Research On Motion Estimation Algorithm Based On HEVC Video Standard

Motion estimation is one of the core technology in video compression coding,and to find faster and robuster motion estimation algorithm is an important topic.Among kinds of motion estimation algorithm,motion estimation based on block is good at balancing the accuracy and complexity of calculation,and it has small amount of calculation and is easy to realize,so that it is adopted in the video coding international standard H.264/AVC and H.265/HEVC.Both of H.264/AVC and H.265/HEVC are block-based hybrid encoding,which is based on block motion estimation.Compared to H.264,HEVC has increased the compression efficiency more than 50%,meanwhile,the computational complexity is three times of H.264.This paper aims to study an algorithm that can reduce computational complexity with no reduction on video quality and HEVC bitrate.This paper mainly does some analysis and research on the classical motion estimation algorithm,and the nearest fast ME algorithms,in order to improve the existing algorithms and achieve a more optimal algorithm.This paper expounds the process of motion estimation in video coding,and then analyzes the main calculation of motion estimation process,analyzes the characteristics of these calculations and address to these calculations making improvement.Through comparing experiments with TZsearch algorithm in HEVC,this paper proof its effectiveness on these improvements.In the end,this paper put forward an improved rotating hexagonal pattern search algorithm based on predictive information.Based on correlation of spatial and temporal motion vector,the algorithm can predict a best search window,an initial search point,and a reasonable threshold for terminating search process.This prediction can effectively reduce the number of search point with almost no losses.During the global search period,this paper puts forward variable step-size rotating-hexagonal search pattern,which search on both horizontal and vertical direction.And if a higher value in precision is searched,then this value will replace the original threshold,so that the ME algorithm gets an adaptive threshold value.Through the contrast experiment with FS and TZsearch algorithm,the following conclusions can be draw:the proposed algorithm in this paper can keep the same search accuracy with the TZsearch and has a big improvement in search speed.The proposed algorithm has been tested with HEVC reference software.Simulation results show that the algorithm achieves up to 44.7% decrease in ME complexity when compared to the fast ME algorithm TZSearch with negligible loss in PSNR and bitrate.