An optimized image encryption framework with chaos theory and EMO approach

Security of digital data is one of the prime concerns of advanced data communication technologies. Images possess a major share of the overall data communication. Image data contains several sensitive pieces of information that are to be protected during transmission, storage, and other stages. There are several algorithms that exist in the literature that addresses this issue. The ever-increasing need for image communication demands sophisticated and robust image encryption approaches that can be effectively applied in real-life scenarios. In this work, a local binary pattern is used to produce the binary image, and the electromagnetism-like optimization approach is used to maximize the transition of 0 and 1 and generate the shuffled image. The shuffled image is decomposed into the constituting bitplanes using any one of three bitplane decomposition techniques. Chaos theory is used to generate some synthetic bitplanes for substitution purposes where the electromagnetism-like optimization process optimizes the uniformity in the histogram. This approach is further secured using a final layer scrambling approach. The proposed approach is tested using both qualitative and quantitative metrics. Moreover, the proposed approach is tested against different types of attacks that prove the practical applicability and robustness of the proposed approach.