期刊文献+

基于免疫克隆选择的块匹配运动估计 被引量:5

Block Motion Estimation Based on Immune Clonal Selection
下载PDF
导出
摘要 运动估计是视频压缩编码中的关键技术.从运动矢量的特点出发,采用搜索点预测、Gray码编码以及有效的迭代终止准则等策略,提出了基于免疫克隆选择的块匹配运动估计.该方法将块匹配运动估计问题的性质与免疫克隆选择算法所具有的全局搜索特性、解的多样性和不易早熟的特点相融合,在能够获得接近全搜索方法所得到的平均峰值信噪比的前提下,使得平均搜索点数大为降低.仿真实验结果表明,在大多数序列上,该算法都比已有的快速搜索算法具有更高的性能和更少的平均搜索点数.同时,该算法适用面广,对大运动和小运动序列都能得到较好的效果. Motion estimation is a key technique in video compress and coding. Based on the analysis of the character of motion vector, a novel block motion estimation based on immune clonal selection (BMEICS) is proposed in this paper with some strategies like prediction of initial search point, Gray encoding and effective stop criteria. BMEICS synthesizes the character of block motion estimation and that of global search, diversity, and no prone to premature in immune clonal selection. It speeds up the process of motion estimation while maintaining the average with little loss. Experimental results show that BMEICS obtains almost the same as the full search algorithm with fewer search points, and outperforms the existing fast block-matching algorithms for most sequences in terms of speed and quality. Furthermore, BMEICS is applicable to all types of video sequences in spite of the degree of motion.
作者 刘芳 潘晓英
出处 《软件学报》 EI CSCD 北大核心 2007年第4期850-860,共11页 Journal of Software
基金 SupportedbytheNationalNaturalScienceFoundationofChinaunderGrantNos.60133010 60372045(国家自然科学基金) theDefensePre-ResearchProjectofthe‘NinthFive-Year-Plan’ofChinaunderGrantNo.51406020104DZ0124(国家"九五"国防预研基金) theKeyScience-TechnologyProjectofHigherEducationofChinaunderGrantNo.0202A022(国家教育部重点项目) theNationalResearchFoundationfortheDoctoralProgramofHigherEducationofChinaunderGrantNo.20030701013(国家教育部博士点基金)
关键词 运动估计 块匹配 免疫克隆选择 绝对误差和准则 搜索窗 motion estimation block-matching immune clonal selection sum of absolute difference search window
  • 相关文献

参考文献1

二级参考文献58

  • 1戴汝为,王珏.关于智能系统的综合集成[J].科学通报,1993,38(14):1249-1256. 被引量:52
  • 2戴汝为,王珏.巨型智能系统的探讨[J].自动化学报,1993,19(6):645-655. 被引量:39
  • 3陆德源.现代免疫学[M].上海:上海科学技术出版社,1998.14-16. 被引量:8
  • 4学科交叉和技术应用专门小组(美).学科交叉和技术应用[R].北京:科学出版社,1994.43. 被引量:1
  • 5M N O Sadiku. Artificial Intelligence [ J ]. IEEE Potentials, 1989, 8(2) :35 - 39. 被引量:1
  • 6R J Patton, C J Lopez-Toribio, F J Uppal. Artificial intelligence approaches to fault diagnosis[ A]. IEE Colloquium on Condition Monitoring :Machinety, External Structures and Health (Ref. No. 1999/034)[ C]. London:The Institute of Electrical Eagineers, 1999.5/1 - 5/18. 被引量:1
  • 7R Orwig, H Chen, D Vogel, et al. A multi-agent view of strategic planning using group support systems and artificial intelligence [J]. Group Decision and Negotiation, 1997,6( 1 ) : 37 - 59. 被引量:1
  • 8A Christopher, Welty, G Peter, Selfridge. Artificial intelligence and software engineering: Breaking the toy mold [ J ]. Automated Software Engineering. 1997,4(3) :255 - 270. 被引量:1
  • 9Donald Gillies. Book review: Artificial intelligence and scientific method [ J]. Journal of Intelligent and Robotic Systems. 1998,22( 1 ) :87-95. 被引量:1
  • 10G Sartor, L Karl Branting. Introduction: Judicial Applications of artificial intelligence [J]. Artificial Intelligence and Law, 1998,6(24) : 105- 110. 被引量:1

共引文献223

同被引文献38

引证文献5

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部