T
E
L
KO
M
N
I
KA
T
e
lec
om
m
u
n
icat
ion
,
Com
p
u
t
i
n
g,
E
lec
t
r
on
ics
an
d
Cont
r
ol
Vol.
1
8
,
No.
1
,
F
e
br
ua
r
y
2020
,
pp.
140
~
147
I
S
S
N:
1693
-
6930,
a
c
c
r
e
dit
e
d
F
ir
s
t
G
r
a
de
by
Ke
me
nr
is
tekdikti
,
De
c
r
e
e
No:
21/E
/KP
T
/2018
DO
I
:
10.
12928/
T
E
L
KO
M
NI
KA
.
v1
8
i
1
.
13039
140
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
A
ugm
ent
ed
real
i
t
y
usi
ng
f
eat
ures
acce
l
erat
ed
s
egm
ent
t
es
t
f
or
propert
y
cat
al
ogue
Rud
y
S
e
t
yad
i
,
I
n
d
r
a
Ranggad
ar
a
Facu
l
t
y
o
f
C
o
mp
u
t
er
Sci
e
n
ce,
Mercu
Bu
a
n
a
U
n
i
v
ers
i
t
y
,
In
d
o
n
e
s
i
a
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
M
a
y
5
,
2019
R
e
vis
e
d
Ju
n
2
3
,
20
19
Ac
c
e
pted
Ju
l
9
,
20
19
Pro
mo
t
i
o
n
a
l
med
i
a
u
s
ed
i
n
t
h
e
m
ark
e
t
i
n
g
o
f
h
o
u
s
i
n
g
u
s
i
n
g
cat
a
l
o
g
s
t
h
a
t
d
i
s
p
l
ay
2
D
i
mag
es
o
f
h
o
u
s
es
fro
m
o
n
e
s
i
d
e
o
f
t
h
e
h
o
u
s
e
mak
e
p
o
t
e
n
t
i
al
cu
s
t
o
mer
s
u
n
ab
l
e
t
o
i
ma
g
i
n
e
t
h
e
d
e
s
i
g
n
o
f
al
l
p
art
s
o
f
t
h
e
h
o
u
s
e.
A
u
g
men
t
ed
Real
i
t
y
ca
n
b
e
u
s
e
d
as
a
n
i
n
t
erac
t
i
v
e
mar
k
et
i
n
g
med
i
a
s
o
t
h
a
t
i
t
can
b
e
u
s
ed
t
o
d
i
s
p
l
ay
h
o
mes
i
n
3
D
s
o
t
h
a
t
t
h
ey
a
p
p
ear
m
o
re
real
fr
o
m
al
l
s
i
d
es
s
o
t
h
a
t
p
ro
s
p
ec
t
i
v
e
cu
s
t
o
mer
s
can
co
n
s
i
d
er
t
h
e
t
y
p
e
o
f
h
o
u
s
e
t
o
b
e
ch
o
s
en
.
D
ev
e
l
o
p
men
t
o
f
t
h
i
s
ap
p
l
i
cat
i
o
n
u
s
i
n
g
t
h
e
Mu
l
t
i
med
i
a
D
ev
e
l
o
p
men
t
L
i
fe
Cy
cl
e.
A
p
p
l
i
ca
t
i
o
n
d
ev
e
l
o
p
men
t
u
s
es
t
h
e
F
A
ST
a
l
g
o
r
i
t
h
m
a
s
d
e
t
ect
i
o
n
o
f
h
o
me
ca
t
al
o
g
mar
k
ers
t
o
d
ef
i
n
e
h
o
w
w
e
l
l
i
mag
e
s
ca
n
b
e
d
et
ec
t
ed
a
n
d
t
rack
e
d
.
T
h
e
FA
S
T
al
g
o
r
i
t
h
m
w
i
l
l
cal
cu
l
at
e
e
v
ery
p
i
x
el
o
n
t
h
e
t
ar
g
e
t
i
mag
e
i
n
d
et
erm
i
n
i
n
g
t
h
e
co
r
n
er
w
h
en
s
can
n
i
n
g
t
h
e
h
o
me
ca
t
al
o
g
t
h
e
n
i
t
w
i
l
l
p
r
o
d
u
ce
a
3
D
o
b
j
ec
t
h
o
me
t
o
s
ee
t
h
e
rea
l
s
h
a
p
e
d
e
s
i
g
n
o
f
t
h
e
h
o
u
s
e
.
K
e
y
w
o
r
d
s
:
Augme
nted
r
e
a
li
ty
F
AST
P
r
ope
r
ty
Unity
Vuf
or
ia
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e
.
C
or
r
e
s
pon
din
g
A
u
th
or
:
R
udy
S
e
tyadi
,
F
a
c
ult
y
of
C
omput
e
r
S
c
ienc
e
,
M
e
r
c
u
B
ua
na
Unive
r
s
it
y,
R
a
ya
M
e
r
uya
S
e
lata
n
S
t.
,
Ke
mbanga
n,
J
a
ka
r
ta
116
50,
I
ndone
s
ia.
E
mail:
41814120140@s
tudent.
mer
c
ubua
na
.
a
c
.
id
1.
I
NT
RODU
C
T
I
ON
Hous
ing
r
e
pr
e
s
e
nts
the
mos
t
ba
s
ic
of
human
ne
e
d
a
nd
it
ha
s
a
p
r
of
ound
im
pa
c
t
on
the
he
a
lt
h,
we
lf
a
r
e
,
a
nd
pr
oduc
ti
vit
y
o
f
indi
v
iduals
[
1]
.
Hous
e
s
a
s
s
he
lt
e
r
s
s
hould
be
a
bl
e
to
f
ulf
i
ll
the
s
pa
c
e
r
e
quir
e
ments
f
or
a
c
ti
vit
ies
of
their
r
e
s
idents
[
2
]
.
T
he
na
tur
e
o
f
ho
us
ing
include
s
not
only
hous
e
s
f
r
om
the
phys
ica
l
s
ide
of
the
buil
ding
but
include
s
a
ll
s
uppor
ti
ng
f
a
c
il
it
ies
both
ins
ide
a
nd
ou
ts
ide
[
3
]
.
Along
with
the
hig
h
mar
ke
t
de
mand
f
or
the
p
r
ope
r
ty
bus
ines
s
[
4]
,
many
c
ompani
e
s
of
f
e
r
their
p
r
oduc
ts
us
ing
va
r
ious
methods
.
I
n
the
bus
ines
s
wor
ld,
c
ompetit
ion
be
twe
e
n
c
ompanie
s
is
incr
e
a
s
ing,
ther
e
f
or
e
e
ve
r
y
c
ompany
is
r
e
quir
e
d
to
make
c
ha
nge
s
to
a
c
hieve
a
s
uc
c
e
s
s
in
doing
bus
i
ne
s
s
[
5]
.
P
r
omot
ion
is
one
of
the
c
omponents
tha
t
f
or
m
a
mar
ke
ti
ng
mi
x
a
nd
mar
ke
ti
ng
a
c
ti
vit
ies
will
f
oc
us
on
c
omm
unica
ti
ng
[
6]
.
E
xa
mpl
e
s
of
p
r
omot
ional
media
that
a
r
e
c
a
r
r
ied
out
is
to
of
f
e
r
homes
to
c
ons
umer
s
by
s
howing
a
c
a
talog
that
c
ontains
in
f
or
mation
o
r
e
xpl
a
na
ti
ons
a
bout
the
a
ppe
a
r
a
nc
e
of
2D
hous
e
s
im
a
ge
s
a
nd
the
de
s
ign
of
the
s
ize
of
the
hous
e
.
W
hil
e
doing
pr
ope
r
ty
mar
ke
ti
ng,
the
obs
tac
les
f
a
c
e
d
by
s
a
les
mar
ke
ti
ng
whe
n
pr
os
pe
c
ti
ve
c
us
tom
e
r
s
a
r
e
c
hoos
ing
a
hous
e
a
r
e
s
ti
ll
c
onf
us
e
d
a
bout
the
s
ha
pe
of
the
hous
e
be
ing
buil
t.
T
his
is
be
c
a
us
e
the
h
ous
e
that
is
dis
playe
d
f
r
om
a
c
a
talog
that
only
c
ontains
a
pic
tur
e
of
one
s
ide
of
the
hous
e
doe
s
not
ye
t
look
de
ta
il
e
d
a
nd
r
e
a
l,
bo
th
outs
ide
a
nd
ins
ide
the
hous
e
.
C
ons
u
mer
s
who
wa
nt
to
s
e
e
the
loca
ti
on
of
the
hous
e
dir
e
c
tl
y,
s
ometim
e
s
do
not
ha
ve
ti
me
due
to
he
a
vy
a
c
ti
vit
y,
c
oupled
with
the
loca
ti
on
of
the
hous
e
is
f
a
r
f
r
om
the
loca
ti
on
of
c
ons
umer
s
a
t
thi
s
ti
me.
