Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
9
, No
.
3
,
J
un
e
201
9
, pp.
1814
~
18
21
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
9
i
3
.
pp
1814
-
18
21
1814
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Develop
ment of
a portab
le comm
un
ity v
id
eo surv
eillance s
ystem
S.
F
ak
h
ar
A
.
G
,
A.
Fa
uz
an
K
,
M.
Saad
H, R. A
ff
e
ndi H
,
K.
H. Fe
n
Facul
t
y
of
El
e
ct
r
ic
a
l
&
E
le
c
tronic
Eng
ine
e
ring
T
ec
hnolog
y
,
Univ
ersit
i
Te
kn
ika
l
Malay
s
ia Mel
ak
a
(UTe
M)
,
Ma
lay
sia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
A
ug
11
, 201
8
Re
vised N
ov 20, 2
01
8
Accepte
d Dec
11, 201
8
In
2016,
a
c
rime
ra
te
h
as
be
en
evi
d
ent
l
y
inc
r
e
asing
par
ti
cu
la
rl
y
in
Kual
a
Lumpur
areas,
i
ncl
uding
rep
orts
on
house
br
ea
k
-
ins,
c
ar
the
fts
,
m
otorcy
cle
the
fts
and
robbe
r
y
.
One
w
a
y
of
det
err
ing
such
c
ase
s
is
b
y
insta
l
li
ng
CCT
V
m
onit
oring
s
y
st
em
in
pre
m
ises
such
as
hous
es
or
shops
,
bu
t
t
his
usua
l
l
y
req
uire
s
expe
nsi
ve
equi
pm
en
t
a
nd
insta
l
la
t
ion
f
ee
s.
In
thi
s
pa
p
er
a
cheape
r
al
t
ern
ative
of
a
porta
bl
e
comm
unity
vid
eo
surve
il
la
n
ce
s
y
st
em
running
on
Raspber
r
y
Pi
3
u
ti
lizing
OpenCV
is
pre
sen
te
d
.
Th
e
s
y
st
em
will
de
t
ec
t
m
oti
on
base
d
on
image
su
btra
ction
a
lg
orit
hm
and
imm
edi
a
te
l
y
inform
users
when
int
rude
rs
are
detec
t
ed
b
y
s
endi
ng
a
li
ve
vid
eo
f
eed
to
a
Te
l
egr
am
group
chat
,
as
well
as
sound
the
buzz
er
al
a
rm
on
the
Rasp
ber
r
y
Pi
.
Additionally
,
a
n
y
Te
l
egr
am
group
m
embers
ca
n
r
eque
st
images
a
nd
rec
or
d
ed
vid
eos
from
the
s
y
stem
at
an
y
t
ime
b
y
send
ing
a
get
r
eque
st
i
n
Te
l
egr
am
which
will
b
e
handl
ed
b
y
T
ele
gra
m
Bot.
Thi
s
s
y
stem
uses
th
e
Pi
NoIR
ca
m
er
a
m
odule
as
the
image
ac
qu
isit
ion
device
equipped
with
a
36
LE
D i
nfra
red
il
lu
m
ina
tor
for
night
vision
c
a
pab
il
i
t
y
.
In
add
it
ion
to
th
e
P
y
t
hon
la
nguage,
OpenCV,
a
computer
vision
sim
ula
ti
on
fro
m
Inte
l
is
al
so
used
for
image
proc
essing
ta
sks
.
Th
e
p
erf
o
rm
anc
e
ana
l
y
sis
of
th
e
completed
s
y
st
em
is
al
so
pre
sent
ed
computat
ion
al
complexi
t
y
wh
il
e
offe
ring
i
m
prove
d
fle
xi
bil
ity
.
The
per
form
anc
e
t
ime
is
al
so
pre
sent
ed,
where
the
w
hole
proc
ess
is
run
with
a
noti
c
ea
bl
e
3
sec
onds de
lay
in
getting
th
e
f
inal
out
put
.
Ke
yw
or
d
s
:
Im
age p
r
ocessi
ng
Op
e
nCV
Ra
sp
be
rr
y
pi
Tel
egr
am
b
ot
Vide
o
s
urveil
la
nce
Copyright
©
201
9
Instit
ut
e
o
f
Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Sh
am
su
l Fak
ha
r
A
bd
Gan
i
,
Faculty
of
Ele
c
tric
al
&
Ele
ct
r
on
ic
En
gin
ee
ri
ng Tec
hnology
,
Un
i
ver
sit
i Te
knikal M
al
ay
sia
Mel
aka (UTe
M),
Hang T
ua
h
Jay
a, 76
100 D
ur
ia
n
T
unggal
, Me
la
ka,
Ma
la
ysi
a.
Em
a
il
: sha
m
su
lfak
har@utem
.ed
u.m
y
1.
INTROD
U
CTION
Roya
l
Ma
la
ysia
Po
li
ce
(PD
RM
)
sug
geste
d
that
i
ns
ta
ll
at
ion
of
surv
ei
ll
ance
syst
em
would
dete
r
po
te
ntial
bur
glaries
as
well
as
facil
it
at
e
po
li
ce
inv
est
i
gation
in
ob
ta
ini
ng
e
vid
e
nce
a
nd
fa
ci
al
identific
at
ion
of
su
s
pects
in
th
e
even
t
that
it
did
happe
n.
Howev
e
r,
ba
sic
su
r
veill
ance
syst
e
m
is
s
ti
ll
pr
ic
ed
rel
at
ively
exp
e
ns
i
ve
in
Ma
la
ysi
a,
app
r
ox
im
at
el
y
R
M
1400
f
or
a
ba
sic
m
on
it
or
i
ng
syst
e
m
and
a
t
le
ast
RM
18
00
wi
t
h
add
e
d
li
ve
m
ob
il
e
vie
wing.
Th
us
,
a
l
ow
-
cost
vid
e
o
s
urveil
la
nce
sy
stem
is
pr
opose
d
to
i
ncr
eas
e
the
afforda
bili
ty
a
nd
wi
den
the
us
er
ba
se
for
su
c
h
syst
em
.
This
pro
j
ect
is
to
st
ud
y
a
nd
i
m
ple
m
ent
a
low
-
c
os
t
vid
e
o
s
urveil
la
nce
syst
em
based
on
Ra
s
pb
e
rry
Pi
3
by
usi
ng
i
m
age
-
base
d
m
ot
ion
de
te
ct
ion
al
gorithm
a
s
well
as Tele
gram
as the
m
ob
il
e ap
plica
ti
on
.
