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.
8
, No
.
6
,
Decem
ber
201
8
, p
p.
4204
~
4211
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
8
i
6
.
pp
4204
-
42
11
4204
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
A Surve
y on Mul
timedia
Content
Protecti
on Mech
anisms
Gottum
ukk
al
a
Him
a
Bi
nd
u
1
,
Ch
in
t
a
An
u
radha
2
,
P
atn
al
a
S
.
R
.
Chan
dra Mur
t
y
3
1
,3
Depa
rtment
of
Com
pute
r
Sci
en
ce
&
Engi
ne
eri
n
g,
Univer
si
t
y
Co
ll
eg
e
of
Engi
n
eering
&
T
ec
hnolo
g
y
,
Ach
ar
y
a
Naga
rjuna Univ
ersity
,
Indi
a
2
Depa
rtment of
Com
pute
r
Scie
n
ce
& Engi
ne
eri
n
g,
V.
R.
Siddhartha
Eng
ine
e
ring Colle
ge
,
India
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Dec
20
, 201
7
Re
vised
Jun
4
,
201
8
Accepte
d
J
ul
29
, 2
01
8
Cloud
computin
g
has
emerge
d
t
o
infl
uen
ce
m
ultim
edi
a
cont
en
t
p
rovide
rs
like
Disne
y
to
r
ender
the
ir
m
ult
imed
ia
servi
ce
s.
W
he
n
cont
ent
prov
id
ers
use
the
publi
c
c
loud
,
there
are
cha
n
ce
s to
have
pir
at
ed co
pie
s furt
her
le
ad
i
ng
to
a
loss
in
rev
enue
s.
At
t
he
sam
e
t
ime,
technologica
l
advance
m
ent
s
reg
ard
ing
conten
t
rec
ording
and
h
osting
m
ade
it
e
as
y
to
duplicate
genui
ne
m
ultim
edi
a
obj
ec
ts
.
Thi
s
proble
m
has
inc
rea
s
ed
wit
h
inc
rea
sed
usa
ge
of
a
cl
oud
p
la
tform
for
ren
der
ing
m
ult
i
m
edi
a
con
te
nt
t
o
users
acros
s
the
g
lobe
.
Th
er
efo
re
it
is
essenti
a
l
to
ha
ve
m
ec
hani
sm
s
to
det
ect
video
cop
y
,
discov
er
cop
y
right
infri
ngement
of
m
ult
imedia
co
nte
nt
and
prot
e
ct
the
in
te
rests
of
genui
ne
cont
en
t
provide
r
s.
I
t
is
a
ch
allen
ging
and
comput
at
ion
al
l
y
expe
ns
ive
probl
em
to
be
addr
essed
c
onsideri
ng
the
expon
ential
g
ro
wth
of
m
ult
ime
dia
con
te
nt
over
the
i
nt
ern
e
t
.
In
thi
s
pape
r,
we
surve
y
ed
m
ult
imedia
-
c
ont
ent
prote
ction
m
ec
hani
sm
s
which
throw
li
ght
o
n
diffe
r
ent
kinds
of
m
ult
imedi
a,
m
ult
imedia
cont
en
t
m
odifi
cation
m
et
hods
,
an
d
te
chn
ique
s
to
prote
c
t
intellect
u
al
prop
e
r
t
y
from
abuse
an
d
cop
y
righ
t
inf
ringe
m
ent
.
It
a
lso
foc
uses
on
challe
ng
es
invol
ved
in
prot
ec
t
ing
m
ult
imedia
content
and
t
he
rese
ar
ch
gaps
in
the
area
of
cl
oud
-
b
ase
d
m
ult
imedia
content
pro
te
c
ti
on
.
Ke
yw
or
d:
C
loud
-
base
d
m
ultim
edia
con
te
nt
Mult
i
m
edia
M
ultim
edia con
te
nt
protect
io
n
Vi
de
o
c
op
y
det
ect
ion
Copyright
©
201
8
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
:
Go
tt
um
ukkala
Him
a Bi
nd
u
,
Dep
a
rtm
ent
of
Com
pu
te
r
Scie
nce & E
ng
i
nee
rin
g,
Un
i
ver
sit
y C
ol
le
ge
of E
ngine
erin
g
&
Tec
hnology,
Ach
a
rya
Nag
a
r
j
una
U
niv
e
rsity
, Gun
t
ur
, An
dhra
Pr
a
des
h, I
ndia
.
Em
a
il
: gh
i
m
abindu19
@g
m
ai
l
.co
m
1.
INTROD
U
CTION
Waterm
ark
in
g
te
chn
iq
ues
ha
ve
bee
n
a
rou
nd
f
or
c
on
te
nt
protect
ion
or
protect
ion
of
intel
le
ct
ua
l
pro
per
ty
in
the
real
world.
W
at
erm
ark
ing
is
a
process
of
in
serti
ng
a
disti
nc
t
patte
rn
into
the
co
ntent
of
vid
e
o
wh
ic
h
is
la
te
r
us
e
d
for
c
opy
detect
ion
.
T
he
diff
e
r
e
nt
a
sp
ect
s
of
wate
rm
ark
ing
a
nd
how
it
is
us
ef
ul
f
or
intel
le
ct
ual
prop
e
rty
pr
otect
ion
on
the
inte
rn
et
is
ex
plain
ed
i
n
[1
]
a
nd
dig
it
al
wate
rm
ark
i
ng
sc
hem
e
s
f
or
m
ul
tim
edia
con
te
nt
pr
otect
ion
us
i
ng
dif
fer
e
nt
ap
proac
hes
su
c
h
as
asy
m
m
et
ric
fing
er
pri
nting
protoc
ol
s,
zer
o
-
knowle
dge
protoc
ols,
com
m
itm
ent
sche
m
e
s,
and
ho
m
omor
phic
encr
ypt
ion
in
[
2
]
.
In
t
eresti
ng
ly
m
a
t
chin
g
te
chn
iq
ues
co
m
ple
m
ent
waterm
ark
ing
te
ch
niques.
T
he
m
at
ching
te
ch
ni
qu
e
s
incl
ud
e
m
otion
directi
on,
m
otion
m
at
ching
,
or
din
al
intensit
y
sign
at
ure,
a
nd
colo
r
histo
gr
a
m
sign
at
ur
e
a
re
ex
plained
i
n
[3
]
.
