Research And Implementation Of Digital Image Encryption Algorithm Based On Chaos

With the rapid development of computer technology and network technology, computer applications are becoming increasingly widespread, and the human society has entered into a highly information-oriented era. Digital image as an important carrier of information, compared to traditional text, audio, etc., because of its vivid, intuitive and visual characteristics, has taken more attention from people. However, due to bulky data capacity, high redundancy, low entropy and high correlation among pixels of an image, traditional encryption algorithms, such as AES, DES, RSA and so on, are found unsuitable for image encryption. Therefore, researching and developing more secure and efficient digital image encryption algorithms has become a hot topic in academia and industry.The chaotic system is a deterministic nonlinear system, which possesses varied features, such as high sensitivity to initial conditions, ergodicity, randomness and control parameters and so on that happened to coincide with the concepts of diffusion, encryption rounds, pseudo random signal and secret key for cryptography, so with the development of the chaotic cryptography, digital image encryption algorithm based on chaotic theory has been widely concerned.This paper mainly studies the design and implementation of digital image encryption algorithms based on chaotic theory. First of all, this paper briefly discusses the basis of modern cryptography and chaotic theory, analyzes the research status of digital image encryption algorithm based on chaotic theory, and expounds the relationship between chaotic theory and cryptography. Then, several common and typical are briefly introduced, and a new three-dimensional autonomous chaotic system is constructed in the light of the Lorenz and Chen systems while building as long as the introduction of different nonlinear terms can produce different chaotic sequences, also its chaotic characteristic is verified from different aspects. Next, as for the periodicity of Arnold transform which leads to a certain degree of limitations in its application for image encryption algorithms, linking the iterative number of Logistic mapping with gray value of image pixel to get the control parameters of Arnold transform, and after each scrambling through the exchange of pixels to change the number of iterations Logistic mapping, thus making the control parameters of each Arnold transform is different, so as to avoid the periodicity of Arnold transform, at the same time can enhance the relevance between permutation and diffusion. Finally, a novel color image encryption algorithm based on the new three dimensional chaotic system, DNA sequence and a Secure Hash Algorithm SHA-256, which uses SHA-256 to calculate the hash value of the color image pixels that is served as the one-time secret keys and applied to produce the initial values of the new three dimensional chaotic system. The diffusion process is based on using three chaotic sequences that produced by the new three dimensional chaotic system to shuffle R, G, B components of the image. The confusion process is based on using Addition algebra operation of DAN sequence to scramble the pixel values of image R, G, B components and then using DNA decoding rules to decode the scrambled images and recovering the RBG image which is the finally ciphered image. The experimental results and security analysis show that the two algorithms largely enhances the security of the ciphered image and has a large key space and a high key sensitivity, and it is able to resist against the statistical and exhaustive attacks.