T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
ni
cs a
nd
Co
ntr
o
l
Vo
l
.
23
,
No
.
6
,
Dec
em
b
er
20
25
,
p
p
.
1
4
7
6
~1
494
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
23
i
6
.
27211
1476
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l
.
uad
.
ac
.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Wa
ter
m
a
r
k
ing
o
n
sprea
d
m
ul
ti
-
fr
a
m
e da
ta v
ideo
us
ing
discrete
w
a
v
elet
trans
form
hy
brid and
fra
m
e
ratio
m
ess
a
g
e
v
a
ria
nce
I
lha
m
F
ir
m
a
n A
s
ha
ri,
Sa
rw
o
no
Su
t
ikn
o
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
c
s
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
I
n
d
u
st
r
i
a
l
T
e
c
h
n
o
l
o
g
y
,
I
n
st
i
t
u
t
T
e
k
n
o
l
o
g
i
S
u
ma
t
e
r
a
,
L
a
mp
u
n
g
S
e
l
a
t
a
n
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
12
,
2025
R
ev
i
s
ed
Sep
23
,
2
0
2
5
A
cc
ep
ted
Oct
19
,
2
0
2
5
T
h
e
e
x
p
o
n
e
n
ti
a
l
g
ro
w
th
o
f
v
id
e
o
sh
a
rin
g
d
e
m
a
n
d
s
se
c
u
re
a
n
d
im
p
e
rc
e
p
ti
b
le
w
a
t
e
r
m
a
r
k
in
g
m
e
th
o
d
s
.
T
h
is
stu
d
y
p
re
se
n
ts
a
v
id
e
o
w
a
ter
m
a
r
k
in
g
fra
m
e
w
o
rk
u
sin
g
d
isc
re
te
w
a
v
e
let
tran
sf
o
r
m
(DWT
)
w
it
h
h
y
b
rid
su
b
-
b
a
n
d
e
m
b
e
d
d
i
n
g
a
n
d
m
u
lt
i
-
f
r
a
m
e
a
ll
o
c
a
ti
o
n
t
o
b
a
lan
c
e
im
p
e
rc
e
p
ti
b
il
it
y
,
c
a
p
a
c
it
y
,
a
n
d
r
o
b
u
st
n
e
ss
.
W
a
ter
m
a
rk
b
it
s
a
re
a
d
a
p
ti
v
e
ly
d
istri
b
u
ted
a
c
ro
ss
lo
w
-
lo
w
(LL
),
lo
w
-
h
ig
h
(
L
H),
a
n
d
h
ig
h
-
lo
w
(HL
)
su
b
-
b
a
n
d
s
o
f
se
lec
ted
f
ra
m
e
s,
w
it
h
u
n
c
o
m
p
re
ss
e
d
a
u
d
io
v
id
e
o
i
n
terle
a
v
e
(A
V
I)
e
n
su
rin
g
c
o
e
f
f
icie
n
t
in
teg
rit
y
.
Ex
p
e
rim
e
n
ts
o
n
6
4
0
×
3
6
0
v
id
e
o
s
sh
o
w
L
L
-
o
n
l
y
e
m
b
e
d
d
in
g
a
c
h
iev
e
s
h
ig
h
im
p
e
rc
e
p
ti
b
il
it
y
(
p
e
a
k
sig
n
a
l
-
to
-
n
o
ise
ra
ti
o
(
P
S
N
R
)
>
3
8
.
6
d
B,
stru
c
t
u
ra
l
sim
il
a
rit
y
in
d
e
x
m
e
a
su
re
(
S
S
IM
)
≥
0
.
9
9
4
5
,
a
n
d
b
i
t
e
rro
r
ra
te
(
BER
)
=
0
),
w
h
il
e
L
L
-
d
o
m
in
a
n
t
h
y
b
rid
s in
c
re
a
se
c
a
p
a
c
it
y
w
it
h
slig
h
t
ro
b
u
st
n
e
ss
trad
e
-
o
f
fs
.
Em
b
e
d
d
in
g
in
L
H
a
n
d
h
ig
h
-
h
ig
h
(
HH
)
su
b
-
b
a
n
d
s
ra
ise
s
d
isto
rti
o
n
v
u
ln
e
ra
b
i
li
ty
.
Un
d
e
r
c
ro
p
p
i
n
g
,
BER
rise
s
f
ro
m
0
.
0
0
5
t
o
0
.
2
0
5
(
0
–
5
0
%
),
a
n
d
n
o
rm
a
li
z
e
d
c
o
rre
lat
io
n
(
NC
)
d
ro
p
s
f
ro
m
0
.
9
9
8
to
0
.
8
0
2
,
re
m
a
in
in
g
a
c
c
e
p
tab
le
f
o
r
≤3
0
%
c
ro
p
p
in
g
.
T
h
e
sc
h
e
m
e
re
sists
jo
in
t
p
h
o
t
o
g
ra
p
h
ic
e
x
p
e
rts’
g
ro
u
p
(JP
EG
)
c
o
m
p
re
ss
i
o
n
q
u
a
li
ty
f
a
c
to
r
2
0
–
8
0
(Q2
0
–
Q
8
0
)
,
re
siz
in
g
(≥7
0
%
),
a
n
d
m
il
d
G
a
u
ss
ian
b
lu
r
(3
×
3
),
m
a
in
tain
in
g
e
f
f
icie
n
t
d
e
c
o
d
in
g
u
n
d
e
r
h
ig
h
e
r
p
a
y
lo
a
d
s
.
F
u
tu
re
w
o
rk
m
a
y
a
p
p
ly
e
rro
r
-
c
o
rre
c
ti
o
n
c
o
d
in
g
a
n
d
re
d
u
n
d
a
n
c
y
-
a
wa
re
e
m
b
e
d
d
in
g
f
o
r
im
p
ro
v
e
d
re
sili
e
n
c
e
.
Ov
e
ra
ll
,
th
e
p
r
o
p
o
se
d
m
e
th
o
d
o
f
fe
rs
a
se
c
u
re
,
a
d
a
p
ti
v
e
,
a
n
d
e
ff
ici
e
n
t
so
lu
ti
o
n
f
o
r
v
id
e
o
a
u
th
e
n
ti
c
a
ti
o
n
a
n
d
c
o
v
e
rt
c
o
m
m
u
n
ica
ti
o
n
.
K
ey
w
o
r
d
s
:
Dis
cr
ete
w
av
ele
t tr
an
s
f
o
r
m
Fra
m
e
r
atio
Sp
r
ea
d
m
u
lti
-
f
r
a
m
e
Vid
eo
W
ater
m
ar
k
i
n
g
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
:
I
lh
a
m
Fir
m
a
n
A
s
h
ar
i
Dep
ar
t
m
en
t o
f
I
n
f
o
r
m
atic
s
E
n
g
in
ee
r
i
n
g
,
Facu
lt
y
o
f
I
n
d
u
s
tr
ia
l T
ec
h
n
o
lo
g
y
,
I
n
s
tit
u
t T
ek
n
o
lo
g
i S
u
m
a
ter
a
St
.
T
er
u
s
an
R
y
ac
u
d
u
,
L
a
m
p
u
n
g
Selata
n
,
I
n
d
o
n
esia
E
m
ail:
f
ir
m
a
n
.
as
h
ar
i
@
i
f
.
iter
a
.
a
c
.
id
1.
I
NT
RO
D
UCT
I
O
N
I
n
th
e
co
n
te
m
p
o
r
ar
y
d
ig
ita
l
la
n
d
s
ca
p
e,
th
e
ex
p
o
n
e
n
tia
l
g
r
o
w
t
h
o
f
v
id
eo
co
n
ten
t
d
i
s
tr
ib
u
t
io
n
ac
r
o
s
s
n
atio
n
al
a
n
d
in
ter
n
atio
n
al
p
lat
f
o
r
m
s
h
as
h
ei
g
h
ten
ed
t
h
e
d
e
m
an
d
f
o
r
r
o
b
u
s
t c
o
n
te
n
t p
r
o
tectio
n
m
ec
h
an
is
m
s
[
1
]
.
As
v
id
eo
b
ec
o
m
es
a
d
o
m
i
n
an
t
m
ed
iu
m
f
o
r
co
m
m
u
n
icat
io
n
,
ed
u
ca
tio
n
,
e
n
ter
tain
m
e
n
t
,
an
d
co
m
m
er
cial
ex
ch
a
n
g
e
,
it
also
in
cr
ea
s
in
g
l
y
b
ec
o
m
e
s
a
tar
g
et
f
o
r
u
n
a
u
th
o
r
ized
d
u
p
licatio
n
,
ta
m
p
er
in
g
,
an
d
m
is
u
s
e
[
2
]
.
T
h
is
g
r
o
w
i
n
g
v
u
l
n
er
ab
ilit
y
u
n
d
er
s
co
r
es
th
e
cr
itical
n
ee
d
f
o
r
ad
v
an
ce
d
tech
n
iq
u
e
s
ca
p
ab
le
o
f
s
af
e
g
u
ar
d
in
g
in
tellect
u
al
p
r
o
p
er
ty
a
n
d
v
er
i
f
y
in
g
co
n
te
n
t a
u
t
h
en
t
icit
y
.
Dig
ital
w
ater
m
ar
k
i
n
g
h
as
e
m
e
r
g
ed
as
a
v
iab
le
tech
n
o
lo
g
ical
r
esp
o
n
s
e
to
th
ese
ch
allen
g
es,
o
f
f
er
i
n
g
a
m
et
h
o
d
to
em
b
ed
i
m
p
er
ce
p
tib
le
id
en
tific
atio
n
d
ata
w
it
h
i
n
m
u
lti
m
ed
ia
co
n
te
n
t
[
3
]
.
