T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
18
,
No
.
2
,
A
p
r
il 2
0
2
0
,
p
p
.
1
1
0
5
~
1
1
1
2
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
8
i2
.
1
4
8
4
2
1105
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
M
a
tching
a
lg
o
rithm per
forma
nce
a
na
ly
sis
for a
utoca
libra
tio
n method o
f
ste
re
o
vis
io
n
Ra
den Ar
ief
Set
y
a
wa
n
1
,
Rud
y
S
oe
no
k
o
2
,
M
o
ch
Ag
us
Cho
iro
n
3
,
P
a
nca
M
ud
j
ira
ha
rdj
o
4
1,
4
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
De
p
a
rtm
e
n
t
,
Bra
wijay
a
Un
i
v
e
rsity
,
In
d
o
n
e
sia
2
,
3
M
e
c
h
a
n
ica
l
En
g
i
n
e
e
rin
g
De
p
a
r
tme
n
t
,
Bra
wijay
a
Un
i
v
e
rsity
,
In
d
o
n
e
sia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
1
5
,
2
0
1
9
R
ev
is
ed
Dec
1
7
,
20
19
Acc
ep
ted
Feb
2
1
,
2
0
2
0
S
tere
o
v
isio
n
is
o
n
e
o
f
th
e
in
tere
stin
g
re
se
a
rc
h
to
p
ics
in
th
e
c
o
m
p
u
ter
v
isio
n
field
.
Two
c
a
m
e
ra
s
a
r
e
u
se
d
to
g
e
n
e
ra
te
a
d
isp
a
rit
y
m
a
p
,
re
su
lt
in
g
in
th
e
d
e
p
t
h
e
stim
a
ti
o
n
.
Ca
m
e
ra
c
a
li
b
ra
ti
o
n
is
t
h
e
m
o
st
imp
o
rtan
t
ste
p
in
ste
re
o
v
isio
n
.
T
h
e
c
a
li
b
ra
ti
o
n
ste
p
is
u
se
d
to
g
e
n
e
ra
te
a
n
in
tr
in
sic
p
a
ra
m
e
ter
o
f
e
a
c
h
c
a
m
e
r
a
to
g
e
t
a
b
e
tt
e
r
d
isp
a
rit
y
m
a
p
.
I
n
g
e
n
e
ra
l,
th
e
c
a
li
b
ra
ti
o
n
p
ro
c
e
ss
is
d
o
n
e
m
a
n
u
a
ll
y
b
y
u
sin
g
a
c
h
e
ss
b
o
a
rd
p
a
tt
e
rn
,
b
u
t
t
h
is
p
ro
c
e
ss
is
a
n
e
x
h
a
u
stin
g
tas
k
.
S
e
lf
-
c
a
li
b
r
a
ti
o
n
is
a
n
imp
o
rtan
t
a
b
il
it
y
r
e
q
u
ired
to
o
v
e
rc
o
m
e
th
is
p
r
o
b
lem
.
S
e
lf
-
c
a
li
b
ra
ti
o
n
re
q
u
ired
a
ro
b
u
st
a
n
d
g
o
o
d
m
a
tch
in
g
a
lg
o
rit
h
m
to
fin
d
th
e
k
e
y
fe
a
tu
re
b
e
twe
e
n
ima
g
e
s
a
s
r
e
fe
re
n
c
e
.
T
h
e
p
u
r
p
o
se
o
f
th
is
p
a
p
e
r
is
t
o
a
n
a
ly
z
e
t
h
e
p
e
rfo
rm
a
n
c
e
o
f
th
re
e
m
a
tch
in
g
a
lg
o
rit
h
m
s
fo
r
th
e
a
u
t
o
c
a
li
b
ra
ti
o
n
p
ro
c
e
ss
.
Th
e
m
a
tch
in
g
a
lg
o
r
it
h
m
s
u
se
d
i
n
th
is
re
se
a
rc
h
a
re
S
IF
T,
S
UR
F
,
a
n
d
ORB.
Th
e
re
su
lt
sh
o
ws
th
a
t
S
IF
T
p
e
rfo
rm
s
b
e
tt
e
r
th
a
n
o
th
e
r
m
e
th
o
d
s.
K
ey
w
o
r
d
s
:
Au
to
ca
lib
r
atio
n
I
m
ag
e
m
atch
in
g
Ster
eo
v
is
io
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
R
ad
en
Ar
ief
Sety
awa
n
,
Dep
ar
tm
en
t
E
lectr
ical
E
n
g
i
n
e
er
in
g
,
B
r
awijay
a
Un
iv
er
s
ity
,
167
MT
Har
y
o
n
o
St.,
Ma
lan
g
6
5
1
4
5
I
n
d
o
n
esia
.
E
m
ail:
r
ar
ief
@
u
b
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
Vis
io
n
-
b
ased
m
ea
s
u
r
em
en
t
h
a
s
b
ee
n
o
n
e
o
f
th
e
m
o
s
t
in
ter
es
tin
g
r
esear
ch
to
p
ics
in
th
e
last
d
ec
ad
es.
Ma
n
y
a
p
p
licatio
n
s
h
av
e
b
ee
n
d
ev
elo
p
e
d
u
s
in
g
v
is
io
n
-
b
ased
m
ea
s
u
r
em
en
t
[
1
]
.
T
h
e
two
m
a
jo
r
m
eth
o
d
s
o
f
3
D
m
ea
s
u
r
em
en
t
ca
n
b
e
ca
teg
o
r
iz
ed
in
to
ac
tiv
e
a
n
d
p
ass
iv
e
m
e
th
o
d
s
.
Stru
ctu
r
e
d
illu
m
in
atio
n
o
r
laser
is
u
s
ed
in
th
e
ac
tiv
e
m
ea
s
u
r
em
e
n
t.
T
h
is
m
eth
o
d
is
n
o
t
ap
p
licab
le
i
n
m
an
y
ca
s
es.
T
h
e
p
ass
iv
e
3
D
m
e
asu
r
em
en
t
is
b
ased
o
n
s
ter
eo
v
is
io
n
a
n
d
p
r
o
v
id
es
m
o
r
e
a
d
v
an
tag
es
t
h
an
ac
tiv
e
m
ea
s
u
r
em
en
t.
I
t
r
eq
u
ir
es
s
im
p
ler
in
s
tr
u
m
en
tatio
n
,
o
f
f
er
in
g
h
ig
h
er
a
p
p
licab
ilit
y
in
m
an
y
en
v
i
r
o
n
m
e
n
ts
.
Ho
w
ev
er
,
th
e
m
ajo
r
is
s
u
e
f
o
r
p
a
s
s
iv
e
m
ea
s
u
r
em
en
t
is
th
e
d
if
f
icu
lty
in
f
in
d
in
g
ac
c
u
r
ate
co
r
r
esp
o
n
d
e
n
ce
b
etwe
en
s
ter
eo
im
ag
es [
2
]
.
Ster
eo
ca
lib
r
atio
n
is
th
e
m
o
s
t
im
p
o
r
tan
t
s
tep
to
f
in
d
a
c
o
r
r
esp
o
n
d
en
ce
p
o
in
t.
C
am
er
a
ca
l
ib
r
atio
n
is
r
eq
u
ir
ed
to
en
s
u
r
e
t
h
at
b
o
t
h
c
am
er
as
ar
e
in
p
e
r
f
ec
t
p
o
s
itio
n
an
d
to
r
em
o
v
e
d
is
to
r
tio
n
.
T
r
ad
itio
n
ally
,
ca
m
er
a
ca
lib
r
atio
n
is
p
er
f
o
r
m
ed
u
s
in
g
th
e
s
tan
d
ar
d
ch
ess
-
b
o
a
r
d
p
ictu
r
e
[
3
]
.
H
o
wev
er
,
m
u
c
h
wo
r
k
is
r
eq
u
ir
e
d
in
th
e
s
e
lf
-
ca
lib
r
atio
n
m
eth
o
d
s
.
Ster
eo
s
elf
-
ca
lib
r
atio
n
r
ef
er
s
to
th
e
a
u
to
m
atic
d
eter
m
in
atio
n
o
f
s
ter
eo
ca
m
er
a
p
ar
am
eter
s
f
r
o
m
im
ag
e
s
eq
u
e
n
ce
s
.
Self
-
ca
lib
r
atio
n
is
a
n
im
p
o
r
tan
t
ab
ilit
y
r
e
q
u
ir
e
d
f
o
r
t
h
e
in
tr
o
d
u
ctio
n
o
f
s
ter
eo
ca
m
er
as
in
to
th
e
m
ar
k
et.
Ma
n
y
wo
r
k
s
h
a
v
e
b
ee
n
p
u
b
lis
h
ed
with
th
i
s
m
eth
o
d
[
4
-
1
1
]
.
