Research On Robustness Of Digital Image Steganography Algorithm

With the rapid development of network technology and smart phones,it is a common phenomenon for people to share pictures and other multimedia data through social networks.In the field of data hiding,steganography is the act that embedding secret information into a normal multimedia data and then transmitting it to the recipient to achieve covert communication.However,due to the limitation of bandwidth and storage,the general social network performs some lossy processing such as compression and scaling on the transmitted data,which causes the receiver can not recover the secret messages embedded in the image completely.Based on such practical problems,the robustness of steganography has gradually attracted the attention of scholars.The first work of this dissertation is a robust steganography scheme based on the latest robust steganography scheme and the characteristics of lossy channels.The second work is a comment to experimentally verify the security and robustness of a steganography scheme proposed by Paul et al.The main contents are as follows:(1)Robust adaptive steganography based on dither modulation and modification with recompressionThis dissertation proposes a modification with re-compression scheme to improve the robust steganography algorithms based on dither modulation.At first,This work adopts the dither modulation algorithm to embed the secret messages into cover images,and then we adjusts the unstable DCT(Discrete Cosine Transform,DCT)coefficients in the image with the modification with re-compression scheme,the extraction error rate of secret messages are greatly reduced.The receiver of the stego image extracts the secret messages with the mapping rules of the dither modulation algorithm and the decoding algorithm.The experimental results show that when the payload is not less than 0.03 bpnzac,the robustness and security of our scheme are greatly improved compared with the latest papers,and the experimental results are slightly lower than the latest schemes when the payload is low.(2)On the security and robustness of multi-bit image steganography algorithmsPaul et al.proposed an adaptive steganography algorithm(multi-bit image steganography algorithm),but Paul et al.did not adopt a more reasonable and effective steganalysis algorithm to detect the security of the algorithm.At the same time,Paul et al.adopted a unreasonable experimental method to get a conclusion that the algorithm is robust.In terms of security,a more effective Spatial Rich Models(SRM)is adopted to detect the security of the algorithm.As for robustness,Stir Mark is adopted to attack the stego images embedded with secret messages and extract the secret messages from the attacked stego images.Experimental results show that the algorithm has poor detection resistance performance,the algorithm has security flaws.The extraction accuracy of secret information is maintained at about 50%,this means,the algorithm is not robust at all.The error rates of extracting secret messages from the attacked stego images are maintained at about 50%,which means the algorithm is not robust at all.This dissertation introduces the development process and research significance of robust steganography,and summarizes the research results in this field.This dissertation analysis the current technical difficulties and research directions.