Docstoc

Digital Watermarking

Document Sample
Digital Watermarking Powered By Docstoc
					             Digital Watermarking




                        by

              Chaelynne M. Wolak
             wolakcha@scsi.nova.edu




A paper submitted in fulfillment of the requirements
        for DISS 780 – Assignment Twelve




   School of Computer and Information Sciences
          Nova Southeastern University

                     July 2000
                                        Abstract
Copyright abuse is the motivating factor in developing new encryption technologies. One
such technology is digital watermarking. The focus of this proposed research paper will
detail digital watermarking for multimedia applications. It will be complimented with a
ten minute PowerPoint presentation. Areas that will be covered are definition of digital
watermarking, purpose, techniques, and types of watermarking attacks. Lastly, the
current laws in place for digital copyright will be briefly detailed along with a review of
the fair use guidelines and the future of digital watermarking technology.




                                             ii
                             Table of Contents
Abstract                       ii

Problem Statement and Goal     1

Relevance and Significance     2

Barriers and Issues            2

Approach                       3

Annotated Bibliography         4




                                    iii
                                                                                             1


                                 Digital Watermarking

Problem Statement and Goal

       The desire for the availability of information and quick distribution has been a

major factor in the development of new technology in the last decade (Zhao, Koch, &

Luo, 1998). There is the increased use of multimedia across the Internet. Multimedia

distribution has become an important way to deliver services to people around the world

(Arn, Gatlin, & Kordsmeier, 1998, December). It is commonly applied in Internet

marketing campaigns and electronic commerce web sites.

       Due to the growing usage of multimedia content on the Internet, serious issues

have emerged. Counterfeiting, forgery, fraud, and pirating of this content are rising (Lan

& Tewfik, 1999, October 30 - November 5). Virtually anyone with a sound card, scanner,

video frame grabbers, or multimedia authoring systems allow them to incorporate

copyrighted material into presentations, web designs, and Internet marketing campaigns.

Consequently, copyright abuse is rampant among multimedia users, who are rarely

caught (Gatlin, Arn, & Kordsmeier, 1999, July/August).

       This copyright abuse is the motivating factor in developing new encryption

technologies (Zhao et al., 1998). One such technology is digital watermarking. The focus

of this proposed research paper will detail digital watermarking for multimedia

applications. It will be complimented with a ten minute PowerPoint presentation. Areas

that will be covered are definition of digital watermarking, purpose, techniques, and types

of watermarking attacks. Lastly, the current laws in place for digital copyright will be

briefly detailed along with a review of the fair use guidelines and the future of digital
                                                                                            2


watermarking technology. This research paper and PowerPoint presentation can be used

to further one’s understanding of a type of multimedia encryption technique.

Relevance and Significance

       Multimedia usage has developed from the need in satisfying human desires in

societies. It is this desire that relies on communication, personal interaction, and

entertainment (Lan & Tewfik, 1999, October 30 - November 5). Thus, the emphasis has

mostly been placed on making information available and on transmitting and

manipulating that information. Protecting information and content has not received the

attention that it deserves (Zhao et al., 1998).

       It has only been recently that fair use guidelines and copyright laws regarding

multimedia usage have been established. Any extensive research in this area is non-

existent (Arn et al., 1998, December). However, even the current copyright laws are

inadequate for dealing with all this digital data (Memon & Wong, 1998). Digital

watermarking seems to be the only potential encryption technology to provide protection

even after data is decrypted (Zhao et al., 1998).

Barriers and Issues

       Several barriers prevent digital watermarking from being effective and

widespread. First, there is not a foolproof protection scheme while making the

watermarks imperceptible. Thus, absolute robustness is impossible (Zhao et al., 1998).

       Second, it is difficult to offer an “off-the-shelf” solution to the mass market.

Although, there is significant interest in this technology from OEMs and system

integrators, there is no infrastructure or protocols. Besides, the legal status of watermarks
                                                                                                3


used as evidence in lawsuits involving intellectual property has not been tested (Zhao et

al., 1998).

        Lastly, out of fear of piracy, many professional photographers, artists, and other

content creators still do not put their work out in digital format. Thus, the market for this

new encryption technology is not widespread. Usually a new technology goes through a

dormancy period before widespread adoption (Zhao et al., 1998).

