A Robust Zero-watermarking Algorithm Against Geometric Attacks with Perceptual Hashing

Taking into account the information security problem in medical image, the paper put forward a new robust zerowatermarking algorithm with three-dimensional (3D) discrete Fourier transform (DFT) and perceptual hashing. The medical volume data was made by 3D DFT. The zero-watermarking algorithm selected the real part of 3D DFT coefficients (4*4*4) to generate the watermarking extraction key sequence (64-bit). Meanwhile, it employed Legendre chaotic neural network scrambling to achieve the two-encryption, and boosted the robustness of the algorithm, so that it can better resist conventional attacks and geometric attacks. The zero-watermarking algorithm can solve the contradiction between the watermarking imperceptibility and robustness. The experimental results show that the proposed algorithm has superduper robustness.