P
ar
ticu
lar
l
y
in
t
h
e
d
o
m
ai
n
o
f
v
id
eo
,
r
ed
,
g
r
ee
n
,
b
lu
e
(
R
GB
)
w
ater
m
ar
k
i
n
g
tec
h
n
iq
u
e
s
h
a
v
e
a
ttra
cted
s
i
g
n
if
ican
t
at
ten
tio
n
d
u
e
to
th
e
ir
p
o
ten
tial
to
p
r
eser
v
e
v
is
u
al
f
id
elit
y
w
h
ile
p
r
o
v
id
in
g
la
y
er
s
o
f
s
ec
u
r
it
y
a
n
d
tr
ac
ea
b
ilit
y
[
4
]
,
[
5
]
.
Ho
w
e
v
er
,
d
esp
ite
its
p
r
o
m
is
e,
w
a
ter
m
ar
k
in
g
i
n
v
id
eo
r
e
m
ai
n
s
a
co
m
p
le
x
t
ask
,
co
n
f
r
o
n
ted
w
i
th
i
s
s
u
es
s
u
c
h
as
p
er
ce
p
tu
a
l
d
eg
r
ad
atio
n
,
f
r
a
g
ilit
y
u
n
d
er
co
m
m
o
n
s
ig
n
al
a
ttack
s
(
co
m
p
r
es
s
io
n
,
s
ca
lin
g
,
f
r
a
m
e
m
a
n
ip
u
lat
io
n
)
,
an
d
t
h
e
n
ee
d
f
o
r
ad
ap
tiv
e
e
m
b
ed
d
in
g
s
ch
e
m
es t
h
at
b
alan
ce
r
o
b
u
s
tn
e
s
s
a
n
d
in
v
is
ib
il
it
y
[
6
]
,
[
7
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t
E
l
C
o
n
tr
o
l
W
a
te
r
ma
r
kin
g
o
n
s
p
r
ea
d
mu
lt
i
-
fr
a
me
d
a
ta
vid
eo
u
s
in
g
d
is
crete
w
a
ve
let
tr
a
n
s
fo
r
m
…
(
I
lh
a
m
F
ir
ma
n
A
s
h
a
r
i
)
1477
T
o
ad
d
r
ess
th
ese
co
n
ce
r
n
s
,
co
n
te
m
p
o
r
ar
y
r
esear
ch
h
a
s
in
cr
e
asin
g
l
y
e
x
p
lo
r
ed
h
y
b
r
id
an
d
t
r
an
s
f
o
r
m
-
b
ased
w
a
ter
m
ar
k
in
g
ap
p
r
o
ac
h
es
.
Am
o
n
g
th
e
s
e,
th
e
d
is
cr
ete
w
a
v
elet
tr
an
s
f
o
r
m
(
DW
T
)
h
as
p
r
o
v
en
esp
ec
iall
y
ef
f
ec
tiv
e,
o
w
i
n
g
to
its
m
u
lti
-
r
eso
lu
tio
n
r
ep
r
esen
tatio
n
an
d
f
r
eq
u
en
c
y
lo
ca
lizatio
n
p
r
o
p
er
tie
s
[
8
]
,
[
9
]
.
DW
T
-
b
ased
s
ch
e
m
es
allo
w
w
ater
m
ar
k
s
to
b
e
e
m
b
ed
d
ed
in
c
o
ef
f
icie
n
t
s
le
s
s
p
r
o
n
e
to
p
er
ce
p
tu
al
lo
s
s
,
th
er
eb
y
en
h
a
n
ci
n
g
in
v
is
ib
ilit
y
a
n
d
r
o
b
u
s
t
n
ess
[
1
0
]
.
Steg
a
n
o
g
r
ap
h
y
,
in
t
h
i
s
co
n
te
x
t,
r
ef
er
s
to
e
m
b
ed
d
in
g
h
id
d
en
in
f
o
r
m
atio
n
w
it
h
i
n
d
ig
ital
m
e
d
ia
s
u
ch
as
i
m
a
g
es,
au
d
io
,
an
d
v
id
eo
w
it
h
o
u
t
attr
ac
tin
g
atten
tio
n
,
d
if
f
er
i
n
g
f
r
o
m
cr
y
p
to
g
r
ap
h
y
w
h
ic
h
f
o
cu
s
es
o
n
en
cr
y
p
tio
n
[
1
1
]
.
An
e
f
f
ec
t
iv
e
s
y
s
te
m
m
u
s
t
u
lti
m
atel
y
p
r
o
v
id
e
h
i
g
h
ca
p
ac
it
y
,
i
m
p
er
ce
p
tib
ilit
y
,
a
n
d
r
esil
ien
c
e
ag
ain
s
t si
g
n
al
p
r
o
ce
s
s
in
g
o
r
d
elib
er
ate
attac
k
s
[
1
2
]
,
[
1
3
]
.
T
w
o
m
aj
o
r
m
et
h
o
d
o
lo
g
ical
b
r
an
ch
e
s
in
s
te
g
an
o
g
r
ap
h
y
a
n
d
w
ater
m
ar
k
i
n
g
ar
e
th
e
s
p
atia
l
d
o
m
a
in
a
n
d
tr
an
s
f
o
r
m
d
o
m
ai
n
tec
h
n
iq
u
es
[
1
4
]
,
[
1
5
]
.
Sp
atial
d
o
m
ain
m
et
h
o
d
s
o
p
er
ate
d
ir
ec
tly
o
n
p
ix
e
l
v
alu
e
s
o
r
te
m
p
o
r
al
s
a
m
p
les,
o
f
ten
o
f
f
er
i
n
g
s
i
m
p
l
icit
y
b
u
t
li
m
ited
r
o
b
u
s
t
n
es
s
[
1
6
]
.
T
r
an
s
f
o
r
m
d
o
m
ai
n
ap
p
r
o
ac
h
es,
b
y
co
n
tr
ast
,
e
m
b
ed
d
ata
in
tr
an
s
f
o
r
m
ed
r
ep
r
esen
tatio
n
s
s
u
ch
as
f
r
eq
u
e
n
c
y
o
r
w
av
e
let
d
o
m
ain
s
,
th
er
eb
y
o
f
f
er
in
g
en
h
a
n
ce
d
d
u
r
ab
ilit
y
an
d
i
m
p
r
o
v
ed
i
m
p
e
r
ce
p
tib
ilit
y
[
1
7
]
.
R
ec
en
t
s
tu
d
i
es
h
i
g
h
l
ig
h
t
d
ee
p
lear
n
in
g
as
a
p
o
w
er
f
u
l
d
ir
ec
tio
n
f
o
r
v
id
eo
w
ater
m
ar
k
in
g
b
u
t
als
o
ex
p
o
s
e
k
ey
li
m
itatio
n
s
.
B
is
tr
o
ń
an
d
P
io
tr
o
w
s
k
i
[
1
8
]
in
tr
o
d
u
ce
d
a
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
w
o
r
k
(
C
N
N)
-
b
ased
en
tr
o
p
y
-
d
r
iv
e
n
m
ap
p
er
th
at
i
m
p
r
o
v
ed
r
o
b
u
s
t
n
es
s
a
g
ai
n
s
t
co
m
p
r
ess
io
n
a
n
d
g
eo
m
etr
ic
d
is
to
r
tio
n
s
,
y
et
its
e
m
b
ed
d
in
g
ca
p
ac
it
y
w
a
s
r
estr
icted
an
d
co
m
p
u
tat
io
n
al
d
e
m
an
d
r
e
m
ai
n
ed
h
i
g
h
.
L
i
k
e
w
is
e,
Ma
n
s
o
u
r
et
a
l
.
[
1
9
]
p
r
o
p
o
s
ed
a
m
o
s
aic
g
e
n
er
atio
n
ap
p
r
o
ac
h
th
at
d
e
m
o
n
s
tr
ated
s
tr
o
n
g
r
esi
s
tan
ce
to
te
m
p
o
r
al
an
d
co
llu
s
io
n
attac
k
s
,
th
o
u
g
h
o
cc
asio
n
al
ar
tif
ac
t
s
r
e
d
u
ce
d
i
m
p
er
ce
p
tib
ilit
y
.
T
h
e
d
ec
is
io
n
to
u
s
e
a
u
d
io
v
id
eo
in
ter
lea
v
e
(
A
VI
)
f
o
r
m
at
as
t
h
e
v
id
eo
ca
r
r
ier
s
te
m
s
f
r
o
m
it
s
u
n
co
m
p
r
e
s
s
ed
o
r
lig
h
tl
y
co
m
p
r
ess
ed
s
tr
u
ct
u
r
e,
w
h
ic
h
r
etai
n
s
g
r
ea
ter
p
ix
el
-
le
v
el
f
id
elit
y
co
m
p
ar
ed
to
h
ea
v
il
y
co
m
p
r
es
s
ed
f
o
r
m
at
s
li
k
e
MP
4
(
m
o
v
i
n
g
p
ict
u
r
e
ex
p
er
ts
g
r
o
u
p
(
MP
E
G
)
-
4
P
a
r
t
1
4
)
[
2
0
]
,
[
2
1
]
.
T
h
is
ch
ar
ac
ter
is
tic
is
p
ar
ticu
lar
l
y
v
al
u
ab
le
f
o
r
tr
an
s
f
o
r
m
-
d
o
m
ai
n
w
ater
m
ar
k
i
n
g
,
w
h
er
e
m
i
n
u
te
alter
atio
n
s
i
n
c
o
ef
f
icie
n
t v
al
u
es
ar
e
s
en
s
iti
v
e
to
r
e
-
co
m
p
r
ess
io
n
ar
t
if
ac
ts
.
A
VI
also
o
f
f
er
s
f
r
a
m
e
-
by
-
f
r
a
m
e
ac
ce
s
s
ib
ili
t
y
,
w
h
ich
i
s
cr
itical
f
o
r
p
r
ec
is
e
co
n
tr
o
l
d
u
r
in
g
w
ater
m
ar
k
in
s
e
r
tio
n
an
d
ex
tr
ac
tio
n
ac
r
o
s
s
m
u
ltip
le
f
r
a
m
es
.
