f
ic
a
nt
s
o
c
ia
l
va
lu
e
a
nd
is
c
ons
id
e
r
e
d
a
hi
ghl
y
e
s
te
e
m
e
d
c
om
pone
nt
of
th
e
di
e
t
[
12]
.
P
oul
tr
y
c
ons
ti
tu
te
s
th
e
m
a
jo
r
it
y
of
m
e
a
t
c
ons
um
pt
io
n i
n M
or
oc
c
o, be
c
a
us
e
of
i
ts
l
ow
e
r
pr
ic
e
[
13]
.
T
o e
n
ha
nc
e
a
c
c
ur
a
c
y,
w
e
i
n
c
lu
d
e
d F
a
youmi
a
s
w
e
ll
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
C
om
par
at
iv
e
analy
s
is
of
c
onv
ol
ut
io
nal
ne
u
r
al
ne
tw
or
k
ar
c
hi
te
c
tu
r
e
s
f
or
poult
r
y
…
(
Sal
m
a
Se
k
h
r
a
)
3717
r
e
s
ul
ti
ng
in
a
da
ta
s
e
t
c
ove
r
in
g
f
our
poul
tr
y
ty
p
e
s
.
T
h
e
da
t
a
s
e
t
w
a
s
c
om
pi
le
d
by
pur
c
ha
s
in
g
va
r
i
ous
c
ut
s
of
tu
r
ke
y,
c
hi
c
ke
n
,
F
a
y
oum
i,
a
nd
c
hi
c
ke
n
f
a
r
m
e
r
f
r
o
m
a
m
a
r
ke
t
i
n
M
e
kne
s
,
M
or
o
c
c
o
.
W
e
u
s
e
d
a
1
6
-
m
e
g
a
pi
x
e
l
di
gi
ta
l
c
a
m
e
r
a
phone
(
H
ua
w
e
i
Y
9
P
r
im
e
)
a
lo
ng
w
it
h
a
phot
ogr
a
phy
L
E
D
box
,
a
s
s
how
n
in
F
ig
ur
e
1
,
to
c
a
pt
ur
e
hi
gh
-
qua
li
ty
im
a
ge
s
.
A
ddi
ti
ona
ll
y,
w
e
e
m
pl
oye
d
a
pr
ogr
a
m
f
or
im
a
ge
c
r
oppi
ng
to
a
ugm
e
nt
th
e
da
ta
s
e
t.
I
m
a
ge
s
w
e
r
e
t
a
ke
n
of
di
f
f
e
r
e
nt
pa
r
ts
of
th
e
bi
r
d,
in
c
l
udi
ng
th
e
th
ig
h,
dr
um
s
ti
c
k,
w
in
g,
br
e
a
s
t,
a
nd
ne
c
k.
T
hi
s
a
ppr
oa
c
h
a
ll
ow
e
d
us
to
c
a
pt
ur
e
de
ta
il
e
d
a
tt
r
ib
ut
e
s
a
nd
te
xt
ur
e
s
c
r
uc
ia
l
f
or
c
la
s
s
if
ic
a
ti
on.
T
he
im
a
ge
s
w
e
r
e
c
a
pt
ur
e
d
ove
r
a
pe
r
io
d
of
20
da
y
s
,
e
ns
ur
in
g
di
ve
r
s
it
y
in
th
e
da
ta
s
e
t.
T
he
or
ig
in
a
l
im
a
ge
di
m
e
ns
io
ns
w
e
r
e
4608
×
3456
pi
xe
l
s
,
w
hi
c
h
w
e
r
e
r
e
s
i
z
e
d
to
40
8
×
306
pi
xe
ls
to
e
ns
ur
e
c
om
p
a
ti
bi
li
ty
w
it
h
th
e
m
ode
l
a
nd
a
c
c
om
m
od
a
te
s
to
r
a
ge
li
m
it
a
ti
ons
on
G
oogl
e
D
r
iv
e
.
W
e
f
oc
us
e
d
on
pr
e
s
e
r
vi
ng
c
ol
or
a
nd
te
xt
ur
e
de
ta
il
s
,
c
r
uc
ia
l
f
or
a
c
c
ur
a
te
c
la
s
s
if
ic
a
ti
on
[
14]
,
[
15]
,
a
nd
z
oom
e
d
in
on
f
e
a
tu
r
e
s
a
nd
a
tt
r
ib
ut
e
s
of
th
e
im
a
ge
s
.
S
a
m
pl
e
i
m
a
ge
s
f
r
om
our
da
ta
s
e
t
a
r
e
de
pi
c
te
d i
n F
ig
ur
e
2.
F
ig
ur
e
1
.
P
hot
ogr
a
phy L
ED
box
F
ig
ur
e
2
.
D
a
ta
s
e
t
s
a
m
pl
e
s
[
1]
5.
D
A
T
A
S
E
T
A
U
G
M
E
N
T
A
T
I
O
N
T
he
tr
a
in
in
g
d
a
ta
s
e
t
c
ons
ti
tu
te
s
a
pi
vot
a
l
c
om
pone
nt
,
c
om
pr
is
in
g
a
to
ta
l
of
746
im
a
ge
s
th
a
t
ha
ve
be
e
n
c
a
te
gor
iz
e
d
in
to
f
our
di
s
ti
nc
t
c
la
s
s
e
s
.
R
e
c
ogni
z
in
g
th
e
s
i
gni
f
ic
a
nc
e
of
da
ta
a
ugm
e
nt
a
ti
on
in
e
nha
nc
in
g
m
ode
l
ge
ne
r
a
li
z
a
ti
on
a
nd
pe
r
f
or
m
a
nc
e
,
w
e
to
ok
m
e
a
s
ur
e
s
to
e
nr
ic
h
th
e
da
ta
s
e
t
f
ur
th
e
r
.
L
e
ve
r
a
gi
ng
a
c
odi
ng
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3715
-
3723
3718
pr
ogr
a
m
de
ve
lo
pe
d,
w
e
im
pl
e
m
e
nt
e
d
a
ugm
e
nt
a
ti
on
te
c
hni
q
ue
s
to
di
ve
r
s
if
y
th
e
da
ta
s
e
t.
T
hr
ough
th
is
a
ugm
e
nt
a
ti
on
pr
oc
e
s
s
,
e
a
c
h
or
ig
in
a
l
im
a
ge
w
a
s
tr
a
ns
f
or
m
e
d
in
to
e
ig
ht
va
r
ia
ti
ons
,
in
c
lu
di
ng
r
ot
a
ti
ons
a
nd
hor
iz
ont
a
l
m
ir
r
o
r
in
g.
T
hi
s
a
ugm
e
nt
a
ti
on
s
tr
a
te
gy
s
ig
ni
f
ic
a
nt
ly
bol
s
te
r
e
d
th
e
da
ta
s
e
t,
e
xpa
ndi
ng
it
to
a
to
ta
l
of
7614
im
a
ge
s
,
a
s
vi
s
ua
ll
y
de
pi
c
te
d
in
F
ig
ur
e
3.
T
hi
s
a
ugm
e
nt
a
ti
on
not
onl
y
a
m
pl
if
ie
s
th
e
da
ta
s
e
t'
s
vol
um
e
but
a
ls
o
e
nr
ic
he
s
it
s
di
ve
r
s
it
y,
e
m
pow
e
r
in
g
our
m
od
e
ls
w
it
h
a
m
or
e
c
om
pr
e
he
ns
iv
e
unde
r
s
ta
ndi
ng
of
th
e
va
r
ia
bi
li
ty
w
it
hi
n t
he
da
ta
s
e
t,
t
he
r
e
by e
nha
nc
in
g t
he
ir
r
obus
tn
e
s
s
a
nd c
la
s
s
if
ic
a
ti
on a
c
c
ur
a
c
y.
F
ig
ur
e
3
.
E
xa
m
pl
e
of
a
n i
m
a
ge
a
ugm
e
nt
a
ti
on
6.
T
H
E
M
O
D
E
L
S
T
he
s
tu
dy
ut
il
iz
e
d
a
di
ve
r
s
e
a
r
r
a
y
of
pr
e
-
tr
a
in
e
d
C
N
N
m
ode
l
s
t
o
pr
e
pr
oc
e
s
s
a
nd
c
la
s
s
if
y
th
e
poul
tr
y
m
e
a
t
da
ta
s
e
t.
in
F
ig
ur
e
4
,
a
n
ove
r
vi
e
w
of
C
N
N
a
r
c
hi
te
c
tu
r
e
is
s
pe
c
if
ie
d.