This
proj
ect
use
s
Ra
spber
ry
Pi
3
Mo
del
B,
release
d
i
n
2016
a
nd
costi
ng
a
m
ere
RM
170
for
the
sing
le
boar
d
c
om
pu
te
r.
As
de
picte
d
in
Tab
le
1,
t
he
proc
essor
has
ha
d
m
ajo
r
up
gr
a
de
fr
om
the
pr
evio
us
ver
si
o
n,
f
ro
m
32
-
bit
ARMv8
to
64
-
bit
Qu
a
d
Co
re
Br
oad
c
om
,
eff
ect
ively
increasin
g
the
process
or
s
pe
ed
f
ro
m
900MHz
on
th
e
Pi
2
to u
p
t
o
1.2GHz o
n
t
he
Pi
3.
I
n
a
ddit
ion
to
that,
it
is
a
lso
now
e
quip
pe
d
with buil
t
-
in W
i
-
Fi
chip
a
nd
Bl
ueto
oth
L
ow
E
nergy
(BLE
)
c
hip
,
w
h
ic
h
faci
li
ta
te
s
ou
r
sur
ve
il
la
nce
syst
e
m
to
connect
to
W
i
-
Fi
without
need
i
ng a
ny addit
io
na
l don
gle [1]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Develo
pm
e
nt
of
a
porta
ble comm
unit
y video
su
rvei
ll
an
ce
sy
ste
m
(
Shams
ul
Fak
ha
r
A
bd Gan
i
)
1815
Table
1.
Ra
s
pb
err
y Pi
3
Sp
eci
ficat
ion
Detai
ls
sp
ecif
icatio
n
d
etails
So
C
Bro
ad
co
m
BCM2
8
3
7
CPU
4
× ARM
Co
rtex
-
A
5
3
,
1
.2GHz
GPU
Bro
ad
co
m
Video
Co
re
IV
RAM
1
GB LPDDR2
(90
0
M
Hz)
Netwo
rkin
g
1
0
/1
0
0
E
th
ernet, 2.
4
GHz 80
2
.11
n
wir
eless
Blu
eto
o
th
Blu
eto
o
th
4.1
Clas
sic, Blu
eto
o
th
L
o
w E
n
ergy
Sto
rage
Extern
al
m
icroSD
GPIO
40
-
p
in
head
er,
po
p
u
lated
Po
rts
HDM
I,
3.5
m
m
an
a
lo
g
u
e aud
io
-
v
id
e
o
jack, 4×
USB 2
.0, E
th
ernet,
Ca
m
e
ra
S
erial
I
n
t
erface
(CSI)
,
Disp
la
y
Se
r
ial I
n
terface
(
DS
I)
An
8
m
egap
i
xe
ls
Ra
sp
be
rr
y
Pi
cam
era
is
attached
t
o
the
on
-
boar
d
Ra
s
pberr
y
Pi
cam
era
co
nn
ect
or,
and
this
cre
at
es
an
i
m
age
captur
e
syst
em
wit
h
em
bed
ded
c
om
pu
ti
ng
that
can
extract
in
for
m
at
ion
fr
om
i
m
ages
without t
he ne
ed
f
or
a
n
exte
r
nal p
r
ocessi
ng u
nit. Co
ns
i
der
i
ng
t
he
re
qu
i
re
m
ents o
f
im
age p
r
ocessin
g
c
om
par
ed
to
the
Ra
spbe
rr
y
Pi’s
proce
ssing
m
od
ule
and
it
s
pe
rip
he
rals,
it
is
decided
that
the
s
yst
e
m
is
capab
le
on
execu
ti
ng t
he
t
asks
s
pecifie
d.
Fo
r
t
he
rem
ote
interface,
Te
le
gr
am
Bot
will
be
us
ed
a
s
it
enab
le
s
m
ac
hin
e
a
nd
us
e
r
to
interface
seam
le
ssly
.
The
Tel
egr
am
ap
p
ca
n
be
in
sta
ll
ed
on
inte
nded
us
er
’s
m
obil
e
phon
e
,
the
n
it
is
rea
dy
t
o
go
.
Tel
egr
am
Bot
has
a
rtific
ia
l
intel
li
gen
ce
feat
ur
es
li
ke
re
plyi
ng
a
m
essage
with
im
ages
via
Im
age
Bot.
T
he
bo
t
is
al
so
desig
ne
d
f
or
pr
ogram
m
ing
pur
po
se
s
and
ca
n
com
m
un
ic
at
e
with
oth
e
r
Bots.
Ac
cordin
g
to
Tel
egr
am
,
the
m
essages
and
requests
se
nt
to
Bots
are
connecte
d
with
an
interm
ediary
serv
e
r
a
nd
a
re
li
nk
e
d
via
H
T
TPS
interface
whic
h
c
om
es w
it
h Tel
egr
am
A
PI
[2
]
.
Op
e
nCV,
a
n
open
sour
ce
co
m
pu
te
r
visio
n
li
br
ary de
velo
pe
d
by
vis
ual
in
te
racti
on
g
r
oup
from
In
te
l
is
us
e
d
as
di
gital
i
m
age
pr
oce
ssing
li
brary.
It
al
lows
Ra
s
pb
e
rr
y
Pi
to
ha
nd
le
im
age
pr
oces
sin
g
al
go
rithm
s
direc
tl
y from
i
ns
ide
the
Pi en
vir
on
m
ent [
3].
Earli
er
w
ork
by
Patoli
ya
,
M
ehta
an
d
Pate
l
propose
a
syst
e
m
con
sist
ing
of
night
visi
on
wireles
s
ca
m
era to
trans
m
it
v
ideos
is p
rop
os
ed [
4]. I
t
us
es
Ardu
i
no a
s the contr
oller
w
hic
h
is co
nn
ect
ed
to se
rvo m
oto
r
for
cam
era
mo
vem
ent
and
Bl
uetoo
t
h
m
o
du
le
f
or
tra
nsm
itti
ng
i
m
ages
to
Android
phones.
H
owe
ver,
the
Bl
uetoo
t
h
sig
na
l
is
fo
und
to
be
not
sta
ble
wh
e
n
com
par
e
d
to
W
i
-
Fi.
D
ong,
Yang,
W
ang
a
nd
X
u
pr
opos
e
d
ano
t
her
sur
veill
ance
syst
em
fo
r
el
der
ly
w
it
h
fall
an
d
i
nacti
vity
dete
ct
ion. Th
e p
r
oject
u
s
es
Ra
spber
ry
P
i
as
the
con
t
ro
ll
er
util
i
zi
ng
Gau
s
sia
n
Mi
xtu
re
Mo
de
l
fo
r
backg
round
s
ub
t
racti
on
m
et
ho
d
w
hich
will
sen
d
em
ai
l
or
SMS
to
i
nfo
r
m
fa
m
il
y
m
e
m
ber
s
shou
l
d
an
i
ncide
nt
occurs
[
5].
