EE 5359Fall 2010 PROJECT PROPOSAL DIGITAL WATERMARKING Abrar Ahmed Syed 1000 61 4216 [email protected] Under the guidance of Dr. K. R. Rao.

EE 5359 Fall 2010

PROJECT PROPOSALDIGITAL WATERMARKING

Abrar Ahmed Syed1000 61 [email protected]

Under the guidance ofDr. K. R. Rao

Key Points

• What is digital watermarking ?• Why do we need digital watermarking ?• What are the different types of watermarking ?• What are its characteristics and requirements ?• What are the different techniques and schemes

used to watermark ?• Two forms of frequency domain watermarking

techniques in detail.

Key Points (Contd.)• What are the different types of attacks it is

susceptible to ?• One attack implementation ?• What are the ways of counter-attacking a

watermarking attack ?• What are the different laws and principles governing

watermarking ?• What are its drawbacks ?• What is its future perspectives ?

Watermarking• Embedding a digital signal (audio, video or image)

with information which cannot be removed easily is called digital watermarking. Figure 1 shows the block diagram of embedding digital watermark.

Figure 1: Block diagram of embedding digital watermark

Attacking Function

Detecting and Retrieving Function

Signal

Any Embedding Function ‘E’

Signal

Watermarking• As advances are made in field of communication,

it became necessary to cipher and decipher.• This led to discovery of stenography and

watermarking.• Stenography: Hiding information over cover.• Watermarking: Hiding information related to

cover.• Embedding is done by manipulating contents in

the signal itself and is made imperceptible.

Encoding For example sake, figure 2 shows the block diagram for audio watermark encoding.

Figure 2: Encoding block diagram of audio watermarking technique

FramingSpectral Analysis

DC Carrier Removal

Watermark Addition

Original Signal

Watermarked Signal

DecodingFor example sake, figure 3 shows the block diagram of audio watermark decoding.

Figure 3: Decoding block diagram of audio watermarking.

FramingSpectral Analysis

Watermark Processing

Watermarked Signal

Original Signal

Applications?• Ownership assertion: Watermarking is used to

establish ownership over the content

• Fingerprinting: In fingerprinting, watermarking is used to avoid illegal distribution of media publicly

• Authentication and integrity verification: Content which is protected by key verification should not be accessible without authentication.

Applications?• Content labeling: Bits embedded in data giving

extra information

• Usage control: To limit copies creation of copyrighted data, by blocking using watermark.

• Content protection: Visible watermark block is used for this purpose.

No universal technique to satisfy all of these.(Source: [2] Memon, N., & Wong, P. W. (1998). Protecting digital media content. Communications of the ACM, 41(7), 35-43.)

Classification or Types• Visible: Any text or logo to verify or hide content

Fw=(1-α)F+ α*W [12]

Fw = Watermarked Imageα= constant 0<= α<=1 (If 0, no watermark or if 1, watermark is present) F=original image W=watermark

# Watermarking done by using Bytescout watermarking software [20]

Classification or Types• Invisible: Hidden in the signal or content. Cannot be

perceived by human eye or ear. Usually used for authentication or security• Robust: Cannot be manipulated without

disturbing the host signal.

# Watermarking done by using Bytescout watermarking software [20]

Classification or Types• Fragile: Fails with even the slightest mod.• Public: Resists benign transformation but fails

malignant ones.• Capacity: Based on length of the embedded signal

- Zero-Bit: It denotes 1 or 0 based on whether watermark is present or absent.

Also called Italic zero bit.- N-Bit: Its N-Bit long. (m=m1, … m=mn ,

with n = | m | ) or M = {0,1}n Is modulated. Also called non-zero bit. [1]

Classification or Types

• Perceptibility: If the presence of watermark is evident in the host signal

• Imperceptibility: If original host and marked host are indistinguishable.

Classification or TypesPerceptible and Imperceptible Example [15]

(a) Original Image (b)Watermark Image(c) Perceptible Image (d)Imperceptible Image

Classification or Types• Embedding techniques:

-Spread spectrum: In this case, watermark is obtained by additive modification. It is robust.It has low information capacity due to host interference.-Quantization: Marking is done by quantizing. It is low robust. It has more data capacity-Amplitude modification: In this marked signal is embedded by additive modification. Similar to spread spectrum but is particularly done in spatial domain.

Techniques and Schemes

• Spatial domain: In this technique randomly selected subsets are modified. It is not reliable when subjected to filtering or lossy compressions- LSB coding: Least significant bit issubstituted with watermark- Predictive coding: Co-relation between adjacent images is used. It can be further improved by adding constant to difference in adjacent values.

