Applicable Image Security Based on Computational Genetic Approach and Self-Adaptive Substitution

Abstract

With the emergence of new information technology fields and various usages of internet applications, securing massive amounts and multiple varieties of multimedia data has become a significant challenge. Hence, it is essential to investigate and develop new cryptographic algorithms to ensure the protection and privacy of multimedia data at rest and during transmission through the communication channel while attempting to overcome the limitations of traditional cryptographic methods. This paper suggests a novel image encryption and decryption technique based on a computational genetic approach and self-adaptive chaotic substitution. The proposed encryption method depends mainly on four steps: sequence generation, diffusion, confusion, and optimization. The USC-SIPI image dataset is being utilized to prove the correctness and effectiveness of our approach in defending against different attack types. Furthermore, the proposed technique could be capable of withstanding attacks while achieving an information entropy of 7. 999, Number of Pixel Change Rate of 99.62%, and a Unified Average Change in Intensity of 33.54%, respectively. The results of security testing and analysis revealed that our image security method is strongly recommended for modern communications security applications.