| The development of the Internet has driven the publication and dissemination of multimedia products such as images,audio and video,but the information security problems that followed have plagued every user of multimedia information.As the most important part of information security,the security of video information in storage and transmission has always been the focus of attention.In the storage and transmission of video,information often exists in the form of coding.The most widely used open source standard is the hybrid coding H.264/AVC coding standard.With this as the background,the protection of video information by digital watermarking under the coding standard of H.264 has become the most valuable research topic.In this paper,the coded video is taken as the research object,and two kinds of algorithm ideas are proposed for the existing problems in the video coding domain watermark.The two algorithms can reversibly recover the characteristics of the carrier information by using reversible information hiding,and solve the problems of video quality degradation and increased encoding due to the change of carrier information in video coding,and verify the watermarking performance of the code combining the reversible idea through experiments..(1)In this paper,the problem of distortion and coding efficiency caused by the existing DCT coefficient-based watermarking algorithm in H.264 encoding process is proposed.A watermarking algorithm based on DCT coefficients for reversible information hiding is proposed.The algorithm embeds mercury in the high frequency of the DCT coefficient which has little influence on the image quality.By adopting the reversible information hiding algorithm for the DCT coefficient,the coefficient can hide the information and realize the lossless recovery at the decoding end,avoiding the coefficient change.The resulting decoded image quality is degraded.The simulation results show that the reversible information hiding algorithm using DCT coefficients with reversible thought has good properties.Compared with the traditional video coding information hiding algorithm,this algorithm has obvious advantages in encoding traffic and decoding video quality.(2)In this paper,the existing motion vector based video information hiding algorithm has the problem of local optimality of vector destruction.A reversible information hiding algorithm based on video motion vector is proposed.The algorithm fine-tunes the difference between the predicted vector and the motion vector obtained by the EPZS algorithm,and embeds the watermark in the least significant bit of the difference.By using the vector difference extension,the proposed algorithm achieves the lossless recovery of the motion vector at the decoding end,so that the information embedding does not affect the local optimality of the vector,and the decoded video quality is not affected by the motion vector change.Compared with the traditional information hiding algorithm based on motion vector,the algorithm does not introduce the increase of code traffic at the same time of embedding the watermark,and embeds the watermark in the motion vector to solve the loss of transmission coefficient in the DCT coefficient embedded in the watermark.The problem caused.The experimental results show that the algorithm has good performance in controlling the video stream and the video quality of the decoder. |
