作者
Hao Peng,Zhong Chen,Xuan Deng,Zhiwen Zheng,Shuanglong Zou,Ming Yao,Jie Jiang,Songlin Xie
摘要
Abstract With the massive generation and sharing of image data, achieving secure and efficient transmission has become an urgent challenge. However, existing image encryption methods often struggle to simultaneously ensure high security, high compression efficiency, and high reconstruction quality. To address this issue, this paper proposes a novel two-dimensional (2D) chaotic map and integrates block index permutation, Tetris bit scrambling, and Feistel-based inverse diffusion into a joint compression-encryption scheme. Firstly, A novel 2D chaotic map with dual variables, designated as 2D-SLTM (2D Sine-Logistic-Tent Map). A suite of experiments, such as Lyapunov index, sample entropy, and NIST SP800-22 test, verified that it has higher randomness and a wider range of hyperchaotic parameters. Secondly, in the compression stage, the scheme achieves efficient compression through DCT-based sparsification and a chaotic measurement matrix. Thirdly, it utilizes inter-block index scrambling, Tetris geometric scrambling, and Feistel structure diffusion operations in the encryption stage, which significantly disrupts the image’s statistical characteristics and pixel correlation. Finally, at the compression ratio of 0.5, the method achieves high-fidelity reconstruction and strong security. The reconstructed images exhibit excellent quality, with a PSNR above 30 dB and an SSIM greater than 0.90. Meanwhile, the encryption demonstrates remarkable resistance to differential attacks, as evidenced by an NPCR of about 99.60% and a UACI of 33.46%. These experimental results demonstrate the scheme’s robustness against brute-force attacks, statistical analysis, and differential attacks, thereby establishing its suitability for secure image transmission and storage.