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
.
5
,
Octo
b
er
2
0
2
0
,
p
p
.
2
4
4
7
~
2
4
5
3
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
i5
.
1
6
6
3
4
2447
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
Abno
rma
l activ
it
y
det
ec
tion in su
r
v
eilla
nce video
sc
enes
J
wa
n J
a
m
a
l A
li,
Na
rj
is
M
ez
a
a
l Sha
t
i,
M
et
ha
q T
a
lib
G
a
a
t
a
Co
ll
e
g
e
o
f
S
c
ien
c
e
,
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
,
Al
-
M
u
sta
n
si
riy
a
Un
i
v
e
rsity
,
Ira
q
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
7
,
2
0
2
0
R
ev
is
ed
Ap
r
2
5
,
2
0
2
0
Acc
ep
ted
May
8
,
2
0
2
0
Au
to
m
a
ted
d
e
tec
ti
o
n
o
f
a
b
n
o
r
m
a
l
a
c
ti
v
it
y
a
ss
u
m
e
s
a
si
g
n
ifi
c
a
n
t
tas
k
in
su
rv
e
il
lan
c
e
a
p
p
li
c
a
ti
o
n
s.
Th
is
p
a
p
e
r
p
re
se
n
ts
a
n
i
n
telli
g
e
n
t
fra
m
e
wo
rk
v
i
d
e
o
su
rv
e
il
lan
c
e
t
o
d
e
tec
t
a
b
n
o
rm
a
l
h
u
m
a
n
a
c
ti
v
i
ty
in
a
n
a
c
a
d
e
m
ic
e
n
v
iro
n
m
e
n
t
th
a
t
tak
e
s
in
t
o
a
c
c
o
u
n
t
th
e
se
c
u
r
it
y
a
n
d
e
m
e
rg
e
n
c
y
a
sp
e
c
ts
b
y
fo
c
u
si
n
g
o
n
th
re
e
a
b
n
o
rm
a
l
a
c
ti
v
it
ies
(fa
ll
i
n
g
,
b
o
x
i
n
g
a
n
d
wa
v
in
g
).
Th
is
fra
m
e
wo
rk
d
e
si
g
n
e
d
to
c
o
n
sist o
f
th
e
two
e
ss
e
n
ti
a
l
p
ro
c
e
ss
e
s: t
h
e
first
o
n
e
is
a
trac
k
in
g
s
y
ste
m
th
a
t
c
a
n
fo
ll
o
w
targ
e
ts
with
i
d
e
n
ti
f
y
s
e
ts
o
f
fe
a
tu
re
s
t
o
u
n
d
e
rsta
n
d
h
u
m
a
n
a
c
ti
v
it
y
a
n
d
m
e
a
su
re
d
e
sc
rip
ti
v
e
in
f
o
rm
a
ti
o
n
o
f
e
a
c
h
targ
e
t.
Th
e
se
c
o
n
d
o
n
e
is
a
d
e
c
isio
n
sy
ste
m
t
h
a
t
c
a
n
re
a
li
z
e
if
th
e
a
c
ti
v
i
ty
o
f
t
h
e
targ
e
t
trac
k
is "
n
o
rm
a
l"
o
r
"
a
b
n
o
rm
a
l” t
h
e
n
e
n
e
rg
izin
g
a
larm
wh
e
n
re
c
o
g
n
ize
d
a
b
n
o
rm
a
l
a
c
t
iv
it
ies
.
K
ey
w
o
r
d
s
:
An
o
m
aly
d
etec
tio
n
Mo
tio
n
o
b
ject
d
etec
tio
n
R
ea
l
-
tim
e
p
r
o
ce
s
s
in
g
T
r
ac
k
in
g
Vid
eo
s
u
r
v
eillan
ce
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
:
J
wan
J
am
al
Ali,
C
o
lleg
e
o
f
Scien
ce
,
Dep
ar
tm
e
n
t o
f
C
o
m
p
u
ter
Scien
ce
,
Al
-
Mu
s
tan
s
ir
iy
a
Un
iv
er
s
ity
,
B
ag
h
d
ad
,
I
r
aq
.
E
m
ail:
jwan
.
jam
al4
8
4
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
An
o
m
aly
d
ef
i
n
ed
as
a
b
eh
av
io
r
th
at
d
ev
iates
b
ased
o
n
wh
at
v
iewe
d
as
n
o
r
m
al
o
r
s
tan
d
ar
d
ac
co
r
d
in
g
to
th
e
d
o
m
ai
n
.
T
h
e
an
aly
s
is
o
f
th
e
ab
n
o
r
m
al
ac
tiv
ity
in
th
e
v
id
eo
s
eq
u
en
ce
h
as a
ttra
cted
r
em
ar
k
ab
le
an
d
r
ap
i
d
atten
tio
n
in
th
e
f
ield
o
f
v
id
eo
s
u
r
v
eillan
ce
d
u
e
to
its
ef
f
icien
cy
b
y
f
ilter
in
g
a
lar
g
e
n
u
m
b
er
o
f
u
n
n
ec
ess
ar
y
in
f
o
r
m
atio
n
a
n
d
th
e
av
ailab
ilit
y
o
f
m
an
y
h
u
m
an
an
d
m
ater
ial
r
eso
u
r
ce
s
[
1
]
.
Vid
eo
s
u
r
v
eillan
ce
s
y
s
tem
s
h
av
e
b
ec
o
m
e
o
n
e
o
f
th
e
m
o
s
t
p
r
o
m
in
e
n
t
r
esear
ch
i
n
s
ec
u
r
ity
an
d
em
er
g
e
n
cy
ap
p
licatio
n
s
in
v
o
lv
in
g
th
e
id
en
tific
atio
n
a
n
d
class
if
icatio
n
o
f
h
u
m
a
n
ac
tiv
ities
in
to
n
o
r
m
al
o
r
ab
n
o
r
m
al
ac
tiv
ities
.
I
n
s
p
ite
o
f
th
e
f
ac
t
th
at
th
er
e
ar
e
m
an
y
p
r
o
p
o
s
ed
alg
o
r
ith
m
s
an
d
tech
n
iq
u
es
in
th
is
m
o
tiv
atin
g
to
p
ic,
r
esear
ch
in
th
is
d
o
m
ain
s
till
in
s
u
f
f
icien
cy
t
o
two
s
ig
n
i
f
ican
t
th
in
g
s
:
d
atasets
th
at
p
u
b
lic
ac
ce
s
s
ib
le
an
d
g
e
n
er
al
c
o
m
p
ar
ativ
e
ass
ess
m
en
t.
Als
o
,
litt
le
ch
an
g
es
in
illu
m
in
atio
n
,
s
h
ad
o
ws,
n
u
m
b
er
o
f
o
b
j
ec
ts
th
at
ap
p
ea
r
in
th
e
s
ce
n
e
ca
n
tu
r
n
th
e
lo
o
k
o
f
th
e
s
ce
n
e
ca
u
s
in
g
a
lo
t
o
f
f
alse
alar
m
s
r
atio
,
lo
s
s
o
f
o
b
ject
t
r
ac
k
in
g
,
a
n
d
o
t
h
er
r
ela
ted
p
r
o
b
lem
s
th
at
m
ay
n
o
t
r
esp
o
n
d
in
r
ea
l
-
tim
e
to
a
n
y
s
u
s
p
icio
u
s
ev
en
t.
th
er
ef
o
r
e,
s
u
ch
s
y
s
tem
s
h
av
e
a
wid
e
ar
ea
o
f
r
esear
ch
to
p
r
o
m
o
te
s
o
m
e
o
f
th
eir
ca
p
ab
ilit
ies
[
1
]
.
T
h
e
ter
m
"su
r
v
eillan
ce
"
is
th
e
ac
tiv
ity
to
lo
o
k
o
u
t f
o
r
.
Su
r
v
ei
llan
ce
is
th
e
d
em
o
n
s
tr
atio
n
o
f
o
b
s
er
v
in
g
tire
less
an
d
tr
a
n
s
ien
t
o
b
jects
in
s
id
e
a
ce
r
tain
en
v
ir
o
n
m
en
t.
I
t
tar
g
ets
to
f
o
r
m
u
late
u
s
ef
u
l
d
ata
f
r
o
m
a
lar
g
e
o
f
"v
id
eo
s
"
o
v
er
all
b
y
o
b
s
er
v
ati
o
n
ca
m
er
as
u
s
in
g
:
d
etec
tio
n
,
s
eg
m
en
tatio
n
,
i
n
ter
est
o
b
jects
tr
ac
k
in
g
,
a
u
to
m
atic
r
ec
o
g
n
ize
an
d
u
n
d
er
s
tan
d
th
ei
r
b
eh
av
io
r
.
