Scholarlink

<p>The limited resources of sensor nodes, coupled with the growing volume of data and real-time business demands, pose challenges for secure image transmission in Internet of Things scenarios. To address these problems, we propose a lightweight image-secure communication scheme based on compressed sensing that simultaneously performs image compression and encryption. Firstly, a new 3-D square chaotic map is designed to generate measurement matrices with a smaller sampling interval, and the outer product method is proposed to further enhance its generation efficiency. Secondly, we propose a new image sparsification method based on sparse threshold estimation, which improves the quality of the reconstructed image by approximately 6% compared with other latest methods. Thirdly, we propose a novel batch compressed sensing scheme that effectively balances reconstruction quality and costs by dividing images into multiple batches and dynamically adjusting batch size. Simulation and analysis validate the benefits of this scheme in terms of compression performance, efficiency, and resource overhead, as well as its resilience against common attacks.</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

Image secure communication system based on sparse threshold estimation and batch compressed sensing for IoT

  • 期刊论文

  • 发表日期：

    2026年6月2日

查看全文
  • Wenhao Liu,
  • Chengqing Li,
  • Kehui Sun,
  • Huihai Wang

摘要

The limited resources of sensor nodes, coupled with the growing volume of data and real-time business demands, pose challenges for secure image transmission in Internet of Things scenarios. To address these problems, we propose a lightweight image-secure communication scheme based on compressed sensing that simultaneously performs image compression and encryption. Firstly, a new 3-D square chaotic map is designed to generate measurement matrices with a smaller sampling interval, and the outer product method is proposed to further enhance its generation efficiency. Secondly, we propose a new image sparsification method based on sparse threshold estimation, which improves the quality of the reconstructed image by approximately 6% compared with other latest methods. Thirdly, we propose a novel batch compressed sensing scheme that effectively balances reconstruction quality and costs by dividing images into multiple batches and dynamically adjusting batch size. Simulation and analysis validate the benefits of this scheme in terms of compression performance, efficiency, and resource overhead, as well as its resilience against common attacks.