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
.
13
, N
o.
4
,
D
e
c
e
m
be
r
2024
, pp.
4241
~
4248
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
13
.i
4
.pp
4241
-
4248
4241
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
Im
age
an
al
ysi
s f
or
c
l
ass
i
f
yi
n
g c
of
f
e
e
b
e
an
q
u
al
i
t
y u
si
n
g a
m
u
l
t
i
-
f
e
at
u
r
e
an
d
m
ac
h
i
n
e
l
e
ar
n
i
n
g ap
p
r
oac
h
A
n
in
d
it
a S
e
p
t
ia
r
in
i
1
, H
am
d
an
i
H
am
d
an
i
1
, A
j
i
E
r
y B
u
r
h
an
d
e
n
y
2
, D
am
ar
N
u
r
c
ah
yon
o
3
,
S
u
r
ya E
k
a P
r
iy
at
n
a
4
1
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
c
s
, F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng, M
ul
a
w
a
r
m
a
n U
ni
ve
r
s
i
t
y, S
a
m
a
r
i
nda
, I
ndone
s
i
a
2
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
E
ngi
ne
e
r
i
ng,
F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng, M
ul
a
w
a
r
m
a
n
U
ni
ve
r
s
i
t
y, S
a
m
a
r
i
nda
, I
ndone
s
i
a
3
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
on
T
e
c
hnol
ogy, P
ol
i
t
e
kni
k N
e
ge
r
i
S
a
m
a
r
i
nda
, S
a
m
a
r
i
nda
, I
ndone
s
i
a
4
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
on
T
e
c
hnol
ogy,
F
a
c
ul
t
y of
D
a
'
w
a
h a
nd C
om
m
uni
c
a
t
i
on S
c
i
e
nc
e
s
,
A
nt
a
s
a
r
i
S
t
a
t
e
I
s
l
a
m
i
c
U
ni
ve
r
s
i
t
y,
B
a
nj
a
r
m
a
s
i
n, I
ndone
s
i
a
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
N
ov 27, 2023
R
e
vi
s
e
d
F
e
b 11, 2024
A
c
c
e
pt
e
d
F
e
b 28, 2024
Price
and
customer
satisfaction
depend
on
coffee
bean
quality.
The
coffee
industry
must
analyze
coffee
bean
qu
ality.
Global
demand
for
robusta
coffee
is
high.
Coffee
industry
professionals
mostly
understand
coffee
bean
q
uality.
Thus,
an
image
analysis
using
a
computer
vision
-
based
approach
for
classify
ing
robusta
coffee
bean
quality
is
required.
Image
acquisition,
region
of
interest
(ROI)
detection,
pre
-
processing,
segmentation,
f
eature
extr
action,
feature selection
, and classi
fication
are covered in
this s
tudy.
A multi
-
feature
derived
based
on
color,
shape,
and
texture
features
was
employed
in
feature
extractio
n,
followed
by
feature
selectio
n
using
principal
component
a
nalysis
(
PCA
)
.
Several
machine
-
learning
methods
classified
the
coff
ee
bean
s.
The
method
performance
was
assessed
using
precision,
recall,
and
accurac
y.
The
selected
features
using
the
backpropaga
tion
neural
network
(
BPNN)
classifi
er outperfo
rmed oth
ers with
98.54%
accuracy.
K
e
y
w
o
r
d
s
:
C
of
f
e
e
be
a
ns
F
e
a
tu
r
e
s
s
e
le
c
ti
on
K
-
m
e
a
ns
M
a
c
hi
ne
l
e
a
r
ni
ng
P
r
in
c
ip
a
l
c
om
pone
nt
a
na
ly
s
is
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
:
A
ni
ndi
ta
S
e
pt
ia
r
in
i
D
e
pa
r
tm
e
nt
of
I
nf
or
m
a
ti
c
s
,
F
a
c
ul
ty
of
E
ngi
ne
e
r
in
g, M
ul
a
w
a
r
m
a
n U
ni
ve
r
s
it
y
St
. S
a
m
ba
li
ung
,
N
o. 9, S
a
m
a
r
in
da
, I
ndone
s
ia
E
m
a
il
:
a
ni
ndi
ta
@
unm
ul
.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
he
ut
il
iz
a
ti
on
of
c
om
put
e
r
s
a
nd
a
s
s
oc
i
a
te
d
te
c
hnol
ogi
e
s
is
s
e
e
in
g
f
a
s
t
e
xpa
n
s
io
n
a
nd
di
ve
r
s
if
ic
a
ti
on.
T
he
a
ppl
ic
a
ti
on
of
th
is
i
s
be
in
g
obs
e
r
ve
d
in
th
e
f
ie
ld
of
a
gr
ic
ul
tu
r
e
.
T
he
r
e
e
xi
s
t
m
ul
ti
pl
e
in
s
ta
nc
e
s
w
he
r
e
in
c
om
put
e
r
s
ha
v
e
be
e
n
e
m
pl
oye
d
in
th
e
a
gr
ic
u
lt
ur
a
l
s
e
c
to
r
,
e
nc
om
p
a
s
s
in
g
th
e
m
oni
to
r
in
g
of
f
r
ui
t
r
ip
e
ne
s
s
[
1]
,
[
2
]
,
la
nd
m
a
na
ge
m
e
nt
[
3
]
,
a
nd
pl
a
nt
de
ve
l
opm
e
nt
[
4
]
,
[
5
]
.
C
of
f
e
e
,
a
s
one
of
th
e
m
os
t
w
id
e
ly
c
ons
um
e
d
be
ve
r
a
ge
s
gl
oba
ll
y,
hol
ds
s
ig
ni
f
ic
a
nt
im
por
ta
nc
e
a
s
a
n
e
c
onomi
c
c
om
m
odi
ty
.
T
he
gl
oba
l
popula
r
it
y
of
c
of
f
e
e
c
a
n
be
a
tt
r
ib
ut
e
d
to
it
s
s
ti
m
ul
a
ti
ng
pr
ope
r
ti
e
s
a
nd
th
e
pr
e
f
e
r
e
nc
e
f
or
it
s
bi
tt
e
r
f
la
vor
.
C
of
f
e
e
s
e
r
ve
s
a
s
a
s
ubs
ta
nt
ia
l
pr
ovi
de
r
of
c
a
f
f
e
in
e
f
or
a
c
ons
id
e
r
a
bl
e
num
be
r
of
in
di
vi
dua
ls
.
W
hi
le
pr
e
vi
ous
r
e
s
e
a
r
c
h
ha
s
e
s
ta
bl
is
he
d
a
c
onne
c
ti
on
b
e
twe
e
n
c
of
f
e
e
a
nd
c
a
f
f
e
in
e
in
ta
ke
a
nd
a
dv
e
r
s
e
he
a
lt
h
e
f
f
e
c
ts
,
r
e
c
e
nt
s
tu
di
e
s
ha
ve
pr
e
s
e
nt
e
d e
vi
de
nc
e
s
ugge
s
ti
ng t
ha
t
th
e
c
om
pounds
f
ound in c
of
f
e
e
, s
uc
h a
s
c
a
f
f
e
in
e
, c
hl
or
oge
ni
c
a
c
id
s
,
ka
hw
e
ol
,
c
a
f
e
s
to
l,
a
nd
v
a
r
io
us
m
ic
r
onut
r
ie
nt
s
(
s
uc
h
a
s
m
a
gne
s
iu
m
,
pot
a
s
s
iu
m
,
a
nd
phos
phor
us
)
,
m
a
y
e
nha
nc
e
th
e
im
m
une
s
y
s
te
m
a
nd
pr
ovi
de
pr
ot
e
c
ti
on
a
ga
in
s
t
th
e
de
ve
lo
pm
e
nt
of
c
ondi
ti
ons
s
uc
h
a
s
obe
s
it
y,
di
a
be
te
s
, ne
ur
ol
ogi
c
a
l
di
s
e
a
s
e
s
, os
t
e
opor
os
is
, a
nd pa
n
c
r
e
a
ti
c
c
a
nc
e
r
[
6
]
.
T
he
c
of
f
e
e
in
dus
tr
y
va
lu
e
s
qua
li
ty
be
c
a
u
s
e
of
th
e
r
e
la
ti
ons
hi
p
be
twe
e
n
c
of
f
e
e
be
a
n
s
c
a
r
c
it
y,
m
one
ta
r
y
c
om
pe
ns
a
ti
on,
a
nd
c
on
s
um
e
r
ha
ppi
ne
s
s
.
R
obu
s
ta
c
of
f
e
e
be
a
ns
,
w
id
e
ly
gr
ow
n,
ha
v
e
a
di
s
ti
nc
t
ta
s
te
a
nd
a
r
om
a
.
Q
ua
li
ty
of
r
obus
ta
c
of
f
e
e
be
a
ns
de
pe
nds
on
s
oi
l
m
a
ke
up,
c
li
m
a
te
,
a
nd
pr
oc
e
s
s
in
g
m
e
th
od.
C
of
f
e
e
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
.
13
,
N
o.
4
,
D
e
c
e
m
be
r
2024
:
4241
-
4248
4242
pr
ic
e
s
de
pe
nd
on
be
a
n
qua
li
ty
.
I
t
is
c
r
uc
ia
l
to
not
e
th
a
t
not
a
ll
gr
ow
e
r
s
a
nd
c
of
f
e
e
s
hop
ow
n
e
r
s
c
a
n
id
e
nt
if
y
c
of
f
e
e
be
a
n
qua
li
ty
.
T
hus
,
e
r
r
or
s
m
a
y
oc
c
ur
w
he
n
th
e
y
la
c
k
t
hi
s
e
xpe
r
ti
s
e
.
G
r
a
di
ng
is
ti
m
e
-
c
ons
um
in
g
a
nd
pr
oduc
e
s
in
c
ons
i
s
te
nt
out
c
om
e
s
.
D
ue
to
vi
s
ua
l
p
e
r
c
e
pt
io
n
li
m
it
s
,
f
a
ti
gue
,
a
nd
c
of
f
e
e
qua
li
ty
e
va
lu
a
ti
on
di
f
f
e
r
e
nc
e
s
,
th
e
s
e
in
c
on
s
is
te
n
c
ie
s
oc
c
ur
.
V
is
ua
l
c
h
a
r
a
c
te
r
is
ti
c
s
a
r
e
of
te
n
us
e
d
to
e
va
lu
a
te
r
obus
ta
c
of
f
e
e
be
a
ns
.
I
n
th
is
s
it
ua
ti
on,
c
om
put
e
r
vi
s
io
n
m
a
y
w
or
k.
I
t
e
xt
r
a
c
ts
r
obus
ta
c
of
f
e
e
be
a
n
vi
s
ua
l
tr
a
it
s
th
a
t
hi
ghl
y
pr
e
di
c
t
qua
li
ty
. C
ol
or
, f
or
m
, a
nd t
e
xt
ur
e
m
a
y be
ne
e
de
d f
or
t
hi
s
pr
oc
e
dur
e
.
N
um
e
r
o
u
s
r
e
s
e
a
r
c
h i
nv
e
s
ti
ga
ti
o
ns
h
a
v
e
b
e
e
n
c
o
nd
uc
te
d i
n c
om
p
ut
e
r
vi
s
i
on
, f
o
c
u
s
in
g
on
t
h
e
a
p
pl
i
c
a
ti
on
of
f
oo
d
pr
o
c
e
s
s
in
g
.
T
h
e
s
e
s
t
udi
e
s
e
nc
om
p
a
s
s
a
r
a
n
ge
of
f
oo
d
it
e
m
s
, s
uc
h
a
s
b
a
n
a
n
a
[
7]
,
h
on
e
y
[
8]
,
d
a
t
e
f
r
ui
t
[
9]
,
pa
l
m
oi
l
[
10]
,
a
nd c
of
f
e
e
[
1
1]
,
[
12]
. T
he
c
o
n
s
tr
u
c
t
io
n of
t
hi
s
s
y
s
t
e
m
i
nvo
lv
e
s
s
e
v
e
r
a
l
g
e
n
e
r
a
l
p
r
o
c
e
s
s
e
s
,
n
a
m
e
ly
pr
e
-
pr
o
c
e
s
s
i
ng
,
s
e
gm
e
n
ta
t
io
n,
f
e
a
tu
r
e
e
xt
r
a
c
ti
o
n,
a
nd
c
la
s
s
if
i
c
a
t
io
n
[
13
]
.
