A block feature matching algorithm using predictive search （BFMPS） algorithm that is used for motion estimation in such the international standards as H.261

H.263

MPEG1

MPEG2

and HDTV

is proposed in this paper.Based on the spatial characteristics of center biased distance between motion vector and corresponding prediction vector and temporal correlation of motion vectors

this method can significantly reduces the search times.The BFMPS method uses the simple and effective block feature matching function for block matching

and thus the corresponding complexity of block matching computation

data fetching

and memory management is greatly reduced.Simulation results show that the proposed BFMPS algorithm is efficient to reduce search times and computational complexity of motion estimation.Comparisons of BFMPS with other fast search algorithms are also given in detail.