F
igur
e
1
s
how
da
ta
on
pr
os
pe
c
ti
ve
c
us
tom
e
r
s
who
vis
it
the
loca
ti
on
of
th
e
hous
e
.
F
r
om
the
da
ta
in
F
igur
e
1
s
hows
the
number
o
f
p
r
o
s
pe
c
ti
ve
c
us
tom
e
r
s
who
vis
it
hous
ing
loca
ti
ons
.
P
r
os
pe
c
ti
ve
c
ons
umer
s
look
dir
e
c
tl
y
a
t
the
hous
ing
loca
ti
on
be
c
a
us
e
they
wa
nt
to
s
e
e
the
s
ha
pe
a
nd
s
ize
of
the
hous
e
in
r
e
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
A
ugme
nted
r
e
ali
ty
us
ing
featur
e
s
ac
c
e
ler
ated
s
e
gm
e
nt
tes
t
for
pr
ope
r
ty
c
atal
ogue
(
R
udy
Se
tyadi
)
141
ter
ms
c
ompar
e
d
to
jus
t
looki
ng
a
t
the
pictur
e
s
in
t
he
c
a
talog.
I
mage
s
c
ontaine
d
in
the
c
a
talog
a
r
e
o
nly
f
r
om
the
f
r
ont
s
ide
of
the
hous
e
a
nd
the
im
a
ge
of
hous
e
plans
.
Augme
nted
r
e
a
li
ty
is
a
tec
h
nology
that
c
ombi
n
e
s
two
-
dim
e
ns
ional
or
thr
e
e
-
dim
e
ns
ional
vir
tual
objec
ts
int
o
a
r
e
a
l
e
nvir
onment
in
r
e
a
l
ti
me
[
7]
s
o
t
he
r
e
is
no
bounda
r
y
be
twe
e
n
the
r
e
a
l
wor
ld
a
nd
th
e
vir
tual
wor
ld
[
8
]
.
T
he
de
ve
lopm
e
nt
of
a
ugmente
d
r
e
a
li
t
y
a
lr
e
a
dy
e
xis
ts
in
s
e
ve
r
a
l
s
e
c
tor
s
s
uc
h
a
s
m
e
dic
[
9]
,
e
duc
a
ti
ons
[
10]
,
mi
li
tar
y
[
11]
,
ga
mes
[
12]
a
nd
ma
nuf
a
c
tur
e
r
[
13]
.
T
he
Augme
nted
R
e
a
li
ty
s
ys
tem
will
s
c
a
n
the
e
nvir
onment
whic
h
ha
s
a
mar
ke
r
that
will
be
im
a
ge
d
by
a
vir
tual
objec
t
.
A
ma
r
ke
r
is
a
pa
tt
e
r
n
made
in
the
f
or
m
of
im
a
ge
s
a
nd
c
a
n
be
r
e
c
ognize
d
by
opti
c
a
l
de
vice
s
or
c
a
mer
a
s
in
the
a
ugmente
d
r
e
a
li
ty
method
[
14]
.
In
thi
s
s
ys
tem
c
a
mer
a
c
onti
nuous
ly
s
na
ps
hot
s
the
tar
ge
t
objec
t
a
nd
pr
oc
e
s
s
the
im
a
ge
to
e
s
ti
mate
the
pos
it
ion,
or
ienta
ti
on
a
nd
moveme
nt
o
f
the
vis
ua
li
z
a
ti
on
dis
p
lay
with
r
e
s
pe
c
t
to
the
tar
ge
t
objec
t
[
15]
.
T
he
de
ve
lopm
e
nt
of
Augme
nted
R
e
a
li
ty
tec
hnology
in
the
home
c
a
talog,
will
make
c
ons
umer
s
s
e
e
the
3D
dis
play
o
f
hous
e
s
c
ontaine
d
in
the
c
a
talog
s
o
that
the
dis
playe
d
hou
s
e
will
look
mor
e
de
tailed
a
nd
r
e
a
l
f
r
om
a
ll
s
ides
by
us
ing
br
oc
hur
e
s
a
s
mar
ke
r
s
.
T
his
is
c
onf
ir
med
by
Vitono
[
16]
ha
s
e
xa
mi
ne
d
the
us
e
of
the
F
AST
a
lgor
it
hm
in
mar
ke
r
les
s
a
ugmente
d
r
e
a
li
ty
to
us
e
it
a
s
a
n
inf
or
mation
media
mus
e
um
c
oll
e
c
ti
on.
T
he
n
Z
uli
[
17
]
a
ls
o
us
e
s
th
e
F
AST
a
lgor
it
hm
a
s
a
3D
inf
o
r
mation
media
a
t
the
Unive
r
s
it
y
of
S
a
tya
Ne
ga
r
a
I
ndone
s
ia.
T
he
n
in
2018
S
e
ti
a
wa
n
[
14]
a
ls
o
e
xa
mi
ne
d
the
us
e
of
a
ugmente
d
r
e
a
li
ty
w
it
h
the
F
AS
T
a
lgo
r
it
hm
in
the
a
blut
ion
manua
l.
T
he
n
Adiwijaya
[
18]
us
ing
F
AS
T
c
or
ne
r
d
e
tec
ti
on
to
a
ugmente
d
r
e
a
li
ty
im
pleme
ntation
f
o
r
int
e
r
a
c
ti
ve
b
r
oc
hur
e
.
Als
o
in
Na
inggol
a
n
[
19]
T
he
im
pleme
ntation
of
a
u
gmente
d
r
e
a
li
ty
in
the
lea
r
ning
media
of
int
r
oduc
ti
o
n
a
nim
a
l
c
a
n
give
the
vis
ua
l
inf
o
r
mation
int
e
r
a
c
ti
ve
ly.
F
r
om
p
r
e
vious
r
e
s
e
a
r
c
h,
the
de
ve
lopm
e
nt
of
thi
s
r
e
s
e
a
r
c
h
c
a
r
r
ied
ou
t
a
n
idea
to
a
pply
the
F
AST
a
lgor
it
hm
to
a
ugmente
d
r
e
a
li
ty
in
it
s
us
e
a
s
a
hom
e
c
a
talog.
T
he
F
AS
T
a
lgor
it
hm
is
im
p
leme
nte
d
in
s
c
a
nning
AR
mar
ke
r
s
a
s
de
tec
ti
on
o
f
ho
me
c
a
talog
mar
ke
r
s
to
de
f
ine
how
we
ll
im
a
ge
s
c
a
n
be
de
tec
ted
a
nd
tr
a
c
ke
d.
Dif
f
e
r
e
nt
f
r
om
p
r
e
vious
r
e
s
e
a
r
c
h,
r
e
s
e
a
r
c
he
r
s
wil
l
a
dd
the
home
tour
f
e
a
tur
e
a
s
a
n
a
ddit
ional
mu
lt
im
e
dia
e
xpe
r
ienc
e
that
s
ur
r
ounds
the
hous
e
a
nd
goe
s
int
o
t
he
hous
e
to
s
e
e
in
de
tail
a
nd
mor
e
r
e
a
l.
W
it
h
the
a
p
pli
c
a
ti
on
of
the
a
ugmente
d
r
e
a
li
ty,
the
us
e
r
c
a
n
mor
e
e
a
s
il
y
ge
t
the
inf
or
mation
a
bout
the
home
in
the
f
or
m
o
f
3
D
models
by
a
im
ing
the
c
a
mer
a
to
the
mar
ke
r
on
the
c
a
talog.
F
igur
e
1.
Da
ta
on
pr
os
pe
c
ti
ve
c
us
tom
e
r
vis
it
s
to
r
e
s
idential
loca
ti
ons
2.
RE
S
E
AR
CH
M
E
T
HO
D
2.
1.
M
u
lt
im
e
d
ia
d
e
ve
lop
m
e
n
t
l
if
e
c
yc
le
T
his
s
tudy
us
e
s
the
M
DL
C
(
M
ult
im
e
dia
De
ve
lopm
e
nt
L
if
e
C
yc
le)
method
.
T
his
method
is
e
a
s
ier
to
unde
r
s
tand
a
nd
im
pleme
nt,
the
s
teps
a
r
e
c
lea
r
a
nd
e
a
s
y
to
f
oll
ow
,
s
tr
uc
tu
r
e
d
a
nd
s
e
que
nti
a
l
logi
c
a
ll
y,
a
nd
c
a
n
be
us
e
d
by
s
mall
de
ve
loper
s
.
