[1]叶佐义.基于H.264的运动估计和EPZS算法的研究[J].计算机技术与发展,2010,(12):107-110.
 YE Zuo-yi.Research Based on H.264 Motion Estimation and EPZS Algorithm[J].,2010,(12):107-110.
点击复制

基于H.264的运动估计和EPZS算法的研究()
分享到:

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2010年12期
页码:
107-110
栏目:
智能、算法、系统工程
出版日期:
1900-01-01

文章信息/Info

Title:
Research Based on H.264 Motion Estimation and EPZS Algorithm
文章编号:
1673-629X(2010)12-0107-04
作者:
叶佐义
华南师范大学计算机学院
Author(s):
YE Zuo-yi
Computer College,South China Normal University
关键词:
EPZS块匹配运动估计
Keywords:
EPZS block matching motion estimation
分类号:
TP301.6
文献标志码:
A
摘要:
在视频序列中,相邻帧之间存在很大的相关性和冗余信息。运动估计与运动补偿减少相邻帧间的冗余度,在视频编码中起着重要的作用。H.264是一种高性能的视频编码技术,其最大的优点是具有很高的压缩比率。在具有高压缩比的同时,还拥有高质量流畅的图像,然而,也带来了高度的计算复杂度,尤其是运动估计模块中。通过分析经典运动估计全搜索算法与EPZS算法后,在此基础上对EPZS算法的进行改进,提出了增强搜索模板的使用条件。通过实验表明,在原来的基础上运动估计的时间减少了1.2%以上,对H.264的实时编码具有一定的意义
Abstract:
A great correlation and redundant information exist between two adjacent frames in the video sequence.Motion estimation and motion compensation reduce the redundancy of adjacent frames that plays an important role in the video encoding.H.264 is a high performance video encoding technology,and its biggest advantage is to have a high compression ratio.The high-quality image is also its advantage the same with its high compression ratio,however,it also brings a high level of computation complexity,especially in motion estimation module.Analyze the classical motion estimation full search algorithm and EPZS algorithm to improve the EPZS algorithm.On this basis,an enhanced search template condition is proposed to use.The result of our work shows that the time of motion estimation reduce more than 1.2% compared with that of former,it has some value for H.264 real-time encoding

相似文献/References:

[1]李淳 马力妮.基于菱形搜索的改进的运动估计算法研究[J].计算机技术与发展,2008,(11):117.
 LI Chun,MA Li-ni.A Study of Improved Motion Estimation Algorithms Based on Diamond Search[J].,2008,(12):117.
[2]丁西明 段汉根 吴长勤.一种简单的块匹配图像修复算法[J].计算机技术与发展,2010,(11):80.
 DING Xi-ming,DUAN Han-gen,WU Chang-qin.Image Inpainting Based on Simple Block Matching[J].,2010,(12):80.

备注/Memo

备注/Memo:
广东省科技计划资助项目(2007B010400068)叶佐义(1984-),男,硕士研究生,研究方向为视频编码
更新日期/Last Update: 1900-01-01