| With the rapid development of computer technology,image,video and other multimedia information more and more need for transmission on the network.The image information often involves the privacy of personal information,the company's strategic interests,the government's military secrets,resulting in the security of image information.Image encryption is the most direct and primary problem of image information security.It can not only serve as a pretreatment step of digital image watermarking,but also a relatively independent branch of information security.This paper briefly discusses the basic theoretical knowledge of image encryption,chaotic theory and multi-core parallel computing technology,improved the coupled map lattice model,the nonlinear coupled tent map lattice model real number and the integer domain were constructed,and the model is designed based on the integer bit level image diffusion and confusion encryption algorithm and parallel image encryption algorithm DNA encoding and multi core technology based on.The main contents are as follows:(1).Coupled tent map lattice model based on the introduction of Arnold cat map instead of the adjacent coupling as the coupling model between the lattice,and the abolition of the coupling coefficient of the model,the coupling method is constructed for nonlinear coupled tent map lattice model real domain,this model can quickly generate parallel multidimensional pseudo random sequence,and multidimensional pseudo random sequence also have the uniformity and independence of the two characteristics,thus overcoming the fundamental flaw in the design of chaotic pseudorandom sequence generator in the existing.(2).Using integer Tent mapping as nonlinear function,nonlinear coupled map lattice model for the tent of the real domain integral,nonlinear coupled chaotic model of integer domain,the integer domain model inherits the parallel iterative characteristics and coupling between the diffusion mechanism of the real domain model in space,can quickly generate parallel iteration multidimensional integer pseudo-random sequence,and solves the problem of computing data truncation process,greatly improves the speed of model.The cryptographic properties of the model are analyzed from the aspects of mutual information,difference characteristics and uniform distribution characteristics.Simulation results show that the model can generate uniformly distributed and mutually independent multidimensional integer pseudo random sequences in a fast and parallel way.(3).Based on the nonlinear coupled chaotic model of integer domain,a bit level image encryption algorithm is designed to make full use of the basic idea of confusion and diffusion.Multidimensional integer pseudo-random sequences generated by integer nonlinear coupled chaotic model and iterative multi line parallel image pixel value summation,modulo operation,so as to fully confuse the image pixel values,obtained intermediate ciphertext;the intermediate ciphertext bits are grouped according to the mapping model and the cat,the intermediate ciphertext bit group respectively.With different parameters of scrambling and diffusion,which avoids the problem of short period cat map exists,and greatly improves the complexity of encryption algorithm.(4).Based on integer nonlinear coupled spatiotemporal chaos model,a multicore parallel color image encryption algorithm is designed by using DNA encoding technology and multi-core parallel computing technology.The integer pseudo-random sequence and image pixel values of fast parallel XOR using data parallel loop method,and the intermediate ciphertext using the DNA encoding rules of different complementary operation,then we use task parallel method to do DNA addition for RGB trichromatic components of color images,and then generate the final ciphertext.Parallel image encryption algorithm is implemented by.NET4.0.The encryption effect of algorithm is analyzed from the aspects of algorithm speed,statistical attack analysis and differential attack analysis. |
PDF Full Text Request