Bl
oc
k
ci
ph
e
r
al
gorithm
us
ed
in
[4
]
f
or
m
ultim
edia
con
te
nt
pr
otect
io
n.
V
ideo
fi
ng
e
r
pr
in
ti
ng
is
al
so
us
e
d
to
ide
ntify
vid
eo
s
un
i
qu
el
y.
Vi
de
o
fin
gerpr
i
nt
i
s
a
vecto
r
wh
i
ch
ca
n
c
ha
racteri
ze
an
d
uniq
uely
ide
ntify
a
vide
o
f
ro
m
ano
t
her
vid
e
o
[5
]
.
F
ull
-
le
ngth
vid
e
o
f
ing
e
rprintin
g
[
6
]
an
d
detect
io
n
of
onli
ne
ab
us
e
of
im
ages
[
7
]
a
re
tw
o
im
portant
ty
pes
of
resear
ch
that
play
a
vital
ro
le
in
pr
otect
ing
intel
le
ct
ual
pro
per
ti
es.
These
t
wo
i
ncide
ntall
y
equ
ip
pe
d
with
Un
it
ed
S
t
at
e Patents.
Con
ce
rn
i
ng
c
onte
nt
-
base
d
co
py
detect
io
n
of
m
ultim
edia
obj
ect
s,
in
de
xing
of
ref
e
ren
ce
sign
at
ur
es
of
vid
e
os
or
fin
ge
rprints
of
vide
os
play
s
a
vital
ro
le
.
The
re
cent
tren
ds
in
interact
ive
m
u
lt
i
m
edia
com
p
utin
g
include
m
ultim
edia
co
ntent
searc
hin
g
,
in
dex
i
ng,
visu
al
iz
at
ion
,
i
ntell
i
gen
t
inf
orm
ati
on
ext
racti
on,
dig
it
al
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
A Survey
on M
ulti
med
ia
Co
nt
ent Pr
otect
ion
Mech
an
is
ms
(
Go
tt
um
ukka
l
a Him
a
Bi
ndu)
4205
m
anag
em
ent,
m
ul
tim
edia
com
m
un
ic
at
ion
s,
dig
it
al
sig
nal
processi
ng,
im
age,
a
udio
an
d
vid
e
o
proces
sing,
a
nd
m
ul
tim
edia
con
te
nt
protect
io
n
[
8
]
.
M
ultim
edia
co
ntent
s
har
e
d
i
n
onli
ne
so
ci
al
n
et
w
orks
(OSNs
)
i
s
al
so
grow
i
ng
ra
pi
dly.
Prote
ct
ing
s
uch
co
nte
nt
ha
s
issues
a
nd
c
ounterm
easur
e
s
as
e
xplore
d
in
[
9
]
.
Di
gital
r
igh
ts
m
anag
em
ent
is
an
oth
e
r
im
po
r
ta
nt
issue
a
bo
ut
m
ultim
edia
con
te
nt
pr
otect
ion
.
T
he
im
ple
m
entat
ion
of
novel
DRM
te
chn
i
ques
ba
sed
on
m
ob
il
e
andr
oid
te
rm
inal
pro
po
se
d
in
[
1
0
]
a
nd
us
a
ge
of
bu
ye
r
-
fr
i
end
ly
waterm
ark
ing
protoc
ols in
[1
1
]
h
a
ve bee
n p
rop
os
ed
to su
pport
the
pro
te
ct
ion
of
c
op
yri
ghte
d
dig
it
al
c
onte
nts.
The
rem
ai
nd
er
of
the
pa
per
is
struct
ur
e
d
i
nt
o
di
ff
e
ren
t
se
ct
ion
s
t
hat
pro
vid
e
i
ns
ig
hts
on
var
i
ou
s
te
ch
niques
us
e
d
f
or m
ultim
edia con
te
nt
protect
ion.
Table
1
.
Acro
nym
s
Acron
y
m
Descripti
o
n
HDFS
Had
o
o
p
Distribu
ted
File
S
y
ste
m
CBCD
Co
n
ten
t
-
Bas
ed
Copy
Dete
ctio
n
MM
MV
Mean o
f
the Magn
itu
d
es o
f
Motio
n
V
ecto
rs
MPM
V
Mean o
f
the Ph
ase
An
g
les o
f
M
o
tio
n
Vectors
DRM
Dig
ital Rig
h
ts Ma
n
ag
e
m
en
t
LSH
Locality
Sens
itiv
e
Hash
in
g
SIFT
Scale I
n
v
ari
an
t Fe
atu
re
T
rans
f
o
r
m
OSN
On
lin
e Social Net
wo
rk
M2M
Mob
ile 2 Mob
ile
2.
VID
E
O FI
NGE
RPR
I
NTI
N
G FO
R CO
N
TE
NT BASE
D
VID
E
O I
D
ENTIFIC
ATI
ON
Lee
and
Y
oo
(
2008)
[5
]
us
e
d
the
con
ce
pt
of
vid
e
o
fi
ng
e
r
pri
ntin
g.
T
hey
pro
posed
a
m
ec
han
ism
fo
r
c
on
te
nt
-
based
vid
e
o
identific
at
ion
us
i
ng
the
fing
e
rprintin
g
con
ce
pt.
The
ov
e
r
view
of
th
e
syst
e
m
is
pr
esente
d
in
F
ig
ure 1
.
It h
as tw
o
im
po
rt
ant phases
kn
own
as
fin
gerpr
i
nt ex
t
racti
on a
nd f
in
ge
rprint
m
at
ching
. T
he
form
er
is
us
ed
t
o
obta
in
a
fin
gerpr
i
nt
fr
om
giv
en
m
ul
tim
edia
ob
j
ect
w
hile
the
la
tt
er
is
us
ed
to
com
par
e
tw
o
vid
e
os
us
in
g
t
heir
c
orr
esp
onding
fin
ge
rprints.
.
Fig
ure
1
.
O
verview
of
vid
e
o
f
ing
e
rprintin
g m
et
ho
d
The
procedu
re
us
e
d
f
or
fin
gerpr
i
nt
extracti
on
is
il
lustrate
d
in
Fig
ur
e
2.
Fi
r
st
of
al
l,
the
giv
en
vid
e
o
is
div
ide
d
i
nto
r
esam
pled
fr
a
m
es
and
the
n
co
nv
e
rted
to
the
gr
ey
scal
e
fr
am
es.
