Media hashing is an alternative to achieve many applications previously accomplished with watermarking. The major disadvantage of the existing media hashing technologies is their poor resistance to geometric attacks. In this paper, our aim is to propose a geometry-invariant image hashing scheme, which can be employed for copy detection and content authentication of digital images. Our scheme is mainly composed of two components: (i) mesh-based robust hash generation and (ii) hash database construction for error-resilient and fast matching. In addition, we further investigate the issues of robustness, error analyses, complexity, granularity, and scalability for the proposed image hashing system. Exhaustive experimental results obtained from benchmark attacks have con rmed the performance of the proposed method.
Ramarathnam Venkatesan, S.-M. Koon, Mariusz H. Jak