I
t
ca
n
g
u
ar
a
n
tee
m
ain
ten
an
ce
-
f
r
ee
an
d
th
e
lo
n
g
-
ter
m
o
p
er
atio
n
,
as
th
e
en
v
ir
o
n
m
en
tal
co
n
d
itio
n
s
m
ay
ch
an
g
e
th
e
ca
m
er
a
p
o
s
itio
n
.
Sp
ec
ial
ex
p
er
tis
e
is
r
eq
u
ir
e
d
to
d
o
th
e
o
f
f
lin
e
ca
lib
r
atio
n
.
Self
-
ca
lib
r
atio
n
m
ay
r
e
d
u
ce
r
eg
u
lar
o
f
f
lin
e
ca
lib
r
atio
n
tim
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
18
,
No
.
2
,
Ap
r
il 2
0
2
0
:
1
1
0
5
-
1
1
1
2
1106
an
d
c
o
s
t.
E
v
en
if
h
u
m
a
n
ey
es
h
av
e
d
if
f
er
en
t
ch
ar
ac
te
r
is
tics
with
m
in
u
s
/p
lu
s
/cy
lin
d
r
ical
p
r
o
p
er
ties
,
th
e
h
u
m
a
n
b
r
ain
ca
n
a
u
to
m
atica
lly
a
d
ju
s
t.
C
o
n
s
eq
u
en
tly
,
th
e
h
u
m
a
n
b
ein
g
will
h
a
v
e
n
o
d
if
f
icu
lt
ies
in
m
er
g
in
g
two
v
is
io
n
s
f
r
o
m
th
e
lef
t
an
d
r
ig
h
t
ca
m
er
as.
I
n
d
esig
n
in
g
a
s
elf
-
ca
lib
r
atio
n
m
eth
o
d
,
a
m
at
ch
in
g
al
g
o
r
ith
m
is
an
im
p
o
r
ta
n
t to
o
l to
f
in
d
a
co
r
r
esp
o
n
d
en
c
e
p
o
i
n
t b
etwe
en
im
ag
es o
f
two
ca
m
er
as.
T
h
e
m
ain
o
b
jectiv
e
o
f
th
is
p
ap
er
is
to
an
aly
ze
th
e
p
er
f
o
r
m
an
ce
o
f
th
r
e
e
m
atch
i
n
g
al
g
o
r
ith
m
s
f
o
r
th
e
au
to
ca
lib
r
atio
n
p
r
o
ce
s
s
.
T
wo
of
th
e
m
o
s
t
co
m
m
o
n
tech
n
iq
u
es
f
o
r
s
ter
eo
co
r
r
esp
o
n
d
e
n
ce
ar
e
th
e
s
u
m
o
f
ab
s
o
lu
te
d
if
f
er
e
n
ce
s
(
SAD)
an
d
th
e
s
u
m
o
f
s
q
u
ar
e
d
d
if
f
e
r
en
ce
s
(
SS
D)
.
T
h
e
co
r
r
esp
o
n
d
in
g
p
o
in
ts
b
etwe
en
im
ag
es
h
av
e
b
ee
n
o
b
t
ain
e
d
b
y
m
in
im
izin
g
SAD
o
r
SS
D
in
ar
ea
-
b
ased
b
lo
ck
m
atch
in
g
[
1
2
]
.
Ho
wev
er
,
th
ese
two
tech
n
iq
u
es
r
esu
lt
in
lo
w
ac
cu
r
ac
y
as
t
h
eir
m
aj
o
r
d
r
aw
b
ac
k
.
An
im
p
r
o
v
em
e
n
t
b
y
u
s
in
g
s
u
b
-
p
ix
el
b
lo
ck
m
atch
in
g
tech
n
iq
u
es
h
as
b
ee
n
ex
p
lo
r
ed
in
[
4
]
,
b
u
t
th
e
o
b
tain
ed
ac
cu
r
ac
y
was
s
till
n
o
t
en
o
u
g
h
.
R
ec
en
tly
,
th
er
e
h
av
e
b
ee
n
m
an
y
alg
o
r
ith
m
s
p
r
o
p
o
s
ed
o
n
im
ag
e
m
atch
in
g
u
s
in
g
v
ar
io
u
s
tech
n
iq
u
es
[
1
3
]
.
I
n
th
is
wo
r
k
,
a
s
et
o
f
ex
p
er
im
en
ts
d
em
o
n
s
tr
ate
s
th
a
t
th
e
s
ter
eo
v
is
io
n
s
y
s
tem
em
p
lo
y
in
g
th
e
p
r
o
p
o
s
ed
tec
h
n
iq
u
e
ca
n
m
ea
s
u
r
e
3
D
s
u
r
f
ac
es
o
f
f
r
ee
-
f
o
r
m
o
b
jects
with
s
u
b
-
m
m
ac
cu
r
ac
y
.
T
h
r
ee
m
atch
in
g
tech
n
i
q
u
es
u
s
ed
in
th
is
r
esear
ch
ar
e
SIFT
,
SUR
F,
an
d
OR
B
.
T
h
e
m
atch
in
g
al
g
o
r
ith
m
p
r
o
v
id
es
t
h
e
ch
ar
ac
te
r
is
tics
o
f
ea
ch
ca
m
er
a
[
1
4
]
.
I
t
u
s
ed
to
tr
an
s
f
o
r
m
t
h
e
s
ec
o
n
d
im
ag
e
to
p
er
f
o
r
m
au
to
m
atic
s
ter
eo
c
alib
r
atio
n
.
T
h
e
ex
p
lan
atio
n
o
f
ea
ch
alg
o
r
ith
m
is
ex
p
lain
ed
as f
o
llo
ws.
-
SIFT
Scale
in
v
ar
ian
t
f
ea
tu
r
e
tr
an
s
f
o
r
m
(
SIFT
)
is
a
m
atch
in
g
alg
o
r
ith
m
p
r
o
p
o
s
ed
b
y
L
o
we
[
1
5
]
.
T
h
is
alg
o
r
ith
m
wo
r
k
s
v
er
y
well
in
f
in
d
in
g
a
co
r
r
esp
o
n
d
en
ce
p
o
i
n
t
o
f
th
e
im
a
g
e
wh
ich
is
r
o
tated
an
d
tr
an
s
f
o
r
m
ed
.
T
h
is
al
g
o
r
ith
m
co
n
s
is
ts
o
f
f
o
u
r
s
tep
s
.
T
h
e
f
i
r
s
t
s
tep
is
th
e
esti
m
atio
n
o
f
s
ca
le
-
s
p
ac
e
ex
tr
em
a
u
s
in
g
th
e
Dif
f
er
e
n
ce
o
f
Gau
s
s
ian
m
eth
o
d
,
b
ein
g
ex
p
r
ess
u
s
in
g
(
1
)
a
n
d
d
escr
ib
ed
in
Fig
u
r
e
1
.
(
1
)
Fig
u
r
e
1
.
T
h
e
esti
m
atio
n
o
f
s
ca
le
-
s
p
ac
e
ex
tr
em
a
u
s
in
g
D
if
f
e
r
en
ce
o
f
Gau
s
s
ian
m
eth
o
d
I
n
th
e
n
ex
t
s
tep
,
th
e
k
ey
p
o
in
t
ca
n
d
id
ates
ar
e
r
e
f
in
ed
b
y
th
e
elim
in
atio
n
o
f
lo
w
v
alu
e
.
L
ap
lacia
n
o
f
Gau
s
s
ian
σ
2
∇
2
G
is
u
s
ed
s
in
ce
it
p
r
o
d
u
ce
s
th
e
m
o
s
t
s
tab
le
i
m
ag
e
f
ea
tu
r
e
th
an
o
th
e
r
s
.
T
h
e
co
r
r
elatio
n
b
etwe
en
th
e
Dif
f
er
en
ce
o
f
Gau
s
s
ian
an
d
th
e
L
ap
lacia
n
o
f
g
au
s
s
ian
ca
n
b
e
ex
p
r
ess
ed
u
s
in
g
(
2
)
an
d
(
3
)
.
(
2
)
(
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Ma
tch
in
g
a
lg
o
r
ith
m
p
erfo
r
ma
n
ce
a
n
a
lysi
s
fo
r
a
u
to
ca
lib
r
a
tio
n
meth
o
d
o
f ster
eo
(
R
a
d
e
n
A
r
ief
S
etya
w
a
n
)
1107
T
h
e
k
ey
p
o
i
n
t
o
r
ien
tatio
n
is
a
s
s
ig
n
ed
b
y
u
s
in
g
an
i
m
ag
e
g
r
ad
ien
t.
T
h
e
f
in
al
s
tep
is
th
e
c
o
m
p
u
tatio
n
o
f
th
e
lo
ca
l
im
a
g
e
d
escr
ip
to
r
b
ased
o
n
th
e
g
r
a
d
ien
t
an
d
o
r
ie
n
tatio
n
o
f
t
h
e
k
ey
p
o
in
t.
