Research And Implementation Of H.264 Motion Estimation Algorithm

H.264 is the video coding standard of ITU-T and MPEG, which achieves enhanced compression performance and provides a"networks friendly"video representation. Compared to other video coding standards, H.264 has made great progress in coding efficiency. For example, it improves the compression efficiency by a factor of about 50 percent over H.263. However the improvement in compression efficiency of H.264 comes from the dramatic increase in computational complexity.Generally, in motion estimation algorithm, image is first divided into non-overlapping macro blocks, and then according to the given block matching measure and searching strategy, some best-matched blocks are found in the reference or past frame. Motion estimation algorithm is the most time-consuming section of the video codec, accounting for 50 percent of the total video coding times.This article first simply introduces the H.264 video coding standard and its key technologies, and lay emphasis on the motion estimation technology principle. It also thoroughly analyzes some classic algorithm principles, the merits and the shortcomings of the motion estimation. At the same time, the motion estimation algorithm makes the optimization in the search strategy, early exit , best searching zone. Under the premise of guaranteeing picture quality, effectively reduces complexity of time and raises the coding efficiency the motion estimation algorithm. The innovations of this article mainly concentrate on following aspects:1. According to the distribution characteristics, searching area is divided into nine sub-regions, including one detailed searching region and eight coarse searching regions.2. According to search region of left macro block, left up macro block, up macro block, an algorithm of Adaptive searching region is implement.3. According to the motion estimation of past frame in video sequences cost, an algorithm of adaptive threshold is implement.