T
he
s
e
m
ode
ls
e
nc
om
pa
s
s
e
d
a
w
id
e
r
a
nge
of
a
r
c
hi
te
c
tu
r
e
s
a
nd
c
ha
r
a
c
te
r
is
ti
c
s
,
in
c
lu
di
ng
V
G
G
19,
V
G
G
16,
R
e
s
N
e
t5
0,
G
oogl
e
N
e
t,
M
obi
le
N
e
tV1,
M
obi
le
N
e
tV2,
D
e
ns
e
N
e
t,
N
a
s
N
e
t,
E
f
f
ic
ie
nt
N
e
t
,
a
nd
A
le
xN
e
t.
E
a
c
h
of
th
e
s
e
m
ode
ls
ha
s
be
e
n
pr
e
tr
a
in
e
d
on
la
r
ge
-
s
c
a
le
im
a
ge
da
ta
s
e
ts
li
ke
I
m
a
ge
N
e
t,
e
na
bl
in
g
th
e
m
to
le
a
r
n
r
ic
h
r
e
pr
e
s
e
nt
a
ti
ons
of
vi
s
ua
l
f
e
a
tu
r
e
s
,
w
hi
c
h c
a
n be
e
f
f
e
c
ti
ve
ly
t
r
a
ns
f
e
r
r
e
d t
o our
poult
r
y m
e
a
t
c
la
s
s
if
i
c
a
ti
on t
a
s
k.
V
G
G
19,
V
G
G
16
,
a
nd
R
e
s
N
e
t5
0
a
r
e
r
e
now
ne
d
f
or
th
e
ir
de
e
p
a
r
c
hi
te
c
tu
r
e
s
a
nd
s
upe
r
io
r
pe
r
f
or
m
a
nc
e
in
im
a
ge
c
la
s
s
if
ic
a
ti
on
ta
s
ks
[
16]
–
[
18]
.
N
a
s
N
e
t,
on
th
e
ot
he
r
ha
nd,
is
de
s
ig
ne
d
to
be
c
om
pa
c
t,
m
a
ki
ng
it
s
ui
ta
bl
e
f
or
de
pl
oym
e
nt
on
r
e
s
our
c
e
-
c
ons
tr
a
in
e
d
de
vi
c
e
s
w
it
hout
c
om
pr
om
is
in
g
a
c
c
ur
a
c
y.
N
a
s
N
e
t
a
r
c
hi
te
c
tu
r
e
s
of
te
n
a
c
hi
e
ve
s
ta
te
-
of
-
th
e
-
a
r
t
pe
r
f
or
m
a
nc
e
but
r
e
qui
r
e
s
ig
ni
f
ic
a
nt
c
om
put
a
ti
ona
l
r
e
s
our
c
e
s
f
or
th
e
ir
de
s
ig
n
[
19]
.
M
obi
le
N
e
tV2
a
nd
M
obi
le
N
e
tV1
a
r
e
opt
im
iz
e
d
f
or
m
obi
le
a
nd
e
m
be
dde
d
a
ppl
ic
a
ti
ons
,
of
f
e
r
in
g l
ig
ht
w
e
ig
ht
a
r
c
hi
te
c
tu
r
e
s
w
hi
le
m
a
in
ta
in
in
g c
om
pe
ti
ti
v
e
a
c
c
ur
a
c
y
[
20]
–
[
22]
. D
e
ns
e
N
e
t
s
ta
nds
out
f
or
it
s
de
ns
e
ly
c
onn
e
c
te
d l
a
ye
r
s
, pr
om
ot
in
g f
e
a
tu
r
e
r
e
us
e
a
nd f
a
c
il
it
a
ti
ng e
f
f
ic
ie
nt
t
r
a
in
in
g
[
23]
.
E
f
f
ic
ie
nt
N
e
t
is
pa
r
t
of
a
f
a
m
il
y
of
m
ode
ls
th
a
t
s
c
a
le
up
in
c
o
m
pl
e
xi
ty
a
nd
a
c
c
ur
a
c
y
by
ba
la
nc
in
g
ne
twor
k
de
pt
h,
w
id
th
,
a
nd
r
e
s
ol
ut
io
n
[
24]
.
A
le
xN
e
t,
one
of
th
e
pi
one
e
r
in
g
C
N
N
a
r
c
hi
te
c
tu
r
e
s
,
in
tr
oduc
e
d
ke
y
c
onc
e
pt
s
s
uc
h
a
s
c
onvolut
io
na
l
l
a
ye
r
s
a
nd
r
e
c
ti
f
ie
d
li
ne
a
r
uni
ts
(
R
e
L
U
)
,
pa
vi
ng
th
e
w
a
y
f
or
m
ode
r
n
de
e
p
le
a
r
ni
ng
r
e
s
e
a
r
c
h
[
25]
.
G
oogl
e
N
e
t,
w
it
h
it
s
in
c
e
pt
io
n
m
odul
e
s
a
nd
gl
oba
l
a
ve
r
a
ge
pool
in
g,
e
m
pha
s
iz
e
s
bot
h
de
pt
h a
nd c
om
put
a
ti
ona
l
e
f
f
ic
ie
nc
y
[
26]
.
B
y
le
ve
r
a
gi
ng
pr
e
-
tr
a
in
e
d
m
ode
ls
,
w
e
c
a
pi
ta
li
z
e
d
on
th
e
w
e
a
lt
h
of
knowle
dge
th
e
s
e
m
ode
ls
ha
ve
a
c
c
um
ul
a
te
d
dur
in
g
th
e
ir
tr
a
in
in
g
on
di
ve
r
s
e
im
a
ge
da
ta
s
e
ts
.
T
hi
s
a
ppr
oa
c
h
s
ig
ni
f
ic
a
nt
ly
r
e
duc
e
d
th
e
c
om
put
a
ti
ona
l
ove
r
he
a
d
a
nd
ti
m
e
r
e
qui
r
e
d
f
or
tr
a
in
in
g,
a
ll
ow
i
ng
us
to
f
oc
us
on
f
in
e
-
tu
ni
ng
th
e
m
ode
ls
f
o
r
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on.
M
or
e
ove
r
,
by
e
va
lu
a
ti
ng
th
e
pe
r
f
or
m
a
nc
e
of
th
e
s
e
di
ve
r
s
e
m
ode
l
s
a
c
r
o
s
s
va
r
io
us
m
e
tr
ic
s
s
uc
h
a
s
a
c
c
ur
a
c
y,
pr
e
c
is
io
n,
r
e
c
a
ll
,
a
nd
F
1
-
s
c
or
e
,
w
e
ga
in
e
d
va
lu
a
bl
e
in
s
ig
ht
s
in
to
th
e
ir
e
f
f
e
c
ti
ve
ne
s
s
a
nd s
ui
ta
bi
li
ty
f
or
poult
r
y m
e
a
t
c
la
s
s
if
ic
a
ti
on t
a
s
ks
.
T
hr
o
ug
h
th
is
c
o
m
p
r
e
he
ns
i
ve
a
na
ly
s
is
,
w
e
a
im
e
d
to
id
e
nt
if
y
t
he
m
os
t
e
f
f
e
c
ti
ve
C
N
N
a
r
c
hi
te
c
t
ur
e
f
o
r
po
ul
tr
y
m
e
a
t
c
la
s
s
i
f
i
c
a
t
io
n,
c
ons
id
e
r
in
g
f
a
c
to
r
s
s
u
c
h
a
s
a
c
c
u
r
a
c
y,
e
f
f
ic
i
e
nc
y,
a
nd
s
c
a
la
b
il
it
y.