B
ec
au
s
e
o
f
its
alg
o
r
ith
m
co
m
p
lex
ity
,
SIFT
r
e
q
u
ir
es
a
lar
g
e
co
m
p
u
tatio
n
al
ca
p
ac
it
y
,
ev
en
t
h
o
u
g
h
it
is
v
e
r
y
s
u
itab
le
f
o
r
o
b
jec
t
r
ec
o
g
n
itio
n
ap
p
licatio
n
s
[
1
6
,
1
7
]
.
-
SUR
F
Sp
ee
d
u
p
r
o
b
u
s
t
f
ea
tu
r
e
(
S
UR
F)
tech
n
iq
u
e
p
e
r
f
o
r
m
s
f
aster
th
an
SIFT
[
1
8
]
.
I
n
s
o
m
e
ca
s
es,
i
t
p
er
f
o
r
m
s
with
eq
u
al
q
u
alit
y
to
SIFT
.
SUR
F
tech
n
iq
u
e
is
b
ased
o
n
a
d
escr
ip
to
r
an
d
a
d
etec
to
r
,
wh
ich
is
eq
u
al
to
SIFT
.
I
n
s
tead
o
f
u
s
in
g
th
e
g
au
s
s
ian
av
er
ag
e
o
f
th
e
im
ag
es,
SUR
F
u
s
e
s
s
q
u
ar
es
f
o
r
ap
p
r
o
x
im
atio
n
.
I
t
em
p
lo
y
s
th
e
Hess
ian
m
atr
ix
-
b
ased
B
lo
b
d
etec
to
r
to
f
in
d
t
h
e
p
o
in
t
o
f
i
n
ter
est.
W
av
elet
r
esp
o
n
s
e
is
u
s
ed
f
o
r
o
r
ien
tatio
n
ass
ig
n
m
en
t
b
y
ap
p
ly
in
g
g
au
s
s
ian
weig
h
t.
SUR
F
f
ea
tu
r
e
d
escr
ip
to
r
is
g
en
er
ated
b
y
th
e
wav
elet
r
esp
o
n
s
e
o
f
th
e
s
u
b
r
eg
i
o
n
.
T
h
e
s
u
b
r
eg
io
n
is
th
e
d
iv
is
io
n
o
f
th
e
n
eig
h
b
o
r
ar
o
u
n
d
th
e
k
e
y
p
o
in
t.
T
wo
p
o
in
ts
will f
o
r
m
a
co
r
r
esp
o
n
d
en
ce
(
m
atch
)
if
th
e
y
th
e
s
am
e
co
n
tr
a
s
t,
g
en
er
ated
f
r
o
m
L
a
p
lacia
n
.
-
OR
B
Or
ien
ted
FAST
an
d
r
o
tated
B
R
I
E
F
(
OR
B
)
h
as
b
ee
n
p
r
o
p
o
s
ed
b
y
R
u
b
lee
,
et
al
.
[
1
9
]
.
I
t
i
s
an
o
th
er
alter
n
ativ
e
f
o
r
SIFT
.
OR
B
is
a
co
m
b
in
atio
n
o
f
th
e
FAST
k
ey
p
o
in
t
an
d
th
e
B
R
I
E
F
d
escr
ip
to
r
.
T
h
e
FAST
is
u
s
ed
to
d
eter
m
i
n
e
th
e
k
e
y
p
o
i
n
t
[
2
0
]
.
I
n
th
e
n
e
x
t
s
tep
,
Har
r
is
co
r
n
er
is
u
s
ed
to
f
in
d
th
e
t
o
p
N
p
o
i
n
t.
FAST
co
m
p
u
tes
th
e
in
te
n
s
ity
-
weig
h
ted
ce
n
tr
o
id
,
lo
ca
ted
at
t
h
e
c
en
ter
.
T
h
e
o
r
ien
tatio
n
is
o
b
ta
in
ed
b
y
th
e
v
ec
t
o
r
d
ir
ec
tio
n
to
th
e
ce
n
tr
o
i
d
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
p
u
r
p
o
s
e
o
f
th
is
r
esear
ch
is
to
f
in
d
th
e
b
est
alg
o
r
ith
m
f
o
r
th
e
au
to
-
ca
lib
r
atio
n
o
f
s
ter
eo
v
is
io
n
.
T
h
e
f
ir
s
t
s
tep
o
f
c
alib
r
atio
n
is
th
e
f
i
n
d
in
g
o
f
th
e
co
r
r
esp
o
n
d
in
g
p
o
in
ts
b
etwe
en
two
im
ag
e
s
.
T
h
e
ac
c
u
r
ac
y
o
f
th
is
s
tep
d
eter
m
in
es
th
e
ac
cu
r
ac
y
o
f
s
ter
eo
v
is
io
n
.
T
h
e
o
b
j
ec
t
o
f
th
is
r
esear
ch
is
a
m
icr
o
s
co
p
ic
o
b
ject
with
th
e
s
ize
o
f
a
f
ew
m
illi
m
et
er
s
.
T
h
e
d
is
p
ar
ity
o
f
th
e
p
o
in
ts
is
co
n
v
er
ted
i
n
to
th
e
in
tr
in
s
ic
p
ar
am
eter
o
f
th
e
ca
m
e
r
a.
T
h
e
m
eth
o
d
u
s
ed
in
th
is
r
esear
ch
is
d
escr
ib
ed
in
Fig
u
r
e
2
.
T
h
e
s
ter
eo
im
ag
e
h
as b
ee
n
p
r
o
d
u
ce
d
u
s
in
g
two
ca
m
er
as.
I
n
o
r
d
er
to
h
an
d
le
th
e
v
er
y
n
ar
r
o
w
v
iew
a
r
e
a
ca
u
s
ed
b
y
th
e
s
m
all
-
s
ize
o
b
jects,
th
e
co
n
v
e
r
g
ed
ca
m
er
a
s
etu
p
is
u
s
ed
.
I
t
is
h
ar
d
to
p
u
t
o
b
jects
in
th
e
o
v
er
lap
p
e
d
ar
ea
if
p
ar
allel
ca
m
er
as
ar
e
u
s
ed
.
T
h
e
h
is
to
g
r
a
m
eq
u
ali
za
tio
n
s
t
ep
s
ar
e
r
e
q
u
ir
ed
s
in
ce
th
e
illu
m
in
atio
n
o
f
ea
c
h
im
ag
e
o
r
ca
m
er
a
co
lo
r
ch
a
r
ac
ter
p
o
s
s
ib
ly
d
if
f
er
en
t
[
2
1
]
.
T
o
r
ed
u
ce
th
e
n
o
is
es,
th
e
co
m
b
i
n
atio
n
o
f
Gau
s
s
ian
an
d
m
ed
iu
m
f
ilter
ap
p
lied
.
B
o
th
f
ilter
s
ar
e
p
r
o
p
o
s
ed
t
o
im
p
r
o
v
e
th
e
im
a
g
e
q
u
ality
[
2
2
]
.
Ga
u
s
s
ian
ca
n
b
e
ex
p
r
ess
ed
in
(
4
)
.
W
h
ile
th
e
m
ed
ian
f
ilter
ex
p
r
ess
ed
in
(
5
)
.
A
c
o
m
b
i
n
atio
n
o
f
b
o
th
f
ilter
ex
p
r
ess
ed
u
s
in
g
(
6
)
.
(
4
)
(
5
)
(
6
)
T
h
e
r
esu
lt
o
f
h
is
to
g
r
am
eq
u
ali
za
tio
n
is
p
r
o
ce
s
s
ed
u
s
in
g
a
f
e
atu
r
e
ex
tr
ac
tio
n
al
g
o
r
ith
m
.
T
h
r
ee
f
ea
tu
r
e
ex
tr
ac
t
io
n
al
g
o
r
ith
m
s
ar
e
u
s
ed
to
f
in
d
th
e
m
atch
co
r
r
e
s
p
o
n
d
en
ce
p
o
in
t
o
n
ea
ch
s
et
o
f
im
ag
es
[
2
0
]
.
T
h
e
m
atch
co
r
r
esp
o
n
d
en
ce
p
o
in
t u
s
e
d
f
o
r
th
e
r
ec
tific
atio
n
p
r
o
ce
s
s
[
2
3
]
.
Dis
tan
ce
b
etwe
en
ea
ch
co
r
r
esp
o
n
d
in
g
p
o
in
t
is
u
s
ed
to
e
x
tr
ac
t
th
e
s
ter
eo
p
ar
am
eter
.
T
h
e
o
u
tp
u
t
o
f
th
is
m
eth
o
d
is
th
e
s
ter
eo
ca
lib
r
atio
n
p
ar
am
eter
[
2
4
]
.
T
h
e
r
esu
lt o
f
t
h
is
p
r
o
ce
s
s
ca
n
b
e
tr
a
n
s
f
o
r
m
e
d
in
to
a
3
D
s
u
r
f
ac
e.