It
is
done
as
t
he
gr
ay
scal
e
i
m
pr
oves
the
rob
us
tness
of
fin
gerpr
i
nt
ext
racti
on.
T
he
r
esi
zed
f
ram
es
are
the
n
par
t
it
ion
ed
i
nto
m
ulti
ple
blo
c
ks
.
A
fter
w
ard
,
for
each
bl
ock
,
the
ce
ntr
oid
of
gradie
nt
or
ie
ntati
ons
is
com
pu
te
d.
Th
en
fin
gerp
rint
vect
or
is
ob
ta
ine
d
w
hich
c
on
ta
in
s
com
p
act
featur
es
of
the
vide
o
cl
ip
w
hich
is
us
ed
to
i
de
ntify
vid
e
o
uniq
uely
.
Fing
e
r
pr
int
m
a
tc
hin
g
is
a
n
im
po
rtant
ph
a
s
e
in
the
pro
posed
syst
em
w
hich
is
res
ponsi
ble
for
extra
ct
ing
a
fin
gerpr
i
nt
fro
m
qu
ery
vid
e
o
and
m
at
ches
it
with
that
of
a
vid
e
o
in
th
e
databas
e
.
T
he
ir
em
pirical
resu
lt
s
rev
eal
e
d
that t
he fin
gerpr
i
nt
m
at
ching
was a
ble to o
utp
e
rfor
m
o
ther
f
eat
ur
es
co
nce
rn
i
ng
vid
e
o fin
gerp
rintin
g.
Lu
(
2009)
[
1
2
]
al
so
ex
plored
vid
e
o
fi
ng
e
rpri
nting
f
or
vid
e
o
cop
y
detect
ion.
The
Me
tric
s
us
e
d
f
or
the
m
echan
ism
include
m
at
ching
eff
ic
ie
ncy,
lo
w
com
plexity
,
com
pact,
discri
m
inati
on
,
a
nd
rob
us
tness
.
Dif
fer
e
nt
al
gorithm
s
are
ex
plo
re
d
nam
el
y
sp
at
ia
l
sign
at
ures,
te
m
poral
sig
natu
res,
colo
r
sig
natu
re
s,
tra
ns
f
or
m
-
dom
ai
n
sign
at
ur
es
, a
nd f
in
gerp
rint m
a
tc
hin
g.
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4204
-
4211
4206
Figure
2
.
Illust
rates the
proce
dure
of f
i
ng
e
rprint e
xtracti
on
3.
EFFE
CTIVE
AND SC
ALA
BL
E VID
EO
COP
Y DET
ECTIO
N
Liu
et
al
.
(2
010)
[
13
]
pro
pose
d
an
al
gori
thm
kn
ow
n
as
con
te
nt
-
base
d
cop
y
detect
io
n
(CBC
D)
al
gorithm
as
s
how
n
in
Fig
ure
2.
The
qu
e
r
y
exa
m
ples
use
d
in
this
syst
e
m
include
ei
ther
par
t
of
refe
ren
ce
vid
e
os
directl
y
or
par
t
of
ref
e
ren
ce
vid
e
os
e
m
bed
ded
int
o
oth
e
r
vi
deo
s
.
As
s
how
n
in
F
igure
3,
ther
e
are
tw
o
input
query
vid
eo
cl
i
ps.
T
he
y
are
us
e
d
to
te
st
the
cop
y
detect
ion
m
echan
ism
e
m
plo
ye
d
by
us
i
ng
CB
CD
al
gorithm
.
The
ref
ere
nce
vid
e
o
is
flipp
e
d
in
qu
e
ry
exam
ple
2.
Exam
ple
1
con
ta
in
s
ref
e
re
nce
vi
deo
em
bed
de
d
in
s
om
e reg
ion.
The
al
gorithm
pr
e
sente
d
in
Figure
4
perform
s
var
io
us
st
eps
to
ha
ve
co
ntent
-
base
d
c
opy
detect
io
n.
Wh
e
n
qu
e
ry
vi
deo
is
gi
ven
as
input,
co
nt
ent
-
based
sam
pling
is
perfor
m
ed
first.
The
n
the
que
ry
vid
eo
is
su
bject
e
d
to
tr
ansfo
rm
ation
detect
ion
a
nd
norm
al
iz
a
ti
on
.
Finall
y,
SI
FT
extracti
on
is
carrie
d
out,
a
nd
LS
H
com
pu
ta
ti
on
is
done
.
T
hese
ste
ps
a
re
al
s
o
ca
rr
ie
d
out
with
a
ref
e
re
nce
vide
o
to
wh
ic
h
th
e
query
vid
e
o
need
t
o
be
com
par
ed
.
Additi
on
al
ly
,
the
LS
H
in
dex
i
ng
is
gen
e
rated
for
ref
e
re
nce
vid
e
o
an
d
sa
ve
d
it
to
the
data
base
f
or
reu
se
.
A
fter
L
SH
c
om
p
utati
on
,
the
query
vid
eo
is
s
ubj
ect
e
d
to
keyfr
am
e
le
vel
qu
e
ry,
ke
yfram
e
le
vel
qu
ery
ref
inem
ent;
keyfr
am
e
le
vel
resu
lt
m
erg
es
,
vid
e
o
le
vel
re
su
lt
fu
sio
n,
vid
eo
sc
or
e
nor
m
al
iz
ation
an
d
finall
y
CB
CD r
es
ults
are
gen
e
rated
.
On
ce
the
resu
l
ts
are
ge
ne
rated
,
t
hey
are
us
e
d
to
m
ake
wel
l
-
inf
or
m
ed
dec
isi
on
s.
The
det
ect
ion
rat
e
and
acc
ur
acy
of
the
CB
CD
al
gorithm
sh
owe
d
good
perf
orm
ance.
T
he
al
gorithm
is
scal
able
as
well
.
T
asdem
ir
and
Ce
ti
n
(
2010)
[
14
]
us
e
d
m
otion
vecto
r
base
d
featu
r
es
fo
r
vid
e
o
cop
y
detect
io
n.
They
wer
e
use
d
f
or
reli
able v
e
rific
at
ion
of sig
nat
ur
es
ab
out
m
ul
tim
edia con
te
nt
as p
a
rt of CB
CD.
Figure
3
.