T
h
ey
ar
e
s
ig
n
if
ica
n
t
to
o
ls
th
at
s
u
p
p
o
r
t
h
u
m
an
s
b
y
ex
p
an
d
in
g
th
eir
co
g
n
itiv
e
ab
ilit
ies an
d
r
ea
s
o
n
i
n
g
ab
o
u
t d
if
f
er
en
t in
ter
esti
n
g
s
itu
atio
n
s
[
2
]
.
−
R
esear
ch
AI
M
T
h
e
g
o
al
o
f
th
is
p
a
p
er
is
to
d
e
s
ig
n
a
s
y
s
tem
th
at
h
as
th
e
a
b
ilit
y
to
d
etec
t
h
u
m
an
s
,
t
r
ac
k
in
g
b
eh
av
io
r
,
an
d
class
if
icatio
n
as
a
n
o
r
m
al
ac
tiv
ity
(
walk
in
g
)
o
r
a
b
n
o
r
m
al
ac
tiv
ities
(
f
allin
g
,
b
o
x
in
g
an
d
wa
v
in
g
)
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
.
5
,
Octo
b
e
r
2
0
2
0
:
2
4
4
7
-
2453
2448
f
r
o
m
v
id
eo
s
o
f
m
o
n
ito
r
in
g
s
t
u
d
en
ts
in
a
n
ac
ad
e
m
ic
en
v
ir
o
n
m
en
t
t
h
at
tak
es
in
to
ac
co
u
n
t
th
e
s
ec
u
r
ity
an
d
em
er
g
en
cy
asp
ec
ts
.
T
h
is
p
ap
er
p
r
o
p
o
s
es
way
s
to
co
m
p
lete
th
e
tr
ac
k
in
g
an
d
class
if
icatio
n
o
f
h
u
m
an
ac
tiv
ity
b
y
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
u
s
in
g
t
h
e
Gau
s
s
ian
m
ix
tu
r
e
m
o
d
el
(
GM
M)
f
o
r
d
etec
tin
g
th
e
m
o
v
in
g
o
b
jects.
T
h
en
u
s
ed
th
e
f
u
zz
y
C
-
m
ea
n
s
clu
s
ter
in
g
(
FC
M)
tech
n
iq
u
e
to
s
eg
m
en
t
th
e
im
a
g
es
f
o
r
f
u
r
th
er
cu
s
to
m
ize
th
e
o
b
ject
a
n
d
d
is
tin
g
u
is
h
it
f
r
o
m
ad
jace
n
t
o
b
jects.
C
o
m
b
in
ed
(
HARR
I
S
-
SIFT
)
alg
o
r
ith
m
s
to
g
eth
e
r
to
ex
tr
ac
t
f
ea
tu
r
es
an
d
th
e
Kalm
an
f
ilter
t
o
tr
ac
k
in
g
t
ar
g
ets.
Fin
ally
,
K
-
n
ea
r
est
n
eig
h
b
o
r
(
KNN)
u
s
ed
t
o
class
if
y
th
e
ac
tiv
ities
"n
o
r
m
al
an
d
a
b
n
o
r
m
al
”.
W
h
en
r
ec
o
g
n
i
zin
g
o
n
e
o
f
th
e
a
b
n
o
r
m
al
ac
tiv
ities
th
e
s
y
s
tem
g
en
er
ates
s
o
u
n
d
alar
m
s
t
o
id
e
n
tify
th
is
ac
tiv
ity
with
a
r
ed
lab
el
a
r
o
u
n
d
th
e
p
er
s
o
n
o
r
p
e
r
s
o
n
s
in
v
o
lv
ed
.
W
h
ils
t
th
e
y
ello
w
la
b
el
id
en
tifie
d
ar
o
u
n
d
th
e
p
er
s
o
n
s
th
at
th
e
s
y
s
tem
r
ec
o
g
n
ized
as n
o
r
m
al
ac
tiv
ity
.
−
L
iter
atu
r
e
r
ev
iew
I
n
[
3
]
Pro
p
o
s
ed
a
s
y
s
tem
to
d
etec
t
u
n
u
s
u
al
ac
tiv
ity
in
r
ea
l
-
tim
e
v
id
eo
s
u
r
v
eillan
ce
.
T
h
is
p
ap
er
p
r
o
p
o
s
ed
a
m
eth
o
d
t
o
co
n
s
i
s
t
o
f
th
r
ee
ess
en
tial
p
r
o
ce
s
s
es.
First,
b
ac
k
g
r
o
u
n
d
s
u
b
tr
ac
tio
n
u
s
ed
to
d
etec
t
th
e
m
o
tio
n
.
Seco
n
d
,
s
k
eleto
n
izatio
n
alg
o
r
ith
m
s
ap
p
lied
.
Fin
ally
,
u
n
u
s
u
al
ev
en
t
d
etec
tio
n
b
y
m
atch
in
g
th
e
Sk
eleto
n
im
ag
e
f
r
am
e
with
a
r
ef
er
en
ce
im
ag
e
f
r
am
e
in
t
h
e
d
atab
ase
an
d
s
ettin
g
a
r
ed
b
o
x
o
n
th
e
f
r
am
e
.
I
n
[
4
]
Pro
p
o
s
ed
o
f
I
SS
to
d
etec
t
h
u
m
an
b
eh
a
v
io
r
in
a
u
n
iv
er
s
al
en
v
ir
o
n
m
en
t
b
y
u
s
in
g
tem
p
o
r
al
-
d
if
f
er
en
cin
g
f
o
r
m
o
v
in
g
o
b
ject
d
etec
tio
n
an
d
G
au
s
s
ian
f
u
n
ctio
n
to
lo
ca
te
m
o
tio
n
s
r
eg
io
n
.
T
h
is
m
eth
o
d
u
s
ed
a
f
ilter
,
its
s
h
ap
e
m
o
d
el
b
ased
o
n
e
q
u
atio
n
n
a
m
ed
(
OM
E
GA)
to
ig
n
o
r
e
n
o
n
-
h
u
m
an
o
b
jects
an
d
a
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e
(
SVM)
to
class
if
y
o
b
jects
in
to
n
o
r
m
al
an
d
ab
n
o
r
m
al
b
e
h
av
io
r
,
an
d
u
s
e
m
o
d
el
f
o
r
r
etr
ie
v
in
g
th
e
o
b
ject
d
et
ec
ted
f
r
o
m
th
e
d
ataset
to
i
d
en
tific
atio
n
b
y
u
s
ed
o
f
co
n
ten
t
-
b
ased
im
a
g
e
r
etr
iev
al
(
C
B
I
R
)
.
I
n
[
5
]
Pro
p
o
s
ed
a
s
ec
u
r
ity
s
y
s
tem
to
th
e
d
etec
tio
n
o
f
an
a
n
o
m
aly
m
o
tio
n
b
y
class
if
y
in
g
d
if
f
er
e
n
t
m
o
tio
n
s
u
s
in
g
th
e
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e
(
SVM)
class
if
ier
.
T
h
is
was
d
o
n
e
b
y
u
s
in
g
th
e
b
ac
k
g
r
o
u
n
d
s
u
b
tr
ac
tio
n
to
d
etec
t
th
e
m
o
v
in
g
o
b
jects
an
d
a
Kalm
an
f
ilter
f
o
r
tr
ac
k
in
g
.
I
n
[
6
]
Pro
p
o
s
ed
a
s
y
s
tem
to
d
etec
t
o
b
ject
awa
r
e
ab
n
o
r
m
al
ac
tiv
ity
b
ased
o
n
b
lo
ck
f
o
r
e
g
r
o
u
n
d
s
eg
m
en
tatio
n
to
r
estrict
th
e
an
aly
s
is
o
f
m
o
v
in
g
o
b
jects
f
r
o
m
th
e
s
p
a
r
s
e
m
atr
ix
.
T
h
e
o
b
jects
ar
e
th
en
r
e
p
r
esen
ted
u
s
in
g
th
e
tr
ajec
to
r
ies an
d
th
en
t
h
e
h
is
to
g
r
am
is
b
u
ilt.
2.
DATAS
E
T
T
h
e
v
id
eo
s
u
r
v
eillan
ce
d
ataset
in
clu
d
es
f
o
u
r
class
es
o
f
ac
tiv
ities
r
ec
o
r
d
ed
b
y
a
ca
m
er
a.
T
h
en
s
ep
ar
ated
in
to
n
o
r
m
al
ac
tiv
ities
an
d
ab
n
o
r
m
al
ac
tiv
ities
.
