www.ietdl.

org
Published in IET Image Processing
Received on 15th August 2010
Revised on 28th January 2011
doi: 10.1049/iet-ipr.2010.0347

ISSN 1751-9659

Wavelet-based oblivious image watermarking

scheme using genetic algorithm
K. Ramanjaneyulu1 K. Rajarajeswari2
1

Department of Electronics and Communications Engineering, Bapatla Engineering College, Bapatla, Acharya Nagarjuna
University, Andhra Pradesh, India
2
Department of Electronics and Communications Engineering, College of Engineering, Andhra University, Visakhapatnam,
Andhra Pradesh, India
E-mail: kongara_r@yahoo.com

Abstract: In this work, a robust and oblivious image watermarking scheme based on discrete wavelet transform (DWT) for copyright
protection is presented. The original unmarked image is not required for watermark extraction. Third-level DWT is applied to the
original cover image. Third- and second-level horizontal detail sub-band (LH2 and LH3) coefcients are grouped into different
blocks. Grouping of the coefcients is done in such a way that each block should contain one coefcient from LH3 sub-band
and four coefcients from LH2 sub-band. In each block, the rst minimum and the second minimum are identied and modied
according to the watermark bit. After watermark insertion, inverse DWT is applied to the sub-bands with modied coefcients to
obtain the watermarked image. For watermark extraction, a threshold-based decoder is designed. Embedding and extraction
process are characterised with parameters and genetic algorithm is used for parameter optimisation. Optimisation is to maximise
the values of peak signal-to-noise ratio of the watermarked image and normalised cross correlation of the extracted watermark.
The performance of the proposed scheme is compared with the existing schemes and signicant improvement is observed.
Experimental results show that, this algorithm is highly robust for many image attacks on the watermarked image.

Introduction

Major problem associated with digital multimedia data

transfer over internet is copyright protection. This is
because digital media are easy to copy and transmit. Many
researchers are also aware of the issues like copyright
protection, authentication, proof of ownership etc., related
to multimedia. Many solutions have been proposed to solve
those problems and are available in the literature.
Watermarking technique is one of the popular solutions to
those issues. Watermarking technique embeds specic
information called watermark into the original media so that
it is not easily perceptible; that is, the viewer cannot see
any information embedded in the contents. In the case of
dispute over the owner of the multimedia data, embedded
watermark can be extracted and it can be used to identify
the owner because the watermark contains the required
information. There are three important issues in the
watermarking system. First, the embedded watermark
should not degrade the quality of the image and should be
perceptually invisible to maintain its protective secrecy.
Second, the watermark must be robust enough to resist
common image processing attacks and not be easily
removable; only the owner of the image ought to be able to
extract the watermark. Third, the blindness is necessary if it
is difcult to obtain the original image and watermark at
time of extraction. A watermarking technique is referred to
as blind (also referred to as oblivious or public) if the
original image and watermark are not needed during
364
& The Institution of Engineering and Technology 2012

extraction. The process of digital watermarking involves

the modication of the original multimedia data to
embed a watermark containing key information such as
authentication or copyright codes. The embedding method
must leave the original data perceptually unchanged, yet
should impose modications which can be detected by
using an appropriate extraction algorithm. Common types
of signals to watermark are images, music clips and digital
video. Application of the digital watermarking to still
images is considered in this work. If the watermark
survives for many image attacks and is still extractable
from the watermarked media then the watermarking
technique is called robust. The major technical challenge is
to design a highly robust digital watermarking technique
which protects copyright of the media owner by making the
process of watermark removal tedious and costly [1].
Watermarking schemes can be classied into several
categories according to their applications, embedding
domain and characteristics. Based on embedding domain,
watermarking schemes are classied into three categories:
spatial domain, transform domain and hybrid domain.
Algorithms based on the transform domain are more robust
than the schemes based on spatial domain and hybrid
domain. Watermarking algorithms based on discrete cosine
transform (DCT) [2 5], discrete wavelet transform (DWT)
[6 9], discrete Hadamard transform [10, 11], singular value
decomposition [12] and discrete Fourier transform [13, 14]
are some of the transform domain methods available in the
literature.
IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364 373
doi: 10.1049/iet-ipr.2010.0347

