Research On Image Encryption Algorithm Based On Chaos And Dynamic DNA

With the advent of WeChat business and short video era,big data,traffic,popularity,etc.are flooding all aspects of the Internet,facilitating people's lives,and at the same time,a series of practical information security problems are derived.As one of the information carriers in the era of big data,images play an important role in business,military affairs and people's livelihood,so they are widely used.Therefore,how to ensure the security of image information and prevent it from being stolen and abused by illegal attackers is a topic worthy of study and of practical significance.Image encryption has been widely studied as a means of image information protection.However,traditional image encryption algorithms have some disadvantages,such as limited use of scenes and poor anti-noise ability,which leads to low encryption efficiency and poor performance.In recent years,chaotic system has attracted much attention in the field of image encryption because of its sensitivity to initial values and difficulty in prediction.Related technologies such as DNA computing and DNA encoding are also suitable for the field of image encryption because of their high parallelism and strong storage capacity.Therefore,related technologies as follows are studied,and two color image encryption algorithms are proposed in this thesis:(1)At present,the fixed scrambling-diffusion framework of some algorithms easily leads to the invalid scrambling of solid-color pictures,and the encryption effect can only be realized by diffusion operation,which cannot resist the attack of selecting plaintext.To solve this problem,a color image encryption algorithm based on Hopfield network and dynamic DNA is proposed.In the algorithm,pretreatment before encryption is introduced,and the initial image is partitioned by XOR to change the visual information of the image.Furthermore,a three-level indexing dynamic DNA coding method controlled by chaotic sequence is proposed to further improve the encryption performance.At the same time,the diffusion method of substitution between alkali layers in DNA domain is adopted to increase the coupling of encryption among three color channels;Finally,the DNA algorithm is used to further homogenize the pixels and complete the encryption work.Simulation results and theoretical analysis prove that the algorithm can significantly improve the encryption performance of solid-color pictures,and has strong ability to resist the attack of selecting plaintext.At the same time,it has high information entropy,low correlation coefficient between pixels and other indicators,and can also resist noise attacks and cropping attacks to a certain extent.(2)A color image encryption algorithm based on DNA-like behavior and elliptic curve is developed,which adopts symmetric encryption and asymmetric encryption process at the same time and has higher security.The algorithm mainly proposes the DNA cross-row permutation algorithm and the DNA cyclic complement diffusion algorithm by imitating DNA behavior,and uses the DNA cyclic complement combined with elliptic curve to achieve better diffusion performance.The algorithm generally adopts the encryption framework of "lightly scrambling-joint diffusion-rearrangingsecondary diffusion",which ensures good encryption performance on the basis of high security.Through security analysis and experimental simulation,the algorithm can only complete decryption on the basis of fully mastering the symmetric key and private key,which has high security performance,and also has good performance in information entropy,histogram,pixel correlation coefficient and anti-noise performance.