T
h
e
n
o
r
m
al
ac
tiv
ities
in
clu
d
e
o
n
e
class
(
walk
in
g
)
an
d
th
e
ab
n
o
r
m
al
ac
tiv
ities
in
clu
d
e
th
r
ee
class
es
(
f
a
llin
g
,
b
o
x
in
g
an
d
wav
in
g
)
f
r
o
m
v
a
r
io
u
s
s
ce
n
ar
io
s
at
d
if
f
e
r
en
t
tim
es
(
m
o
r
n
in
g
,
af
ter
n
o
o
n
)
,
s
u
n
n
y
an
d
clo
u
d
y
wea
th
er
,
a
n
d
d
if
f
er
en
t
d
is
tan
ce
b
etwe
en
th
e
ca
m
er
a
an
d
p
er
s
o
n
s
.
Fig
u
r
e
1
s
h
o
wn
f
r
am
es a
s
ex
a
m
p
les o
f
th
ese
ac
tiv
ities
.
T
h
e
ac
tiv
ities
p
er
f
o
r
m
ed
in
o
u
r
in
d
o
o
r
ac
ad
e
m
ic
d
ep
ar
tm
e
n
t
with
a
s
tatic
ca
m
er
a
(
L
o
g
itech
HD
Pro
W
eb
ca
m
C
9
2
0
)
,
wh
ich
co
n
tai
n
s
1
2
3
v
i
d
eo
s
(
8
0
tr
ain
in
g
,
4
3
test
)
.
All
with
AVI
f
ile
f
o
r
m
at.
T
h
e
v
id
e
o
s
th
at
u
s
ed
f
o
r
tr
ai
n
in
g
a
n
d
test
in
g
h
av
e
th
e
f
o
llo
win
g
m
etr
ics:
−
Fra
m
es p
er
Seco
n
d
(
f
p
s
)
: 3
0
−
R
eso
lu
tio
n
f
r
am
e:
4
8
0
x
3
6
0
−
T
o
tal
T
r
ain
in
g
Fra
m
es: 2
0
4
5
3
f
r
am
es
−
T
o
tal
T
esti
n
g
Fra
m
es: 1
0
7
9
3
f
r
am
es
Fig
u
r
e
1
.
E
x
am
p
les o
f
s
o
m
e
th
e
n
o
r
m
al
a
n
d
ab
n
o
r
m
al
o
f
th
e
h
u
m
an
ac
tiv
ities
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
A
b
n
o
r
ma
l a
ctivity
d
etec
tio
n
i
n
s
u
r
ve
illa
n
ce
vid
eo
s
ce
n
es
(
Jwa
n
Ja
ma
l A
li
)
2449
3.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
f
o
c
u
s
es
o
n
m
o
n
ito
r
in
g
h
u
m
a
n
in
an
ac
ad
em
ic
en
v
ir
o
n
m
en
t
t
h
at
h
as
th
e
ab
ilit
y
to
d
etec
t
th
e
m
o
v
in
g
o
b
jects
f
r
o
m
v
id
eo
b
y
s
ep
ar
atin
g
th
e
f
o
r
eg
r
o
u
n
d
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
,
ex
tr
ac
t
f
ea
tu
r
es
o
f
m
o
tio
n
,
t
r
ac
k
in
g
,
an
d
r
ec
o
g
n
it
io
n
o
f
n
o
r
m
al
ac
tiv
ities
:
(
walk
in
g
)
a
n
d
a
b
n
o
r
m
al
ac
tiv
ities
:
(
f
allin
g
,
b
o
x
in
g
an
d
wav
in
g
)
.
Fig
u
r
e
2
illu
s
tr
ated
th
e
p
r
o
p
o
s
ed
s
y
s
tem
.
T
h
e
ab
n
o
r
m
al
ac
tiv
ity
d
etec
tio
n
s
y
s
tem
s
tep
ca
n
b
e
s
u
m
m
ar
ized
as th
e
f
o
llo
win
g
s
tep
s
:
Step
1
: I
n
p
u
t
v
id
eo
(
r
aw
d
ata)
f
r
o
m
th
e
ca
m
er
a
s
en
s
o
r
.
Step
2
: Pr
e
-
Pro
ce
s
s
in
g
with
two
s
tag
es:
−
C
o
n
v
er
t v
id
e
o
to
f
r
am
es
−
Mo
tio
n
d
etec
tio
n
I
n
th
e
first
s
tag
e,
v
id
eo
clip
s
ar
e
co
n
v
er
ted
in
t
o
s
eq
u
en
tial
f
r
a
m
es
an
d
th
e
tar
g
et
d
etec
tio
n
b
y
th
e
GM
M
alg
o
r
ith
m
in
t
h
e
s
ec
o
n
d
s
tag
e.
Step
3
: U
s
e
f
u
zz
y
C
-
m
ea
n
s
(
F
C
M)
clu
s
ter
in
g
to
m
o
r
e
ac
cu
r
a
te
an
aly
s
is
.
Step
4
: T
h
e
f
ea
tu
r
es
o
f
tar
g
et
s
f
o
r
ea
ch
s
eq
u
en
tial f
r
am
e
ar
e
ex
tr
ac
ted
u
s
in
g
(
Har
r
is
-
SIFT
)
f
ea
tu
r
es.
Step
5
: U
s
e
Kalm
an
f
ilter
tr
ac
k
in
g
to
tr
ac
k
tar
g
ets.
Step
6
: Reco
g
n
itio
n
o
f
th
e
ac
ti
v
ities
an
d
class
if
icatio
n
in
to
n
o
r
m
al
o
r
ab
n
o
r
m
al
u
s
in
g
th
e
KNN
alg
o
r
ith
m
.
Step
7
:
Gen
er
ate
alar
m
wh
e
n
r
ec
o
g
n
izin
g
o
n
e
o
f
th
e
a
b
n
o
r
m
al
ac
tiv
ities
.
Fig
u
r
e
2
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
d
iag
r
am
3
.
1
.
P
re
pro
ce
s
s
ing
T
h
e
p
r
im
a
r
y
p
r
ep
r
o
ce
s
s
in
g
b
asis
is
to
en
h
an
ce
im
a
g
e
q
u
ality
th
at
p
r
e
v
en
ts
u
n
wan
ted
d
is
to
r
tio
n
.
Ma
k
in
g
it
ap
p
r
o
p
r
iate
f
o
r
h
u
m
an
in
ter
p
r
etatio
n
an
d
m
ac
h
in
e
p
er
ce
p
tio
n
th
at
im
p
r
o
v
es
th
e
ex
tr
ac
tio
n
o
f
th
e
im
p
o
r
tan
t
o
b
ject
f
ea
tu
r
es
f
o
r
f
u
r
th
er
p
r
o
ce
s
s
in
g
.
T
h
er
ef
o
r
e,
co
n
s
id
er
e
d
to
b
e
o
n
e
o
f
th
e
ess
en
tial
an
d
n
ec
ess
ar
y
p
h
ases
o
f
th
e
v
id
e
o
s
u
r
v
eillan
ce
s
y
s
tem
s
[
7
]
.
Vid
eo
s
ar
e
s
eq
u
en
ce
s
o
f
im
ag
es
ev
er
y
o
n
e
o
f
th
ese
im
ag
es
ca
lled
a
f
r
am
e,
s
h
o
we
d
in
q
u
ick
en
o
u
g
h
f
r
e
q
u
en
c
y
f
o
r
th
ese
r
ea
s
o
n
s
th
e
ey
es
o
f
h
u
m
an
s
ca
n
p
e
r
ce
p
t
th
e
co
n
g
r
u
ity
o
f
its
co
n
ten
t.
All
tech
n
iq
u
es
o
f
im
ag
e
p
r
o
c
ess
in
g
ca
n
b
e
ab
le
to
ap
p
ly
in
d
iv
id
u
al
f
r
am
es
th
at
m
an
ip
u
latio
n
a
n
d
a
n
aly
s
is
,
in
p
ar
ticu
lar
,
to
e
n
h
a
n
ce
its
q
u
ality
with
in
th
e
lim
its
o
f
th
e
wo
r
k
an
d
p
u
r
p
o
s
e
o
f
t
h
e
p
r
o
ce
s
s
in
g
[
8
]
.
Fig
u
r
e
3
s
h
o
ws
th
e
v
id
eo
f
r
a
m
es
o
f
d
if
f
er
en
t
s
ce
n
ar
io
s
,
an
d
t
h
e
f
r
am
e
s
eq
u
en
ce
o
f
v
id
e
o
s
d
ataset
ar
e
v
is
ib
le.
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
.
5
,
Octo
b
e
r
2
0
2
0
:
2
4
4
7
-
2453
2450
Fig
u
r
e
3
.
Fra
m
es seq
u
e
n
ce
o
f
v
id
eo
s
d
ataset
3.
1
.
1
.