Ac
c
or
ding
to
L
uther
i
n
Nur
a
ji
z
a
h
[
20
]
the
de
ve
lopm
e
nt
of
the
mul
ti
medi
a
method
wa
s
c
a
r
r
ied
out
ba
s
e
d
on
s
ix
s
tage
s
,
a
r
r
a
nge
d
s
ys
tema
ti
c
a
ll
y
a
s
f
oll
ows
:
−
T
he
f
i
r
s
t
s
tep
of
thi
s
method
is
C
onc
e
pt.
I
t
s
tar
ts
with
de
ter
mi
ning
goa
ls
,
us
e
r
s
,
types
of
mul
ti
medi
a
a
nd
ge
ne
r
a
l
s
pe
c
if
ica
ti
ons
.
T
he
pur
pos
e
of
the
a
ppli
c
a
ti
on
is
tailor
e
d
to
the
ne
e
ds
of
or
ga
niza
ti
ons
that
ne
e
d
it
s
o
that
mul
ti
media
in
f
or
m
a
ti
on
is
c
onve
ye
d.
−
T
he
s
e
c
ond
s
tep
is
De
s
ign,
a
matu
r
e
c
onc
e
pt
will
th
e
n
de
s
c
r
ibe
wha
t
will
be
done
a
t
th
is
s
tage
.
T
he
pu
r
pos
e
of
the
de
s
ign
pha
s
e
is
to
make
s
p
e
c
if
ica
ti
ons
r
e
ga
r
ding
the
s
tyl
e
,
s
ha
pe
,
a
ppe
a
r
a
nc
e
a
nd
mate
r
ial
r
e
quir
e
ments
f
o
r
de
ve
lopm
e
nt.
F
or
the
de
s
ign
of
m
a
r
ke
r
s
that
will
us
e
f
r
om
the
e
xis
ti
ng
home
c
a
talog
a
nd
f
or
the
de
s
ign
of
3D
objec
ts
c
a
n
be
de
s
c
r
ibed
with
the
s
ha
pe
of
the
hous
e
ba
s
e
d
on
s
pe
c
if
ica
ti
ons
on
m
a
nua
l
da
ta
r
e
tr
ieva
l.
−
T
he
thi
r
d
s
tep
is
M
a
ter
ial
C
oll
e
c
ti
ng,
na
mely
c
oll
e
c
ti
on
in
a
c
c
or
da
nc
e
with
the
mate
r
ials
ne
e
de
d
to
be
wor
ke
d
on,
a
mong
o
ther
s
,
im
a
ge
s
,
c
li
p
a
r
t,
icons
a
nd
3D
f
or
ms
of
int
e
r
io
r
de
s
ign
in
the
hous
e
a
nd
other
objec
ts
outs
ide
the
home
s
uc
h
a
s
tr
e
e
s
a
nd
plants
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
1
8
,
No
.
1
,
F
e
br
ua
r
y
2020
:
140
-
147
142
−
T
he
f
our
th
s
tep
is
As
s
e
mbl
y,
c
a
r
r
ied
out
making
the
wh
ole
mul
ti
media
mate
r
ial.
M
a
king
a
n
a
ppli
c
a
ti
on
ba
s
e
d
on
the
de
s
ign
that
ha
s
be
e
n
made
a
nd
the
mate
r
ial
that
ha
s
be
e
n
c
oll
e
c
ted.
At
thi
s
s
tage
us
e
the
B
lende
r
s
of
twa
r
e
to
c
r
e
a
te
3D
ob
jec
ts
.
Unity
3d
a
nd
Vuf
or
ia
f
o
r
a
ugmente
d
r
e
a
li
ty
.
T
he
f
i
f
th
s
tep
is
T
e
s
ti
ng,
the
tr
ial
is
done
a
f
ter
the
c
ompl
e
ti
on
o
f
t
he
manuf
a
c
tur
ing
pha
s
e
by
r
unning
the
a
ppli
c
a
ti
o
n
a
nd
tes
ti
ng
is
done
to
s
e
e
whe
ther
ther
e
is
a
n
e
r
r
or
o
r
n
ot.
T
e
s
ted
us
ing
the
F
AST
a
lgo
r
it
hm
to
tes
t
the
ma
r
ke
r
s
us
e
d
in
AR
.
−
T
he
las
t
s
tep
is
Dis
tr
ibut
ion.
I
f
the
tes
ti
ng
pha
s
e
h
a
s
be
e
n
c
ompl
e
ted
withou
t
e
r
r
or
,
the
s
ys
tem
is
r
e
a
dy
to
be
dis
tr
ibut
e
d
or
a
ppli
e
d
a
c
c
or
ding
to
the
pur
pos
e
o
f
thi
s
r
e
s
e
a
r
c
h.
T
he
a
ppli
c
a
ti
on
will
then
be
ope
r
a
ted
on
a
n
Andr
oid
de
vice
f
or
late
r
us
e
.
T
he
e
va
luation
is
a
ls
o
ve
r
y
much
ne
e
de
d
f
or
pr
oduc
t
de
ve
lopm
e
nt
that
ha
s
be
e
n
made
to
be
be
tt
e
r
a
s
a
n
input
f
o
r
the
s
tage
c
onc
e
pt
in
the
ne
xt
de
ve
lopm
e
nt.
T
he
s
ix
s
teps
of
the
mul
ti
media
de
ve
lopm
e
nt
li
f
e
c
yc
le
method
a
r
r
a
nge
d
s
ys
tema
ti
c
a
ll
y
a
r
e
s
hown
in
F
igur
e
2.
T
he
c
onc
e
pt
s
tage
mus
t
indee
d
be
the
f
i
r
s
t
thi
ng
done
[
21
]
.
F
igur
e
2
.
M
DL
C
s
tage
[
20]
2.
2.
F
AST
(
f
e
at
u
r
e
s
f
r
om
ac
c
e
ler
at
e
d
s
e
gm
e
n
t
t
e
s
t
)
algorit
h
m
Vuf
or
ia
us
ing
F
AS
T
(
F
e
a
tur
e
s
f
r
om
Ac
c
e
ler
a
ted
S
e
gment
T
e
s
t)
Algo
r
it
hm
C
or
ne
r
De
tec
ti
on
to
de
f
ine
how
we
ll
im
a
ge
s
c
a
n
be
de
tec
ted
a
nd
tr
a
c
ke
d
us
ing
[
14]
.
Vuf
or
ia
is
a
n
Augme
nted
R
e
a
li
ty
S
of
twa
r
e
De
ve
lopm
e
nt
Kit
(
S
DK
)
f
or
mobi
le
de
vice
s
t
ha
t
a
ll
ows
the
c
r
e
a
ti
on
of
Augme
nted
R
e
a
li
ty
a
ppli
c
a
ti
ons
[
17]
.
T
he
a
dva
ntage
s
of
us
ing
the
F
AST
a
lgor
it
hm
a
r
e
a
r
a
pid
ope
r
a
ti
on
a
nd
low
c
omput
a
ti
ons
c
ompar
e
d
to
other
c
or
ne
r
de
tec
tor
s
[
22]
with
the
c
ons
e
que
nc
e
of
r
e
du
c
ing
the
a
c
c
ur
a
c
y
of
a
ngle
de
tec
ti
on
[
17
]
.
I
n
Vuf
o
r
ia
ther
e
a
r
e
r
a
ti
ngs
d
is
playe
d
in
the
T
a
r
ge
t
M
a
na
ge
r
[
16]
a
nd
r
e
tur
n
f
o
r
e
a
c
h
upload
ta
r
ge
t
via
the
we
b
AP
I
.
R
a
ti
ngs
r
a
nge
f
r
o
m
0
to
5
f
o
r
e
a
c
h
pictur
e
given.
T
he
high
e
r
the
r
a
ti
ng
of
the
tar
ge
t
im
a
ge
,
the
s
tr
onge
r
the
d
e
tec
ti
on
a
nd
tr
a
c
king
c
a
pa
bil
it
ies
it
c
ontains
.
F
igur
e
3
s
how
1
-
s
tar
r
a
ti
ng
indi
c
a
tes
that
the
tar
ge
t
is
ha
r
d
to
t
r
a
c
ke
d
a
t
a
ll
by
the
Augme
nted
R
e
a
li
ty
s
ys
tem,
while
the
5
-
s
tar
r
a
ti
ng
s
hows
that
a
n
im
a
ge
i
s
e
a
s
il
y
tr
a
c
ke
d
by
a
n
Augme
nted
R
e
a
li
ty
s
ys
tem
be
c
a
us
e
many
of
ins
e
r
t
point
s
or
c
or
ne
r
s
on
a
n
im
a
ge
.
F
igur
e
3
.