Approach

        The goal is the research paper that will detail digital watermarking for multimedia

applications. It will be complimented with a ten minute PowerPoint presentation. Areas

that will be covered are definition of digital watermarking, purpose, techniques, and types

of watermarking attacks. Lastly, the current laws in place for digital copyright will be

briefly detailed along with a review of the fair use guidelines and the future of digital

watermarking technology.

        The first step in accomplishing this is performing a thorough search into digital

watermarking. Next, the preliminary proposal will be written. It will complete the review

of literature and methodology of this paper. In addition, it will detail the production of the

PowerPoint presentation. During the production of the PowerPoint presentation, two

individuals will be asked for their input of it. Comments and observations from these

individuals will be incorporated into the final design. All of this information will be

placed into the final research paper.

        Lastly, the PowerPoint presentation will be completed as well as this research

paper on digital watermarking. It will be completed in February 2001 for the final

requirement for DISS 880.
                                                                                              4


Annotated Bibliography

Arn, J., Gatlin, R., & Kordsmeier, W. (1998, December). Multimedia copyright laws
       and guidelines: Take the test. Business Communication Quarterly, 32-39.

These authors are from the University of Central Arkansas. They discuss the issue of
using multimedia content for classroom teaching. Since copyright guidelines affecting
multimedia production have only been recently established, a survey was conducted to
determine if members of the Association of Business Communication understood these
latest guidelines. Results from this survey indicated an average grade of 71.082 percent
with a standard margin of error of 1.706.

The authors further detail the fair use guidelines. In addition, they provide examples of
permitted uses of copyrighted multimedia materials as it relates to an education setting.
This information can be used for the conclusion of the research paper describing fair use
guidelines when using multimedia content.

Bartolini, F., Bini, G., Cappellini, V., Fringuelli, A., Meucci, G., Piva, A., & Barni,
       M. (1999, June 7 - 11). Enforcement of copyright laws for multimedia through
       blind, detectable, reversible watermarking. Paper presented at the Proceedings
       of the IEEE international conference on multimedia computing and systems,
       Florence, Italy.

All authors are from the University of Florence with the exception of M. Barni, who is
from the University of Siena. They propose a new watermark technique based on the
decoding process. A new system called the Electronic Copyright Management System
(ECMS) is presented using this watermark technique to help enforce copyright laws in
open-network environments.

The technique works like this. As the digital information travels across the network, no
matter how many times it has changed hands, the focus is put on the single piece of data
and the information borne by the watermark. At any time, the person who is using this
digital information can demonstrate he/she is making a legal use of the information by
reading the watermark. In addition, a mechanism is provided that enables authorities to
prove the illegal use of any digital information.

Berghel, H. (1998). Digital watermarking makes it mark. netWorker: The craft of
      network computing, 2(4), 30-39.

This author is a professor of computer science at the University of Arkansas. He details
the history of watermarking back to the Middle Ages. Due to the huge popularity of the
World Wide Web and the commercialization of offering multimedia resources through
these digital networks for a profit, there is now a strong interest in protecting ownership
rights.
                                                                                           5


The author describes the purpose of watermarking, the litmus test for watermarks, the
techniques, and the limitations of it. He also goes on to briefly describe the future of
digital watermarking. This article will be helpful in the areas of defining watermarking
and its purpose as well as what the future holds for it.

Busch, C., Funk, W., & Wolthusen, S. (1999, January, February). Digital
       watermarking: From concepts to real-time video applications. IEEE
       Computer Graphics & Applications, 19(1), 25-35.

The authors are from the Fraunhofer Institute for Computer Graphics. They present an
algorithm for watermarking and monitoring video streams in a TV-broadcasting
environment. In addition, it has the capability of surviving MPEG-2 compression of high
quality, real-world video sequences without degrading the watermark's quality.

The algorithm is frame-based. Thus, during the watermark embedding and retrieval, the
algorithm separately processes each frame of the uncompressed video stream. This is
similar to the Koch-Zhao algorithm, but it is slightly modified. The modification includes
new discrete cosine transform (DCT) and inverse DCT routines that are optimized for
real-time implementation.

