期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

基于OpenStack云架构的尺度不变特征变换算法 被引量：2

Scale-invariant feature transform based on OpenStack cloud framework
下载PDF
导出
摘要 在OpenStack计算架构基础上,部署并解决了尺度不变特征变换(SIFT)特征提取在单一计算节点中计算效率的低下的问题。在保持计算结果精度的前提下,降低了系统计算资源负载,对大量SIFT计算请求进行实现,通过Nova以及Swift项目实现动态规划计算节点和面向对象存储,保证了原算法计算的精度,同时降低20%以上的系统负载,达到预期效果。 The computational efficiency of Scale-Invariant Feature Transform （SIFT） algorithm for computing in a single node is low. For maintaining the accuracy of the premise, and the system calculating resource load was downgraded. A large number of SIFT calculation requests were implemented. Through Nova and Swift project, compute nodes dynamic programming and object-oriented storage were realized. The accuracy of the original algorithm was ensured, reducing the load on the system more than 20% as expected.
作者 曲进男 唐政 王帅群
机构地区 同济大学电子与信息工程学院 富士大学电子与信息学院
出处 《计算机应用》 CSCD 北大核心 2014年第A01期90-92,123,共4页 journal of Computer Applications
关键词 云计算 尺度不变特征变换 OPENSTACK 基础设施即服务 cloud computing Scale-Invariant Feature Transform （SILT） OpenStack Infrastructure as a Service （IaaS）
分类号 TP301.6 [自动化与计算机技术—计算机系统结构]
  • 相关文献

参考文献14

  • 1YANG D, LIU L, ZHU F, et al. A parallel analysis on Scale Invari- ant Feature Transform (SIFT) algorithm [C]// Advanced Parallel Processing Technologies. Heidelberg: Springer Berlin Heidelberg, 2011:26 -27.
  • 2CERBELAUD D, GARG S, HUYLEBROECK J. Opening the clouds: qualitative overview of the state-of-the-art open-source VM- based cloud management platforms [ C]// Proceedings of the 10th ACM International Conference on Middleware. New York: Spring- er-Verlag, 2009: 22.
  • 3LOWED G. Distinctive image features from scale-invariant key- points [ J]. International Journal of Computer Vision, 2004, 60 (2): 91-110.
  • 4LOWED G. Object recognition from local scale-invariant features [ C]// ICCV'99: Proceedings of the International Conference on Computer Vision. Washington, DC: IEEE Computer Society, 1999:1150 - 1157.
  • 5LOWED G. Local feature view clustering for 3D object recognition [ C]//IEEE Conference on Computer Vision and Pattern Recogni- tion. Piscataway: IEEE, 2011:682-688.
  • 6WIND S. Open source cloud computing management platforms: in- troduction, comparison and recommendations for implementation [ C]// Proceedings of the 2011 IEEE Conference on Open Systems. New York: Spfinger-Verling, 2011 : 175 - 179.
  • 7OpenNebula Home Page[ EB/OL]. [ 2013 - 06 - 01 ]. http:// www. opennebula.org/.
  • 8NGUYEN B M, TRAN V, HLUCHY L. Abstraction approach for developing and delivering cloud-based services[ C]//Proceedings of 2012 International Conference on Computer Systems and Industrial Informatics. Piscataway: IEEE, 2012:1 -6.
  • 9BARHAM P, DRAGVIC B, FRASER K. Xen and the art of virtual- ization [J]. ACM SIGOPS Operating Systems Review, 2003, 37 (5): 164-177.
  • 10ITO M, OIKAWA S. Improving real-time performance of a virtual machine monitor based system[ C]// SEUS'08: Proceedings of the 6th IFIP WG 10.2 International Workshop on Software Technolo- gies for Embedded and Ubiquitous Systems, LNCS 5257. Berlin: Springer, 2008:114 - 125.

同被引文献41

引证文献2

二级引证文献7

计算机应用
2014年 第A01期

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