Sam
ple q
ue
ry
vid
e
os (a)
Que
ry e
xa
m
ple
1(
b) Que
ry exam
ple 2
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
A Survey
on M
ulti
med
ia
Co
nt
ent Pr
otect
ion
Mech
an
is
ms
(
Go
tt
um
ukka
l
a Him
a
Bi
ndu)
4207
Figure
4
.
O
verview
of CB
CD
algorit
hm
4.
DETE
CTING
3
-
D VI
DEO
COPIE
S
Kho
dab
a
khs
hi
and
Hef
ee
da
(
2013)
[15]
pro
po
s
ed
a
novel
con
te
nt
-
based
cop
y
detect
ion,
especial
ly
for
3
-
D
vid
e
os.
First
of
al
l
th
e
syst
e
m
gen
er
at
es
visu
al
sig
na
tures
for
the
gi
ven
3
-
D
vid
e
os.
The
se
sig
nat
ur
es
are
m
ai
ntained
in
a
database
.
They
a
re
kn
own
as
re
fer
e
nc
e
sign
at
ur
es
.
T
h
e
qu
e
ry
vi
de
o
is
the
n
c
ompare
d
against
the
in
de
xed
database
ref
e
ren
ces
for
cop
y
detect
io
n.
The
syst
em
i
s
prov
e
d
to
be
com
pu
ta
ti
on
al
ly
and
stora
ge
-
wise
ef
fici
ent.
Th
ey
nam
ed
their
pro
po
s
ed
syst
em
as
Sp
id
er.
The
high
-
le
vel
ove
rv
ie
w
of
vi
deo
copy
detect
ion sy
ste
m
is show
n
in
Figure
5.
Figure
5
.
3D
vi
deo co
py
detec
ti
on
syst
em
The
pe
ople
or
organ
iz
at
io
ns
who
ow
n
m
u
lt
i
m
edia
con
te
nt
are
kn
own
as
con
te
nt
ow
ner
s
.
Vide
o
ho
sti
ng
sit
es
a
re
w
ebsite
s
w
her
e
vi
deo
s
ar
e
hoste
d
.
F
or
instance
,
YouT
ub
e
is
on
e
of
the
hosti
ng
we
bs
it
es
.
Vide
o
c
op
y
de
te
ct
ion
is
the
proces
s
of
c
om
par
in
g
ori
gi
nal
vid
e
o
a
nd
pirat
ed
c
op
y
a
nd
de
te
ct
ing
a
c
op
y
of
t
he
vid
e
o
.
T
he
m
e
thod
use
s
data
set
s
pro
vid
e
d
by
Mo
bile3DT
V,
Mi
cr
osoft,
and
Y
ouTu
be.
The
qu
e
ry
vi
de
os
ar
e
furth
e
r
di
vid
e
d
int
o
thr
ee
c
at
egories.
Ty
pe
1
qu
e
ry
vi
de
os
or
near
duplica
te
or
part
of
ref
e
ren
ce
vid
e
os
.
Ty
pe 2
cat
eg
or
y are
par
t
of r
e
fer
e
nce
vid
e
os
e
m
bed
de
d
i
nto othe
r vide
os
.
The
ty
pe
3
vid
eo
s
are
the
vi
deo
s
that
c
on
t
ai
n
no
par
ts
of
or
igi
nal
ref
e
r
ence
vi
d
eos
.
P
recisi
on
a
nd
recall
are
us
e
d
to
evaluate
the
syst
e
m
.
Ye
et
al
.
(2
016)
[
16
]
f
ocused
on
m
ob
il
e
to
m
ob
il
e
(M2M)
com
m
un
ic
at
ion
f
or
sec
ur
e
m
ul
tim
edia
con
te
nt
distrib
ution
.
Th
ey
pr
opos
e
d
a
fr
a
m
ewo
r
k
for
con
te
nt
distrib
ution
w
hich
is
sho
wn
in
Fig
ur
e
6.
T
he
c
on
te
nt
pr
ot
ect
ion
m
echan
ism
is
bu
il
t
into
a
sp
eci
al
m
achine.
The
sp
eci
al
de
vice
al
so
pro
vid
es
an
in
dex
of
con
te
nt
to
sup
port
faster
sear
ch
an
d
identifi
cat
ion
of
vid
e
os.
Th
e
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4204
-
4211
4208
M2M
net
wor
k
can
ha
ve
c
on
te
nt
distri
bu
ti
on
ca
pa
bili
ti
es
i
n
a
secu
re
e
nv
iro
nm
ent.
De
li
ver
y
of
m
ultim
edia
con
te
nt ove
r
th
e
I
ntern
et
is
explo
red
in
[17].
Figure
6
.
O
verview
of sec
ur
e
m
ul
tim
edia con
te
nt
distrib
ution f
ram
ewo
r
k
5.
CLOUD
-
BAS
ED
MU
LT
I
M
EDIA C
ONT
ENT PR
OTE
CTIO
N
S
YS
T
EM
Hef
ee
da
et
al
.
(20
15)
[
18]
pr
opos
e
d
a
syst
e
m
fo
r
la
rg
e
-
sc
al
e
m
ultim
edia
co
ntent
protect
ion
.
It
wa
s
bu
il
t
for
protec
ti
ng
dif
fer
e
nt
ki
nd
s
of
m
edia
su
c
h
as
m
us
ic
cl
ips,
songs,
a
ud
i
o
cl
ips,
im
a
ges,
2
-
D
vi
deos
an
d
3
-
D
vi
deos.
T
heir
syst
em
c
an
be
dep
l
oyed
ei
ther
in
public
or
pr
ivate
cl
oud.
T
hey
pro
po
se
d
tw
o
novel
com
po
ne
nts
suc
h
as
a
m
e
thod
f
or
sig
natu
re
gen
e
rati
on
for
m
ult
i
m
edia
c
on
te
nt
an
d
a
di
stribu
te
d
m
at
c
hing
eng
i
ne
that
is used t
o protec
t
m
ul
tim
edia o
bject
s.
The
syst
em
o
ver
vie
w
is
pr
ese
nted
in
Figure
8.
Figure
8
.
Cl
oud
-
base
d
m
ultim
edia con
te
nt
protect
ion sy
stem
Wh
en
co
nte
nt
owne
rs
su
c
h
a
s
Pixa
r
or
Dis
ney
hosts
ne
w
m
ul
tim
edia
con
te
nt
ov
e
r
t
he
internet
,
th
e
con
te
nt
ref
e
re
nce
reg
ist
rati
on
is
m
ade
by
gen
erati
ng
si
gn
at
ur
es
a
nd
storing
them
i
n
a
distrib
uted
ind
e
x
whe
re
by
i
m
pl
e
m
enting
ob
j
e
ct
m
at
ching
and
qu
e
ry
proc
essing.
