Machine Learning-Aided Cloud-Based Robust Image Cryptography with Lupus Coyote Optimization for Enhanced Performance

To ensure the confidentiality of the data that are transferred over cloud networks including sensitive information like medical images and money transaction is significant and requires severe security mechanism to protect the personal information from hackers as well as malicious users as breaches in security affect the privacy and reputation of the user. Hence, an image-based cryptographic method is developed for improving security performance. Here, the time-bound encryption approach is established for protecting confidential medical images which is implemented in e-healthcare. Further, it traverses through a Lupus Coyote optimization (LCO) algorithm which utilizes the systematic chaotic maps (CM), since it performs based on the parameters which may be either discrete or consistent time parameters, and the inception of chaotic signals prevents the prohibited accesses. The proposed LCO optimizes the process of pixel assembling by determining the right parameter values for formulating the chaotic patterns to provide a robust cryptography method. Finally, the permutation as well as the properties of diffusion is controlled in the entire system to deal with the complexity of control pixel shuffling as well as the operations of substitution. Additionally, the correlation between neighboring pixels is interrupted which adds complexity to extracting the information by the attackers and the diffusion algorithm assists in eliminating the storage issue. Thus, the performance of the LCO-based time bound image encryption using the CM model is improved than the other systems, in which the performance metrics such as cosine similarity (CS), histogram correlation (HC), mean square error (MSE), peak signal to noise ratio (PSNR), root mean square error (RMSE), and structural similarity index measure (SSIM) attained the values of 86.53, 0.935, 4.12, 39.51, 4.54, and 0.94 dB, respectively, at number of images 50 with the size of population 250.