B
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
C
om
par
at
iv
e
analy
s
is
of
c
onv
ol
ut
io
nal
ne
u
r
al
ne
tw
or
k
ar
c
hi
te
c
tu
r
e
s
f
or
poult
r
y
…
(
Sal
m
a
Se
k
h
r
a
)
3719
e
va
l
ua
t
in
g
th
e
te
n
p
r
om
in
e
nt
a
r
c
h
it
e
c
tu
r
e
s
-
V
G
G
19
,
V
G
G
16,
R
e
s
N
e
t5
0,
G
o
og
le
N
e
t,
M
ob
i
le
N
e
tV1
,
M
obi
le
N
e
tV
2,
D
e
ns
e
N
e
t,
N
a
s
N
e
t
,
E
f
f
ic
ie
nt
N
e
t
,
a
nd
A
le
x
N
e
t
-
w
e
s
o
ug
ht
to
de
te
r
m
in
e
w
hi
c
h
m
ode
l
pe
r
f
o
r
m
s
b
e
s
t
in
d
is
t
in
gu
is
hi
ng
be
twe
e
n
c
hi
c
ke
n
,
tu
r
ke
y,
f
a
y
oum
i,
a
n
d
c
hi
c
ke
n
f
a
r
m
e
r
m
e
a
t.
O
u
r
f
in
d
in
gs
r
e
ve
a
l
t
ha
t
c
e
r
ta
i
n
a
r
c
h
it
e
c
tu
r
e
s
,
s
uc
h
a
s
D
e
ns
e
N
e
t
a
n
d
V
G
G
16
,
de
m
ons
t
r
a
te
d
s
up
e
r
io
r
a
c
c
ur
a
c
y
a
n
d
c
om
p
ut
a
t
io
n
a
l
e
f
f
ic
ie
nc
y,
m
a
ki
ng
t
he
m
w
e
l
l
-
s
ui
te
d
f
o
r
r
e
a
l
-
w
or
ld
a
ppl
ic
a
ti
ons
i
n
po
ul
tr
y
m
e
a
t
c
la
s
s
i
f
ic
a
t
io
n.
T
he
s
e
r
e
s
ul
ts
c
o
nt
r
i
bu
te
t
o
t
he
a
d
va
nc
e
m
e
n
t
o
f
r
e
s
e
a
r
c
h
i
n
l
e
ve
r
a
g
in
g
de
e
p
le
a
r
ni
ng
te
c
h
ni
que
s
f
or
f
oo
d
q
ua
l
it
y
a
s
s
e
s
s
m
e
n
t
a
nd
a
u
th
e
nt
ic
a
t
io
n,
w
it
h
po
te
n
ti
a
l
a
pp
li
c
a
ti
ons
in
i
ndus
t
r
ie
s
in
v
ol
ve
d
i
n
p
ou
lt
r
y
m
e
a
t
p
r
od
uc
t
io
n a
nd
di
s
t
r
ib
ut
io
n.
F
ig
ur
e
4
.
H
ig
h l
e
ve
l
de
s
ig
n of
c
onvolut
io
n a
r
c
hi
te
c
tu
r
e
7.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
I
n
th
is
s
e
c
ti
on,
w
e
pr
e
s
e
nt
a
nd
di
s
c
us
s
th
e
r
e
s
ul
ts
of
our
c
om
pa
r
a
ti
ve
a
na
ly
s
is
of
te
n
C
N
N
m
ode
ls
f
or
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on.
T
he
m
ode
l
s
e
va
lu
a
te
d
in
c
lu
de
V
G
G
19,
V
G
G
16,
R
e
s
N
e
t5
0,
G
oogl
e
N
e
t,
M
obi
le
N
e
tV1,
M
obi
le
N
e
tV2,
D
e
n
s
e
N
e
t,
N
a
s
N
e
t,
E
f
f
ic
ie
nt
N
e
t
,
a
nd
A
le
xN
e
t.
W
e
a
na
ly
s
e
d
th
e
ir
pe
r
f
or
m
a
nc
e
ba
s
e
d on a
c
c
ur
a
c
y a
nd l
os
s
m
e
tr
ic
s
t
o de
te
r
m
in
e
t
he
ir
s
ui
ta
bi
li
t
y f
or
c
la
s
s
if
yi
ng poult
r
y m
e
a
t.
7.1.
P
e
r
f
or
m
an
c
e
c
o
m
p
ar
is
on
T
he
r
e
s
ul
ts
in
T
a
bl
e
1
pr
e
s
e
nt
a
nua
nc
e
d
b
a
la
nc
e
be
t
w
e
e
n
c
om
put
a
ti
ona
l
e
f
f
ic
ie
nc
y
a
nd
c
la
s
s
if
ic
a
ti
on
a
c
c
ur
a
c
y a
c
r
os
s
t
he
e
va
lu
a
t
e
d C
N
N
m
ode
ls
. D
e
n
s
e
N
e
t
e
m
e
r
ge
s
a
s
a
f
r
ont
r
unne
r
, s
how
c
a
s
in
g a
n
out
s
ta
ndi
ng
a
c
c
ur
a
c
y
of
98%
w
it
h
a
n
e
xc
e
pt
io
na
ll
y
lo
w
lo
s
s
of
0.03.
T
hi
s
r
e
m
a
r
ka
bl
e
pe
r
f
o
r
m
a
nc
e
c
a
n
be
a
tt
r
ib
ut
e
d
to
D
e
ns
e
N
e
t'
s
de
ns
e
c
onne
c
ti
vi
ty
pa
tt
e
r
n,
w
hi
c
h
p
r
om
ot
e
s
f
e
a
tu
r
e
r
e
us
e
a
nd
gr
a
di
e
nt
f
lo
w
,
a
s
hi
ghl
ig
ht
e
d
by
[
27]
.
T
hi
s
f
in
di
ng
unde
r
s
c
or
e
s
th
e
im
por
ta
nc
e
of
in
tr
ic
a
te
ne
twor
k
a
r
c
hi
te
c
tu
r
e
s
in
a
c
hi
e
vi
ng
hi
gh a
c
c
ur
a
c
y i
n i
m
a
ge
c
la
s
s
if
ic
a
ti
on t
a
s
k
s
.
T
r
a
di
ti
ona
l
a
r
c
hi
te
c
tu
r
e
s
s
uc
h
a
s
V
G
G
19,
V
G
G
16
,
a
nd
R
e
s
N
e
t5
0
e
xhi
bi
t
c
om
pe
ti
ti
ve
pe
r
f
or
m
a
nc
e
,
w
it
h
a
c
c
ur
a
c
ie
s
of
95%
,
97%
a
nd
91%
.
T
he
s
e
m
ode
l
s
,
r
e
no
w
ne
d
f
or
th
e
ir
de
e
p
a
r
c
hi
te
c
tu
r
e
s
a
nd
r
e
s
id
ua
l
c
onne
c
ti
ons
,
e
xc
e
l
in
c
a
pt
ur
in
g
in
tr
ic
a
te
f
e
a
tu
r
e
s
w
it
hi
n
poul
tr
y
m
e
a
t
im
a
ge
s
[
28]
.
H
ow
e
ve
r
,
th
e
ir
de
e
pe
r
s
tr
uc
tu
r
e
s
m
a
y e
nt
a
il
hi
ghe
r
c
om
put
a
ti
ona
l
ove
r
he
a
ds
dur
in
g both t
r
a
in
in
g a
nd
in
f
e
r
e
nc
e
pha
s
e
s
, ne
c
e
s
s
it
a
ti
ng
c
a
r
e
f
ul
c
ons
id
e
r
a
ti
on i
n r
e
s
our
c
e
-
c
ons
tr
a
in
e
d e
nvi
r
onm
e
nt
s
.
T
a
bl
e
1
.
T
he
a
c
c
ur
a
c
y a
nd l
os
s
va
lu
e
s
of
t
e
n m
ode
l
s
M
ode
l
A
c
c
ur
a
c
y
(%)
L
os
s
V
G
G
19
95
0.14
V
G
G
16
97
0.11
R
e
s
N
e
t
50
91
0.12
M
obi
l
e
N
e
t
V
2
94
0.20
M
obi
l
e
N
e
t
V
1
95
0.21
D
e
ns
e
N
e
t
98
0.03
E
f
f
i
c
i
e
nt
N
e
t
95
0.21
A
l
e
xN
e
t
77
0.30
N
a
s
N
e
t
95
0.16
G
oogl
e
N
e
t
95
0.11
C
onve
r
s
e
ly
,
li
ght
w
e
ig
ht
a
r
c
hi
te
c
tu
r
e
s
li
ke
N
a
s
N
e
t
,
G
oogl
e
N
e
t
,
a
nd
M
obi
le
N
e
tV1
d
e
m
ons
tr
a
te
c
om
m
e
nda
bl
e
c
om
put
a
ti
ona
l
e
f
f
ic
ie
nc
y.
