Journal of King Saud University: Computer and Information Sciences (Feb 2023)

A novel color image privacy-preserving method: Combining breadth and depth visual encryption with chaotic system

  • Dezhi An,
  • Dawei Hao,
  • Ruoyu Zhao,
  • Jun Lu,
  • Yan Li,
  • Yushu Zhang

Journal volume & issue
Vol. 35, no. 2
pp. 576 – 589

Abstract

Read online

As the saying goes, one image is worth ten thousand words. Because of the rich visual information carried by the image itself, previewers can easily find what they need. On the other hand, the large amount of detail information carried by the image may invade the privacy of the parties, which needs to be protected. Traditional image encryption schemes can protect privacy, but they indiscriminately erase all visual information, which also gives up the visual usability of the image. Motivated by this, we propose a novel color image privacy-preserving method, which can be regarded as a variant of thumbnail-preserving encryption (TPE). The scheme preserves the rough visual information of the image, and considers the information redundancy in the depth direction for the first time compared with other TPE schemes. Specifically, we propose a TPE scheme combining breadth and depth, and a matching encryption framework is proposed, which is divided into three steps: cube permutation, block substitution, and block permutation. Meanwhile, a prototype is constructed by using a chaotic system and sufficient experiments are carried out with it. The experimental results show that this scheme has excellent properties, can well balance privacy and availability, and has high time efficiency.

Keywords