Research On Digital Image Encryption Algorithm Based On Multi Chaotic System

In recent years,with the rapid development of communication and computer technology in the world,a large number of communication needs based on multimedia data have shown exponential growth,such as image,audio,video and other information expression.Image has become the most commonly used information carrier in human social activities because of its intuitive and vivid characteristics.However,the transmission of image information is often in an open communication environment.Under this premise,how to ensure the safe transmission and access of images has become one of the hotspots in the field of information security.At present,image encryption technology is considered to be an effective method to protect image information from leakage,and has achieved remarkable results.The emergence of chaos theory opens another door in the field of image encryption.Due to its inherent characteristics,researchers have found that it is very suitable for application in the field of image encryption.Therefore,since the end of the last century,a large number of algorithms based on chaotic systems have appeared in image encryption algorithms,and many valuable research results have been achieved.Scholars give full play to the advantages of chaotic systems in image encryption technology,and on this basis,they continue to learn the advantages of other theories,so as to achieve the purpose of improving image encryption effect.The main research content and innovation of this paper is to analyze the dynamic behavior of two neuron models,and obtain that the neuron model will be in a chaotic state under certain parameters.Therefore,it can be applied to image encryption algorithms,and combined with excellent methods in other fields,two image encryption algorithms are designed:(1)Aiming at many problems such as low security and low algorithm complexity in the encryption algorithm using a single chaotic system,this paper designs a digital image encryption algorithm combining multiple chaotic systems and DNA operation.In this algorithm,firstly,the R,G and B components of the original image are divided into three matrices,and then the random matrix is constructed by Logistic mapping.The DNA encoding,DNA operation and DNA decoding operations are performed on the block matrix and the random matrix respectively.Secondly,the image matrix is replaced by different sequences obtained by Logistic mapping iteration.Finally,the image components are merged to obtain a ciphertext image,thereby completing image encryption.And the chaotic sequence generated by the ML neuron chaotic system dynamically determines all the operations based on DNA in the algorithm.In this paper,the encryption algorithm is simulated,and its security is analyzed subjectively and objectively.The data shows that the security of this encryption algorithm is higher,and it has obvious advantages compared with other algorithms.(2)Aiming at many problems such as poor security and lower key space size in the encryption algorithm using low-dimensional chaotic system,this paper designs a digital image encryption algorithm combining three-dimensional chaotic system and Fibonacci matrix.In this algorithm,image encryption is completed by combining scrambling and diffusion.Firstly,the initial value of the system is calculated,and then a new vector is generated by HR neuron chaotic system iteration.Secondly,the sequence is sorted to change the pixel position of the original image.Finally,on this basis,the image is divided into sub-blocks of equal size,and then the pixel value of the image sub-block is changed by using the Fibonacci Q matrix.After two rounds of scrambling diffusion steps,the encrypted image is obtained.In this paper,the algorithm is verified by simulation experiments,and the security of the algorithm is analyzed according to a series of security evaluation indicators,mainly including histogram before and after image encryption,global entropy and local entropy,plaintext sensitivity,noise robustness and anti-shear attack analysis.The results of each evaluation index show that the algorithm not only gives full play to the advantages of HR neuron chaotic system,but also has a large key space,which can resist various attacks and is not easy to be cracked.