Fig
u
r
e
2
.
T
h
e
d
is
tan
ce
m
ea
s
u
r
em
en
t p
r
o
ce
d
u
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
18
,
No
.
2
,
Ap
r
il 2
0
2
0
:
1
1
0
5
-
1
1
1
2
1108
T
wo
in
d
u
s
tr
ial
s
tan
d
ar
d
HD
ca
m
er
a
is
u
s
ed
in
th
i
s
r
ese
ar
ch
.
T
h
ese
ca
m
er
as
ar
e
eq
u
ip
p
ed
with
a
1
0
0
x
len
s
to
en
lar
g
e
th
e
o
b
ject
s
ize.
T
wo
ca
p
t
u
r
ed
im
a
g
es
f
r
o
m
b
o
t
h
ca
m
er
as
a
r
e
t
h
en
co
m
p
ar
e
d
an
d
ev
alu
ated
u
s
in
g
t
h
e
m
atch
in
g
alg
o
r
ith
m
to
f
in
d
th
e
co
r
r
esp
o
n
d
in
g
p
o
in
t.
Fig
u
r
e
3
s
h
o
ws
t
h
e
ca
m
er
a
s
etu
p
a
n
d
th
e
o
b
ject
s
ize.
Fig
u
r
e
3
.
T
h
e
ca
m
er
as set
-
u
p
an
d
th
e
o
b
ject
s
ize
A
m
illi
m
eter
tem
p
late
is
u
s
ed
to
m
ea
s
u
r
e
th
e
s
ize
o
f
th
e
o
b
ject
a
n
d
as
a
r
ef
er
en
ce
o
f
th
e
au
to
-
ca
lib
r
atio
n
.
T
h
e
d
im
en
s
io
n
o
f
th
e
o
b
ject
is
s
h
o
w
n
in
Fig
u
r
e
4
(
a
)
,
wh
er
ea
s
Fig
u
r
e
4
(
b
)
r
ep
r
esen
ts
f
iv
e
p
air
s
o
f
im
ag
e
s
ets
wh
ich
ar
e
g
en
er
ated
u
s
in
g
th
e
s
y
s
tem
f
o
r
test
in
g
p
u
r
p
o
s
e.
E
ac
h
s
e
t
is
co
m
p
ar
ed
u
s
in
g
th
r
ee
m
atch
in
g
alg
o
r
ith
m
s
:
SIFT
,
SUR
F,
an
d
OR
B
.
S
e
t
1
S
e
t
2
S
e
t
3
S
e
t
4
S
e
t
5
(
a)
(
b
)
Fig
u
r
e
4
.
(
a
)
T
h
e
d
im
en
s
io
n
o
f
th
e
o
b
ject
an
d
(
b
)
th
e
d
atasets
u
s
ed
in
th
e
r
esear
ch
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
T
h
e
ex
ec
u
tio
n
o
f
SIFT
,
SUR
F,
an
d
OR
B
o
n
ea
ch
p
air
o
f
im
ag
e
s
ets
h
as
b
ee
n
p
e
r
f
o
r
m
ed
to
f
in
d
th
e
b
est
m
eth
o
d
f
o
r
im
ag
e
m
at
ch
in
g
.
I
n
th
e
o
b
tain
ed
r
esu
lts
,
th
e
g
r
ee
n
lin
e
in
d
icate
s
th
e
co
r
r
esp
o
n
d
e
n
ce
p
o
in
t
b
etwe
en
th
e
lef
t
an
d
r
ig
h
t
im
a
g
es.
T
h
e
n
u
m
b
e
r
o
f
co
n
n
ec
ted
lin
es
s
h
o
ws
th
e
n
u
m
b
er
o
f
m
atch
ed
p
o
in
ts
.
Ho
w
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Ma
tch
in
g
a
lg
o
r
ith
m
p
erfo
r
ma
n
ce
a
n
a
lysi
s
fo
r
a
u
to
ca
lib
r
a
tio
n
meth
o
d
o
f ster
eo
(
R
a
d
e
n
A
r
ief
S
etya
w
a
n
)
1109
ev
er
,
ea
ch
alg
o
r
ith
m
s
till
r
esu
lted
in
an
er
r
o
r
if
t
h
e
alg
o
r
ith
m
f
ailed
to
m
atch
th
e
co
r
r
ec
t
p
o
in
ts
.
T
h
e
r
esu
lt
o
f
th
is
m
atch
in
g
is
u
s
ed
to
g
e
n
er
ate
th
e
ca
lib
r
atio
n
p
ar
am
eter
o
f
s
ter
eo
v
is
io
n
.
3
.
1
.
M
a
t
ching
r
esu
lt
s
us
i
ng
SI
F
T
,
SURF
,
a
nd
O
RB
T
h
e
r
esu
lt
o
f
im
p
lem
en
tin
g
th
e
SIFT
,
SUR
F
an
d
OR
B
alg
o
r
ith
m
s
o
n
th
e
ca
p
tu
r
e
d
o
b
ject
ar
e
g
iv
en
r
esp
ec
tiv
ely
in
Fig
u
r
e
s
5
(
a
-
c)
.
As
s
ee
n
i
n
th
e
im
ag
e
s
et
1
a
n
d
5
o
f
Fig
u
r
e
5
(
a
)
,
o
n
ly
a
f
e
w
lin
es
h
av
e
b
ee
n
g
en
er
ated
b
y
th
e
SIFT
alg
o
r
it
h
m
.
T
h
e
b
ac
k
g
r
o
u
n
d
h
as
v
er
y
h
ig
h
s
im
ilar
ities
b
etwe
en
im
ag
es.
T
h
e
r
esu
lt
o
f
SUR
F
alg
o
r
ith
m
im
p
lem
en
tatio
n
g
iv
e
n
in
Fig
u
r
e
5
(
b
)
in
d
icate
s
th
at
o
n
th
e
im
ag
e
s
et
1
th
er
e
h
av
e
b
ee
n
o
n
ly
a
f
ew
lin
es
g
en
er
ated
b
y
an
alg
o
r
ith
m
an
d
s
o
m
e
lin
es
in
d
icate
d
a
m
ajo
r
er
r
o
r
.
T
h
e
r
est
o
f
th
e
im
ag
e
s
ets
s
h
o
ws th
e
co
r
r
ec
t c
o
r
r
esp
o
n
d
i
n
g
p
o
in
ts
.
T
h
e
r
esu
lt o
f
im
p
le
m
en
tin
g
th
e
OR
B
alg
o
r
ith
m
s
h
o
wn
in
Fig
u
r
e
5
(
c
)
also
in
d
icate
s
th
at
th
er
e
h
av
e
b
ee
n
o
n
l
y
a
f
ew
lin
es
g
e
n
er
ate
d
b
y
th
e
alg
o
r
ith
m
o
n
th
e
i
m
a
g
e
s
et
1
,
with
s
o
m
e
lin
es in
d
icate
d
m
ajo
r
e
r
r
o
r
.
T
h
e
f
o
u
r
o
th
er
im
a
g
e
s
ets in
d
icate
d
th
e
co
r
r
ec
t c
o
r
r
esp
o
n
d
i
n
g
p
o
in
ts
.
T
h
e
co
m
p
ar
is
o
n
o
f
th
e
m
atch
in
g
r
esu
lts
u
s
in
g
SIFT
,
SUR
F,
an
d
OR
B
tech
n
iq
u
es
is
p
r
esen
ted
in
T
ab
le
1
.
I
t
in
d
icate
s
m
atch
in
g
ac
cu
r
ac
y
o
f
th
e
th
r
ee
alg
o
r
ith
m
s
SIFT
,
SUR
F,
an
d
OR
B
.
I
t
ca
n
b
e
k
n
o
wn
f
r
o
m
th
e
tab
le
th
at
th
e
SIFT
alg
o
r
ith
m
g
iv
es
th
e
h
ig
h
est
av
er
ag
e
p
er
ce
n
tag
e
ac
cu
r
ac
y
.
Ho
wev
e
r
,
th
e
p
er
ce
n
tag
e
o
f
co
r
r
ec
t
lin
es
v
ar
ies
d
ep
en
d
in
g
o
n
th
e
im
ag
e
ch
ar
ac
ter
is
tics
.
Fo
r
th
e
im
ag
e
with
h
ig
h
s
im
il
ar
ities
,
SU
R
F
f
ai
led
to
g
iv
e
a
g
o
o
d
r
esu
lt,
wh
er
ea
s
OR
B
co
u
ld
g
en
er
ate
m
an
y
lin
es,
b
u
t w
ith
h
ig
h
er
r
o
r
r
ates.
1
2
3
4
5
(
a)
(
b
)
(
c)
Fig
u
r
e
5
.