www.ietdl.org
Image watermarking is the process of inserting an image
called watermark in another image called cover image.
Outcome of this insertion process is called watermarked
image. The insertion process must be in such way that
watermark is extractable from the watermarked image.
The main problem in digital image watermarking is
survival of the watermark when the watermarked image is
subjected to image processing operations (image attacks)
like compression, low-pass ltering, high-pass ltering,
histogram equalisation etc. Watermarking schemes are called
robust if the survival capability of the watermark is high. If
the extraction process does not require watermark or cover
image then the scheme is called oblivious watermarking.
Several algorithms are available in the literature for robust
and oblivious image watermarking. But, there are some
limitations in the existing algorithms. Designing a robust and
oblivious watermarking system is still a challenging problem.
Watermark embedding in the cover image requires
modication of some features of the cover image according
to the symbols of the watermark. Selected features of the
cover image can be pixels or transform coefcients. If the
watermarking algorithm modies the cover image pixels
then it is called spatial domain watermarking. Transform
domain watermarking algorithms modies the transform
coefcients of the cover image. Spatial domain algorithms
are simple and data embedding capacity is more. But they
are not robust against image attacks. On the other hand,
transform domain algorithms are robust against image
attacks and the amount of robustness varies from attack to
attack. No single algorithm is available with which
watermark survives for all possible image attacks. Hence,
there is a scope for improvement in the excising watermark
algorithms and an attempt is made to improve the
performance of some of the existing transform domainbased robust and oblivious watermarking algorithms.
Lin et al. [8] proposed an algorithm based on DWT. Binary
watermark of size 32 16 is embedded in the 8-bit greyscale cover image of size 512 512. Cover image
is decomposed using fourth-level DWT. LH4 sub-band is
selected for watermark embedding. Selected sub-band
is divided into variable size blocks. Number of blocks
required is equal to the number of watermark bits. In each
block of LH4 , one bit is embedded by modulating the
maximum wavelet coefcient according to the watermark
bit. Value of the maximum coefcient is increased for
embedding a 1. To embed a 0, maximum coefcient value
is decreased. But the modied value must be greater than or
equal to the second maximum in that block. This
requirement is essential for watermark extraction. Some
parameters are used to obtain control over the embedding
and extraction process. Limitations of this method are low
data embedding capacity and the difculty of adjusting the
parameters when there is a change in watermarking
requirements or change in input images.
Another DWT-based algorithm was proposed by Lin et al.
[15], in which there is more scope for increasing the
information carrying capacity of the cover image. Binary
watermark of size 32 16 is embedded in the 8-bit grey
scale cover image of size 512 512. LH4 and LH3 subbands are selected for watermarking. Fixed length blocks
are formed with the coefcients of the selected sub-bands.
There is a constraint in the formation of the blocks. In each
block, one coefcient from LH4 and four coefcients from
LH3 must be included. One watermark bit can be embedded
in each block by modulating the rst minimum and second
minimum in that block according to the watermark bit.
IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364373
doi: 10.1049/iet-ipr.2010.0347

With this framework, maximum of 1024 bits can be

embedded. If the input images (cover image, watermark)
are changed then the parameters of the scheme need to be
adjusted for satisfying the peak signal-to-noise ratio (PSNR)
requirement. In that paper, the authors showed that
performance of their method is superior in comparison with
the methods of Wang and Lin [16], Li et al. [17] and Lien
and Lin [18].
This paper presents a robust and oblivious watermarking
scheme in the transform domain. In general, any twodimensional multi-resolution transform can be used. The
performance of the proposed scheme might change slightly
from transform to transform in terms of PSNR and NCC
and those changes can be analysed experimentally. In this
work, DWT is used. Third-level DWT is applied to the
original cover image. Third-and second-level horizontal
detail sub-band (LH2 and LH3) coefcients are grouped into
different blocks. Grouping of the coefcients is done in
such a way that each block should contain one coefcient
from LH3 sub-band and four coefcients (2 2 nonoverlapping block) from LH2 sub-band. A distance vector is
computed using the coefcients of these blocks. Each
element in the distance vector represents the difference
between the second minimum and the rst minimum in a
block. To embed a watermark bit 0, rst minimum and
second minimum in a block are assigned the same value
and that value lies between rst minimum and second
minimum. Exact value will be selected using genetic
algorithm (GA) for the specied watermarking
requirements. To embed a watermark bit 1, rst minimum
value is decreased by a value which is the maximum of the
mean value of the distance vector and a constant. GA is
used to choose the constant value. After embedding all the
watermark bits, inverse DWT is applied to the sub-bands
including modied sub-bands to obtain the watermarked
image. For watermark extraction, a threshold-based
statistical decoder is designed. In that, a distance vector is
calculated from the possibly attacked watermarked image
and by comparing each element of the distance vector with
the threshold, a decision is taken in favour of one of the
binary symbol. The scheme is characterised with parameters
and GA is used for parameter optimisation. Optimisation is
required to satisfy the conicting requirements of PSNR and
normalised cross correlation (NCC). Experimental results
show that, this algorithm is highly robust for many image
attacks on the watermarked image. The performance of the
proposed method is tested with watermark images of
different sizes. To embed a 64 64 watermark, LH2 and
LH3 are used. But the coefcients of LH3 and LH4 subbands are used for embedding both 32 32 and 32 16
binary watermarks. The proposed method is compared with
the existing methods [15 18] [in which 32 16 size binary
watermark (512 bits) is embedded]. The proposed method
can effectively embed a maximum of 2048 watermark bits
(64 64 size binary image) without compromising the
quality (NCC) of the extracted watermark image.
Remainder of the paper is organised as follows: In Section
2, introduction to GA and its application to watermarking are
described. The proposed scheme is presented in Section 3.
Experimental results are given in Section 4. Conclusions are
presented in Section 5.

