HAIE: a hybrid adaptive image encryption algorithm using Chaos and DNA computing

In this paper, a distinctive Hybrid Adaptive Image Encryption (HAIE) scheme is proposed that utilizes some statistical parameters of the plain image like mean and variance to modify the initial conditions and control parameters of the chaotic system. The proposed scheme involves two one-dimensional chaotic maps; logistic map and tent map to generate the pseudo-random sequences. The reason for the scheme being distinctive is that the outcome of the confusion phase, i.e. the permuted image is a hybrid one. Half of the plain image is permuted using the logistic map and the other half using the tent map. The two half-permuted images are concatenated after being encoded into DNA sequences using a DNA coding rule. Another random sequence generated using the logistic map is also encoded into a DNA sequence using the same rule. Finally, in the diffusion phase, the two DNA sequences are operated using the DNA addition operation and then decoded using the same DNA decoding rule followed by a XOR operation. The various experimental results like NPCR (Number of Pixel Changing Rate), UACI (Unified Average Changing Intensity), entropy, and correlation coefficients are calculated. For all the test images used, NPCR and UACI values are closer to their ideal values of 99.61% and 33.46%, respectively, entropy is also approximately equal to 8, and likewise, correlation coefficients are closer to zero in the encrypted images. Some other parameters like SSIM (Structural Similarity) and PSNR (Peak Signal to Noise Ratio) are also calculated and lie in their expected range. The scheme is also subjected to noise and cropping attacks. It is observed that the scheme is highly robust against these attacks. Additionally, being adaptive, the proposed algorithm is resilient to chosen and known plaintext attacks. All these experimental observations indicate that the proposed scheme has not only a good encryption effect but can resist various attacks as well.