Wh
e
n
pirated
c
op
ie
s
are
fou
nd
over
th
e
i
ntern
et
,
the
qu
ery
sig
natu
res
are
gen
e
rated
a
nd
m
at
ched
wi
th
the
sig
natu
r
es
sto
red
in
the
distri
bu
te
d
in
de
x
to
detect
vi
olati
on
s.
W
ei
et
al
. (201
4)
[
19
]
propo
s
e
d
a
sc
hem
e
f
or
secu
rity
and
pri
vacy o
f
both
c
om
pu
ta
ti
ons
a
nd
stora
ge
in
the
c
loud
.
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
A Survey
on M
ulti
med
ia
Co
nt
ent Pr
otect
ion
Mech
an
is
ms
(
Go
tt
um
ukka
l
a Him
a
Bi
ndu)
4209
5.1.
S
ign
at
u
re
C
re
at
i
on
The
syst
em
su
pport
s
dif
fer
e
nt
kin
ds
of
m
e
dia
f
or
a
si
gnat
ur
e
gen
e
rati
on.
In
fact,
it
s
upports
th
e
creati
on
of
c
om
po
sit
e
sign
at
ur
e
w
hich
ca
n
hav
e
on
e
or
m
or
e
of
t
he
el
e
m
ents
su
c
h
as
visu
al
si
gnat
ure,
a
ud
i
o
sign
at
ur
e,
de
pt
h
signa
ture
,
and
m
et
adata
.
The
f
ollo
wing
are
t
he
im
p
or
ta
nt
ste
ps
i
nvolv
e
d
in
si
gn
at
ur
e
gen
e
rati
on.
1.
Com
pu
ti
ng
vis
ual d
e
script
or
s
for
im
ages.
2.
D
ividi
ng eac
h im
age in
to
bloc
ks
.
3.
Ma
tc
hin
g vis
ua
l descri
ptors
usi
ng Eu
cl
id
ean
d
ist
ance
4.
Bl
ock
disp
a
rity
co
m
pu
ta
ti
on
5.
Com
pu
te
sign
a
ture
D
i
L
-
D
j
R
=
√
1
−
1
)
2
+
⋯
…
…
…
…
…
+
(
−
)
2
(1)
√
(
(
−
)
/
)
2
+
(
(
−
)
/
)
2
)
(2)
Eq
uations (
1) a
nd (2) are
us
e
d t
o
m
at
ch
visu
a
l descri
ptors a
nd c
om
pu
ti
ng
bl
ock d
is
par
it
y.
5.2.
Distribu
ted
Matc
hing
Eng
in
e
The
distri
bu
te
d
m
a
tc
hin
g
en
gin
e
is
m
ade
up
of
obj
ect
m
a
tc
hin
g
a
nd
distrib
uted
in
dex
com
po
ne
nts
.
Mult
i
m
edia
ob
j
ect
s
are
cha
ra
ct
erized
by
m
a
ny
featur
e
s
co
ntainin
g
h
ig
h
dim
ension
s.
F
or
insta
nce,
an
i
m
age
can
be
re
pr
e
se
nted
by
10
0
-
200
S
IF
T
descri
pto
rs
.
I
n
each
descr
ipt
or,
th
ere
m
igh
t
be
up
to
128
dim
e
ns
io
ns
.
Howe
ver,
this
diff
e
rs
f
ro
m
each
m
ultim
edia
obj
ect
.
A
m
atch
in
g
en
gin
e
i
s
con
st
ru
ct
e
d
that
descr
i
bes
obj
ect
m
at
chi
ng
lo
gic
and
distri
bu
te
d
in
dex
tree
th
at
ho
ld
s
sig
nat
ur
es
of
m
ultimed
ia
ob
j
ect
s.
T
her
e
a
re
th
ree
ste
ps
involve
d
in
ob
je
ct
m
at
ching
. Fi
rst of all
que
ry
data set
is
pa
rtit
ion
ed
. For
each
data
point
,
K
-
nea
rest
nei
ghbors
are
fou
nd
.
A
fterw
a
rd
,
ap
plica
ti
on
sp
eci
fic
obj
ect
m
at
ching
is
carried
out.
The
pr
eci
si
on
(3)
an
d
av
erage
pr
eci
sio
n
(
4)
a
re
us
ed
to
e
va
luate
the
syst
e
m
by
cal
culat
i
ng
the
acc
urac
y
of
the
K
-
nea
rest
nei
ghbors
for
a
po
i
nt and
over
al
l t
he
points i
n qu
e
ry set.
Pr
eci
sio
n @
K
(p)=
∑
{
<
=
}
=
1
(
3)
Av
e
ra
ge
P
recis
ion
@
K=
∑
{
@
(
)
}
|
|
=
1
|
|
(
4)
Their
em
pirical
resu
lt
s
rev
e
al
ed
that
si
gnat
ur
e
for
3
-
D
vid
e
os
show
e
d
high
acc
ur
a
cy
reg
a
rd
i
ng
pr
eci
sio
n
a
nd
re
cal
l. Th
ei
r
syst
e
m
is
sti
l
l t
o
be
i
m
pr
ov
e
d
to
su
pp
or
t q
uick verificat
io
n
of
sh
ort
v
ide
o
se
gm
ents
and
li
ve
st
rea
m
ing
vid
e
os
f
or
c
on
te
nt
prot
ect
ion
.
Sim
il
ar
kind
of
resea
r
ch
is
m
ade
by
Nihar
i
ka
an
d
Sahoo
(20
16) [20]
for
cl
oud
-
base
d
m
ultim
edi
a con
t
ent pr
otect
ion
syst
e
m
.
6.
SUMM
A
RY
OF
MU
LT
I
M
EDIA C
ONT
ENT PR
OTE
CTIO
N MET
HODS
Table
2
s
hows
a
su
m
m
ary
of
the
researc
h
t
hat
pro
vid
es
i
nsi
gh
ts
int
o
dif
f
eren
t
te
ch
niqu
es
e
m
plo
ye
d
for
m
ultim
edia
content
protec
ti
on
.
Table
2.