On
t
h
e
o
th
er
h
an
d
,
th
e
w
ater
m
ar
k
p
ay
lo
ad
s
el
ec
ted
in
t
h
is
s
t
u
d
y
i
s
a
p
lain
te
x
t
m
es
s
ag
e
s
a
v
ed
as
a
“
.
t
x
t
”
f
ile
.
T
h
e
r
atio
n
ale
f
o
r
u
s
i
n
g
p
lain
te
x
t
l
ies
in
its
u
b
iq
u
it
y
,
lo
w
co
m
p
u
ta
tio
n
al
o
v
er
h
ea
d
,
an
d
r
elev
a
n
ce
to
p
r
ac
tical
w
ate
r
m
ar
k
i
n
g
u
s
e
ca
s
es
s
u
c
h
as
o
w
n
er
s
h
ip
clai
m
s
,
m
etad
ata
e
m
b
ed
d
i
n
g
,
a
n
d
co
n
f
id
e
n
tial a
n
n
o
tatio
n
[
2
2
]
.
T
h
e
n
o
v
elt
y
o
f
th
is
s
tu
d
y
lie
s
in
t
h
e
h
y
b
r
id
s
u
b
-
b
an
d
w
ater
m
ar
k
i
n
g
s
tr
ateg
y
ap
p
l
ied
o
v
er
m
u
ltip
le
v
id
eo
f
r
a
m
e
s
,
w
h
er
e
w
a
ter
m
ar
k
b
its
ar
e
ad
ap
tiv
el
y
d
is
tr
ib
u
te
d
ac
r
o
s
s
s
elec
ted
DW
T
s
u
b
-
b
an
d
s
(
lo
w
-
lo
w
(
LL
)
,
lo
w
-
h
ig
h
(
LH
)
,
h
ig
h
-
lo
w
(
HL
)
)
ac
co
r
d
in
g
to
f
r
am
e
p
e
r
c
e
p
tu
a
l
an
d
s
t
r
u
ct
u
r
al
c
h
a
r
a
c
t
e
r
i
s
tic
s
.
T
h
i
s
m
u
l
ti
-
f
r
am
e
a
l
l
o
c
a
ti
o
n
d
i
s
p
e
r
s
es
t
h
e
p
ay
l
o
ad
t
em
p
o
r
al
ly
,
r
e
d
u
c
in
g
d
e
t
e
c
tab
i
l
i
ty
,
m
it
ig
at
in
g
p
e
r
c
e
p
tu
al
a
r
ti
f
ac
t
s
,
an
d
in
c
r
ea
s
in
g
r
e
s
i
li
en
c
e
a
g
a
in
s
t
f
r
am
e
l
o
s
s
,
c
r
o
p
p
i
n
g
,
o
r
t
em
p
o
r
a
l
a
t
t
a
ck
s
.
I
n
c
o
n
t
r
as
t
t
o
d
e
e
p
l
e
a
r
n
in
g
-
b
a
s
e
d
m
et
h
o
d
s
,
t
h
e
p
r
o
p
o
s
e
d
f
r
e
q
u
en
cy
-
d
o
m
a
in
a
p
p
r
o
a
c
h
a
ch
i
ev
e
s
s
u
p
e
r
i
o
r
im
p
e
r
c
e
p
t
i
b
i
li
ty
(
p
e
ak
s
ig
n
a
l
-
to
-
n
o
is
e
r
a
t
i
o
(
PS
NR
)
>
3
8
d
B
,
s
t
r
u
c
tu
r
al
s
im
il
a
r
ity
i
n
d
ex
m
e
asu
r
e
(
SS
I
M
)
>
0
.
9
9
i
n
L
L
-
o
n
ly
)
w
i
th
lig
h
tw
ei
g
h
t
c
o
m
p
u
t
at
i
o
n
.
W
h
i
le
m
i
d
-
an
d
h
ig
h
-
f
r
e
q
u
e
n
cy
b
an
d
s
(
L
H
/
h
ig
h
-
h
ig
h
(
HH
)
)
r
em
ai
n
v
u
ln
er
ab
le
to
d
is
to
r
tio
n
,
th
e
f
r
a
m
e
w
o
r
k
p
r
o
v
id
es
a
n
ef
f
icien
t a
n
d
p
r
ac
tical
alter
n
ati
v
e
f
o
r
s
ec
u
r
e
v
id
eo
au
t
h
en
t
icat
io
n
an
d
co
v
er
t c
o
m
m
u
n
icat
io
n
.
T
o
r
ig
o
r
o
u
s
l
y
v
alid
ate
t
h
e
p
r
o
p
o
s
ed
tech
n
iq
u
e,
a
co
m
p
r
e
h
en
s
i
v
e
e
v
alu
a
tio
n
w
a
s
co
n
d
u
cted
ac
r
o
s
s
m
u
ltip
le
p
er
f
o
r
m
a
n
ce
ax
es
[
2
3
]
–
[
2
5
]
.
E
m
b
ed
d
in
g
ca
p
a
cit
y
w
as
m
ea
s
u
r
ed
f
o
r
ea
ch
DW
T
s
u
b
b
an
d
co
n
f
i
g
u
r
atio
n
,
d
e
m
o
n
s
tr
atin
g
h
o
w
t
h
e
h
y
b
r
id
allo
ca
tio
n
i
m
p
ac
ts
to
tal
b
its
em
b
ed
d
ed
p
er
f
r
a
m
e
[
2
6
]
,
[
2
7
]
.
C
o
m
p
u
tatio
n
al
e
f
f
ic
ien
c
y
w
a
s
ass
es
s
ed
b
y
ca
lcu
lat
in
g
en
c
o
d
in
g
an
d
d
ec
o
d
in
g
r
u
n
ti
m
es
f
o
r
v
ar
io
u
s
f
r
a
m
e
m
es
s
ag
e
r
atio
s
a
n
d
s
u
b
b
an
d
co
m
b
i
n
atio
n
s
[
2
8
]
.
T
h
e
m
eth
o
d
w
a
s
s
u
b
j
ec
ted
to
r
ea
l
-
w
o
r
ld
s
ig
n
a
l
d
is
to
r
tio
n
s
in
cl
u
d
in
g
lo
s
s
y
MP
E
G
co
m
p
r
ess
io
n
,
Gau
s
s
ia
n
b
lu
r
,
r
esizin
g
o
p
er
atio
n
s
,
an
d
ad
d
itiv
e
n
o
is
e
to
ex
a
m
i
n
e
r
o
b
u
s
tn
es
s
[
2
9
]
.
A
cc
u
r
ac
y
was
m
ea
s
u
r
ed
u
s
i
n
g
th
e
b
it
er
r
o
r
r
ate
(
B
E
R
)
to
ca
p
tu
r
e
b
it
-
lev
el
d
is
cr
ep
an
cies
b
et
w
ee
n
o
r
ig
in
a
l
an
d
ex
tr
ac
te
d
m
e
s
s
a
g
es
[
3
0
]
.
P
SNR
w
as
c
o
m
p
u
ted
to
q
u
an
ti
f
y
p
er
ce
p
tu
al
d
eg
r
ad
atio
n
,
an
d
SS
I
M
w
as
u
s
ed
to
ev
al
u
ate
v
i
s
u
al
s
i
m
ilar
it
y
p
er
f
r
a
m
e
[
3
1
]
,
[
3
2
]
.
Fu
r
t
h
er
,
co
m
p
ar
ativ
e
p
er
f
o
r
m
an
ce
w
as
ex
p
lo
r
ed
b
y
s
i
m
u
lat
in
g
d
if
f
er
en
t
h
y
b
r
id
s
u
b
b
an
d
co
n
f
i
g
u
r
atio
n
s
(
L
L
-
o
n
l
y
,
L
H
-
o
n
l
y
,
H
L
-
o
n
l
y
,
an
d
m
i
x
ed
h
y
b
r
id
s
)
an
d
r
ec
o
r
d
in
g
th
eir
r
esp
e
ctiv
e
e
n
co
d
in
g
ti
m
e
,
d
ec
o
d
in
g
ti
m
e,
an
d
b
it
ca
p
ac
it
y
.
E
x
p
er
i
m
en
t
s
al
s
o
in
cl
u
d
ed
p
er
f
o
r
m
a
n
ce
c
u
r
v
e
s
r
elati
n
g
c
ap
ac
it
y
an
d
r
u
n
ti
m
e
to
ch
a
n
g
e
s
in
t
h
e
f
r
a
m
e
m
e
s
s
a
g
e
r
atio
.
A
ttack
-
s
p
ec
if
ic
e
v
alu
a
tio
n
s
w
er
e
co
n
d
u
cted
t
o
ass
ess
h
o
w
ea
ch
p
er
tu
r
b
atio
n
(
co
m
p
r
ess
io
n
,
r
es
ize,
n
o
is
e,
b
l
u
r
)
in
f
l
u
en
ce
d
d
e
co
d
in
g
s
u
cc
e
s
s
[
3
3
]
.
Me
tr
ics
s
u
ch
as
B
E
R
a
n
d
t
h
e
n
u
m
b
er
o
f
s
u
cc
e
s
s
f
u
ll
y
d
ec
o
d
ed
tex
t
lin
es
w
er
e
r
ep
o
r
ted
p
o
s
t
-
at
tack
,
w
it
h
v
i
s
u
a
l
p
lo
ts
illu
s
tr
atin
g
d
eg
r
ad
atio
n
tr
en
d
s
.
T
h
ese
co
m
p
r
eh
e
n
s
iv
e
an
d
r
ep
r
o
d
u
cib
le
ev
alu
atio
n
s
af
f
ir
m
t
h
e
r
eliab
ilit
y
,
ad
ap
tab
ilit
y
,
an
d
p
r
ac
tical
u
s
ab
ilit
y
o
f
t
h
e
p
r
o
p
o
s
ed
m
u
l
tif
r
a
m
e
w
ater
m
ar
k
i
n
g
s
c
h
e
m
e
b
ased
o
n
DW
T
h
y
b
r
id
s
u
b
b
a
n
d
e
m
b
ed
d
in
g
a
n
d
d
y
n
a
m
ic
f
r
a
m
e
allo
ca
tio
n
.
