Designing Three-Dimensional Cellular Automata Based Video Authentication With an Optical Integral Imaging Generated Memory-Distributed Watermark

Abstract

Integral imaging for image watermarking is attracting a great deal of interest as it provides high processing speed and data redundancy property. Traditional video watermarking algorithms are mostly based on discrete transform domains, such as the discrete cosine transform and discrete wavelet transform. Most of these algorithms generally provide one transform plane for watermarking. In this paper, we propose a video watermarking by employing three-dimensional (3-D) cellular automata (CA) transform and perfect-reconstruction integral imaging (PRII). Compared to the existing works, the main strength of this paper is to utilize a 3-D CA transform to produce multiple transform planes for watermark embedding, so that greatly improve the security; decompose watermark into a series of elemental images using integral imaging to enhance the robustness; and employ perfect-reconstruction filter into the iterative back-projection process thus resolving the ringing artifacts problems. Experimental results show that the proposed PRII-based watermarking method outperforms other similar watermarking methods and is robust to a wide range of attacks, e.g., Gaussian noise, salt and pepper noise, frame clipping, JPEG compression, and median filter attack.