A New Multi Chaos-Based Compression Sensing Image Encryption

The advancements in technology have substantially grown the size of image data. Traditional image encryption algorithms have limited capabilities to deal with the emerging challenges in big data, including compression and noise toleration. An image encryption method that is based on chaotic maps and orthogonal matrix is proposed in this study. The proposed scheme is built on the intriguing characteristics of an orthogonal matrix. Gram Schmidt disperses the values of pixels in a plaintext image by generating a random orthogonal matrix using logistic chaotic map. Following the diffusion process, a block-wise random permutation of the data is performed using multi-chaos. The proposed scheme provides sufficient security and resilience to JPEG compression and channel noise through a series of experiments and security evaluations. It enables Partial Encryption (PE) for faster processing as well as complete encryption for increased security. The higher values of the number of pixels change rates and unified average change intensity confirm the security of the encryption scheme. In contrast to other schemes, the proposed approach can perform full and partial encryption depending on security requirements.