Chandramouli, R., & Memon, N. (2000, March 27 - 29). How many pixels to
     watermark? Paper presented at the Proceedings of the international
     conference of information technology: Coding and computing, Las Vegas,
     Nevada.

R. Chandramouli is from Iowa State University and N. Memon is from the Polytechnic
University. These authors present a watermark technique based on the watermark
detection method. By developing this technique, the number of watermark pixels can be
reduced almost by a factor of two. This reduction leads to robust watermarking systems.
In addition, there is significant cost savings since the need for redundant watermarking
systems can be reduced.

By decreasing the amount of watermark pixels, there is an increased immunity to host
signal attacks such as cropping. The watermark detection algorithm used to develop this
new technique is called sequential. Their new watermark technique was tested on a fixed
sample size. The results showed that the sequential detection is superior, making it the
preferred candidate for watermark detection in image/video databases.

Collberg, C., & Thomborson, C. (1999, January 20 - 22). Software watermarking:
       Models and dynamic embeddings. Paper presented at the Proceedings of the
       26th ACM SIGPLAN-SIGACT on principles of programming languages,
       San Antonio, Texas.

Both authors are from the department of Computer Science at the University of
Auckland. They discuss the attacks on watermarking systems. There are five main
                                                                                                6


attacks. They are statistical, distortive, collusive, cropping and additive. The authors also
discuss countermeasures against these attacks.

The authors explain various software watermarking approaches. Software watermarking
is the process of embedding a large number into a program. Thus, the number can be
reliably retrieved after the program has been attacked. In addition, the embedding of the
number is imperceptible to those that are trying to attack. This software approach is a
new watermarking technique.

Cox, I., Miller, M., & Bloom, J. (2000, March 27 -29). Watermarking applications
        and their properties. Paper presented at the Proceedings of the international
        conference on information technology: Coding and computing, Las Vegas,
        Nevada.

These authors are from the NEC Research Institute. They present various digital
watermarking applications. There are seven applications of watermarking. They are
broadcast monitoring, owner identification, proof of ownership, authentication,
transactional watermarks, copy control, and covert communication. Each of these
applications may have different watermark requirements. Therefore, these authors feel
that there should not be a single set of standards for all watermarking systems.

This work contrasts the work done by Fred Mintzer, Gordon Braudaway, and Alan Bell
(1998). Those authors feel that for applications that do not depend strictly on standard
techniques, opportunities for standards would better serve the user community. However,
the authors from NEC Research Institute claim one size does not fit all.

Dittmann, J., Mukherjee, A., & Steinebach, M. (2000, March 27 - 29). Media-
      independent watermarking classification and the need for combining digital
      video and audio watermarking for media authentication. Paper presented at
      the Proceedings of the international conference on information technology:
      Coding and computing, Las Vegas, Nevada.

These authors are from the German National Research Center for Information
Technology. The most important parameters of digital watermarks are robustness,
security, transparency, complexity, capacity, and verification. Based on these parameters,
the authors suggest a watermarking solution that is robust and guarantees authenticity and
integrity. Their technique is called content-fragile watermarking.

Content-fragile watermarking uses a robust watermark to embed content information. It is
this embedded information that can be compared to the actual work. If any changes are
made to the work, a warning message is displayed. The fragility portion of this technique
is about losing equality of extracted and embedded content. This is another proposal for
watermarking techniques.
                                                                                           7


Dittmann, J., Stabenau, M., & Steinmetz, R. (1998, September 13 - 16). Robust
      MPEG video watermarking technologies. Paper presented at the Proceedings
      of the 6th ACM international conference on multimedia, Bristol, United
      Kingdom.

The authors from the German National Research Center for Information and Darmstadt
University of Technology present two watermarking techniques for MPEG video. They
discuss the requirements for MPEG video watermarking such as robustness against high
compression rates, robustness against scaling, labeling of every single video frame,
ensuring correct decoding of the frame sequences, and runtime, performance for
streaming video. Two algorithms are discussed such as the Zhao Koch and the Fridrich.

