Hybrid video watermarking technique by using dwt & pca

International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN
INTERNATIONAL JOURNAL OF ELECTRONICS AND
0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME
COMMUNICATION ENGINEERING & TECHNOLOGY (IJECET)
ISSN 0976 – 6464(Print)
ISSN 0976 – 6472(Online)
Volume 4, Issue 2, March – April, 2013, pp. 172-179
IJECET
© IAEME: www.iaeme.com/ijecet.asp
Journal Impact Factor (2013): 5.8896 (Calculated by GISI) ©IAEME
www.jifactor.com

HYBRID VIDEO WATERMARKING TECHNIQUE BY USING DWT &
PCA

1
Mr. N. R. Bamane, 2Dr. Mrs. S. B. Patil , 3Prof. B. S. Patil, 4Prof. R. K. Undegaonkar
1
Dr. JJMCOE, JSP.
2
Head, Department of Electronics Engg. Dr. JJMCOE, JSP.
3
Head, Department of Information Technology PVPIT Budhgaon, Sangli
4
Trinity College of Engineering & research, Pune

ABSTRACT

Security and copyright protection are becoming important issues in multimedia
applications and services, as Past few years have witnessed rapid growth in Digital video
watermarking. Watermarking techniques have been proposed for these purposes in which the
copyright information is embedded into multimedia data in order to protect the ownership.
This paper presents a novel technique for embedding a binary logo watermark into video
frames. The proposed scheme is an imperceptible and a robust hybrid video watermarking
scheme. PCA is applied to each block of the two bands (LL – HH) which result from Discrete
Wavelet transform of every video frame. The watermark is embedded into the principal
components of the LL blocks and HH blocks in different ways. In this paper, a
comprehensive approach for digital video watermarking is introduced, where a binary
watermark image is embedded into the video frames. The proposed scheme is tested using a
number of video sequences. Experimental results show high imperceptibility where there is
no noticeable difference between the watermarked video frames and the original frames.
Combining the two transforms improved the performance of the watermark algorithm. The
scheme can be tested by applying various attacks.

Keywords- Digital Video Watermarking, Copyright protection, Discrete wavelet transform,
Principal component analysis, Binary logo watermark.

172


1. INTRODUCTION

Recently, the users of networks, especially the world wide web are increasing rapidly.
The reproduction, manipulation and the distribution of digital multimedia (images, audio and
video) via networks become faster and easier. Hence, the owners and creators of the digital
products are concerned about illegal copying of their products. As a result, security and
copyright protection are becoming important issues in multimedia applications and services.
In the Past years, Watermarking techniques have been proposed for these purposes in which
the copyright information is embedded into multimedia data in order to protect the
ownership. Research is now being focused on watermarking schemes to protect multimedia
content. Digital watermarking is a technology that can serve this purpose. A large number of
watermarking schemes have been proposed to hide copyright marks and other information in
digital images, video, audio and other multimedia objects.
In the literature, different digital video watermarking algorithms have been proposed.
Some techniques embed watermark in the spatial domain by modifying the pixel values in
each frame but these methods are not robust to attacks and common signal distortions. In
contrast, other techniques are more robust to distortions when they add the watermark in the
frequency domain. In these types of schemes, the watermark is embedded by modifying the
transform coefficients of the frames of the video sequence. The most commonly used
transforms are the Discrete Fourier Transform (DFT), the Discrete Cosine Transform (DCT),
and the Discrete Wavelet Transform. (DWT). Several researches concentrated on using DWT
because of its multire solution characteristics, it provides both spatial and frequency domain
characteristics so it is compatible with the Human Visual System (HVS).
The recent trend is to combine the DWT with other algorithms to increase robustness
and invisibility. In this paper, we propose an imperceptible and robust video watermarking
algorithm based on DWT and PCA. DWT is more computationally efficient than other
transform methods because of its excellent localization properties which provide the
compatibility with the Human Visual System (HVS). This paper is organized as follows:
section 2 presents the proposed watermarking scheme. Section 3 introduces the experimental
results and the conclusion is given in section 4.

2. PROPOSED VIDEO WATERMARKING TECHNIQUE

The proposed hybrid watermarking scheme is based on combining two
transformations; the DWT and the PCA. The block diagrams of embedding algorithms are
shown in Fig.2. In our method, video frames are taken as the input, and watermark is
embedded in each frame by altering the wavelet coefficients of selected DWT sub bands,
followed by performing the PCA transformation on the selected sub bands.