R
ashid
,
Ab
ir
,
S
houro
ve
a
nd
Mun
ta
ha
pro
po
se
d
a
ho
m
e
secur
it
y
syst
e
m
wh
ic
h
ide
ntify
visit
ors
by
i
m
age
proces
sing.
T
he
pro
j
e
ct
us
es
Ra
s
pbe
rr
y
Pi
as
the
con
t
ro
ll
er
with
O
pe
nC
V
li
br
ary
f
or
ba
ckgr
ound
sub
tract
ion
m
et
ho
d
an
d
al
so
P
oc
ketsp
hinx,
a
s
peech
recog
niti
on
s
of
tware t
hat runs
n
at
ively
on
Raspber
ry Pi fo
r
vo
ic
e
rec
ogniti
on [6].
2.
PROP
OSE
D DESIG
N
2
.
1.
Bl
ock
di
ag
r
am
The bloc
k diag
ram
in
Figu
r
e
1
s
hows
t
he ove
rall
syst
e
m
co
m
po
nen
ts i
nc
lu
ding the Ras
pb
err
y Pi
3
as
the
central
plat
form
fo
r
the
syst
e
m
.
An
8
m
e
gap
i
xels
night
visio
n
pi
cam
e
ra
with
in
fr
a
re
d
il
lum
inators
act
as
the
im
age
acqu
isi
ti
on
dev
ic
e
to
the
syst
em
.
Wh
en
t
he
syst
e
m
detect
s
pos
sible
intr
us
io
n,
the
buzze
r
wil
l
e
m
i
t
al
ert
so
un
d
an
d
ad
diti
on
al
ly
an
al
ert
m
essa
ge
will
be
sent
fr
om
the
Ra
spber
ry
Pi
3
to
the
gro
up
m
e
mb
ers
i
n
Tel
egr
am
app
.
Additi
on
al
ly
,
a
ny
Tel
eg
ram
gr
ou
p
m
e
m
ber
s
can
r
eq
uest
im
ages
a
nd
rec
orded
vi
deos
f
rom
the
syst
e
m
at any tim
e b
y sen
ding
a g
et
re
qu
est
i
n
Tel
e
gr
am
w
hi
ch
will
b
e
handled by Tel
e
gra
m
Bot.
Fig
ure
1
.
Syst
em
b
lock
d
ia
gr
a
m
2
.
2
.
Motio
n
de
tect
ion
algorithm
An
sa
ri,
Sedky,
Sh
a
rm
a
and
Ty
agi
sug
gest
that
m
otion
detect
ion
al
go
rithm
can
be
set
to
detect
ph
ysi
cal
m
ov
em
ent
of
obj
ec
ts
in
a
giv
e
n
range
by
det
erm
ining
obj
e
ct
m
ov
em
ent
on
s
ucces
sive
an
d
pro
gr
essi
ve
i
m
ages.
T
he
al
gorithm
co
m
par
es
on
ly
the
gr
ey
fr
am
es
afte
r
c
onve
rting
c
olour
im
ages
fr
om
RGB
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
3
,
June
201
9
:
1814
-
182
1
1816
patte
rn
t
o
gr
ey
scal
e
[7
]
,
[8
]
.
J
yothi
an
d
Vard
han
sug
gest
th
at
m
otion
dete
ct
ion
al
gorith
m
app
li
es
pri
nc
iples
of
diff
e
re
ncin
g
f
r
a
m
e
wh
ere
t
he
al
gorithm
com
par
es
pix
el
s
on
how
t
hey
change
locat
i
on
after
each
fr
a
m
e
[9
]
.
A
s
tu
dy
cond
uc
te
d
by
Islam
,
Nah
iy
an
a
nd
K
iber
al
so
a
gr
ee
d
on
the
c
on
ce
pt
that
consi
de
rin
g
fr
am
es
of
vid
e
o
as
im
ages,
m
o
ti
on
ca
n
be
de
te
ct
ed
by
pe
rfor
m
ing
a
bs
ol
ut
e
i
m
age
dif
fe
ren
ce
[
10]
,
[
11]
.
T
her
e
is
a
m
or
e
reli
able
m
et
hod
base
d
on
ba
ckgr
ound
subtract
ion
a
nd
fr
a
m
e
diff
ere
nce,
a
fu
sio
n
to
m
or
e
pr
eci
sel
y
detect
m
ot
ion
.
Liu,
Fan
a
nd
Wa
ng
desig
ne
d
a
te
ch
nique
ba
sed
on
backg
rou
nd
subtract
ion
an
d
t
hr
ee
-
f
ram
e
diff
e
re
nce
to
de
te
ct
m
otion
s
unde
r
water
by prod
ucin
g
a
dif
f
eren
t
im
age
to
determ
ine
pix
e
ls
fr
om
non
-
m
ov
i
ng
o
bject
s
[
12]
.
T
he
f
unct
ion
of
three
f
ram
e
diff
ere
nce
is
le
s
s
sensiti
ve
to
backg
rou
nd
no
ise
,
an
d
the
a
ut
hors
pro
ved
t
hat
the
tw
o
c
om
bin
ed
m
et
ho
ds
work
well
in
their
e
xp
e
ri
m
ent.
This
proj
ect
will
al
so
a
pp
ly
backg
rou
nd subtracti
on
with
fr
am
e d
iffe
rence
for
t
he
m
otion detect
io
n.
The
aut
hors
a
gr
ee
d
that
se
pa
rati
on
process
betwee
n
f
or
e
gro
und
an
d
ba
ckgr
ound
is
a
key
facto
r
in
com
pu
te
r
visio
n
a
pp
li
cat
ions,
for
e
xam
ple
in
s
urveil
la
nce
us
e,
m
otion
c
a
pture,
act
io
n
r
ecognizi
ng
a
nd
vid
e
o
com
pr
essin
g.
Howe
ver,
the
acce
pta
ble
se
pa
rati
on
proces
s
faces
m
any
chall
enges.
Fa
ct
or
s
that
res
ul
te
d
in
fail
ur
e
in
m
oti
on
d
et
ect
ion
in
cl
ud
es
s
udde
n
changes
in
il
lu
m
inati
on
li
ke
cl
oud
m
ov
e
m
ent,
m
otion
change
li
ke
wav
i
ng
br
a
nc
he
s
of
trees
,
cam
era
sh
ake
,
and
al
s
o
bac
kgr
ound
ge
om
et
r
y
chan
ge
l
ike
sh
a
dows
an
d
parke
d
cars.