T
he
y
a
c
hi
e
ve
a
n
a
c
c
u
r
a
c
y
of
95%
.
T
he
s
e
m
ode
ls
,
c
ha
r
a
c
te
r
iz
e
d
by
th
e
ir
c
om
pa
c
t
a
r
c
hi
te
c
tu
r
e
s
a
nd
pa
r
a
m
e
te
r
-
e
f
f
ic
ie
nt
de
s
ig
ns
,
of
f
e
r
pr
om
is
in
g
s
ol
ut
io
ns
f
or
de
pl
oym
e
nt
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3715
-
3723
3720
r
e
s
our
c
e
-
c
ons
tr
a
in
e
d
e
nvi
r
onm
e
nt
s
[
29]
,
[
30]
.
M
ode
l
s
w
it
h
s
ha
ll
ow
e
r
a
r
c
hi
te
c
tu
r
e
s
,
s
uc
h
a
s
A
le
xN
e
t
a
nd
M
obi
le
N
e
tV2,
s
tr
uggl
e
to
a
c
hi
e
ve
c
om
pa
r
a
bl
e
a
c
c
ur
a
c
ie
s
,
w
it
h
A
le
xN
e
t
a
c
hi
e
vi
ng
77%
a
c
c
ur
a
c
y
a
nd
M
obi
le
N
e
tV2
a
c
hi
e
vi
ng
94%
.
T
he
s
e
m
ode
l
s
,
w
it
h
th
e
ir
s
im
pl
e
r
a
r
c
hi
te
c
tu
r
e
s
,
m
a
y
e
nc
ount
e
r
di
f
f
ic
ul
ti
e
s
in
c
a
pt
ur
in
g i
nt
r
ic
a
te
f
e
a
tu
r
e
s
pr
e
s
e
nt
i
n pou
lt
r
y
m
e
a
t
im
a
ge
s
, l
e
a
di
ng t
o r
e
duc
e
d pe
r
f
or
m
a
nc
e
c
om
pa
r
e
d t
o t
he
i
r
c
ount
e
r
pa
r
ts
[
31]
.
O
ur
c
om
pa
r
a
ti
ve
a
na
ly
s
is
s
he
ds
li
ght
on
t
he
s
tr
e
ngt
hs
a
nd
li
m
it
a
ti
ons
of
va
r
io
us
C
N
N
m
ode
ls
f
or
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on.
B
y
unde
r
s
ta
ndi
ng
th
e
s
e
nua
n
c
e
s
,
pr
a
c
ti
ti
one
r
s
c
a
n
m
a
ke
in
f
or
m
e
d
de
c
is
io
ns
w
he
n
s
e
le
c
ti
ng mode
ls
f
or
r
e
a
l
-
w
or
ld
a
ppl
ic
a
ti
ons
i
n t
he
poult
r
y i
ndus
tr
y, ul
ti
m
a
te
ly
c
ont
r
ib
ut
in
g t
o
m
or
e
e
f
f
ic
ie
nt
a
nd a
c
c
ur
a
te
c
la
s
s
if
ic
a
ti
on s
ys
t
e
m
s
.
7.2.
T
r
ai
n
in
g
c
u
r
ve
s
T
he
s
e
c
ur
ve
s
,
a
s
s
how
n
in
F
ig
ur
e
5
,
p
r
ovi
de
in
s
ig
ht
s
in
to
th
e
tr
a
in
in
g
a
nd
va
li
da
ti
on
pr
ogr
e
s
s
of
e
a
c
h
m
ode
l,
il
lu
s
tr
a
ti
ng
how
a
c
c
ur
a
c
y
im
pr
ove
s
a
s
s
how
n
in
F
ig
ur
e
s
5(
a
)
a
nd
5
(
b)
a
nd
lo
s
s
de
c
r
e
a
s
e
s
a
s
s
how
n
in
F
ig
ur
e
s
5(
c
)
a
nd
5
(
d)
ove
r
s
uc
c
e
s
s
iv
e
e
poc
h
s
of
tr
a
in
in
g.
F
or
in
s
ta
nc
e
,
de
e
pe
r
a
r
c
hi
te
c
tu
r
e
s
li
ke
D
e
ns
e
N
e
t,
V
G
G
16,
R
e
s
N
e
t5
0
,
a
nd
E
f
f
ic
ie
nt
N
e
t
e
xhi
bi
t
s
m
oot
he
r
c
onve
r
ge
nc
e
c
ur
ve
s
,
in
di
c
a
ti
ng
s
ta
bl
e
le
a
r
ni
ng
dyna
m
ic
s
a
nd
e
f
f
e
c
ti
ve
f
e
a
tu
r
e
e
xt
r
a
c
ti
on,
w
hi
le
s
ha
ll
ow
e
r
m
ode
ls
li
ke
A
le
xN
e
t
s
how
m
or
e
f
lu
c
tu
a
ti
ons
,
r
e
f
le
c
ti
ng
c
ha
ll
e
nge
s
in
c
a
pt
ur
in
g
c
om
pl
e
x
p
a
tt
e
r
ns
.
S
uc
h
vi
s
ua
li
z
a
ti
ons
of
f
e
r
a
de
e
pe
r
unde
r
s
ta
ndi
ng
of
th
e
le
a
r
ni
ng
dyna
m
ic
s
a
nd
c
onve
r
ge
nc
e
be
ha
vi
or
of
th
e
C
N
N
m
ode
ls
,
c
om
pl
e
m
e
nt
in
g
th
e
qua
nt
it
a
ti
ve
e
va
lu
a
ti
on
of
th
e
ir
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
.
B
y
a
na
ly
s
in
g
th
e
s
e
c
ur
ve
s
,
w
e
c
a
n
id
e
nt
if
y
m
ode
ls
th
a
t
not
onl
y
a
c
hi
e
ve
hi
gh
a
c
c
ur
a
c
y
but
a
l
s
o
de
m
ons
tr
a
te
c
on
s
is
te
nt
a
nd
r
e
li
a
bl
e
tr
a
in
in
g
be
ha
vi
or
,
w
hi
c
h
is
c
r
uc
ia
l
f
or
r
e
a
l
-
w
or
ld
de
pl
oym
e
nt
i
n poult
r
y m
e
a
t
c
la
s
s
if
ic
a
ti
on
t
a
s
ks
.
(
a
)
(
b)
(
c
)
(
d)
F
ig
ur
e
5. I
ns
ig
ht
s
i
nt
o t
he
t
r
a
in
in
g a
nd va
li
da
ti
on pr
ogr
e
s
s
of
e
a
c
h m
ode
l
:
(
a
)
t
r
a
in
in
g a
c
c
ur
a
c
y
c
ur
ve
s
f
or
e
a
c
h m
ode
l,
(
b)
va
l
id
a
ti
on
a
c
c
ur
a
c
y
c
ur
ve
s
f
or
e
a
c
h
m
od
e
l,
(
c
)
t
r
a
i
ni
n
g
lo
s
s
c
ur
ve
s
f
or
e
a
c
h m
od
e
l
,
a
nd
(
d)
v
a
li
da
ti
o
n
lo
s
s
c
ur
ve
s
f
or
e
a
c
h m
od
e
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
C
om
par
at
iv
e
analy
s
is
of
c
onv
ol
ut
io
nal
ne
u
r
al
ne
tw
or
k
ar
c
hi
te
c
tu
r
e
s
f
or
poult
r
y
…
(
Sal
m
a
Se
k
h
r
a
)
3721
8.
F
U
T
U
R
E
D
I
R
E
C
T
I
O
N
S
L
ooki
ng
a
he
a
d,
f
ut
ur
e
r
e
s
e
a
r
c
h
in
poul
tr
y
m
e
a
t
c
la
s
s
if
ic
a
ti
on
c
oul
d
e
xpl
or
e
e
xpa
ndi
ng
th
e
s
c
ope
to
in
c
lu
de
m
in
c
e
d
m
e
a
t
a
na
ly
s
is
.