2.1 DISCRETE WAVELET TRANSFORM

The DWT is used in a wide variety of signal processing applications. 2-D discrete
wavelet transform (DWT) decomposes an image or a video frame into sub images, 3 details
and 1 approximation. The approximation sub image is lower resolution approximation image
(LL) however the details sub images are horizontal (HL), vertical (LH) and diagonal (HH)
detail components. The process can then be repeated to compute multiple "scale" wavelet
decomposition. The main advantage of the wavelet transform is its compatibility with a
model aspect of the HVS as compared to the FFT or DCT. This allows us to use higher

173


energy watermarks in regions that the HVS is known to be less sensitive, such as the high
resolution detail bands. Embedding watermarks in these regions allow us to increase the
robustness of our watermark without any visible impact on the image quality. In the proposed
algorithm, sub-bands LL and HH from resolution level 2 of the wavelet transform of the
frame are chosen for the embedding process. The following fig.1 shows the selected DWT
bands which used in our proposed algorithm.

Fig. a fig. b

Fig.1 DWT sub-bands in (a) level 1, (b) level 2.

Embedding the watermark in low frequencies obtained by wavelet decomposition increases
the robustness against attacks like filtering, lossy compression and geometric distortions
while making the scheme more sensitive to contrast adjustment, gamma correction, and
histogram equalization. Embedding the watermark in high frequency sub-bands makes the
watermark more imperceptible while embedding in low frequencies makes it more robust
against a variety of attacks.

2.2 PRINCIPAL COMPONENT ANALYSIS

PCA is an optimal unitary transformation that projects the data on a new coordinate
system such that the greatest data variation data comes lies on the first principal component,
the second greatest variation on the second principal component, and so on. This
transformation Orthogonalizes the components of the input data vectors so that they are
completely de-correlated. The resulting orthogonal components called (principal
components) are ordered such that most of the energy is concentrated into the first several
principal components. Due to the excellent energy compaction property, components that
contribute the least variation in the data set are eliminated without much loss of information.
Unlike other linear transformations, the PCA does not have a fixed set of basis functions but
it has basis functions which depend on the data set.

Fig.2. Watermark Embedding process.

174


The PCA approach is applied to the transform coefficients of wavelet sub band Iθ
where θ represents (LL or HH) as shown in the following steps:

Step1: The wavelet subband Iθ with NxN dimension is subdivided into nxn non overlapping
blocks (the block size should be appropriate to the sub band size) where the number of blocks
is given by k = NxN/nxn.
Step 2: Each block in LL band can be processed by method1 and each block in HH band can
be processed by method2 as following:
method 1: Consider each block like a vector; data vectors can be expressed as: Iθ =(Iθ1, Iθ2,
Iθ3,……., Iθk)T , where vector Iθi represents block number i with n2 dimension.
method 2: Each block can be considered as 2D array Bθ =( Bθ1, Bθ2, Bθ3,……., Bθk)T ,
where array Bθi represents block number i with size nxn.
Step 3: For each block, the covariance matrix Ci of the zero mean block A is calculated as:
Ci = Ai AiT (1)
where T denotes the matrix transpose operation, and A is defined by :
method 1: for a vector block as Ai=E(Iθi –mi).
method 2: for 2D array block as Ai=E(Bθi –mi).
where mi is the mean of block and E denotes expectation operation.
Step 4: Each block is transformed into PCA components by calculating the eigenvectors
(basis function) corresponding to eigenvalues of the covariance matrix:
Ci Ф = λi Ф (2)
where Ф is the matrix of eigenvectors and λ is the matrix of eigenvalues defined for:
method 1: for a vector block as Ф = (e1 ,e2 ,e3 ,…., enxn) and λi =(λ1, λ2, λ3,… λnxn).
method 2: for 2D array block as Ф = (e1 ,e2 ,e3 ,…., en) and λi = (λ1, λ2, λ3,…… λn).
Ф vectors are sorted in descending order according to λi , where ( λ1 ≥ λ2 ≥ λ3 ≥ …..≥ λn or
(λnxn)). The matrix Ф is an orthogonal matrix called basis function of PCA (PCA
eigenimages)
Step 5: Calculate the PCA components of the block. The PCA transforms the correlated
block into uncorrelated coefficients by taking the inner product of the block with the basis
functions Ф: Yi = ФT Ai (3)
where Yi is the PC block which represents the principle component of block i.
Step 6: Apply inverse PCA on the modified PCA components to obtain the modified wavelet
coefficients. The inversion can be performed by the equation: Ai = Ф Yi (4)

2.3 WATERMARK EMBEDDING

The proposed watermarking process shown in Fig. 2 is briefly described in the
following steps:

Step 1: Divide video into frames and convert 2Nx2N RGB frames into YUV components.
Step 2: For each frame, choose the luminance Y component and apply the DWT to
decompose the Y frame into four multi-resolution sub bands NxN: LL , HL , LH , and HH .
Step 3: Divide the two sub bands LL and HH into n x n non-overlapping blocks.
Step 4: Apply PCA to each block in the chosen subbands LL by using method1 and HH
by using method2
Step 5: Convert the 32x32 binary watermark logo into a vector W = { w1, w2 , ……. ,
w32x32} of '0's and '1's.

