International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169

Volume: 5 Issue: 6 1491 1496

_______________________________________________________________________________________________
Implementation of Adaptive Digital Image Watermarking Technique in Discrete
Wavelet Transform for Copyright Protection

Naina R. Dhanwate1, Y. A. Sadawarte2

PG Student, Department of Electronics Engineering, B. D. College of Engineering, Wardha, India1
Assistant Professor, Department of Electronics Engineering, B. D. College of Engineering, Wardha, India2

Abstract: Digital watermarking is used for content protection, content authentication, copyright management and tamper detection. With the use
of sophisticated signal / image processing algorithms, manipulations and duplications of audio, images and videos is much easier. Hence
copyright of the content has become an urgent and important issue in this world. In this paper, digital image watermarking technique based on
discrete wavelet transform and encryptionis proposed. Watermark embedding procedure algorithm using DWT coefficients, scaling factor and
scrambling matrix is proposed. The technique results in PNSR greater than 50 dB and is resistance to noise, geometric and compression attack.
The proposed technique may be applied for copyright and content authentication applications.

Keywords: Image Watermarking, Discrete Wavelet Transform, Encryption, PSNR, Copyright Protection.

__________________________________________________*****_________________________________________________

I. INTRODUCTION Steganography is an area which is, more or less, a Hide-&-

Seek game. Some important data or information is hidden in
The growth of high speed computer networks and World
another medium. The cover medium has no relationship
Wide Web (WWW) have explored means of new business,
with the data or information hidden. Data or information
scientific, entertainment and social opportunities in the form
which is hidden is not encrypted also [1]-[3]. The key issue
of electronic publishing and advertising, massaging, real-
in a steganography system becomes that no one should
time information delivery, data sharing, collaboration among
suspect that a particular medium is carrying any hidden data
computers, product ordering, transaction processing, digital
or information. We can extend the steganography concept
repositories and libraries, web newspapers and magazines,
for the authentication of digital multimedia data. Digital
network video and audio, personal communication and lots
multimedia data which has to be protected is now the cover
more. The cost effectiveness of selling softwares in the
medium and then we can hide the copyright data into it.
form of digital images and video sequences by transmission
Characteristics of Watermarking Schemes [3] are
over WWW is greatly enhanced due to the improvement in
imperceptibility, robustness, fragility, resilient to common
technology. Digital data can be stored efficiently and with a
signal processing operations, resilient to geometric
very high quality, and it can be manipulated very easily
operations, robust to subterfuge attacks and
using computers. Furthermore, digital data can be
unambiguousness. Applications of Watermarking
transmitted in a fast and inexpensive way through data
Techniques [3] are copyright protection, copy protection,
communication networks without losing quality. Digital
temper detection, broadcast monitoring, fingerprinting and
media offer several distinct advantages over analog media.
annotation applications.
The quality of digital audio, images and video signals are
higher than that of their analog counterparts. Editing is easy Watermarking techniques can be categorized as spatial
because one can access the exact discrete locations that need watermarking can also be applied using colour separation. In
to be changed. Copying is simple with no loss of fidelity. A this way, the watermark appears in only one of the colour
copy of a digital media is identical to the original. bands. This renders the watermark visibly subtle such that it
is difficult to detect under regular viewing. However, the
With digital multimedia distribution over World Wide Web,
mark appears immediately when the colours are separated
authentications are more threatened than ever due to the
for printing. This renders the document useless for the
possibility of unlimited copying. For digital data, copyright
printer unless the watermark can be removed from the
enforcement and content verification are very difficult tasks.
colour band. This approach is used commercially for
One solution would be to restrict access to the data using
journalists to inspect digital pictures from a photo-stock
some encryption techniques. However, encryption does not
house before buying unmarked versions. Watermarking can
provide overall protection. Once the encrypted data are
be applied in the frequency domain (and other transform
decrypted, they can be freely distributed or manipulated.
domains) by first applying a transform like the Fast Fourier
Steganography deals with the methods of embedding data
Transform (FFT) [4] [6]. DCT based watermarking
within a medium (host or cover medium) in an
techniques are more robust compared to simple spatial
imperceptible way. All forms of digital data (still images,
domain watermarking techniques. Such algorithms are
audio, video, text documents and multimedia documents)
robust against simple image processing operations like low
can be used as a cover medium for information hiding.
1491
IJRITCC | June 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 6 1491 1496
_______________________________________________________________________________________________
pass filtering, brightness and contrast adjustment, blurring In this paper, digital image watermarking based on discrete
etc. However, they are difficult to implement and are wavelet transform and encryption is proposed. Watermark
computationally more expensive. At the same time they are embedding algorithm is devised and results show better
weak against geometric attacks like rotation, scaling, PSNR and resistance to attack. The paper is organized as
cropping etc. section 2 describes the embedding and extraction algorithm,
section 3 illustrates results and concluded in section 4.