No
is
e
a
nd
im
a
g
es e
nh
a
ncem
ent
No
is
e
is
an
u
n
d
esira
b
le
ef
f
ec
t
o
n
p
ix
el
v
alu
es a
n
d
will p
r
esen
t r
ath
er
d
if
f
er
e
n
t in
ten
s
ity
v
al
u
es with
in
d
ig
ital
im
ag
es
s
u
ch
as
b
ac
k
g
r
o
u
n
d
n
o
is
e
an
d
b
l
u
r
r
ed
o
b
ject
s
th
at
alwa
y
s
p
r
esen
t
th
r
o
u
g
h
im
ag
e
ac
q
u
is
itio
n
,
tr
an
s
m
is
s
io
n
,
en
co
d
in
g
a
n
d
p
r
o
ce
s
s
in
g
s
tep
s
[
9
]
.
−
C
o
n
tr
ast
en
h
an
ce
m
en
t
C
o
n
tr
ast
is
a
s
ig
n
if
ican
t
f
ac
to
r
in
ea
ch
s
u
b
jectiv
e
ass
ess
m
en
t
o
f
im
ag
e
q
u
ality
,
wh
e
r
ea
s
co
n
tr
ast
is
th
e
v
ar
iatio
n
in
v
is
u
al
p
r
o
p
er
t
ies
th
at
m
ak
e
an
o
b
ject
id
en
ti
f
iab
le
f
r
o
m
o
th
er
o
b
jects
an
d
b
ac
k
g
r
o
u
n
d
.
C
o
lo
r
im
ag
e
co
n
tr
ast
en
h
an
ce
m
en
t
is
d
o
n
e
b
y
c
o
n
v
e
r
tin
g
an
im
ag
e
t
o
an
o
th
er
c
o
lo
r
s
p
ac
e
(
L
*
a*
b
*
)
o
n
L
*
(
lu
m
in
o
s
ity
lay
er
)
th
e
co
n
tr
ast
ad
ju
s
tm
en
t
ac
co
m
p
lis
h
ed
af
ter
th
at
tr
an
s
f
o
r
m
ed
th
e
im
ag
e
b
ac
k
to
th
e
co
lo
r
s
p
ac
e
(
R
GB
)
.
T
h
e
p
ix
els in
ten
s
ity
im
p
ac
ts
th
e
lu
m
in
o
s
ity
,
wh
er
ea
s
r
ea
l c
o
l
o
r
s
p
r
eser
v
ed
[
1
0
]
.
−
Sp
atial
f
ilter
in
g
Sp
atial
f
ilter
in
g
m
o
d
if
ies
th
e
im
ag
e
b
y
e
x
ch
an
g
in
g
th
e
v
al
u
e
o
f
ea
c
h
b
y
p
ix
el
with
th
e
f
u
n
ctio
n
o
f
th
e
v
alu
es o
f
t
h
at
p
ix
el
an
d
it'
s
n
ea
r
b
y
[
1
1
]
.
B
y
u
s
in
g
th
e
f
o
ll
o
win
g
s
y
n
tax
:
h
=
f
s
p
ec
ial
(
'
d
is
k
’
,
r
ad
iu
s
)
3
.
1
.
2
.
B
a
ck
g
r
o
un
d
s
ub
t
ra
ct
i
o
n
A
tech
n
iq
u
e
u
s
ed
to
d
etec
t
m
o
tio
n
in
th
e
v
id
eo
s
ce
n
es
b
y
u
s
in
g
th
e
v
id
eo
f
r
am
es
to
s
u
b
tr
ac
t
th
e
cu
r
r
en
t
f
r
am
e
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
m
o
d
el
th
at
p
r
ev
io
u
s
ly
o
b
tain
ed
a
n
d
class
if
ied
th
r
o
u
g
h
two
m
ai
n
s
tep
s
:
b
ac
k
g
r
o
u
n
d
m
o
d
elin
g
an
d
f
o
r
eg
r
o
u
n
d
e
x
tr
ac
tio
n
[
1
2
]
.
Gau
s
s
ian
m
ix
tu
r
e
m
o
d
el
(
GM
M)
is
o
n
e
o
f
th
e
p
o
p
u
lar
an
d
r
o
b
u
s
t
tech
n
iq
u
es
f
o
r
co
n
s
tr
u
ct
a
b
ac
k
g
r
o
u
n
d
m
o
d
el
to
s
eg
m
en
t
m
o
v
in
g
o
b
jects
f
r
o
m
t
h
e
b
ac
k
g
r
o
u
n
d
t
h
at
ass
ig
n
s
to
a
p
r
o
b
a
b
ilit
y
d
en
s
ity
f
u
n
ctio
n
r
ep
r
esen
ted
as
a
s
u
m
o
f
Gau
s
s
ian
d
en
s
ities
.
B
ased
o
n
v
ar
ian
ce
an
d
p
er
s
is
ten
ce
,
G
au
s
s
ian
s
ca
teg
o
r
ized
as "f
o
r
e
g
r
o
u
n
d
"
an
d
"b
ac
k
g
r
o
u
n
d
".
T
h
e
v
alu
e
o
f
p
ix
els if
n
o
t
r
ep
r
es
en
t th
e
b
ac
k
g
r
o
u
n
d
d
is
tr
ib
u
tio
n
s
th
en
tak
e
in
to
th
e
f
o
r
eg
r
o
u
n
d
th
is
will
b
e
d
o
n
e
u
n
til
it
is
Gau
s
s
ian
with
co
n
s
i
s
ten
t
an
d
s
u
f
f
icien
t
ev
id
en
ce
s
u
p
p
o
r
ted
[
1
3
]
.
Acc
u
r
ate
f
o
r
e
g
r
o
u
n
d
d
etec
tio
n
is
a
d
if
ficu
lt
task
in
r
ea
l
-
ti
m
e
b
ec
au
s
e
th
e
r
ea
l
-
w
o
r
ld
o
f
th
e
v
id
eo
f
r
am
es
s
eq
u
en
ce
s
i
n
clu
d
e
m
a
n
y
cr
itical
s
itu
atio
n
s
.
T
h
e
m
ain
ch
allen
g
e
to
ch
an
g
e
d
ete
ctio
n
alg
o
r
ith
m
s
is
th
e
ca
s
tin
g
s
h
ad
o
ws
ac
c
o
m
p
an
y
in
g
f
o
r
eg
r
o
u
n
d
o
b
jects.
Fo
r
t
h
o
s
e
r
ea
s
o
n
s
,
th
e
f
o
r
eg
r
o
u
n
d
m
ask
was
p
r
o
ce
s
s
ed
u
s
in
g
a
3
×3
m
ed
iu
m
f
ilter
with
a
m
o
r
p
h
o
lo
g
ical
o
p
er
ati
o
n
(
o
p
en
in
g
an
d
clo
s
in
g
)
.
I
t
is
o
f
ten
u
s
ed
s
in
ce
it
is
en
o
u
g
h
to
elim
in
ate
n
o
is
e
an
d
at
th
e
s
am
e
tim
e
p
r
eser
v
e
th
e
ed
g
es.
See
Fig
u
r
e
4
th
e
ex
p
er
im
en
t
p
r
o
ce
s
s
in
g
f
r
am
es
in
in
d
o
o
r
ac
tiv
ities
a
s
s
h
o
wed
in
:
(
a)
th
e
f
r
am
e
with
a
m
o
v
in
g
o
f
t
h
e
o
b
jec
t,
(
b
)
th
e
co
n
tr
ast
en
h
an
ce
m
e
n
t,
(
c)
th
e
f
o
r
eg
r
o
u
n
d
d
etec
ted
with
th
e
s
h
ad
o
w
,
an
d
(
d
)
s
h
o
we
d
th
e
f
o
r
e
g
r
o
u
n
d
m
ask
af
ter
th
e
p
r
o
ce
s
s
in
g
.
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
A
b
n
o
r
ma
l a
ctivity
d
etec
tio
n
i
n
s
u
r
ve
illa
n
ce
vid
eo
s
ce
n
es
(
Jwa
n
Ja
ma
l A
li
)
2451
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
4
.
T
h
e
ex
p
er
im
en
t p
r
o
c
ess
in
g
f
r
am
es in
in
d
o
o
r
ac
tiv
it
ies:
(
a)
f
r
am
e
with
a
m
o
v
in
g
o
b
ject
,
(
b
)
co
n
tr
ast
en
h
an
ce
m
en
t
,
(
c
)
f
o
r
eg
r
o
u
n
d
d
etec
ted
with
s
h
a
d
o
w
,
an
d
(
d
)
f
o
r
eg
r
o
u
n
d
m
ask
af
ter
p
r
o
ce
s
s
ed
3
.2
.
I
ma
g
e
s
eg
m
ent
a
t
io
n
Dig
ital
im
ag
e
p
ar
titi
o
n
in
g
i
n
t
o
m
u
ltip
le
p
ar
ts
(
g
r
o
u
p
s
o
f
p
i
x
els)
r
ef
er
r
ed
t
o
as
im
ag
e
s
eg
m
en
tatio
n
.
