| With the rapid development of digital communication technology, the mature of image compression techniques and the publication of international standards, audio-video communication has came into human's daily life and works. The technology of digital audio-video coding is general foundations for the industries of digital audio and video, it has extensive applications in the high definition digital TV, 3G video communication, etc. The source coding technology of the second generation is coming in our country. Compress technology of higher efficiency, better audio and video quality, lower cost is badly needed for digital audio and video industry. A key issues of Internet multimedia is the synchronized transmission of audio and video. In order to resolve the problem, it is important to improve compression and decompression performance of digital video under the finite bandwidth of network.Pervasive, seamless, high-quality digital video has been the goal of companies, researchers and standards groups over the last two decades. As we all known, the R, G, B components of a frame describes the same texture and margin of the image. Each one of them can almost describe the information of the whole image except color. If encoding them separately, it will not take enough advantage of the correlation between the components. This has led to the development of a number of key international standards for image and video compression, including the MPEG-1, MPEG-2, MPEG-4 and H.26x series of standards. Most of these standards are based on the method of inter-frame motion compensation and two-dimensional discrete cosine transform (2D-DCT) and encode and describe the color video in YCbCr 4:2:0 format, which want to take advantage of human visual system (HVS) to save bit expense by decreasing the resolution of two color difference components. Then, the gray image encoding method is separately applied to each component.This thesis is part of the International Cooperation Project of National Science Foundation of China"The Mixed Compression and Synchronized Transmission of Audio and Video Based on Multidimensional Matrix". An embedding scheme for mixed compression of audio and video has been put forward in the project. At transmitter, a multi-matrix theory model has been established and the color video is presented into a four dimensional matrix model and the four-dimensional discrete cosine transform (4DM-DCT) is applied to reduce the correlation between color components and adjacent pixels. And then audio is embedded as hiding signal into DCT coefficients of video images, the compression and coding is applied to the video signal in which audio signal has been embedded. On the other hand, at receiver, the audio signal will be extracted from video signal, then reconstruct the audio and video signal, respectively. The experimental results show good performance for restored video and audio signal after several comparison experiments with the PC.At the meantime, This paper has proposed and realized the scheme for mixed compression of audio and video based on the common use multimedia processor-TMS320DM642 with the characteristics of it. The research of this paper can realize the application of the scheme for mixed compression of audio and video encoder based on microprocessor DSP platform, and apply the scheme for mixed compression of audio and video to other portable terminals besides PC, thus promote the popularization and application of the mixed compression of digital audio and video standard.This paper mainly works on the basis of research and improvement of the scheme for mixed compression of audio and video algorithm, mixed compression of audio and video encoder scheme proposed and realizing on TMS320DM642 DSP platform. Regard this design plan as the centre , this paper proposes the corresponding improved algorithm based on the primary audio and video algorithm, Which reduce the operation amount of mixed compression of audio and video while guaranteeing quality of the picture in any sense. Finally we realize mixed compression of audio and video algorithm on the digital signal processor DM642 platform. Through verifying the mixed compression of audio and video method, while keeping very high picture quality comparing with higher compression to QCIF form picture of the TV quality of the meeting, having realized the audio part basically.The research contents and achievements of this paper are summarized as follows:Firstly, this thesis presents a mixed compression of audio and video entropy coding system model based on four-dimensional matrix, which contains four components: preprocessor of first step, the scheme for mixed compression of audio and video encoding processor, preprocessor of secondary step and the scheme for mixed compression of audio and video decoding processor.Secondly, the audio hiding part of mixed compression of audio and video entropy coding algorithm based on four-dimensional matrix is designed and improved by C language. In the audio signal embedding part of media synchronization method.An adaptive embedding way has been found synchronously, which enhance the resumptive quality of the picture and hidden efficiency in any sense. Experimental results show that the proposed algorithm is better than the traditional method of 2D-DCT/motion and H.264 algorithm standard.Thirdly,this paper use its own algorithm to realize the automatic allocation between the internal storage and external storage by C language in the storage allocating part, which enhance the using efficiency of momery space. Meanwhile, this paper has realized the transplantation of mixed compression of audio and video entropy coding algorithm procedure based on digital processor TMS320DM642 platform. The corresponding optimization for mixed compression of audio and video algorithm with DSP/BIOS system operating and Reference Framework 5 from TI make the coding speed of it satisfied the requirement. Through further optimizing the whole system model, the method having realized the audio part basically.The optimization of parameters which affect performance of the encoder and decoder are studied by experiments. The optimum scheme is obtained through comparing and analyzing the threshold and quantification weight for of prediction modes. The larger threshold brings more precise prediction. Whereas, smaller block sizes lead to increased complexity and an increase in the number of the additional information. Experimental results show that it was no use to improve PSNR of the reconstructed image by enhancing the threshold. So the threshold should not be much larger. Another parameter which affects performance of the encoder and decoder is quantification weight. If the quantification weight is larger, the quantity of output data will be less, but the visual quality will be poorer. Experiment results show that when quantification weight is in small-value range, the compression ratio can be enhanced by enlarging quantification weight, however, when quantification weight is in large-value range will cause serious degradation of reconstructed image while the compression ratio is improved.Overall, this thesis research and improve the mixed compression of audio and video algorithm, propose and realize the scheme on TMS320DM642 of TI platform. It has certain value to the application of mixed compression of audio and video entropy coding algorithm and DSP realizing research. |
PDF Full Text Request