2.
M
E
T
H
O
D
Fig
u
r
e
1
p
r
esen
ts
th
e
e
n
d
-
to
-
en
d
w
o
r
k
f
lo
w
o
f
th
e
s
u
b
b
a
n
d
-
b
ased
v
id
eo
w
ater
m
ar
k
in
g
s
y
s
te
m
,
en
co
m
p
as
s
in
g
b
o
th
e
m
b
ed
d
in
g
a
n
d
ex
tr
ac
tio
n
.
I
n
t
h
e
e
m
b
e
d
d
in
g
p
h
a
s
e,
t
h
e
s
ec
r
et
m
es
s
a
g
e
is
b
in
ar
ized
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l
.
23
,
No
.
6
,
Dec
em
b
er
20
25
:
1
4
7
6
-
1494
1478
d
is
p
er
s
ed
ac
r
o
s
s
s
elec
ted
f
r
a
m
es
b
y
m
o
d
if
y
i
n
g
leas
t
s
ig
n
i
f
ic
an
t
b
it
(
L
SB
)
co
ef
f
icie
n
t
s
w
i
t
h
in
DW
T
s
u
b
b
an
d
s
.
T
h
e
m
o
d
if
i
ed
f
r
a
m
e
s
ar
e
th
e
n
r
ec
o
n
s
tr
u
cted
to
g
e
n
er
ate
th
e
s
teg
o
v
id
eo
w
h
ile
p
r
eser
v
in
g
v
i
s
u
al
q
u
a
lit
y
.
Du
r
in
g
th
e
e
x
tr
ac
t
io
n
s
ta
g
e,
tar
g
et
f
r
a
m
e
s
ar
e
id
en
tif
ied
an
d
d
ec
o
m
p
o
s
ed
th
r
o
u
g
h
DW
T
to
r
etr
iev
e
th
e
e
m
b
ed
d
ed
b
its
.
T
h
e
r
e
c
o
v
e
r
e
d
b
i
t
s
t
r
ea
m
i
s
t
h
e
n
v
al
i
d
a
t
e
d
u
s
i
n
g
an
en
d
o
f
f
i
le
(
E
O
F
)
m
a
r
k
e
r
a
n
d
c
o
n
v
e
r
t
e
d
b
a
ck
i
n
t
o
Am
e
r
i
c
an
S
t
an
d
a
r
d
C
o
d
e
f
o
r
I
n
f
o
r
m
at
i
o
n
I
n
t
e
r
ch
an
g
e
(
A
S
C
I
I
)
te
x
t
t
o
r
e
c
o
n
s
t
r
u
c
t
th
e
h
id
d
e
n
m
e
s
s
ag
e
.
T
h
i
s
s
t
r
u
c
tu
r
e
d
p
r
o
c
e
s
s
en
s
u
r
es
ef
f
ici
e
n
t
d
a
t
a
h
i
d
in
g
w
ith
a
b
a
la
n
c
e
o
f
im
p
e
r
c
e
p
ti
b
i
li
ty
,
r
o
b
u
s
tn
es
s
,
a
n
d
c
a
p
a
ci
ty
.
2
.
1
.
E
nco
de
Fig
u
r
e
2
p
r
esen
ts
a
co
m
p
r
e
h
en
s
iv
e
o
v
er
v
ie
w
o
f
th
e
f
r
a
m
e
s
el
ec
tio
n
p
r
o
ce
s
s
in
t
h
e
p
r
o
p
o
s
ed
s
u
b
b
an
d
-
b
ased
v
id
eo
w
ater
m
ar
k
i
n
g
s
y
s
te
m
.
T
h
e
d
iag
r
a
m
s
y
s
te
m
a
ticall
y
d
ep
icts
h
o
w
t
h
e
s
ec
r
et
m
es
s
a
g
e,
v
id
eo
p
r
o
p
er
ties
,
an
d
em
b
ed
d
in
g
p
ar
a
m
eter
s
in
ter
ac
t
to
d
eter
m
i
n
e
th
e
tar
g
et
f
r
a
m
es
.
T
h
is
s
tr
u
ct
u
r
ed
ap
p
r
o
ac
h
en
s
u
r
es
th
at
t
h
e
m
es
s
a
g
e
is
e
m
b
ed
d
ed
ef
f
ic
ien
tl
y
w
h
ile
m
ai
n
tai
n
i
n
g
v
id
eo
q
u
al
it
y
.
A
d
d
itio
n
all
y
,
t
h
e
d
u
al
f
r
a
m
e
s
elec
tio
n
m
o
d
e
r
an
d
o
m
o
r
s
eq
u
en
t
ial
p
r
o
v
id
es
f
le
x
ib
ilit
y
t
o
o
p
tim
ize
b
et
w
ee
n
r
o
b
u
s
t
n
ess
an
d
co
m
p
u
tatio
n
al
co
s
t
.
Fig
u
r
e
1
.
Ov
er
v
ie
w
o
f
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
Fig
u
r
e
2
.
Fra
m
e
s
elec
t
io
n
p
r
o
ce
s
s
i
n
s
u
b
b
an
d
-
b
ased
v
id
eo
s
teg
a
n
o
g
r
ap
h
y
s
y
s
te
m
T
h
e
d
etailed
p
r
o
ce
d
u
r
e
f
o
r
d
e
ter
m
i
n
in
g
w
h
ic
h
v
id
eo
f
r
a
m
e
s
w
ill
ca
r
r
y
t
h
e
e
m
b
ed
d
ed
w
a
ter
m
ar
k
i
s
f
o
r
m
alize
d
i
n
A
l
g
o
r
ith
m
1
,
w
h
ic
h
o
u
tl
in
e
s
th
e
s
tep
s
f
o
r
co
n
v
er
ti
n
g
th
e
m
e
s
s
a
g
e
to
b
in
ar
y
,
ca
lc
u
lati
n
g
e
m
b
ed
d
in
g
ca
p
ac
it
y
,
an
d
s
elec
tin
g
tar
g
et
f
r
a
m
e
s
b
ased
o
n
p
r
ed
ef
in
ed
o
r
r
an
d
o
m
m
o
d
es
.
Algorithm 1
.
G
enerate
t
arget
f
rames (
function to select target frames for embedding
)
Input: secretMessage, inputVideo, embeddingParams
Output: List of FrameIndex
1: Con
vert secretMessage to binary:
2: For each character mᵢ in secretMessage:
3: Convert mᵢ to 8
-
bit binary
→
bᵢ
4: Concatenate all bᵢ and append 32
-
bit EOF marker
5: Compute message length: L = 8n + 32
6: Read video properties from inputVideo:
7: Extract total
frames (N), height (H), and width (W)
8: Set embedding parameters from embeddingParams:
9: α = frame message ratio
10: r_HL, r_HH
,
r_LL, r_LH = sub
-
band allocation ratios
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t
E
l
C
o
n
tr
o
l
W
a
te
r
ma
r
kin
g
o
n
s
p
r
ea
d
mu
lt
i
-
fr
a
me
d
a
ta
vid
eo
u
s
in
g
d
is
crete
w
a
ve
let
tr
a
n
s
fo
r
m
…
(
I
lh
a
m
F
ir
ma
n
A
s
h
a
r
i
)
1479
11: Calculate sub
-
band capacities:
12: C_LL =
floor (
(H/2 × W/2) × α ×
r_LL)
13: C_LH =
floor (
(H/2 × W/2) × α × r_LH)
14: C_HL =
floor (
(H/2 × W/2) × α × r_HL)
15: C_HH =
floor (
(H/2 × W/2) × α × r_HH)
16: C_total = C_LL + C_LH + C_HL + C_HH
17: Compute required number of frames:
18: N_req =
ceil (
L / C_
total)
19: Select frame indices based on selection mode:
20: If mode = "SEQUENTIAL":
21: targetFrames = [0, 1,
.
.
.
, N_req
-
1]
22: Else if mode = "RANDOM":
23: targetFrames = random_sample
([0, 1,
.
.
.
, N
-
1], N_req, seed=s)
2
4: Ret
urn targetFrames
2
.
1
.
1
.
F
ra
m
e
-
ba
s
ed
re
la
t
iv
e
m
o
t
io
n v
ec
t
o
r
ca
lcula
t
io
n
I
n
t
h
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
,
th
e
f
r
a
m
e
-
b
ased
r
elat
iv
e
m
o
tio
n
v
ec
to
r
(
FR
M
V)
is
e
m
p
lo
y
ed
t
o
q
u
an
ti
f
y
te
m
p
o
r
al
s
tab
ilit
y
ac
r
o
s
s
co
n
s
ec
u
ti
v
e
v
id
eo
f
r
a
m
es
an
d
to
g
u
id
e
th
e
s
elec
t
io
n
o
f
e
m
b
ed
d
i
n
g
r
eg
io
n
s
.
Fo
r
ea
ch
b
lo
ck
in
th
e
c
u
r
r
en
t
f
r
a
m
e,
a
m
o
tio
n
v
ec
to
r
(
,
)
=
(
,
)
is
est
i
m
a
ted
b
y
b
lo
ck
m
atc
h
i
n
g
w
it
h
t
h
e
r
ef
er
en
ce
f
r
a
m
e,
an
d
t
h
e
r
elati
v
e
m
o
tio
n
v
ec
to
r
is
t
h
en
d
e
f
i
n
ed
as c
an
b
e
s
ee
n
in
(
1
)
.
(
,
)
=
(
,
)
−
−
1
(
,
)
(
1
)
T
h
e
o
v
er
all
FR
MV
v
alu
e
f
o
r
f
r
a
m
e
is
ca
lcu
lated
as
th
e
m
ea
n
m
a
g
n
i
tu
d
e
o
f
th
ese
r
elati
v
e
v
ec
to
r
s
,
ex
p
r
ess
ed
as c
an
b
e
s
ee
n
i
n
(
2
)
.