C
om
m
on
pr
e
-
pr
o
c
e
s
s
in
g
t
a
s
k
s
of
t
e
n
in
vo
lv
e
s
c
a
l
in
g
[
14]
a
nd
c
o
nv
e
r
ti
ng
c
ol
or
s
p
a
c
e
s
[
15]
.
T
h
e
O
t
s
u
th
r
e
s
hol
di
ng
m
e
t
hod
[
13]
,
K
-
m
e
a
ns
c
l
u
s
t
e
r
in
g
a
lg
or
it
hm
[
1
6]
,
a
nd
e
dg
e
de
t
e
c
ti
on
a
ppr
oa
c
h
[
1
7]
w
e
r
e
s
ub
s
e
q
ue
n
tl
y
e
m
p
lo
y
e
d
,
a
l
on
g
w
it
h
m
a
n
y
e
s
t
a
bl
i
s
h
e
d
s
e
gm
e
n
ta
t
io
n
m
e
th
od
ol
o
gi
e
s
.
T
he
e
xt
r
a
c
t
a
bl
e
f
e
a
t
ur
e
s
th
a
t
c
a
n
b
e
c
on
s
id
e
r
e
d
f
o
r
e
di
bl
e
s
e
nc
om
p
a
s
s
c
ol
or
[
10]
,
s
h
a
p
e
[
1
8]
,
a
nd
t
e
x
tu
r
e
[
1
1]
.
M
or
e
o
ve
r
,
n
a
ïv
e
B
a
y
e
s
(
N
B
)
[
1
0]
,
k
-
n
e
a
r
e
s
t
n
e
i
gh
bor
(
K
N
N
)
[
19]
,
a
nd
s
up
por
t
ve
c
to
r
m
a
c
hi
ne
s
(
S
V
M
)
[
10]
a
r
e
f
r
e
qu
e
n
tl
y
ut
il
i
z
e
d
in
th
e
c
l
a
s
s
if
i
c
a
ti
o
n p
r
o
c
e
s
s
.
R
e
c
e
nt
s
tu
di
e
s
ha
v
e
us
e
d
m
a
c
hi
ne
le
a
r
ni
ng
to
c
la
s
s
if
y
c
of
f
e
e
be
a
ns
a
c
r
os
s
a
gr
ic
ul
tu
r
a
l
s
it
ua
ti
ons
.
C
ol
or
a
nd
s
ha
pe
he
lp
e
d
id
e
nt
if
y
hi
gh
-
qua
li
ty
be
a
ns
.
T
he
in
ve
s
ti
ga
ti
on
us
e
d
im
a
ge
pr
oc
e
s
s
in
g
a
nd
m
a
c
hi
ne
le
a
r
ni
ng
on
a
n
A
r
dui
no
m
e
ga
boa
r
d.
E
s
s
e
nt
ia
l
c
r
it
e
r
ia
w
e
r
e
a
s
s
e
s
s
e
d
to
de
te
r
m
in
e
hi
gh
-
qua
li
ty
gr
e
e
n
c
of
f
e
e
be
a
ns
.
K
N
N
w
a
s
us
e
d
to
e
va
lu
a
t
e
c
of
f
e
e
be
a
ns
a
nd
c
la
s
s
if
y
t
he
m
by
de
f
e
c
t
ty
pe
.
L
ogi
c
,
im
a
g
e
pr
oc
e
s
s
in
g,
a
nd s
upe
r
vi
s
e
d l
e
a
r
ni
ng a
lg
or
it
hm
s
a
r
e
e
xe
c
ut
e
d a
nd c
od
e
d on the
A
r
dui
no boa
r
d. T
he
m
a
c
hi
ne
vi
s
io
n s
ys
te
m
ha
s
a
n
a
ve
r
a
ge
a
c
c
ur
a
c
y
of
94.79%
f
or
qua
li
ty
a
nd
95.78%
f
or
de
f
e
c
t
-
ty
pe
e
va
lu
a
ti
on.
H
o
w
e
ve
r
,
lo
ng
b
e
r
r
y
be
a
n
c
la
s
s
if
ic
a
ti
on
w
a
s
98.05%
a
c
c
ur
a
te
[
2
0
]
.
S
ubs
e
que
nt
ly
,
a
va
r
ie
ty
of
m
a
c
hi
ne
le
a
r
ni
ng
m
e
th
odol
ogi
e
s
s
uc
h
a
s
S
V
M
,
de
e
p
ne
ur
a
l
ne
twor
ks
(
D
N
N
)
,
a
nd
r
a
ndom
f
or
e
s
t
(
R
F
)
w
e
r
e
ut
il
iz
e
d
to
e
va
lu
a
te
th
e
s
ig
ni
f
ic
a
nc
e
of
s
ha
pe
a
nd c
ol
or
c
ha
r
a
c
te
r
is
ti
c
s
in
th
e
a
s
s
e
s
s
m
e
n
t
of
f
a
ul
ts
in
c
of
f
e
e
be
a
ns
.
T
he
da
ta
pr
e
s
e
nt
e
d
in
th
e
s
tu
dy
hi
ghl
ig
ht
s
th
e
s
ig
ni
f
ic
a
nc
e
of
c
ol
or
de
s
c
r
ip
to
r
s
in
t
he
c
la
s
s
if
ic
a
ti
on
of
f
a
ul
ts
in
c
of
f
e
e
be
a
ns
.
T
he
c
la
s
s
if
ic
a
ti
on
m
od
e
ls
c
ons
id
e
r
th
e
m
os
t
s
ig
ni
f
ic
a
nt
f
e
a
tu
r
e
s
obt
a
in
e
d
f
r
om
th
e
a
ve
r
a
ge
G
va
lu
e
of
th
e
c
om
pone
nt
i
n t
he
R
G
B
c
ol
or
s
pa
c
e
a
nd t
he
a
ve
r
a
ge
V
va
lu
e
i
n
th
e
H
S
V
c
ol
or
s
pa
c
e
. A
ll
t
he
c
la
s
s
if
ie
r
m
ode
ls
e
xhi
bi
te
d c
om
pa
r
a
bl
e
pe
r
f
or
m
a
nc
e
, w
it
h t
he
be
s
t
a
c
c
ur
a
c
y va
lu
e
a
bove
88%
[
1
2
]
.
S
e
ve
r
a
l
e
f
f
or
ts
w
e
r
e
pr
e
s
e
nt
e
d
in
or
de
r
to
id
e
nt
if
y
a
nd
c
a
te
gor
iz
e
c
of
f
e
e
f
r
ui
ts
,
a
s
w
e
ll
a
s
to
m
a
p
th
e
s
ta
ge
of
m
a
tu
r
a
ti
on
of
th
e
s
e
f
r
ui
ts
dur
in
g
th
e
ha
r
ve
s
t
pr
oc
e
s
s
.
T
he
m
e
th
odol
ogy
w
a
s
e
xe
c
ut
e
d
ut
il
iz
in
g
th
e
D
a
r
kne
t
f
r
a
m
e
w
or
k.
T
he
Y
O
L
O
v3
-
ti
ny
obj
e
c
t
id
e
nt
if
ic
a
ti
on
s
ys
te
m
id
e
nt
if
ie
d
a
nd
c
a
te
gor
iz
e
d
c
of
f
e
e
f
r
ui
t.
T
he
c
ol
le
c
ti
on
c
ont
a
in
s
90
vi
de
os
f
r
om
th
e
2020
a
r
a
bi
c
a
c
of
f
e
e
(
C
a
tu
a
í
144)
ha
r
ve
s
t,
s
hot
a
t
a
c
of
f
e
e
ha
r
ve
s
te
r
'
s
di
s
c
ha
r
ge
c
onve
yor
te
r
m
in
a
ti
on
poi
nt
.
A
bus
in
e
s
s
a
r
e
a
in
P
a
to
s
de
M
in
a
s
,
M
in
a
s
G
e
r
a
is
,
B
r
a
z
il
hos
te
d
th
e
r
e
c
or
di
ngs
.
T
h
e
m
ode
l
p
e
r
f
or
m
e
d
be
s
t
a
t
a
r
ound
3
300t
h
it
e
r
a
ti
on
w
it
h
a
n
800
×
800
-
pi
xe
l
im
a
ge
in
put
.
T
he
m
ode
l
ha
d
84%
m
e
a
n
a
v
e
r
a
ge
pr
e
c
is
io
n
(
m
A
P
)
,
82%
F
1
-
s
c
or
e
,
83%
pr
e
c
is
io
n,
a
nd
82%
r
e
c
a
ll
in
th
e
va
li
da
ti
on
s
e
t.
T
he
pr
e
c
is
io
n
va
lu
e
s
f
or
unr
ip
e
,
r
ip
e
,
a
nd
ov
e
r
r
ip
e
c
of
f
e
e
f
r
ui
ts
w
e
r
e
86%
,
85%
,
a
nd
80%
,
r
e
s
pe
c
ti
ve
ly
[
2
1
]
.
A
not
he
r
s
tu
dy
us
e
d
a
c
onvolut
io
na
l
ne
twor
k
on
a
n
in
e
xpe
ns
iv
e
m
ic
r
o
-
c
ont
r
ol
le
r
boa
r
d
to
c
la
s
s
if
y
c
of
f
e
e
le
a
f
di
s
e
a
s
e
s
lo
c
a
ll
y
w
it
hout
th
e
in
te
r
ne
t.
E
a
r
ly
di
a
gnos
is
of
c
of
f
e
e
pl
a
nt
di
s
e
a
s
e
s
w
a
s
c
r
uc
ia
l
f
or
opt
im
a
l
out
put
a
nd
pr
oduc
ti
on
qua
li
ty
.
T
w
o
da
ta
s
e
ts
a
nd
de
ve
lo
pm
e
nt
boa
r
d
im
a
ge
s
w
e
r
e
u
s
e
d
in
th
is
in
ve
s
ti
ga
ti
on.
T
h
e
c
ol
le
c
ti
on
in
c
lu
de
d
a
r
ound
6000
im
a
ge
s
f
r
om
s
ix
s
ic
kne
s
s
c
la
s
s
e
s
.
T
he
in
c
or
por
a
te
d
c
a
s
c
a
d
e
a
nd
s
in
gl
e
-
s
ta
ge
s
y
s
te
m
s
w
e
r
e
98%
a
nd
96%
a
c
c
ur
a
te
,
r
e
s
pe
c
ti
ve
ly
.
T
h
e
s
e
f
in
di
ngs
im
pl
y
th
a
t
th
e
s
e
s
tr
uc
tu
r
e
s
de
te
c
t
c
of
f
e
e
pl
a
nt
a
ti
on dis
e
a
s
e
s
[
2
2
]
.
T
hi
s
s
tu
dy
pr
e
s
e
nt
s
a
pr
opos
e
d
m
e
th
od
f
or
c
la
s
s
if
yi
ng
c
of
f
e
e
b
e
a
n
qua
li
ty
ba
s
e
d
on c
om
put
e
r
vi
s
io
n
te
c
hni
que
s
.
T
he
m
e
th
od
ut
il
iz
e
s
c
ol
or
,
s
ha
p
e
,
a
nd
te
xt
ur
e
da
t
a
e
xt
r
a
c
te
d
f
r
om
th
e
R
G
B
,
H
S
V
,
a
nd
L
*a
*b
c
ol
or
s
pa
c
e
s
.
T
he
B
P
w
a
s
e
m
pl
oy
e
d
a
s
th
e
c
la
s
s
if
ie
r
in
th
is
w
or
k.
T
he
obj
e
c
ti
ve
of
th
is
m
e
th
od
w
a
s
to
a
s
c
e
r
ta
in
th
e
c
la
s
s
if
ic
a
ti
on
of
c
of
f
e
e
be
a
n
s
a
c
c
or
di
ng
to
th
e
ir
qua
li
ty
by
ut
il
iz
in
g
im
a
ge
da
ta
.
T
he
qu
a
li
ty
ty
pe
s
w
e
r
e
c
la
s
s
if
ie
d i
nt
o f
our
c
la
s
s
e
s
:
in
ta
c
t,
pe
r
f
or
a
te
d, w
r
in
kl
e
d, a
nd c
r
a
c
ke
d.