T
he
a
ugmenta
ble
r
a
ti
ng
on
t
he
tar
ge
t
m
a
na
ge
r
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
A
ugme
nted
r
e
ali
ty
us
ing
featur
e
s
ac
c
e
ler
ated
s
e
gm
e
nt
tes
t
for
pr
ope
r
ty
c
atal
ogue
(
R
udy
Se
tyadi
)
143
F
AST
C
or
ne
r
De
tec
ti
on
is
de
tec
ti
on
by
look
ing
f
o
r
ins
e
r
t
poin
ts
or
c
or
ne
r
s
on
a
n
im
a
ge
[
14]
.
T
he
pr
oc
e
s
s
of
de
ter
m
ini
ng
the
a
ngle
is
by
c
ha
nging
the
im
a
ge
to
gr
a
ys
c
a
le
a
nd
r
unning
the
a
l
gor
it
hm.
T
he
c
r
it
e
r
ion
of
the
c
or
ne
r
s
hould
be
mor
e
r
e
laxe
d
to
block
thi
s
br
oa
d
tr
ial.
A
pixel's
c
r
it
e
r
ia
mus
t
be
a
c
or
ne
r
ba
s
e
d
on
a
n
a
c
c
e
l
e
r
a
ted
s
e
gment
tes
t
(
AST
)
whic
h
ther
e
mus
t
e
xis
t
a
t
lea
s
t
S
pixels
that
ha
ve
mor
e
br
il
li
a
nt
c
ir
c
le
c
onne
c
ti
on
o
r
da
r
ke
r
than
a
thr
e
s
hold.
Othe
r
va
lues
of
16
pixels
a
r
e
dis
r
e
ga
r
de
d.
S
o
the
va
lue
of
S
c
a
n
be
us
e
d
to
de
ter
mi
ne
the
de
tec
ted
c
or
ne
r
a
t
the
ma
xim
um
a
ngle
[
23
]
.
F
AST
C
or
ne
r
De
tec
ti
on
s
tar
ti
ng
with
de
ter
mi
ning
a
pixel
a
t
c
oor
dinate
s
(
,
)
in
the
im
a
ge
a
nd
c
ompar
e
the
int
e
ns
it
y
p
ixel
with
4
p
ixels
f
r
om
16
pixels
B
r
e
s
e
nha
m
c
ir
c
le
s
ur
r
ounding
the
pixel
,
whos
e
r
a
dius
is
3
[
24]
.
F
ir
s
t
p
ixel
loca
ted
in
the
c
oor
dinate
s
(
,
−
3
)
,
a
s
e
c
ond
pixel
loca
ted
in
the
c
oor
dinate
s
(
+
3
,
)
,
a
thi
r
d
pixel
loca
ted
in
the
c
oor
dinate
s
(
,
+
3
)
,
a
nd
a
f
our
th
pixel
loca
ted
in
the
c
oor
dinate
s
(
−
3
,
)
.
As
il
lus
tr
a
ted
in
F
igu
r
e
4
in
pixel
num
be
r
1
,
5
,
9
,
a
nd
13.
I
f
the
int
e
ns
it
y
pixel
is
be
low
o
r
a
bove
than
the
int
e
ns
it
y
of
a
t
lea
s
t
thr
e
e
pixels
f
r
om
1
,
5,
9
,
a
nd
13
plus
a
T
hr
e
s
hold
int
e
ns
it
y
,
then
it
c
a
n
be
s
a
id
that
pixel
is
a
n
int
e
r
e
s
t
point
[
23]
.
I
t
is
de
f
ined
in
(
1)
:
=
{
1
,
<
−
>
+
0
,
ℎ
(
1)
a
f
ter
that
pixel
will
be
a
t
the
c
oo
r
dinate
s
(
+
1
,
)
a
nd
r
e
pe
a
t
the
p
r
oc
e
dur
e
by
c
ompar
ing
the
int
e
ns
it
y
with
the
other
f
ou
r
pixels
s
ur
r
ou
nding.
T
he
s
a
me
pr
oc
e
dur
e
c
a
n
it
e
r
a
te
f
o
r
th
e
whole
im
a
ge
's
pixels
.
F
igur
e
4
.
F
AS
T
De
tec
tor
[
22]
I
n
the
mac
hine
lea
r
ning
a
ppr
oa
c
h,
F
AST
a
lgo
r
it
hm
c
ompar
ing
s
ixt
e
e
n
pixel
that
s
ur
r
ounding
the
c
a
ndidate
int
e
r
e
s
t
point
(
c
or
ne
r
)
pixel
—
a
s
i
ll
us
tr
a
ted
in
F
igur
e
4
—
the
c
o
r
ne
r
de
tec
tor
de
f
ine
pi
xe
l
a
s
a
c
or
ne
r
i
f
the
r
e
e
xis
ts
a
s
e
t
of
c
onti
guous
pix
e
ls
in
the
c
i
r
c
le,
whic
h
a
r
e
b
r
ight
e
r
than
the
int
e
ns
it
y
o
f
the
c
a
ndidate
c
or
ne
r
plus
a
thr
e
s
hold
,
o
r
a
r
e
da
r
ke
r
than
mi
nus
the
[
25
]
.
T
he
is
c
hos
e
n
a
s
1
2
be
c
a
us
e
it
a
ll
ows
a
high
-
s
pe
e
d
tes
t
that
c
a
n
be
us
e
d
to
e
xc
lude
a
ve
r
y
lar
ge
number
of
non
-
c
or
ne
r
s
.
I
t
d
e
f
ined
in
thi
s
e
qua
ti
on:
→
=
{
,
→
≤
−
(
da
r
ke
r
)
,
−
<
→
<
+
(
s
im
il
a
r
)
,
−
≤
→
(
br
ight
e
r
)
(
2)
whe
r
e
is
the
int
e
ns
it
y
of
p,
→
is
the
int
e
ns
it
y
of
the
s
i
xtee
n
pixels
that
s
ur
r
ounding
the
e
xa
mpl
e
pixel
a
n
d
is
a
thr
e
s
hold.
I
f
→
is
e
qua
l
to
d
,
the
pixel
be
longs
to
the
da
r
ke
r
gr
oup;
if
→
is
e
qua
l
to
s
,
the
pixel
be
longs
to
the
s
im
il
a
r
gr
oup;
if
→
is
e
qua
l
to
b
,
th
e
pixel
be
longs
to
the
b
r
ight
e
r
gr
oup
.
I
f
ther
e
a
r
e
12
c
onti
guous
pixels
that
be
long
to
the
da
r
ke
r
or
br
ig
hter
gr
oup
,
p
is
c
las
s
if
ied
a
s
a
c
or
ne
r
.
3.
RE
S
UL
T
S
A
ND
AN
AL
YSI
S
3.
1.
T
e
s
t
in
g
F
AST
alg
h
or
it
m
T
he
de
ve
lopm
e
nt
of
thi
s
a
ugmente
d
r
e
a
li
ty
a
ppli
c
a
t
ion
us
e
s
a
home
c
a
talog
a
s
a
mar
ke
r
that
will
br
ing
up
the
vir
tual
f
o
r
m
of
a
3D
home.
T
he
r
e
a
r
e
3
ma
r
ke
r
s
us
e
d
in
a
c
c
or
da
nc
e
with
the
number
of
types
of
hous
e
s
be
ing
mar
ke
ted.
T
he
c
a
talog
im
a
ge
will
be
tar
ge
te
d
if
a
ppr
op
r
iate
the
r
e
quir
e
ments
of
Vu
f
or
ia
by
upl
oa
ding
it
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
1
8
,
No
.
1
,
F
e
br
ua
r
y
2020
:
140
-
147
144
to
the
Vu
f
or
ia
we
bs
it
e
.
Vuf
or
ia
is
a
ble
to
r
e
c
ognize
a
nd
tr
a
c
k
tar
ge
ts
by
a
na
lyzing
the
c
ont
r
a
s
t
of
the
f
e
a
tur
e
s
or
de
tec
ted
c
or
ne
r
s
e
e
n
in
the
c
a
mer
a
.
A
f
e
a
tur
e
i
s
a
s
ha
r
p,
s
piked,
de
tailed
de
tail
in
the
im
a
ge
.
T
h
e
wa
y
to
de
ter
mi
ne
t
he
f
e
a
tu
r
e
s
us
e
d
by
Vuf
or
ia
is
the
F
A
S
T
c
or
ne
r
de
tec
ti
on
a
lgo
r
it
hm.
T
he
leve
l
o
f
e
a
s
e
of
im
a
ge
tr
a
c
king
that
is
us
e
d
a
s
a
mar
ke
r
is
g
iven
a
r
a
ti
ng
by
Vuf
or
ia
c
a
ll
e
d
a
n
a
ugmenta
ble
r
a
ti
ng
.
T
h
e
higher
the
r
a
ti
ng,
the
e
a
s
ier
it
will
be
to
tr
a
c
k.
T
he
C
a
tal
og
de
s
ign
that
will
be
us
e
a
s
mar
ke
r
s
s
how
in
F
ig
ur
e
5.
