A Cloud Image Data Protection Algorithm with Multilevel Encryption Scheme and Automated- Selection Mechanism

In this paper, we present a cloud image data protection algorithm with a multilevel encryption scheme and automated-selection mechanism to maintain the privacy of cloud data contents. This algorithm is also useful for the protection of personal or commercial data uploaded to the cloud server for real-time applications, monitoring, and transmission. Fundamental and well-known in cryptography, the confusion–diffusion scheme, as well as an automated-selection mechanism (sliding pixel window) were selected as the main motor of the proposed algorithm to cipher images. First, a sliding pixel window is selected to expedite a two-stepped process, whether in small or big images. The confusion stage was designed to drastically change data from plain image to cipher image. The conversion of pixels from decimal to binary and their vertical and horizontal relocation were performed to help in this stage, not only by randomly moving bits, but also by changing the pixel values when they returned to their corresponding decimal values. Meanwhile, the diffusion stage was designed to destroy all possible existing patterns in the sliding pixel window after the confusion stage. Two hyperchaotic systems, together with a logistic map (multilevel scheme), produce pseudorandom numbers to separately conceal the original data of each subplain image through first- and second-level encryption processes. The two-stepped algorithm was designed to be easily implemented by practitioners. Furthermore, the experimental analysis demonstrates the effectiveness and feasibility of the proposed encryption algorithm after being tested using the benchmark “Lena” image, as well as the “Bruce Lee” image, the latter of which is completely different to the first one, statistically speaking.