Genetic algorithm

GAs were rst developed by John Holland [19]. GA is one of

the best optimisation tools available in the literature. It is
365

& The Institution of Engineering and Technology 2012

www.ietdl.org
widely used to solve the problems in various scientic and
engineering applications [20]. GA process can be described
based on ve functional units. They are a random number
generator, tness evaluation unit and genetic operators for
reproduction, crossover and mutation operations. Random
number generator generates a set of number strings called
population. Each string represents a solution to the
optimisation problem. For each string, a tness value is
computed by the evaluation unit. A tness value is a
measure of the goodness of the solution. The objective of
the genetic operators is to transform the set of strings into
sets with higher tness values. The reproduction operator
performs a natural selection function known as seeded
selection. Individual strings are copied from one set
(generation of solutions) to the next according to their
tness values. The probability of a string being selected for
the next generation increases with the tness value. The
crossover operator chooses pairs of strings at random and
produces new pairs. The mutation operator randomly
changes the values of bits in a string. A phase of the
algorithm consists of applying the evaluation, reproduction,
crossover and mutation operations. A new generation of
solutions is produced with each phase of the algorithm.
Completion of optimisation process depends on termination
criterion. Termination criterion can be specied in terms of
number of generations, specied time interval etc.
Watermarking problem can be viewed as an optimisation
problem. In this work, GA is used for solving the
optimisation problem. PSNR and NCC are the two
important characteristic parameters of a watermarking
system. The amount of distortion introduced to the host
image during embedding process is inversely proportional
to PSNR. NCC indicates the amount of similarity between
original watermark and extracted watermark. Hence, both
PSNR and NCC values must be as large as possible for a
good watermarking system. But PSNR and NCC are related
in such way that maximisation of PSNR decreases the value
of NCC. Hence, the watermarking scheme is characterised
with parameters and GA is used to nd the optimum values
of parameters to obtain the specied performance of the
watermarking system in terms of PSNR and NCC.

Proposed scheme

In this section, proposed scheme is described in three subsections. Sub-section 3.1 deals with watermark embedding
procedure, watermark extraction is explained in Sub-section
3.2 and the application of GA for determining the optimum
parameters of the scheme is given in Sub-section 3.3.
3.1

Watermark embedding technique

This sub-section describes the embedding procedure to

embed a binary watermark into the 8-bit grey-scale cover
image. Consider the cover image of size 512 512 and
transform it using the third-level DWT. This transformation
produces ten sub-bands. One or more sub-bands can be
used for watermark embedding. Depending upon the
specic watermarking requirements, sub-bands will be
selected. The requirements considered in this work are
imperceptibility, robustness and data embedding capacity.
For satisfying those requirements and based on the reasons
in [8, 15] and [21], both LH3 and LH2 are chosen for
embedding 64 64 size watermark image.
Coefcients of LH3 and LH2 are grouped into various
blocks. Grouping is in such a way that each block should
366
& The Institution of Engineering and Technology 2012

contain one coefcient from LH3 and four coefcients from

LH2 . In each block, one bit of information is embedded by
modifying the two coefcients in that block. The two
coefcients in a block are the rst minimum and the second
minimum.
Steps of the embedding process are as follows:
1. Decompose the original cover image into various subbands using the third-level DWT.
2. Organise the coefcients of LH2 and LH3 into veelement blocks (4096 blocks). One coefcient from LH3
and the four non-overlapping coefcients from LH2 are
included in each block.
3. Let mini and seci are the rst minimum and the second
minimum, respectively, in the ith block. Let di denotes the
difference between seci and mini . Compute di for all i 1
to 4096.
4. Compute the mean value of the above differences and
identify it with dmean .
5. Consider a binary watermark of size 64 64 (4096 bits).
Let wi represents the ith watermark bit.
To embed a bit 0 (wi 0), modify the values of mini and
seci to the following equations
= mini + k1 di
minnew
i

(1)

= minnew
secnew
i
i

(2)

where k1 is the scaling factor used to modify the rst

minimum.
To embed a bit 1 (wi 1), modify mini using the following
equation
= mini , if di max(dmean , k2)
minnew
i
= mini e, otherwise

(3)

where e k2 + dmean/max(di , k3), k2 and k3 are the scaling

factors (or parameters) used to control the perceptible
quality (PSNR) of the watermarked image.
6. After embedding all the watermark bits, combine the
blocks and form the modied LH2 and LH3 sub-bands.
Then with these modied sub-bands and with the remaining
unmodied sub bands, apply inverse DWT to obtain the
watermarked image.
3.2

Watermark extraction process

Possibly attacked watermarked image and one parameter (k4)

are the inputs required for the extraction process. The
parameter k4 can be considered as the required key in the
extraction process.
Extraction of watermark is as follows:
1. Decompose the possibly attacked watermarked image into
sub-bands using the third-level DWT.
2. Organise the coefcients of LH2 and LH3 into veelement blocks. One coefcient from LH3 and the four nonoverlapping coefcients from LH2 are included in each block.
3. Let minoi and secoi are the rst minimum and the second
minimum, respectively, in the ith block. Let dio denotes the
difference between minoi and secoi . Compute dio = secoi minoi
for all i 1 to 4096 and arrange them in a vector, {dio }.
o
o
4. Compute dmean
, distd
mean and standard deviation values
o
of the vector {di }, where, i 1 to 4096.
IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364 373
doi: 10.1049/iet-ipr.2010.0347

www.ietdl.org
5. Compute the following
o
dio dmean
,
o
dstd

for all i = 1 to 4096

(4)

o
fmean
= mean(fio ),

for all i = 1 to 4096

(5)

fi o =

T=

o
dmean
k4
o
10 fmean

(6)

where T is the threshold and k4 is the parameter used to

control the NCC value.
6. Extract the watermark using the following rule
wi = 1,

if dio T ,

= 0, othewise

for all i = 1 to 4096

(7)