2.
M
A
T
E
R
I
A
L
S
A
N
D
M
E
T
H
O
D
S
T
he
a
ppr
oa
c
h
pr
e
di
c
te
d
th
e
qua
li
ty
c
la
s
s
of
a
ll
r
obus
ta
c
of
f
e
e
be
a
n
phot
os
.
I
ts
m
a
in
pr
oc
e
s
s
e
s
w
e
r
e
r
e
gi
on
of
in
te
r
e
s
t
(
R
O
I
)
de
te
c
ti
on,
pr
e
-
pr
oc
e
s
s
in
g,
f
e
a
tu
r
e
e
xt
r
a
c
ti
on,
s
e
le
c
ti
on,
a
nd
c
la
s
s
if
ic
a
ti
on.
T
he
m
e
th
od
ha
s
two
pha
s
e
s
:
tr
a
in
in
g
a
nd
t
e
s
ti
ng.
T
r
a
in
in
g
a
nd
te
s
ti
ng
s
e
ts
pr
ovi
de
d
in
put
f
or
e
a
c
h
pha
s
e
.
B
ot
h
pha
s
e
s
w
e
r
e
ha
ndl
e
d
di
f
f
e
r
e
nt
ly
.
R
O
I
de
te
c
ti
on
a
s
s
ig
ne
d
th
e
c
of
f
e
e
be
a
n
a
r
e
a
to
th
e
im
a
ge
us
in
g
K
-
m
e
a
ns
c
lu
s
te
r
in
g.
T
he
tr
a
in
in
g
s
te
p
pr
e
-
pr
oc
e
s
s
e
d
R
G
B
da
ta
in
to
gr
a
ys
c
a
le
,
H
S
V
,
a
nd
L
*a
*b.
A
f
te
r
w
a
r
d,
c
ol
or
,
te
xt
ur
e
,
a
nd
s
ha
p
e
w
e
r
e
u
s
e
d
to
e
xt
r
a
c
t
f
e
a
tu
r
e
s
.
S
ubs
e
que
nt
ly
,
th
e
f
e
a
tu
r
e
s
e
le
c
ti
on
pr
oc
e
dur
e
w
a
s
u
s
e
d
to
c
hoos
e
th
e
m
os
t
s
ig
ni
f
ic
a
nt
f
e
a
tu
r
e
s
a
nd
s
im
pl
if
y
c
la
s
s
if
ic
a
ti
on.
P
r
e
-
pr
oc
e
s
s
in
g
m
e
r
e
ly
c
onve
r
te
d
R
G
B
to
th
e
f
e
a
tu
r
e
s
e
le
c
ti
on
c
ol
or
s
pa
c
e
dur
in
g
te
s
ti
ng.
P
r
in
c
ip
a
l
c
om
p
one
nt
a
na
ly
s
i
s
(
P
C
A
)
ge
ne
r
a
te
d
th
e
s
e
le
c
te
d
f
e
a
tu
r
e
s
in
th
e
pr
opos
e
d
te
c
hni
que
.
T
he
f
e
a
tu
r
e
s
e
l
e
c
ti
on
r
e
s
u
lt
w
a
s
u
s
e
d
to
a
ppl
y
th
e
e
xt
r
a
c
te
d
f
e
a
tu
r
e
.
A
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
I
m
age
analy
s
is
f
or
c
la
s
s
if
y
in
g c
of
fe
e
be
an quali
ty
us
in
g a multi
-
fe
at
ur
e
…
(
A
ni
ndi
ta
Se
pt
ia
r
in
i
)
4243
pr
e
di
c
ti
on
c
la
s
s
(
in
ta
c
t/
pe
r
f
or
a
te
d/
w
r
in
kl
e
d/
c
r
a
c
ke
d)
w
a
s
de
te
r
m
in
e
d
f
r
om
s
e
le
c
te
d
f
e
a
tu
r
e
s
in
th
e
f
in
a
l
s
te
p.
F
ig
ur
e
1
il
lu
s
tr
a
te
s
t
he
r
obus
ta
c
of
f
e
e
be
a
n qua
li
ty
c
la
s
s
if
ic
a
ti
o
n.
F
ig
ur
e
1
.
O
ve
r
vi
e
w
of
a
ll
s
te
ps
i
n t
he
pr
opos
e
d m
e
th
od f
or
qua
li
ty
c
la
s
s
if
ic
a
ti
on of
r
obus
ta
c
of
f
e
e
be
a
n
2.1.
D
at
as
e
t
T
h
e
d
a
t
a
s
e
t
in
th
i
s
s
t
ud
y
w
a
s
i
m
a
g
e
s
of
r
o
b
u
s
t
a
c
of
f
e
e
b
e
a
n
s
.
J
P
E
G
i
m
a
g
e
s
w
e
r
e
ta
k
e
n
w
i
t
h
a
X
i
a
om
i
5
A
s
m
a
r
tp
h
on
e
'
s
i
n
bu
il
t
c
a
m
e
r
a
. T
h
e
c
of
f
e
e
b
e
a
n w
a
s
pl
a
c
e
d
o
n
a
w
hi
te
b
a
c
kg
r
o
un
d i
n t
h
e
c
e
nt
e
r
o
f
a
2
8
×
19
×
1
8
c
m
s
t
u
di
o
m
i
ni
b
ox
.
A
1
0
c
m
g
a
p
b
e
t
w
e
e
n
t
h
e
c
a
m
e
r
a
a
n
d
t
h
e
c
of
f
e
e
be
a
n
s
w
a
s
m
a
i
n
ta
in
e
d
b
y
d
e
l
ib
e
r
a
t
e
l
y
p
o
s
it
io
ni
n
g
a
nd
or
i
e
n
ti
ng
th
e
c
a
m
e
r
a
.
S
m
a
r
t
ph
o
ne
c
a
m
e
r
a
s
a
r
e
1
3
-
m
e
g
a
pi
x
e
l
.
T
h
e
im
a
ge
h
a
s
d
im
e
n
s
i
on
s
of
1
5
60
×
1
56
0
pi
x
e
l
s
.
T
he
d
a
t
a
s
e
t
h
a
d
14
40
c
of
f
e
e
b
e
a
n
im
a
g
e
s
,
36
0
e
a
c
h
c
l
a
s
s
.
I
t
w
a
s
di
vi
d
e
d
i
nt
o
f
ou
r
c
l
a
s
s
e
s
:
i
n
ta
c
t,
pe
r
f
or
a
t
e
d,
w
r
i
nk
l
e
d
,
or
c
r
a
c
ke
d
,
w
i
th
th
e
e
x
a
m
pl
e
i
m
a
g
e
s
h
o
w
n
in
F
ig
ur
e
s
2(
a
)
to
2
(
d
)
,
r
e
s
pe
c
ti
v
e
l
y
.
(
a
)
(
b)
(
c
)
(
d)
F
ig
ur
e
2
.
E
xa
m
pl
e
s
of
c
of
f
e
e
be
a
n i
m
a
ge
s
w
it
h va
r
io
us
qua
li
ty
ty
pe
s
:
(
a
)
i
nt
a
c
t,
(
b)
pe
r
f
or
a
te
d, (
c
)
w
r
in
k
le
d,
a
nd (
d)
c
r
a
c
ke
d
2.2.
R
e
gi
on
of
i
n
t
e
r
e
s
t
d
e
t
e
c
t
io
n
R
O
I
de
te
c
ti
on
a
tt
e
m
pt
s
to
ge
n
e
r
a
te
a
s
ub
-
im
a
ge
m
o
s
tl
y
of
th
e
c
of
f
e
e
be
a
n
a
r
e
a
.
D
ur
in
g
th
is
s
ta
g
e
,
th
e
in
it
ia
l
r
e
s
ol
ut
io
n
of
th
e
im
a
ge
w
a
s
r
e
duc
e
d
f
r
om
1560×
1560
pi
xe
ls
to
500×
500
pi
x
e
ls
[
3]
to
m
in
im
iz
e
th
e
c
om
put
a
ti
ona
l
ti
m
e
. S
ubs
e
que
nt
ly
, a
c
ol
or
s
pa
c
e
c
onve
r
s
io
n w
a
s
pe
r
f
or
m
e
d f
r
om
R
G
B
t
o
L
*a
*b;
t
hi
s
e
na
bl
e
d
th
e
s
ys
t
e
m
to
a
c
c
ur
a
te
ly
di
f
f
e
r
e
nt
ia
te
b
e
twe
e
n
obj
e
c
t
a
nd
b
a
c
kgr
ound
r
e
gi
ons
in
va
r
io
us
s
c
e
na
r
io
s
.
T
h
e
ut
il
iz
a
ti
on of
L
*a
*b c
ol
or
s
pa
c
e
s
ne
c
e
s
s
it
a
te
s
a
c
onve
r
s
io
n pr
oc
e
dur
e
t
ha
t
r
e
li
e
s
on t
he
va
lu
e
s
w
it
hi
n t
he
R
G
B
c
ol
or
s
pa
c
e
, w
hi
c
h a
r
e
e
xpl
ic
it
ly
de
f
in
e
d a
s
i
n [
23]
.
T
he
r
e
s
ul
t
of
th
e
c
onve
r
s
io
n
of
a
n
or
ig
in
a
l
im
a
g
e
in
R
G
B
F
ig
ur
e
3(
a
)
to
L
*a
*b
c
ol
or
s
p
a
c
e
is
de
pi
c
te
d
in
F
ig
ur
e
3(
b
)
.
F
u
r
th
e
r
m
or
e
,
by
e
m
pl
oyi
ng
th
e
c
lu
s
te
r
in
g
w
it
h
th
e
K
-
m
e
a
ns
a
lg
or
it
hm
[
16]
,
th
e
a
r
e
a
of
c
of
f
e
e
be
a
ns
w
a
s
a
ppr
oxi
m
a
te
d.
D
ue
to
th
e
di
vi
s
io
n
of
th
e
im
a
ge
'
s
a
r
e
a
in
to
two
di
s
ti
nc
t
r
e
gi
ons
—
th
e
c
of
f
e
e
be
a
n
r
e
gi
on
a
nd
th
e
ba
c
kgr
ound
r
e
gi
on
—
th
e
va
lu
e
of
K
w
a
s
s
e
t
in
to
two.
T
he
s
t
e
ps
of
th
e
K
-
m
e
a
ns
a
lg
or
it
hm
a
r
e
de
f
in
e
d a
s
f
ol
lo
w
s
[
24]
:
−
S
te
p 1:
i
ni
ti
a
li
z
e
numbe
r
of
c
lu
s
te
r
k a
nd c
e
nt
r
e
.
−
S
te
p
2:
F
or
e
a
c
h
pi
xe
l
of
a
n
im
a
ge
,
c
a
lc
ul
a
t
e
th
e
E
uc
li
de
a
n
di
s
ta
nc
e
d,
b
e
twe
e
n
th
e
c
e
nt
r
e
a
nd
e
a
c
h
pi
xe
l
of
a
n i
m
a
ge
us
in
g t
he
r
e
la
ti
on give
n.
=
‖
(
,
)
−
‖
(
1
)
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
.
13
,
N
o.
4
,
D
e
c
e
m
be
r
2024
:
4241
-
4248
4244
−
S
te
p 3:
A
s
s
ig
n a
ll
t
he
pi
xe
ls
t
o t
he
n
e
a
r
e
s
t
c
e
nt
r
e
b
a
s
e
d on di
s
ta
nc
e
d.
−
S
te
p 4:
A
f
te
r
a
ll
pi
xe
ls
ha
ve
be
e
n a
s
s
ig
ne
d, r
e
c
a
l
c
ul
a
te
ne
w
po
s
it
io
n of
t
he
c
e
nt
r
e
us
in
g t
he
r
e
la
ti
on give
n:
=
1
∑
∑
(
,
)
(
2
)
−
S
te
p 4:
R
e
pe
a
t
th
e
pr
oc
e
s
s
unt
il
i
t
s
a
ti
s
f
ie
s
t
h
e
t
ol
e
r
a
nc
e
or
e
r
r
or
va
lu
e
.
−
S
te
p
5
:
R
e
s
ha
pe
t
he
c
lu
s
te
r
pi
xe
ls
i
nt
o i
m
a
ge
.