T
h
e
o
b
jectiv
e
is
to
f
ac
ilit
ate
(
a
n
d
/o
r
)
m
o
d
if
y
in
g
t
h
e
r
eg
io
n
s
o
f
in
ter
est
in
th
e
im
ag
e
t
o
b
e
s
i
m
p
ler
an
aly
ze
d
an
d
m
o
r
e
b
en
e
f
icial
p
r
o
ce
s
s
ed
esp
ec
ially
af
ter
f
o
r
eg
r
o
u
n
d
ex
t
r
a
ctio
n
wh
en
an
in
ac
c
u
r
ate
m
er
g
er
o
cc
u
r
s
b
etwe
en
p
eo
p
le
an
d
th
e
o
p
ac
ity
o
f
th
e
o
b
ject.
f
u
zz
y
C
-
m
ea
n
s
(
FC
M)
is
a
s
o
f
t
clu
s
ter
in
g
alg
o
r
ith
m
wh
er
e
ea
ch
elem
en
t
ca
n
h
av
e
a
p
lace
with
m
o
r
e
th
an
o
n
e
g
ath
er
in
g
,
th
u
s
(
FC
M)
ca
n
b
e
ex
tr
em
ely
f
ast
b
ec
au
s
e
o
f
th
e
n
u
m
b
er
o
f
iter
at
io
n
s
d
em
an
d
t
o
o
b
tain
a
s
p
ec
if
ic
clu
s
ter
in
g
p
r
ac
tice
id
e
n
tity
to
th
e
d
e
m
an
d
in
g
ac
cu
r
a
cy
[
1
4
]
.
3
.3
.
M
o
t
io
n
t
r
a
ck
ing
a
nd
f
e
a
t
ure
ex
t
ra
c
t
io
n
Fo
r
tr
ac
k
in
g
an
y
o
b
ject,
f
ea
t
u
r
e
ex
tr
ac
tio
n
p
lay
s
a
s
ig
n
if
i
ca
n
t
r
o
le.
C
o
m
b
in
ed
(
Har
r
is
an
d
Scale
I
n
v
ar
ian
t
Featu
r
e
T
r
an
s
f
o
r
m
(
SIFT
)
)
wh
ich
is
p
r
o
p
o
s
ed
in
th
is
s
y
s
tem
to
ex
tr
ac
t
f
ea
tu
r
es
o
f
o
b
jects
th
at
r
ed
u
ctio
n
o
f
d
im
e
n
s
io
n
al
ity
an
d
r
ed
u
ce
th
e
n
u
m
b
e
r
o
f
r
eso
u
r
ce
s
n
ee
d
ed
to
d
escr
ib
e
a
lar
g
e
r
an
g
e
o
f
d
ata
[
1
5
-
1
7
]
.
T
r
ac
k
i
n
g
o
b
jects
d
etec
ted
in
f
r
am
es seq
u
en
ce
a
n
d
m
atch
in
g
th
em
is
a
cr
itical
s
t
ag
e
o
f
in
tellig
en
t
s
ec
u
r
ity
s
y
s
tem
s
b
ec
au
s
e
o
f
th
e
ab
ilit
y
t
o
e
x
tr
ac
t
th
e
o
b
jects
an
d
a
n
aly
ze
th
eir
b
eh
av
i
o
r
[
1
8
]
.
Kalm
a
n
f
ilter
is
a
m
ath
em
atica
l
an
d
r
ec
u
r
s
iv
e
f
ilter
.
T
h
e
wo
r
d
"f
ilter
"
is
u
tili
ze
d
b
ec
au
s
e
it
is
th
e
p
r
o
ce
s
s
o
f
r
et
u
r
n
s
th
e
b
est
esti
m
ate
o
f
n
o
is
y
d
ata
a
"
f
ilter
o
u
t"
th
e
n
o
is
e.
I
t
is
an
esti
m
ate
ac
q
u
ir
ed
b
y
in
te
g
r
ated
b
o
t
h
"p
r
ed
ictio
n
"
an
d
"c
o
r
r
ec
tio
n
"
[
1
9
,
20]
.
T
h
e
m
ajo
r
o
b
jectiv
es to
u
s
e
a
Kalm
an
filt
er
ar
e
as f
o
llo
ws:
−
Pre
d
ict
th
e
f
u
tu
r
e
lo
ca
tio
n
o
f
o
b
jects.
−
R
ed
u
ce
n
o
is
e
d
u
e
t
o
in
ac
cu
r
at
e
d
etec
tio
n
.
−
Ass
o
ciate
m
u
ltip
le
o
b
jects f
o
r
th
eir
tr
ac
k
s
.
3
.4
.
Act
i
o
n
re
co
g
nitio
n a
nd
cla
s
s
if
ica
t
io
n
Af
ter
ex
tr
ac
tio
n
f
ea
tu
r
es,
th
e
class
if
icatio
n
alg
o
r
ith
m
a
p
p
lied
to
im
ag
es
o
r
v
id
eo
s
[
2
0
,
2
1
]
.
C
las
s
ificatio
n
is
th
e
d
ata
an
aly
s
is
p
r
o
ce
s
s
b
y
a
s
et
o
f
tr
ain
in
g
d
ata
to
fin
d
a
m
o
d
el
th
at
d
is
tin
g
u
is
h
es
an
d
d
escr
ib
es
d
ata
class
es
an
d
th
u
s
r
ec
o
g
n
itio
n
o
f
ac
tiv
ities
th
at
u
s
ed
f
o
r
th
e
o
b
jectiv
e
o
f
ab
n
o
r
m
ality
d
etec
tio
n
[
1
]
.
L
o
w
co
m
p
lex
ity
a
n
d
s
im
p
le
y
e
t
ef
f
icien
t
in
m
an
y
ca
s
es
class
if
ier
alg
o
r
ith
m
n
am
ed
"K
-
Nea
r
est
Neig
h
b
o
r
s
(
KNN)
".
T
h
is
alg
o
r
ith
m
class
if
ied
b
y
m
atch
in
g
th
e
an
o
n
y
m
o
u
s
d
ata
with
g
r
o
u
p
s
o
f
s
im
ilar
tr
ain
ed
d
ata
u
s
in
g
th
e
"E
u
clid
ea
n
Dis
tan
ce
"
as
a
s
im
ilar
ity
m
ea
s
u
r
e.
t
o
p
r
ev
e
n
t
th
e
lar
g
er
r
an
g
e
attr
ib
u
tes
f
r
o
m
o
v
er
r
id
in
g
o
f
th
e
s
m
aller
r
an
g
e
attr
ib
u
tes,
attr
ib
u
te
v
alu
es
ar
e
s
et
[
2
2
,
2
3
]
.
T
h
e
class
if
icatio
n
o
f
(
KNN)
,
m
ea
n
t th
at
th
e
m
o
s
t
an
o
n
y
m
o
u
s
p
atter
n
ass
ig
n
ed
t
h
e
m
o
s
t
p
r
e
v
alen
t
class
es
am
o
n
g
th
e
n
ea
r
est
n
eig
h
b
o
r
s
’
cl
ass
es.
I
n
[
2
4
,
2
5
]
i
f
th
er
e
ar
e
two
class
es
lin
k
ed
,
th
e
lin
k
o
f
th
e
l
o
west
av
er
ag
e
d
is
tan
ce
ass
ig
n
ed
to
th
e
an
o
n
y
m
o
u
s
p
atter
n
[
2
6
,
2
7
]
.
4.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S AN
D
D
I
SC
USS
I
O
N
T
h
e
r
an
d
o
m
n
a
t
u
r
e
an
d
h
ig
h
c
o
m
p
lex
ity
o
f
h
u
m
an
m
o
v
em
en
ts
,
in
ad
d
itio
n
to
,
th
e
is
s
u
e
o
f
in
ad
eq
u
ac
y
d
atasets
th
at
p
u
b
licly
ac
ce
s
s
ib
le
lead
s
to
d
if
f
ic
u
lty
an
d
tr
i
ck
y
o
f
class
if
y
in
g
s
u
s
p
icio
u
s
h
u
m
an
ac
tiv
ities
o
f
in
ter
est
an
d
m
ak
es
th
e
d
etec
tio
n
o
f
s
u
s
p
icio
u
s
ac
tiv
ities
is
c
h
allen
g
in
g
.
T
h
ese
is
s
u
es
lead
i
n
g
to
in
co
n
s
is
ten
cies
in
ex
p
er
im
en
tal
r
esu
lts
in
s
ev
e
r
al
p
ap
e
r
s
in
liter
atu
r
e.