=
1
∑
∥
(
,
)
∥
=
1
∑
√
(
(
,
)
−
−
1
(
,
)
)
2
+
(
(
,
)
−
−
1
(
,
)
)
2
(
,
)
(
,
)
∈
(
2
)
w
h
er
e
d
en
o
tes
th
e
to
tal
n
u
m
b
er
o
f
b
lo
ck
s
.
A
lo
w
er
FR
MV
in
d
icate
s
h
ig
h
er
te
m
p
o
r
al
co
n
s
is
te
n
c
y
a
n
d
th
u
s
p
r
o
v
id
es
m
o
r
e
r
eliab
le
ca
n
d
id
ates f
o
r
w
ater
m
ar
k
e
m
b
ed
d
in
g
.
F
ig
u
r
e
3
i
l
lu
s
t
r
a
te
s
t
h
e
c
o
m
p
l
et
e
f
r
am
e
em
b
e
d
d
i
n
g
p
r
o
c
e
d
u
r
e
im
p
lem
en
t
e
d
in
th
e
p
r
o
p
o
s
e
d
s
u
b
b
an
d
-
b
a
s
e
d
v
i
d
e
o
w
a
t
e
r
m
a
r
k
i
n
g
s
y
s
tem
,
le
v
e
r
a
g
in
g
th
e
D
W
T
.
T
h
e
p
r
o
c
e
s
s
o
f
em
b
e
d
d
in
g
s
e
c
r
et
b
i
ts
in
t
o
s
e
l
e
c
te
d
v
i
d
e
o
f
r
am
es
u
s
in
g
D
W
T
-
b
a
s
e
d
s
u
b
b
a
n
d
m
o
d
if
i
ca
t
i
o
n
is
o
u
tl
in
e
d
in
A
lg
o
r
i
th
m
2
,
w
h
ich
d
e
t
ai
l
s
th
e
t
r
a
n
s
f
o
r
m
a
t
i
o
n
,
b
it
d
i
s
t
r
i
b
u
ti
o
n
,
c
o
ef
f
i
c
i
en
t
m
o
d
if
ic
a
t
i
o
n
,
an
d
r
e
c
o
n
s
t
r
u
ct
i
o
n
s
te
p
s
f
o
r
g
en
e
r
at
in
g
s
t
eg
o
v
i
d
e
o
c
o
n
t
en
t
.
Fig
u
r
e
3
.
Fra
m
e
e
m
b
ed
d
in
g
p
r
o
ce
d
u
r
e
u
s
in
g
DW
T
in
s
u
b
b
an
d
-
b
ased
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l
.
23
,
No
.
6
,
Dec
em
b
er
20
25
:
1
4
7
6
-
1494
1480
Algorithm 2
.
E
mbed
message
in
frames
(
function to
embed secret bits in target frames
)
Input: targetFrames,
videoFrames, messageBits, embeddingParams
Output: stegoVideo
1: For each frame fᵢ in videoFrames:
2: If fᵢ
∈
targetFrames then
3: Convert fᵢ from BGR to YCrCb
4: Extract Y,
Cr, and Cb channels
5: Apply 1
-
level DWT on Y →
obtain LL, LH, HL, HH
6
: Distribute messageBits based on subband capacity:
7
: Bits_LL = messageBits[0 : C_LL]
8
: Bits_LH = messageBits[C_LL : C_LL + C_LH]
9
:
Bits_HL = messageBits[C_LL + C_LH : C_LL + C_LH + C
_HL]
1
0
: Flatten LL matrix to 1D array
1
1
: For each coefficient cᵢ in LL:
1
2
: I =
floor (
cᵢ × 255)
1
3
: F = (cᵢ × 255)
-
I
1
4
: I_new = (I AND 0xFE)
OR b
ₖ
// Embed bit b
ₖ
1
5
: cᵢ_new = (I_n
ew + F) / 255
1
6
: end for
17
: Reshape modified LL to 2D
18
: Reconstruct Y_stego using
IDWT (
LL_new, LH, HL, HH)
19
: Convert Y_stego × 255 to uint8 and clip to [0, 255]
2
0
: Merge with original Cr and Cb
2
1
: Convert YCrCb back to BGR
2
2
: Write stego frame to output
2
3
: else
2
4
: Write original frame to output
2
5
: end if
2
6
: end for
27
: Return stegoVideo
2
.
2
.
Dec
o
de
T
h
e
d
ec
o
d
in
g
p
h
ase
b
eg
i
n
s
b
y
id
en
ti
f
y
i
n
g
an
d
v
alid
ati
n
g
f
r
a
m
es
w
i
th
in
t
h
e
s
teg
o
v
id
eo
th
at
p
o
ten
tiall
y
co
n
tain
h
id
d
en
i
n
f
o
r
m
atio
n
as
illu
s
tr
ated
in
F
ig
u
r
e
4
.
T
h
is
p
r
o
ce
s
s
in
v
o
l
v
es
a
n
al
y
zi
n
g
v
i
d
eo
p
r
o
p
er
ties
an
d
s
elec
ti
n
g
s
p
ec
i
f
ic
f
r
a
m
e
s
eith
er
th
r
o
u
g
h
p
r
ed
ef
in
ed
p
ar
a
m
e
te
r
s
o
r
b
y
au
to
m
a
ticall
y
d
ete
ctin
g
u
n
iq
u
e
f
r
a
m
e
f
ea
t
u
r
es
.
On
ce
t
h
e
ca
n
d
id
ate
f
r
a
m
es
ar
e
id
en
ti
f
ied
,
th
e
y
ar
e
v
alid
ated
to
en
s
u
r
e
o
n
l
y
t
h
e
c
o
r
r
ec
t
tar
g
et
f
r
a
m
es
ar
e
u
s
ed
f
o
r
ex
tr
ac
ti
n
g
t
h
e
h
id
d
en
m
es
s
a
g
e
.
Fig
u
r
e
4
.
T
a
r
g
et
f
r
a
m
e
v
alid
at
io
n
p
r
o
ce
s
s
in
s
te
g
o
v
id
eo
an
a
l
y
s
i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t
E
l
C
o
n
tr
o
l
W
a
te
r
ma
r
kin
g
o
n
s
p
r
ea
d
mu
lt
i
-
fr
a
me
d
a
ta
vid
eo
u
s
in
g
d
is
crete
w
a
ve
let
tr
a
n
s
fo
r
m
…
(
I
lh
a
m
F
ir
ma
n
A
s
h
a
r
i
)
1481
T
h
e
p
r
o
ce
d
u
r
e
f
o
r
id
en
tify
i
n
g
w
h
ic
h
f
r
a
m
e
s
i
n
a
s
teg
o
v
id
eo
ar
e
elig
ib
le
f
o
r
d
ata
ex
tr
ac
tio
n
is
d
etailed
in
A
l
g
o
r
ith
m
3
,
w
h
ich
o
u
tli
n
es
v
alid
atio
n
s
tep
s
b
ased
o
n
p
r
ed
ef
i
n
ed
m
etad
ata
o
r
au
to
-
d
etec
t
io
n
u
s
i
n
g
s
tati
s
tical
an
al
y
s
is
o
f
f
r
a
m
e
f
e
a
tu
r
es
.
Algorithm 3
.
I
dentify
valid
target
frames (function to determine valid frames for
extraction)
Input: stegoVideo, knownParams (optional)
Output: List of Valid Target Frames
1: Read stegoVideo
as sequence of frames:
2: V = {F
₁
, F
₂
,
.
.
.
,
F
ₙ
}
3
: Extract video properties:
4:
-
N = total frames = duration × frame rate
5:
-
W, H = frame width and height
6: Load embedding parameters (if known):
7:
-
α = frame_message_ratio
8:
-
R = {r_LL, r_LH, r_HL, r_HH}, with
∑
rᵢ = 1
9: Det
ermine selection mode:
10: If mode = "Predefined":
11: T = {t
₁
, t
₂
,
.
.
.
, t_k} from metadata or input
12: For each tᵢ in T:
13: If 0
≤
tᵢ < N and rᵢ > 0:
14:
Add tᵢ to validTargetFrames
15: Else if mode =
"Auto
-
Detect":
16: For each frame Fᵢ in V:
17: Compute statistical features (e
.
g
.
, DWT energy)
18: Compute global mean μ and std deviation σ
19: For each Fᵢ:
20: If
Feature (
Fᵢ) >
μ +
k
σ:
21:
Add i to validTargetFrames
22: Return validTargetFrames
On
ce
v
alid
tar
g
et
f
r
a
m
e
s
h
a
v
e
b
ee
n
id
en
tif
ied
,
th
e
s
y
s
te
m
p
r
o
ce
ed
s
w
it
h
ex
tr
ac
ti
n
g
th
e
e
m
b
ed
d
ed
m
es
s
ag
e
f
r
o
m
ea
ch
f
r
a
m
e
b
y
r
ev
er
s
i
n
g
t
h
e
e
m
b
ed
d
in
g
o
p
er
atio
n
s
.
Fi
g
u
r
e
5
illu
s
tr
ates
t
h
e
b
it
ex
tr
ac
tio
n
p
r
o
ce
s
s
p
er
f
o
r
m
ed
o
n
tar
g
et
f
r
a
m
es
w
i
th
i
n
th
e
p
r
o
p
o
s
ed
DW
T
-
b
ased
v
id
eo
w
ater
m
ar
k
i
n
g
f
r
a
m
e
w
o
r
k
.
Fig
u
r
e
5
.
B
it
ex
tr
ac
tio
n
p
r
o
ce
s
s
f
r
o
m
tar
g
et
f
r
a
m
es
i
n
DW
T
-
b
ased
v
id
eo
w
ater
m
ar
k
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l
.
23
,
No
.