E
x
p
er
im
e
n
t r
esu
lts
o
f
m
atch
in
g
al
g
o
r
ith
m
u
s
in
g
:
(a
)
SIFT
,
(
b
)
SUR
F
,
an
d
(
c
)
OR
B
As
s
ee
n
i
n
Fig
u
r
e
5
(
a
)
,
th
e
i
m
ag
e
s
et
1
a
n
d
s
et
5
o
n
ly
h
a
v
e
a
f
ew
lin
es
h
av
e
b
ee
n
g
en
er
ated
b
y
th
e
SIFT
alg
o
r
ith
m
.
T
h
e
b
ac
k
g
r
o
u
n
d
h
as
v
er
y
h
ig
h
s
im
ilar
ities
b
etwe
en
im
ag
es.
I
n
co
n
tr
ast,
th
e
r
esu
lts
o
f
th
e
SUR
F
alg
o
r
ith
m
(
Fig
u
r
e
5
(
b
)
),
wh
ic
h
ap
p
lied
t
o
th
e
im
ag
e
s
et
1
,
o
n
ly
a
f
ew
l
in
es
g
en
er
ated
b
y
th
e
alg
o
r
ith
m
.
T
h
e
r
est
o
f
th
e
im
ag
e
s
ets
s
h
o
ws
th
e
c
o
r
r
ec
t
co
r
r
esp
o
n
d
in
g
p
o
i
n
ts
.
T
h
e
r
esu
lt
o
f
im
p
lem
en
tin
g
th
e
OR
B
alg
o
r
ith
m
s
h
o
wn
in
Fig
u
r
e
5
(
c
)
also
in
d
icate
s
th
a
t
th
er
e
h
av
e
b
ee
n
o
n
ly
a
f
ew
lin
es
g
en
er
ated
b
y
th
e
alg
o
r
ith
m
o
n
th
e
im
a
g
e
s
et
1
,
with
s
o
m
e
lin
es
in
d
icate
d
m
ajo
r
er
r
o
r
.
Par
allel
lin
es,
wh
ich
g
r
o
u
p
to
g
et
h
er
to
f
o
r
m
a
th
ick
er
im
ag
e
s
y
m
b
o
lize
ac
cu
r
ac
y
.
I
n
co
n
tr
ast,
lin
e
o
u
t
o
f
p
ar
allel,
cr
is
s
cr
o
s
s
in
g
ea
ch
o
th
er
cr
ea
tin
g
a
d
is
p
er
s
ed
im
ag
e
s
ig
n
if
y
in
ac
cu
r
ac
y
.
T
h
e
co
m
p
ar
is
o
n
o
f
th
e
m
atch
in
g
r
esu
lts
u
s
in
g
SIFT
,
SUR
F,
an
d
OR
B
tech
n
iq
u
es
is
p
r
esen
ted
in
T
ab
le
1
.
I
t
in
d
icate
s
m
atc
h
in
g
ac
cu
r
ac
y
o
f
th
e
th
r
ee
alg
o
r
i
th
m
s
SIFT
,
SUR
F,
an
d
OR
B
.
As
ca
n
b
e
s
ee
n
in
th
e
tab
le
,
th
e
SIFT
alg
o
r
ith
m
g
iv
es
th
e
h
ig
h
est
av
er
ag
e
ac
cu
r
ac
y
p
er
ce
n
tag
e
.
Ho
wev
er
,
th
e
p
er
ce
n
tag
e
o
f
co
r
r
ec
t
lin
es
v
ar
ies
d
ep
en
d
in
g
o
n
th
e
im
ag
e
ch
ar
ac
ter
is
tics
.
Fo
r
th
e
im
ag
e
with
h
ig
h
s
im
il
ar
ities
,
SU
R
F
f
ai
led
to
g
iv
e
a
g
o
o
d
r
esu
lt,
wh
er
ea
s
OR
B
co
u
ld
g
en
er
ate
m
an
y
lin
es,
b
u
t
with
h
ig
h
er
r
o
r
r
ates.
T
h
e
r
esu
lt
in
T
ab
le
1
co
m
p
a
r
ed
with
th
e
r
esu
lt
f
r
o
m
Kar
am
i
et.
all
[
1
3
]
with
th
e
ca
s
e
o
f
v
a
r
y
in
g
in
ten
s
ity
s
h
o
wn
in
T
ab
le
2
.
I
t
s
h
o
ws
th
at
in
b
o
th
wo
r
k
s
,
SIFT
p
er
f
o
r
m
s
b
etter
th
an
o
t
h
er
m
eth
o
d
s
.
T
ab
le
3
s
h
o
ws
th
e
co
m
p
ar
is
o
n
o
f
th
e
c
o
m
p
u
tatio
n
al
tim
e
o
f
ea
ch
al
g
o
r
ith
m
.
I
t
s
h
o
ws
th
at
th
e
SIFT
m
eth
o
d
r
eq
u
ir
e
d
a
lo
n
g
e
r
tim
e
th
a
n
th
e
o
th
e
r
s
d
u
e
to
th
e
c
o
m
p
lex
alg
o
r
ith
m
co
m
p
u
tatio
n
.
SIFT
r
eq
u
ir
e
d
a
lo
n
g
e
r
tim
e
wh
e
n
th
e
im
ag
e
h
a
d
h
i
g
h
s
im
ilar
ities
in
its
tex
tu
r
e
.
Fig
u
r
e
6
in
d
icate
s
th
at
t
h
e
OR
B
alg
o
r
ith
m
h
as
th
e
f
astes
t
co
m
p
u
tatio
n
tim
e
f
o
r
all
im
a
g
es
s
ets.
I
t
tak
es
less
th
an
0
.
5
s
p
r
o
ce
s
s
in
g
tim
e.
Ho
wev
er
,
th
e
OR
B
alg
o
r
ith
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
18
,
No
.
2
,
Ap
r
il 2
0
2
0
:
1
1
0
5
-
1
1
1
2
1110
g
iv
es
less
m
atch
in
g
r
ates
co
m
p
ar
ed
with
o
th
er
m
eth
o
d
s
.
T
h
e
lin
e
ch
ar
t
in
Fig
u
r
e
6
also
in
d
icate
s
th
at
th
e
co
m
p
lex
ity
o
f
th
e
im
ag
es
lin
ea
r
with
th
e
co
m
p
u
tatio
n
tim
e.
I
m
a
g
e
s
et
1
an
d
5
g
iv
e
th
e
lo
n
g
est
co
m
p
u
tatio
n
tim
e
th
an
o
th
er
i
m
ag
es set b
ec
au
s
e
o
f
th
eir
c
o
m
p
lex
ity
.
T
ab
le
1
.
C
o
m
p
a
r
is
o
n
o
f
th
e
m
atch
in
g
r
esu
lts
u
s
in
g
th
e
SIFT
,
SUR
F,
an
d
OR
B
tech
n
iq
u
es
N
o
.
S
I
F
T
S
U
R
F
OR
B
Li
n
e
s
C
o
r
r
e
c
t
P
o
i
n
t
%
C
o
r
r
e
c
t
Li
n
e
s
C
o
r
r
e
c
t
P
o
i
n
t
%
C
o
r
r
e
c
t
Li
n
e
s
C
o
r
r
e
c
t
P
o
i
n
t
%
C
o
r
r
e
c
t
1
15
14
9
3
.
3
3
%
14
2
1
4
.
2
9
%
78
24
3
0
.
7
7
%
2
1
5
0
1
2
0
8
0
.
0
0
%
4
4
3
4
3
0
9
7
.
0
7
%
89
70
7
8
.
6
5
%
3
4
0
0
3
5
6
8
9
.
0
0
%
2
7
8
2
5
6
9
2
.
0
9
%
2
5
4
2
2
4
8
8
.
1
9
%
4
3
4
5
3
2
1
9
3
.
0
4
%
6
0
0
4
6
7
7
7
.
8
3
%
3
4
5
1
5
6
4
5
.
2
2
%
5
20
16
8
0
.
0
0
%
1
2
5
1
1
2
8
9
.
6
0
%
3
7
5
1
5
3
4
0
.
8
0
%
A
v
e
r
a
g
e
8
7
.
0
8
%
A
v
e
r
a
g
e
7
4
.
1
7
%
A
v
e
r
a
g
e
5
6
.
7
3
%
T
ab
le
2
.
C
o
m
p
a
r
is
o
n
o
f
th
e
m
atch
in
g
r
esu
lts
b
etwe
en
Kar
a
m
i a
n
d
th
is
wo
r
k
M
e
t
h
o
d
M
a
t
c
h
R
a
t
e
(
%)
K
a
r
a
mi
Th
i
s
W
o
r
k
S
I
F
T
7
6
.
7
8
7
.
0
8
S
U
R
F
7
2
.
6
7
4
.
1
7
O
R
B
6
3
.
6
5
6
.
7
3
T
ab
le
3
.