175


Step 6: Embed the logo into LL and HH bands by different ways. For LL band, the
watermark bits are embedded with strength α1 into the first principle component of each PC
block Yi. From equation (3) , for the PC block Y1, Y2, Y3,……., Yk, we can define YI =
(Y1(1), Y2(1), Y3(1),……., Yk(1))T and the embedding equation:
YI ' = YI + α1 W (5)
Step 7: For HH band, use two pseudorandom sequences (PNS); p0 and p1 with different
keys k1 and k2 to embed the watermark bit w '0' and '1' respectively [10,11]. So, we can
represent Wm as follows:

(6)
when bit w=0, embed p0 with strength α2 to the mid-band coefficient of PC block Yi
and when bit w=1, embed p1 with strength α2 to the mid-band coefficients of PC block Yi .
If YB includes the mid-band coefficients then the embedding equation is
YB ' = YB + α2 Wm (7)
Step 8: Apply inverse PCA on the modified PCA components of the two bands to obtain the
modified wavelet coefficients.
Step 9: Apply the inverse DWT to produce the watermarked luminance component of the
frame. Then reconstruct the watermarked frame

3. EXPERIMENTAL RESULTS

A number of video sequences are used for testing the proposed scheme for example
the foreman video sequence. For evaluating the performance of any watermarking system,
Peak Signal to Noise Ratio (PSNR) is used as a common measure of the visual quality of the
watermarking system. To calculate the PSNR, first the Mean Square Error (MSE) between
the original and watermarked frame is computed as follows:

Where M, N are the size of the frame, and I(i, j), I'(i, j) are the pixel values at location
(i, j) of the original and watermarked frames. Then, PSNR is defined as

The luminance component of the first 100 frames of the foreman video sequence are
watermarked. The frame size is 256x256. The watermark is a binary image with
size 32x32. The original sampled frame and its corresponding watermarked frame are shown
in Fig. 3. The measured PSNR is 44.0975 db and the watermarked frame appears
visually identical to the original. The value of PSNR is constant over all the tested frames
which means that the error between the original and watermarked frames is very low so high
visual quality is obtained. Fig. 4 shows the original watermark and the extracted watermark
from LL band and HH band where no attacks were applied. The measured value of NC is 1
for both LL band and HH band, i.e. the extracted watermark is identical to the original and
exact extraction is obtained.

176


Fig. 3 (a) Original frame, (b) Watermarked frame (PSNR = 44.0975db).

Fig.4. Binary logo Watermark

Fig.5 GUI used to execute the Experiment.

Experimental results show high imperceptibility where there is no noticeable
difference between the watermarked video frames and the original frames. Combining the
two transforms improved the performance of the watermark algorithm. The scheme can be
tested by applying various attacks.
To measure the robustness of our proposed scheme, the watermarked frame can be
subjected to a variety of attacks such as gamma correction, contrast adjustment, histogram
equalization, and jpeg compression.

4. CONCLUSION

A hybrid video watermarking scheme has been proposed in this paper. The algorithm
is implemented using 2-level DWT in conjunction with PCA transform. This scheme is
imperceptible and robust against several attacks. A binary watermark has been embedded into
LL and HH bands of level 2 of DWT block based PCA. The proposed scheme has a good
performance compared with previous schemes. As a future work, embedding the watermark
into higher levels of the wavelet transform will be investigated. Collecting other
transformations together to enhancement the performance of the proposed scheme against
geometric attacks will be studded.