The intended audience is for those in the business of video production. A final video
production is based on one's individual ideas and unique intellectual creation. However,
with digital representation of the video, problems of unauthorized taping, reading,
manipulating, or removing of data can be experienced. This article provides additional
information on the types of watermarking techniques.

Fread, J., Stalnaker, A., & Baughman, L. (2000, April 28). Digimarc and
       creativepro.com launch marcstation digitial watermarking e-service. Electric
       Library: Business Wire [Online]. Available: http://www.elibrary.com [2000,
       July 6].

John Fread and Laura Baughman are from Digimarc Corporation. Anne Stalnaker is from
Creativepro.com. Digimarc Corporation develops watermarking techniques and
applications. Creativepro.com is an online resource for creative professionals. New
services from Creativepro.com allow members to quickly and easily embed watermarks
into their works. This is done via Digimarc's newest watermarking technique called
MarcStation.

MarcStation allow users to embed their digital images with their unique Digimarc ID that
links to vital information such as copyright ownership, usage restrictions, and licensing
instructions. The watermark is intricately woven into the fabric of the image. Thus, it
stays with that image wherever it may be distributed. This is a vital component in
identifying, tracking, and marketing content on the Web.

Gatlin, R., Arn, J., & Kordsmeier, W. (1999, July/August). AACSB deans'
       understanding of multimedia copyright laws and guidelines. Journal of
       Education for Business, 74(6), 368-371.

These authors are from the University of Central Arkansas. Another questionnaire was
conducted to measure AACSB deans' awareness regarding copyright laws and the usage
of multimedia content. The grade from this questionnaire was 68.655 percent with a
standard margin of error of 1.086.
                                                                                            8


This study compared to one done earlier with the members of the Association for
Business Communication prove that there is a lack of understanding of the copyright
guidelines for multimedia production. Here again, a discussion of the fair use guidelines
was presented.

Gerhardt-Powals, J., & Powals, M. (1999, June 27 - 30). The digital millennium
      copyright act: An international assault on fair use? Paper presented at the
      Proceedings of the 4th annual SIGCSE/SIGCUE on innovation and
      technology in computer science education, Krakow, Poland.

Jill Gerhardt-Powals, Ph.D is from Richard Stockton College of New Jersey. Matthew
Powals is an attorney at law. Both of these authors discuss the Digital Millennium
Copyright Act (called The Act) signed in November 4, 1998 by president Clinton. The
Act includes new rules for downloading, sharing, or simply viewing copyrighted
materials online.

This new law is not widely accepted. The software and entertainment industries feel they
can now market their products online without the fear of piracy. However, the academics,
educators, and researchers view this new law as a set back. The traditional access and use
of information, no matter what the intentions, is now illegal.

Grimm, R. (2000, February 7). Verance corporation awarded eleventh digital
     watermarking patent. Electric Library: Business Wire [Online]. Available:
     http://www.elibrary.com [2000, July 6].

Randy Grimm is from Verance Corporation. Verance Corporation just received its 11th
patent for digital watermarking technology. This latest patent addresses ways in which a
Verance watermark is embedded into broadcasted content. Verance claims this newest
watermark can generate increased consumer awareness and purchase behavior in
response to targeted commercials and programming.

This work is related to the article on Digimarc, who developed an "off-the-shelf" solution
for members at Creativepro.com. Both companies, Verance and Digimarc, are playing
key roles in developing multimedia data protection applications that can be used on the
Internet.

Kahng, A., Lach, J., Mangione-Smith, W., Mantik, S., Markov, I., Potkonjak, M.,
     Tucker, P., Wang, H., & Wolfe, G. (1998, June 15 - 19). Watermarking
     technique for intellectual property protection. Paper presented at the
     Proceedings of the 35th annual conference on design automation conference,
     San Francisco, California.

These authors are from the computer science and electrical engineering departments of
the University of California Los Angeles and the University of California San Diego.
They have proposed a new watermarking technique for intellectual property called the
canonical technique. This technique is centered on the use of constraints to sign the
                                                                                                 9


output of a given design synthesis. In addition, this watermarking technique can be
applied not only to software but also to hardware.

Watermarking for intellectual property imposes a much stronger constraint than one just
for images. It must remain functionally correct at all times. The key behind this technique
is imposing a set of additional constraints during the design and implementation of
intellectual property, so that it is uniquely encoded with the signature of the author.