T
o
incr
e
a
s
e
the
r
a
ti
ng
to
it
s
maximum
va
lue,
you
c
a
n
c
r
e
a
te
a
de
s
ign
that
is
r
ich
in
f
e
a
tur
e
s
,
good
c
ont
r
a
s
t,
s
moot
h
dis
tr
ibut
ion
of
f
e
a
tur
e
s
,
a
nd
a
voidi
ng
r
e
pe
ti
ti
ve
p
a
tt
e
r
ns
in
one
im
a
ge
.
F
igur
e
6
s
how
s
the
r
e
s
ult
s
of
tes
ti
ng
the
home
c
a
talog
mar
ke
r
de
tec
ti
on
f
o
r
e
a
c
h
type
of
hous
e
ba
s
e
d
on
the
a
ugmenta
ble
r
a
ti
ng
of
Vuf
or
ia
.
F
igur
e
5
.
C
a
talog
de
s
ign
a
s
mar
ke
r
s
F
igur
e
6.
M
a
r
ke
r
r
a
ti
ng
a
ugmenta
ble
f
r
om
vuf
or
ia
T
he
a
ugmenta
ble
r
a
ti
ng
s
hows
a
high
r
a
ti
ng
o
f
5
s
tar
s
indi
c
a
ti
ng
that
the
e
xis
ti
ng
mar
ke
r
i
mage
s
a
r
e
e
a
s
il
y
tr
a
c
ke
d
by
the
Augme
nted
R
e
a
li
ty
s
y
s
tem.
T
his
is
be
c
a
us
e
c
a
t
a
log
im
a
ge
s
c
ontain
many
f
e
a
tur
e
s
of
mar
ke
r
objec
ts
(
c
ompl
e
x
pa
tt
e
r
ns
a
nd
de
tails
)
with
good
c
ontr
a
s
t
a
nd
s
moot
h
dis
tr
ibut
ion
of
f
e
a
tur
e
s
.
T
he
a
lgor
it
hm
us
e
d
in
Vuf
o
r
ia
is
the
F
AST
c
o
r
ne
r
de
tec
ti
on
a
lgor
it
hm.
T
he
wa
y
it
wor
ks
is
to
de
ter
mi
ne
one
by
one
a
ll
the
pixels
in
the
im
a
ge
a
nd
then
de
ter
m
ine
whe
ther
the
poin
t
is
a
c
or
ne
r
o
r
not
.
An
e
xa
m
ple
of
a
n
e
xpe
r
im
e
nt
c
onduc
ted
is
a
s
il
lus
tr
a
ted
in
F
igur
e
7.
I
n
F
igur
e
7
s
e
lec
t
a
pixel
a
s
with
c
oor
dinate
s
(
,
)
f
or
e
xa
mpl
e
in
the
ma
r
ke
r
im
a
ge
,
a
nd
then
c
ompar
e
the
int
e
ns
it
y
p
ixel
with
4
pixels
f
r
om
1
6
pixels
s
ur
r
ounding
the
pixel
.
F
i
r
s
t
pixel
numbe
r
1
in
F
igur
e
7
loca
ted
in
the
c
oor
dinate
s
(
,
+
3
)
,
s
e
c
ond
pi
xe
l
is
number
5
loca
ted
in
the
c
oor
dinate
s
(
+
3
,
)
,
thi
r
d
pixel
is
nu
mber
9
loca
ted
in
the
c
oor
dina
tes
(
,
−
3
)
,
a
nd
f
ou
r
th
pixel
is
nu
mber
13
loca
ted
in
the
c
oor
dinate
s
(
−
3
,
).
T
he
ne
xt
s
tep
is
c
ompar
ing
pix
e
l
with
pixel
1,
5
,
9,
a
nd
13.
I
f
the
int
e
ns
it
y
pixel
is
be
low
or
a
bove
than
the
int
e
ns
it
y
of
a
t
lea
s
t
thr
e
e
pixels
f
r
om
1,
5
,
9,
a
nd
13
p
lus
a
T
hr
e
s
hold
int
e
n
s
it
y,
then
it
c
a
n
be
s
a
id
that
p
ixel
is
a
n
int
e
r
e
s
t
(
c
o
r
ne
r
)
.
I
n
thi
s
c
a
s
e
,
the
int
e
ns
it
y
o
f
pixel
is
br
ight
e
r
than
1
,
5
a
nd
13
whic
h
mea
n
pixel
is
a
c
or
ne
r
a
c
c
or
ding
to
(
1)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
A
ugme
nted
r
e
ali
ty
us
ing
featur
e
s
ac
c
e
ler
ated
s
e
gm
e
nt
tes
t
for
pr
ope
r
ty
c
atal
ogue
(
R
udy
Se
tyadi
)
145
F
igur
e
7.
F
AS
T
a
lgor
it
hm
on
a
mar
ke
r
B
e
c
a
us
e
of
the
lar
ge
number
of
pixel
point
s
that
mus
t
be
tes
ted,
to
f
a
c
il
it
a
te
tes
ti
ng
o
f
the
F
AS
T
a
lgor
it
hm,
the
ne
xt
tes
t
is
to
look
a
t
the
a
ngles
de
tec
ted
us
ing
ope
n
s
our
c
e
M
AT
L
AB
f
or
F
AST
a
lgo
r
it
hm
by
E
dwa
r
d
R
os
ten.
A
pixel
is
de
f
ined
a
s
a
c
o
r
ne
r
i
f
(
i
n
a
c
ir
c
le
s
ur
r
ounding
the
pixel
)
,
N
o
r
mo
r
e
c
onti
guo
us
pixels
a
r
e
a
ll
s
igni
f
ica
ntl
y
br
ight
e
r
then
o
r
a
ll
s
igni
f
ica
nt
ly
da
r
ke
r
than
the
c
e
nter
pixel.
T
he
or
de
r
ing
of
q
ue
s
ti
ons
us
e
d
to
c
las
s
if
y
a
pixel
is
lea
r
ne
d
us
ing
the
I
D3
a
lgor
it
hm.
T
his
de
tec
tor
ha
s
b
e
e
n
s
hown
to
e
xhib
it
a
high
de
gr
e
e
of
r
e
pe
a
tabili
ty
.
S
im
il
a
r
r
e
s
ult
s
a
r
e
ind
ica
ted
by
a
ll
mar
ke
r
s
.
T
he
gr
e
e
n
dots
s
how
in
F
i
gur
e
8
is
the
Non
-
maximally
s
uppr
e
s
s
e
d
c
or
ne
r
s
.
T
he
in
ter
e
s
t
point
s
we
r
e
ge
ne
r
a
ted
with
a
n
ope
n
s
our
c
e
M
AT
L
AB
c
ode
f
or
the
F
AST
a
lg
or
it
h
m
by
E
dwa
r
d
R
os
ten.
F
igur
e
8.
M
a
r
ke
r
with
int
e
r
e
s
t
point
de
tec
ted
3.
2.
A
u
gm
e
n
t
e
d
r
e
ali
t
y
im
p
lem
e
n
t
at
ion
T
he
a
ugmente
d
r
e
a
li
ty
a
ppli
c
a
ti
on
a
s
a
home
c
a
talog
that
ha
s
be
e
n
made
us
ing
the
a
ppli
c
a
ti
on
o
f
the
F
AST
a
lgor
it
hm
on
ma
r
ke
r
de
tec
ti
on
is
ope
r
a
t
e
d
on
mobi
le
with
the
Andr
oid
ope
r
a
ti
ng
s
ys
tem.
T
he
r
e
a
r
e
5
main
menus
,
na
mely
I
nt
r
oduc
ti
on
to
inf
or
mation
a
bou
t
the
hous
e
,
AR
c
a
mer
a
to
de
tec
t
c
a
talogs
,
Ho
me
T
our
s
f
or
the
e
xpe
r
ienc
e
of
s
e
e
ing
a
nd
go
a
r
ound
the
hous
e
f
r
o
m
the
ou
ts
ide
a
nd
int
o
the
ins
ide,
Guide
f
or
in
f
or
mation
on
how
to
us
e
a
nd
the
late
s
t
I
nf
or
mation
a
bout
th
e
a
ppli
c
a
ti
on
a
s
s
hown
in
F
igur
e
9
.
Augm
e
nted
r
e
a
li
ty
that
will
c
ome
out
whe
n
de
tec
ti
ng
home
c
a
talog
mar
ke
r
s
c
a
n
be
s
e
e
n
in
F
igur
e
10
.
T
he
3D
dis
play
c
a
n
be
e
nlar
ge
d
a
nd
r
otate
d
s
o
that
the
a
ppe
a
r
a
nc
e
of
the
hous
e
c
a
n
be
s
e
e
n
f
r
om
a
ll
s
ides
.