T
he
refor
e
,
im
age
pr
e
-
processin
g
is
c
r
uc
ia
l
to
reduce
error
i
n
m
otion
detect
ion
.
S
m
al
l
chan
ges
in
pix
e
l
will
easi
ly
le
a
d
to
fail
ur
e
in
m
ot
ion
detect
ion.
For
exam
ple,
no
ise
al
wa
ys
happen
in
t
he
bac
kgr
ound
and
it
changes
t
he
va
lue
of
pix
el
s.
Fu
rt
her
ex
plained
by
the
aut
hors,
noise
re
s
ults
in
diff
e
re
nc
e
in
pi
xels
with
the
correct
p
i
xel
va
lue in a
n
im
age.
Co
ns
ide
r
E
quat
ion
(
1) for
a
n
im
age f
with
scene s
and
noise
n
:
(
,
)
=
(
,
)
+
(
,
)
(1)
The
e
qu
at
io
n
e
xp
la
ine
d
that
a
ny
acq
uire
d
im
age
can
be
a
ff
ect
ed
by
sce
ne
an
d
no
ise
.
I
f
there
is
no
change
i
n
inte
ns
it
y,
no
cha
nge
is
obse
rv
e
d.
But
if
one
pixe
l
is
br
ig
hter
,
and
has
highe
r
pix
el
values
t
han
it
s
neig
hbours
,
th
is
will
resu
lt
in
noise
.
T
he
a
uthors
sta
te
d
that
im
age
pr
e
-
proces
si
ng
s
ho
uld
be
do
ne
be
fore
act
ual im
age p
ro
ces
sin
g
to
suppress
disto
rtio
ns
.
3.
IMPLEME
N
TATION
3
.
1.
Ha
rdw
are d
e
s
ign
The
ha
rdwar
e
will
be
r
un
on
a
Ra
s
pb
e
rr
y
Pi
3
as
t
he
ce
ntral
c
om
pu
te
r
with
Ra
spbia
n
Jes
sie
O
S
instal
le
d
an
d
a
dd
it
io
nal
per
ip
her
al
s
of
a
n
8
m
egap
ixels
ca
m
era
to
ca
ptur
e
i
m
age,
a
nd
a
buzzer
as
a
n
a
ud
i
ble
al
arm
.
The
syst
e
m
is
intended
to
be
porta
ble
so
it
is
po
wer
e
d
by
c
onne
ct
ing
it
to
a
m
ob
il
e
powe
r
bank.
Figure
2
sho
w
s
the
prot
otype
of
the
vid
e
o
s
urveil
la
nce
sys
tem
in
de
velo
pm
ent
m
od
e,
w
hile
Fi
gure
3
s
hows
the
prototyp
e
in
norm
al
ru
nni
ng
m
od
e, w
it
h
the
LCD
m
on
it
or,
keyb
oa
rd
a
nd
m
ou
se
rem
ov
e
d.
Fig
ur
e 4
sh
ows
the pr
oto
ty
pe r
unning i
n pit
ch
b
la
ck
d
a
r
kn
es
s w
it
h o
nly t
he
I
R LE
Ds
v
isi
bl
e.
Fig
ure
2.
Syst
em
in d
e
velo
pme
nt
m
od
e
Fig
ure
3.
Syst
em
in nor
m
al
m
od
e
Fig
ure
4.
Syst
em
in p
it
ch blac
k
3.2.
So
f
tware
desi
gn
In
a
ddit
ion
t
o
t
he
Pyt
ho
n
pro
gr
am
m
ing
la
nguag
e
that
is
na
ti
vely
us
ed
i
n
Ra
sp
be
rr
y
Pi
e
nv
i
ronm
ent,
this
syst
e
m
wil
l
al
so
us
e
O
pe
nCV
,
a
c
om
pu
te
r
visio
n
sim
ul
at
ion
f
ro
m
In
te
l
to
help
with
i
m
age
pr
e
-
pro
c
essin
g
that i
s d
esi
gne
d wit
h
c
rite
rias of r
e
source
op
tim
iz
at
ion
, low
pow
e
r
c
onsu
m
ption an
d
im
prov
e
d spee
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Develo
pm
e
nt
of
a
porta
ble comm
unit
y video
su
rvei
ll
an
ce
sy
ste
m
(
Shams
ul
Fak
ha
r
A
bd Gan
i
)
1817
4.
E
X
PERI
MEN
TAL
RES
UL
TS
A
ND
D
IS
CUSSIO
N
4
.
1.
Night
v
isi
on
o
ut
pu
t
Figure
5(
a
)
s
hows
the
im
age
captu
re
f
ro
m
the
syst
em
du
ring
day
ti
m
e.
T
he
col
our
of
t
he
i
m
age
can
be
see
n
to
be
sli
gh
tl
y
pink
i
n
col
or
beca
use
of
t
he
discar
ded
IR
filt
er
.
Figure
5
(
b)
shows
the
im
age
captu
re
durin
g
ni
gh
t
ti
m
e
witho
ut
ni
gh
t
visi
on
tu
rned
on
w
her
e
the
i
m
age
is
in
total
pitch
black
,
w
hile
F
igure
5
(c
)
sh
ows
the
im
a
ge
ca
pture
when
the
IR
il
lu
m
inator
is
tu
r
ne
d
on
w
he
re
w
e
can
see
s
ome
obj
ect
s
in
t
he
i
m
age
al
beit t
he dar
k l
igh
ti
ng.
(a)
(
b)
(c)
Fig
ure
5. Sam
ple captu
red im
age
Figure
6
(a
)
s
how
s
the
im
age
captu
re
of
f
r
om
the
syst
em
,
and
since
the
durin
g
t
he
next
few
f
ram
e
captu
res
there
are
no
cha
ng
es
in
the
vid
eo
,
the
i
m
age
diff
e
ren
ce
in
Fi
gur
e
6
(b)
an
d
Fig
ur
e
6
(c
)
resu
lt
s
in
a
black
im
age, w
hich
is
esse
ntial
ly
n
o
m
otion
de
te
ct
ed.
(a)
(b)
(c)
Fig
ure
6.
Sam
ple captu
red im
age
Figure
7
(a
)
s
how
s
w
hat
happens
w
he
n
a
n
obj
ect
m
ov
es
i
nto
t
he
picture
durin
g
dayt
im
e
wh
ic
h
is
or
i
gin
al
ly
as
il
lustrate
d
i
n
Fi
gure
6(
a
).
