Research On Deep Space Image Compression And Encryption Based On Compressive Sensing

With the rapid development of aerospace technology,image acquisition in deep space exploration occupies an increasingly important position.Deep-space image data has a huge amount of data,heavy transmission load,and limited hardware conditions.Therefore,existing image compression and encryption technologies have been unable to meet the efficiency and security of obtaining deep-space image information.This dissertation focuses on this issue and studies compressive sensing-based image compression encryption algorithm.Compression and encryption algorithm based on compressive sensing has the advantages of both compression and encryption,low computational complexity of compressing encryption and ciphertext robustness,and it is easy to implement in software and hardware.However,there are unavoidable drawbacks when compressive sensing is used for encryption,leaking plaintext image energy information,and the measurement matrix as a key cannot be reused.From the perspective of theoretical research and practicality,this paper studies the encryption algorithms based on compressed sensing.The main research work is as follows:(1)In-depth study of compressive sensing theory for image compression encryption.Summarizes the advantages and disadvantages of Compressed Sensing combined with other encryption algorithms.At the same time,the security of the design of compressive sensing based image compression encryption method is analyzed,which provides a solid theoretical basis for the research and implementation of compressed sensing combined with chaotic systems and fractional Fourier transform.(2)First of all,this paper analyzes the problem of excessive key consumption when compressed sensing is used for encryption,and designs a measurement matrix construction algorithm based on chaotic systems,reducing the size of the key to only two parameters,and ensuring that it is large enough.Key space.Secondly,we study the weaknesses of the encryption method based on compressive sensing which can not resist the choice of plaintext attack,and combine the counter mode to use different measurement matrices for different images so as to achieve "one time and one secret." Then taking into account the huge amount of image data,a block parallel and sampling rate adaptive method is proposed to improve the compression ratio and compression encryption efficiency.Finally,the encryption method for compressed sensing will leak the problem of plaintext energy information,and proposes an encryption method that combines diffusion.(3)Design and implementation of image compression encryption method based on compressed sensing and Fourier transform.In order to improve the security of the compressed sensing encryption method and ensure that the combined encryption method does not damage the robustness of the compressed sensing ciphertext sensitivity,a compressed encryption method combining compressive sensing and fractional Fourier transform is proposed.The plaintext real part and the imaginary part are encrypted by a fractional Fourier double random transformation of the complex field.Finally,the security of the image compression encryption method is tested and analyzed.In this paper,the theory of compressive sensing and image compression encryption technology are studied in depth,and the image compression and encryption methods with emphasis on compression performance and high security are proposed.Finally,tests and analysis are conducted through experiments.