After extracting all the watermark bits, reshape them as

64 64 size binary watermark.
3.3

Optimisation of parameters using GA

As described in Section 2, GA can be used for watermarking

applications [20] based on the fact that effective
watermarking has two conicting requirements, PSNR and
NCC. These two requirements are related to each other and
therefore the watermarking algorithm described above must
be optimised. Optimisation search space and the tness
function are described as follows.

5. Write a function to extract the watermark from the attacked

watermarked image (with one or more specic attacks) as per
the procedure explained in Section 3.2. Function should
return the NCC value for the extracted watermark.
6. Write another function by using the parameters of the
scheme, embed and extract functions described in the
previous two steps for calculating the tness value. Fitness
function is dened in (8), which is the most generalised one
and it can be changed according to requirements.
7. Run GA to maximise the tness function. After the
termination of GA, we obtain the optimum values for the
parameters.
8. Using the parameter values obtained from the previous
step, calculate the optimum value of PSNR for an
unattacked watermarked image and NCC values for the
extracted watermarks with various attacks.
9. Use the obtained optimum values of PSNR and NCCs (with
various attacks) to describe the performance of the scheme.
Note:
Refer Figs. 1 3 for ow charts of embedding algorithm,
extraction algorithm and parameter optimisation, respectively.
Steps 1 3 represent the initialisation of the GA-training
parameters.

Search space: The values of the four scaling factors (k1, k2,
k3 and k4) are the key that, if chosen properly, will result in
optimal imperceptible and robust watermarking. It is the
role of the GA to nd such values, where the GAs search
space must include all possible values for the four scaling
factors. The GA is an iterative procedure that achieves
optimisation in a given search space using the genetic
operators (selection, reproduction, crossover and mutation)
and a tness function.
The tness function: Two common performance evaluation
metrics are combined to form the tness function, PSNR
and NCC. The tness function is formed by combining the
two metrics as follows.
1
(NCCk,l ak )
P k=1
p

fitl = PSNRl +

(8)

where l denotes GA generation number, p denotes the total

number of attacks used in the optimisation process, NCCk,l
represents NCC value with attack k and ak represents
weighting factor for NCC. PSNR and NCC are dened in
Section 4.
Optimisation of parameters is as follows:
1. Specify an initial range for all the parameters (or scaling
factors) used in the scheme.
2. Dene the population size, crossover rate, mutation rate
and number of iterations.
3. Specify the termination criteria.
4. Write a function to embed a binary watermark into the
grey-level host image following the steps in Section 3.1.
Function should return the PSNR value of the obtained
watermarked image.
IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364373
doi: 10.1049/iet-ipr.2010.0347

Fig. 1 Flow chart for embedding algorithm

367

& The Institution of Engineering and Technology 2012

www.ietdl.org

Fig. 4 Lena image

Fig. 5 Barbara image

Fig. 2 Flow chart for extraction algorithm

Fig. 6 Watermark image

Fitness function tl of (8) is the most generalised one and it

can be changed according to the watermarking requirements
for some intended application.
Fitness functions used in this work are given in the
Experimental results section.

Experimental results

The PSNR is used to evaluate the quality of an attacked image

in comparison with the original image. PSNR is dened as
Fig. 7 Watermarked Lena
PSNR 40.1818 dB

follows
PSNR = 10 log10

255 255
dB (9)
M N
1/(M N ) x=1 y=1 [f (i, j) g(i, j)]2

Fig. 3 Flow chart for GA-based watermark embedding

where M and N are the height and width of the image,

respectively. f (i, j) and g(i, j) are the pixel values located at
coordinates (i, j) of the original image and the attacked
image, respectively. After extracting the watermark, the
NCC is computed using the original watermark and the
extracted watermark to judge the existence of the watermark

368
& The Institution of Engineering and Technology 2012

IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364 373

doi: 10.1049/iet-ipr.2010.0347

www.ietdl.org
Table 1

Results of GA-based optimisation with cover image Lena for different GA generations

No. of Gen. and iterations

1 (20)
2 (40)
3 (60)
4 (80)
5 (100)

Fitness value

PSNR in dB

NCC

8.8375
7.5815
6.5152
6.911
6.9008

37.0694
36.8027
36.4752
35.8554
35.9857

0.7047
0.7808
0.8505
0.8617
0.8557

Parameters [k1, k2, k3, k4]

[0.7705,
[0.9049,
[0.8535,
[0.2790,
[0.3455,

24.0026, 0.5579, 3.4828]

25.8589, 0.8510, 3.2509]
28.8097, 2.2475, 1.4246]
29.5332, 0.5579, 3.2654]
28.4320, 0.5386, 2.5172]

