Scholarlink

<p>With the rapid growth of image-centric transmission in mobile Internet, cloud computing, and the Internet of Things (IoT), protecting image confidentiality under bandwidth and edge-computing constraints has become increasingly important. This paper presents an adaptive image encryption framework that couples compressed sensing (CS) with multi-layer chaotic dynamics and the semi-tensor product (STP). The proposed pipeline performs key-dependent blockwise orthogonal preprocessing, transform-domain scrambling and diffusion driven by Logistic/Tent/Hénon/Lorenz systems, and compressive measurement in an integrated manner. An entropy-guided sampling strategy adjusts the measurement rate according to image complexity, improving the compression–reconstruction trade-off across heterogeneous contents. Moreover, STP is employed to implement structured measurement operations, which reduces measurement-matrix storage and lowers computational overhead. Experimental results indicate that the scheme exhibits favorable statistical behaviors and competitive reconstruction quality under adaptive compression ratios.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive compressed sensing image encryption scheme integrating multi-layer chaos and semi-tensor product

  • 期刊论文

  • 发表日期：

    2026年5月6日

查看全文
  • Maolin Zhang,
  • Mingwen Zheng,
  • Yanping Zhang

摘要

With the rapid growth of image-centric transmission in mobile Internet, cloud computing, and the Internet of Things (IoT), protecting image confidentiality under bandwidth and edge-computing constraints has become increasingly important. This paper presents an adaptive image encryption framework that couples compressed sensing (CS) with multi-layer chaotic dynamics and the semi-tensor product (STP). The proposed pipeline performs key-dependent blockwise orthogonal preprocessing, transform-domain scrambling and diffusion driven by Logistic/Tent/Hénon/Lorenz systems, and compressive measurement in an integrated manner. An entropy-guided sampling strategy adjusts the measurement rate according to image complexity, improving the compression–reconstruction trade-off across heterogeneous contents. Moreover, STP is employed to implement structured measurement operations, which reduces measurement-matrix storage and lowers computational overhead. Experimental results indicate that the scheme exhibits favorable statistical behaviors and competitive reconstruction quality under adaptive compression ratios.