Image Encryption Using Value-Differencing Transformation and Modified ZigZag Transformation

Image encryption is an efficient technique to protect the contents of an image. However, many existing image encryption algorithms have low efficiency and are weak to resist many commonly used security attacks such as the chosen-plaintext attack. To address these issues, in this paper, we propose a new image encryption scheme using value-differencing transformation (VDT) and modified ZigZag transformation. First, we propose a new transform called VDT. It can split a plain image into four subbands based on value-differencing and this can greatly help to diffuse the image pixels. Second, we develop a modified ZigZag transformation that can separate the image pixels with a much higher efficiency. Besides, to improve the ability to defense the chosen-plaintext attack, we extract part information of plain image as parameters to encrypt the image. Notice that these information don’t need when decrypting the image so that the proposed image encryption scheme is a symmetric key encryption algorithm. Simulation results and security analysis show that the proposed image encryption scheme can encrypt different kinds of plain images into unrecognized cipher images with a high security level, and it can outperform several existing image encryption schemes.