Attack: JPEG-30 fitness function: (40-PSNR) + 20(1-NCC). Parameter value ranges: [0.10.9; 525; 0.51.5; 1.0 6.0]. Cover image: Lena
(Fig. 4)
Optimisation is against the JPEG compression attack with quality factor 30. The other details of GA are also provided in the table

Table 2

Results of GA-based optimisation with cover image Barbara for different GA generations

No. of Gen. and iterations

1 (20)
2 (40)
3 (60)
4 (80)
5 (100)

Fitness value

PSNR in dB

NCC

7.3524
8.5484
8.4653
6.8279
6.6154

37.8881
38.7404
37.7985
37.5731
37.8163

0.7380
0.6356
0.6868
0.7800
0.7784

Parameters [k1, k2, k3, k4]

[0.8434,
[0.5949,
[0.7880,
[1.2214,
[1.2647,

23.3791, 1.3875, 1.8837]

19.1600, 0.8423, 1.1163]
23.3381, 1.2155, 3.2244]
23.0954, 1.2430, 1.4644]
22.8179, 1.8970, 1.5637]

Attack: JPEG-30 fitness function: (40-PSNR) + 20(1-NCC). Parameter value ranges: [0.10.9; 5 25; 0.5 1.5; 1.0 6.0]. Cover image: Barbara
(Fig. 5)
Optimisation is against the JPEG compression attack with quality factor 30. The other details of GA are also provided in the table

Table 3

Results of GA-based optimisation with cover image Lena for different GA generations

No. of Gen. and iterations

1 (20)
2 (40)
3 (6)
4 (80)
5 (100)

Fitness value

PSNR in dB

NCC

6.4278
6.2082
6.6096
6.3445
6.2409

39.6086
40.1818
39.6354
40.0465
39.4262

0.6982
0.6987
0.6878
0.6851
0.7166

Parameters [k1, k2, k3, k4]

[0.4130,
[0.7818,
[0.3651,
[0.7501,
[0.7616,

18.265, 1.0078, 1.4852]

16.9749, 1.2095, 2.0162]
18.7261, 1.1723, 2.4957]
16.8966, 0.9034, 2.2378]
18.1968, 0.9536, 1.9285]

Attack: Sharpening filter fitness function: (40-PSNR) + 20(1-NCC). Parameter value ranges: [0.1, 0.9; 5, 25; 0.5, 1.5; 1.0, 6.0]. Cover image:
Lena (Fig. 4)
Optimisation is against the sharpening filter attack. The other details of GA are also provided in the table

Table 4

Results of GA-based optimisation with cover image Barbara for different GA generations

No. of Gen. and iterations

1 (20)
2 (40)
3 (60)
4 (80)
5 (100)

Fitness value

PSNR in dB

NCC

7.4072
7.4928
7.3545
7.7781
7.6679

38.4928
38.9112
38.6964
39.4015
39.7540

0.7050
0.6798
0.6975
0.6410
0.6289

Parameters [k1, k2, k3, k4]

[0.8311,
[0.8864,
[0.8123,
[0.5961,
[0.6590,

20.9822, 1.4933, 1.5092]

19.7283, 1.4537, 2.2639]
20.1866, 1.3221, 1.9582]
18.8951, 1.4678, 2.1066]
16.8123, 0.9625, 1.6944]

Attack: Sharpening filter fitness function: (40-PSNR) + 20(1-NCC). Parameter value ranges: [0.1, 0.9; 5, 25; 0.5, 1.5; 1.0, 6.0]. Cover image:
Barbara (Fig. 5)
Optimisation is against the sharpening filter attack. The other details of GA are also provided in the table

and to measure the correctness of an extracted watermark.

NCC is dened as follows (see (10))
where m and n are the height and width of the watermark,
respectively. The symbols w(i, j) and w o(i, j) are the bits
located at the coordinates (i, j) of the original watermark

and the extracted watermark, respectively. The symbols

wmean and womean are the mean values of the original
watermark and the extracted watermark, respectively.
Lena image and Barbara image, both of size 512 512,
are used as the cover images and are shown in Figs. 4 and

m n

o
o
i=1
j=1 [w(i, j) wmean ][w (i, j) wmean ]
NCC = 
 
 

m
n
m
n
2
o
o
2
i=1
j=1 [w(i, j) wmean ]
i=1
j=1 [w (i, j) wmean ]

IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364373

doi: 10.1049/iet-ipr.2010.0347

(10)