6
,
Dec
em
b
er
20
25
:
1
4
7
6
-
1494
1482
T
h
e
p
r
o
ce
s
s
o
f
r
etr
iev
i
n
g
e
m
b
ed
d
ed
b
its
f
r
o
m
t
h
e
s
elec
ted
ta
r
g
et
f
r
a
m
e
s
is
p
r
ese
n
ted
i
n
A
l
g
o
r
ith
m
4
,
w
h
ic
h
d
escr
ib
es
th
e
s
tep
s
f
o
r
lu
m
in
a
n
ce
e
x
tr
ac
tio
n
,
DW
T
d
ec
o
m
p
o
s
i
tio
n
,
an
d
L
SB
-
b
ased
b
it
r
ec
o
v
er
y
f
r
o
m
th
e
L
L
,
L
H,
a
n
d
HL
s
u
b
b
an
d
s
.
Algorithm 4
.
E
xtract
bits
from
frames (function to recover embedded bits)
Input: targetFrame
s, stegoVideo
Output: Bitstream B
1: Initialize empty bit collector:
2: B ←
∅
3: For each frame index tᵢ in targetFrames:
4: Read frame F
ₜ
ᵢ
fr
o
m
stegoVideo
5: Convert F
ₜ
ᵢ to YCr
Cb color space
6: Extract Y channel: Y =
ψ(
F
ₜ
ᵢ)
7: Apply 2D Discrete Wavelet Transform:
8: [LL, LH, HL, HH] = DWT
(Y)
9: For each subband S
∈
{LL, LH, HL}:
10: Flatten subband: S_flat = fl
atten(S)
11: For each coefficient c in S_flat:
12: I = floor(c × 255)
13: b = I mod 2 // Extract LSB
14: Append b to
B
15: Return bitstream B
Fig
u
r
e
6
d
e
p
icts
th
e
m
es
s
ag
e
r
ec
o
n
s
tr
u
ctio
n
p
r
o
ce
s
s
f
r
o
m
th
e
ex
tr
ac
ted
b
its
tr
ea
m
in
t
h
e
p
r
o
p
o
s
ed
DW
T
-
b
ased
w
ater
m
ar
k
in
g
s
y
s
te
m
.
Fig
u
r
e
6
.
Me
s
s
ag
e
r
ec
o
n
s
tr
u
ct
io
n
f
r
o
m
e
x
tr
ac
ted
b
its
tr
ea
m
i
n
DW
T
-
b
ased
w
ater
m
ar
k
i
n
g
s
y
s
te
m
Th
e
f
in
al
p
h
a
s
e
o
f
t
h
e
d
ec
o
d
in
g
p
r
o
ce
s
s
i
s
d
etailed
in
A
l
g
o
r
ith
m
5
,
w
h
ich
o
u
tli
n
es
t
h
e
s
t
ep
s
f
o
r
lo
ca
tin
g
th
e
E
OF
m
ar
k
er
,
v
alid
ati
n
g
a
n
d
tr
u
n
ca
t
in
g
t
h
e
b
its
tr
ea
m
,
ali
g
n
in
g
to
b
y
te
b
o
u
n
d
ar
ies,
an
d
co
n
v
er
ti
n
g
b
i
n
ar
y
s
eg
m
e
n
ts
i
n
to
A
S
C
I
I
ch
ar
ac
ter
s
to
r
ec
o
n
s
tr
uc
t t
h
e
h
id
d
en
m
ess
a
g
e
.
Algorithm 5
.
D
ecode
bitstream
to
message (function to recover hidden message from
bitstream)
Input: Bitstream B = {b
₁
, b
₂
,
.
.
.
, b
ₙ
}
Output: Decoded ASCII Message M or Error
1: Define EOF marker:
2: ε =
"1111111111111111111111
1111111110" (32 bits)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t
E
l
C
o
n
tr
o
l
W
a
te
r
ma
r
kin
g
o
n
s
p
r
ea
d
mu
lt
i
-
fr
a
me
d
a
ta
vid
eo
u
s
in
g
d
is
crete
w
a
ve
let
tr
a
n
s
fo
r
m
…
(
I
lh
a
m
F
ir
ma
n
A
s
h
a
r
i
)
1483
3: Locate EOF marker in B:
4: For index p = 0 to n
-
32:
5: If B[p : p + 32] == ε:
6: EOF_found = True
7: Break
8: If EOF_found and p ≤ n
-
32 then:
9: Truncate bit
stream before EOF:
10:
B
’
= {b
₁
,
.
.
.
, b_{p
-
1}}
11: Align to byte boundary:
12: B
’’
= {b
₁
,
.
.
.
, b_{8 ×
⌊
p/8
⌋
}}
13: Convert each 8
-
bit segment to character:
14: For j = 0 to (|B
’’
| / 8)
-
1:
15: S_j
= {b_{8j+1},
.
.
.
, b_{8j+8}}
16:
c_j = ∑_{k=0}^{7} b_{8j+k+1} × 2^{7
-
k}
17: If c_j
∈
[32, 126]:
18: Append chr(c_j) to M
19: Return M // Final decoded message
20: Else:
21: Return Error: Corrupted or incompl
ete data
2
.
3
.
P
er
f
o
rm
a
nce
m
et
rics
T
o
ev
alu
ate
t
h
e
s
y
s
te
m
p
er
f
o
r
m
an
ce
q
u
a
n
ti
tativ
e
l
y
,
s
e
v
er
a
l
s
tan
d
ar
d
m
etr
ics
w
er
e
ap
p
li
ed
.
T
h
ese
m
etr
ics
m
ea
s
u
r
e
i
m
p
er
ce
p
tib
il
it
y
,
r
o
b
u
s
t
n
es
s
,
a
n
d
co
m
p
u
tati
o
n
al
ef
f
ic
ien
c
y
.
T
o
g
eth
er
,
t
h
e
y
p
r
o
v
id
e
a
r
eliab
le
ass
es
s
m
en
t o
f
th
e
p
r
o
p
o
s
ed
w
ater
m
ar
k
i
n
g
f
r
a
m
e
w
o
r
k
.
2
.
3
.
1
.
P
ea
k
s
ig
na
l
-
to
-
no
is
e
ra
t
io
Me
asu
r
es t
h
e
d
is
to
r
tio
n
b
et
w
e
en
th
e
o
r
ig
in
a
l
an
d
s
teg
o
f
r
a
m
e
[
3
4
]
,
[
3
5
]
.
T
h
e
P
SNR
ca
n
b
e
ca
lcu
lated
u
s
i
n
g
th
e
f
o
llo
w
i
n
g
f
o
r
m
u
la,
as
ca
n
b
e
s
ee
n
i
n
(
3
)
.
A
h
ig
h
e
r
P
SNR
v
alu
e
(
t
y
p
icall
y
>
4
0
d
B
)
in
d
icate
s
b
etter
f
id
elit
y
an
d
m
in
i
m
al
v
i
s
u
al
d
e
g
r
ad
atio
n
.
=
2
0
.
10
(
√
)
(
3
)
w
h
er
e:
=2
5
5
,
m
ax
i
m
u
m
p
o
s
s
ib
le
p
ix
el
v
alu
e
f
o
r
8
-
b
it i
m
ag
e
s
MSE
=
1
.
∑
∑
[
(
,
]
−
(
,
)
]
2
=
1
=
1
,
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
MSE
)
b
et
w
ee
n
t
h
e
o
r
ig
i
n
al
i
m
ag
e
an
d
th
e
s
te
g
o
i
m
a
g
e
.
2
.
3
.
2
.
S
t
ruct
ura
l si
m
ila
rit
y
ind
ex
m
ea
s
ure
Ass
es
s
es
t
h
e
v
is
u
al
s
i
m
i
lar
it
y
b
et
w
ee
n
th
e
o
r
i
g
in
al
a
n
d
s
te
g
o
f
r
a
m
e,
co
n
s
id
er
in
g
s
tr
u
ct
u
r
al
i
n
f
o
r
m
atio
n
(
lu
m
in
a
n
ce
,
co
n
tr
ast,
tex
t
u
r
e)
[3
4
]
,
[
3
6
]
.
T
h
e
co
m
p
u
tat
io
n
o
f
SS
I
M
is
b
ased
o
n
a
co
m
p
ar
is
o
n
o
f
lo
ca
l
p
atter
n
s
o
f
p
ix
el
in
ten
s
i
tie
s
th
at
h
a
v
e
b
ee
n
n
o
r
m
alize
d
f
o
r
lu
m
i
n
a
n
c
e
an
d
co
n
tr
ast
.
T
h
e
m
at
h
e
m
a
t
ical
f
o
r
m
u
la
tio
n
o
f
SS
I
M
is
p
r
esen
ted
i
n
(
4
)
.
(
,
)
=
(
2
+
1
)
(
2
+
2
)
(
2
+
2
+
1
)
(
2
+
2
+
2
)
(
4
)
w
h
er
e:
ar
e
th
e
m
ea
n
in
ten
s
ities
2
,
2
ar
e
th
e
v
ar
ia
n
ce
s
is
th
e
co
v
ar
ia
n
ce
1
,
2
ar
e
s
m
all
co
n
s
tan
ts
to
av
o
id
d
iv
is
io
n
b
y
ze
r
o
SS
I
M
v
al
u
es r
a
n
g
e
f
r
o
m
0
to
1
,
w
it
h
1
in
d
icatin
g
p
er
f
ec
t
s
i
m
ilar
it
y
2
.
3
.
3
.
B
it
er
r
o
r
ra
t
e
C
alcu
lated
as
th
e
r
atio
o
f
d
if
f
er
in
g
b
its
b
et
w
ee
n
t
h
e
o
r
ig
in
al
an
d
ex
tr
ac
ted
w
ater
m
ar
k
.
B
E
R
q
u
an
ti
f
ie
s
th
e
p
r
o
p
o
r
tio
n
o
f
b
its
th
at
w
er
e
in
co
r
r
ec
tl
y
d
ec
o
d
ed
[
3
7
]
.
