Research On Reversible Data Hiding Algorithm In Encrypted Images Base On Multi Strategy Labelcoding

Digital image is an important carrier of data transmission and information interaction.The security of digital image in cloud storage will make the research of reversible data hiding in encrypted image(RDH-EI)become a hot spot.In this thesis,we mainly use the pixel prediction scheme,using different ways to label and compress the prediction results.Combined with a variety of reversible information hiding algorithms based on prediction and coding compression,two completely reversible data hiding schemes for encrypted images are proposed in terms of prediction accuracy,embedded capacity of secret data,and whether it can be completely decrypted and recovered.The main contents of this thesis are as follows.This thesis propose a RDH-EI algorithm based on 3-MSB coding and scrambling.In the image preprocessing stage,3-MSB fixed length coding is used to mark the high bit,then XOR the lower five bits of the image,and then scrambling for encryption.3-MSB fixed length coding label reduces the process of Huffman coding table storage and encoding and decoding,reduces the complexity of the algorithm,and uses XOR mark image low 5 bits and pixel scrambling to encrypt,which can resist ciphertext only attack and improve the security of the algorithm.The experimental results show that the algorithm can extract secret information without error,and restore the original image losslessly.The embedding capacity of secret data can reach 3.061 bpp,and the average running time is only 6.090 seconds,which is 2.977 seconds faster than the Yin algorithm.This thesis propose a RDH-EI algorithm based on differential mark Huffman coding.In the image preprocessing stage,the scheme of retaining two pixels is used to predict,and the predicted value is compared with each bit of the original pixel value.Huffman coding is used to compress the 1-256 different possibilities generated by the comparison,and more redundant space is obtained.At the same time,the scheme does not need to use the mark graph for secret information Embedding avoids the problem of leaking the original image information through the tag graph,and improves the security of the algorithm.The experimental results show that the algorithm can extract the secret information correctly and restore the original image losslessly.The embedding capacity of secret information can reach 3.477 bpp,which is better than the existing similar algorithms.