C
o
m
p
u
tatio
n
al
tim
e
u
s
in
g
th
e
SIFT
,
SUR
F,
an
d
OR
B
tech
n
iq
u
es
I
mag
e
S
e
t
C
o
m
p
u
t
a
t
i
o
n
a
l
T
i
me
S
I
F
T
S
U
R
F
O
R
B
1
2
.
1
1
4
0
.
9
2
6
0
.
0
5
2
2
1
.
1
4
9
0
.
7
7
7
0
.
0
3
9
3
0
.
7
8
8
0
.
6
0
.
0
3
3
4
0
.
8
5
8
0
.
5
7
6
0
.
0
3
3
5
1
.
3
6
1
.
1
4
9
0
.
0
7
5
Fig
u
r
e
6
.
C
o
m
p
u
tatio
n
al
tim
e
co
m
p
ar
is
o
n
c
h
ar
t
3
.
2
.
I
m
a
g
e
re
ct
if
ic
a
t
io
n
T
h
e
m
atch
in
g
p
o
in
t
f
r
o
m
p
r
e
v
io
u
s
s
tep
s
is
u
s
ed
f
o
r
r
ec
tify
in
g
th
e
im
a
g
es.
T
h
e
d
if
f
er
e
n
c
e
p
o
s
itio
n
b
etwe
en
s
o
u
r
ce
an
d
d
esti
n
atio
n
p
o
in
t
u
s
ed
as
a
r
ef
er
en
c
e
f
o
r
tr
an
s
f
o
r
m
atio
n
.
Fig
u
r
e
s
7
a
a
n
d
7
b
s
h
o
w
a
d
is
to
r
ted
im
ag
e
f
r
o
m
lef
t a
n
d
r
ig
h
t c
am
e
r
a
.
Fig
u
r
e
7
a
u
s
ed
as th
e
r
ef
er
en
ce
,
wh
ile
th
e
Fig
u
r
e
7
b
is
th
e
o
b
ject
o
f
tr
an
s
f
o
r
m
atio
n
.
T
h
e
r
es
u
lt
o
f
th
e
im
ag
e
tr
an
s
f
o
r
m
atio
n
o
f
Fig
u
r
e
7
b
d
is
p
lay
ed
in
Fig
u
r
e
7
c.
T
h
is
tr
an
s
f
o
r
m
atio
n
b
ased
o
n
t
h
e
h
o
m
o
g
r
a
p
h
y
e
q
u
atio
n
to
r
e
d
u
ce
d
is
to
r
tio
n
[
2
5
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Ma
tch
in
g
a
lg
o
r
ith
m
p
erfo
r
ma
n
ce
a
n
a
lysi
s
fo
r
a
u
to
ca
lib
r
a
tio
n
meth
o
d
o
f ster
eo
(
R
a
d
e
n
A
r
ief
S
etya
w
a
n
)
1111
(
a)
(
b
)
(
c)
Fig
u
r
e
7
.
R
ec
tific
atio
n
r
esu
lt (
a
)
l
ef
t I
m
ag
e
as r
e
f
er
en
ce
(
b
)
r
ig
h
t i
m
ag
e
(
c
)
th
e
r
esu
lt o
f
r
ec
tific
atio
n
3.
3
.
3
D
s
urf
a
ce
g
ener
a
t
io
n
T
h
e
m
atch
in
g
p
r
o
ce
s
s
r
esu
lts
in
th
e
d
is
tan
ce
b
etwe
en
p
o
in
ts
.
Usi
n
g
th
e
d
is
tan
ce
v
alu
es,
a
3
D
s
u
r
f
ac
e
o
b
ject
ca
n
b
e
g
en
er
ated
b
y
p
r
o
jectin
g
th
e
m
o
n
to
th
e
z
-
ax
i
s
[
2
6
,
27]
.
Dis
tan
ce
v
alu
e
b
e
twee
n
b
o
th
im
ag
es
ass
ig
n
ed
as
th
e
d
ep
th
v
alu
e.
I
f
th
e
d
is
tan
ce
is
s
m
all,
th
e
o
b
ject
is
clo
s
er
to
th
e
ca
m
er
a,
an
d
v
ice
v
er
s
a.
Dep
th
v
alu
e
f
o
r
ea
ch
p
ix
el
t
h
an
c
o
n
v
er
ted
to
g
r
a
y
s
ca
le
to
d
i
s
tin
g
u
is
h
th
e
d
ep
th
o
f
p
o
in
t
.
Fig
u
r
e
8
s
h
o
ws
th
e
g
en
er
ated
d
is
p
ar
ity
m
a
p
o
f
th
e
d
ataset
u
s
in
g
SIFT
Ad
ju
s
tm
en
t.
C
o
r
r
elate
d
p
o
in
t
p
r
o
d
u
c
es
b
y
SIF
T
is
u
s
ed
to
ca
lcu
late
th
e
s
ter
eo
ca
m
er
a
p
ar
am
eter
s
.
T
h
e
r
esu
lt
s
h
o
ws
th
at
th
e
alg
o
r
ith
m
s
u
cc
ess
f
u
ll
y
g
en
e
r
ates
m
atch
s
ter
eo
,
h
o
we
v
er
,
t
h
e
n
o
is
y
o
u
tp
u
t
is
a
b
it
ch
allen
g
in
g
.
Usi
n
g
th
e
d
e
p
th
v
alu
e
as
z
-
ax
is
p
r
o
d
u
ce
3
d
v
iew
as
s
h
o
wn
in
Fig
u
r
e
9
t
h
e
al
g
o
r
ith
m
s
u
cc
ess
f
u
lly
p
r
o
d
u
c
es
3
D
r
ec
o
n
s
tr
u
ctio
n
,
b
u
t
th
e
n
o
is
es
r
ed
u
ce
im
ag
e
q
u
ality
.
(
a)
(
b
)
(
c)
Fig
u
r
e
8
.
Gen
e
r
ated
d
e
p
th
v
al
u
e
b
ased
o
n
SIFT
m
atch
in
g
alg
o
r
ith
m
(
a)
(
b
)
(
c)
Fig
u
r
e
9
.
3
D
s
u
r
f
ac
e
r
ec
o
n
s
tr
u
ctio
n
4.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
th
r
ee
d
if
f
er
e
n
t
im
ag
e
m
atch
in
g
tech
n
iq
u
e
s
,
SIFT
,
SUR
F,
an
d
OR
B
,
f
o
r
s
ter
eo
au
to
ca
lib
r
atio
n
s
y
s
tem
h
av
e
b
ee
n
co
m
p
ar
ed
.
SIFT
in
d
icate
s
th
e
b
est
p
er
f
o
r
m
a
n
ce
in
m
o
s
t
s
ce
n
ar
io
s
u
n
d
er
co
n
s
id
er
atio
n
.
I
n
th
e
s
p
ec
ial
c
ase,
wh
en
th
e
im
a
g
es
co
n
tain
m
u
ltip
le
h
ig
h
s
im
ilar
ities
tex
t
u
r
e,
SUR
F
f
ailed
to
g
iv
e
g
o
o
d
r
esu
lt
s
.
I
n
th
e
OR
B
im
p
lem
en
tatio
n
,
th
e
f
ea
tu
r
es
a
r
e
m
o
s
tly
co
n
ce
n
tr
ated
in
o
b
je
cts
at
th
e
ce
n
ter
o
f
th
e
im
ag
e.
W
h
ile
SIFT
an
d
SUR
F,
th
e
f
ea
tu
r
es
ar
e
d
is
tr
ib
u
ted
o
v
er
th
e
im
ag
e
.
T
h
e
3
D
r
e
co
n
s
tr
u
ctio
n
im
ag
e
h
as
s
u
cc
ess
f
u
lly
g
en
er
ated
,
b
u
t
th
e
n
o
is
e
r
e
d
u
ce
s
th
e
q
u
ality
o
f
th
e
im
ag
es.
Fo
r
f
u
tu
r
e
wo
r
k
,
a
g
o
o
d
f
ilter
in
g
alg
o
r
ith
m
r
e
q
u
ir
ed
f
o
r
a
b
etter
r
esu
lt,
with
o
u
t scar
if
y
in
g
th
e
d
etails o
f
im
ag
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
18
,
No
.
2
,
Ap
r
il 2
0
2
0
:
1
1
0
5
-
1
1
1
2
1112
RE
F
E
R
E
NC
E
S
[1
]
J.
C.
H.
Lo
n
d
o
n
o
,
“
S
u
rfa
c
e
ro
u
g
h
n
e
ss
e
stim
a
ti
o
n
b
y
3
D
ste
re
o
S
EM
re
c
o
n
stru
c
ti
o
n
,
”
Un
i
v
e
r
sid
a
d
Na
c
io
n
a
l
d
e
Co
l
o
m
b
ia,
pp.
7
0
,
2
0
1
5
.
[2
]
T.
S
h
ib
a
h
a
ra
a
n
d
T.