Kankanhalli, M., & Ramakrishnan, K. (1998, September 13 - 16). Content based
     watermarking of images. Paper presented at the Proceedings of the 6th ACM
     international conference on multimedia, Bristol, United Kingdom.

The authors from the National University of Singapore and Indian Institute of Science
describe an approach to watermarking images. There are two classifications of image
watermarking. They are spatial domain and transform domain methods. These methods
are used against the 16 different image watermark attacks described in the article.

A new image watermark using the advanced spatial domain method is proposed by the
authors. It is called the human visual system (HVS). This new method is based on
classifying a region of the image based on its sensitivity or tolerance to noise. This is just
one of the newest forms of watermarking techniques.

Lan, T., & Tewfik, A. (1999, October 30 - November 5). Fraud detection and self
      embedding. Paper presented at the Proceedings of the seventh ACM
      international conference (part 2) on multimedia, Orlando, FL.

Both authors are from the Electrical and Computer Engineering department of the
University of Minnesota. The issue these authors address is fraud detection in multimedia
distribution. They discuss three high capacity data embedding watermarking systems
such as private, semi-private, and public.

Several tests were conducted to determine the robustness, capacity, and visual distortion
of these type of high capacity data embedding system. This information further enhances
the knowledge in watermarking techniques.

Memon, N., & Wong, P. W. (1998). Protecting digital media content.
     Communications of the ACM, 41(7), 35-43.

Nasir Memon is an associate professor of computer science at Northern Illinois
University. Ping Wah Wong is the manager of the Internet Imaging Operation at Hewlett
Packard. Both authors feel that the current copyright laws are inadequate for dealing with
digital data. Therefore, there is an increase interest in copyright protection such as digital
watermarking.

The article details the purpose of having a watermark signal such as ownership assertion,
fingerprinting, authentication and integrity verification, content labeling, usage control,
                                                                                            10


and content protection. In addition, the article explains the watermark insertion process
and techniques. This article can be used to explain why digital watermarking is popular
due to all of its purposes and some of the latest watermarking techniques.

Mintzer, F., Braudaway, G., & Bell, A. (1998). Opportunities for watermarking
      standards. Communications of the ACM, 41(7), 57-64.

The authors who are employed at IBM discuss the opportunities for watermarking
standards. One of the barriers to digital watermarking is not being able to provide an "off-
the-shelf" solution. This is because there are many integrators and companies with
application specific watermarking techniques. In addition, watermarking is such a new
field that it is often misunderstood or not accepted.

The authors benchmark various watermark techniques for robustness, unintentional
attacks, and intentional attacks. They have narrowed the field to three main application
areas where the technical requirements are similar. This benchmarked data helps
determine the best approach to digital watermarking.

Mohanty, S. P., Ramakrishnan, K. R., & Kankanhalli, M. (1999, October 30 -
     November 5). A dual watermarking technique for images. Paper presented at
     the Proceedings of the seventh ACM international conference (part 2) on
     multimedia, Orlando, Florida.

Saraju Mohanty and K. R. Ramakrishnan are from the Indian Institute of Science. Mohan
Kankanhalli is from the National University of Singapore. These authors developed a
dual watermarking technique. This technique places a visible and invisible watermark on
an image. The dual watermark serves two main purposes. First, it establishes the owner's
right to the image. Second, it detects the intentional and unintentional tampering of the
image. This technique can work for images in color or in the gray scale.

Even if the visible watermark is removed (by an attack), there is the invisible one as the
backup. The visible watermark is inserted into the original image while the invisible
watermark is added to it. Therefore, it is a watermark within a watermark creating a dual
watermarked image. This is another method of developing robust watermarking
techniques.

Nikolaidis, N., & Pitas, I. (1999, June 7 - 11). Digital image watermarking: An
       overview. Paper presented at the Proceedings of the IEEE international
       conference on multimedia computing and systems, Florence, Italy.

