论文部分内容阅读
本文提出了一种新的块运动估计匹配准则函数——块特征匹配(BFM)函数,可以用于视频压缩的一些国际标准,如H.261,H.263,MPEG1,MPEG2,HDTV的编解码器中。在这些视频压缩国际标准中视频系统编码器的复杂性最主要取决于运动估计算法。实时的块匹配运动估计的VLSI实现需要考虑以下几个方面:在给定搜索域内运动搜索的复杂度;每次块匹配运算的匹配计算复杂度;每次块匹配运算需要从帧存读取到运动估计处理器的数据量大小;实时硬件实现的适用性.仿真表明BFM算法非常简单有效,可以大大降低相应的块匹配计算复杂度、匹配运算时数据传输时间.BFM函数便于并行实现,从而可以有效地缩短视频编码器的编码时间。本文还详细地给出了BFM函数与其它常用匹配准则函数的比较结果.
This paper presents a new block motion estimation matching criterion function - block feature matching (BFM) function, which can be used in some international standards of video compression, such as H.261, H.263, MPEG1, MPEG2, HDTV codec In the device. The complexity of video system encoders in these video compression international standards depends primarily on the motion estimation algorithm. Real-time block matching motion estimation VLSI implementation needs to consider the following aspects: the complexity of motion search in a given search domain; the matching computational complexity of each block matching operation; each block matching operation needs to be read from the frame memory The size of the motion estimation processor and the applicability of the real-time hardware.The simulation shows that the BFM algorithm is very simple and effective, which can greatly reduce the computational complexity of block matching and match the data transmission time in computing.BFM function is easy to implement in parallel, Effectively shorten the video encoder encoding time. The paper also gives a detailed comparison between the BFM function and other commonly used matching criteria functions.