369

& The Institution of Engineering and Technology 2012

www.ietdl.org
5, respectively. The watermark image is of 64 64 size,
which is a binary logo as shown in Fig. 6. Watermarked
Lena image is shown in Fig. 7. Two-dimensional DWT with
Haar wavelet lters is used. Parameters of the scheme are
k1, k2, k3 and k4. Perceptibility of the watermarked image is
controlled by k1, k2 and k3. Parameter k4 controls the
quality of the extracted watermark. GA is used to nd the
optimum values of these parameters for some target values
for PSNR, NCC and one or more specic image attacks. For
testing the robustness of the proposed scheme MATLAB 7.0
and Checkmark 1.2 [22] are used. For all the attacks of
Checkmark 1.2 and MATLAB 7.0, a window size of 3 3
is taken. A built-in function related to genetic algorithm
available in MATLAB 7.0 is used. A population size of 20
chromosomes, a crossover rate of 0.8 and a Gaussian
mutation function (with a scale 1.0 and shrink 1.0) are used.
Various attacks used to test the robustness of the proposed
watermark are JPEG compression, median lter, Gaussian
lter, average lter, image sharpening, histogram
equalisation, resizing, cropping, Gaussian noise, salt &
pepper noise, rotation, Wiener lter, grey-scale inversion,
gamma correction, soft thresholding, template removal,
trimmed mean alpha, bit plane removal, row & column
copying and row & column blanking.
Table 1 shows the results of GA optimisation process. Lena
image is used as a cover image (Fig. 4) and the image shown
in Fig. 6 is used as watermark. The attack used in the
optimisation is JPEG compression with a quality factor 30.
Fitness function is [abs ((40-PSNR) +20(1-NCC))], where
abs refers to absolute value. Search space for the
optimisation process can be specied by providing the
range for each parameter. 0.1 0.9 is the range given for k1.
Similarly, the ranges for k2(5 25), k3(0.5 1.5) and
k4(1.0 6.0) are specied. PSNR value of the watermarked
image depends on k1, k2 and k3. By changing these
parameter values, PSNR value can be adjusted to any
desired value. NCC value of the extracted watermark
depends on k4 value. PSNR and NCC are related. NCC
decreases as PSNR increases. But the requirement is that
both must be as large as possible. Hence, it is very difcult
to adjust the parameter values for satisfying watermarking
requirement. Using the above specied tness function,
PSNR is targeted to 40 dB and NCC is targeted to 1.0. GA
is used to minimise the value of the tness function. As the
tness function value decreases, PSNR approaches 40 and
NCC approaches 1.0. As explained in Section 2,
optimisation process will be terminated according to the
specied termination criteria. In this work, the number of
GA generations is used as the termination criteria.
Increasing GA generations can be stopped if there is no
improvement in both PSNR and NCC.
Table 2 shows the results under the conditions specied for
the results of Table 1 except for the change in the cover
image. The cover image is the Barbara image shown in
Fig. 5. Similarly, Table 3 shows the results with sharpening
lter attack and Lena as the cover image. The results of
Table 4 are with sharpening lter attack and the Barbara
cover image.
Fig. 8 shows the NCC values of extracted watermarks from
the attacked watermarked images. Watermark image is shown
in Fig. 6. PSNR value of the unattacked watermarked image
(Fig. 7) is 40.1818 dB.
Parameters used in the embedding process are k1, k2 and
k3. The threshold parameter k4 is used in the extraction
process. The four parameter values are 0.7818, 16.9749,
1.2095 and 2.0162. Those values are obtained using GA
370

& The Institution of Engineering and Technology 2012

Fig. 8 Extracted watermark images from the watermarked Lena

(Fig. 7, PSNR 40.1818 dB)
NCC values are also indicated

with [abs ((40 2 PSNR) + 20(1 2 NCC))] as tness

function and with sharpening lter attack (refer Table 3).
Table 5 shows the results of GA with Lena cover image and
with watermark images of different sizes. The results indicate
that PSNR of the watermarked image decreases with the
number of watermark bits. This is because the amount of
distortion introduced by the watermarking process will
increase with the number of bits that are embedded.
IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364 373
doi: 10.1049/iet-ipr.2010.0347

www.ietdl.org
Table 5

Results of GA-based optimisation with cover image Lena for different size watermark images

Watermark image
32 16 binary logo (512 bits)
32 32 binary logo (1024 bits)
64 64 binary logo (4096 bits)

PSNR in dB

NCC1 with JPEG (QF 10)

NCC2 with median filter

Scaling factors [k1, k2, k3, k4]

44.9526
41.2708
37.5575

0.5527
0.5580
0.1375

0.9101
0.9139
0.8737

[0.6395, 28.4554, 1.7893, 3.0186]

[0.6091, 28.5795, 0.7091, 3.6058]
[0.6120, 26.1193 2.3710, 3.3454]

Attack: JPEG (10), Median filter (3 3), Fitness function: (44-PSNR) + 20(2-NCC1-NCC2), Initial range for parameters: [0.10.95; 26 30;
0.1 2.5;1.06.0], Number of GA iterations: 20, Cover image: Lena
Optimisation is against the JPEG compression (QF 10) and median filter (3 3 window) attacks. The inputs given for GA are provided in
first row of the table

Table 6

NCC coefficients of the extracted watermarks with the original watermark

JPEG quality
factor (QF)

(a) 32 16 binary logo

PSNR 44.9526 dB, k1 0.6395,
k2 28.4554, k3 1.7893, k4 3.0186

(b) 32 32 binary logo

PSNR 41.2708 dB, k1 0.6091,
k2 28.5795, k3 0.7091, k4 3.6058

(c) 64 64 binary logo

PSNR 37.5575 dB, k1 0.6120,
k2 26.1193, k3 2.3710, k4 3.3454

NCC

NCC

NCC

0.5527
0.6446
0.7609
0.8791
0.9256
0.9555
0.9853
0.9853
1.0000
1.0000
1.0000
1.0000
1.0000