Fig
ur
e
7
(a)
is
the
‘
F
ram
e
Delt
a
’
w
hich
is
the
pr
e
-
th
res
ho
l
d
i
m
age,
wh
ic
h
in
Fig
ure
7
(
b)
is
s
how
n
hav
i
ng
grey
colo
r
f
ram
e
diff
ere
nces
.
Fig
ure
7(
c
)
is
t
he
‘
Thr
e
sh
’
wh
ic
h
is
the
thres
ho
l
ded im
age, sh
owin
g
wh
it
e
blob a
ga
inst
blac
k
backgro
und,
w
hich
is
cl
assifi
ed
as
m
ot
ion
detect
ed.
(a)
(b)
(c)
Fig
ure
7.
Sam
ple captu
red im
age
Figure
8
rec
rea
te
s
the
exp
eri
m
ent
do
ne
in
Figure
7,
bu
t
the
i
m
age
captu
re
is
done
duri
ng
night
tim
e
and
in
t
otal
da
rkness.
A
n
im
age
is
ca
pture
d
with
f
ram
e
diff
e
re
nce
tha
n
the
previ
ous
captu
re
as
show
n
i
n
Figure
8(
a
), p
r
e
-
thre
shold
as i
n
Fi
gure
8(b
),
and the
res
ult i
s a
detect
ed
m
otion o
f w
hite blo
b
in
Fig
ure
8(
c
).
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
3
,
June
201
9
:
1814
-
182
1
1818
(a)
(b)
(c)
Fig
ure
8.
Sam
ple captu
red im
age
4
.
2
.
Int
er
act
i
on
b
e
tween
h
ardwa
re and
s
oftwa
re
On
ce
m
otion
detect
s
possibl
e
intrude
r,
a
n
i
m
age
captu
r
e
is
autom
at
icall
y
sent
to
a
pr
e
de
fine
d
Tel
egr
am
gr
ou
p
chat
to
gethe
r
with
a
3
se
c
onds
vid
e
o
of
the
incide
nt.
Figure
9(a)
s
hows
t
he
gro
up
chat
con
ta
ini
ng
fa
m
ily
m
e
m
ber
s
as
well
as
th
e
Ra
spber
ry
P
i
as
a
us
e
r.
Fi
gure
9(b)
sho
ws
the
al
e
rt
m
essage
no
ti
ficat
io
n
ha
s
ar
rive
d
int
o
t
he
Tel
e
gr
am
gro
up,
a
nd
Fi
gure
9(
c
)
s
hows
the
re
port
fro
m
Ra
sp
ber
ry
P
i
with
captu
red
im
age
as
well
as
a
3
seco
nds
vid
e
o
that
trigg
e
rs
the
al
arm
wh
ic
h
can
be
viewe
d
an
d
dow
nlo
a
ded
by
anyo
ne
in
that
par
ti
cula
r
Tel
e
gr
am
g
r
oup.
Figure
9. Tel
eg
ram
n
otific
at
ion
from
Raspb
e
rr
y Pi
4
.
3
.
Rea
c
tio
n
t
im
e
f
r
om
Pi
to Te
le
gra
m
v
ers
us
d
ist
an
ce
Ba
sed
on
the
e
xp
e
rim
ent,
it
is
found
th
at
rea
ct
ion
tim
e
to
transm
it
picture
and
vid
e
o
fro
m
Ra
sp
berry
Pi
to
Tel
eg
ra
m
is
no
t
af
fec
te
d
by
distanc
e
as
detai
le
d
i
n
Ta
ble
2
(
da
y
tim
e)
and
T
able
3
(
night
tim
e)
.
The
reacti
on
ti
m
e
is
on
ly
aff
e
ct
ed
by
the
s
pe
ed
of
the
a
vaila
ble
wifi
c
onne
ct
ion
.
W
it
hin
the
set
distanc
e
of
8
m
et
ers,
the
m
otion
detect
io
n
syst
e
m
wo
rk
s
ver
y
well
duri
ng
day
tim
e.
Re
su
lt
s
in
Tab
le
3
sh
ows
tha
t
nig
ht
su
r
veill
ance
is
no
t
avail
able
f
or
dista
nces
ov
er
4
m
et
ers
du
e
to
the
infr
a
re
d
il
lu
m
inator
li
gh
ti
ng
w
hich
i
s
on
ly
eff
ect
ive
wit
hin
5
m
et
ers
ide
al
ly
.
Thu
s,
ob
je
ct
s
m
ov
ing
be
tween
4
m
et
e
rs
to
8
m
et
ers
durin
g
ni
gh
t
ti
m
e
can
be
see
n from
r
ecorde
d vide
o,
bu
t t
he
m
otion
will
n
ot
be det
ect
ed.
Table
2
.
Re
act
i
on
T
im
e V
ersus
D
ist
ance
(
D
a
y
T
i
m
e
)
Distan
ce (
m
)
Reactio
n
T
i
m
e
(
s
)
1
2
2
1
3
2
4
1
5
2
6
2
7
2
8
1
Table
3
.
Re
act
i
on
T
im
e V
ersus
D
ist
ance
(
N
i
gh
t
T
i
m
e
)
Distan
ce (
m
)
Reactio
n
T
i
m
e
(
s
)
1
2
2
2
3
1
4
2
5
-
6
-
7
-
8
-
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Develo
pm
e
nt
of
a
porta
ble comm
unit
y video
su
rvei
ll
an
ce
sy
ste
m
(
Shams
ul
Fak
ha
r
A
bd Gan
i
)
1819
Figure
10
il
lustrate
s
the
reacti
on
tim
e
in
se
conds
ve
rs
us
di
sta
nce
in
m
eter
s
for
day
ti
m
e
and
night
tim
e
analy
sis
of
the
syst
em
resp
ect
ively
.
It
is
obse
rv
e
d
t
hat
gi
ve
n
a
st
able
wi
fi
sig
na
l
is
avail
able
f
or
t
he
Ra
sp
be
rr
y
Pi
t
o
piggybac
k
on,
any
m
otion
detect
ion
al
ert
no
ti
ficat
i
on
wi
ll
be
delivered
to
Tel
egr
am
a
pp
in
us
er
’s
m
ob
il
e
phone
within
1 t
o 2 s
ec
onds.
Figure
10. Rea
ct
ion
ti
m
e v
ers
us
distance a
na
ly
sis
4
.
4
.