Su
m
m
ary o
f
Me
th
ods
f
or
M
ultim
edia Co
ntent P
ro
t
ect
ion
Au
th
o
r
&
Y
ear
Techn
iq
u
e
Ad
v
an
tag
es
Li
m
itat
io
n
s
Re
m
arks
Ha
m
p
ap
u
r,
Hy
u
n
,
an
d
Bo
lle (
2
0
0
2
)
[
3
]
Seq
u
en
ce
m
atch
in
g
tech
n
iq
u
es f
o
r
co
p
y
d
etectio
n
Detectio
n
of
cop
ied
m
o
v
ie
clips
Ind
ex
in
g
sch
e
m
es
f
o
r
p
arallel
co
n
v
o
lu
tio
n
are
y
et to
b
e
i
m
p
l
e
m
en
ted
.
Co
lo
u
r
an
d
inten
sit
y
b
ased
sig
n
atu
res
are
u
sed
.
Lee
an
d
Yoo
(20
0
8
)
[
5
]
No
v
el vid
eo
f
in
g
erprint
in
g
m
et
h
o
d
Perf
o
r
m
s b
ette
r
th
a
n
ex
istin
g
on
es.
Ro
b
u
stn
es
s ag
ain
st
trans
f
o
r
m
atio
n
s
is n
o
t
y
et evalu
ated
.
Help
s in
con
ten
t
-
b
ased
v
id
eo
id
en
tif
icatio
n
.
Tasd
e
m
ir
and
Cetin
(
2
0
1
0
)
[
1
4
]
Vector
-
b
ased
f
eatu
re
set f
o
r
co
n
ten
t
-
b
ased
co
p
y
detectio
n
(CBC
D)
Featu
re
sets
repres
en
t
v
id
eo
f
o
r
d
etectin
g
co
p
y
detectio
n
.
-
Mean o
f
the Magn
itu
d
es
o
f
M
o
tio
n
Vectors
(M
M
MV
)
an
d
M
e
an
of
th
e Phas
e Ang
les o
f
Motio
n
Vectors
(M
P
MV)
ar
e
exp
lo
ited
.
Metois
et
al
.
(20
1
1
)
[
13
]
Detectio
n
techn
iq
u
e to
f
in
d
an
o
n
li
n
e
ab
u
s
e of
i
m
ag
es
Ch
arac
teriz
atio
n
of
i
m
ag
es an
d
thu
s
d
etectio
n
of
abu
se
-
Un
ited
Nation
s Pat
en
t
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4204
-
4211
4210
Au
th
o
r
&
Y
ear
Techn
iq
u
e
Ad
v
an
tag
es
Li
m
itat
io
n
s
Re
m
arks
Saraswathi
&
Ven
k
atesu
lu
(20
1
2
)
[
4
]
Blo
ck
ciph
er
alg
o
rith
m
Multi
m
ed
ia
co
n
ten
t
p
rotectio
n
with
en
cry
p
tio
n
-
Sh
o
ws b
etter
p
erfo
r
m
an
ce than
DES
alg
o
rith
m
.
Iof
f
e (
2
0
1
2
)
[
7
]
Fu
ll
-
len
g
th
v
id
eo
f
in
g
erprint
in
g
Ch
arac
teriz
es en
tir
e
d
u
ration
of
the
v
id
eo
an
d
su
p
p
o
rts
n
ear
-
d
u
p
licate
d
etectio
n
.
-
Un
ited
Nation
s Pat
en
t
Kh
o
d
ab
ak
sh
i an
d
Hef
eeda (20
1
3
)
[
1
5
]
No
v
el
co
n
ten
t
-
b
ased
co
p
y
detectio
n
f
o
r
3
D
v
id
eo
s.
Hig
h
precisio
n
and
reca
ll
-
3
D f
o
r
m
ats o
f
vid
eo
s
are
su
p
p
o
rted
.
W
ei
et al.
(
2
0
1
4
)
[
1
9
]
Privacy
cheatin
g
d
isco
u
rage
m
en
t
an
d
secu
re
co
m
p
u
tatio
n
au
d
itin
g
pro
to
co
l
Su
p
p
o
rts secu
re
sto
rage and
secu
re
co
m
p
u
tatio
n
as
we
ll.
Linear prog
ra
m
co
m
p
u
tatio
n
and
data
m
in
in
g
m
o
d
els
are
no
t
y
et f
o
r
m
a
lized
.
A
testb
ed
k
n
o
wn
as
SecHDFS
is u
sed
f
o
r
th
e
e
m
p
irical
stu
d
y
.
Zhan
g
et
al.
(20
1
4
)
[
1
0
]
No
v
el dig
ital r
ig
h
ts
m
a
n
ag
e
m
en
t
(DR
M)
tech
n
iq
u
e.
Mob
ile
m
u
lti
m
ed
i
a
co
n
ten
t is pro
tected
u
sin
g
DRM.
-
Protectio
n
o
f
co
p
y
righ
ted
con
ten
ts in
a
m
o
b
ile
en
v
iron
m
e
n
t.
Haf
eeda et al.
(20
1
5
)
[
1
8
]
Clo
u
d
-
b
ased
m
u
lti
m
ed
ia
con
ten
t
p
rotectio
n
sy
ste
m
with
sig
n
atu
re
g
en
eration
an
d
dis
tribu
ted
m
a
tch
in
g
eng
in
e.
Su
p
p
o
rts diff
erent
ty
p
es o
f
m
u
lti
m
ed
i
a
co
n
ten
t.
Batch
pro
cess
in
g
,
su
p
p
o
rt
for
m
u
lti
-
v
iew
p
lu
s d
ep
th
vid
e
o
s
are
n
o
t exp
lo
red
.
The d
istrib
u
ted
in
d
ex
h
elp
s in
o
b
ject
m
at
ch
in
g
an
d
qu
ery
pro
cess
i
n
g
.
7.
CONCL
US
I
O
NS
A
ND FUT
UR
E
WO
RK
Of
la
te
m
ultimed
ia
con
te
nt
grow
i
ng
e
xpone
ntial
ly
le
d
to
the
e
m
erg
ence
of
the
cl
ou
d
w
he
re
people
in
gen
e
ral
an
d
m
ul
tim
edia
con
te
nt
pr
ov
i
ders
can
sto
re
a
nd
retrieve
la
r
ge
volum
es
of
m
ultim
e
dia
con
te
nt.
