Hologram authentication based on a secure watermarking algorithm using cellular automata

Abstract

A secure watermarking algorithm for hologram authentication is presented in this paper. The algorithm exploits the noise-like feature of holograms to randomly embed a watermark in the domain of the discrete cosine transform with marginal degradation in transparency. The pseudo random number (PRN) generators based on a cellular automata algorithm with asymmetrical and nonlocal connections are used for the random hiding. Each client has its own unique PRN generators for enhancing the watermark security. In the proposed algorithm, watermarks are also randomly generated to eliminate the requirements of prestoring watermarks in the clients and servers. An authentication scheme is then proposed for the algorithm with random watermark generation and hiding.

Review on optical image hiding and watermarking techniques

“…The experimental results indicate that a very small amount of bit errors in the hologram can be detected. To enhance the security, the DCT coefficients for watermark embedding are randomly selected by a pseudo random number (PRN) generator based on cellular automata [65], rather than deterministic locations [64].…”

An asymmetric hybrid watermarking mechanism using hyperchaotic system and random decomposition in 2D Non-separable linear canonical domain

“…Extensive research has also been carried out in optical watermarking area were work has been done in Fourier domain (Sheng et al, 2009), gyrator wavelet domain (Abuturab, 2015), gyrator domain (Li, 2014;Yadav et al, 2015;Liansheng et al, 2017); fractional Fourier transform domain (Guo, Liu and Liucora, 2011;Lang and Zhang, 2014); wavelet domain (Mendlovic and Konforti, 1993;Javidi et al, 2006;Okman and Akar, 2007;Li et al, 2008Li et al, , 2009Hwang, Chan and Cheng, 2014;Mehra and Nishchal, 2015); kinoform transform (Deng,Yang and Xie,804 Pankaj Rakheja et al 2011); using orthogonal transforms: discrete cosine transform (DCT), discrete wavelet transform (DWT), Hadamard-Walsh (Ishikawa, Uehira and Yanaka, 2012); in DCT domain (Starchenko, 2011); based on 5-DWT, fast Fourier transform (FFT) & singular value decomposition (SVD) (Mittal, Bisen and Gupta, 2017). Here in all these transforms number of independent variables were limited as gyrator transform and fractional Fourier transform has one independent variable, Fresnel transform has two independent variables whereas there is no independent variable in Fourier transform.…”