L
et
b
e
th
e
i
-
th
b
it
o
f
th
e
o
r
ig
i
n
al
m
es
s
ag
e
a
n
d
̂
b
e
th
e
co
r
r
esp
o
n
d
in
g
e
x
tr
ac
ted
b
it
.
T
h
e
f
o
r
m
al
d
e
f
in
itio
n
o
f
B
E
R
is
g
iv
e
n
i
n
(
5
)
.
=
1
∑
1
(
≠
=
1
̂
)
(
5
)
w
h
er
e
is
t
h
e
to
tal
n
u
m
b
er
o
f
b
i
ts
,
an
d
1(
⋅
)
is
th
e
i
n
d
icato
r
f
u
n
ctio
n
.
A
B
E
R
o
f
0
in
d
icate
s
p
e
r
f
ec
t e
x
tr
ac
tio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l
.
23
,
No
.
6
,
Dec
em
b
er
20
25
:
1
4
7
6
-
1494
1484
2
.
3
.
4
.
E
nco
din
g
/
deco
din
g
t
i
me
T
h
e
to
tal
p
r
o
ce
s
s
in
g
ti
m
e
is
r
e
co
r
d
ed
t
o
ev
alu
ate
alg
o
r
it
h
m
ic
ef
f
icien
c
y
[
3
8
]
.
T
h
e
to
tal
p
r
o
ce
s
s
i
n
g
ti
m
e
is
co
m
p
u
ted
b
y
r
ec
o
r
d
in
g
t
h
e
ti
m
e
s
ta
m
p
s
at
t
h
e
b
eg
i
n
n
i
n
g
an
d
en
d
o
f
th
e
o
p
er
atio
n
.
T
h
i
s
is
m
at
h
e
m
atica
ll
y
ex
p
r
ess
ed
i
n
(
6
)
.
=
−
(
6
)
w
h
er
e
an
d
d
en
o
te
ti
m
esta
m
p
s
i
n
s
ec
o
n
d
s
o
r
m
illi
s
ec
o
n
d
s
c
ap
tu
r
ed
b
ef
o
r
e
an
d
a
f
ter
t
h
e
o
p
er
atio
n
.
R
u
n
ti
m
e
is
t
y
p
icall
y
r
ep
o
r
ted
i
n
m
i
llis
ec
o
n
d
s
(
m
s
)
.
2
.
3
.
5
.
At
t
a
ck
s
i
m
ula
t
io
ns
T
o
v
alid
ate
w
ater
m
ar
k
r
esi
lien
ce
,
f
o
u
r
d
is
ti
n
ct
attac
k
m
o
d
els
w
er
e
i
m
p
le
m
e
n
ted
:
a.
No
is
e
atta
ck
(
G
a
u
s
s
ian
n
o
is
e
i
n
j
ec
tio
n
)
I
n
t
h
is
a
ttack
,
ad
d
iti
v
e
w
h
i
te
G
au
s
s
ian
n
o
i
s
e
(
A
W
GN)
i
s
i
n
tr
o
d
u
ce
d
to
ea
ch
p
i
x
el
o
f
e
v
er
y
v
i
d
eo
f
r
a
m
e
to
s
i
m
u
late
s
e
n
s
o
r
n
o
i
s
e
o
r
ch
an
n
el
i
n
ter
f
er
en
ce
[
3
9
]
.
T
h
is
p
r
o
ce
s
s
is
m
at
h
e
m
atica
l
l
y
d
ef
in
ed
in
(
7
)
:
′
(
,
,
)
=
(
,
,
)
+
(
0
,
2
)
(
7
)
w
h
er
e:
′
(
,
,
)
is
th
e
o
r
ig
i
n
al
p
ix
e
l a
t c
o
o
r
d
in
ate
(
,
)
an
d
co
lo
r
ch
an
n
el
(
0
,
2
)
d
en
o
tes ze
r
o
-
m
ea
n
Ga
u
s
s
ian
n
o
is
e
w
it
h
s
ta
n
d
ar
d
d
ev
iatio
n
σ
,
C
lip
p
in
g
is
ap
p
lied
to
en
s
u
r
e
p
ix
el
v
a
lu
e
s
r
e
m
ai
n
w
it
h
i
n
[
0
,
2
5
5
]
.
b.
R
esize
a
ttac
k
T
o
em
u
late
r
eso
l
u
tio
n
lo
s
s
a
n
d
in
ter
p
o
latio
n
n
o
is
e,
ea
c
h
f
r
am
e
w
as
d
o
w
n
s
a
m
p
led
b
y
a
f
a
cto
r
an
d
th
en
r
es
to
r
ed
to
its
o
r
ig
in
al
s
iz
e
[
4
0
]
,
[
4
1
]
.
T
h
e
en
tire
p
r
o
ce
s
s
ca
n
b
e
f
o
r
m
a
ll
y
r
ep
r
esen
ted
a
s
f
o
llo
w
s
,
as
s
h
o
w
n
in
(
8
)
an
d
(
9
)
.
−
Do
w
n
s
ca
li
n
g
:
=
(
,
.
,
.
)
(
8
)
−
Up
s
ca
lin
g
:
=
(
,
,
)
(
9
)
w
h
er
e:
=
0
.
5
(
.
3
.
,
50%
)
,
d
en
o
te
th
e
o
r
ig
i
n
al
h
ei
g
h
t a
n
d
w
id
t
h
,
B
ilin
ea
r
i
n
ter
p
o
latio
n
w
a
s
u
s
ed
f
o
r
b
o
th
s
tep
s
.
c.
B
lu
r
attac
k
(
lo
w
-
p
as
s
f
ilter
in
g
)
B
lu
r
attac
k
s
,
i
n
te
n
tio
n
al
l
y
o
b
s
cu
r
e
i
m
a
g
e
d
etails,
w
h
ic
h
in
t
u
r
n
a
f
f
ec
ts
t
h
e
w
ater
m
ar
k
’
s
v
i
s
ib
ilit
y
a
n
d
in
te
g
r
it
y
[
4
2
]
.
B
lu
r
r
in
g
s
i
m
u
la
tes
o
p
tical
o
r
co
m
p
r
ess
io
n
-
in
d
u
ce
d
lo
w
-
p
ass
f
ilter
in
g
[
4
3
]
.
A
Gau
s
s
ia
n
b
lu
r
w
it
h
a
5
×5
k
er
n
el
w
a
s
ap
p
l
ied
to
ea
ch
f
r
a
m
e
.
T
h
e
b
lu
r
r
in
g
o
p
er
ati
o
n
is
m
a
th
e
m
atica
l
l
y
r
ep
r
esen
t
ed
in
(
10
)
.
′
(
,
)
=
∑
∑
(
,
)
.
(
+
,
+
)
2
=
−
2
2
=
−
1
(
10
)
T
h
e
w
ei
g
h
tin
g
f
u
n
ct
io
n
i
n
(
8
)
co
r
r
esp
o
n
d
s
to
th
e
n
o
r
m
alize
d
Gau
s
s
ia
n
k
er
n
el
,
w
h
ic
h
d
ete
r
m
i
n
es
t
h
e
in
f
lu
e
n
ce
o
f
n
ei
g
h
b
o
r
in
g
p
ix
el
s
d
u
r
in
g
t
h
e
b
lu
r
r
in
g
p
r
o
ce
s
s
.
T
h
is
k
er
n
el
i
s
s
y
m
m
etr
ic
a
n
d
ce
n
ter
ed
,
en
s
u
r
in
g
is
o
tr
o
p
ic
s
m
o
o
th
i
n
g
o
f
t
h
e
i
m
a
g
e
.
T
h
e
s
h
ap
e
an
d
in
te
n
s
it
y
d
i
s
tr
ib
u
tio
n
o
f
th
e
Ga
u
s
s
ian
k
er
n
el
ar
e
g
o
v
er
n
ed
b
y
th
e
s
ta
n
d
ar
d
d
ev
iatio
n
σ
.
T
h
e
m
at
h
e
m
a
tical
f
o
r
m
u
latio
n
o
f
t
h
e
Gau
s
s
ia
n
k
er
n
el
i
s
p
r
esen
te
d
in
(
9
)
.
(
,
)
=
1
2
2
−
2
+
2
2
2
(
11
)
w
it
h
k
er
n
el
s
ize
5
×5
an
d
d
e
f
au
lt
=1
.
0
.
T
h
is
o
p
er
atio
n
r
e
m
o
v
e
s
h
ig
h
-
f
r
eq
u
en
c
y
co
m
p
o
n
en
t
s
,
p
o
ten
tiall
y
af
f
ec
tin
g
t
h
e
s
u
b
b
an
d
co
ef
f
ici
en
ts
i
n
DW
T
w
h
er
e
w
ater
m
ar
k
b
i
ts
ar
e
e
m
b
ed
d
ed
.
d.
C
o
m
p
r
ess
io
n
attac
k
T
h
is
p
r
o
ce
s
s
o
f
co
m
p
r
es
s
io
n
l
ea
d
s
to
d
ata
lo
s
s
,
w
h
ich
ca
n
c
o
n
s
eq
u
e
n
tl
y
ca
u
s
e
t
h
e
w
ater
m
ar
k
to
b
e
co
r
r
u
p
ted
o
r
co
m
p
letel
y
lo
s
t
[
4
4
]
.
T
h
e
v
id
eo
w
a
s
r
e
-
e
n
co
d
ed
u
s
in
g
th
e
m
o
tio
n
j
o
in
t
p
h
o
to
g
r
ap
h
ic
ex
p
er
ts
’
g
r
o
u
p
(
MJ
P
G)
co
d
ec
w
it
h
r
ed
u
ce
d
q
u
alit
y
s
etti
n
g
s
.
T
h
is
attac
k
s
i
m
u
late
s
lo
s
s
y
tr
an
s
m
is
s
io
n
o
r
s
to
r
ag
e
s
ce
n
ar
io
s
[
4
5
]
,
[
4
6
]
.
