Research On Three Digital Image Encryption And Decryption Algorithms And Development Of GUI Tool

Image encryption technology is one of the most direct and effective ways to ensure the security of digital image transmission and storage.This paper comprehensively applies combinatorial mathematics,deep learning theories and techniques,researches and proposes three new digital image encryption and decryption algorithms,and develops a graphical digital image encryption and decryption tool DIEDT.The primary research works of the thesis are given as follows:1.Propose an image encryption and decryption algorithm FMSS based on the square theorem and magic square.In order to alleviate the limitation of variable scale scrambling ability of magic square scrambling method,the block rule based on the square theorem is introduced,and the encryption effect is improved by block,scrambling,transposing,changing shape,splicing and other operations.The experimental results show that FMSS can effectively encrypt and decrypt the image,and the encryption effect is good.2.The image encryption and decryption algorithm NMF&FCNN based on Nonnegative Matrix Factorization and Fully Connected Neural Network is proposed.In order to cover up the image information,the original image is decomposed into nonnegative matrix and synthesized reversely to obtain the encrypted image with highly concentrated histogram.The fully connected neural network establishes the mapping relationship between the encrypted image and the plain image to complete the lossless decryption.The experimental results show that the scheme is feasible and the robustness and security are improved.3.An image encryption and decryption algorithm CnnEnc based on Convolutional Neural Network is proposed.In order to obtain the encrypted image full of noise,the main features of plain image are extracted by preprocessing network,and the encrypted network fuses it with the randomly generated noise image.The decryption network separates the noise and restores the original plaintext image.The experimental results show that the encryption scheme is feasible,secure and fast,but the algorithm is sensitive to a wide range of changes.4.A graphical Digital Image Encryption and Decryption Tool(DIEDT)is developed by using Python and Py Qt5,which includes many practical function modules such as digital image processing,image encryption and decryption,key management,etc.It provides convenience for research and use,enriches the tools in related fields.The research contributions of this paper are as follows: According to the characteristics of the square theorem,this paper proposes an encryption algorithm with stronger variable scale scrambling ability;proposes an image encryption and decryption scheme with stronger robustness and security by combining non-negative matrix factorization and fully connected neural network;proposes an encryption and decryption scheme by using convolution neural network to extract image features,and verifies the feasibility of the method through experiments;and develops a graphical digital image encryption and decryption scheme Tools.