With the advent of multimedia technology and the popularity of
Internet communications, there has been great interest in using
digital watermarks for the purpose of copy protection,
content authentication, and data restoration.
Digital watermarking technology allows the content
owner to embed a secret signature, i.e., watermark,
into the host content for many applications.
For digital watermarking, a major challenge lies in the
confident verification of the embedded watermark,
even after the watermarked content undergoes various forms
of unintentional or malicious modification.
Block diagrams of a typical watermark encoder and decoder, respectively:
In previous work, a more robust digital
image watermarking algorithm -- equalized spectrum watermarking -- was proposed.
Experimental results demonstrated improved robustness compared to
conventional spread-spectrum watermarking techniques.
In this follow-up project, we incorporated further improvements into the
algorithm.
For transform-domain algorithms, the discrete cosine
transform (DCT) is commonly
used for transforming the host image before embedding the watermark.
Due to the low-pass characteristic of most images, the DCT coefficients
generally vary in amplitude throughout the image spectrum.
This low-pass nature is an advantage
for many transform coders, but it does not facilitate a reliable watermark
extraction for many watermarking algorithms.
With equalized-spectrum watermarking,
a simple, invertible permutation operator is used to equalize the
transform coefficients before watermarking.
In this project, we reduced the perceptibility of
equalized-spectrum watermarking by using human
visual models. The visual mask was generated using Watson's model,
which is a block-based discrete cosine transform based visual
model. The generated mask was applied to the watermarked image as
a post-watermarking process. An existing masking equation, which
was commonly used for spatial masks, was modified such that it
could be used for the block-based discrete cosine based visual
mask. The proposed method not only reduced the perceptibility of
the watermark, but also made the system more robust.
Publications:
-
Neema K. Shetty and Jeffrey J. Rodriguez,
"Equalized-Spectrum Watermarking Using Perceptual Modeling,"
2006 IEEE Southwest Symp. on Image Analysis and Interpretation,
March 26-28, 2006, Denver, Denver, CO, pp. 1-5. [ PDF ]