Study On Synchronization Method Of Stereo Audio And Stereo Video Based On Video Embedded By Audio

With the notable improvement of the computer technology and network technology, especially broadband communication technology, Internet technology and data compression technology significantly, multimedia communications technology has made rapid development. Multimedia applications have been deep into all aspects of our lives, and people use multimedia information as a means of communication more and more in the information exchange process.Modern video technology is developing into"digital","HD"and"stereo". Stereo video that provides users with three-dimensional perception of three-dimensional scene has broad applications, such as stereo television, stereo conference, distance education, aviation navigation, remote medical diagnosis, etc. A key technology to support stereo video applications is the synchronization technology of stereo audio and stereo video.With the continuous development of digital television system, for different applications, a series of digital video coding standards have been developed and continuously improved, Joint Video Team (JVT) consisting of ISO/IEC and ITU-T has formulated the latest video coding standard-H.264. H.264 has inherited the advantages of the past standards, and uses a lot of new technologies, making its coding performance superior to other standard. Compared with the previous standards, H.264 has a higher compression ratio and better channel adaptability, therefore the stereo video coding based on H.264 standard is more likely to achieve a breakthrough in the storage and transmission for stereo video. Based on H.264 standard, a stereo video coding platform with the combination of motion compensation prediction and disparity compensation prediction has been built on the basis of single-view coding.According to the feature of the stereo video coding platform based on H.264 standard, a synchronization method of stereo audio and stereo video based on video embedded by audio is proposed in this paper. The basic idea of the algorithm is to make the compressed stereo audio stream embedded into the stereo video coding system based on H.264 standard to ensure the stereo audio and stereo video"always synchronized"during the transmission, storage, receiver decoding and playback. The synchronization method of stereo audio and stereo video proposed in this paper should follow three design principles: (1) Audio is preferential; (2) Minimize the impact on the video image quality; (3) Control the embedding overhead.In this paper, we do not use the time stamp technology of MPEG-2 standard, and propose a method of stereo audio and stereo video hybrid signal synchronous compression coding based on video embedded by audio. At the sender, the stereo audio data as hidden message are embedded into the transform coefficients of the stereo video, then the stereo video which have been embedded by stereo audio data are encoded on the stereo video coding platform based on H.264 standard. During the transmission, only the stereo video signals hidden the stereo audio signals are sent. At the receiver, stereo audio data are extracted from the decoded stereo video according to the embedding rule then stereo video and stereo audio are reconstructed respectively. Thus stereo audio and stereo video can reach the receiver synchronously, and ultimately can be broadcast simultaneously. Stereo audio are hidden in stereo video to encode synchronously, so stereo audio and stereo video achieve synchronous transmission under the condition of not using the time stamp and saving the transmission resource, and overcome the impact of delay, jitter and network congestion during the network transmission.Simulation experiments are implemented in VC++6.0 to prove the proposed synchronization method of stereo audio and stereo video based on video embedded by audio. Select three different formats stereo video sequences as test sequences, and analyze the experimental results. Compare the stereo audio data extracted at the decoder with the stereo audio data embedded at the encoder, we can find that both of them are the same, it can indicate that the proposed embedding and extraction method can guarantee stereo audio data embedded and extracted without any distortion. It can be seen from the PSNR comparison chart of Y component that the two embedding methods proposed in this paper have less impact on the quality of stereo video image, and the decline magnitude of the PSNR of Y component is far less than 0.1dB on average. On the subjective compare the 30th frame decoded image which is encoded singly with the 30th frame decoded image which is encoded synchronously, we can not find any difference with human eyes from either Y component image contrast or YUV image contrast. In the embedding overhead, the embedding overhead of each of the two embedding methods is small, much smaller than 3% which is the channel overhead of MPEG-2 standard, so stereo audio and stereo video can achieve synchronous transmission in the case of low overhead.Experimental results show that the synchronization method of stereo audio and stereo video based on video embedded by audio proposed in this paper can achieve synchronous transmission of stereo audio and stereo video by stereo audio and stereo video encoded synchronously, and thus the simultaneous playback, and the impact on the quality of stereo video image is small, and the overhead due to embedding is small, and the anticipant research purpose is achieved.