Both authors are from the Artificial Intelligence and Information Analysis Laboratory at
the Aristotle University of Thessaloniki. These authors present an overview to digital
watermarking. Image watermarking is a new challenging field that involves principles
from diverse disciplines like communications, signal processing, encryption, and
steganography. Even though there have been many watermarking techniques developed,
                                                                                              11


there is not one that is robust enough to all possible attacks or to the image processing
operation.

Due to the financial implications of the new copyright laws, there is an increased effort to
research a viable digital watermarking solution. However, for this solution to be accepted
worldwide, the legal status of it must be clarified and a trusted central authority has to be
established who maintains the regulations and operation of the watermarking framework.

Praun, E., Hoppe, H., & Finkelstein, A. (1999, August 8 - 13). Robust mesh
      watermarking. Paper presented at the Proceedings of the SIGGRAPH 1999
      annual conference on computer graphics, Los Angeles, California.

These authors from Princeton University address the problem of robustly watermarking
3D models. They present a method of encoding information in the model geometry by
displacing the vertices. This method is imperceptible to the human eye. The watermark is
hidden within the significant features of the 3D model. It can be identified using a multi-
solution approach.

This method has proven robust to various watermark attacks. The attacks ranged from
vertex reordering, the addition of noise, cropping, smoothing, simplification, and
insertion of a second watermark. The work done by these authors was also supported by
Microsoft Research and the National Science Foundation.

Samuelson, P. (1999). Good news and bad news on the intellectual property front.
     Communications of the ACM, 42(3), 19-24.

Pamela Samuelson is a professor of Information Management and of Law at the
University of California. She details the three legal initiatives as they relate to digital
information. They are the following: Digital Millennium Copyright Act, the Collections
of Information Antipiracy Act, and the Article 2B of the Uniform Commercial Code.

She warns that even though digital technology may pose threats such as fraud, piracy,
duplicating, etc., that policymakers should be careful not to overreact. By adopting
overbroad or imbalanced rules may benefit only some industries, but it is at the cost of
others. It could prove detrimental to those in emerging markets and the public.

Voyatzis, G., & Pitas, I. (1999, January/February). Protecting digital-image
      copyrights: A framework. IEEE Computer Graphics & Applications, 19(1),
      18-24.

Both authors are from the University of Thessaloniki. The demand for watermarks to
remain robust under digital imaging operations has not been successful. This is a
remaining problem for many watermarking techniques. The authors present a framework
for future algorithms and techniques for digital watermarking.
                                                                                           12


The basic watermarking procedures such as generation, embedding, and detection are
discussed. In addition, watermark demands such as perceptual quality preservation,
trustworthy detection, computational efficiency, and robustness to digital processing are
detailed. Lastly, the intentional attacks on watermarks are described. These attacks
include extraction, detection, statistical extraction, multiple watermarking, watermarking
with arbitrary keys, and private key loss.

Yeung, M., Yeo, B., & Holliman, M. (1998, November/December). Digital
      watermarks: Shedding light on the invisible. IEEE Micro, 18(6), 32-41.

These authors are from the Microcomputer Research Labs of Intel Corporation. They
discuss the purpose of digital watermarking, watermarking techniques, and digital
imaging devices. The five main purposes are evidence of ownership, fingerprinting,
tracing and infringement, copy control, and labeling and metadata insertion. The
discussion of techniques includes spatial domain watermarking, transform domain
watermarking, and the fragile watermarking technique.

Lastly, digital imaging devices can create watermark schemes. In fact, the authors believe
they provide stronger protection when the watermark is incorporated into the image
capture device. Examples of digital imaging devices include digital cameras and
scanners.

Zhao, J., Koch, E., & Luo, C. (1998). In business today and tomorrow.
       Communications of the ACM, 41(7), 67-71.

The authors, whom are the President and co-founders of MediaSec Technologies, focus
on digital watermarking encryption technology. Protecting information and content has
not received the attention it deserves compared to making information available anytime.
The authors go on to define digital watermarking and how it can be used in business
opportunities. In addition, they also discuss the various companies who specialize in this
technology.

The intended audience is for those who are in the business of digital and online
publication and broadcasting. This article provides the foundation for the introduction
and relevance. In addition, it can also be used for finding other sources that specialize in
the area of digital watermarking.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:25
posted:11/8/2011
language:English
pages:15