Year: 2005
Type: dissertation
Abstract: Digital fingerprinting is a method by which a copyright owner can uniquely \nembed a buyer-dependent, inconspicuous serial number (representing the fingerprint) \ninto every copy of digital data that is legally sold. The buyer of a legal copy is \nthen deterred from distributing further copies, because the unique fingerprint can be \nused to trace back the origin of the piracy. The major challenge in fingerprinting is \ncollusion, an attack in which a coalition of pirates compare several of their uniquely \nfingerprinted copies for the purpose of detecting and removing the fingerprints. \nThe objectives of this work are two-fold. First, we investigate the need for robustness \nagainst large coalitions of pirates by introducing the concept of a malicious \ndistributor that has been overlooked in prior work. A novel fingerprinting code that \nhas superior codeword length in comparison to existing work under this novel malicious \ndistributor scenario is developed. In addition, ideas presented in the proposed \nfingerprinting design can easily be applied to existing fingerprinting schemes, making \nthem more robust to collusion attacks. \nSecond, a new framework termed Joint Source Fingerprinting that integrates the \nprocesses of watermarking and codebook design is introduced. The need for this new \nparadigm is motivated by the fact that existing fingerprinting methods result in a \nperceptually undistorted multimedia after collusion is applied. In contrast, the new \nparadigm equates the process of collusion amongst a coalition of pirates, to degrading \nthe perceptual characteristics, and hence commercial value of the multimedia in question. \nThus by enforcing that the process of collusion diminishes the commercial value \nof the content, the pirates are deterred from attacking the fingerprints. A fingerprinting \nalgorithm for video as well as an efficient means of broadcasting or distributing \nfingerprinted video is also presented. Simulation results are provided to verify our \ntheoretical and empirical observations.
Author: William Luh
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2255589506