Accur
acy
t
es
t,
s
ensiti
vity
t
es
t
an
d
s
peci
fici
ty
t
es
t
Table
4
an
d
T
able
5
show
s
the
accu
racy
te
st
resu
lt
s
for
da
y
tim
e
and
nig
ht
tim
e
su
rv
ei
ll
ance
te
ste
d
directl
y
fo
r
10
consecuti
ve
da
ys.
Accord
i
ng
to
Table
4,
day
tim
e
su
rv
ei
ll
ance
sco
red
95.
42%,
w
hich
is
m
or
e
accurate
t
han
a
t
night
ti
m
e
wi
th
sc
or
e
of
89.
82%.
T
he
pro
je
ct
shows
hi
gh
accu
racy,
a
nd
the
syst
em
is
a
ble
t
o
diff
e
re
ntiat
e intru
sio
n
a
nd
nor
m
al
cases corr
ect
ly
.
Table
4.
T
est
for
D
ay
T
im
e
S
urveil
la
nce
Tr
ials
(Da
y
)
Tr
u
e
Po
sitiv
e
Tr
u
e
Neg
ativ
e
False
Po
sitiv
e
False
Neg
ativ
e
Accurac
y
Test
Sen
sitiv
ity
Test
Sp
ecif
icity
Test
1
st
5
4
1
0
90%
100%
80%
2
nd
5
5
1
1
8
3
.33
%
8
3
.33
%
8
3
.33
%
3
rd
5
4
1
0
90%
100%
80%
4
th
5
4
0
0
100%
100%
100%
5
th
5
5
1
0
9
0
.91
%
100%
8
3
.33
%
6
th
5
5
0
0
100%
100%
100%
7
th
5
5
0
0
100%
100%
100%
8
th
5
5
0
0
100%
100%
100%
9
th
5
5
0
0
100%
100%
100%
10
th
5
5
0
0
100%
100%
100%
1
st
4
4
2
1
7
2
.73
%
100%
6
6
.66
%
2
nd
4
4
1
1
80
%
80%
80
%
3
rd
4
5
2
0
8
1
.81
%
100%
7
1
.43
%
4
th
4
5
1
1
8
1
.81
%
8
0%
8
3
.33
%
5
th
5
5
1
1
9
0
.91
%
8
3
.33
%
8
3
.33
%
6
th
5
5
0
0
100%
100%
100%
7
th
5
5
0
0
100%
100%
100%
8
th
5
5
0
0
100%
100%
100%
9
th
5
5
1
0
9
0
.91
%
100%
8
3
.33
%
10
th
5
5
0
0
100%
100%
100%
Ba
sed
on
se
ns
i
ti
vity
te
st,
day
tim
e
su
rv
ei
ll
ance
sc
or
e
d
98.
33%,
w
hic
h
is
m
or
e
sensiti
ve
to
m
otion
than
in
night
t
i
m
e
with
92
.
33%.
F
or
s
pe
ci
f
ic
it
y
te
st,
day
tim
e
su
rv
ei
ll
an
ce
scor
e
d
92.67%,
w
hich
is
m
or
e
sp
eci
fic
t
han
in
night
ti
m
e
with
86.
81%.
T
he
syst
e
m
scor
e
s
hi
gh
pe
rce
ntage
a
nd
is
a
ble
to
determ
ine
norm
al
cases co
rr
ect
ly
.
Figure
11
s
umm
arizes
the
te
st
resu
lt
s
of
m
e
asur
i
ng
acc
ura
cy
,
se
ns
it
ivit
y,
and
s
pecifici
ty
du
ri
ng
day
tim
e,
wh
il
e
Figure
12
s
umm
arizes
the
sam
e
te
st
bu
t
do
ne
durin
g
night
tim
e.
It
is
fo
und
that
night
tim
e
su
r
veill
ance
pe
rfor
m
s
poorer
than
durin
g
da
y
tim
e
m
a
inly
beca
us
e
of
the
in
suffici
ent
night
visio
n
l
igh
t
ing
pro
vid
e
d
by
th
e
infr
a
re
d
il
lum
inator
caus
i
ng
the
f
ram
e
diff
ere
nce
to
be
too
sm
al
l
and
ne
glect
ed
in
gr
e
ysc
al
e
i
m
ages
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
3
,
June
201
9
:
1814
-
182
1
1820
Figure
11
.
Acc
ur
acy
,
s
en
sit
ivit
y,
s
pecifici
ty
ver
s
us
d
ay
s
(
d
ay
t
i
m
e)
Figure
12. Acc
ur
acy
,
sensi
ti
vity
, s
pecifici
ty
v
ers
us
days
(
night t
im
e)
5.
CONCL
US
I
O
N
The
de
vel
op
m
ent
of
a
porta
ble
com
m
un
ity
vid
eo
s
urve
il
la
nce
syst
e
m
us
in
g
Ra
spbe
rr
y
Pi
3
is
dem
on
strat
ed
i
n
this
a
rtic
le
,
im
ple
m
enting
Op
e
nCV
as
th
e
cor
e
im
age
processi
ng
s
of
tware.
At
the
correct
set
up
the
syst
e
m
per
form
s
wo
nde
rfull
y,
esp
eci
al
ly
s
ince
the
w
ho
le
syst
e
m
costs
within
a
bu
dget
of
RM
500
com
par
ed
to surv
ei
ll
ance
syst
e
m
s
so
ld
in
t
he
m
ark
et
w
hich
sel
ls
fo
r
ar
ound
RM
1800. The
no
ti
ficat
io
n
s
yst
e
m
al
so
em
plo
ys an
op
e
n so
ur
ce
chat ap
p, Tel
eg
ram
, w
hich
is free a
nd can
b
e
easi
ly
instal
le
d
by the
user.
ACKN
OWLE
DGE
MENTS
Th
e
aut
hors
w
ou
l
d
li
ke
to
th
ank
Un
i
ver
sit
i
Tekn
i
kal
Ma
la
ysi
a
Me
la
ka
(U
TeM
)
f
or
pr
ov
i
ding
the
su
pp
or
t
nee
de
d
to
com
plete
the
w
ork
he
rein
via
gra
nt
nu
m
ber
:
RAGS
/
1/2015/ICT
01/FTK/
03
/B
00
1
21
a
nd
PJP/2
016/P
KA/
FT
K
-
C
ACT/S
01512
.
REFERE
NCE
S
[
1
]
Hos
sain
N.,
Kab
ir
M.
T
.
,
R
ahman
T.
R
.
,
Hos
sen
M.