T
he
r
e
s
ul
ti
ng
im
a
g
e
of
th
e
K
-
m
e
a
n
s
a
lg
or
it
hm
is
s
how
n
in
F
ig
ur
e
3(
c
)
.
A
f
te
r
w
a
r
d,
a
m
or
phol
ogi
c
a
l
ope
r
a
ti
on
w
a
s
a
ppl
ie
d
us
in
g
di
la
ti
on;
he
nc
e
,
th
e
c
of
f
e
e
be
a
n
a
r
e
a
a
ppr
oa
c
he
s
th
e
or
ig
in
a
l,
a
nd
th
e
r
e
s
ul
t
is
de
pi
c
te
d
in
F
ig
ur
e
3(
d)
.
S
ubs
e
que
nt
ly
,
th
e
s
e
tt
in
g
of
th
e
c
of
f
e
e
be
a
n
a
r
e
a
w
a
s
c
a
r
r
ie
d
out
a
s
th
e
gr
ound
f
o
r
de
f
in
in
g t
he
R
O
I
i
m
a
ge
bounda
r
y ba
s
e
d on the
ye
ll
ow
box,
a
s
s
how
n i
n F
ig
ur
e
3(
e
)
. A
c
c
or
di
ngl
y,
t
he
f
o
r
m
e
d
R
O
I
i
m
a
ge
s
i
n bi
na
r
y a
nd R
G
B
c
ol
or
s
pa
c
e
a
r
e
s
how
n i
n F
ig
ur
e
s
3(
f
)
a
nd 3(
g)
.
(
a
)
(
b)
(
c
)
(
d)
(
e
)
(f)
(
g)
F
ig
ur
e
3. T
he
r
e
s
ul
ti
ng i
m
a
ge
of
e
a
c
h pr
oc
e
s
s
i
n R
O
I
de
te
c
ti
on:
(
a
)
or
ig
in
a
l
im
a
ge
i
n R
G
B
c
ol
or
s
pa
c
e
,
(
b)
L
*a
*b c
ol
or
s
pa
c
e
, (
c
)
K
-
m
e
a
ns
c
lu
s
te
r
in
g, (
d)
m
or
phol
ogi
c
a
l
ope
r
a
ti
on, (
e
)
s
e
tt
in
g t
he
a
r
e
a
of
R
O
I
im
a
ge
, a
nd (
f
)
R
O
I
i
m
a
ge
2.3.
P
r
e
-
p
r
oc
e
s
s
in
g
T
hi
s
pr
oc
e
dur
e
ge
ne
r
a
te
d
p
a
r
a
m
e
te
r
va
lu
e
s
f
or
f
e
a
tu
r
e
e
xt
r
a
c
ti
on.
T
hi
s
s
tu
dy
e
x
a
m
in
e
d
c
ol
or
,
te
xt
ur
e
,
a
nd
s
h
a
pe
.
R
G
B
im
a
ge
s
m
us
t
b
e
c
onve
r
te
d
to
L
*a
*b
a
nd
H
S
V
to
c
r
e
a
te
c
ol
or
f
e
a
tu
r
e
s
,
R
G
B
im
a
g
e
s
to
gr
a
ys
c
a
le
to
c
r
e
a
te
te
xt
ur
e
f
e
a
tu
r
e
s
,
a
nd
bi
na
r
y
im
a
ge
s
t
o
bui
ld
f
or
m
f
e
a
tu
r
e
s
.
I
n
or
de
r
to
im
pr
ove
c
la
s
s
if
ic
a
ti
on
r
e
s
ul
ts
, t
he
c
ol
or
s
pa
c
e
m
us
t
be
c
ha
nge
d dur
in
g pr
e
-
pr
oc
e
s
s
in
g. A
gr
ic
ul
tu
r
a
l
r
e
s
e
a
r
c
h us
e
s
R
G
B
f
or
obj
e
c
t
c
la
s
s
if
ic
a
ti
on.
S
om
e
in
v
e
s
ti
ga
ti
ons
ha
ve
e
m
pl
oye
d
L
*a
*b
a
nd
H
S
V
c
ol
or
s
pa
c
e
s
.
U
s
in
g
di
f
f
e
r
e
nt
c
ol
or
s
pa
c
e
s
r
e
qui
r
e
s
a
c
onve
r
s
io
n
te
c
hni
que
th
a
t
u
s
e
s
R
G
B
va
lu
e
s
[
2
3
]
.
In
(
3
)
-
(
6
)
de
f
in
e
R
G
B
-
to
-
L
*a
*b
c
onve
r
s
io
n.
I
n
H
S
V
c
ol
or
s
pa
c
e
,
in
(
3
)
-
(
4
)
c
a
lc
ul
a
te
hue
(
H
)
a
nd
th
e
n
s
a
tu
r
a
ti
on
(
S
)
a
nd
va
lu
e
(
V
)
.
S
a
nd
V
va
lu
e
s
w
e
r
e
c
om
put
e
d u
s
in
g
a
s
(
5
)
a
nd (
6
).
=
{
,
≤
360
−
,
>
(
3
)
w
he
r
e
:
=
−
1
{
1
2
[
(
−
)
+
(
−
)
]
[
(
−
)
2
+
(
−
)
(
−
]
1
/
2
}
(
4
)
=
{
0
,
m
a
x
(
,
,
)
=
0
1
−
min
(
,
,
)
ma
x
(
,
,
)
,
ℎ
(
5
)
=
m
a
x
(
,
,
)
(
6
)
F
ur
th
e
r
m
or
e
,
c
onve
r
ti
ng
th
e
R
G
B
im
a
ge
to
a
gr
a
ys
c
a
le
im
a
ge
w
a
s
ne
e
de
d;
he
nc
e
,
th
is
w
or
k
a
ppl
ie
d
te
xt
ur
e
f
e
a
tu
r
e
s
.
T
he
s
e
f
e
a
tu
r
e
p
a
r
a
m
e
te
r
s
w
il
l
l
a
te
r
be
u
s
e
d
a
s
in
put
f
or
th
e
c
la
s
s
if
ic
a
ti
on
pr
oc
e
s
s
.
R
G
B
c
onve
r
s
io
n t
o gr
a
ys
c
a
le
i
s
c
a
r
r
ie
d out t
o pr
oduc
e
i
nt
e
ns
it
y (
I
)
va
lu
e
s
us
in
g (
7
)
[
11]
.
=
1
3
(
+
+
)
(
7
)
2.4.
F
e
at
u
r
e
e
xt
r
ac
t
io
n
C
of
f
e
e
be
a
n
im
a
ge
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
r
e
tr
ie
ve
s
c
ol
or
,
te
xt
ur
e
,
a
nd
s
ha
pe
in
f
or
m
a
ti
on.
S
om
e
s
tu
di
e
s
a
na
ly
z
e
one
a
s
pe
c
t,
w
hi
le
ot
he
r
s
a
n
a
ly
z
e
s
e
ve
r
a
l.
T
hi
s
s
tu
dy
a
na
ly
z
e
s
th
r
e
e
a
ppr
oa
c
he
s
f
or
e
xt
r
a
c
ti
ng
th
r
e
e
-
c
ol
or
f
e
a
tu
r
e
s
us
in
g
R
G
B
,
H
S
V
,
a
nd
L
*a
*b
c
ol
or
m
ode
l
s
ta
ti
s
ti
c
a
l
va
lu
e
s
.
R
G
B
,
H
S
V
,
a
nd
L
*a
*b
a
r
e
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
I
m
age
analy
s
is
f
or
c
la
s
s
if
y
in
g c
of
fe
e
be
an quali
ty
us
in
g a multi
-
fe
at
ur
e
…
(
A
ni
ndi
ta
Se
pt
ia
r
in
i
)
4245
us
e
f
ul
c
ol
or
c
ha
r
a
c
te
r
is
ti
c
s
in
m
a
ny
a
ppl
ic
a
ti
ons
.
C
onve
r
ti
ng
R
G
B
to
H
S
V
a
nd
L
*a
*b
li
m
i
ts
c
ol
or
s
pa
c
e
di
m
e
ns
io
ns
a
nd
f
e
a
tu
r
e
s
.
T
e
xt
ur
e
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
us
in
g
th
e
gr
a
y
le
ve
l
c
o
-
oc
c
ur
r
e
nc
e
m
a
tr
ix
(
G
L
C
M
)
f
ol
lo
w
s
.
O
ur
f
o
r
m
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
a
ppr
oa
c
h
us
e
s
s
ta
ti
s
ti
c
a
l
c
ha
r
a
c
te
r
is
ti
c
s
a
nd
s
ha
pe
di
s
ta
n
c
e
in
th
e
bi
na
r
y
pi
c
tu
r
e
.
T
a
bl
e
1
li
s
ts
m
e
th
od
f
e
a
tu
r
e
c
ount
s
. A
ddi
ng
f
e
a
tu
r
e
s
do
e
s
n'
t
ne
c
e
s
s
a
r
il
y
e
nha
nc
e
m
ode
l
pe
r
f
or
m
a
nc
e
.
T
hus
, a
c
c
ur
a
te
c
la
s
s
if
ic
a
ti
on r
e
qui
r
e
s
c
a
r
e
f
ul
f
e
a
tu
r
e
s
e
le
c
ti
on.
T
a
bl
e
1.
T
he
numbe
r
of
f
e
a
tu
r
e
s
T
ype
of
f
e
a
t
ur
e
s
M
e
t
hod
N
um
be
r
of
F
e
a
t
ur
e
s
C
ol
or
R
G
B
M
ode
l
3
H
S
V
M
ode
l
3
L
*a
*b M
ode
l
3
T
e
xt
ur
e
S
t
a
t
i
s
t
i
c
F
e
a
t
ur
e
of
G
L
C
M
84
S
ha
pe
A
r
e
a
-
B
a
s
e
d
2
2.4.1.
C
ol
or
f
e
at
u
r
e
e
xt
r
a
c
t
io
n
T
he
H
S
V
,
R
G
B
,
a
nd
L
*a
*b
c
ol
or
s
pa
c
e
s
ha
v
e
be
e
n
us
e
d
a
s
c
ol
or
f
e
a
tu
r
e
s
to
di
f
f
e
r
e
nt
ia
te
va
r
io
us
obj
e
c
ts
[
11]
.
In
(
8
)
us
e
s
th
e
m
e
a
n
to
e
xa
m
in
e
e
a
c
h
c
ol
or
m
ode
l
c
ha
nne
l'
s
s
ta
ti
s
ti
c
a
l
pr
ope
r
ti
e
s
.
W
h
e
r
e
μ
is
th
e
a
ve
r
a
ge
c
ol
or
c
ha
nne
l.
T
hi
s
s
tu
dy
s
e
e
ks
H
S
V
,
R
G
B
,
or
L
*a
*
b
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
m
e
th
ods
f
or
c
of
f
e
e
be
a
n
da
ta
a
na
ly
s
is
.
T
hi
s
s
tu
dy
got
r
e
d,
gr
e
e
n,
a
nd
bl
ue
c
ol
or
va
lu
e
s
f
r
om
th
e
R
G
B
im
a
ge
,
hue
,
s
a
tu
r
a
ti
on,
a
n
d
va
lu
e
va
lu
e
s
f
r
om
th
e
H
S
V
im
a
ge
,
a
nd
li
ght
ne
s
s
(
L
)
a
nd
c
ol
or
-
oppone
nt
di
m
e
ns
io
ns
(
a
a
nd
b)
in
di
c
a
ti
ng
r
e
dne
s
s
–
gr
e
e
nne
s
s
a
nd blue
ne
s
s
–
ye
ll
ow
ne
s
s
f
r
om
t
he
L
*a
*b i
m
a
ge
.
=
1
∑
∑
(
)
=
1
=
1
(
8
)
2.4.2.
T
e
xt
u
r
e
f
e
at
u
r
e
e
xt
r
ac
t
io
n
C
of
f
e
e
be
a
n
te
xt
ur
e
s
c
a
n
be
id
e
nt
if
ie
d
by
te
xt
ur
e
.
C
of
f
e
e
be
a
n
f
ib
e
r
ha
s
di
s
ti
nc
t
vi
s
ua
l
a
nd
te
xt
ur
e
pr
ope
r
ti
e
s
.