Con
te
nt
pro
vi
der
s
w
ho
are
storing
m
ultim
edia
con
te
nt
in
public
cl
oud
m
igh
t
lose
rev
e
nues
w
he
n
thei
r
le
gitim
at
e
con
te
nt
gets
pirate
d
il
le
gally
.
The
rati
on
al
e
be
hi
nd
this
is
that
te
chn
ol
og
ic
al
adv
a
ncem
ents
in
the
com
pu
ti
ng
w
orl
d
m
ade
the
c
on
te
nt
du
plica
ti
on
a
nd
hosti
ng
easi
er
.
T
hu
s
there
is
e
ver
y
increasi
ng
t
hreat
to
le
gitim
at
e
m
ul
tim
edia
con
te
nt
ov
e
r
the
cl
oud
.
P
ro
te
ct
in
g
su
c
h
intel
le
ct
ual
pr
ope
rty
need
s
to
be
giv
en
par
am
ount
i
m
portance
.
The
refor
e
it
is
i
nev
it
able
to
hav
e
a
m
or
e
so
phist
ic
at
ed
m
e
chan
ism
t
hat
can
dynam
ic
al
l
y
pr
otect
m
ultim
e
dia
co
ntent.
H
ow
e
ve
r,
in
a
distrib
uted
e
nv
iro
nm
ent
it
is
chall
eng
i
ng
to
have
com
pu
ta
ti
on
al
ly
intensive
op
erati
on
s
.
I
n
t
hi
s
pap
e
r,
we
m
ake
a
re
view
of
t
he
pres
ent
sta
te
-
of
-
t
he
-
art
of
m
et
ho
ds
a
vaila
ble
f
or
m
ultim
ed
ia
-
c
on
te
nt
pr
otect
ion
.
The
insig
hts
of
the
pap
e
r
al
s
o
incl
ud
e
oppo
rtu
niti
es
and
chall
enges
in
t
he
protect
ion
of
rig
hts
of
le
gitim
at
e
us
ers
of
the
c
onte
nt.
I
n
f
ut
ur
e,
we
intend
to
pro
pose
a
nd
i
m
ple
m
ent a soph
ist
ic
at
ed
cl
oud
-
based m
echan
ism
f
or
prote
ct
ing
m
u
lt
i
m
ed
ia
co
nte
nt.
REFERE
NCE
S
[1]
H.
E.
Sur
y
av
anshi,
e
t
al
.
,
“
Digital
Im
age
W
ater
m
ark
ing
in
W
av
el
e
t
Dom
ai
n
,
”
I
nte
rnational
Jou
rnal
of
Elec
tri
ca
l
and
Computer
E
ngine
ering
(
IJECE)
,
v
ol
/i
ss
ue:
3
(
1
)
,
pp
.
1
-
6
,
201
3
.
[2]
T
.
Bia
n
chi
and
A
.
Piva
,
“
Secur
e
W
at
ermarki
ng
for
Multi
m
edi
a
Conte
nt
Protect
ion:
A
rev
ie
w
o
f
it
s
bene
fit
s
an
d
open
issues
,
”
IE
EE
,
vol
/
issue:
30
(
2
)
,
pp
.
1
-
23
,
20
13.
[3]
A
.
Ham
papur
,
e
t
al.
,
“
Com
par
ison
of
Sequence
Matc
hing
Tech
nique
s
for
Vide
o
Cop
y
De
tection
,
”
S
torage
an
d
Re
tri
ev
al
for M
e
dia
Databases
,
v
ol.
4676
,
pp
.
194
-
201
,
2002
.
[4]
P.
V
.
Sara
sw
at
hi
and
M.
Venka
te
sulu
,
”A
Bloc
k
Ciphe
r
Algorit
hm
for
Multi
m
edi
a
Conte
nt
Protec
ti
on
wit
h
Random
Subs
ti
tut
ion
using
Binar
y
Tr
ee
Tr
ave
rs
al
,
”
Journal
of
Computer
Sci
en
ce
,
vol
/i
ss
ue:
9
(
8
)
,
pp.
154
1
-
15
46
,
2012
.
[5]
S
.
Le
e
and
C
.
D
.
Yoo,
“
Robust
Video
Fingerprinting
for
Conte
n
t
-
Based
Video
I
dent
ifica
ti
on
,
”
I
EE
E
transact
ion
s
on
ci
rcu
it
s and
systems f
or v
ide
o
te
chno
logy
,
vol
/is
sue:
18
(
7
)
,
pp.
983
-
988
,
2008
.
[6]
S
.
Ioffe
,
“
Full
-
L
engt
h
Vid
eo
Fin
ger
printing
,
”
Un
it
ed
States P
a
te
n
t
,
pp
. 1
-
11
,
2012
.
[7]
E
.
Me
toi
s,
et al
.
,
“
Dete
c
ti
ng
Onl
i
ne
Abus
e
In
Im
a
ges
,
”
Uni
te
d
States P
at
ent
,
pp
.
1
-
10
,
2011
.
[8]
R
.
Bouta
ba
,
et
a
l.
,
“
Recent
tre
nd
s
in
int
era
ctive
m
ult
imedia
computing
for
the
in
dustr
y
,
”
Cluste
r
Computing
,
vol.
17,
pp
.
723
-
726
,
2014
.
[9]
C
.
Patsaki
s
and
A
.
Zi
gom
it
ros,
“
Achil
l
ea
s
Papag
eor
giou
and
Agus
ti
Solana
s
Pri
vacy
and
Secur
ity
for
Multi
m
edia
Conte
nt
share
d
o
n
OS
Ns
,
”
Iss
ues
and
Counte
rm
ea
sur
es,
The
Britis
h
Computer
Soc
i
et
y
,
pp
.
1
-
18
,
20
14
.
[10]
Z
.
Zh
ang,
et
al.
,
“
A
novel
appr
oac
h
to
right
s
sharing
-
e
nabl
ing
dig
it
a
l
right
s
m
ana
gement
for
m
obil
e
m
ult
imedia
,
”
Sp
ringer
Scienc
e
,
p
p.
1
-
17
,
2014
.
[11]
F
.
Fratt
ol
il
lo
,
“
A
Digit
al
R
ights
Mana
gement
S
y
stem
Based
on
Cloud
,
”
TE
LK
OMNIKA
Tele
communic
a
ti
o
n
Computing
E
le
c
t
ronics
and
Cont
rol
,
v
ol
/i
ss
ue:
15
(
2
)
,
pp
.