I
nc
or
por
a
ti
ng
m
in
c
e
d
m
e
a
t
c
l
a
s
s
if
ic
a
ti
on
po
s
e
s
uni
que
c
ha
ll
e
nge
s
,
s
uc
h
a
s
di
s
ti
ngui
s
hi
ng
be
twe
e
n
m
e
a
t
pa
r
ti
c
le
s
a
nd
gr
e
a
s
e
c
ont
e
nt
.
F
ut
ur
e
r
e
s
e
a
r
c
h
c
oul
d
ut
il
iz
e
a
dva
nc
e
d
im
a
ge
pr
oc
e
s
s
in
g
a
nd
m
a
c
hi
ne
le
a
r
ni
ng
te
c
hni
que
s
to
a
c
hi
e
v
e
pr
e
c
is
e
c
la
s
s
if
ic
a
ti
on
of
m
in
c
e
d
poul
tr
y
m
e
a
t,
w
hi
le
a
ls
o
e
s
ti
m
a
ti
ng
th
e
pr
opor
ti
on
of
m
e
a
t
a
nd
f
a
t
in
th
e
s
a
m
pl
e
s
.
T
hi
s
e
xt
e
ns
io
n
w
oul
d
not
onl
y
e
nha
nc
e
th
e
a
ppl
ic
a
bi
li
ty
of
C
N
N
s
in
poul
tr
y
m
e
a
t
a
na
ly
s
is
but
a
ls
o
pr
ov
id
e
va
lu
a
bl
e
in
s
ig
ht
s
f
or
th
e
f
ood
pr
oc
e
s
s
in
g
in
dus
tr
y, pa
r
ti
c
ul
a
r
ly
i
n qua
li
ty
c
ont
r
ol
a
nd pr
oduc
t
f
o
r
m
ul
a
ti
on
.
9.
C
O
N
C
L
U
S
I
O
N
T
he
s
t
ud
y
pr
ov
id
e
s
va
lu
a
bl
e
i
ns
i
gh
ts
in
t
o
th
e
a
ppl
ic
a
ti
on
o
f
C
N
N
s
f
o
r
po
ul
tr
y
m
e
a
t
c
la
s
s
i
f
ic
a
ti
o
n,
de
m
o
ns
t
r
a
ti
ng
th
e
f
e
a
s
i
bi
li
ty
o
f
a
u
to
m
a
ti
n
g
th
is
t
a
s
k
th
r
oug
h
a
c
om
pa
r
a
t
iv
e
a
n
a
ly
s
is
of
te
n
C
N
N
a
r
c
h
it
e
c
t
ur
e
s
e
va
l
ua
te
d
o
n
a
s
ta
n
da
r
di
z
e
d
da
ta
s
e
t.
W
e
s
h
ow
t
ha
t
w
h
il
e
de
e
p
e
r
a
r
c
h
i
te
c
t
ur
e
s
li
ke
D
e
ns
e
N
e
t
a
nd
V
G
G
16
o
f
f
e
r
h
ig
he
r
a
c
c
u
r
a
c
y
a
n
d
r
o
bus
tn
e
s
s
,
li
ght
w
e
i
ght
m
ode
ls
s
uc
h
a
s
M
ob
il
e
N
e
t
V
2
p
r
ov
id
e
a
m
o
r
e
c
om
pu
ta
t
io
na
l
ly
e
f
f
ic
i
e
nt
s
o
lu
ti
on,
m
a
ki
ng
t
he
m
s
u
it
a
bl
e
f
o
r
r
e
s
o
ur
c
e
-
c
ons
tr
a
i
ne
d
e
n
vi
r
o
nm
e
nt
s
.
M
or
e
ove
r
,
da
t
a
a
ug
m
e
nt
a
ti
on
te
c
h
ni
que
s
a
r
e
e
s
s
e
nt
ia
l
f
or
i
m
p
r
ov
in
g
t
he
ge
ne
r
a
l
iz
a
ti
on
a
bi
li
ty
o
f
C
N
N
m
ode
ls
,
t
he
r
e
b
y
e
nha
nc
i
ng
t
he
i
r
e
f
f
e
c
ti
ve
ne
s
s
i
n
r
e
a
l
-
w
o
r
ld
a
ppl
ic
a
ti
ons
.
T
he
r
e
s
ul
ts
o
f
th
is
s
tu
d
y
s
upp
o
r
t
th
e
a
dva
nc
e
m
e
n
t
of
in
te
ll
ig
e
nt
f
ood
p
r
oc
e
s
s
in
g
s
ys
te
m
s
th
a
t
c
a
n
a
c
c
u
r
a
te
ly
c
la
s
s
i
f
y
pou
l
tr
y
m
e
a
t
,
c
ont
r
ib
ut
in
g
t
o
i
m
p
r
ov
e
d
f
oo
d
s
a
f
e
ty
a
n
d
qua
li
ty
a
s
s
u
r
a
nc
e
w
i
th
in
t
he
in
dus
tr
y.
M
ov
in
g
f
or
w
a
r
d,
f
u
r
t
he
r
r
e
s
e
a
r
c
h
is
w
a
r
r
a
n
te
d
t
o
e
xp
lo
r
e
a
d
va
n
c
e
d
t
e
c
hn
iq
ue
s
s
uc
h
a
s
t
r
a
ns
f
e
r
le
a
r
n
in
g,
d
om
a
in
a
da
p
ta
t
io
n,
a
nd
e
ns
e
m
bl
e
m
e
t
hods
t
o
e
nha
nc
e
th
e
pe
r
f
o
r
m
a
nc
e
a
nd
s
c
a
la
bi
li
ty
o
f
C
N
N
m
od
e
ls
f
or
f
oo
d
c
la
s
s
i
f
ic
a
t
io
n
ta
s
ks
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
hi
s
r
e
s
e
a
r
c
h
w
a
s
s
uppor
te
d
by
th
e
S
pe
c
tr
om
e
tr
y,
M
a
te
r
ia
ls
a
nd
A
r
c
he
om
a
te
r
ia
ls
L
a
bor
a
to
r
y
(
L
A
S
M
A
R
)
, F
a
c
ul
ty
of
S
c
ie
nc
e
s
, M
oul
a
y I
s
m
a
il
U
ni
ve
r
s
it
y,
M
e
kne
s
, M
or
oc
c
o.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
S
a
lm
a
S
e
khr
a
✓
✓
✓
✓
✓
✓
✓
✓
M
oha
m
m
e
d H
a
bi
b
✓
✓
✓
✓
✓
A
di
l
T
a
nnouc
he
✓
✓
✓
✓
✓
Y
ous
s
e
f
O
une
jj
a
r
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
T
he
a
ut
hor
s
de
c
la
r
e
t
ha
t
th
e
r
e
a
r
e
no c
onf
li
c
ts
of
i
nt
e
r
e
s
t
r
e
ga
r
d
in
g t
he
publi
c
a
ti
on of
t
hi
s
a
r
ti
c
le
.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
ta
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
o
pe
nl
y
a
va
il
a
bl
e
a
t
G
oogl
e
D
r
iv
e
:
ht
tp
s
:/
/d
r
iv
e
.google
.c
om
/d
r
iv
e
/f
ol
de
r
s
/1
js
8X
q0K
20y3vG
iq
W
1y
P
5oT
yY
X
c
K
F
21gk?
us
p
=
dr
iv
e
_l
in
k.
R
E
F
E
R
E
N
C
E
S
[
1]
S
.
S
a
l
m
a
,
M
.
H
a
bi
b,
A
.
T
a
nnouc
he
,
a
nd
Y
.
O
une
j
j
a
r
,
“
P
oul
t
r
y
m
e
a
t
c
l
a
s
s
i
f
i
c
a
t
i
on
us
i
ng
M
obi
l
e
N
e
t
V
2
pr
e
t
r
a
i
ne
d
m
ode
l
,”
R
e
v
u
e
d’
I
nt
e
l
l
i
ge
nc
e
A
r
t
i
f
i
c
i
e
l
l
e
, vol
. 37, no. 2, pp. 275
–
280, A
pr
. 2023, doi
:
10.18280
/
r
i
a
.370204.
[
2]
C
a
l
vi
n,
G
.
B
.
P
ut
r
a
,
a
nd
E
.