Techniques and Schemes- Correlation based: Pseudo random noise (PN) with a pattern W(x, y) is added

to an image according toIw(x,y)=I(x,y)+ K * W(x,y) [12]

Iw(x,y) = Watermarked imageI(x,y)=Original imagek=gain factor

At the decoder the correlation between the random noise and the image is found. If the value exceeds

threshold, watermark is detected.

Techniques and Schemes- Patchwork: First ever watermarking scheme.

Image divided into two subsets. An operation is applied to these two subsets in the

opposite direction. If a[i] is the value of a sample at I in subset ‘A’ which is increased and b[i] is the value of the same sample in the subset ‘B’ which is decreased, then the difference

∑a[i]-b[i]= 2N for watermarked images = 0 otherwise1<=N<=∞

Techniques and Schemes

• Frequency domain: Values of lower frequency coefficients are altered. This technique is applied to the whole image.

- Discrete cosine transform (DCT): Converts data in spatial domain into cosine with different amplitudes in frequency domain. [11][12][18]

- Discrete wavelet transform (DWT): It decomposes signal into set of basic wavelets. Lower frequencies are then altered at different resolutions [11][12][18]

Attacks

“Digital watermarking is not as secure as date encryption. Therefore, digital watermarking is not immune to hacker attacks”.[5] The following are few forms of attacks•Basic: Take advantage of limitations in design

of embedding technique•Robustness: Attempts to diminish or remove presence of watermark•Presentation: Modifies the content of the host signal to detect watermark

Attacks

• Intrepretation: They find a situation where ownership certification is prevented.• Implementation: Attacks the detection software.• Removal: Includes denoising, lossy

compression, quantization, remodulation, collusion and averaging attack. Removes watermark from cover signal.• Geometrical: Instead of removing they distort

the watermark using spatial or temporal alteration of stego data.

Attacks• Cryptographic: Brute force attacks are used for

exhaustive search.

• Protocol, estimation, perceptual remodulation, copy and benchmarking

• Wavelet Based- Active: Hacker removes or spoils watermark- Passive: Hacker just identifies, does not damage- Forgery: Forges new watermark

Attacks• Wavelet based

- Collusion: Hacks different copies with different watermarks and joins them to make one single watermark.- Distortive: Hacker applies distortive transformation to make the watermark undetectable

Counter - Attacks

• Power spectrum condition (PSC)- When original signal is given, its power spectrum can be varied. The PSC can be shaped to resist estimation based attack.• Noise visibility function (NVF)

- Watermark is added in the form of noise. De-noising is used to derive noise. This noise becomes the watermark.

Privacy Laws• Privacy by design• Avoid embedding independently useful identifying

information directly in watermark • Provide notice to end users• Control access to reading capability• Respond appropriately when algorithms are compromised • Provide security and access controls for back end databases‑• Limit uses for secondary purposes• Provide reasonable access and correction procedures for

personally identifiable information

Project Goal• An image ‘A’ will be selected from a set of images of various sizes and color scales.• A watermark ‘B’ will be selected from a set of images of various sizes and color

scales.• The MSB of ‘B’ will be read and will be written on the LSB of ‘A’.• Thus, ‘A’ will be watermarked with ‘B’ resulting in a combined image ‘C’. The

technique used will be LSB technique.• LSB technique is a form of Spatial Domain Technique.• Noise will be added to image ‘C’ resulting in image ‘D’.• Both the images ‘C’ and ‘D’ will be compressed .• The compression will be done in the following steps.• The image is broken into 8X8 blocks of pixels. [19]• From left to right and top to bottom discrete cosine transform is applied to each

block• Blocks are compressed through quantization.• For reconstruction of image, decompression is done by inverse discrete cosine

transform. [11][12]• Once this is done, the whole process will be reversed, i.e. The compressed image will

be decompressed, the noise will be removed, the watermark will be removed, and the original watermark and the original image will be obtained.

• An analysis will be done on the original image before starting of any process and the final retrieved image at the end of the entire process.