Augme
nted
r
e
a
li
ty
us
e
s
mar
ke
r
de
tec
ti
on
us
ing
th
e
F
AST
a
lgo
r
it
hm
method
in
Vuf
o
r
ia.
T
he
mar
ke
r
im
a
ge
is
r
ich
in
f
e
a
tur
e
s
a
nd
s
pr
e
a
d
e
ve
nly
.
E
xis
ti
ng
home
br
oc
hur
e
s
a
ppr
op
r
iate
thes
e
r
e
qui
r
e
ments
s
o
that
ther
e
is
no
ne
e
d
f
or
a
lot
o
f
modi
f
ica
ti
ons
.
W
he
n
u
pl
oa
de
d
in
Vuf
or
ia,
c
a
talog
de
s
ign
r
e
c
e
ived
a
n
a
ug
menta
ble
r
a
ti
ng
with
a
s
c
or
e
o
f
5.
A
high
s
c
or
e
p
r
ovides
e
a
s
y
Vuf
or
ia
in
de
tec
ti
ng
im
a
ge
s
s
o
that
it
will
make
it
e
a
s
ier
to
dis
play
e
xis
ti
ng
3D
a
nim
a
ti
ons
.
Af
ter
be
ing
downloa
de
d
a
ga
in
f
r
om
Vuf
or
ia
a
nd
made
i
n
Unity3D.
T
he
wa
ys
of
Vuf
or
ia
wor
king
whe
n
the
a
ppli
c
a
ti
on
ope
ns
the
c
a
mer
a
,
the
tas
k
of
the
c
a
mer
a
is
to
c
a
ptur
e
im
a
ge
s
in
r
e
a
l
ti
me
a
nd
then
t
r
a
c
king
a
nd
de
tec
ti
n
g
c
a
ptur
e
d
c
a
mer
a
objec
ts
will
be
c
a
r
r
ied
out
by
tr
a
c
ke
r
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
1
8
,
No
.
1
,
F
e
br
ua
r
y
2020
:
140
-
147
146
Vuf
or
ia
c
ontaining
c
omput
e
r
vi
s
ion
a
lgor
it
h
ms
na
mely
the
F
AST
a
lgor
i
thm
.
T
he
F
AST
a
lgor
it
hm
will
wor
k
to
c
he
c
k
im
a
ge
s
by
wor
king
a
c
c
or
ding
to
(
1)
.
T
he
im
a
ge
that
is
vis
ibl
e
in
the
c
a
mer
a
will
be
de
tec
ted
a
ga
ins
t
e
a
c
h
pixel
a
nd
de
ter
mi
ne
whe
ther
it
include
s
a
c
or
ne
r
or
not.
De
tec
ti
ng
th
e
c
or
ne
r
that
ha
s
be
e
n
done
make
s
a
pa
tt
e
r
n
then
it
will
do
the
matc
hing
with
the
da
tab
a
s
e
im
a
ge
that
is
on
vuf
or
ia.
W
he
n
it
f
inds
a
n
im
a
ge
matc
h,
the
r
e
s
ult
s
of
de
tec
ti
on
will
then
be
us
e
d
to
s
tar
t
r
e
nde
r
ing
a
pr
e
de
ter
mi
ne
d
3D
objec
t
that
is
the
hous
e
f
r
o
m
the
c
a
talog
s
c
a
n.
T
he
n
the
r
e
nde
r
ing
r
e
s
ult
s
will
be
dis
playe
d
via
mobi
le
de
vice
s
in
r
e
a
l
-
ti
me.
T
he
a
dde
d
f
e
a
tur
e
is
that
ther
e
is
a
home
tour
t
o
f
e
e
l
a
r
e
a
l
e
xpe
r
ienc
e
s
uc
h
a
s
wa
lki
ng
a
r
ound
the
hous
e
to
look
ins
ide
the
c
ontents
of
the
hous
e
c
a
n
be
s
e
e
n
in
F
igur
e
11.
W
he
n
c
hoos
ing
the
ho
me
tour
menu,
the
hous
e
will
a
ppe
a
r
a
c
c
or
ding
to
the
type
you
ha
ve
pr
e
vious
ly
c
hos
e
n.
Us
e
r
s
c
a
n
tr
a
ve
l
by
movi
ng
the
butt
on
to
wa
lk
int
o
the
hous
e
a
nd
s
e
e
the
c
on
tents
of
e
a
c
h
e
xis
ti
ng
r
oom
a
c
c
ompanie
d
by
e
xa
mpl
e
s
of
i
nter
ior
de
s
ign.
F
igur
e
11
.
Dis
play
of
the
home
tour
4.
CONC
L
USI
ON
T
he
de
tec
ti
on
of
mar
ke
r
a
ugmente
d
r
e
a
li
ty
wit
h
the
a
ppli
c
a
ti
on
o
f
the
F
AS
T
c
or
ne
r
de
tec
ti
on
a
lgor
it
hm
us
e
d
in
Vu
f
or
ia
S
DK
is
us
e
f
ul
a
s
a
tr
igg
e
r
f
or
a
ugmente
d
r
e
a
li
ty
.
T
he
wor
king
method
of
t
he
F
AST
a
lgor
it
hm
is
to
e
xa
mi
ne
e
a
c
h
pixel
in
a
n
im
a
ge
a
nd
de
ter
m
ine
whe
ther
the
pixel
is
a
n
in
ter
e
s
t
point
(
c
or
ne
r
)
or
not.
T
he
c
ount
us
e
d
f
r
om
the
e
xa
mpl
e
pixel
is
tes
t
e
d
c
ompar
e
d
with
s
ixt
e
e
n
pixels
a
r
ound
it
then
de
t
e
r
mi
ne
s
f
our
point
s
.
F
r
o
m
thes
e
f
ou
r
point
s
,
if
ther
e
a
r
e
a
t
l
e
a
s
t
thr
e
e
da
r
ke
r
or
br
igh
ter
one
s
,
the
e
xa
mpl
e
pix
e
l
that
is
in
the
mi
ddle
is
a
n
a
ngle.
I
n
Vuf
or
ia
pixels
,
whic
h
is
a
n
int
e
r
e
s
t
point
,
it
is
c
a
ll
e
d
f
e
a
tur
e
s
.
T
he
us
e
o
f
im
a
ge
s
a
s
mar
ke
r
s
mus
t
ha
ve
many
f
e
a
tur
e
s
,
c
ontr
a
s
t
im
a
ge
s
,
a
nd
s
moot
h
dis
tr
ibut
ion
o
f
f
e
a
tur
e
s
s
o
that
mar
ke
r
s
ha
ving
a
high
a
ugmenta
ble
r
a
ti
ng
will
make
it
e
a
s
ier
f
or
Vu
f
or
ia
to
de
tec
t
im
a
ge
s
.
T
he
F
AST
a
l
gor
it
hm
pe
r
f
or
ms
matc
hing
f
e
a
tur
e
s
that
e
xis
t
in
the
tar
ge
t
im
a
ge
with
the
da
taba
s
e
.
T
he
n
the
a
ppli
c
a
ti
on
wi
ll
r
e
nde
r
3D
objec
ts
.
I
f
the
de
tec
ti
on
of
a
s
uit
a
ble
tar
ge
t
im
a
ge
w
il
l
r
e
nde
r
the
3D
objec
t
of
the
hous
e
.
T
he
r
e
s
ult
s
of
vis
ua
li
z
ing
the
hous
e
in
3D
a
r
e
e
a
s
ier
to
e
xpl
a
in
the
or
igi
na
l
a
ppe
a
r
a
nc
e
of
the
hous
e
c
om
pa
r
e
d
to
the
a
ppe
a
r
a
nc
e
of
the
photo.
T
his
a
ppli
c
a
ti
on
f
a
c
il
it
a
tes
s
a
les
mar
ke
ti
ng
a
s
a
n
e
xplana
ti
on
to
pr
os
pe
c
ti
ve
c
us
tom
e
r
s
who
a
r
e
c
onf
us
e
d
a
bout
the
or
igi
na
l
de
s
c
r
ipt
ion
of
the
hous
e
both
f
r
om
outs
ide
the
home
a
nd
ins
ide
F
igur
e
9.
Dis
play
of
the
main
menu
F
igur
e
10.
Dis
play
of
a
ugmente
d
r
e
a
li
ty
a
f
ter
mar
ke
r
de
tec
ted
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
A
ugme
nted
r
e
ali
ty
us
ing
featur
e
s
ac
c
e
ler
ated
s
e
gm
e
nt
tes
t
for
pr
ope
r
ty
c
atal
ogue
(
R
udy
Se
tyadi
)
147
the
hous
e
a
nd
by
us
ing
the
ho
me
tou
r
menu
pr
os
pe
c
ti
ve
c
us
tom
e
r
s
c
a
n
f
e
e
l
the
e
xpe
r
ienc
e
of
goin
g
a
r
ound
the
hous
e
in
a
nim
a
ti
on.