A
o
k
i,
“
A
S
u
b
-
P
ix
e
l
S
tere
o
Co
rre
sp
o
n
d
e
n
c
e
T
e
c
h
n
iq
u
e
Ba
se
d
On
1
d
P
h
a
se
-
On
ly
Co
rre
lati
o
n
,”
2
0
0
7
IE
EE
I
n
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
Im
a
g
e
Pr
o
c
e
ss
in
g
,
S
a
n
A
n
to
n
i
o
, TX
,
p
p
.
V
-
2
2
1
-
V
–
2
2
4
,
2
0
0
7
.
[3
]
R.
A.
S
e
t
y
a
wa
n
,
R.
S
u
n
o
k
o
,
M
.
A.
Ch
o
ir
o
n
,
a
n
d
P
.
M
.
Ra
h
a
rd
j
o
,
“
Im
p
lem
e
n
tatio
n
o
f
ste
re
o
v
is
io
n
se
m
i
-
g
lo
b
a
l
b
lo
c
k
m
a
tch
in
g
m
e
t
h
o
d
s
fo
r
d
i
sta
n
c
e
m
e
a
su
re
m
e
n
t,
”
In
d
o
n
e
s.
J
.
El
e
c
tr.
E
n
g
.
C
o
mp
u
t.
S
c
i
.,
v
o
l.
1
2
,
n
o
.
2
,
p
p
.
5
8
5
–
5
9
1
,
2
0
1
8
.
[4
]
T.
Da
n
g
,
C.
Ho
ffm
a
n
n
,
a
n
d
C
.
S
ti
ll
e
r,
“
C
o
n
ti
n
u
o
u
s
ste
re
o
se
lf
-
c
a
li
b
ra
ti
o
n
b
y
c
a
m
e
ra
p
a
ra
m
e
ter
trac
k
in
g
,
”
IEE
E
T
ra
n
s
a
c
ti
o
n
s
Ima
g
e
Pr
o
c
e
ss
.
,
v
o
l
.
1
8
,
n
o
.
7
,
p
p
.
1
5
3
6
–
1
5
5
0
,
Ju
l
y
2
0
0
9
.
[5
]
H.
Wan
g
e
t
a
l.
,
“
An
a
u
to
m
a
ti
c
se
lf
-
c
a
li
b
ra
ti
o
n
a
p
p
r
o
a
c
h
f
o
r
wi
d
e
b
a
se
li
n
e
ste
re
o
c
a
m
e
ra
s
u
sin
g
se
a
su
rfa
c
e
ima
g
e
s,”
2
0
1
5
I
EE
E
I
n
ter
n
a
ti
o
n
a
l
W
o
rk
sh
o
p
o
n
A
d
v
a
n
c
e
d
Ro
b
o
ti
c
s
a
n
d
it
s
S
o
c
ia
l
Im
p
a
c
ts
(AR
S
O),
L
y
o
n
,
p
p
.
1
-
6
,
2
0
1
5
.
[6
]
Z.
Xia
o
,
J.
Li
a
n
g
,
D.
Yu
,
Z.
Tan
g
,
a
n
d
A.
As
u
n
d
i,
“
An
a
c
c
u
ra
te
ste
re
o
v
isio
n
sy
ste
m
u
sin
g
c
ro
ss
-
sh
a
p
e
d
targ
e
t
se
lf
-
c
a
li
b
ra
ti
o
n
m
e
th
o
d
b
a
se
d
o
n
p
h
o
to
g
ra
m
m
e
try
,
”
Op
ti
c
s
a
n
d
L
a
se
rs
in
En
g
i
n
e
e
rin
g
,
v
o
l.
4
8
,
n
o
.
1
2
,
p
p
.
1
2
5
2
–
1
2
6
1
,
De
c
e
m
b
e
r
2
0
1
0
.
[7
]
S
.
S
h
a
h
,
N.
Bh
u
iy
a
n
,
a
n
d
O.
O.
Kh
a
li
fa
,
“
Eff
ic
ien
t
3
D
ste
re
o
v
i
sio
n
sta
b
il
iza
ti
o
n
f
o
r
m
u
lt
i
-
c
a
m
e
ra
v
iew
p
o
in
ts,
”
Bu
ll
e
ti
n
o
f
El
e
c
trica
l
E
n
g
in
e
e
rin
g
a
n
d
I
n
fo
rm
a
t
ics
,
v
o
l.
8
,
n
o
.
3
,
p
p
.
8
8
2
–
8
8
9
,
S
e
p
tem
b
e
r
2
0
1
9
.
[8
]
H.
Yu
a
n
d
Y.
Wan
g
,
“
An
imp
ro
v
e
d
se
lf
-
c
a
li
b
ra
ti
o
n
m
e
th
o
d
fo
r
a
c
t
iv
e
ste
re
o
c
a
m
e
r
a
,
”
2
0
0
6
6
th
W
o
r
ld
Co
n
g
re
ss
o
n
In
telli
g
e
n
t
Co
n
tro
l
a
n
d
Au
t
o
ma
t
io
n
,
Da
l
ia
n
,
p
p
.
5
1
8
6
-
5
1
9
0
,
2
0
0
6
.
[9
]
B.
H.
Ya
n
g
a
n
d
F
.
L.
Li
a
n
,
“
C
o
n
ti
n
u
o
u
s
se
lf
-
c
a
li
b
ra
t
io
n
o
f
ste
re
o
c
a
m
e
ra
b
a
se
d
o
n
p
a
rti
c
le
fil
ter
fra
m
e
wo
rk
fo
r
e
n
d
o
sc
o
p
ic
v
id
e
o
s
,
”
2
0
1
6
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
In
d
u
stria
l
T
e
c
h
n
o
lo
g
y
(ICIT
),
Taip
e
i,
p
p
.
1
4
7
6
-
1
4
8
1
,
2
0
1
6
.
[1
0
]
I.
El
Ba
tt
e
o
u
i,
K.
S
a
t
o
ri,
a
n
d
A
.
S
a
a
id
i
,
“
A
S
tere
o
P
la
n
a
r
S
e
lf
-
Ca
li
b
ra
ti
o
n
M
e
t
h
o
d
o
f
a
Ca
m
e
ra
with
Va
riab
le
In
tri
n
sic
P
a
ra
m
e
ters
,
”
2
0
1
6
1
3
t
h
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Co
mp
u
ter
Gr
a
p
h
ics
,
Ima
g
in
g
a
n
d
V
isu
a
li
za
ti
o
n
(CGiV),
Be
n
i
M
e
ll
a
l,
p
p
.
6
-
1
1
,
2
0
1
6
.
[1
1
]
Y.
Hu
,
Q.
Ch
e
n
,
S
.
F
e
n
g
,
T.
Ta
o
,
A.
As
u
n
d
i,
a
n
d
C.
Zu
o
,
“
A
n
e
w
m
icro
sc
o
p
ic
tele
c
e
n
tri
c
ste
re
o
v
isio
n
sy
ste
m
Ca
li
b
ra
ti
o
n
,
re
c
ti
fica
ti
o
n
,
a
n
d
th
re
e
-
d
ime
n
sio
n
a
l
re
c
o
n
stru
c
ti
o
n
,
”
Op
ti
c
s
a
n
d
L
a
se
rs
in
E
n
g
i
n
e
e
rin
g
,
v
o
l.
1
1
3
,
p
p
.
1
4
–
2
2
,
F
e
b
ru
a
r
y
2
0
1
9
.
[1
2
]
D.
S
c
h
a
rste
in
a
n
d
R
.
S
z
e
li
sk
i,
“
A
Tax
o
n
o
m
y
a
n
d
E
v
a
lu
a
ti
o
n
o
f
De
n
se
Tw
o
-
F
ra
m
e
S
tere
o
Co
rre
sp
o
n
d
e
n
c
e
Alg
o
rit
h
m
s,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
Vi
si
o
n
,
v
o
l.
4
7
,
n
o
.
1
-
3
,
p
p
.
7
–
4
2
,
Ap
r
il
2
0
0
2
.
[1
3
]
E.
Ka
ra
m
i,
S
.
P
ra
sa
d
,
a
n
d
M
.
S
h
e
h
a
ta,
“
Im
a
g
e
M
a
tch
in
g
Us
in
g
S
I
F
T
,
S
UR
F
,
BRI
EF
a
n
d
ORB:
P
e
rfo
rm
a
n
c
e
Co
m
p
a
riso
n
fo
r
Disto
rte
d
Im
a
g
e
s,”
Co
n
fer
e
n
c
e
:
2
0
1
5
Ne
wf
o
u
n
d
l
a
n
d
El
e
c
trica
l
a
n
d
Co
m
p
u
te
r
En
g
in
e
e
rin
g
Co
n
fer
e
n
c
e
,
2
0
1
5
.
[1
4
]
M
.
Ok
ti
a
n
a
,
F
.
