A Robust Content in DCT Domain for Image Authentication

Discrete cosine transform (DCT) coefficients of a natural image can be statistically modeled as the generalized Gaussian distribution (GGD) whose shape parameters can be estimated by the maximum likelihood (ML) estimation method. This paper presents a novel image authentication scheme based on the robust content in DCT domain, where a new concept robust feature is extracted using the modified ML principle. Our simulation results show that the proposed method can resist most content-preserving operations such as JPEG compression, brightness enhancement, histogram equalization, filtering, scaling, and rotation with small angles.