T
o
v
ali
d
ate
th
e
p
r
o
p
o
s
ed
s
y
s
tem
p
e
r
f
o
r
m
an
ce
,
we
ap
p
lied
th
e
ap
p
r
o
ac
h
to
a
v
id
eo
s
eq
u
en
ce
in
r
ea
l
-
tim
e
th
at
r
ec
o
r
d
ed
at
o
u
r
ac
ad
em
ic
d
ep
ar
t
m
en
t
in
an
in
d
o
o
r
en
v
ir
o
n
m
en
t
with
th
e
s
tatic
ca
m
er
a
(
L
o
g
itech
HD
Pro
W
eb
c
am
C
9
2
0
)
.
Fo
r
th
is
p
u
r
p
o
s
e,
MA
T
L
AB
(
R
2
0
1
9
a)
to
o
l u
s
ed
an
d
im
p
lem
en
te
d
o
n
a
PC
with
a
2
.
8
0
GHz
p
r
o
ce
s
s
o
r
co
r
e
i7
an
d
1
6
.
0
GB
R
AM
.
I
n
f
ac
t,
th
at
i
n
m
o
s
t
o
f
th
e
tr
a
ck
in
g
s
ce
n
ar
io
s
,
th
e
r
e
is
n
o
t
m
u
ch
ch
a
n
g
e
in
th
e
o
b
ject'
s
a
r
ticu
latio
n
b
etwe
en
th
e
s
u
cc
ess
iv
e
f
r
am
es.
So
,
af
ter
th
e
ex
p
er
im
en
ts
,
we
h
av
e
ad
o
p
ted
in
o
u
r
s
y
s
tem
th
e
b
asis
o
f
p
r
o
ce
s
s
in
g
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
.
5
,
Octo
b
e
r
2
0
2
0
:
2
4
4
7
-
2453
2452
ev
er
y
th
r
ee
f
r
am
es
in
s
tead
o
f
t
h
e
h
ig
h
c
o
m
p
u
tatio
n
al
o
f
f
r
am
e
-
to
-
f
r
am
e
f
ea
tu
r
e
m
atc
h
in
g
th
at
r
eq
u
ir
es
ex
ten
s
iv
e
p
r
o
ce
s
s
in
g
to
r
u
n
in
r
ea
l
-
tim
e.
T
h
e
ex
p
e
r
im
en
tal
r
esu
lts
o
f
t
h
e
s
y
s
tem
s
h
o
wed
ac
h
iev
e
d
p
r
o
m
is
in
g
r
esu
lts
in
m
o
s
t te
s
t c
o
n
d
itio
n
s
.
T
o
e
v
alu
ate
th
e
s
y
s
tem
p
er
f
o
r
m
s
th
e
f
o
llo
win
g
m
etr
ics u
s
ed
:
Acc
u
r
ac
y
(
AC
C
)
A
C
C
=
.
.
ℎ
(
1
)
Dete
ctio
n
R
ate
(
DR
)
DR
=
T
r
ue
Po
s
i
t
i
v
e
T
r
ue
Po
s
i
t
i
v
e
+
F
al
s
eN
eg
at
i
v
e
(
2
)
Fals
e
Alar
m
R
ate
(
FA
R)
FAR
=
F
al
s
e
Po
s
i
t
i
v
e
T
r
ue
N
eg
at
i
v
e
+
F
al
s
e
Po
s
i
t
i
v
e
(
3
)
T
h
e
ex
p
er
im
e
n
tal
r
esu
lt
o
f
ac
tio
n
class
if
icatio
n
is
as
s
h
o
wn
in
Fig
u
r
e
5
.
T
ab
le
1
Sh
o
ws
s
am
p
les
o
f
th
e
ex
p
er
im
en
tal
r
esu
lts
th
at
wer
e
lis
t
in
g
to
s
h
o
w
th
e
q
u
an
titativ
e
r
esu
lts
as
s
o
ciate
d
with
ac
tiv
ity
in
ter
est b
eh
av
io
r
.
Fig
u
r
e
5
.
R
esu
lt o
f
ac
tio
n
class
if
icatio
n
in
a
d
if
f
er
e
n
t scen
e
T
ab
le
1
.
E
x
p
er
im
en
tal
r
esu
lt
N
u
mb
e
r
o
f
p
e
o
p
l
e
B
e
h
a
v
i
o
r
A
c
c
u
r
a
c
y
%
D
e
t
e
c
t
i
o
n
R
a
t
e
%
F
a
l
se
A
l
a
r
m
R
a
t
e
%
1
n
o
r
m
a
l
9
9
.
1
5
%
9
9
.
2
8
%
2
.
0
5
%
1
a
b
n
o
r
m
a
l
9
7
.
1
4
%
9
8
.
0
7
%
3
.
0
5
%
2
N
o
r
mal
a
n
d
a
b
n
o
r
ma
l
9
6
.
5
0
%
9
8
.
1
8
%
4
.
1
2
%
2
n
o
r
m
a
l
9
7
.
0
5
%
9
6
.
0
4
%
5
.
8
4
%
2
a
b
n
o
r
m
a
l
9
8
.
8
0
%
9
8
.
7
8
%
4
.
8
8
%
mu
l
t
i
p
l
e
n
o
r
m
a
l
9
4
.
5
6
%
9
6
.
5
7
%
8
.
5
4
%
mu
l
t
i
p
l
e
a
b
n
o
r
m
a
l
9
4
.
7
9
%
9
5
.
5
9
%
9
.
5
9
%
mu
l
t
i
p
l
e
N
o
r
mal
a
n
d
a
b
n
o
r
ma
l
9
3
.
9
6
%
9
5
.
5
6
%
9
.
4
5
%
5.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
SCO
P
E
T
h
is
wo
r
k
p
r
o
p
o
s
es
an
au
to
m
ated
r
ea
l
-
tim
e
v
id
eo
s
u
r
v
eillan
ce
s
y
s
tem
to
tr
ac
k
tar
g
ets
an
d
r
ec
o
g
n
ize
n
o
r
m
al
an
d
ab
n
o
r
m
al
h
u
m
an
ac
tiv
ity
in
an
ac
ad
em
ic
f
ield
th
at
tak
es
in
to
ac
co
u
n
t
th
e
s
ec
u
r
ity
an
d
e
m
er
g
e
n
c
y
asp
ec
ts
.
T
h
e
test
o
f
r
esu
lts
o
f
th
e
r
ea
l
-
tim
e
v
id
eo
s
eq
u
en
ce
s
d
e
m
o
n
s
tr
ate
th
e
ef
f
ec
tiv
e
n
ess
o
f
d
esig
n
to
r
ec
o
g
n
ize
an
d
ev
alu
ate
h
u
m
a
n
ac
tiv
ity
d
u
e
to
th
e
h
ig
h
v
alu
es
o
f
ac
c
u
r
a
cy
,
d
etec
tio
n
r
ate
wh
ile
a
lo
w
f
alse
alar
m
r
ate.
As
f
u
tu
r
e
wo
r
k
class
if
ied
m
o
r
e
s
u
s
p
icio
u
s
ac
tiv
ities
th
at
ca
n
id
en
tify
ac
cu
r
ate
ac
tio
n
s
.
Mo
r
e
ad
v
an
ce
d
alg
o
r
ith
m
s
d
esig
n
ed
f
o
r
r
ea
l
-
ti
m
e
v
id
eo
s
u
r
v
eillan
ce
.
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
A
b
n
o
r
ma
l a
ctivity
d
etec
tio
n
i
n
s
u
r
ve
illa
n
ce
vid
eo
s
ce
n
es
(
Jwa
n
Ja
ma
l A
li
)
2453
ACK
NO
WL
E
DG
M
E
N
T
T
h
e
au
th
o
r
th
an
k
s
th
e
Dep
ar
tm
en
t
o
f
C
o
m
p
u
ter
Scien
ce
,
C
o
lleg
e
o
f
Scien
ce
,
Mu
s
tan
s
ir
iy
ah
Un
iv
er
s
ity
,
f
o
r
s
u
p
p
o
r
tin
g
th
is
wo
r
k
.
RE
F
E
R
E
NC
E
S
[1
]
Al
-
Dh
a
m
a
ri
A.
A.,
S
u
d
irma
n
R.
,
M
a
h
m
o
o
d
N.
H.
,
“
Ab
n
o
rm
a
l
Be
h
a
v
io
r
De
tec
ti
o
n
in
Au
t
o
m
a
ted
S
u
r
v
e
il
lan
c
e
Vid
e
o
s:
A Rev
iew
,
”
J
o
u
rn
a
l
o
f
T
h
e
o
re
ti
c
a
l
a
n
d
Ap
p
li
e
d
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
v
o
l.
9
5
,
n
o
.
1
9
,
p
p
.
5
2
4
5
-
52
63
,
2
0
1
7
.
[2
]
S
a
h
a
sri
M
.
,
C
.
