Joint video compression and encryption using parallel compressive sensing and improved chaotic maps

Abstract

In this paper, we propose a method for joint compression and encryption of video using parallel compressive sensing (PCS) and improved chaotic maps. The method is shown to be computationally efficient without compromising the quality of the recovered video. The processing of the video frames are carried out at group of pictures (GOP) level which consists of both reference and non-reference frames. We apply DCT for sparsification of the reference frames and permute the DCT coefficients to improve the quality of the reconstructed video. We introduce a Logistic Tent Infinite Collapse Map (LT-ICM) which has a higher chaotic range and complex behavior, unpredictability, and sensitivity towards initial values and controlling parameters. The measurement matrix for PCS is generated by using one-dimensional LT-ICM. A novel substitution method is introduced using circular shift, xor, and modular addition operations. We finally shuffle the substituted frames using frame dependent two-dimensional LT-ICM chaotic sequence. Experimental analysis including comparisons with state-of-the-art approaches establishes the effectiveness of the proposed solution.