M
J
P
G
co
m
p
r
e
s
s
e
s
ea
ch
f
r
a
m
e
in
d
ep
en
d
en
tl
y
as
a
J
P
E
G
im
a
g
e,
ap
p
ly
in
g
d
is
cr
ete
co
s
i
n
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t
E
l
C
o
n
tr
o
l
W
a
te
r
ma
r
kin
g
o
n
s
p
r
ea
d
mu
lt
i
-
fr
a
me
d
a
ta
vid
eo
u
s
in
g
d
is
crete
w
a
ve
let
tr
a
n
s
fo
r
m
…
(
I
lh
a
m
F
ir
ma
n
A
s
h
a
r
i
)
1485
tr
an
s
f
o
r
m
(
D
C
T
)
-
b
ased
q
u
an
ti
za
tio
n
w
h
ich
ca
n
d
is
to
r
t
t
h
e
e
m
b
ed
d
ed
w
ater
m
ar
k
.
T
h
e
co
m
p
r
e
s
s
io
n
p
r
o
ce
s
s
is
m
at
h
e
m
a
ticall
y
ex
p
r
ess
ed
i
n
(
1
2
)
.
=
(
,
)
(
1
2
)
w
h
er
e
[
1
,
100
]
an
d
in
t
h
i
s
s
t
u
d
y
a
d
ef
au
lt
q
u
a
lit
y
o
f
=5
0
w
as
u
s
ed
.
T
h
is
attac
k
is
e
s
p
ec
iall
y
r
ele
v
an
t
to
ass
es
s
r
o
b
u
s
t
n
ess
a
g
ai
n
s
t
f
o
r
m
at
co
n
v
er
s
io
n
an
d
b
an
d
w
id
th
-
r
ed
u
ctio
n
tec
h
n
iq
u
es
.
e.
C
r
o
p
p
in
g
attac
k
C
r
o
p
p
in
g
attac
k
is
a
g
eo
m
e
tr
ic
m
an
ip
u
latio
n
w
h
er
e
p
ar
ts
o
f
a
v
id
eo
f
r
a
m
e
ar
e
r
e
m
o
v
ed
,
w
h
ic
h
m
a
y
d
is
ca
r
d
em
b
ed
d
ed
w
ater
m
ar
k
in
f
o
r
m
at
io
n
.
L
et
′
(
,
)
d
en
o
te
a
w
ater
m
ar
k
ed
f
r
a
m
e
o
f
s
ize
×
,
an
d
(
,
)
∈
{
0
,
1
}
a
b
in
ar
y
m
a
s
k
r
ep
r
esen
tin
g
p
r
eser
v
ed
r
eg
io
n
s
.
T
h
e
cr
o
p
p
ed
f
r
a
m
e
i
s
d
ef
i
n
ed
as (
1
3
)
.
̃
(
,
)
=
′
(
,
)
.
(
,
)
(
1
3
)
T
h
e
cr
o
p
p
in
g
r
atio
,
in
d
icatin
g
th
e
p
r
o
p
o
r
tio
n
o
f
r
em
o
v
ed
p
ix
els,
is
d
ef
i
n
ed
as (
1
4
)
.
=
1
−
∥
∥
0
×
(
1
4
)
T
o
m
ea
s
u
r
e
s
i
m
ilar
it
y
b
et
w
e
en
th
e
ex
tr
ac
ted
w
a
ter
m
ar
k
̂
an
d
th
e
o
r
ig
in
al
,
th
e
n
o
r
m
a
lized
co
r
r
elatio
n
(
NC
)
is
u
s
ed
i
n
(
1
5
)
.
=
∑
(
−
̅
)
(
̂
−
̂
̅
)
=
1
√
∑
(
−
̅
)
2
√
∑
(
̂
−
̅
̂
)
2
=
1
=
1
(
1
5
)
w
h
er
e
is
t
h
e
w
ater
m
ar
k
len
g
t
h
,
an
d
̅
,
̅
̂
ar
e
th
e
m
ea
n
v
al
u
es o
f
th
e
o
r
ig
in
al
a
n
d
ex
tr
ac
ted
w
a
ter
m
ar
k
s
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
p
r
esen
t
s
a
co
m
p
r
eh
en
s
i
v
e
an
al
y
s
i
s
o
f
t
h
e
ex
p
er
i
m
en
tal
o
u
tco
m
es
o
b
tain
ed
f
r
o
m
th
e
h
y
b
r
id
s
u
b
b
an
d
-
b
ased
v
id
eo
w
at
er
m
ar
k
i
n
g
f
r
a
m
e
w
o
r
k
ap
p
lied
to
m
u
lti
-
f
r
a
m
e
e
m
b
ed
d
i
n
g
.
T
h
e
p
r
o
p
o
s
ed
w
ater
m
ar
k
i
n
g
alg
o
r
it
h
m
w
as
i
m
p
le
m
e
n
ted
u
s
i
n
g
P
y
t
h
o
n
p
r
o
g
r
a
m
m
i
n
g
lan
g
u
a
g
e
w
it
h
i
n
th
e
Vis
u
al
S
tu
d
io
C
o
d
e
en
v
ir
o
n
m
e
nt
.
All
e
x
p
er
i
m
e
n
ts
w
er
e
e
x
ec
u
ted
o
n
a
n
A
p
p
le
M
ac
B
o
o
k
P
r
o
eq
u
ip
p
e
d
w
it
h
a
n
M1
P
r
o
p
r
o
ce
s
s
o
r
,
1
6
GB
R
A
M,
an
d
5
1
2
GB
SS
D
s
to
r
ag
e,
en
s
u
r
i
n
g
s
u
f
f
icien
t
co
m
p
u
tatio
n
al
ca
p
ac
it
y
f
o
r
h
an
d
lin
g
v
id
eo
d
ata
.
T
o
ev
alu
a
te
r
o
b
u
s
t
n
ess
a
n
d
i
m
p
e
r
ce
p
tib
ilit
y
,
f
o
u
r
u
n
co
m
p
r
ess
e
d
v
id
eo
s
eq
u
en
ce
s
w
it
h
a
r
es
o
lu
tio
n
o
f
6
4
0
×3
6
0
p
ix
els
w
er
e
s
elec
ted
as
h
o
s
t
m
ed
ia
.
T
h
e
s
ec
r
et
m
es
s
ag
e
s
w
er
e
g
en
er
ated
as
A
S
C
I
I
tex
t
200
w
o
r
d
s
,
r
ep
r
esen
ti
n
g
p
r
ac
tical
p
ay
lo
ad
s
izes
f
o
r
co
p
y
r
i
g
h
t
o
r
au
t
h
e
n
ticatio
n
i
n
f
o
r
m
at
io
n
.
Du
r
i
n
g
th
e
e
v
alu
a
t
io
n
,
b
o
th
o
b
j
ec
tiv
e
m
etr
ics
s
u
ch
a
s
P
SNR
,
SS
I
M,
an
d
NC
,
a
s
w
ell
as
B
E
R
,
w
er
e
u
s
ed
to
as
s
es
s
t
h
e
v
is
u
al
q
u
a
lit
y
a
n
d
r
o
b
u
s
t
n
es
s
o
f
th
e
e
x
tr
ac
ted
w
ater
m
ar
k
.
T
h
is
e
x
p
er
i
m
e
n
tal
co
n
f
ig
u
r
atio
n
p
r
o
v
id
es
s
u
f
f
icien
t
d
etail
to
allo
w
r
ep
licatio
n
b
y
o
th
er
r
esear
ch
er
s
w
o
r
k
i
n
g
o
n
v
i
d
eo
w
ater
m
ar
k
i
n
g
s
y
s
te
m
s
.
Fig
u
r
e
7
(
a)
s
h
o
w
s
t
h
e
i
n
p
u
t
v
id
eo
(
Vid
eo
1
.
av
i
)
w
it
h
a
r
eso
lu
t
io
n
o
f
6
4
0
×3
6
0
,
w
h
ile
Fi
g
u
r
e
7
(
b
)
p
r
esen
ts
(
Vid
eo
2
.
av
i
)
w
it
h
t
h
e
s
a
m
e
r
es
o
lu
tio
n
.
B
o
th
v
id
eo
s
w
er
e
u
s
ed
as
h
o
s
t
m
ed
ia
f
o
r
em
b
ed
d
i
n
g
e
x
p
er
i
m
e
n
ts
.
T
h
e
r
esu
lt
s
ar
e
d
is
cu
s
s
e
d
in
ter
m
s
o
f
i
m
p
er
ce
p
tib
ilit
y
,
co
m
p
u
tatio
n
al
ef
f
icie
n
c
y
,
r
o
b
u
s
t
n
es
s
,
an
d
ca
p
ac
it
y
.
(
a)
(
b
)
Fig
u
r
e
7
.
I
n
p
u
t
v
id
eo
:
(
a)
V
id
eo
1
.
av
i
,
r
eso
lu
tio
n
: 6
4
0
×
360
,
s
i
ze
: 3
.
4
MB
an
d
(
b
)
V
id
e
o
2
.
av
i
,
r
eso
lu
tio
n
:
640
×
3
6
0
,
s
ize:
4
.
1
MB
3
.
1
.
E
v
a
lua
t
i
o
n o
f
s
t
eg
o
v
ide
o
qu
a
lity
us
ing
s
ub
-
ba
n
d v
a
ria
t
io
ns
T
h
e
s
teg
o
-
v
id
eo
q
u
al
it
y
w
as
e
v
alu
a
ted
ac
r
o
s
s
s
u
b
-
b
an
d
r
atio
s
etti
n
g
s
u
s
i
n
g
B
E
R
,
P
SNR
,
a
n
d
SS
I
M
o
n
Vid
eo
1
an
d
Vid
e
o
2
.
LL
-
o
n
l
y
e
m
b
ed
d
in
g
y
ield
ed
th
e
b
est
r
o
b
u
s
tn
e
s
s
–
i
m
p
er
ce
p
tib
ilit
y
tr
ad
e
-
o
f
f
:
Vid
eo
1
Evaluation Warning : The document was created with Spire.PDF for Python.