RE
F
E
RE
NC
E
S
[1
]
N
.
W
i
c
k
ramaarac
h
ch
i
,
“D
et
erm
i
n
a
n
t
s
o
f
Ren
t
al
V
al
u
e
f
o
r
Res
i
d
e
n
t
i
al
Pr
o
p
er
t
i
e
s
:
A
L
an
d
O
w
n
er’s
Per
s
p
ec
t
i
v
e
fo
r
Bo
ard
i
n
g
H
o
me
s
,
”
B
u
i
l
t
-
E
n
vi
r
o
n
m
e
n
t
S
r
i
La
n
k
a
,
v
o
l
.
1
2
,
n
o
.
1
,
p
p
.
1
0
–
2
2
,
2
0
1
6
.
[2
]
L
.
Mau
l
i
an
i
a
n
d
W
.
Su
d
arw
a
t
i
,
“Pe
n
g
ar
u
h
M
o
d
u
l
Be
s
ar
an
Ru
a
n
g
T
erh
a
d
ap
T
a
t
a
Ru
a
n
g
R
u
mah
Sa
n
g
a
t
Se
d
erh
a
n
a,
”
NA
LA
R
s
,
v
o
l
.
1
7
,
n
o
.
2
,
p
p
.
1
3
5
–
1
4
4
,
2
0
1
8
.
[3
]
T
.
H
.
K
u
s
u
m
a,
“A
n
al
y
s
i
s
o
f
Fact
o
rs
A
ffec
t
i
n
g
H
o
u
s
e
O
w
n
ers
h
i
p
i
n
In
d
o
n
es
i
a,
”
A
m
.
R
es
.
J.
B
u
s
.
M
a
n
a
g
.
,
v
o
l
.
4
,
n
o
.
1
,
p
p
.
1
–
1
2
,
2
0
18.
[4
]
F.
T
.
Mao
l
u
d
y
o
a
n
d
A
.
A
p
r
i
an
i
n
g
s
i
h
,
“Fac
t
o
r
s
In
f
l
u
e
n
ci
n
g
C
o
n
s
u
mer
B
u
y
i
n
g
In
t
en
t
i
o
n
f
o
r
H
o
u
s
i
n
g
U
n
i
t
i
n
D
ep
o
k
,
”
J.
B
u
s
.
M
a
n
a
g
.
,
v
o
l
.
4
,
n
o
.
4
,
p
p
.
4
8
4
–
4
9
3
,
2
0
1
5
.
[5
]
Y
.
S.
Sari
,
“A
n
al
y
s
i
s
an
d
D
es
i
g
n
o
f
In
f
o
rmat
i
o
n
Sy
s
t
em
Cas
h
Pu
rc
h
as
e
o
f
J
el
i
t
a
S
p
rey
w
i
t
h
O
b
j
ec
t
O
ri
e
n
t
ed
Met
h
o
d
o
l
o
g
y
,
”
v
o
l
.
7
,
n
o
.
7
,
p
p
.
4
5
–
5
7
,
2
0
1
8
.
[6
]
A
s
w
i
n
a
n
d
S
y
ah
ar
u
d
d
i
n
,
“Pen
g
ar
u
h
B
i
ay
a
Pro
m
o
s
i
T
er
h
ad
ap
Pen
i
n
g
k
a
t
an
Pe
n
j
u
al
a
n
Ru
ma
h
Pad
a
Peru
ma
h
an
G
r
an
d
A
ro
e
p
al
a
D
i
Mak
a
s
s
ar,
”
J.
Iq
t
i
s
a
d
u
n
a
,
v
o
l
.
1
,
n
o
.
2
,
p
p
.
1
–
1
7
,
2
0
1
5
.
[7
]
M.
O
zd
emi
r,
C.
Sah
i
n
,
S.
A
rcag
o
k
,
an
d
M.
K
.
D
emi
r,
“T
h
e
E
ffect
o
f
A
u
g
men
t
ed
Real
i
t
y
A
p
p
l
i
ca
t
i
o
n
s
i
n
t
h
e
L
earn
i
n
g
Pro
ces
s
:
A
Me
t
aA
n
al
y
s
i
s
St
u
d
y
,
”
E
u
r
a
s
i
a
n
J.
E
d
u
c.
R
e
s
.
,
v
o
l
.
1
8
,
n
o
.
A
p
r
i
l
,
p
p
.
1
6
5
–
1
8
6
,
2
0
1
8
.
[8
]
B.
A
ri
fi
t
ama,
P
a
n
d
u
a
n
M
u
d
a
h
M
e
m
b
u
a
t
A
u
g
m
e
n
t
e
d
R
e
a
l
i
t
y
,
1
s
t
ed
.
Y
o
g
y
ak
ar
t
a:
A
n
d
i
,
2
0
1
7
.
[9
]
V
.
D
.
D
eo
l
e
k
ar
an
d
P.
M.
D
es
h
mu
k
h
,
“Cas
e
St
u
d
y
o
f
A
u
g
men
t
ed
Real
i
t
y
A
p
p
l
i
cat
i
o
n
s
i
n
Med
i
cal
Fi
e
l
d
,
”
In
t
.
J.
Tr
en
d
S
ci
.
R
e
s
.
D
ev.
,
v
o
l
.
2
,
n
o
.
4
,
p
p
.
2
6
9
1
–
2
6
9
4
,
2
0
1
8
.
[1
0
]
A
.
H
.
Safar,
A
.
A
.
A
l
-
J
afar,
an
d
Z
.
H
.
A
l
-
Y
o
u
s
ef
i
,
“T
h
e
E
ffect
i
v
e
n
es
s
o
f
U
s
i
n
g
A
u
g
men
t
ed
Real
i
t
y
A
p
p
s
i
n
T
eac
h
i
n
g
t
h
e
E
n
g
l
i
s
h
A
l
p
h
a
b
et
t
o
K
i
n
d
er
g
art
e
n
Ch
i
l
d
ren
:
A
Cas
e
St
u
d
y
i
n
t
h
e
St
a
t
e
o
f
K
u
w
a
i
t
,
”
E
u
r
a
s
i
a
J.
M
a
t
h
.
S
c
i
.
Tech
n
o
l
.
E
d
u
c.
,
v
o
l
.
1
3
,
n
o
.
2
,
p
p
.
4
1
7
–
4
4
0
,
2
0
1
7
.
[1
1
]
X
.
Y
o
u
,
W
.
Z
h
an
g
,
M.
Ma,
C.
D
en
g
,
an
d
J
.
Y
a
n
g
,
“Su
rv
ey
o
n
U
rb
a
n
W
arfare
A
u
g
me
n
t
e
d
Real
i
t
y
,
”
IS
P
R
S
In
t
.
J.
G
eo
-
I
n
f
o
r
m
a
t
i
o
n
,
v
o
l
.
7
,
n
o
.
2
,
p
p
.
1
–
1
6
,
2
0
1
8
.
[1
2
]
Y
.
A
.
Sek
h
a
v
at
,
“K
i
o
s
k
A
R:
A
n
A
u
g
me
n
t
e
d
Real
i
t
y
G
ame
as
a
N
ew
Bu
s
i
n
es
s
Mo
d
el
t
o
Pres
e
n
t
A
rt
w
o
r
k
s
,
”
In
t
.
J.
Co
m
p
u
t
.
G
a
m
es
Tech
n
o
l
.
,
v
o
l
.
2
0
1
6
,
p
p
.
1
–
1
2
,
2
0
1
6
.
[1
3
]
D
.
Sah
i
n
an
d
A
.
T
o
g
ay
,
“A
u
g
me
n
t
e
d
Real
i
t
y
A
p
p
l
i
cat
i
o
n
s
i
n
Pro
d
u
ct
D
e
s
i
g
n
Pro
ces
s
,
”
New
Tr
en
d
s
Is
s
u
e
s
P
r
o
c.
H
u
m
a
n
i
t
.
S
o
c.
S
c
i
.
,
v
o
l
.
2
,
n
o
.
1
,
p
p
.
1
1
5
–
1
2
5
,
2
0
1
7
.
[1
4
]
E
.
Set
i
a
w
an
,
U
.
Sy
ar
i
p
u
d
i
n
,
a
n
d
Y
.
A
.
G
erh
a
n
a,
“Imp
l
e
men
t
a
s
i
T
ek
n
o
l
o
g
i
A
u
g
men
t
ed
Real
i
t
y
p
a
d
a
B
u
k
u
Pan
d
u
an
W
u
d
h
u
Berb
as
i
s
M
o
b
i
l
e
A
n
d
r
o
i
d
,
”
J.