S.
and
Sal
a
uddin
F.,
"
A
Re
al
-
T
ime
Surveil
l
anc
e
Min
i
-
Rove
r
Based
on
Open
CV
-
Py
thon
-
JA
VA
Us
ing
Raspb
err
y
Pi
2,
"
Pro
ce
ed
ings
-
5th
I
EE
E
In
te
rnation
al
Confe
ren
ce
o
n
Control
Syst
em
,
Computing
and
Engi
ne
ering,
pp.
476
–
481,
Nov
2
016.
[
2
]
Job
J.,
Nar
esh
V.
and
Chandras
eka
ran
,
K
.
,
"
A
Modifie
d
Se
cur
e
Version
of
th
e
T
el
eg
ram
Protocol
(MTProto)
,
"
IEE
E
In
te
rnatio
nal
Confe
ren
ce
on
El
e
ct
ronics
,
Computing
and
Comm
unic
ati
on
Technol
ogi
es,
C
ONECCT
2015
,
pp.
1
–
6
,
2016
.
[
3
]
W
.
F.
Aba
y
a
,
J
.
Basa
M
.
S
y
,
A.
C.
Ab
ad
and
E
.
P.
Dadios
,
"
Lo
w
C
ost
S
m
art
S
ec
uri
t
y
C
amer
a
with
N
ight
V
isi
on
C
apa
bilit
y
U
sin
g
Raspber
r
y
P
i
and
OpenCV
,
"
Inte
rnat
ional
Confe
renc
e
o
n
Hum
anoid,
Nanote
chnol
og
y,
Information
Te
chnol
ogy
,
Com
mu
nic
ati
on
and
Control,
Env
i
ronm
ent
and
Manage
ment
(
HNICEM)
,
Palaw
an,
pp.
1
-
6
,
2014.
[
4
]
Patol
i
y
a
J,
M
ehta
H.
and
Pat
el
H
.
,
"
Arduino
Cont
roll
ed
W
ar
Fie
ld
Sp
y
Robot
Us
ing
Night
Vision
W
ire
le
ss
Camera
and
Android
Applicati
on
,
"
NUiCONE
2015
-
5th
Nirma
Unive
rs
it
y
I
nte
rnat
io
nal
Confe
renc
e
on
Engi
nee
rin
g
,
pp.
3
–
7
,
2016
.
[
5
]
Dong
Q,
Yang
Y,
W
ang
H
and
Xu
J.
H,
"
Fal
l
A
la
rm
and
Ina
c
ti
v
ity
De
te
c
ti
on
S
ystem
Design
and
Im
ple
m
ent
ation
on
Raspber
r
y
Pi
,
"
I
nte
rnational
Co
nfe
renc
e
on
Advance
d
Comm
uni
cat
ion
Technol
o
gy,
IC
ACT
,
pp
.
3
8
2
–
386,
2015
.
[
6
]
Rashid
M.
T,
Ab
ir
I.
K,
Shourove
N.
S,
Muntaha
R.
and
Rhaman
M.
K,
"
Inte
ll
igent
Intrusion
Preve
nti
on
S
y
st
em
fo
r
Hous
ehol
ds
Based
on
S
y
stem
-
On
-
Chip
Com
puter
,
"
Canadian
Confe
renc
e
on
Elec
tric
al
and
Comp
ute
r
Eng
ine
erin
g
,
Oct
2016.
[
7
]
A.
N.
Ans
ari
,
M.
Sedk
y
,
N.
S
har
m
a
and
A.
T
y
ag
i,
"A
n
Int
er
net
of
th
ings
ap
proa
ch
for
m
otion
det
e
ct
ion
usi
ng
Raspber
r
y
Pi,"
Proce
ed
ings
of
2015
Inte
rnatio
nal
Confe
ren
ce
on
Inte
l
li
g
ent
C
omputing
and
I
nte
rnet
o
f
Thing
s
,
Harbi
n,
pp
.
131
-
134
,
2015
.
[
8
]
Ham
za
h
R.
A,
Abd
Gani
S.
F,
Kadm
in
A.
F,
Ham
id
M.
S,
S
al
am
S.
and
T.
W
ook
T.
M.
F
,
"
Disparity
M
a
p
Esti
m
at
ion
Us
e
s
Bloc
k
Mat
ch
ing
Algorit
hm
and
Bil
atera
l
Filt
er
,
"
Int
ernational
Confe
ren
c
e
on
Informati
on
Technol
ogy
Syst
ems and
Innov
a
t
ion
(
ICITSI)
,
pp.
151
–
154,
2017.
[
9
]
S.
N.
J
y
oth
i
and
K
.
V.
Vardha
n
,
"
Design
and
I
m
ple
m
ent
at
ion
of
R
eal
T
ime
S
ec
uri
t
y
S
urve
illa
nce
S
y
st
em
U
sing
IoT,
"
In
te
rnatio
nal
Conf
ere
nce
on
Comm
unic
ation and
E
lectroni
cs
Syste
ms
(
ICCES)
,
Coim
bat
ore
,
pp
.
1
-
5
,
2016
.
[
1
0
]
M.
Z.
Islam,
K.
M.
T.
Nahi
y
a
n
and
M.
A.
Kibe
r,
"A
M
oti
on
D
e
tecti
on
A
lgorit
hm
for
V
ide
o
-
P
ol
y
som
nogra
ph
y
to
D
ia
gnose
S
le
ep
D
isorder
,
"
18th
Inte
rnational
Confe
renc
e
on
Computer
and
Information
Technol
ogy
(
ICCIT
)
,
Dhaka
,
pp.
272
-
275
,
2015
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Develo
pm
e
nt
of
a
porta
ble comm
unit
y video
su
rvei
ll
an
ce
sy
ste
m
(
Shams
ul
Fak
ha
r
A
bd Gan
i
)
1821
[
1
1
]
Ham
za
h
R.
A,
Ham
id
M.
S,
Kadm
in
A.
F.
and
Abd
Gani
S.
F,
"
Im
prove
m
ent
o
f
Stere
o
Corre
sponding
Algorit
h
m
Based
on
Su
m
o
f
Abs
olut
e
Diffe
ren
ce
s
and
Edge
Preserving
Filt
e
r,
"
IEE
E
Inte
rna
ti
onal
Confe
ren
c
e
on
Signal
and
Image
Proc
essing
Applications (
ICSIPA
)
,
pp.
222
–
225,
2017
.
[
1
2
]
W
ang
Z,
Li
u
S.
and
Fan
Q
.
,
"
Cloud
-
base
d
Plat
form
for
E
m
bedde
d
W
ire
l
ess
Video
Surveil
l
ance
S
y
s
te
m
,
"
Proce
ed
ings
-
In
te
rnational
Conf
ere
nce on
Comp
utat
ional and
In
f
orm
ati
on
Scienc
es
,
pp
.