T
he
c
ol
or
c
ha
r
a
c
te
r
is
ti
c
s
of
th
e
pe
r
f
or
a
te
d,
w
r
in
kl
e
d,
c
r
a
c
ke
d
c
la
s
s
a
r
e
s
im
il
a
r
to
th
os
e
of
th
e
in
ta
c
t
c
la
s
s
,
m
a
ki
ng
id
e
nt
if
ic
a
ti
on
di
f
f
ic
ul
t.
O
nl
y
G
L
C
M
is
us
e
d
to
e
xt
r
a
c
t
te
xt
ur
e
f
e
a
tu
r
e
s
in
th
is
in
ve
s
ti
ga
ti
on. G
L
C
M
ha
s
be
e
n us
e
d f
or
t
e
xt
ur
e
f
e
a
tu
r
e
e
xt
r
a
c
ti
on w
it
h good r
e
s
ul
ts
.
C
a
lc
ul
a
ti
ng
th
e
pr
oba
bi
li
ty
of
th
e
a
dj
a
c
e
nc
y
r
e
la
ti
ons
hi
p
be
tw
e
e
n
two
pi
xe
ls
a
t
a
s
p
e
c
if
ic
di
s
ta
nc
e
a
nd
a
ngl
e
or
ie
nt
a
ti
on
yi
e
ld
s
G
L
C
M
[
1]
.
C
a
lc
ul
a
te
th
e
im
a
ge
'
s
s
ta
ti
s
ti
c
a
l
a
tt
r
ib
ut
e
s
a
f
te
r
c
ol
le
c
ti
ng
th
e
co
-
oc
c
ur
r
e
nc
e
m
a
tr
ix
. G
L
C
M
s
ta
ti
s
ti
c
a
l
f
e
a
tu
r
e
s
e
xi
s
t
f
or
f
our
a
ngl
e
s
(
0
∘
, 45
∘
, 90
∘
, a
nd 135
∘
)
a
nd one
di
s
ta
nc
e
(
1
pi
xe
l)
.
G
L
C
M
(
i,
j
)
is
th
e
jo
in
t
pr
oba
bi
li
ty
di
s
tr
ib
ut
io
n
of
a
pi
xe
l
pa
ir
w
it
h
gr
a
y
le
ve
ls
i
a
nd
j
.
I
m
a
ge
gr
a
y
le
ve
l
de
te
r
m
in
e
s
G
L
C
M
m
a
tr
ix
r
ow
s
a
nd
c
ol
um
ns
.
L
is
th
e
c
om
put
e
d
gr
a
y
le
ve
ls
m
in
us
1.
T
he
gr
a
ys
c
a
le
va
lu
e
of
a
n
im
a
ge
be
twe
e
n
0
a
nd
255
[
7]
.
T
he
ty
pe
s
of
f
e
a
tu
r
e
s
us
e
d
in
G
L
C
M
in
th
is
r
e
s
e
a
r
c
h
in
c
lu
de
:
a
ut
o
c
or
r
e
la
ti
on,
c
lu
s
te
r
pr
om
in
e
nc
e
,
c
lu
s
te
r
s
h
a
de
,
c
ont
r
a
s
t,
c
or
r
e
la
ti
on,
di
f
f
e
r
e
nc
e
e
nt
r
opy,
di
f
f
e
r
e
nc
e
va
r
ia
nc
e
,
di
s
s
im
il
a
r
it
y,
e
ne
r
gy,
e
nt
r
opy
,
I
D
M
,
in
f
or
m
a
ti
on
m
e
a
s
ur
e
s
of
c
or
r
e
la
ti
on
1,
a
nd
2,
in
ve
r
s
e
di
f
f
e
r
e
nc
e
,
m
a
xi
m
um
pr
oba
bi
li
ty
, s
um
a
ve
r
a
ge
s
, s
um
e
nt
r
opy, s
um
of
s
qua
r
e
s
va
r
ia
nc
e
s
, s
um
va
r
ia
nc
e
, I
D
M
nor
m
a
li
z
e
d,
in
ve
r
s
e
di
f
f
e
r
e
nc
e
nor
m
a
li
z
e
d
.
2.4.3.
S
h
ap
e
f
e
at
u
r
e
e
xt
r
ac
t
io
n
T
he
k
-
m
e
a
ns
m
e
th
od
w
a
s
e
m
pl
oye
d
to
c
onve
r
t
th
e
c
of
f
e
e
be
a
n
s
im
a
ge
in
to
a
bi
na
r
y
im
a
ge
in
or
de
r
to
r
e
m
ove
noi
s
e
in
th
e
s
ha
pe
of
th
e
c
of
f
e
e
bodi
e
s
.
T
he
goa
l
of
th
e
K
-
m
e
a
ns
a
lg
or
it
hm
is
to
c
lu
s
te
r
obj
e
c
ts
by
gr
oupi
ng
th
e
m
w
it
h
th
e
K
poi
nt
s
th
a
t
a
r
e
c
lo
s
e
s
t
to
th
e
m
in
t
he
s
pa
c
e
.
T
he
va
lu
e
s
of
c
lu
s
te
r
c
e
nt
r
oi
ds
a
r
e
upda
te
d
it
e
r
a
ti
ve
ly
unt
il
th
e
opt
im
a
l
c
lu
s
te
r
in
g
r
e
s
ul
ts
a
r
e
a
c
hi
e
ve
d.
V
a
r
io
us
s
ha
pe
pa
r
a
m
e
te
r
s
,
s
uc
h
a
s
e
c
c
e
nt
r
ic
it
y (
e
)
a
nd pe
r
im
e
te
r
(
p)
, a
r
e
c
a
lc
ul
a
te
d t
o a
s
s
e
s
s
t
he
c
ha
r
a
c
te
r
is
ti
c
s
of
c
of
f
e
e
be
a
n
s
ha
pe
[
20]
.
=
√
1
−
2
2
(
9
)
=
2
√
(
2
+
2
)
2
(
1
0
)
2.5.
F
e
at
u
r
e
s
e
le
c
t
io
n
T
he
a
na
ly
s
is
s
houl
d
c
ons
id
e
r
th
e
to
ta
l
a
m
ount
of
f
e
a
tu
r
e
s
in
or
de
r
to
id
e
nt
if
y
th
e
m
os
t
be
ne
f
ic
ia
l
or
hi
ghl
y
di
s
c
r
im
in
a
ti
ve
f
e
a
tu
r
e
s
w
it
hi
n
th
e
ut
il
iz
e
d
da
t
a
s
e
t.
T
he
pr
e
s
e
nt
s
tu
dy
e
m
pl
oye
d
P
C
A
a
s
a
m
e
th
od
f
or
c
onduc
ti
ng
f
e
a
tu
r
e
s
e
le
c
ti
on.
P
C
A
is
a
w
e
ll
-
e
s
ta
bl
is
he
d
te
c
h
ni
que
in
th
e
f
ie
ld
of
pa
tt
e
r
n
r
e
c
ogni
ti
on
a
n
d
c
om
put
e
r
vi
s
io
n.
I
t
s
e
r
ve
s
a
s
a
s
ta
nda
r
d
m
e
th
od
f
or
f
e
a
tu
r
e
e
xt
r
a
c
ti
on
a
nd
da
ta
r
e
pr
e
s
e
nt
a
ti
on,
c
om
m
onl
y
e
m
pl
oye
d
to
id
e
nt
if
y
a
nd
r
e
c
ogni
z
e
obj
e
c
ts
.
A
s
ta
ti
s
ti
c
a
l
m
e
th
od
r
e
duc
e
d
th
e
num
be
r
of
di
m
e
ns
io
ns
in
da
ta
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
.
13
,
N
o.
4
,
D
e
c
e
m
be
r
2024
:
4241
-
4248
4246
s
e
ts
w
it
h
m
a
ny
f
a
c
to
r
s
.
I
t
m
a
ke
s
obj
e
c
t
r
e
c
ogni
ti
on
w
or
k
be
tt
e
r
a
nd
ha
s
be
e
n
s
how
n
to
lo
w
e
r
a
nd
r
a
is
e
th
e
a
c
c
ur
a
c
y va
lu
e
[
25]
.
2.6.
C
la
s
s
if
i
c
at
io
n
D
a
ta
is
c
la
s
s
if
ie
d
by
c
la
s
s
if
ic
a
ti
on.
M
a
c
hi
ne
le
a
r
ni
ng
ha
s
s
e
ve
r
a
l
pl
a
nt
obj
e
c
ts
us
e
s
.
B
y
r
e
s
e
a
r
c
hi
ng
a
lg
or
it
hm
s
a
nd
us
in
g
da
ta
to
f
or
e
c
a
s
t,
m
a
c
hi
ne
le
a
r
ni
ng
a
ut
om
a
te
s
ope
r
a
ti
ons
.
T
he
a
lg
or
it
hm
us
e
s
a
m
ode
l
to
e
s
ti
m
a
te
da
ta
a
nd
m
a
ke
ju
dgm
e
nt
s
ba
s
e
d
on
s
a
m
pl
e
in
pu
t
in
s
te
a
d
th
a
n
f
ol
lo
w
in
g
f
ix
e
d
in
s
tr
uc
ti
ons
.
M
a
th
e
m
a
ti
c
a
l
a
nd
s
ta
ti
s
ti
c
a
l
m
ode
l
s
pr
e
di
c
te
d
unknown
da
ta
u
s
in
g
tr
a
in
in
g
da
ta
.
T
hi
s
s
tu
dy
c
la
s
s
if
ie
d
c
of
f
e
e
be
a
n
qua
li
ty
us
in
g
m
a
c
hi
ne
le
a
r
ni
ng.
B
a
c
kpr
opa
ga
ti
on
n
e
ur
a
l
ne
twor
k
(
B
P
N
N
)
,
li
ne
a
r
di
s
c
r
im
in
a
nt
a
na
ly
s
i
s
(
L
D
A
)
,
K
N
N
,
NB
,
a
nd
S
V
M
w
e
r
e
us
e
d.
P
r
e
vi
ous
r
e
s
e
a
r
c
h
on
num
e
r
ous
pl
a
nt
s
pe
c
im
e
ns
us
e
d
th
e
s
e
m
e
th
odol
ogi
e
s
[
19]
, [
25]
.
2.7.
P
e
r
f
or
m
an
c
e
m
e
as
u
r
e
m
e
n
t
F
e
a
tu
r
e
s
e
le
c
ti
on
a
nd
di
f
f
e
r
e
nt
c
la
s
s
if
ic
a
ti
on
a
ppr
oa
c
h
e
s
a
r
e
us
e
d
to
e
va
lu
a
t
e
th
e
pr
opos
e
d
m
e
th
od'
s
pe
r
f
or
m
a
nc
e
a
nd
de
te
r
m
in
e
th
e
m
os
t
a
ppr
opr
ia
te
a
nd
r
obus
t
m
e
th
od
f
or
th
e
da
ta
s
e
t.
T
hi
s
s
tu
dy
e
va
lu
a
te
s
1140
r
obus
ta
be
a
n
phot
os
(
360
in
ta
c
t,
360
pe
r
f
or
a
te
d,
360
w
r
in
kl
e
d,
a
nd
360
c
r
a
c
ke
d
).
T
he
pe
r
f
or
m
a
nc
e
of
th
e
c
la
s
s
if
ic
a
ti
on
m
e
th
od
w
a
s
a
s
s
e
s
s
e
d
u
s
in
g
th
r
e
e
in
di
c
a
to
r
s
:
a
c
c
ur
a
c
y
[
25]
,
w
hi
c
h
w
e
r
e
c
a
lc
ul
a
te
d
ba
s
e
d
on
th
e
m
ul
ti
c
la
s
s
c
onf
us
io
n m
a
tr
ix
. T
h
e
pa
r
a
m
e
te
r
s
a
r
e
d
e
f
in
e
d by (
1
1
)
[
25]
.
=
×
100
(1
1
)
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
he
to
ta
l
num
be
r
of
1440
im
a
ge
s
c
ons
is
te
d
of
f
our
c
la
s
s
e
s
,
in
c
lu
di
ng
in
ta
c
t,
pe
r
f
or
a
te
d,
w
r
in
kl
e
d,
a
nd
c
r
a
c
ke
d,
us
e
d
to
e
va
lu
a
te
th
e
m
e
th
od.