P
r
a
ka
s
a
,
“
C
l
a
s
s
i
f
i
c
a
t
i
on
of
c
hi
c
ke
n
m
e
a
t
f
r
e
s
hne
s
s
us
i
ng
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
k
a
l
gor
i
t
hm
s
,”
i
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
5
,
O
c
to
be
r
20
25
:
3715
-
3723
3722
2020
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
I
nnov
at
i
on
and
I
nt
e
l
l
i
ge
nc
e
f
or
I
nf
or
m
at
i
c
s
,
C
om
put
i
ng
and
T
e
c
hnol
ogi
e
s
,
3I
C
T
2020
,
D
e
c
. 2020, pp. 1
–
6
,
doi
:
10.1109/
3I
C
T
51146.2020.9312018.
[
3]
N
.
S
.
N
.
A
bd
A
z
i
z
,
S
.
M
ohd
D
a
ud,
R
.
A
.
D
z
i
ya
uddi
n,
M
.
Z
.
A
da
m
,
a
nd
A
.
A
z
i
z
a
n,
“
A
r
e
vi
e
w
on
c
om
put
e
r
vi
s
i
on
t
e
c
hnol
ogy
f
or
m
oni
t
or
i
ng
poul
t
r
y
f
a
r
m
-
a
ppl
i
c
a
t
i
on,
ha
r
dw
a
r
e
,
a
nd
s
of
t
w
a
r
e
,”
I
E
E
E
A
c
c
e
s
s
,
vol
.
9,
pp.
12431
–
12445,
2021,
doi
:
10.1109/
A
C
C
E
S
S
.2020.3047818.
[
4]
H
.
M
oha
m
m
e
d,
A
.
T
a
nnouc
he
,
a
nd
Y
.
O
un
e
j
j
a
r
,
“
W
e
e
d
de
t
e
c
t
i
on
i
n
pe
a
c
ul
t
i
va
t
i
on
w
i
t
h
t
he
f
a
s
t
e
r
R
C
N
N
R
e
s
N
e
t
5
0
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
k,”
R
e
v
ue
d’
I
nt
e
l
l
i
ge
nc
e
A
r
t
i
f
i
c
i
e
l
l
e
, vol
. 36, no. 1, pp. 13
–
18, F
e
b. 2022, doi
:
10.18280/
r
i
a
.360102.
[
5]
A
.
T
a
nnouc
he
,
K
.
S
ba
i
,
M
.
R
a
hm
oune
,
R
.
A
gounoune
,
a
nd
A
.
R
a
hm
a
ni
,
“
R
e
a
l
t
i
m
e
w
e
e
d
de
t
e
c
t
i
on
us
i
ng
a
boos
t
e
d
c
a
s
c
a
de
of
s
i
m
pl
e
f
e
a
t
ur
e
s
,”
I
nt
e
r
nat
i
onal
J
ou
r
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
6,
no.
6,
pp.
2755
–
2765,
D
e
c
.
2016,
doi
:
10.11591/
i
j
e
c
e
.v6i
6.11878.
[
6]
M
.
H
a
bi
b,
S
.
S
e
khr
a
,
A
.
T
a
nnouc
he
,
a
nd
Y
.
O
une
j
j
a
r
,
“
T
he
i
de
nt
i
f
i
c
a
t
i
on
o
f
w
e
e
ds
a
nd
c
r
ops
us
i
ng
t
he
popul
a
r
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
ks
,”
i
n
D
i
gi
t
al
T
e
c
hnol
ogi
e
s
and
A
ppl
i
c
at
i
ons
.
C
h
a
m
,
S
w
i
t
z
e
r
l
a
nd:
S
pr
i
nge
r
,
2023,
pp.
484
–
493,
doi
:
10.1007/
978
-
3
-
031
-
29857
-
8_49.
[
7]
A
.
T
a
nnouc
he
,
A
.
G
a
ga
,
M
.
B
out
a
l
l
i
ne
,
a
nd
S
.
B
e
l
houi
de
g,
“
W
e
e
d
s
de
t
e
c
t
i
on
e
f
f
i
c
i
e
nc
y
t
hr
ough
di
f
f
e
r
e
nt
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
ks
t
e
c
hnol
ogy,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
12,
no.
1,
pp.
1048
–
1055,
F
e
b.
2022,
doi
:
10.11591/
i
j
e
c
e
.v12i
1.pp1048
-
1055.
[
8]
D
.
S
e
o
e
t
al
.
,
“
I
de
nt
i
f
i
c
a
t
i
on
of
t
a
r
ge
t
c
hi
c
ke
n
popul
a
t
i
on
s
by
m
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
s
u
s
i
ng
t
he
m
i
ni
m
um
num
be
r
of
S
N
P
s
,
”
A
ni
m
al
s
, vol
. 11, no. 1, pp. 1
–
18, J
a
n. 2021, doi
:
10.3390/
a
ni
11010241.
[
9]
I
.
T
om
a
s
e
vi
c
e
t
al
.
,
“
E
va
l
ua
t
i
on
of
pou
l
t
r
y
m
e
a
t
c
ol
our
us
i
ng
c
om
put
e
r
vi
s
i
on
s
ys
t
e
m
a
nd
c
ol
our
i
m
e
t
e
r
:
i
s
t
he
r
e
a
di
f
f
e
r
e
nc
e
?
,”
B
r
i
t
i
s
h F
ood J
our
nal
, vol
. 121, no. 5, pp. 1078
–
1087, J
un. 2019, doi
:
10.1108/
B
F
J
-
06
-
2018
-
0376.
[
10]
C
. Y
e
e
t
al
.
,
“
A
n e
xpe
r
i
m
e
nt
a
l
s
t
udy of
s
t
unne
d
s
t
a
t
e
de
t
e
c
t
i
on f
or
br
oi
l
e
r
c
hi
c
ke
ns
us
i
ng
a
n i
m
pr
ove
d c
onvol
ut
i
on ne
ur
a
l
ne
t
w
or
k
a
l
gor
i
t
hm
,”
C
om
put
e
r
s
and E
l
e
c
t
r
oni
c
s
i
n A
gr
i
c
ul
t
ur
e
, vol
. 170,
M
a
r
. 2020, doi
:
10.1016/
j
.c
om
pa
g.2020.105284.
[
11]
N
.
D
a
r
a
pa
ne
ni
,
B
.
K
r
i
s
hna
m
ur
t
hy,
a
nd
A
.
R
.
P
a
dur
i
,
“
C
onvol
ut
i
on
ne
ur
a
l
ne
t
w
or
ks
:
a
c
om
pa
r
a
t
i
ve
s
t
udy
f
or
i
m
a
ge
c
l
a
s
s
i
f
i
c
a
t
i
on,”
i
n
2020
I
E
E
E
15t
h
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
I
ndus
t
r
i
al
an
d
I
nf
or
m
at
i
on
Sy
s
t
e
m
s
,
I
C
I
I
S
2020
-
P
r
oc
e
e
di
ngs
,
N
ov. 2020, pp. 327
–
332
,
doi
:
10.1109/
I
C
I
I
S
51140.2020.9342667.
[
12]
M
.
S
.
D
e
oul
a
e
t
al
.
,
“
C
ons
um
pt
i
on
of
m
e
a
t
,
t
r
a
di
t
i
ona
l
a
nd
m
od
e
r
n
pr
oc
e
s
s
e
d
m
e
a
t
a
nd
c
ol
or
e
c
t
a
l
c
a
n
c
e
r
r
i
s
k
a
m
ong
t
he
M
or
oc
c
a
n
popul
a
t
i
on:
a
l
a
r
ge
-
s
c
a
l
e
c
a
s
e
–
c
ont
r
ol
s
t
udy,”
I
nt
e
r
nat
i
onal
J
our
nal
of
C
anc
e
r
,
vol
.
146,
no.
5,
pp.
1333
–
1345,
M
a
r
. 2020, doi
:
10.1002/
i
j
c
.32689.
[
13]
N
. C
ohe
n, H
.
E
nna
j
i
, B
.
B
ouc
hr
i
f
, M
. H
a
s
s
a
r
, a
nd H
.