1335
–
13
38,
2013
.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Sh
ams
ul
Fakh
ar
bin
Ab
d
Gani
gra
duated
from
Univer
siti
Malay
sia
Perli
s
(Uni
MA
P)
in
Bac
helor
of
Engi
ne
eri
ng
(
Com
pute
r
Engi
n
ee
ring)
wi
th
hon
ours
in
2006
and
la
t
er
re
ceive
d
h
i
s
Master
'
s
d
egr
e
e
in
Int
ern
et
&
W
eb
Com
puti
ng
in
2015
from
Ro
y
al
Melbour
ne
Insti
tut
e
of
Te
chno
log
y
(R
MIT)
Aus
tra
li
a
.
He
st
art
ed
his
ca
r
ee
r
as
an
R
&D
e
lec
troni
c
engi
n
ee
r
spec
ializing
in
s
oftwa
re
d
esign
f
or
m
et
er
c
luste
r
d
eve
lopment
in
Siemens
VD
O
Autom
oti
ve
Pe
nang
(l
at
e
r
kno
wn
as
Conti
n
en
ta
l
Autom
oti
ve
Mal
a
y
si
a).
Befor
e
l
e
avi
ng
Cont
ine
nt
al
,
Sham
sul
plays
an
a
ct
iv
e
rol
e
i
n
Mitsubishi
Fus
o
proje
c
ts
as
a
sof
twar
e
develope
r
and
al
so
softw
are
project
m
ana
ger
.
Sham
sul
is
now
a
le
ct
ure
r
in
el
e
ct
roni
c and c
om
pute
r
engi
n
eering
t
ec
hnolog
y
depa
rtment
of
F
TK
EE
UT
eM.
Ah
mad
Fau
z
a
n
bin
Kadmin
C
Eng.
P.T
e
ch
c
ur
ren
tly
atta
che
d
with
UTe
M
as
a
rese
arc
her
,
Ahm
ad
Fauza
n
bin
Kad
m
in
CEng.
P.T
ec
h.
h
as
over
1
4
y
e
ars
of
experie
nc
e
in
e
le
c
tr
onic
&
comput
er
engi
ne
eri
ng
fiel
d
with
te
chnic
al
expe
r
t
ise
in
R&D
engi
neering,
computer
vision
&
m
e
dic
a
l
el
e
ct
roni
cs.
He
g
rad
uated
with
a
Bac
h
el
or
Degr
ee
i
n
E
le
c
tronic
Engi
nee
r
ing
fro
m
Univer
siti
Sain
s
Malay
s
ia
(US
M)
and
Master
D
egr
ee
in
Com
pute
r
&
Com
m
unic
ation
Engi
n
ee
ri
ng
from
Univer
siti
Keba
ngsaa
n
Ma
l
a
y
si
a
(UK
M)
.
Pr
evi
ousl
y
,
h
e
wor
ked
with
Meg
ast
ee
l
Sdn
.
Bhd.
,
S
ams
ung
SD
I
(M
)
Sdn.
Bhd.
and
Agensi
Angkasa
Nega
ra
.
He
publ
ished
seve
r
al
t
echnic
a
l
and
engi
n
ee
ring
p
ape
rwor
k
s
in
image
proc
essing
and
m
edi
c
al
el
e
ct
roni
cs.
Mohd
Saad
b
or
n
in
1981,
gra
du
at
ed
from
Multi
m
edi
a
Univer
sit
y
(MM
U)
where
he
rec
e
ive
d
his
B.
Eng
m
aj
oring
in
Com
pute
r
in
2003.
H
e
the
n
continues
working
with
Te
l
ekom
Malay
si
a
Berha
d
an
d
la
t
er
joi
n
ed
Si
e
m
ens
VDO
Auto
m
oti
ve
Penang
(
la
t
er
known a
s Conti
nental
Auto
m
oti
ve
Malay
sia
)
as
a
software
eng
i
nee
r.
He
d
evel
oped
firmw
are
in
m
et
er
c
luste
r
developm
ent
f
or
both
loc
a
l
a
n
d
int
ern
at
ion
al
cus
tomer
in
the
area
of
Control
le
r
Area
Network
a
nd
ECU
dia
gnostic
during
his
tenure
in
the
compan
y
.
In
2014
he
rec
ei
ved
his
M.
En
g
m
aj
oring
in
Com
pute
r
and
Com
m
unic
at
ion
from
Univer
siti
Keba
ngsaa
n
Mal
a
y
s
ia
(UK
M)
.
Curre
n
tly
h
e
is
a
le
c
tur
er
in
Univer
sit
i
Te
knik
al
Ma
lay
s
ia
Mela
ka
teac
h
ing
progra
m
m
ing,
e
m
bedd
ed
s
y
st
em a
nd
m
i
cro
cont
r
oll
er
subjects.
Ros
tam
Affen
di
Ham
z
ah
gra
du
at
ed
from
Univer
siti
Te
knolog
i
Malay
s
ia
(UTM
)
where
he
rec
eived
his
B.
Eng
m
aj
o
ring
in
Elec
tronic
Engi
n
ee
ring
in
2000.
In
2010
he
recei
ved
h
is
M.
Sc.
m
aj
oring
in
El
e
ct
roni
c
S
y
s
tem
Des
ign
Engi
n
ee
ring
from
Univer
siti
Sa
ins
Ma
lay
s
ia
(US
M)
.
I
n
2017,
he
re
ce
i
ved
PhD
m
aj
oring
in
El
ec
t
ronic
Im
agi
ng
from
Uni
ver
siti
Sains
Malay
si
a
(US
M)
.
Curre
ntly
he
is
a
le
c
ture
r
in
Uni
ver
siti
Te
knik
al
Malay
sia
Melaka
teac
h
ing
dig
it
al
elec
troni
cs
and
digital
ima
ge
proc
essing.
Kong
Hui
Fe
n
is
an
under
gra
duat
e
studen
t
from
Univer
siti
Te
kn
ologi
Malay
si
a
Mela
ka
in
Ba
ch
el
or
of
Engi
ne
eri
ng
Te
chno
log
y
(Co
m
pute
r
Engi
n
eering).
She
is
cu
rre
ntly
f
ini
shing
her
l
ast
sem
est
er
int
ern
ship
progr
amm
e
in
Int
el
Pr
ogra
m
m
abl
e
Sol
uti
on
Group Pen
ang.
Evaluation Warning : The document was created with Spire.PDF for Python.