期刊名称：International Journal of Multimedia and Ubiquitous Engineering
印刷版ISSN：1975-0080
出版年度：2015
卷号：10
期号：8
页码：221-230
DOI：10.14257/ijmue.2015.10.8.23
出版社：SERSC
摘要：Based on the optimized three-step search algorithm. Combining threshold judgment as well as local full search, a more efficient motion estimation algorithm is proposed. The algorithm not only inherited the traditional three-step algorithm's quick speed but also kept the advantages of a relatively small amount of calculation, besides it can avoid the local optimum problem in the three-step search algorithm (TSS). In addition, the algorithm combined with the threshold judgment and local full search algorithm, so it also maintains satisfactory visual quality. Comparing the algorithm with TSS and local full search algorithm (LFS). The algorithm has great performance in search points and peak signal-to-noise ratio. Experimental results show that compared with LFS, search points drop by 34.61% ~ 54.47% .While compared with the TSS, the search points only rise by 6.15% ~ 12.21%. The average PSNR of proposed Algorithm is 0.24dB higher than LFS and 3.30dB higher than TSS.