T
he
c
la
s
s
if
ic
a
ti
on
r
e
s
ul
ts
w
e
r
e
c
om
pa
r
e
d
f
or
f
e
a
tu
r
e
s
e
ts
obt
a
in
e
d
w
it
hout
a
nd
w
it
h
f
e
a
tu
r
e
s
e
le
c
ti
on.
T
he
f
e
a
tu
r
e
s
e
t
w
it
hout
f
e
a
t
ur
e
s
e
le
c
ti
on
w
a
s
pe
r
f
or
m
e
d
us
in
g
th
e
f
e
a
tu
r
e
s
f
or
m
e
d
by
G
L
C
M
,
w
hi
c
h
w
e
r
e
a
r
e
a
-
ba
s
e
d
a
nd
c
ol
or
-
ba
s
e
d
(
R
G
B
,
H
S
V
,
a
nd
L
*a
*b)
.
T
he
s
e
le
c
t
e
d
f
e
a
tu
r
e
s
w
e
r
e
a
ppl
ie
d
to
P
C
A
.
T
he
e
xpe
r
im
e
nt
ut
il
iz
e
d
f
iv
e
c
la
s
s
if
ie
r
s
:
B
P
N
N
,
K
N
N
,
L
D
A
,
N
B
,
a
nd
S
V
M
.
T
he
m
e
th
od'
s
pe
r
f
or
m
a
nc
e
w
a
s
a
s
s
e
s
s
e
d
us
in
g
a
c
c
ur
a
c
y
.
T
he
c
om
pa
r
is
on
of
m
e
th
od
pe
r
f
or
m
a
nc
e
w
it
h
di
f
f
e
r
e
nt
c
la
s
s
if
ie
r
s
us
in
g m
ul
ti
-
f
e
a
tu
r
e
i
ndi
c
a
te
d by the
a
c
c
ur
a
c
y va
lu
e
i
s
s
um
m
a
r
iz
e
d i
n T
a
bl
e
2.
T
a
bl
e
2
de
m
on
s
tr
a
te
s
th
e
te
s
ti
ng
of
s
it
ua
ti
on
s
w
it
hout
f
e
a
tu
r
e
s
e
le
c
ti
on.
T
h
e
te
s
ts
w
e
r
e
c
onduc
te
d
us
in
g
te
xt
ur
e
c
ha
r
a
c
te
r
is
ti
c
s
obt
a
in
e
d
by
th
e
G
L
C
M
a
ppr
oa
c
h,
te
xt
ur
e
da
ta
c
om
bi
ne
d
w
it
h
s
ha
pe
f
e
a
tu
r
e
s
(
A
r
e
a
-
ba
s
e
d)
,
a
nd
te
xt
ur
e
f
e
a
tu
r
e
s
c
om
bi
ne
d
w
it
h
H
S
V
,
L
*a
*b
,
a
nd
R
G
B
c
ol
or
s
pa
c
e
s
.
T
he
B
P
N
N
c
l
a
s
s
if
ie
r
a
c
hi
e
ve
d
th
e
m
a
xi
m
um
a
c
c
ur
a
c
y
of
94.83%
w
hi
le
ut
il
iz
in
g
th
e
G
L
C
M
.
S
ubs
e
que
nt
ly
,
th
e
L
D
A
,
K
N
N
,
S
V
M
,
a
nd
N
B
c
la
s
s
if
ie
r
s
yi
e
ld
e
d
de
c
r
e
a
s
in
g
a
c
c
ur
a
c
y
va
lu
e
s
o
f
92.08%
,
85.21%
,
83.54%
,
a
nd
80.83
%
,
r
e
s
pe
c
ti
ve
ly
.
T
he
B
P
N
N
c
l
a
s
s
if
ie
r
a
c
hi
e
ve
d
th
e
m
a
xi
m
um
a
c
c
u
r
a
c
y
of
97.86%
f
o
r
th
e
a
r
e
a
-
ba
s
e
d
f
e
a
tu
r
e
s
e
t,
w
hi
le
th
e
N
B
c
la
s
s
if
ie
r
ha
d
th
e
lo
w
e
s
t
a
c
c
ur
a
c
y
of
58.54%
.
T
he
pe
r
f
or
m
a
nc
e
r
e
li
e
d
on
c
ol
or
a
tt
r
ib
u
te
s
de
r
iv
e
d
f
r
om
th
r
e
e
c
ol
or
s
pa
c
e
s
:
H
S
V
,
L
*a
*b,
a
nd
R
G
B
.
T
h
e
pe
r
f
or
m
a
nc
e
of
th
e
a
ppr
oa
c
h
ut
il
iz
in
g
H
S
V
s
how
e
d
th
a
t
S
V
M
a
tt
a
in
e
d
a
n
id
e
a
l
a
c
c
ur
a
c
y
r
a
te
of
92.92%
.
U
s
in
g
th
e
L
*a
*b
a
nd
R
G
B
f
e
a
tu
r
e
s
e
ts
,
th
e
B
P
N
N
a
c
hi
e
ve
d a
n i
m
pr
e
s
s
iv
e
a
c
c
ur
a
c
y r
a
te
of
97.71%
.
T
a
bl
e
2. P
e
r
f
or
m
a
nc
e
c
om
pa
r
is
on of
t
he
c
la
s
s
if
ie
r
w
it
h va
r
io
us
f
e
a
tu
r
e
s
e
ts
ba
s
e
d on a
c
c
ur
a
c
y va
lu
e
(
%
)
C
l
a
s
s
i
f
i
e
r
W
i
t
hout
f
e
a
t
ur
e
s
s
e
l
e
c
t
i
on
W
i
t
h f
e
a
t
ur
e
s
s
e
l
e
c
t
i
on
G
L
C
M
A
r
e
a
B
a
s
e
d
H
S
V
L
*a
*b
R
G
B
P
C
A
B
P
N
N
94.83
97.86
84.79
97.71
97.71
98.54
KNN
85.21
84.38
85.42
90.83
89.79
90.83
L
D
A
92.08
91.46
77.92
91.46
68.33
80.63
NB
80.83
58.54
53.96
53.54
48.54
55.83
S
V
M
83.54
72.71
92.92
97.50
94.83
97.50
I
n
th
e
r
e
s
ul
ts
obt
a
in
e
d
vi
a
P
C
A
f
e
a
tu
r
e
s
e
le
c
ti
on,
th
e
ba
c
kpr
opa
ga
ti
on
c
la
s
s
if
ie
r
a
tt
a
in
e
d
th
e
m
a
xi
m
um
a
c
c
ur
a
c
y
v
a
lu
e
of
98.54%
.
B
y
c
ont
r
a
s
t,
th
e
NB
c
l
a
s
s
if
ie
r
yi
e
ld
e
d
th
e
lo
w
e
s
t
r
e
s
ul
ts
,
a
c
hi
e
vi
ng
a
n
a
c
c
ur
a
c
y
of
55.83%
.
B
a
c
kpr
opa
ga
ti
on
de
m
ons
tr
a
te
s
hi
gh
a
c
c
u
r
a
c
y
a
c
r
os
s
a
ll
f
e
a
tu
r
e
te
s
t
s
it
ua
ti
ons
w
it
hout
r
e
qui
r
in
g
f
e
a
tu
r
e
s
e
le
c
ti
on.
T
he
L
D
A
a
lg
or
it
hm
a
c
hi
e
v
e
d
th
e
h
ig
he
s
t
a
c
c
ur
a
c
y
r
a
te
of
98.54%
a
c
r
os
s
a
ll
f
our
te
s
t
s
c
e
na
r
io
s
.
I
t
w
a
s
pe
r
f
or
m
e
d
us
in
g
th
e
G
L
C
M
f
e
a
tu
r
e
,
A
r
e
a
-
ba
s
e
d
m
e
th
od,
H
S
V
c
ol
or
s
pa
c
e
va
lu
e
s
,
a
nd
P
C
A
a
ppl
ic
a
ti
on
f
or
f
e
a
tu
r
e
s
e
le
c
ti
on.
T
h
e
NB
c
la
s
s
if
ie
r
is
c
on
s
is
te
nt
ly
ove
r
w
he
lm
e
d
by
e
v
e
r
y
tr
ia
l
s
it
ua
ti
on.
T
he
obs
e
r
ve
d
r
e
s
ul
ts
in
di
c
a
te
th
a
t
a
c
om
bi
na
ti
on
of
te
xt
ur
e
,
s
ha
pe
,
a
nd
c
ol
or
f
e
a
tu
r
e
s
,
f
ol
lo
w
e
d
by
f
e
a
tu
r
e
s
e
le
c
ti
on
to
li
m
it
th
e
num
be
r
of
f
e
a
tu
r
e
s
,
m
ig
ht
le
a
d
to
hi
gh
a
c
c
ur
a
c
y
th
r
oughout
th
e
c
la
s
s
if
ic
a
ti
on
pr
oc
e
s
s
.
T
he
B
P
N
N
c
la
s
s
if
ie
r
pe
r
f
or
m
s
be
tt
e
r
t
ha
n ot
he
r
c
la
s
s
if
ie
r
s
by m
in
im
iz
in
g e
r
r
or
s
i
n e
a
c
h s
c
e
na
r
io
.
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
I
m
age
analy
s
is
f
or
c
la
s
s
if
y
in
g c
of
fe
e
be
an quali
ty
us
in
g a multi
-
fe
at
ur
e
…
(
A
ni
ndi
ta
Se
pt
ia
r
in
i
)
4247
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
c
la
s
s
if
ie
s
r
obus
ta
c
of
f
e
e
be
a
ns
by
qua
li
ty
.
T
h
e
r
e
a
r
e
f
our
ty
pe
s
of
c
of
f
e
e
be
a
ns
:
in
ta
c
t,
pe
r
f
or
a
te
d,
w
r
in
kl
e
d,
a
nd
br
oke
n.
T
hi
s
pr
oc
e
dur
e
in
vol
ve
s
R
O
I
de
te
c
ti
on,
pr
e
-
pr
oc
e
s
s
in
g,
s
e
gm
e
nt
a
ti
on,
f
e
a
tu
r
e
e
xt
r
a
c
ti
on,
s
e
le
c
ti
on
,
a
nd
c
la
s
s
if
ic
a
ti
on.
E
a
c
h
s
te
p
is
done
to
a
c
c
ur
a
te
ly
c
la
s
s
if
y
c
of
f
e
e
be
a
ns
a
nd
de
te
r
m
in
e
th
e
ir
qua
li
ty
.
T
he
s
tu
dy
te
s
te
d
de
s
ig
ns
w
it
h
te
xt
ur
e
,
te
xt
ur
e
w
it
h
s
ha
pe
,
a
nd
te
xt
ur
e
w
it
h
c
ol
or
s
pa
c
e
va
lu
e
s
(
H
S
V
,
L
*a
*b,
a
nd
R
G
B
)
.
B
P
N
N
s
tu
dy
r
out
in
e
ly
out
pe
r
f
or
m
s
ot
he
r
c
of
f
e
e
be
a
n
qua
li
ty
a
s
s
e
s
s
m
e
nt
m
e
th
odol
ogi
e
s
.
I
t
us
e
s
th
e
P
C
A
f
e
a
tu
r
e
s
e
le
c
ti
on
te
c
hni
que
to
ge
t
th
e
be
s
t
r
e
s
ul
ts
on
G
L
C
M
,
a
r
e
a
-
ba
s
e
d, a
nd L
*a
*b f
e
a
tu
r
e
s
w
it
h 98.54%
a
c
c
ur
a
c
y. U
s
in
g s
e
ve
r
a
l
s
c
e
na
r
io
s
a
nd
a
tt
r
ib
ut
e
s
c
a
n i
m
pr
ove
t
he
va
r
ie
ty
a
nd qua
li
ty
of
t
hi
s
r
e
s
e
a
r
c
h.
A
C
K
N
O
WL
E
D
G
E
M
E
N
T
S
T
he
a
ut
hor
w
oul
d
li
ke
to
th
a
nk
th
e
F
a
c
ul
ty
of
E
ngi
ne
e
r
in
g
a
t
M
ul
a
w
a
r
m
a
n
U
ni
ve
r
s
it
y
in
S
a
m
a
r
in
da
,
I
ndone
s
ia
, f
or
pr
ovi
di
ng f
in
a
nc
ia
l
s
uppor
t
f
or
t
he
r
e
s
e
a
r
c
h c
onduc
te
d i
n 2023 (
N
o. 497/UN17.L
1/
H
K
/2
023)
.
