Color Image Encryption Algorithms With Fractional Moments Transform And Nonlinear Permutation Diffusion

With the development of science and technology and the improvement of people's living standard,information security has attracted much attention.Images play a crucial role in information transmission,since they can carry large amounts of data intuitively.However,images are easily stolen and tampered by the attacker during transmission,resulting in information leakage,the receiver can not get accurate information in time.To solve the problems that efficiency and security of image encryption are low during information transmission,the research of efficient and secure color image encryption algorithms possesses practical significance.Two chaotic color image encryption algorithms based on nonlinear permutation diffusion method are proposed.The research contents of this dissertation are as follows:A nonlinear fast color image encryption algorithm is constructed with 3D orthogonal Latin squares,matching matrix and singular value decomposition.A new permutation method is constructed with 3D orthogonal Latin squares which are generated with chaotic sequence and matching matrix.Elements of 3D orthogonal Latin squares are integers,only appear once in the same row and column,size of each element is less than or equal to row number of the image.Matching matrix is generated according to the position of the elements with 3D orthogonal Latin squares.Original image is permutated with the permutation method,a permutation matrix is obtained.Permutation matrix is blocked with chaotic sequence and the block operation,a block matrix is obtained.According to the nonlinear diffusion method,the block matrix is diffused to obtain the ciphertext image.Compared with other existing image encryption algorithms,the efficiency of the proposed image encryption algorithm is significantly improved under ensuring the security of image encryption.Key space analysis,correlation analysis,histogram analysis,information entropy analysis,sensitivity analyses,encryption quality analyses,noise attack analyses,cropping attack analysis,known plaintext attack and chosen plaintext attack analyses,encryption and decryption rate analyses are also carried out.Simulation results and performance analyses verify that the proposed chaotic color image encryption algorithm based on 3D orthogonal Latin square and matching matrix has high security,fast encryption and decryption rate,and can resist a variety of common attacks.Combined with multiple-parameter discrete fractional Tchebyshev moments and nonlinear fractal permutation method,a chaotic color image encryption algorithm is designed.Since key space of the discrete fractional Tchebyshev moments is small and its security is not high,discrete fractional Tchebyshev moments are extended to multiple-parameter discrete fractional Tchebyshev moments.A nonlinear scrambling method is constructed with fractal geometry,chaotic system and DNA coding.Sherpinski triangle is employed in this dissertation.Sherpinski triangle is a large congruent triangle which is composed of several small congruent triangles.In this algorithm,original image is transformed with multiple-parameter discrete fractional Tchebyshev moments,a coefficient matrix is obtained.Coefficient matrix is transformed into a bit matrix,and the bit matrix is encoded with DNA coding rules controlled by chaotic sequence.Permutation method is adopted to scramble the encoded matrix,and the DNA decoding rule controlled by chaotic system is employed to decode the scrambled matrix.Decoded matrix is processed with decimal operation to get the scrambled image.According to the nonlinear diffusion method,the scrambled image is diffused to obtain the ciphertext image.Key space analysis,correlation analysis,histogram analysis,information entropy analysis,sensitivity analyses,encryption quality analyses,noise attack analyses,cropping attack analysis,known plaintext attack and chosen plaintext attack analyses,algorithms comparison analysis are also carried out.Simulation results and performance analyses show that the proposed color image encryption algorithm is feasible and has high security.