A COMPLEXITY-THEORETIC FOUNDATION FOR
DIGITAL IMAGE
WATERMARKING SYSTEMS
This material is based upon work supported by the National
Science Foundation under Grant IIS-0242435.
Any opinions, findings and conclusions or recomendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF).
A paper watermark is an imprint stamped on important paper documents to
guarantee their authenticity. The ease of duplicating, manipulating, and
transmitting information on the World Wide Web has made protection of digital
documents an urgent research topic. The emerging field of digital watermarking
seeks to invent methods of embedding some information (watermark) into
a digital document so that the information is conspicuous and yet difficult
to fake or remove. Digital watermarks can be used for copyright protection,
fingerprinting, digital signatures, advertising, annotations and captioning,
access control, covert communications, and digital warfare.
This project is a continuation of the PI's ongoing research on the foundations of digital watermarking, which has been supported by NSF and DoD.
It aims to develop a firm complexity-theoretic
foundation for digital image watermarking systems which accurately
reflects two fundamental constraints on the insertion and extraction algorithms, namely that they must be fast and also preserve the perceptual fidelity of their inputs. It will allow researchers to characterize
image watermarking systems in a precise manner according to four important criteria: capacity, efficiency, robustness, and transparency.
Specifically, the following activities are planned:
- proving or disproving existence of computationally secure image watermarking systems under various formal models and characterizations of the insertion, extraction, and tampering algorithms used in watermarking systems;
- characterizing the class of algorithms that preserve perceptual fidelity of input images under various distortion metrics that are supported by current theories of vision (these algorithms are called hiding functions);
- designing and teaching a course in digital watermarking at Santa Clara University;
- developing a real-world application using fragile watermarks.
Research
Education
Software
Outreach