Research On Anti-Collusion Digital Fingerprinting Coding And Detection

Rapid developments in network technologies and multimedia processing have facilitated the distribution and sharing of multimedia. It is a great challenge to protect multimedia from illegal alteration, repackaging, and unauthorized redistribution. How to efficiently protect copyright of digital products has become a crucial issue that calls for immediate solution. Digital fingerprinting is an emerging technology that offers proactive post-delivery protection of copyrighted content. It labels each distributed copy with the corresponding user’s forensic mark, which can be used to trace traitors who illegally redistribute their copies and prevent the spread of illegal users, so it can reach the effective copyright protection of digital products.Current digital fingerprint research especially focused on fingerprinting code, collusion attacks, fingerprint detection and distribution agreements and so on. This paper is mainly based on two aspects:fingerprint detection and anti-collusion fingerprinting code, our work and research results include:First, based on this latest of Tardos fingerprinting scheme, we proposed a two-level Tardos fingerprinting scheme. For the reduction of computational costs required for the detection, a two-level Tardos code structure is introduced on the number of operations for calculating correlation scores. One is for group ID, the other is for user ID within each group. Furthermore, using the central limit theorem for correlation scores, we present a proper threshold for detection. The performance evaluation and analysis prove that our proposed scheme is more efficient.Second, theoretical analysis and experiment simulations of hardware detection, it indicates that the detection performance heavily relies on the threshold fixed at the beginning. And this is the main problem of the hard detection. In this paper, from the viewpoint of the two ideals(ensuring detection rate and false detection rate), we gave a method for determining the threshold and improved the threshold detection. Theoretical analysis and simulation results showed that the improved threshold detection is better than the oringal one for the detection of conspirators.