671
-
677
,
2017
.
[12]
J
.
Lu,
“
Video
finge
rprinting
for
cop
y
id
ent
if
ication:
from
rese
arc
h
to
industr
y
ap
pli
c
at
ions
,
”
M
ed
ia
Forensic
s
and
Sec
urit
y
,
vol
/
issue:
725402
(
1
)
,
p
p.
1
-
15
,
2009
.
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
A Survey
on M
ulti
med
ia
Co
nt
ent Pr
otect
ion
Mech
an
is
ms
(
Go
tt
um
ukka
l
a Him
a
Bi
ndu)
4211
[13]
Z
.
Li
u
,
e
t
al.
,
“
E
ffe
ctive
and
Sc
alable
Vid
eo
Copy
Det
ection
,
”
I
nt
ernati
onal
con
fer
enc
e
on
Multimedia
inf
orm
ati
o
n
retrie
va
l
Pages
,
pp.
119
-
128
,
20
10
.
[14]
K
.
T
.
Dem
ir
a
nd
A.
Eni
s
C
.
,
“
Motion
Vec
t
or
Based
Fe
at
u
res
for
Cont
ent
-
Based
Video
Cop
y
De
tecti
on
,
”
Inte
rnational
Co
nfe
renc
e
on
Patt
ern
Recogni
t
ion
,
pp.
3134
-
3137
,
2010
.
[15]
M
.
Hefe
eda
and
N
.
Khodaba
khsh
i
,
“
Spider: A S
y
s
te
m
for
Finding
3D Vide
o
Copi
e
s
,
”
ACM
,
pp
.
1
-
20
,
2013
.
[16]
C
.
Ye,
et
al.
,
“
Secur
e
Multi
m
edi
a
Con
te
nt
Distribut
ion
for
M2M
Comm
uni
ca
t
ion
,
”
Int
ernat
ional
Journal
o
f
Sec
urit
y
and
Its
Appl
ic
a
ti
ons
,
vo
l
/i
ss
ue:
10
(
4
)
,
pp
.
279
-
288
,
2016
.
[17]
F
.
Fund,
et
al.
,
“
Under
a
cl
ou
d
of
unce
rta
inty:
Le
gal
quest
io
ns
aff
ec
ti
ng
Int
ern
et
storag
e
an
d
tra
nsm
ission
of
cop
y
r
ight
-
pro
te
c
te
d
v
ide
o
conten
t
,
”
ACM
,
pp
.
1
-
14
,
2016
.
[18]
M
.
Hefe
eda,
e
t
al.
,
“
Cloud
-
Bas
ed
Multi
m
edi
a
Conte
nt
Prote
ct
i
on
Sy
st
em
,
”
IE
EE
Tr
ansacti
on
s
on
Mult
imedia
,
vol
/i
ss
ue:
17
(
3
)
,
pp.
1
-
14
,
2015
.
[19]
L
.
W
e
i,
et
al
.
,
“
Secur
ity
and
pri
vacy
for
storag
e
and
computatio
n
in
cl
oud
computing
,
”
Informati
on
Scienc
es
,
vol
.
258,
pp
.
371
–
38
6
,
2014
.
[20]
M.
Nihar
ika
and
P.
K
.
Sahoo,
“
Protec
ti
ng
Clou
d
-
Based
Multi
m
edi
a
Cont
ent
usi
ng
3
-
D
Signat
ur
es
,
”
Inte
rnat
iona
l
Journal
of
Ad
va
nce
d
Comput
ing
Techni
qu
e
and
Appl
ic
a
ti
ons
,
vo
l
/i
ss
ue:
4
(
1
)
,
pp.
1
-
4
,
2016
.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Mrs
.
G.Him
a
B
indu
pursuing
P
h.
D.
in
the
Dep
art
m
ent
of
Com
pute
r
scie
n
ce
an
d
Engi
nee
r
ing,
Acha
r
y
a
Nag
arjuna
Univer
si
t
y
.
She
was
awa
rd
e
d
B.
T
ec
h
in
Info
rm
at
ion
T
ec
hnol
og
y
in
2004
an
d
rec
e
ive
d
M.t
ec
h
in
Com
pute
r
S
ci
en
ce
and
Eng
i
nee
ring
in
th
e
y
ea
r
2009.
Her
r
ese
arc
h
in
te
r
ests
inc
lud
e
Secur
ity
.
Mrs
.
Ch.
Anura
dha
was
awa
rde
d
B.
T
ec
h
in
Inf
orm
at
ion
T
ec
hn
olog
y
from
Ach
ar
y
a
Nag
arj
un
a
Univer
sit
y
in
2
007
and
rec
e
iv
ed
M.t
ec
h
in
Com
pute
r
Scie
nc
e
and
Engi
ne
ering
from
J
NTU
Kakina
da
in
th
e
y
ea
r
2014
.
Pre
sently
she
is
w
orking
as
As
sistant
Profess
or
in
D
epa
rtment
of
Com
pute
r
Scie
n
ce
&
Eng
ine
e
ri
ng,
V.
R.
Sidd
har
tha
Eng
ineer
ing
Coll
eg
e
En
gine
er
ing.
Her
rese
arc
h
in
te
rest
s inc
lud
e
Im
age
Proce
ss
ing
and
Data
m
ini
ng
.
Dr
.
P.
Sri
R
ama
Chandr
a
Murt
h
y
was
awa
rde
d
B.
T
ec
h
in
Co
m
pute
r
Scie
n
ce
and
Engi
n
ee
r
ing
from
JN
TU
H
in
2005
and
rec
eived
M.t
e
ch
in
Com
pute
r
Scie
nc
e
and
Engi
ne
eri
n
g
from
Acha
r
y
a
Naga
rjuna
Univ
ersity
in
the
y
e
a
r
2008.
He
was
awa
rde
d
a
doc
to
rat
e
in
th
e
y
e
ar
2013.
Present
l
y
he
is
working
as
As
sistant
Profess
or
in
Depa
rtme
nt
of
Com
pute
r
Scie
nc
e
&
En
gin
ee
ring
,
Acha
r
y
a
Naga
rjuna
Univ
ersity
.
His
rese
arc
h
int
er
ests
in
cl
ude
Dig
ital
I
m
age
Proce
ss
in
g,
Dat
a
Mining
,
Network
Secur
i
t
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.