II. WATERMARK EMBEDDING AND EXTRACTION ALGORITHM

Fig. 1 Watermark embedding procedure

In the last few years wavelet transform has been widely

studied in signal processing in general and image
compression in particular. In some applications wavelet
based watermarking schemes outperforms DCT based
approaches. The wavelet transform decomposes the image
into three spatial directions, i.e. horizontal, vertical and
diagonal. Hence wavelets reflect the anisotropic properties
Fig. 2 Scrambling matrix
of HVS more precisely [2]. Wavelet Transform is
Where alpha is visibility coefficient and x is carrier image
computationally efficient and can be implemented by using
DWT coefficients & y is watermark image DWT
simple filter convolution [7] [8]. Magnitude of DWT
coefficients. Watermark extraction is the reverse process of
coefficients is larger in the lowest bands (LL) at each level
embedding. Wavelet coefficients of the watermark image
of decomposition and is smaller for other bands (HH, LH,
were extracted using equation shown in fig. 3.
HL) [1] [5] [6]. The larger the magnitude of the wavelet
coefficient the more significant it is. Watermark detection at
lower resolutions is computationally effective because at
every successive resolution level there are few frequency
bands involved [5]. High resolution subbands helps to easily
locate edge and textures patterns in an image [1]. Figure 1
shows the method adopted for watermarking images.
Fig. 3 Descrambling matrix
Wavelet coefficients of watermark image were encrypted
Where alpha is visibility coefficient and x is watermarked
using rows and columns rotations using random number
image DWT coefficients & y is original image DWT
generator. It increases security of watermark in case of
coefficients. Watermark extraction procedure is illustrated in
removal of watermark from the original image [9] [12].
fig. 4.
Watermark was embedded using scrambling matrix as
shown in figure 2.

1492
IJRITCC | June 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 6 1491 1496
_______________________________________________________________________________________________

Fig. 4 Watermark extraction

III. RESULTS & DISCUSSION speed. Table 1 shows the results obtained for various input
and watermark images. Algorithms show consistent values
Watermark embedding and extraction algorithm was
of PSNR and CC for all types of images. Fig. 5 shows the
implemented using MATLAB software and executed on
watermarks used in algorithms.
intel i5 processor with 1 GB RAM and 3 GHz processing

Table 1 Watermark parameters

Parameter Water mark Lena Complex Baboon Car Fruits

MSE Apple 0.22 0.22 0.23 0.21 0.17

Rings 0.17 0.16 0.18 0.21 0.16
Name 0.22 0.25 0.25 0.23 0.24
PSNR Apple 54.66 54.69 54.38 54.77 55.67
Rings 55.81 55.9 55.54 54.85 56.04
Name 54.69 54.01 54.06 54.48 54.31
CC Apple 0.99 0.98 0.98 0.99 0.99
Rings 0.99 0.99 0.99 0.99 0.99
Name 0.98 0.98 0.98 0.98 0.98

1493
IJRITCC | June 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 6 1491 1496
_______________________________________________________________________________________________

(a) Apple (b) Rings (c) Name

Fig. 5 Watermark images
Where PSNR is peak signal to noise ratio, mse is mean after watermark embedding and extraction algorithm for two
square error and CC is correlation coefficients. All the watermarks. Fig. 8 shows the results for watermark using
parameters were also obtained by affecting images using noise attack. Obtain results validate the concept of
various attacks such as noise, geometrical and compression. watermark embedding and extraction. Wavelet transform
Table 2 shows the watermark parameters obtained with provides better embedding and extraction procedure.
attack. Fig. 6 and 7 shows the complete results obtained