K
a
r
i
b, “
C
om
p
a
r
a
t
i
ve
s
t
ud
y of
m
i
c
r
obi
ol
ogi
c
a
l
qua
l
i
t
y of
r
a
w
poul
t
r
y m
e
a
t
a
t
va
r
i
ous
s
e
a
s
on
s
a
nd
f
or
di
f
f
e
r
e
nt
s
l
a
ught
e
r
i
ng
pr
oc
e
s
s
e
s
i
n
C
a
s
a
bl
a
nc
a
(
M
or
oc
c
o)
,”
J
our
nal
of
A
ppl
i
e
d
P
oul
t
r
y
R
e
s
e
ar
c
h
,
vol
.
16,
no. 4, pp. 502
–
508, D
e
c
. 2007, doi
:
10.3382/
j
a
pr
.2006
-
00061.
[
14]
X
. F
u a
nd
J
. C
he
n,
“
A
r
e
vi
e
w
of
hype
r
s
pe
c
t
r
a
l
i
m
a
gi
ng f
or
c
hi
c
ke
n
m
e
a
t
s
a
f
e
t
y
a
nd qua
l
i
t
y
e
va
l
ua
t
i
on:
a
ppl
i
c
a
t
i
on, ha
r
dw
a
r
e
, a
n
d
s
of
t
w
a
r
e
,”
C
om
pr
e
h
e
ns
i
v
e
R
e
v
i
e
w
s
i
n
F
ood
Sc
i
e
nc
e
and
F
ood
Saf
e
t
y
,
vol
.
18,
no.
2,
pp.
535
–
547,
M
a
r
.
2019,
doi
:
10.1111/
1541
-
4337.12428.
[
15]
I
.
T
om
a
s
e
vi
c
e
t
al
.
,
“
C
om
pa
r
i
s
on
of
a
c
om
put
e
r
vi
s
i
on
s
ys
t
e
m
v
s
.
t
r
a
di
t
i
ona
l
c
ol
or
i
m
e
t
e
r
f
or
c
ol
or
e
va
l
ua
t
i
on
of
m
e
a
t
pr
oduc
t
s
w
i
t
h va
r
i
ous
phys
i
c
a
l
pr
ope
r
t
i
e
s
,”
M
e
at
Sc
i
e
nc
e
, vol
. 148, pp. 5
–
12, F
e
b. 2019,
doi
:
10.1016/
j
.m
e
a
t
s
c
i
.2018.09.015.
[
16]
R
.
M
a
l
hot
r
a
a
nd
P
.
S
i
ngh,
“
R
e
c
e
nt
a
dva
nc
e
s
i
n
de
e
p
l
e
a
r
ni
ng
m
ode
l
s
:
a
s
y
s
t
e
m
a
t
i
c
l
i
t
e
r
a
t
ur
e
r
e
vi
e
w
,”
M
ul
t
i
m
e
di
a
T
ool
s
and
A
ppl
i
c
at
i
ons
, vol
. 82, no. 29, pp. 44977
–
45060, D
e
c
. 2023, doi
:
10.1007/
s
11042
-
023
-
15295
-
z.
[
17]
M
.
S
ha
f
i
q
a
nd
Z
.
G
u,
“
D
e
e
p
r
e
s
i
dua
l
l
e
a
r
ni
ng
f
or
i
m
a
ge
r
e
c
ogni
t
i
on
:
a
s
ur
ve
y,”
A
ppl
i
e
d
Sc
i
e
nc
e
s
,
vol
.
12,
no.
18,
S
e
p.
2022,
doi
:
10.3390/
a
pp12188972.
[
18]
D
.
T
he
c
ke
da
t
h
a
nd
R
.
R
.
S
e
da
m
ka
r
,
“
D
e
t
e
c
t
i
ng
a
f
f
e
c
t
s
t
a
t
e
s
us
i
ng
V
G
G
16,
R
e
s
N
e
t
50
a
nd
S
E
-
R
e
s
N
e
t
50
n
e
t
w
or
ks
,”
SN
C
om
put
e
r
Sc
i
e
nc
e
, vol
. 1, no. 2, 2020, doi
:
10.1007/
s
42979
-
020
-
0114
-
9.
[
19]
H
.
A
.
S
a
be
r
,
A
.
Y
oune
s
,
M
.
O
s
m
a
n,
a
nd
I
.
E
l
ka
ba
ni
,
“
Q
ur
a
n
r
e
c
i
t
e
r
i
de
nt
i
f
i
c
a
t
i
on
us
i
ng
N
A
S
N
e
t
L
a
r
ge
,”
N
e
u
r
al
C
om
put
i
ng
an
d
A
ppl
i
c
at
i
ons
, vol
. 36, no. 12, pp. 6559
–
6573, 2024, doi
:
10.1007/
s
00521
-
023
-
0
9392
-
1.
[
20]
L
.
Z
ha
ng,
J
.
W
a
ng,
B
.
L
i
,
Y
.
L
i
u,
H
.
Z
ha
ng,
a
nd
Q
.
D
u
a
n,
“
A
M
obi
l
e
N
e
t
V
2
-
S
E
N
e
t
-
ba
s
e
d
m
e
t
hod
f
or
i
de
nt
i
f
yi
ng
f
i
s
h
s
c
hool
f
e
e
di
ng be
ha
vi
or
,”
A
quac
ul
t
ur
al
E
ngi
ne
e
r
i
ng
, vol
. 99, N
ov. 2022, doi
:
10.1016/
j
.a
qua
e
ng.2022.102288.
[
21]
S
. E
l
-
K
ha
m
y, A
. A
l
-
K
a
bba
ny, a
nd S
. E
l
-
B
a
na
, “
G
oi
ng
s
ha
l
l
ow
e
r
w
i
t
h
M
obi
l
e
N
e
t
s
:
on t
he
i
m
pa
c
t
of
w
a
ve
l
e
t
pool
i
n
g,”
i
n
N
at
i
onal
R
adi
o Sc
i
e
nc
e
C
onf
e
r
e
nc
e
, N
R
SC
, P
r
oc
e
e
di
ng
s
, J
ul
. 2021, pp. 126
–
138
,
doi
:
10.1109/
N
R
S
C
52299.2021.9509825.
[
22]
D
.
S
i
nha
a
nd
M
.
E
l
-
S
ha
r
ka
w
y,
“
T
hi
n
M
obi
l
e
N
e
t
:
a
n
e
nha
nc
e
d
M
obi
l
e
N
e
t
a
r
c
hi
t
e
c
t
ur
e
,”
i
n
2019
I
E
E
E
10t
h
A
nnual
U
bi
qui
t
ous
C
om
put
i
ng,
E
l
e
c
t
r
oni
c
s
and
M
obi
l
e
C
om
m
uni
c
at
i
on
C
onf
e
r
e
n
c
e
,
U
E
M
C
O
N
2019
,
O
c
t
.
2019,
pp.
0280
–
0285
,
doi
:
10.1109/
U
E
M
C
O
N
47517.2019.8993089.
[
23]
T
.
C
ha
uha
n,
H
.
P
a
l
i
ve
l
a
,
a
nd
S
.
T
i
w
a
r
i
,
“
O
pt
i
m
i
z
a
t
i
on
a
nd
f
i
ne
-
t
uni
ng
o
f
D
e
n
s
e
N
e
t
m
ode
l
f
or
c
l
a
s
s
i
f
i
c
a
t
i
on
of
C
O
V
I
D
-
19
c
a
s
e
s
i
n
m
e
di
c
a
l
i
m
a
gi
ng,”
I
nt
e
r
nat
i
onal
J
our
nal
of
I
nf
or
m
at
i
on
M
anage
m
e
nt
D
at
a
I
ns
i
ght
s
,
vol
.
1,
no.
2,
N
ov.
2021,
doi
:
10.1016/
j
.j
j
i
m
e
i
.2021.100020.
[
24]
M
.
T
a
n
a
nd
Q
.
V
.
L
e
,
“
E
f
f
i
c
i
e
nt
N
e
t
V
2:
s
m
a
l
l
e
r
m
ode
l
s
a
nd
f
a
s
t
e
r
t
r
a
i
ni
n
g,”
P
r
oc
e
e
di
ngs
of
M
ac
hi
ne
L
e
ar
ni
ng
R
e
s
e
ar
c
h
,
vol
.
139, pp.
10096
–
10106, 2021.
[
25]
S
.
B
.
A
kba
r
,
K
.
T
ha
nupi
l
l
a
i
,
a
nd
S
.