• The tools used in this project’s implementation will be Matlab, Visual Studio, Stirmark and Bytescout. [20]

References• [1] R. Popa, “An analysis of steganographic techniques”, The Politehnica University of Timisoara,

Faculty of Automatics and Computers, Department of Computer Science and Software Engineering, Website: http://ad.informatik.uni-freiburg.de/mitarbeiter/will/dlib_bookmarks/digital-watermarking/popa/popa.pdf, 1998

• [2] N. Memon and P. W. Wong, “Protecting digital media content”, Communications of the ACM,

Vol. 41(7), pp 35-43, 1998 • [3] T. C. Lin and C. M. Lin, “Wavelet based copyright protection scheme for digital

images based on local features”, Information Sciences: an International Journal, Vol. 179(19), Sept. 2009.

• [4] G. Langelaar, I. Setyawan and R.L. Lagendijk, “Watermarking digital image and video data”,

IEEE Signal Processing Magazine, Vol. 17, pp 20-43, Sept. 2000 • [5] M. Yeung, B. Yeo and M. Holliman, “Digital watermarks: shedding light on the invisible”, IEEE

Micro, Vol. 18(6), pp 32-41. Nov. 1998 • [6] P. Vidysagar, S. Han, and E. Chang. "A survey of digital image watermarking techniques”, 3rd

IEEE International Conference on Industrial Informatics (INDIN 2005), edited by Dillon, T., Yu, X. and Chang, E., pp 495-502, Perth, Western Australia, 2005

References• [7] S. Voloshynovskiy, S. Pereira and T. Pun, “Attacks on digital watermarks: classification,

estimation-based attacks and benchmarks,” IEEE Commun. Mag., Vol. 39(8), pp. 2–10, Aug. 01 • [8] X. Jian-hui, W. Li-na and Z. Huan-guo, “Wavelet based denoising attack on image

watermarking”, Wuhan University Journal of Natural Sciences, Vol.10(1), pp. 279-83, Oct. 2005 • [9] A. Khan and A.M. Mirza, “Genetic perceptual shaping: utilizing cover image and conceivable

attack information during watermark embedding”. Inf. Fusion 8, 4, 354-365, Oct. 2007 • [10] I. J. Cox, J. Kilian, T. Leighton and T. Shamoon, “Secure spread spectrum watermarking for

multimedia”, IEEE Transactions on Image Processing, Vol. 6(12), pp 1673–1687, 1997 • [11] N. Ahmed, T. Natarajan, and K. R. Rao, "Discrete cosine transform", IEEE Trans. Computers,

Vol. 23(1), pp. 90-93, Jan. 1974. • [12] V. Britanak, K. R. Rao and P. Yip, “Discrete Cosine Transform: Properties, Algorithms,

Advantages, Applications”, Academic Press Publications, ISBN 978-0-12-373624-6, Boston, 1990.

• [13] M. Arnold, M. Schmucker and S. D. Wolthusen. “Techniques & Application of Digital Watermarking and Content Protection”, Artech House Publications, ISBN:1580531113, Boston, London, 2003

References• [14] I. J. Cox, M. L. Miller, J. A. Bloom, J. Fridrich and T. Kalker, "Digital Watermarking and

Steganography" (Second Edition), Morgan Kaufmann Publications, ISBN: 9780123725851, 2008 • [15] Jeng-Shyang Pan, Hsiang-Cheh Huang and L. C. Jain, “Intelligent Watermarking Techniques-

Series on Innovative Intelligence, Vol. 7”, World Scientific Publication, ISBN:9812387579, 2004 • [16] F. Jin, P. Fieguth, L. Winger, and E. Jernigan, “Adaptive Wiener filtering of noisy images and

image sequences,” Proc. IEEE Int. Conf. on Image Process, pp. 349–352, Sept. 2003 • [17] F. Mintzer and G. W. Braudaway, “If one watermark is good or more better?” IEEE, Int. Conf.

on Acoustics, Speech and Signal Processing- ICASSP, pp. 2067–2070, 1999 • [18] J. Cummins, P. Diskin, S. Lau and R. Parlett, “Stegnography and digital watermarking”,

School of Computer Science, The University of Birmingham, 2004 Website: http://www.cs.bham.ac.uk/~mdr/teaching/modules03/security/students/SS5/Steganography.pdf

• [19] Electrical and Computer Engineering, University of Victoria, British Columbia. Website: http://www.ece.uvic.ca/~aupward/w/watermarking.html

• [20] Bytescout watermarking software for visible and invisible watermarking. Website: http://bytescout.com/products/enduser/watermarking/watermarking.html

•

THANK YOUQUESTIONS & SUGGESTIONS

• Who questions much, shall learn much & retain much -Francis Bacon • Questions are guaranteed in life, answers aren’t. – English Proverb