Research On Image Authentication And Encryption Algorithm Based On Transport-of-intensity Equation

With the development of information technology and the improvement of living standards,images have become an important carrier of information dissemination due to their intuitive and vivid characteristics.Simultaneously,image information also faces severe security issues because of the openness of the Internet.Optical information security technology has remarkable features such as parallel processing,multi-dimensional and multi-parameter capabilities,and can encrypt image information safely and efficiently.In this thesis,the following image encryption and verification algorithms have been proposed by applying the transport of intensity equation base on optical encryption technology:1.An optical multiple-image authentication approach based on transport of intensity equation technique has been proposed.Initially,a phase-encoded plaintext is synthesized with significant blocks chosen from the multiple plaintext images by evaluating their spatial frequency coefficients and then bonded with a random intensity mask generated with Logistic map to constitute the complex amplitude.Then,the complex amplitude is encrypted into a real-valued ciphertext with noise-like distribution by using Fresnel diffraction.In the process of authentication,the phase information containing sparse plaintext is firstly reconstructed by solving transport of intensity equation.The existence of a plaintext image can be identified by calculating the nonlinear correlation between it and its partial data only containing the extracted significant blocks from the phase information by aid of the corresponding binary mask.A set of numerical simulations are carried out to demonstrate the feasibility of the proposed approach.2.An optical image encryption algorithm based on the transport of intensity equation is proposed.The image to be encrypted is recorded as intensity difference distribution through phase truncated Fresnel diffraction and numerical calculation.The intensity difference is decomposed into multiple sub-bands by the three-level discrete wavelet transform,and scrambled by numerous disordered iterative values generated with Logistic chaotic map.With the help of inverse wavelet transform,a random amplitude mask replaces the low-frequency sub-band to synthesize noise-like ciphertext.In decryption process,the intensity difference distribution is reconstructed by the inverse process of the encryption process,and then the plaintext information is quickly recovered using the transport of intensity equation.Through considerable numerical simulation experiments,the effectiveness of the al gorithm is verified.