R
E
F
E
R
E
N
C
E
S
[
1]
M
.
R
.
F
i
ona
,
S
.
T
hom
a
s
,
I
.
J
.
M
a
r
i
a
,
a
nd
B
.
H
a
nna
h,
“
I
de
nt
i
f
i
c
a
t
i
on
o
f
r
i
pe
a
nd
unr
i
pe
c
i
t
r
us
f
r
u
i
t
s
us
i
ng
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
k,”
i
n
J
our
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
S
e
r
i
e
s
, I
O
P
P
ubl
i
s
hi
ng, 2019, doi
:
10.1088/
1742
-
6596/
1362/
1/
012033.
[
2]
S
.
M
une
r
a
,
F
.
H
e
r
ná
nde
z
,
N
.
A
l
e
i
xo
s
,
S
.
C
ub
e
r
o,
a
nd
J
.
B
l
a
s
c
o,
“
M
a
t
ur
i
t
y
m
o
ni
t
or
i
ng
of
i
nt
a
c
t
f
r
ui
t
a
nd
a
r
i
l
s
of
pom
e
gr
a
na
t
e
c
v.
‘
M
ol
l
a
r
de
E
l
c
he
’
us
i
ng
m
a
c
hi
ne
vi
s
i
on
a
nd
c
he
m
om
e
t
r
i
c
s
,”
P
os
t
har
v
e
s
t
B
i
ol
ogy
and
T
e
c
hnol
ogy
,
vol
.
156,
2019,
doi
:
10.1016/
j
.pos
t
ha
r
vbi
o.2019.110936.
[
3]
H
a
m
da
ni
,
A
.
S
e
pt
i
a
r
i
ni
,
a
nd
D
.
M
.
K
ha
i
r
i
na
,
“
M
ode
l
a
s
s
e
s
s
m
e
nt
of
l
a
nd
s
ui
t
a
bi
l
i
t
y
de
c
i
s
i
on
m
a
ki
ng
f
or
oi
l
p
a
l
m
pl
a
nt
a
t
i
on,”
i
n
2016 2nd I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
Sc
i
e
nc
e
i
n
I
nf
or
m
at
i
on T
e
c
hnol
ogy
, I
C
S
I
T
e
c
h 2016:
I
nf
or
m
at
i
on Sc
i
e
nc
e
f
or
G
r
e
e
n Soc
i
e
t
y
and E
nv
i
r
onm
e
nt
, I
E
E
E
, 2017, pp. 109
–
113
, do
i
:
10.1109/
I
C
S
I
T
e
c
h.2016.7852617.
[
4]
A
.
Y
udha
na
,
R
.
U
m
a
r
,
a
nd
F
.
M
.
A
yude
w
i
,
“
T
he
m
oni
t
or
i
ng
of
c
or
n
s
pr
out
s
gr
ow
t
h
us
i
ng
t
he
r
e
gi
on
gr
ow
i
ng
m
e
t
hods
,”
i
n
J
our
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
S
e
r
i
e
s
, I
O
P
P
ubl
i
s
hi
ng, N
ov. 2019, doi
:
10.1088/
1742
-
6596/
1373/
1/
012054.
[
5]
A
.
S
e
z
g
i
n
a
n
d
V
.
K
ü
ç
ü
k
,
“
C
o
m
p
u
t
e
r
s
c
i
e
n
c
e
m
o
n
i
t
o
r
i
n
g
p
l
a
n
t
g
r
o
w
t
h
w
i
t
h
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
m
e
t
h
o
d
s
a
n
d
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
s
u
p
p
o
r
t
e
d
a
g
r
i
c
u
l
t
u
r
e
s
y
s
t
e
m
,
”
i
n
I
n
t
e
r
n
a
t
i
o
n
a
l
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
D
a
t
a
P
r
o
c
e
s
s
i
n
g
S
y
m
p
o
s
i
u
m
,
2
0
2
2
,
p
p
.
1
6
5
-
1
7
6
,
d
o
i
:
1
0
.
5
3
0
7
0
/
b
b
d
.
1
1
7
2
7
7
4
.
[
6]
B
.
A
ç
ı
ka
l
ı
n
a
nd
N
.
S
a
nl
i
e
r
,
“
C
of
f
e
e
a
nd
i
t
s
e
f
f
e
c
t
s
on
t
he
i
m
m
une
s
ys
t
e
m
,”
T
r
e
nds
i
n
F
ood
Sc
i
e
nc
e
&
T
e
c
hnol
ogy
,
vol
.
114,
pp.
625
–
632, A
ug. 2021, doi
:
10.1016/
j
.t
i
f
s
.2021.06.023.
[
7]
E
.
P
i
e
da
d,
J
.
I
.
L
a
r
a
da
,
G
.
J
.
P
oj
a
s
,
a
nd
L
.
V
.
V
F
e
r
r
e
r
,
“
P
os
t
ha
r
ve
s
t
c
l
a
s
s
i
f
i
c
a
t
i
on
of
ba
na
na
(
M
us
a
a
c
um
i
na
t
a
)
us
i
ng
t
i
e
r
-
ba
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng,”
P
os
t
har
v
e
s
t
B
i
ol
ogy
and T
e
c
hnol
ogy
, vol
. 145, pp. 93
–
100, 2018, doi
:
10.1016/
j
.pos
t
ha
r
vbi
o.2018.06.004.
[
8]
A
.
N
ovi
ya
nt
o
a
nd
W
.
H
.
A
bdul
l
a
,
“
H
one
y
bot
a
ni
c
a
l
o
r
i
gi
n
c
l
a
s
s
i
f
i
c
a
t
i
on
us
i
ng
hype
r
s
pe
c
t
r
a
l
i
m
a
gi
ng
a
nd
m
a
c
hi
ne
l
e
a
r
ni
ng,”
J
our
nal
of
F
ood E
ngi
ne
e
r
i
ng
, vol
. 265, 2020, doi
:
10.1016/
j
.j
f
oode
ng.2019.109
684.
[
9]
D
.
Z
ha
ng,
D
.
J
.
L
e
e
,
B
.
J
.
T
i
ppe
t
t
s
,
a
nd
K
.
D
.
L
i
l
l
yw
hi
t
e
,
“
D
a
t
e
m
a
t
ur
i
t
y
a
nd
qua
l
i
t
y
e
va
l
ua
t
i
on
us
i
ng
c
ol
or
di
s
t
r
i
but
i
on
a
na
l
ys
i
s
a
nd ba
c
k pr
oj
e
c
t
i
on,”
J
our
nal
of
F
ood E
ngi
ne
e
r
i
ng
, vol
. 131, pp. 161
–
169, 201
4, doi
:
10.1016/
j
.j
f
oode
ng.2014.02.002.
[
10]
A
. S
e
p
t
i
a
r
i
n
i
,
H
.
H
a
m
da
ni
, T
.
H
a
r
d
i
a
n
t
i
, E
.
W
i
na
r
n
o,
S
. S
uy
a
nt
o,
a
n
d E
.
I
r
w
a
ns
y
a
h, “
P
i
xe
l
qua
nt
i
f
i
c
a
t
i
on
a
n
d
c
ol
o
r
f
e
a
t
ur
e
e
x
t
r
a
c
t
i
on
on
l
e
a
f
i
m
a
ge
s
f
o
r
oi
l
pa
l
m
di
s
e
a
s
e
i
d
e
nt
i
f
i
c
a
t
i
on
,”
i
n
7
t
h
I
n
t
e
r
n
at
i
on
al
C
o
nf
e
r
e
nc
e
on
E
l
e
c
t
r
i
c
a
l
,
E
l
e
c
t
r
o
ni
c
s
a
nd
I
n
f
or
m
at
i
on
E
n
gi
ne
e
r
i
ng
:
T
e
c
hn
ol
o
gi
c
al
B
r
e
ak
t
hr
ou
gh
f
or
G
r
e
a
t
e
r
N
e
w
L
i
f
e
,
2
021
,
pp.
1
–
5
,
d
oi
:
1
0.1
10
9/
I
C
E
E
I
E
526
63
.20
21
.96
16
645
.
[
11]
W
.
G
.
D.
C
os
t
a
,
I
.
D.
P
.
B
a
r
bos
a
,
J
.
E
.
D.
S
ouz
a
,
C
.
D
.
C
r
uz
,
M
.
N
a
s
c
i
m
e
nt
o,
a
nd
A
.
C
.
B
.
D.
O
l
i
ve
i
r
a
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
a
n
d
s
t
a
t
i
s
t
i
c
s
t
o qua
l
i
f
y e
nvi
r
onm
e
nt
s
t
hr
ough m
ul
t
i
-
t
r
a
i
t
s
i
n C
of
f
e
a
a
r
a
bi
c
a
,”
P
L
oS
O
N
E
, vol
. 16, no.
1, pp.
e
0245298
–
e
0245298, J
a
n.
2021, doi
:
10.1371/
j
our
na
l
.pone
.0245298.
[
12]
F
.
F
.
L
.
D
.
S
a
n
t
os
,
J
. T
.
F
. R
os
a
s
, R
. N
. M
a
r
t
i
ns
,
G
.
D.
M
.
A
r
a
ú
j
o
,
L
.
D.
A
. V
i
a
n
a
,
a
n
d
J
.
D.
P
. G
on
ç
a
l
v
e
s
,
“
Q
ua
l
i
t
y
a
s
s
e
s
s
m
e
n
t
o
f
c
o
f
f
e
e
b
e
a
ns
t
h
r
ou
g
h
c
o
m
p
u
t
e
r
v
i
s
i
o
n
a
n
d
m
a
c
h
i
n
e
l
e
a
r
ni
n
g
a
l
g
o
r
i
t
h
m
s
,”
C
o
f
f
e
e
S
c
i
e
n
c
e
,
vo
l
.
1
5
,
n
o
.
1,
p
p
.
1
–
9
,
2
0
2
0,
d
o
i
:
1
0
.
25
1
8
6
/
.
v
1
5
i
.1
7
5
2
.
[
13]
A
.
S
e
p
t
i
a
r
i
n
i
,
H
.
H
a
m
da
n
i
,
A
.
R
i
f
a
ni
,
Z
.
A
r
i
f
i
n
,
N
.
H
i
d
a
y
a
t
,
a
n
d
H
.
I
s
m
a
n
t
o
,
“
M
ul
t
i
-
c
l
a
s
s
s
up
po
r
t
ve
c
t
o
r
m
a
c
h
i
n
e
f
o
r
a
r
a
b
i
c
a
c
o
f
f
e
e
b
e
a
n
r
o
a
s
t
i
ng
g
r
a
de
c
l
a
s
s
i
f
i
c
a
t
i
on
,”
i
n
I
C
O
I
A
C
T
2
02
2
-
5
t
h
I
nt
e
r
n
a
t
i
o
na
l
C
o
n
f
e
r
e
nc
e
o
n I
n
f
or
m
at
i
on
a
n
d C
o
m
m
un
i
c
a
t
i
o
ns
T
e
c
hn
o
l
o
gy
:
A
N
e
w
W
ay
t
o
M
a
k
e
A
I
U
s
e
f
u
l
f
or
E
v
e
r
y
on
e
i
n
t
he
N
e
w
N
or
m
a
l
E
r
a
,
2
02
2
,
pp
.
4
07
–
4
11
,
d
o
i
:
1
0
.1
10
9
/
I
C
O
I
A
C
T
55
5
06
.2
0
22
.9
9
71
89
7
.
[
14]
R
.
S
.
E
l
-
S
a
y
e
d
a
n
d
M
.
N
.
E
l
-
S
a
y
e
d
,
“
C
l
a
s
s
i
f
i
c
a
t
i
on
o
f
v
e
h
i
c
l
e
s
’
t
yp
e
s
us
i
ng
hi
s
t
o
g
r
a
m
o
r
i
e
n
t
e
d
g
r
a
d
i
e
nt
s
:
c
o
m
p
a
r
a
t
i
ve
s
t
ud
y
a
nd
m
o
d
i
f
i
c
a
t
i
o
n,
”
I
A
E
S
I
n
t
e
r
n
at
i
on
al
J
o
ur
n
a
l
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
ge
nc
e
,
v
o
l
.
