A robust object-based video authentication system

We propose an object-based video authentication system, which employs a combination of watermarking, error correction coding (ECC) and digital signature techniques, to protect the authenticity between video objects and their associated backgrounds. In this system, a set of angular radial transformation (ART) coefficients is selected as the feature to represent the video object and the background respectively. ECC and cryptographic hashing are applied to those selected coefficients to generate the robust watermark. This semifragile watermark is then embedded into the objects frame by frame before MPEG4 coding. The experimental results demonstrate that our system is robust against MPEG4 compression, object segmentation errors, and some common object-based video processing such as object translation, rotation, scaling.