A Bit-Level Three-Dimensional Rectangular Image Encryption Algorithm Using New Chaotic Maps
Zhenhui Li, Zhouli He, Feng Huang, Renyuan Jia · Automatic Control and Computer Sciences · 2023
Abstract Some two-dimensional chaotic maps are used for scrambling in image encryption. However, the encryption is often unsafe without destroying image statistical characteristics. Moreover, they can only process square image. Here, it designs a bit-level rectangular image encryption algorithm using three-dimensional chaotic maps to solve the problems. The process can permute and confuse images at the same time. Firstly, a plain image is splitted into eight layers according to the value of pixels, then the eight layers are stretched and connected to a new 8-fold image. Secondly, it introduces and improves a new two-dimensional chaotic map that can deal with rectangular images. Every pixel of a column is placed between two adjacent pixels of corresponding row one by one. When the row is longer than the column, the extra pixels are arranged in the original order. After the process, all the pixels are stretched and connected into a row of pixels. Introducing folding mechanism, the line is further mapped to a new image that is not recognized. Lastly, it uses the maps to permute the bigger image. Finally, the cipher image is combined into an original size cipher image. The process shuffles the pixels’ positions and changes the values. Simulation results validate encryption algorithm has large keys space, no statistical characteristics and fast encryption speed above 1 MB/S. It can achieve enough security requirements in public network.