07-02-2013, 03:52 PM
Hybrid Scheme for Robust Digital Image Watermarking Using Dirty Paper
Trellis Codes
1Hybrid Scheme.pdf (Size: 216.76 KB / Downloads: 25)
Abstract
Digital images can be copied and redistributed without
the loss of fidelity. This leads to issues related with
copyright protection of digital images. Digital watermarking
is used to effectively tackle these security related issues. In
this paper a robust digital image watermarking scheme using
dirty paper trellis codes is proposed. To increase the
robustness and make the scheme survive more attacks the
basic scheme is modified into a hybrid scheme using spread
spectrum watermarking. First, a modified trellis code
wherein a given message is represented by many different
code words is used for embedding the watermark. Further,
another watermark based on spread spectrum is also found
and embedded. This hybrid scheme is more robust and
survives various attacks like additive noise, valumetric
scaling, histogram and de-synchronization.
Introduction
With the growth of multimedia services over the Internet the
fact that digital information can be copied and distributed
without loss of fidelity has led to development of tools for
asserting copyright of the data. Digital watermarking has
gained interest in the past decade for copyrighting of
multimedia data. The goal of watermarking is to hide a
message, either visible or invisible, in the multimedia data
and then access the message using extraction process to
prove the ownership of the data.
Watermarking is the process that embeds and extracts
the copyright information in multimedia data. The embedded
and hidden information may be media number, copyright
message, text, binary image or any other digital data that
protects the copyright of the owner. In order to be effective,
a watermark should have following characteristics
depending on the application.
Dirty Paper Trellis [DPT] Code Based
Watermarking
Figure 1 illustrates the basic idea of informed embedding
where the embedder considers the information available in
the cover work (the image to be watermarked). The
watermark embedding process is split into three steps: first it
encodes a message m as a watermarked pattern Wm and then
modifies this pattern to generate Wa and finally adds this
pattern to the cover image Co, to produce watermarked work
Cw. The modification of the pattern from Wm to Wa is
dependent on the coding process and the actual cover work.
Spread Spectrum Based
Watermarking
At times the main requirement of embedding the watermark
into the host image is that the watermark should be invisible
and robust to all possible intentional and unintentional
attacks. Spread spectrum (SS) based watermarking fulfills
both the requirements up to a great extent and is used for
watermarking digital images. In spread spectrum
communication, one transmits a narrowband signal over a
much larger bandwidth such that signal energy present in
any single frequency is very small. Nevertheless, watermark
verification process knows the location and content of the
watermark, it is possible to concentrate these many weak
signals into a single output with high signal to noise ratio. To
destroy, such a watermark, it requires noise of high
amplitude to be added to all frequency regions. The
embedding procedure involves addition of pseudo random
noise to the watermark pattern and detection utilizes
correlation sum after subtracting the original image from the
marked image [9] [10]. Figure 4 and Figure 5 show the
embedder and the decoder for such a scheme.
Conclusions
A hybrid method is proposed here for copyright
protection of digital images. The scheme embeds two
watermarks; first one by using the dirty paper trellis coding
and the second one with the spread spectrum. The two
techniques of embedding watermarks are found to
complement each other as far as robustness against various
attacks is concerned. The hybrid scheme being a
combination of the two schemes survives all the attacks
survived by each scheme independently.