论文部分内容阅读
为了使带宽不同的用户能通过互联网观看数字电视节目,提出了一种将卫星数字电视网络直播的IP(in ternetprotoco l)电视应用方案,采用运动矢量重估计技术来实现M PEG(m ov ing p icture experts group)-2视频到低码率的M PEG-4视频的实时转换,从而避免用户带宽的限制。提出了一种运动重估计算法,通过建立运动矢量反向查找表,利用相似宏块重叠面积来优化候选运动矢量的搜索过程,用帧内相邻宏块运动矢量进行估计等方法,显著提高视频转码效率并保证质量。实验结果表明:在视频转码过程中,与采用PM V fast(pred ictive m otion vector fie ld adaptive searchtechn ique)搜索算法相比,该算法能节省40%左右的编码时间,而峰值信噪比降低不足0.3 dB,显著提高了转码效率。
In order to enable users with different bandwidths to watch digital TV programs through the Internet, an IP (Internet Protocol Television (IP)) application that broadcasts satellite digital television networks is proposed. Motion vector re-estimation techniques are used to implement M PEG icture experts group) -2 Real-time conversion of video to low bit rate MPEG-4 video to avoid user bandwidth limitations. A motion re-estimation algorithm is proposed to improve the search process of candidate motion vectors by establishing a look-up table of motion vectors, using the overlapping area of similar macroblocks, and estimating the motion vectors of adjacent macroblocks. Transcoding efficiency and quality assurance. The experimental results show that the proposed algorithm can save about 40% coding time and reduce the peak signal-to-noise ratio (SNR) compared with the PM v fast (pred ictive m otion vector adaptive adaptive search technology) search algorithm in video transcoding Less than 0.3 dB, transcoding efficiency is significantly improved.