9
,
n
o.
4,
pp
.
7
0
0
–
71
2
,
20
2
0,
d
o
i
:
10
.1
1
59
1
/
i
j
a
i
.v
9.
i
4.
p
p7
00
-
71
2
.
[
15]
M
.
S
ha
r
i
f
,
M
.
A
.
K
ha
n,
Z
.
I
qba
l
,
M
.
F
.
A
z
a
m
,
M
.
I
.
U
.
L
a
l
i
,
a
nd
M
.
Y
.
J
a
ve
d,
“
D
e
t
e
c
t
i
on
a
nd
c
l
a
s
s
i
f
i
c
a
t
i
on
of
c
i
t
r
us
di
s
e
a
s
e
s
i
n
a
gr
i
c
ul
t
ur
e
ba
s
e
d on opt
i
m
i
z
e
d w
e
i
ght
e
d
s
e
gm
e
nt
a
t
i
on
a
nd f
e
a
t
ur
e
s
e
l
e
c
t
i
on,”
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
.
150,
pp. 220
–
234, 2018, doi
:
10.1016/
j
.c
om
pa
g.2018.04.023.
[
16]
A
.
S
e
pt
i
a
r
i
ni
,
H
.
H
a
m
da
ni
,
S
.
U
.
S
a
r
i
,
H
.
R
a
hm
a
ni
a
H
a
t
t
a
,
N
.
P
us
pi
t
a
s
a
r
i
,
a
nd
W
.
H
a
di
kur
ni
a
w
a
t
i
,
“
I
m
a
ge
pr
oc
e
s
s
i
ng
t
e
c
hni
que
s
f
or
t
om
a
t
o
s
e
gm
e
nt
a
t
i
on
a
ppl
yi
ng
k
-
m
e
a
ns
c
l
us
t
e
r
i
ng
a
nd
e
dge
de
t
e
c
t
i
on
a
ppr
oa
c
h
,”
i
n
2021
I
nt
e
r
nat
i
onal
S
e
m
i
nar
on
M
ac
hi
ne
L
e
ar
ni
ng, O
pt
i
m
i
z
at
i
on, and D
at
a Sc
i
e
nc
e
, I
SM
O
D
E
2021
, I
E
E
E
, 2022, pp. 92
–
96
, doi
:
10.1109/
I
S
M
O
D
E
53584.2022.9742740.
[
17]
J
.
L
u
e
t
al
.
,
“
L
i
ght
w
e
i
ght
gr
e
e
n
c
i
t
r
us
f
r
ui
t
de
t
e
c
t
i
on
m
e
t
hod
f
or
pr
a
c
t
i
c
a
l
e
nvi
r
onm
e
nt
a
l
a
ppl
i
c
a
t
i
ons
,”
C
om
put
e
r
s
an
d
E
l
e
c
t
r
oni
c
s
i
n A
gr
i
c
ul
t
ur
e
, vol
. 215, 2023, doi
:
10.1016/
j
.c
om
pa
g.2023.108205
.
[
18]
J
.
L
i
a
ng,
K
.
H
ua
ng,
H
.
L
e
i
,
Z
.
Z
hong,
Y
.
C
a
i
,
a
nd
Z
.
J
i
a
o,
“
O
c
c
l
us
i
on
-
a
w
a
r
e
f
r
ui
t
s
e
gm
e
nt
a
t
i
on
i
n
c
om
pl
e
x
na
t
ur
a
l
e
nvi
r
onm
e
nt
s
unde
r
s
ha
pe
pr
i
or
,”
C
om
put
e
r
s
and E
l
e
c
t
r
oni
c
s
i
n A
gr
i
c
ul
t
u
r
e
, vol
. 217, 2024,
doi
:
10.1016/
j
.c
om
pa
g.2024.108620.
[
19]
X
.
Y
a
ng,
R
.
Z
ha
ng,
Z
.
Z
ha
i
,
Y
.
P
a
ng,
a
nd
Z
.
J
i
n,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
f
or
c
ul
t
i
v
a
r
c
l
a
s
s
i
f
i
c
a
t
i
on
of
a
pr
i
c
ot
s
(
P
r
unus
a
r
m
e
ni
a
c
a
L
.)
ba
s
e
d on s
h
a
pe
f
e
a
t
ur
e
s
,”
Sc
i
e
nt
i
a H
o
r
t
i
c
ul
t
ur
ae
, vol
. 256, 2019, doi
:
10.1016/
j
.s
c
i
e
nt
a
.2019.05.051.
[
20]
H
.
L
i
,
W
.
S
.
L
e
e
,
a
nd
K
.
W
a
ng,
“
I
de
nt
i
f
yi
ng
bl
ue
be
r
r
y
f
r
ui
t
of
di
f
f
e
r
e
nt
gr
ow
t
h
s
t
a
ge
s
u
s
i
ng
na
t
ur
a
l
out
door
c
ol
or
i
m
a
ge
s
,”
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
. 106, pp. 91
–
101, 2014, doi
:
10.1016/
j
.c
om
pa
g.2014.05.015.
[
21]
G
a
r
c
í
a
,
C
.
B
e
c
e
r
r
a
,
a
nd
H
oyos
,
“
Q
ua
l
i
t
y
a
nd
de
f
e
c
t
i
ns
pe
c
t
i
on
of
gr
e
e
n
c
of
f
e
e
be
a
ns
us
i
ng
a
c
om
put
e
r
vi
s
i
on
s
y
s
t
e
m
,”
A
ppl
i
e
d
Sc
i
e
nc
e
s
, vol
. 9, no. 19, O
c
t
. 2019, doi
:
10.3390/
a
pp9194195.
[
22]
H
.
C
.
B
a
z
a
m
e
,
J
.
P
.
M
ol
i
n,
D
.
A
l
t
ho
f
f
,
a
nd
M
.
M
a
r
t
e
l
l
o,
“
D
e
t
e
c
t
i
on,
c
l
a
s
s
i
f
i
c
a
t
i
on,
a
nd
m
a
ppi
ng
of
c
of
f
e
e
f
r
ui
t
s
dur
i
ng
ha
r
ve
s
t
w
i
t
h c
om
put
e
r
vi
s
i
on,”
C
om
put
e
r
s
and E
l
e
c
t
r
oni
c
s
i
n A
gr
i
c
ul
t
u
r
e
, vol
. 183, 20
21, doi
:
10.1016/
j
.c
om
pa
g.2021.106066.
[
23]
F
.
G
.
-
L
a
m
ont
,
J
.
C
e
r
va
nt
e
s
,
A
.
L
óp
e
z
,
a
nd
L
.
R
odr
i
gue
z
,
“
S
e
gm
e
nt
a
t
i
on
of
i
m
a
ge
s
by
c
ol
or
f
e
a
t
ur
e
s
:
A
s
ur
ve
y,
”
N
e
ur
oc
om
put
i
ng
, vol
. 292, pp. 1
–
27, 2018, doi
:
10.1016/
j
.ne
uc
om
.2018.01.091
.
[
24]
N
.
D
ha
na
c
ha
ndr
a
,
K
.
M
a
ngl
e
m
,
a
nd
Y
.
J
.
C
ha
nu,
“
I
m
a
ge
s
e
gm
e
nt
a
t
i
on
u
s
i
ng
K
-
m
e
a
ns
c
l
us
t
e
r
i
ng
a
l
gor
i
t
hm
a
nd
s
ubt
r
a
c
t
i
v
e
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
.
13
,
N
o.
4
,
D
e
c
e
m
be
r
2024
:
4241
-
4248
4248
c
l
us
t
e
r
i
ng a
l
gor
i
t
hm
,”
P
r
oc
e
di
a C
om
put
e
r
Sc
i
e
n
c
e
, vol
. 54, pp. 764
–
771, 2015,
doi
:
10.1016/
j
.pr
oc
s
.2015.06.090.
[
25]
A
.
S
e
pt
i
a
r
i
ni
,
R
.
S
a
put
r
a
,
A
.
T
e
dj
a
w
a
t
i
,
M
.
W
a
t
i
,
a
nd
H
.
H
a
m
da
ni
,
“
P
a
t
t
e
r
n
r
e
c
ogni
t
i
on
of
s
a
r
ong
f
a
br
i
c
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h
ba
s
e
d
on
c
om
put
e
r
vi
s
i
on
f
or
c
ul
t
ur
a
l
p
r
e
s
e
r
va
t
i
on
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
I
nt
e
l
l
i
ge
nt
E
ngi
ne
e
r
i
ng
and
Sy
s
t
e
m
s
,
vol
.
15, no. 5, pp. 284
–
295, 2022, doi
:
10.22266/
i
j
i
e
s
2022.1031.26.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Anindita
Septiarini
is
a
professor
at
the
Department
of
Informatics
at
Mulawarman
University,
Indonesia.
She
holds
a
Doctoral
degree
in
Computer
Science
from
Gadjah
Mada
University,
Indonesia,
specializing
in
image
analysis.
She
is
also
a
researcher
and
got
a
grant
from
the
Ministry
of
Education,
Culture,
Resear
ch,
and
Technology
o
f
Indonesia
from
2016
until
the
present.
Her
research
interests
li
e
i
n
artificial
intellig
ence,
especiall
y
pattern
recogniti
on,
image
processin
g,
and
computer
vi
sion.
She
has
received
national
awards
such
as
scientific
article
incentives
from
the
Ministr
y
of
E
ducation,
Culture,
Resear
ch,
and
Technol
ogy
of
Indones
ia
in
2017
and
2019.
She
hel
d
severa
l
administra
tive
posts
with
the
Department
of
Informatics,
Mulawarman
University,
I
ndonesia,
from
2018
to
2020,
including
the
head
of
department
and
the
head
of
laboratory
.
She
can
be
contacted
at
email:
anindit
a@
unmul.
ac.id.
Hamdani
Hamdani
is
a
professor
in
Department
of
Informatics
at
Mulawarman
University,
Indonesia.
He
is
a
lecturer
and
researcher
since
2005
at
Mulawarman
University,
Indonesia.
His
research
interests
lie
in
the
field
of
artificial
intellig
ence,
especially
pattern
recognition
,
decision
support
system
,
and
expert
system.
He
r
eceived her
bachelor’s
degree
in
2002
from
Ahmad
Dahlan
University,
Indonesia,
his
master’s
degre
e
in
2009
from
Gadjah
Mada
University,
Indonesia,
and
her
doctoral
degree
in
computer
science
in
2018
from
Gadjah Mada University
Indonesia. He can be contacted at
email:
hamdani@
unmul.
ac.id.
Aji
Ery
Burhandenny
is
an
assistant
professor
in
Department
of
Electrical
Engineering
at
Mulawarman
University,
Indonesia,
specializing
in
So
ftware
Engineering.
His
research
interest
s
lie
in
empirical
software
engineering
particularl
y
understanding
human
factors
in
software
development
activities
.
He
also
made
several
con
tributio
ns
to
internet
of
things,
machine
learning,
and
green
technologie
s
related
topics.
He
can
be
contacted
at
email:
a.burhanden
ny@
ft.unmu
l.ac.id
.
Damar
Nurcahyono
is
an
assistant
professor
at
The
Informatio
n
Technology
Department
of
the
State
Polytechnic
of
Samarinda,
Indonesia,
s
pecializing
in
Software
Engineering.
His
research
interests
lie
in
software
engineering
especially
games
and
animati
on.
He
has
also
made
several
contribu
tions
to
softwar
e
developm
ent,
game
development,
and
environmental
technology
-
related
topics.
He
can
be
contacted
at
email:
damarnc@pol
nes.ac.id
.
Surya
Eka
Priyatna
is
a
lecturer
at
the
Department
of
Information
Engineering,
Antasari
State
Islamic
University,
Indonesia,
specializing
in
mobile
application
design.
His
research
interests
lie
in
human
needs
in
the
use
of
computer
applicat
ions.
He
has
also
made
several
contributions
on
topics
related
to
mobile
applications.
He
c
a
n
be
contacted
at
email:
suryaekapriya
tna@
uin
-
antasari.ac.
id
.
Evaluation Warning : The document was created with Spire.PDF for Python.