Authenticated Encryption to Prevent Cyber-Attacks in Images

The increased usage and availability of multimedia-based applications have led authentication and encryption to gain considerable importance. Cryptography plays an important role in protecting images from theft and alteration. Digital images are used in a large number of applications such as education, defense, medicine, space and industry. This chapter aims at providing a secure authenticated encryption algorithm for the storage and transmission of digital images to avoid cyber threats and attacks. The designed algorithm makes use of the deep convolutional generative adversarial network to test if the image is a fake image originated by the intruder. If found fake exclusive OR operations are performed with the random matrices to confuse the intruder. If the image is not fake, then encryption operations are directly performed on the image. The image is split into two four-bit images and a permutation operation using a logistic map is performed and finally the split images are merged together. Finally, exclusive OR operations are performed on the merged image using the convolution- based round keys generated to generate the concealed image. In addition, authentication is also achieved by calculating the mean of the actual image. The performance analysis shows that the designed technique offers excellent security and also helps in testing the authenticity of the stored images.