177


REFERENCES

[1] M. K. Thakur, V. Saxena, and J. P. Gupta, “A Performance analysis of objective video
quality metrics for digital video watermarking”, 3rd IEEE International Conference on
Computer Science and Information Technology – ICCSIT ‘10, 9-11 July, 2010, pp.12-17,
Chengdu, China.
[2] S. Voloshynovskiy, S. Pereira, and T. Pun, “Watermark attacks”, Erlangen Watermarking
Workshop 99, October 1999.
[3] C.I. Podilchuk and E.J. Delp, “Digital watermarking: algorithms and applications”, IEEE
Signal Processing Magazine, Vol. 18, Issue 4, July 2001, pp. 33-46.
[4] P.W. Chan, M.R Lyu, and R.T. Chin, “A Novel scheme for hybrid digital video
watermarking”, IEEE Transactions on Circuits and Systems For Video Technology, Vol. 15,
No. 12, December 2005.
[5] G. Doërr and J.L. Dugelay, “A guide tour of video watermarking”, Signal Processing:
Image Commun., April 2003, Vol. 18, No. 4, pp. 263–282.
[6] Y. R. Lin, H.Y. Huang and W.H Hsu, “An embedded watermark technique in video for
copyright protection”, 18th International Conference on Pattern Recognition – ICPR ‘06, 20-
24 August 2006, pp. 795- 798, Hong Kong.
[7] C.V. Serdean, M.A. Ambroze., M. Tomlinson, and J.G. Wade, “DWT based video
watermarking for copyright protection, invariant to geometrical attacks”, IEE on Vision,
Image and Signal Processing, Vol. 150, Issue 1, 2003, pp. 51- 58.
[8] R. Chandramouli and N. Memon, “Analysis of LSB based image steganography
techniques”, in Proceedings International Conference on Image Processing, 7-10 October,
2001, Vol. 3, pp. 1019–1022,Thessaloniki, Greece.
[9] G. Langelaar, I. Setyawan, and R. Lagendijk, “Watermarking digital image and video
data”, IEEE Signal Processing Magazine, Vol. 17, No. 9, September 2000, pp. 20– 43.
[10] I. J. Cox, J. Kilian, F. T. Leighton and T. Shamoon, “Secure spread spectrum
watermarking for multimedia”, IEEE Transactions on Image Processing, Vol. 6, Issue 12,
1997, pp. 1673-1687.
[11] C.H. Li and S.S. Wang, “Transform-based watermarking for digital images and video”,
International Conference on Consumer Electronics –ICCE ‘99, 22-24 June, 1999, Los
Angeles, USA.
[12] J. Hussein and A. Mohammed, "Robust video watermarking using multi-band wavelet
transform", International Journal of Computer Science Issues, IJCSI, Vol. 6, Issue 1,
November 2009, pp. 44-49.
[13] T. D. Hien, Y.W. Chen, and Z. Nakao, “A robust digital watermarking technique based
on principal component analysis” International Journal of Computational Intelligence and
Applications, Vol. 4, No. 2, 2004, pp. 138-192.
[14] C.V. Serdean, M.A. Ambroze, M. Tomlinson and J.G. Wade, “DWT Based Video
Watermarking for Copyright Protection, Invariant to Geometrical Attacks”, Proceedings of
the 3rd International Symposium on Communication Systems Networks and Digital Signal
Processing – CSNDSP'02, Stafford, UK, 15-17 July 2002.
[15] Maher El'arbi, Chokri Amar, Henri Nicolas, "Video Watermarking Based on Neural
Networks", IEEE International Conference on Multimedia and Expo, ICME'06, pp.1577-
1580, 2006.
[16] Yang Gaobo; Sun Xingming; Wang Xiaojing, " A Genetic Algorithm based Video
Watermarking in the DWT Domain", IEEE, Digital Object Identifier
10.1109/ICCIAS.2006.295247, pp.1209-1212.

178


[17] Maher El'arbi, M. Ben Amar, C. Nicolas, H. " A Video Watermarking Scheme Resistant
to Geometric Transformations", IEEE International Conference on Image Processing,
ICIP'07, Vol. 5, pp 481– 484, 2007.
[18] Thai Duy Hien, Yen-Wei Chen, Zensho Nakao," PCA Based Digital Watermarking",
KES 2003, LNAI 2773, pp. 1427-1434, 2003.
[19] Yavuz E., Telatar Z., “Digital Watermarking with PCA Based Reference Images”,
ACIVS 2007, Springer-Verlag, Lecture Notes in Computer Science, 4678, pp.1014-1023,
2007.
[20] Xiangui Kang ,WenjunZeng ,and Jiwu Huang, " A Multi-band Wavelet Watermarking
Scheme ", International Journal of Network Security ,Vol 6 ,No 2, pp. 121–126, Mar 2008.
[21] G. B. Khatri and D. S. Chaudhari, “Digital Audio Watermarking Applications and
Techniques”, International journal of Electronics and Communication Engineering &
Technology (IJECET), Volume 4, Issue 2, 2013, pp. 109 - 115, ISSN Print: 0976- 6464,
ISSN Online: 0976 –6472.
[22] Fahd N. Al-Wesabi, Adnan Z. Alsakaf and Kulkarni U. Vasantrao, “A Zero Text
Watermarking Algorithm Based on the Probabilistic Patterns for Content Authentication of
Text Documents”, International journal of Computer Engineering & Technology (IJCET),
Volume 4, Issue 1, 2013, pp. 284 - 300, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.

179

Hybrid video watermarking technique by using dwt & pca

More Related Content

What's hot (18)

Viewers also liked (11)

Similar to Hybrid video watermarking technique by using dwt & pca (20)

More from IAEME Publication (20)

Recently uploaded (20)

Hybrid video watermarking technique by using dwt & pca