Research On Image Chaotic Encryption Algorithm Based On Compressive Sensing

In the environment big data,the loss of personal privacy data is particularly serious.As the main method of information exchange,digital pictures show the characteristics of large data volume,high redundancy,strong pixel correlation,real-time and strict storage methods compared with traditional texts.This makes traditional encryption methods such as DES and AES cannot meet people's needs.Chaotic systems have characteristics such as sensitive dependence on initial conditions,unpredictability,and inherent randomness.These characteristics are highly similar to the confusion and diffusion in cryptography,and meet the basic requirements of cryptography.Therefore,chaotic cryptography is widely used in the neighborhood of secure communication.In addition,in the environment of big data,the amount of data conta ined in image information is also increasing.If the image information is directly encrypted,it will inevitably lead to a large amount of data and increase the amount of data in the transmission and storage of image information.The compressed sensing the ory can not only achieve high-quality signal compression,but also effectively restore the signal from low-dimensional to high-dimensional.It can be seen that applying the compressed sensing theory to the field of image chaos encryption can not only compl ete the encryption of image data,but also effectively reduce the amount of image data transmission and storage.Therefore,this paper proposes two image chaotic encryption algorithms based on compressed sensing.The specific work of this thesis is as belo w:(1)Structuring a two-level key image encryption system using compressed sensing and improved hyperchaotic.The design scheme includes five parts: plaintext image sparseness,coefficient scrambling,compressed sensing measurement,pixel scrambling and two-way diffusion.Among them,the coefficient scrambling and compressed sensing measurement process are controlled by the chaotic sequence produced on the two-dimensional Logistic-Cubic coupled chaotic system(2D-LCCCS);the scrambling and diffusion process are controlled by the chaotic sequence generated by the four-dimensional hyperchaotic system,of which the four-dimensional hyperchaotic system The initial value of the system is related to the chaotic sequence and plaintext information generated by the two-dimensional Logistic-Cubic coupled chaotic system(2D-LCCCS).In the coefficient scrambling stage,a three-dimensional cat mapping is used to scramble the sparse matrix.Simulation experiments and performance analysis show that the algorithm has better anti-differential attack ability,compression performance and robustness.(2)Structuring an image encryption system using compressed sensing and chaotic neural network system.The design plan includes a compressed sensing module,an encryption module,and a cryptographic system module.The cryptographic system module is composed of a chaotic system and a neural network to form a chaotic neural network,which is used to generate the passwords required for the scrambling,compressed sensing measurement,scrambling,and diffusion stages.The initial value in the chaotic neural network is related to the plaintext information.The compressed sensing module includes plaintext image sparseness,scrambling,compression measurement,etc.which is used for compressed sensing measurement of plaintext image information to ensure its compression performance.The encryption module includes scrambling and diffusion processing to ensure its security.Simulation experiments and performance analysis show that the algorithm has better anti-differential attack ability,compression performance and robustness.