Hopfield attractor-trusted neural network: an attack-resistant image encryption

The recent advancement in multimedia technology has undoubtedly made the transmission of objects of information efficiently. Interestingly, images are the prominent and frequent representations communicated across the defence, social, private and aerospace networks. Image ciphering or image encryption is adopted as a secure medium of the confidential image. The utility of soft computing for encryption looks to offer an uncompromising impact in enhancing the metrics. Aligning with neural networks, a Hopfield attractor-based encryption scheme has proposed in this work. The parameter sensitivity, random similarity and learning ability have been instrumental in choosing this attractor for performing confusion and diffusion. The uniqueness of this scheme is the achievement of average entropy of 7.997, average correlation of 0.0047, average NPCR of 99.62 and UACI of 33.43 without using any other chaotic maps, thus proposing attack-resistant image encryption against attackable chaotic maps.