Arn
ia,
Y.
Aw
a
y
,
a
n
d
K.
M
u
n
a
d
i,
“
F
e
a
tu
re
s fo
r
c
r
o
ss
sp
e
c
tral
ima
g
e
m
a
tch
in
g
:
A su
r
v
e
y
,
”
Bu
ll
e
ti
n
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
In
fo
rm
a
t
ics
,
v
o
l
.
7
,
n
o
.
4
,
p
p
.
5
5
2
–
5
6
0
,
De
c
e
m
b
e
r
2
0
1
8
.
[1
5
]
S
.
Ke
y
p
o
i
n
ts
a
n
d
D
.
G
.
Lo
we
,
“
Distin
c
ti
v
e
Im
a
g
e
F
e
a
tu
re
s
fro
m
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
C
o
mp
u
ter
Vi
sio
n
,
v
o
l.
6
0
,
n
o
.
2
,
p
p
.
9
1
–
1
1
0
,
N
o
v
e
m
b
e
r
2
0
0
4
.
[1
6
]
I.
Re
y
Ote
ro
a
n
d
M
.
De
lb
ra
c
io
,
“
An
a
to
m
y
o
f
t
h
e
S
IF
T
M
e
th
o
d
,
”
Ima
g
e
Pro
c
e
ss
in
g
o
n
L
in
e
,
v
o
l.
4
,
p
p
.
3
7
0
–
3
9
6
,
2
0
1
4
.
[1
7
]
C.
Je
a
n
d
H.
M
.
P
a
rk
,
“
Op
ti
m
iz
e
d
h
iera
rc
h
ica
l
b
lo
c
k
m
a
tch
in
g
f
o
r
fa
st
a
n
d
a
c
c
u
ra
te
ima
g
e
re
g
ist
ra
ti
o
n
,
”
S
i
g
n
a
l
Pro
c
e
ss
in
g
:
Ima
g
e
Co
mm
u
n
.
,
v
o
l
.
2
8
,
n
o
.
7
,
p
p
.
7
7
9
–
7
9
1
,
Au
g
u
st
2
0
1
3
.
[1
8
]
H.
Ba
y
,
A.
Ess,
T.
T
u
y
tela
a
rs,
a
n
d
L
.
Va
n
G
o
o
l,
“
S
p
e
e
d
e
d
-
Up
Ro
b
u
st
F
e
a
tu
re
s
(S
URF),
”
Co
mp
u
ter
Vi
sio
n
a
n
d
Ima
g
e
Un
d
e
rs
t
a
n
d
in
g
,
v
o
l
.
1
1
0
,
n
o
.
3
,
p
p
.
3
4
6
–
3
5
9
,
J
u
n
e
2
0
0
8
.
[1
9
]
E.
R
u
b
lee
,
V.
Ra
b
a
u
d
,
K.
Ko
n
o
li
g
e
,
a
n
d
G
.
Bra
d
s
k
i,
“
ORB:
An
e
fficie
n
t
a
lt
e
rn
a
ti
v
e
t
o
S
IF
T
o
r
S
URF
2
0
1
1
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
C
o
mp
u
ter
Vi
si
o
n
,
Ba
rc
e
lo
n
a
,
p
p
.
2
5
6
4
-
2
5
7
1
,
2
0
1
1
.
[2
0
]
O.
An
d
e
rss
o
n
a
n
d
S
.
R.
M
a
r
q
u
e
z
,
“
A
c
o
m
p
a
riso
n
o
f
o
b
jec
t
d
e
tec
ti
o
n
a
lg
o
r
it
h
m
s
u
sin
g
u
n
m
a
n
i
p
u
late
d
tes
ti
n
g
ima
g
e
s: Co
m
p
a
rin
g
S
IF
T
,
KA
ZE
,
AK
AZE
a
n
d
ORB,”
2
0
1
6
.
[2
1
]
Y.
Ch
a
n
g
,
C.
J
u
n
g
,
P
.
Ke
,
H.
S
o
n
g
,
a
n
d
J.
Hw
a
n
g
,
“
Au
to
m
a
ti
c
C
o
n
tras
t
-
Li
m
it
e
d
Ad
a
p
t
iv
e
Histo
g
r
a
m
Eq
u
a
li
z
a
ti
o
n
with
Du
a
l
G
a
m
m
a
Co
rre
c
ti
o
n
,
”
I
EE
E
Acc
e
ss
,
v
o
l
.
6
,
p
p
.
1
1
7
8
2
–
1
1
7
9
2
,
2
0
1
8
.
[2
2
]
B.
Ch
a
n
,
K.
H.
Li
m
,
L.
G
o
p
a
l,
A.
A.
G
o
p
a
lai,
a
n
d
D.
G
o
u
wa
n
d
a
,
“
S
tere
o
v
isi
o
n
h
u
m
a
n
m
o
ti
o
n
d
e
tec
ti
o
n
a
n
d
trac
k
in
g
i
n
u
n
c
o
n
tr
o
ll
e
d
e
n
v
iro
n
m
e
n
t,
”
T
EL
KOM
NIKA
T
e
lec
o
m
mu
n
ica
t
io
n
Co
mp
u
ti
n
g
El
e
c
tro
n
i
c
s
a
n
d
Co
n
tro
l
,
v
o
l.
1
6
,
n
o
.
3
,
p
p
.
9
5
5
–
9
6
4
,
Ju
n
e
2
0
1
8
.
[2
3
]
X.
Li
,
W.
Li
u
,
W.
F
a
n
,
J.
S
u
n
,
a
n
d
N.
S
a
t
o
sh
i
,
“
P
e
rsp
e
c
ti
v
e
c
o
rr
e
c
ti
o
n
u
sin
g
c
a
m
e
ra
in
tri
n
sic
p
a
ra
m
e
ters
,
”
2
0
1
6
IEE
E
1
3
th
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
S
ig
n
a
l
P
ro
c
e
ss
in
g
(IC
S
P),
Ch
e
n
g
d
u
,
p
p
.
8
5
4
-
8
5
8
,
2
0
1
6
.
[2
4
]
C.
Lo
o
p
a
n
d
Z.
Zh
a
n
g
,
“
Co
m
p
u
ti
n
g
re
c
ti
fy
in
g
h
o
m
o
g
ra
p
h
ies
fo
r
st
e
re
o
v
isio
n
,
”
Pro
c
e
e
d
in
g
s.
1
9
9
9
I
EE
E
Co
mp
u
te
r
S
o
c
iety
Co
n
fer
e
n
c
e
o
n
C
o
mp
u
te
r
Vi
sio
n
a
n
d
P
a
tt
e
rn
Rec
o
g
n
it
i
o
n
(Ca
t
.
No
PR
0
0
1
4
9
)
,
F
o
rt
Co
l
l
in
s,
CO,
U
S
A,
v
o
l.
1
,
p
p
.
1
2
5
-
1
3
1
,
1
9
9
9
.
[2
5
]
Ku
k
e
lo
v
a
Z
.
,
He
ll
e
r
J
.
,
Bu
jn
a
k
M
.
,
P
a
jd
la
T
.
,
“
Ra
d
ial
Disto
rt
io
n
Ho
m
o
g
ra
p
h
y
,
”
2
0
1
5
IEE
E
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ter
V
isio
n
a
n
d
Pa
t
ter
n
Rec
o
g
n
it
io
n
(CV
PR
),
B
o
sto
n
,
M
A,
p
p
.
6
3
9
-
6
4
7
,
2
0
1
5
.
[2
6
]
K.
P
e
n
g
,
X.
Ch
e
n
,
D.
Zh
o
u
a
n
d
Y.
Li
u
,
"
3
D
re
c
o
n
stru
c
ti
o
n
b
a
se
d
o
n
S
I
F
T
a
n
d
Ha
rris
fe
a
t
u
re
p
o
in
ts,"
2
0
0
9
IEE
E
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
R
o
b
o
t
ics
a
n
d
Bi
o
mime
ti
c
s (R
OBIO)
,
G
u
il
in
,
p
p
.
9
6
0
-
9
6
4
,
2
0
0
9
.
[2
7
]
K.
Ya
m
a
d
a
a
n
d
A.
Kim
u
ra
,
"
A
p
e
rfo
rm
a
n
c
e
e
v
a
lu
a
ti
o
n
o
f
k
e
y
p
o
i
n
ts
d
e
tec
ti
o
n
m
e
th
o
d
s
S
IF
T
a
n
d
AK
AZE
fo
r
3
D
re
c
o
n
stru
c
ti
o
n
,
"
2
0
1
8
In
ter
n
a
ti
o
n
a
l
W
o
rk
sh
o
p
o
n
Ad
v
a
n
c
e
d
Ima
g
e
T
e
c
h
n
o
l
o
g
y
(IW
AIT
)
,
Ch
ia
n
g
M
a
i
,
p
p
.
1
-
4
,
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.