G
iree
sh
,
“
Ob
jec
t
M
o
ti
o
n
De
tec
ti
o
n
a
n
d
Trac
k
i
n
g
f
o
r
Vid
e
o
S
u
rv
e
il
la
n
c
e
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
rin
g
T
re
n
d
s
a
n
d
T
e
c
h
n
o
l
o
g
y
,
p
p
.
1
6
1
-
1
6
4
,
2
0
1
7
.
[3
]
Ad
h
v
a
r
y
u
A.
,
Ka
lp
e
sh
J.
,
“
Re
a
l
Ti
m
e
Un
u
su
a
l
Ev
e
n
t
De
tec
ti
o
n
in
Vid
e
o
S
e
q
u
e
n
c
e
s,
”
I
n
ter
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
Res
e
a
rc
h
in
Co
mp
u
ter
a
n
d
Co
mm
u
n
ica
t
io
n
En
g
i
n
e
e
rin
g
,
v
o
l
.
4
,
n
o
.
3
,
p
p
.
5
6
5
-
5
7
2
,
2
0
1
5
.
[4
]
Al
-
Na
wa
sh
i
M
.
,
Al
-
Ha
z
a
ime
h
O.
M
.
,
S
a
ra
e
e
M
.
,
“
A
n
o
v
e
l
fra
m
e
wo
rk
f
o
r
i
n
telli
g
e
n
t
s
u
rv
e
il
la
n
c
e
sy
ste
m
b
a
se
d
o
n
a
b
n
o
rm
a
l
h
u
m
a
n
a
c
ti
v
it
y
d
e
tec
ti
o
n
in
a
c
a
d
e
m
ic
e
n
v
iro
n
m
e
n
ts,
”
Ne
u
ra
l
Co
mp
u
ti
n
g
a
n
d
Ap
p
li
c
a
ti
o
n
s,
v
o
l
.
2
8
,
p
p
.
5
6
5
-
5
7
2
,
2
0
1
7
.
[
5
]
Z
a
i
d
i
S
.
,
e
t
a
l
.
,
“
V
i
d
e
o
A
n
o
m
a
l
y
D
e
t
e
c
t
i
o
n
a
n
d
C
l
a
s
s
i
f
i
c
a
t
i
o
n
f
o
r
H
u
m
a
n
A
c
t
i
v
i
t
y
R
e
c
o
g
n
i
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
u
r
r
e
n
t
T
r
e
n
d
s
i
n
C
o
m
p
u
t
e
r
,
E
l
e
c
t
r
i
c
a
l
,
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
(
C
T
C
E
E
C
)
,
p
p
.
5
4
4
-
548,
2017.
[6
]
G
o
g
a
wa
le
R.
C.
,
Ka
k
a
d
e
K
.
A.
,
Ka
le
P
.
N.
,
Ya
d
a
v
A.
S
.
,
Ha
n
wa
te
P
.
S
.
,
“
De
tec
ti
n
g
Ab
n
o
rm
a
l
Ac
ti
v
it
ies
fr
o
m
I
n
p
u
t
Vid
e
o
s
a
n
d
Re
p
o
r
ti
n
g
to
Au
t
h
o
r
it
i
e
s
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
Res
e
a
rc
h
a
n
d
I
n
n
o
v
a
ti
v
e
I
d
e
a
s
i
n
Ed
u
c
a
ti
o
n
,
v
o
l.
5
,
n
o
.
1
,
p
p
.
3
0
1
-
3
0
7
,
2
0
1
9
.
[7
]
Ch
e
n
Y.,
Ch
e
n
A.,
Jia
n
g
Z.
,
Z
h
o
n
g
J.
,
“
S
u
r
v
e
il
lan
c
e
Vid
e
o
P
o
rtrait
Re
c
o
g
n
it
i
o
n
P
re
p
ro
c
e
ss
in
g
Tec
h
n
o
lo
g
y
f
o
r
P
o
li
c
e
Ac
tu
a
l
Co
m
b
a
t,
”
Op
e
n
J
o
u
rn
a
l
o
f
Ap
p
li
e
d
S
c
ien
c
e
s
,
v
o
l.
0
9
,
n
o
.
0
5
,
p
p
.
3
9
4
-
4
0
2
,
2
0
1
9
.
[8
]
S
tu
d
e
n
t
P
.
G
.
,
En
g
i
n
e
e
rin
g
C.
,
Co
ll
e
g
e
B.
V.
M
.
E
.
,
a
n
d
Na
g
a
r
V.
V.
,
“
A
S
u
rv
e
y
o
n
Ob
jec
t
De
tec
ti
o
n
a
n
d
Trac
k
in
g
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
En
g
i
n
e
e
rin
g
a
n
d
Res
e
a
rc
h
De
v
e
l
o
p
me
n
t
,
v
o
l.
3
,
n
o
.
0
1
,
p
p
.
2
9
7
0
-
2
9
7
8
,
2
0
1
8
.
[9
]
Al
a
z
a
wi
S
.
A.,
S
h
a
ti
N.
M
.
,
Ab
b
a
s
A.
H.,
“
Tex
tu
re
fe
a
tu
re
s
e
x
trac
ti
o
n
b
a
se
d
o
n
G
LCM
fo
r
fa
c
e
re
tri
e
v
a
l
sy
ste
m
,
”
Per
io
d
ica
ls
o
f
En
g
in
e
e
rin
g
a
n
d
N
a
tu
ra
l
S
c
ien
c
e
s
,
v
o
l
.
7
,
n
o
.
3
,
p
p
.
1
4
5
9
-
1
4
6
7
,
2
0
1
9
.
[1
0
]
M
a
rien
a
A.
A.
&
S
a
th
ias
e
e
lan
J.
G
.
R.
,
“
Co
n
tras
t
e
n
h
a
n
c
e
m
e
n
t
o
f
g
ra
y
sc
a
le
a
n
d
c
o
lo
r
ima
g
e
s
u
sin
g
a
d
a
p
ti
v
e
tec
h
n
iq
u
e
s,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
E
n
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
v
o
l
.
7
,
n
o
.
2
,
p
p
.
1
-
4,
2
0
1
8
.
[1
1
]
P
é
re
z
-
Be
n
it
o
C.
,
M
o
ril
las
S
.
,
Jo
r
d
á
n
C.
,
C
o
n
e
jero
J.
A.
,
“
S
m
o
o
th
i
n
g
v
s.
sh
a
rp
e
n
i
n
g
o
f
c
o
l
o
u
r
ima
g
e
s:
T
o
g
e
th
e
r
o
r
se
p
a
ra
ted
,
”
Ap
p
l
ied
M
a
t
h
e
ma
ti
c
s
a
n
d
No
n
li
n
e
a
r S
c
ien
c
e
s
,
v
o
l.
2
,
n
o
.
1
,
p
p
.
2
9
9
-
3
1
6
,
2
0
1
7
.
[1
2
]
S
h
a
ti
N.
M
.
,
Ala
z
a
wi
S
.
A.,
Ab
d
u
lb
a
q
i
H.
A.
,
“
Ba
c
k
g
ro
u
n
d
S
u
b
trac
ti
o
n
(B
S
)
u
si
n
g
I
n
sta
n
t
P
ix
e
l
Hist
o
g
ra
m
,
”
J
o
u
rn
a
l
o
f
S
o
u
th
we
st Jia
o
to
n
g
Un
ive
rs
it
y
,
vol
.
5
4
,
no
.
5
,
p
p
.
1
-
6
,
2
0
1
9
.
[1
3
]
G
o
y
a
l
K.
,
S
i
n
g
h
a
i
J.
,
“
Re
v
iew
o
f
b
a
c
k
g
ro
u
n
d
su
b
trac
ti
o
n
m
e
th
o
d
s
u
sin
g
G
a
u
ss
ian
m
ix
tu
re
m
o
d
e
l
fo
r
v
i
d
e
o
su
rv
e
il
lan
c
e
sy
ste
m
s,
”
Arti
fi
c
i
a
l
I
n
telli
g
e
n
c
e
Rev
iew
,
v
o
l
.
5
0
,
n
o
.
2
,
p
p
.
2
4
1
-
2
5
9
,
2
0
1
8
.
[1
4
]
M
a
n
ik
a
n
d
a
n
S
.
S
.
S
.,
“
Ou
t
d
o
o
r
S
c
e
n
e
Im
a
g
e
S
e
g
m
e
n
tatio
n
Ba
se
d
o
n
Ke
rn
e
li
z
e
d
F
u
z
z
y
C
M
e
a
n
s
Clu
ste
rin
g
a
n
d
P
o
m
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
of
C
u
rr
e
n
t
En
g
i
n
e
e
rin
g
a
n
d
S
c
ien
ti
fi
c
Res
e
a
rc
h
,
v
o
l
.
6
,
no
.
3
,
p
p
.
2
0
-
26
,
2
0
1
9
.
[1
5
]
Alsh
ib
a
n
i
D.
R.
,
S
h
a
ti
N.