O
n
l
i
n
e
I
n
f
o
r
m
.
,
v
o
l
.
1
,
n
o
.
1
,
p
p
.
2
8
–
3
3
,
2
0
1
8
.
[1
5
]
D
.
A
mi
n
a
n
d
S.
G
o
v
i
l
k
ar,
“Co
m
p
arat
i
v
e
St
u
d
y
o
f
A
u
g
men
t
e
d
Real
i
t
y
SD
K
’s
,
”
In
t
.
J.
Co
m
p
u
t
.
S
c
i
.
A
p
p
l
.
,
v
o
l
.
5
,
n
o
.
1
,
p
p
.
1
0
–
2
6
,
2
0
1
5
.
[1
6
]
H
.
V
i
t
o
n
o
,
H
.
N
as
u
t
i
o
n
,
an
d
H
.
A
n
ra,
“Im
p
l
eme
n
t
a
s
i
Mark
erl
e
s
s
A
u
g
men
t
ed
Rea
l
i
t
y
Se
b
ag
a
i
Me
d
i
a
I
n
fo
r
mas
i
K
o
l
ek
s
i
Mu
s
eu
m
Ber
b
as
i
s
A
n
d
r
o
i
d
,
”
J.
S
i
s
t
.
d
a
n
Tekn
o
l
.
In
f
.
,
v
o
l
.
2
,
n
o
.
4
,
p
p
.
2
3
9
–
2
4
5
,
2
0
1
6
.
[1
7
]
F.
Z
u
l
i
,
“Ran
can
g
Ban
g
u
n
A
u
g
men
t
ed
D
an
V
i
rt
u
al
Rea
l
i
t
y
Me
n
g
g
u
n
ak
a
n
A
l
g
o
ri
t
ma
Fas
t
Se
b
ag
a
i
Med
i
a
In
fo
rm
as
i
3
D
D
i
U
n
i
v
ers
i
t
a
s
Sat
y
a
N
eg
ara
In
d
o
n
es
i
a,
”
J.
A
l
g
o
r
i
t
m
.
Lo
g
.
d
a
n
Ko
m
p
u
t
a
s
i
,
v
o
l
.
1
,
n
o
.
2
,
p
p
.
9
4
–
1
0
4
,
2
0
1
8
.
[1
8
]
N
.
O
.
A
d
i
w
i
j
a
y
a,
Y
.
W
ah
y
u
,
an
d
C.
P.
A
n
t
o
n
i
u
s
,
“Imp
l
e
men
t
a
s
i
A
u
g
me
n
t
e
d
Real
i
t
y
U
n
t
u
k
Bro
s
u
r
In
t
era
k
t
i
f
D
en
g
an
Met
o
d
e
FA
S
T
Co
r
n
er
D
e
t
ect
i
o
n
(i
n
b
a
h
as
a
:
A
u
g
me
n
t
e
d
Real
i
t
y
Im
p
l
eme
n
t
a
t
i
o
n
f
o
r
In
t
eract
i
v
e
Br
o
ch
u
res
w
i
t
h
t
h
e
FA
ST
Co
r
n
er
D
e
t
ect
i
o
n
Met
h
o
d
),
”
i
n
S
e
m
n
a
s
t
i
k
o
m
,
p
p
.
5
9
8
–
6
0
6
,
2
0
1
6
.
[1
9
]
E
.
R.
N
ai
n
g
g
o
l
a
n
et
a
l
.
,
“T
h
e
Imp
l
emen
t
at
i
o
n
o
f
A
u
g
me
n
t
e
d
Real
i
t
y
a
s
L
earn
i
n
g
Me
d
i
a
i
n
I
n
t
r
o
d
u
ci
n
g
A
n
i
ma
l
s
fo
r
E
arl
y
Ch
i
l
d
h
o
o
d
E
d
u
ca
t
i
o
n
,
”
T
h
e
6
th
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
Cyb
er
a
n
d
IT
S
er
v
i
ce
M
a
n
a
g
em
en
t
(CITS
M
2
0
1
8
)
,
2
0
1
8
.
[2
0
]
S.
N
u
raj
i
zah
,
“Imp
l
eme
n
t
a
s
i
Mu
l
t
i
med
i
a
D
e
v
el
o
p
me
n
t
L
i
fe
Cy
cl
e
Pad
a
A
p
l
i
k
as
i
Pen
g
en
a
l
an
L
ag
u
A
n
ak
-
A
n
a
k
(i
n
B
ah
as
a
:
Imp
l
eme
n
t
a
t
i
o
n
o
f
Mu
l
t
i
me
d
i
a
D
e
v
el
o
p
me
n
t
L
i
fe
Cy
cl
e
i
n
Ch
i
l
d
re
n
's
So
n
g
Reco
g
n
i
t
i
o
n
A
p
p
l
i
cat
i
o
n
)
,
”
J.
P
R
O
S
IS
K
O
,
v
o
l
.
3
,
n
o
.
2
,
p
p
.
1
4
–
1
9
,
2
0
1
6
.
[2
1
]
M.
Mu
s
t
i
k
a,
E
.
P.
A
.
Su
g
ara,
an
d
M.
Prat
i
w
i
,
“Pen
g
emb
an
g
a
n
Med
i
a
Pemb
e
l
aj
a
ran
In
t
erak
t
i
f
d
e
n
g
a
n
Men
g
g
u
n
a
k
an
Met
o
d
e
Mu
l
t
i
me
d
i
a
D
ev
e
l
o
p
men
t
L
i
fe
Cy
cl
e
(
i
n
b
a
h
a
s
a:
D
e
v
el
o
p
me
n
t
o
f
In
t
er
act
i
v
e
L
earn
i
n
g
Me
d
i
a
U
s
i
n
g
t
h
e
Mu
l
t
i
me
d
i
a
D
e
v
el
o
p
me
n
t
L
i
fe
Cy
c
l
e
Met
h
o
d
)
,
”
J.
O
n
l
i
n
e
In
f
o
r
m
.
,
v
o
l
.
2
,
n
o
.
2
,
p
.
1
2
1
,
2
0
1
8
.
[2
2
]
J
.
H
u
a
n
g
,
G
.
Z
h
o
u
,
X
.
Z
h
o
u
,
an
d
R.
Z
h
a
n
g
,
“
A
N
e
w
FPG
A
A
rch
i
t
ec
t
u
re
o
f
FA
S
T
an
d
BRIE
F
A
l
g
o
r
i
t
h
m
fo
r
On
-
b
o
ar
d
Co
rn
er
D
et
ec
t
i
o
n
an
d
Mat
c
h
i
n
g
,
”
S
e
n
s
o
r
s
,
v
o
l
.
1
8
,
n
o
.
4
,
p
p
.
1
–
1
7
,
2
0
1
8
.
[2
3
]
A
.
A
.
K
ari
m
a
n
d
E
.
F.
N
a
s
s
er,
“Imp
r
o
v
eme
n
t
o
f
Co
r
n
er
D
et
ect
i
o
n
A
l
g
o
ri
t
h
m
s
(H
arr
i
s
,
FA
S
T
an
d
S
U
SA
N
)
Ba
s
ed
o
n
Re
d
u
c
t
i
o
n
o
f
Fea
t
u
re
s
Sp
ace
an
d
Co
m
p
l
e
x
i
t
y
T
i
me,
”
E
n
g
i
n
ee
r
i
n
g
Tec
h
n
o
l
ogy
J
o
u
r
n
a
l
,
v
o
l
.
3
5
,
n
o
.
2
,
p
p
.
1
1
2
–
1
1
8
,
2
0
1
7
.
[2
4
]
L
.
Y
u
,
Z
.
Y
u
,
an
d
Y
.
G
o
n
g
,
“A
n
Imp
ro
v
ed
O
RB
A
l
g
o
ri
t
h
m
o
f
E
x
t
ract
i
n
g
an
d
Mat
c
h
i
n
g
Feat
u
res
,
”
In
t
er
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
S
i
g
n
a
l
P
r
o
ces
s
i
n
g
,
Im
a
g
e
P
r
o
ces
s
i
n
g
a
n
d
P
a
t
t
er
n
R
eco
g
n
i
t
i
o
n
,
v
o
l
.
8
,
n
o
.
5
,
p
p
.
1
1
7
–
1
2
6
,
2
0
1
5
.
[2
5
]
E
.
Mu
eg
g
l
er
an
d
D
.
Scaramu
zza,
“Fas
t
E
v
e
n
t
-
b
as
e
d
Co
rn
er
D
et
ect
i
o
n
,
”
B
r
.
M
a
c
h
.
V
i
s
.
Co
n
f
.
,
v
o
l
.
1
,
p
p
.
1
–
1
1
,
2
0
1
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.