Estimated Based On The Movement Of Block-matching Algorithm Research

The requirement of multimedia communication service is increased steadily along with the fast development of digital information technology.Video compression technology is the key technology to realize the application of multimedia communication.A fast motion estimation algorithm based on block matching is not only a core step of video compression schemes,but also is one of effective technologies for motion target creation,recognition and tracking.Therefore, it is a hotspot in the field of target detection and video image processing.Moreover,it possesses of biggish computation in the whole vedio compression encoding and becomes the choke point of video encoding.The performance of the whole encoder is determined directly.Most search algorithms for motion estimation are based on block-matching integer pixel search for the moment and the production is abundant.But motion estimation for sub-pixel still has ample space to improve and farther research.Therefore,to study several fast motion estimation algorithms for block matching has important meanings.In the paper,there are some correlative research in the aspect of amelioration and optimization for fast motion estimation algorithms and a block-matching motion estimation algorithm based on sub-pixel and layered strategy is proposed.It has some academic significance and important value in the engineering.Above all,some basic concepts,meanings and research status in the field of video compression encoding and motion estimation are introduced.Then the emphasis of research is full search and several typical fast motion estimation algorithms,to analyze the features of different algorithms and to summarize some key technique for improving motion estimation search efficiency.A new layered full-search algorithm for the full search is put forward through the analyse of intrinsic relation and exsited problems in these different algorithms,namely to introduce sub-pixel and layered strategy to improve search speed ,efficiency and accuracy in order to improve the real time of full search in the basic of the full search.The motion estimation algorithm proposed in the paper is to adopt the layered idea to reduce the computation and enhance the search efficiency and to bring half-pixel search to improve search accuracy in order to achieve better search effect than before.In the end,for the sake of testing the availability of this algorithm,the performance comparisons and analyses of this algorithm with several general fast algorithms are given in detail by experimentation data in the paper.Results show sufficiently that the algorithm is better than other typical search motion estimation algorithms in the aspect of time and performance.It is not only used in the video encoding and compressing,but also feasible in the motion target detection and video tracking and the practical effect is good.Of course,in the farther research,the new standard based on H.264 can be considered ulteriorly and it is available to put half-pixel in the paper to quarter-pixel motion estimation to improve search accuracy.Besides,to study new fast motion estimation algorithms from the size of block,the design of new search strategy and the apply of new terminal conditions is also promising.