M
.
,
&
Ah
m
e
d
N.
T.
,
“
DN
A G
e
n
e
ti
c
Re
c
o
m
b
in
a
ti
o
n
b
a
se
d
Im
a
g
e
E
n
c
ry
p
ti
o
n
u
sin
g
Ch
a
o
ti
c
M
a
p
s,
”
In
d
ia
n
J
o
u
rn
a
l
o
f
Pu
b
li
c
He
a
lt
h
Res
e
a
rc
h
&
De
v
e
lo
p
me
n
t
,
vol
.
1
0
,
no
.
6
,
2
0
1
9
.
[1
6
]
Aw
a
d
A.
I.
,
Ha
ss
a
b
a
ll
a
h
M
.
,
“
Im
a
g
e
F
e
a
tu
re
De
tec
to
rs
a
n
d
De
sc
rip
to
rs:
F
o
u
n
d
a
ti
o
n
s
a
n
d
Ap
p
li
c
a
ti
o
n
s,
”
S
tu
d
ies
i
n
Co
mp
u
t
a
ti
o
n
a
l
I
n
telli
g
e
n
c
e
,
2
0
2
0
.
[1
7
]
Do
n
g
C.
,
L
iu
J.,
Xu
F
.
,
Li
u
C.
,
“
S
h
ip
d
e
tec
ti
o
n
fr
o
m
o
p
ti
c
a
l
re
m
o
te
se
n
sin
g
ima
g
e
s
u
sin
g
m
u
lt
i
-
sc
a
le
a
n
a
ly
sis
a
n
d
fo
u
rier HOG
d
e
sc
rip
to
r,
”
Rem
o
te
S
e
n
s
in
g
,
v
o
l
.
1
1
,
n
o
.
1
3
,
p
p
.
1
-
1
9
,
2
0
1
9
.
[1
8
]
Wen
L.
,
e
t
a
l.
,
“
M
u
lt
i
-
Ob
jec
t
D
e
tec
ti
o
n
a
n
d
Trac
k
i
n
g
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
r
n
a
l
f
o
r
Res
e
a
rc
h
in
Ap
p
li
e
d
S
c
ien
c
e
En
g
i
n
e
e
rin
g
&
T
e
c
h
n
o
lo
g
y
,
v
o
l.
6
,
p
p
.
8
0
9
-
8
1
3
,
2
0
1
8
.
[1
9
]
Ti
ru
p
a
t
h
a
m
m
a
M
.
,
“
Ob
jec
t
De
tec
ti
o
n
a
n
d
Trac
k
in
g
i
n
v
id
e
o
u
si
n
g
Ka
lma
n
F
il
ter,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Rec
e
n
t
Ad
v
a
n
c
e
s i
n
En
g
i
n
e
e
rin
g
&
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
5
,
n
o
.
5
,
p
p
.
1
0
2
-
1
0
6
,
2
0
1
6
.
[2
0
]
Ku
m
a
r
A.
,
Ku
sh
wa
h
a
S
.
,
“
A F
ra
m
e
wo
rk
fo
r
Hu
m
a
n
Ac
ti
v
it
y
Re
c
o
g
n
it
io
n
u
si
n
g
P
o
se
F
e
a
tu
re
fo
r
Vid
e
o
S
u
rv
e
il
lan
c
e
S
y
ste
m
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Co
mp
u
ter
Ap
p
li
c
a
t
io
n
s
,
v
o
l.
1
,
p
p
.
1
-
4
,
2
0
1
6
.
[2
1
]
Ho
ta
S
.
,
P
a
t
h
a
k
S
.
,
“
KNN
c
las
sifier
-
b
a
se
d
a
p
p
r
o
a
c
h
f
o
r
m
u
lt
i
-
c
las
s
se
n
ti
m
e
n
t
a
n
a
ly
sis
o
f
twit
ter
d
a
ta
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
E
n
g
i
n
e
e
rin
g
&
T
e
c
h
n
o
lo
g
y
,
v
o
l.
7
,
n
o
.
3
,
p
p
.
1
3
7
2
-
1
3
7
5
,
2
0
1
8
.
[2
2
]
Isa
N.
E.
Z.
M
d
,
Am
ir
A.,
Ily
a
s
M
.
Z.
,
Ra
z
a
ll
i
M
.
S
.
,
“
Th
e
P
e
rfo
r
m
a
n
c
e
An
a
ly
sis
o
f
K
-
Ne
a
r
e
st
Ne
ig
h
b
o
rs
(K
-
NN
)
Alg
o
rit
h
m
fo
r
M
o
t
o
r
Im
a
g
e
ry
Clas
sifica
ti
o
n
Ba
se
d
o
n
E
EG
S
ig
n
a
l,
”
M
a
tec
W
e
b
Co
n
f
.
,
v
o
l.
1
4
0
,
2
0
1
7
.
[2
3
]
N.
D.
Zak
i
,
N.
Y.
Ha
sh
im,
Y.
M
.
M
o
h
ial
d
e
n
,
M
.
A.
M
o
h
a
m
m
e
d
,
T.
S
u
ti
k
n
o
,
a
n
d
A.
H.
Ali,
"
A
re
a
l
-
ti
m
e
b
ig
d
a
ta
se
n
ti
m
e
n
t
a
n
a
ly
sis f
o
r
iraq
i
twe
e
ts u
sin
g
sp
a
rk
stre
a
m
in
g
,
"
Bu
l
letin
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
In
f
o
rm
a
ti
c
s
,
v
o
l.
9
,
n
o
.
6
,
p
p
.
1
4
1
1
-
1
4
1
9
,
2
0
2
0
.
[2
4
]
O.
A.
Ha
m
m
o
o
d
,
M
.
N.
M
.
Ka
h
a
r,
W.
A.
Ha
m
m
o
o
d
,
R.
A.
Ha
sa
n
,
M
.
A.
M
o
h
a
m
m
e
d
,
A.
A.
Y
o
o
b
,
e
t
a
l
.
,
"
An
e
ffe
c
ti
v
e
tran
sm
it
p
a
c
k
e
t
c
o
d
in
g
wit
h
tr
u
st
-
b
a
se
d
re
lay
n
o
d
e
s
i
n
VA
NETs,
"
B
u
ll
e
ti
n
o
f
El
e
c
trica
l
E
n
g
i
n
e
e
rin
g
a
n
d
In
f
o
rm
a
ti
c
s
,
v
o
l.
9
,
n
o
.
2
,
p
p
.
6
8
5
-
6
9
7
,
2
0
2
0
.
[2
5
]
A.
F
.
M
.
Ra
ffe
i,
T.
S
u
ti
k
n
o
,
H.
As
m
u
n
i,
R.
Ha
ss
a
n
,
R.
M
.
Oth
m
a
n
,
S
.
Ka
sim
,
e
t
a
l.
,
"
F
u
si
o
n
Iris
a
n
d
P
e
rio
c
u
lar
Re
c
o
g
n
it
i
o
n
s
in
N
o
n
-
C
o
o
p
e
ra
ti
v
e
En
v
ir
o
n
m
e
n
t
,
"
I
n
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
I
n
fo
rm
a
ti
c
s
,
v
o
l.
7
,
n
o
.
3
,
p
p
.
5
4
3
-
5
5
4
,
2
0
1
9
.
[2
6
]
Z.
F
.
H
u
ss
a
in
,
H.
R.
I
b
ra
h
e
e
m
,
M
.
Alsa
jri
,
A.
Hu
ss
e
in
Ali,
M
.
A.
Ism
a
il
,
S
.
Ka
sim
,
e
t
a
l.
,
"
A
n
e
w
m
o
d
e
l
fo
r
iri
s
d
a
ta
se
t
c
las
sifica
ti
o
n
b
a
se
d
o
n
li
n
e
a
r
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
p
a
ra
m
e
ter'
s
o
p
ti
m
iza
ti
o
n
,
"
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
&
Co
mp
u
ter
E
n
g
i
n
e
e
r
in
g
,
v
o
l.
1
0
,
n
o
.
1
,
p
p
.
1
0
7
9
-
10
8
4
,
2
0
2
0
.
[2
7
]
T.
Zh
a
n
g
,
Z.
Ya
n
g
,
W
.
Jia
,
B.
Ya
n
g
,
J.
Ya
n
g
,
a
n
d
X.
He
,
"
A
n
e
w
m
e
th
o
d
f
o
r
v
io
len
c
e
d
e
tec
ti
o
n
i
n
su
rv
e
il
lan
c
e
sc
e
n
e
s,"
M
u
lt
ime
d
ia
T
o
o
ls a
n
d
Ap
p
li
c
a
ti
o
n
s,
v
o
l.
7
5
,
n
o
.
1
2
,
p
p
.
7
3
2
7
-
7
3
4
9
,
2
0
1
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.