Table 2 Watermark parameters with attack

Parameter Lena Baboon
Noise Geometric Compression Noise Geometric Compression
MSE 0.24 0.22 0.23 0.23 0.22 0.21
PSNR 54.31 54.63 54.23 54.41 54.48 54.74
CC 0.98 0.99 0.99 0.98 0.99 0.98

Fig. 6 Watermark algorithm results with watermark apple

1494
IJRITCC | June 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 6 1491 1496
_______________________________________________________________________________________________

Fig. 7 Watermark algorithm results with watermark ring

Fig. 8 Watermark algorithm results after noise attack (salt and pepper noise)

IV. CONCLUSION sequentially processed from low resolution to high

resolution. Proposed watermark embedding and extraction
More robust watermarks could be embedded in the
algorithm results better PSNR of greater than 50 dB and
transform domain of images by modifying the transform
correlation coefficients of greater than 0.97. Moreover,
domain coefficients than spatial domain.Wavelet Transform
MSE, PSNR and CC these three parameters are closely
is computationally efficient and can be implemented by
related to a pattern of watermarking images so any
using simple filter convolution. Wavelet coded image is a
modification of one of these factors influences directly the
multi-resolution description of image. Hence an image can
others. Watermarking algorithms robust against the
be shown at different levels of resolution and can be
1495
IJRITCC | June 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
 International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 6 1491 1496
_______________________________________________________________________________________________
geometrical distortions have been the focus of further
research.

REFERENCES
[1] Xiaosheng Huang, Sujuan Zhao, An Adaptive Digital
Image Watermarking Algorithm Based on Morphological
Haar Wavelet Transform, in International Conference on
Solid State Devices and Materials Science, Published by
Elsevier Physics Procedia 25, 2012, pp 568 575.
[2] PrashantVaidya, et al, Adaptive Digital Watermarking for
Copyright Protection of Digital Images in Wavelet
Domain, in 2nd International Symposium on Computer
Vision & Internet Published by Elsevier Procedia Computer
Science 58, 2015, pp 233 - 240.
[3] Lei Chen and Jiying Zhao, Adaptive Digital Watermarking
Using RDWT and SVD, in IEEE International Symposium
on Haptic, Audio and Visual Environments and Games
(HAVE), 2015.
[4] L. R. Roldan, M. C. Hernndez, J. Chao, M. N. Miyatake
and H. P. Meana, Watermarking-based Color Image
Authentication With Detection And Recovery Capability,
in IEEE Latin America Transactions, Vol. 14, No. 2, Feb.
2016, pp 1050 1057.
[5] Aniket Roy, Arpan Kumar Maiti, KuntalGhosh, A
perception based color image adaptive watermarking
scheme in YCbCr space, in 2nd IEEE International
Conference on Signal Processing and Integrated Networks
(SPIN), 2015.
[6] NavneetYadav, Kulbir Singh, Transform Domain Robust
Image-Adaptive Watermarking: Prevalent Techniques and
Their Evaluation, in IEEE International Conference on
Computing, Communication and Automation, 2015.
[7] DeeptiShukla, NirupamaTiwari and DeepikaDubey,
Survey on Digital Watermarking Techniques, in
International Journal of Signal Processing, Image
Processing and Pattern Recognition Vol.9, No.1, 2016,
pp.239-244.
[8] Hai Tao, et al, Robust Image Watermarking Theories and
Techniques: A Review, in Journal of Applied Research
and Technology, Vol. 12, February 2014, pp 122 138.
[9] Hirak Kumar Maity, Santi P. Maity, Multiple Predictors
based RW Scheme with Adaptive Image Partitioning, in
IEEE International Conference on Advances in Computing,
Communications and Informatics, 2015.
[10] Pushpa Mala S., et al, Digital Image Watermarking
Techniques: A Review, in International Journal of
Computer Science & Security, vol. 9, issue 3, 2015, pp 140
156.
[11] EmanueleMaiorana, et al, High-capacity watermarking of
high dynamic range images, in Springer EURASIP Journal
on Image and Video Processing, 2016.
[12] B. Jagadeesh, et al, Fuzzy Inference System Based Robust
Digital Image Watermarking Technique using Discrete
Cosine Transform, in International Conference on
Information and Communication Technologies, published
by Elsevier Procedia Computer Science 45, 2015, pp 1618
1625.

1496
IJRITCC | June 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________