S
unda
r
a
r
a
j
,
“
C
om
bi
ni
ng
t
he
a
dva
nt
a
ge
s
of
A
l
e
xN
e
t
c
onvol
ut
i
ona
l
de
e
p
ne
ur
a
l
n
e
t
w
or
k
opt
i
m
i
z
e
d
w
i
t
h
a
nophe
l
e
s
s
e
a
r
c
h
a
l
gor
i
t
hm
ba
s
e
d
f
e
a
t
ur
e
e
xt
r
a
c
t
i
on
a
nd
r
a
nd
om
f
or
e
s
t
c
l
a
s
s
i
f
i
e
r
f
or
C
O
V
I
D
-
19
c
l
a
s
s
i
f
i
c
a
t
i
on,”
C
onc
ur
r
e
nc
y
and C
o
m
put
at
i
on:
P
r
ac
t
i
c
e
and E
x
pe
r
i
e
nc
e
, vol
. 34, no. 15, J
ul
. 2022, doi
:
10.1002/
c
pe
.6958.
[
26]
X
.
Z
ha
ng,
N
.
H
a
n,
a
nd
J
.
Z
ha
ng,
“
C
om
pa
r
a
t
i
ve
a
na
l
ys
i
s
of
V
G
G
,
R
e
s
N
e
t
,
a
nd
G
oogL
e
N
e
t
a
r
c
hi
t
e
c
t
ur
e
s
e
va
l
ua
t
i
ng
p
e
r
f
or
m
a
nc
e
,
c
om
put
a
t
i
ona
l
e
f
f
i
c
i
e
nc
y,
a
nd
c
onve
r
ge
nc
e
r
a
t
e
s
,”
A
ppl
i
e
d
and
C
o
m
put
at
i
onal
E
ngi
ne
e
r
i
ng
,
vol
.
44,
no.
1,
pp.
172
–
181
,
M
a
r
. 2024, doi
:
10.54254/
2755
-
2721/
44/
20230676.
[
27]
D
.
K
i
m
,
B
.
H
e
o,
a
nd
D
.
H
a
n,
“
D
e
ns
e
N
e
t
s
r
e
l
oa
d
e
d:
pa
r
a
di
gm
s
hi
f
t
be
yond
R
e
s
N
e
t
s
a
nd
V
i
T
s
,”
i
n
C
o
m
put
e
r
S
c
i
e
nc
e
.
C
ha
m
,
S
w
i
t
z
e
r
l
a
nd:
S
pr
i
nge
r
,
pp. 395
–
415, 2025, doi
:
10.1007/
978
-
3
-
031
-
72646
-
0_23
.
[
28]
R
.
R
a
nf
t
l
,
A
.
B
oc
hkovs
ki
y,
a
nd
V
.
K
ol
t
un,
“
V
i
s
i
on
t
r
a
ns
f
or
m
e
r
s
f
or
de
ns
e
pr
e
di
c
t
i
o
n,”
i
n
P
r
oc
e
e
di
ngs
of
t
he
I
E
E
E
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on C
om
put
e
r
V
i
s
i
on
, O
c
t
. 2021, pp. 12159
–
12168
,
doi
:
10.1109/
I
C
C
V
48922.2021.01196.
[
29]
N
.
Z
a
ka
r
i
a
,
F
.
M
oh
a
m
e
d,
R
.
A
bde
l
gha
ni
,
a
nd
K
.
S
unda
r
a
j
,
“
V
G
G
16,
R
e
s
N
e
t
-
50,
a
nd
G
oog
L
e
N
e
t
de
e
p
l
e
a
r
ni
ng
a
r
c
hi
t
e
c
t
ur
e
f
or
br
e
a
t
hi
ng
s
ound
c
l
a
s
s
i
f
i
c
a
t
i
on:
a
c
om
pa
r
a
t
i
ve
s
t
udy,”
i
n
2021
I
nt
e
r
nat
i
onal
C
onf
e
r
e
n
c
e
on
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
f
or
C
y
be
r
Se
c
ur
i
t
y
Sy
s
t
e
m
s
and P
r
i
v
ac
y
(
A
I
-
C
SP
)
, N
ov. 2021, pp. 1
–
6
,
doi
:
10.1109/
A
I
-
C
S
P
52968.2021.9671124.
[
30]
F
.
S
a
xe
n,
P
.
W
e
r
ne
r
,
S
.
H
a
ndr
i
c
h,
E
.
O
t
hm
a
n,
L
.
D
i
nge
s
,
a
nd
A
.
A
l
-
H
a
m
a
d
i
,
“
F
a
c
e
a
t
t
r
i
but
e
de
t
e
c
t
i
on
w
i
t
h
m
obi
l
e
ne
t
v2
a
nd
na
s
ne
t
-
m
obi
l
e
,”
i
n
I
nt
e
r
nat
i
onal
Sy
m
pos
i
um
on
I
m
age
and
Si
gnal
P
r
oc
e
s
s
i
ng
and
A
nal
y
s
i
s
,
I
SP
A
,
S
e
p.
2019,
pp.
176
–
18
,
doi
:
10.1109/
I
S
P
A
.2019.8868585.
[
31]
A
.
K
r
i
z
he
vs
ky,
I
.
S
ut
s
ke
ve
r
,
a
nd
G
.
E
.
H
i
nt
on,
“
I
m
a
ge
N
e
t
c
l
a
s
s
i
f
i
c
a
t
i
on
w
i
t
h
de
e
p
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
ks
,
”
C
om
m
uni
c
at
i
ons
of
t
he
A
C
M
, vol
. 60, no. 6, pp. 84
–
90, M
a
y 2017, doi
:
10.114
5/
3065386.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
C
om
par
at
iv
e
analy
s
is
of
c
onv
ol
ut
io
nal
ne
u
r
al
ne
tw
or
k
ar
c
hi
te
c
tu
r
e
s
f
or
poult
r
y
…
(
Sal
m
a
Se
k
h
r
a
)
3723
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Sekhra
Salma
is
a
Ph.D.
student
at
the
Spectrometry,
Materials
,
and
Archeomaterials
Laboratory
(LASMAR),
Faculty
of
Sciences,
Mo
ulay
Ismail
University,
Meknes,
Morocco.
Her
research
focuses
on
computer
vision,
deep
learning,
and
their
applicati
ons
in
the
agri
-
food
sector,
with
a
particular
focus
on
meat
au
thentication
and
poultry
meat class
ification.
She ca
n be c
ontact
ed at
email:
sekhrasalma3@gmail.com
.
Mohammed
Habib
is
currently
pursuing
a
Ph.D.
at
the
Spectrome
try,
Materials
,
and
Archeomaterials
Laboratory
(LASMAR),
Faculty
of
Sciences,
Moulay
Ismail
University,
Meknes,
Morocco.
His
main
researc
h
interests
include
computer
vis
ion,
deep
learning,
and
their applic
ation in agr
iculture.
He can be contacted at email:
moh.habib@edu.umi.ac.ma
.
Adil
Tannouche
is
a
faculty
member
at
the
Higher
School
of
Tech
nology,
Sultan
Moulay
Slimane
University,
Béni
Mellal,
Morocco.
He
earned
his
P
h.D.
in
Electronics
and
Embedded
Systems
from
Moulay
Ismail
University,
Meknes.
He
is
a
member
of
the
Laboratory
of
Engineering
and
Applied
Technologies.
His
research
areas
include
machine
vision,
artificial
intelligence,
and
their
applications
in
precision
agric
ulture
and
agro
-
industry.
He can be contacted at email:
tannouche
@
gmail.com
.
Youssef
Ounejjar
earned
his
B.
E
ng.
and
M.S.
degrees
in
Electric
al
Engineering
from
the
Ecole
Nationale
d'
Ingénieurs
de
Sfax,
Tunisia,
in
1996
and
1998,
respectively.
He
completed
his
Ph.D.
in
Electrical
Engineeri
ng
at
the
École
de
Technolog
ie
Supérieur
e,
Montréal,
Canada,
in
2011.
Currently,
he
is
an
Associate
Profe
ssor
at
Moulay
Ismai
l
University, Meknes
, with research interests
focused on
multilevel
pow
er converters.
He can be
contacted
at email
:
ounejjar@
gmail.com
.
Evaluation Warning : The document was created with Spire.PDF for Python.