0.5580
0.7148
0.8360
0.9332
0.9247
0.9673
0.9890
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000

0.1375
0.1711
0.2398
0.4501
0.7650
0.8721
0.9167
0.9397
0.9668
0.9844
1.0000
1.0000
1.0000

10
15
20
25
30
35
40
50
60
70
80
90
100

JPEG compression with different quality factors is the attack applied on the Lena watermarked image for watermark images with different
sizes: (a) 32 16 binary logo, (b) 32 32 binary logo and (c) 64 64 binary logo

Table 7

NCC coefficients of the extracted watermarks with the original watermark

Type of the attack

median filter (3 3)
Gaussian filter (3 3)
variance 0.5
average filter (3 3)
sharpening filter
(3 3)
histogram
equalisation
scaling 50%
cropping 25%
Gaussian noise
variance 0.001
salt & pepper noise
(0.001)
Rotation
58
108
158
308
Wiener filter

(a) 32 16 binary logo PSNR 44.

(b) 32 32 binary logo
(c) 64 64 binary logo
9526 dB, k1 0.6395, k2 28.4554,
PSNR 41.2708 dB, k1 0.6091,
PSNR 37.5575 dB, k1 0.6120,
k3 1.7893, k4 3.0186
k2 28.5795, k3 0.7091, k4 3.6058 k2 26.1193, k3 2.3710, k4 3.3454
NCC

NCC

NCC

0.9101
0.9779

0.9139
0.9817

0.8737
0.9745

0.7568
0.7155

0.8062
0.7068

0.7650
0.8084

0.9560

0.8039

0.8503

0.8792
1. 0000
0.8172

0.8753
0.5539
0.8877

0.7206
0.5539
0.7321

0.9632

0.9362

0.9247

0.7822
0.7118
0.6556
0.5778
0.9481

0.8021
0.7264
0.6688
0.5745
0.9519

0.8021
0.7264
0.6688
0.5745
0.8911

Different attacks are applied on the Lena watermarked image for watermark images with different sizes: (a) 32 16 binary logo,
(b) 32 32 binary logo and (c) 64 64 binary logo
IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364373
doi: 10.1049/iet-ipr.2010.0347

371

& The Institution of Engineering and Technology 2012

www.ietdl.org
Table 8 Comparison of the proposed method with the methods of Lin et al. [15], Lien and Lin [16], Li et al. [17] and Wang and Lin [18] in
terms of NCC
Attack type

Wang et al.
(PSNR 38.2 dB) [18]

median filter
(3 3)
median filter
(4 4)
IPEG (QF 10)
JPEG (QF 20)
JPEG (QF 30)
IPEG (QF 50)
JPEG (QF 70)
JPEG (QF 90)
sharpening
Gaussian filter
Rotation
0.258
0.758
18
20.258
20.758
218
cropping 1/4
scaling
256 256

Li et al.
(PSNR 40.6 dB) [17]

Lien et al.
(PSNR 41.54 dB) [16]

Lin et al.
(PSNR 44.73 dB) [15]

Proposed method
(PSNR 44.95 dB)

0.51

0.35

0.79

0.92

0.91

0.23

0.26

0.51

0.75

0.64

NA
NA
0.15
0.28
0.57
1
0.46
0.64

0.15
0.34
0.52
0.52
0.63
0.78
0.38
0.70

0.17
0.61
0.79
0.89
0.97
1
0.88
0.84

0.33
0.59
0.81
0.95
1
1
0.99
0.96

0.55
0.76
0.93
0.99
1
1
0.72
0.98

0.37
0.26
0.24
0.32
0.24
0.16
NA
NA

0.46
0.36
0.33
0.50
0.29
0.33
0.61
0.35

0.53
0.16
0.07
0.47
0.10
0.16
0.92
0.79

0.61
0.34
0.27
0.65
0.22
0.28
0.60
0.86

0.88
0.87
0.86
0.88
0.88
0.90
1
0.88

Tables 6 and 7 show the performance of the proposed

scheme against various image attacks in terms of NCC
values. One can observe the change in PSNR and NCC as
the number bits embedding into the host image increases.
Table 8 provides the performance comparison with the
existing methods. The performance of the proposed method
is better in comparison with the existing [15 18] schemes
for most of the image attacks shown in the table.

