A color image encryption scheme based on chaotic mapping, chaotic system, and DNA coding

In this paper, an improved two-dimensional chaotic mapping is constructed by introducing a Logistic mapping and a Sine mapping. By studying its Lyapunov and bifurcation diagrams, it is proved that the proposed chaotic mapping has better chaotic properties. A new four-dimensional chaotic system is also proposed, and the richer dynamical behavior of the new system is verified by studying the Lyapunov exponent and bifurcation, SE and C0 complexity, and coexisting attractors of the new system. Then, a new color image encryption algorithm is proposed based on an improved two-dimensional chaotic mapping and a new four-dimensional chaotic system and DNA coding. The improved two-dimensional chaotic mapping and the new four-dimensional chaotic system increase the complexity and randomness of the system, giving the encryption algorithm a higher level of security. In this algorithm, the pixel correlation of the three channels of RGB is first broken by ascending arrangement of disorder, the chaotic sequence generated by the iteration of the four-dimensional chaotic system is applied in the DNA encoding encryption. Finally, a comprehensive evaluation of the algorithm performance is presented. In particular, statistical and information entropy analyses verify that the algorithm in this paper is resistant to statistical attacks based on the frequency of pixel values. The results obtained for the resistance to differential attacks are close to the ideal value, indicating that the algorithm has good resistance to plaintext attacks. By calculating the key space and analysing the key sensitivity, it is verified that the algorithm has a strong resistance to brute-force exhaustive attacks. The decryption algorithm can recover the original image by adding noise to the ciphertext image or cropping part of the region, which proves that the algorithm has strong resistance to interference and strong robustness. Test experiments show that this algorithm has high security and is competitive with other algorithms.