A Novel Image Encryption Scheme Based on Hidden Random Disturbance and Feistel RPMPFrHT Network

In this paper, we propose a novel image encryption scheme based on hidden random disturbance and Feistel RPMPFrHT network, which can improve some common defects of the transform-based methods. At first, we hide some distractive information into the plain image without enlarging the size of the image, which can greatly improve the ability to resist chosen plaintext attack. Then we put the image into Feistel RPMPFrHT network to transform it into two different fractional domain and we will obtain the merits of robustness to noise and data loss. At last we homogenize the numerical distribution and rearrange the pixels by plain-image-dependent chaotic sequences to hide the special statistical characteristics of the image, which can increase the plaintext sensitivity. Simulation results have proved that our proposed encryption scheme shows good performance on most general tests and it makes a great improvement compared with existing image encryption schemes in transform domain.