Conclusions

In this paper, a robust and oblivious watermarking scheme

based on DWT is presented. Watermark is embedded in the
detail sub-band coefcients of the cover image. LH2 and
LH3 sub-band coefcients of the cover image are used to
form ve-element blocks. In each block, the rst minimum
and the second minimum are identied and one watermark
bit is embedded in them. To embed a 0, the difference
between the rst minimum and the second minimum
is reduced to zero and to embed a 1, the difference is
increased to a signicant value. This embedding process
is characterised with four parameters. GA is used to nd the
optimum values for the parameters. Optimisation is to
maximise the values of PSNR and NCC. The performance
of the proposed method is tested with watermark images of
different sizes. To embed a 64 64 watermark, LH2 and
LH3 are used. But the coefcients of LH3 and LH4 subbands are used for embedding both 32 32 and 32 16
binary watermarks. The proposed method is compared with
the existing methods [15 18] [in which 32 16 size
binary watermark (512 bits) is embedded]. Compared with
those methods, the proposed GA-based method is superior
in terms of the embedding capacity and survival to number
of image attacks. It has been observed experimentally that
the performance of the scheme is satisfactory for several
372
& The Institution of Engineering and Technology 2012

image attacks. In addition, the exibility of the proposed

GA-based scheme is demonstrated in xing the parameters
of the scheme.

References

1 Zhao, J.: Look, its not there, Byte, January 1997

2 Cox, I.J., Killian, J., Leighton, T., Shamoon, T.: Secure spread spectrum
watermarking for multimedia, IEEE Trans. Image Process., 1997, 6,
pp. 1673 1687
3 Podilchuk, C.I., Zeng, W.: Image adaptive watermarking using visual
models, IEEE J. Sel. Areas Commun., 1998, 16, pp. 525539
4 Chu, W.C.: DCT based image watermarking using sub sampling, IEEE
Trans. Multimedia, 2005, 5, (1), pp. 3438
5 Wang, Y., Pearmain, A.: Blind image data hiding based on self
reference, Pattern Recognit. Lett., 2004, 25, pp. 1681 1689
6 Barni, M., Bartolini, M., Piva, F.V.: Improved wavelet based
watermarking through pixel-wise masking, IEEE Trans. Image
Process., 2001, 10, pp. 783 791
7 Wang, H.J.M., Su, P.C., Kuo, C.C.J.: Wavelet-based digital image
watermarking, Opt. Express, 1998, 3, (12), pp. 491 496
8 Lin, W.-H., Horng, S.-J., Kao, T.-W., Fan, P., Lee, C.-L., Pan, Y.: An
efcient watermarking method based on signicant difference of
wavelet coefcient quantization, IEEE Trans. Multimedia, 2008, 10,
(5), pp. 746 757
9 Lin, W.H., Wang, Y.R., Horng, S.J., Kao, T.W., Pan, Y.: A blind
watermarking method using maximum wavelet coefcient
quantization, Expert Syst. Appl., 2009, 36, (9), pp. 1150911516
10 Ramanjaneyulu, K., Rajarajeswari, K.: An oblivious image
watermarking scheme using multiple description coding and genetic
algorithm, IJCSNS, Int. J. Comput. Sci. Netw. Secur., 2010, 10, (5),
pp. 167174
11 Ramanjaneyulu, K., Rajarajeswari, K.: An oblivious and robust
multiple image watermarking scheme using genetic algorithm,
Int. J. Multimedia Appl. (IJMA), 2010, 2, (3), pp. 19 38
12 Yongdong, Wu: On the security of an SVD based ownership
watermarking, IEEE Trans. Multimedia, 2005, 7, (4), pp. 624627
13 Premaratne, P.: A novel watermark embedding and detection scheme
for images in DFT domain. Proc. IEE Seventh Int. Conf. on Image
Processing & Applications, 1999, vol. 2, pp. 780 783
IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364 373
doi: 10.1049/iet-ipr.2010.0347

www.ietdl.org
14 Solachidis, V., Pitas, V.: Circularly symmetric watermark embedding in
2-D DFT domain, IEEE Trans. Image Process., 2001, 10,
pp. 1741 1753
15 Lin, W.-H., Wang, Y.-R., Horng, S.-J.: A wavelet-tree-based
watermarking method using distance vector of binary cluster, Expert
Syst. Appl., 2009, 36, pp. 9869 9878
16 Wang, S.H., Lin, Y.P.: Wavelet tree quantization for copyright
protection watermarking, IEEE Trans. Image Process., 2004, 13, (2),
pp. 154165
17 Li, E., Liang, H., Niu, X.: An integer wavelet based multiple
logo-watermarking scheme. Proc. IEEE WCICA, 2006,
pp. 10256 10260

IET Image Process., 2012, Vol. 6, Iss. 4, pp. 364373

doi: 10.1049/iet-ipr.2010.0347

18 Lien, B.K., Lin, W.H.: A watermarking method based on maximum

distance wavelet tree quantization. 19th Conf. on Computer Vision,
Graphics and Image Processing, 2006, pp. 269 276
19 Holland, J.H.: Adaptation in natural and articial systems: an
introductory analysis with applications to biology, control and
articial intelligence (MIT Press, Cambridge, MA, 1992)
20 Al-Haj, A., Abu-Errub, A.: Performance optimization of discrete
wavelets transform based image watermarking using genetic
algorithms, J. Comput. Sci., 2008, 4, (10), pp. 834 841
21 Shapiro, J.M.: Embedded image coding using zerotrees of wavelet
coefcients, IEEE Trans. Signal Process., 1993, 41, (12), pp. 34453462
22 www.http://watermarking.unige.ch/Checkmark/

373

& The Institution of Engineering and Technology 2012