A robust image encryption algorithm based on Chua's circuit and compressive sensing.

• A robust compression-encryption image cryptosystem is proposed. • A modified measurement matrix CADRMM is developed. • A new method of generating diffusion matrix is introduced. • The robustness of the scheme is improved to defend occlusion and noise attacks. • The experimental results demonstrate a good security performance. In terms of Chua's circuit system, compressive sensing (CS) and Haar wavelet, a novel image compression-encryption scheme (CES) is proposed in this paper. Firstly, the plaintext image is decomposed into approximate component and detail components through Haar wavelet. Then the approximate component is diffused by the threshold processing of local binary patterns (LBP) operator-based chaotic sequence which is produced by the combination of Chua's circuit and Logistic map. Next, the Lissajous map is applied to generate the chaos-combined asymptotic deterministic random measurement matrices (CADRMM) which are employed to measure the detail components in different compression ratios. In addition, the combination of mapped approximate and detail components is shuffled by the Logistic map. The experimental results and simulation analysis prove that the proposed cryptosystem is capable of reducing data for transmission and has good security performance under various attacks, especially for the shear and noise attacks. [ABSTRACT FROM AUTHOR]