期刊文献+

人群运动中的视觉显著性研究 被引量:1

Research on Visual Saliency of Crowd Movement
下载PDF
导出
摘要 在公共场所中人们都倾向于以分组的形式进行运动,本文把这种以分组形式运动的若干个行人称为运动群组,具有视觉显著性的人群运动群组是场景理解的重点,其对人群的整体运动也影响最大。本文对运动群组的视觉显著性展开了研究,分别从规模、速度、组内紧致度和变化度4个方面来对运动群组的视觉显著性进行度量,并基于该度量给出了视觉显著性运动群组检测方法。首先,利用光流法对运动人群进行分析得到光流向量;然后通过层次聚类算法对运动人群进行分组;最后,基于本文所给出的度量计算每个群组的视觉显著性,以检测出视觉显著性最高的运动群组。实验表明该方法能够有效地对视觉显著性运动群组进行检测,该研究成果可应用于人群场景理解、人群运动分析和人群场景分类等计算机视觉研究领域。 In public places,pedestrians always move by groups,which are called as motion groups.A motion group with the highest visual salienoy is the focus of the scene understanding.A new measurement of motion group′s visual saliency is defined in this paper,and the measurement includes four descriptors as follows:scale,speed,group compactness and group variation of different frame.Based on these descriptors,a new method is proposed for detecting the highest visual saliency group.Firstly,the optical flow method is used to compute optical flow vectors.Then,hierarchical clustering algorithm is used to group the crowd.Finally,the values of each group′s visual saliency are computed to find the group with the highest visual saliency value.Experimental results show that the proposed method can detect the highest visual saliency groups effectively.The research can be applied to computer visual fields such as crowd scene understanding,crowd motion analysis and crowd scene classification etc.
作者 刘赏 董林芳
出处 《数据采集与处理》 CSCD 北大核心 2017年第5期890-897,共8页 Journal of Data Acquisition and Processing
基金 国家自然科学基金(61502331)资助项目 天津市自然科学基金(15JCQNJC00800)资助项目 中国民航信息技术科研基地开放课题(CAAC-ITRB-201504)资助项目 中央高校科研业务经费项目(3122013C005)资助项目 中国民航大学科研启动项目(2013QD18X)资助项目
关键词 人群运动 视觉显著性 群组规模 组内紧致度 群组变化度 crowd motion visual saliency group scale group tightness group variation
  • 相关文献

参考文献2

二级参考文献47

  • 1Khani F, Hosseini M J, Abin A A, et al. An algorithm for discovering clusters of different densities or shapes in noisy data sets[C]//Processing of the 28th Annual ACM Symposium on Applied Computing. [S. 1. ]: ACM, 2013:144-149.
  • 2Patwary M, Palsetia D, Agrawal A, et al. A new scalable parallel DBSCAN algorithm using the disjoint-set data structure [C]//High Performance Computing, Networking, Storage and Analysis. Salt Lake City, Utah, USA.. IEEE, 2012:10-16.
  • 3Thanh T N, Drab K, Daszykowski M. Revised DBSCAN algorithm to cluster data with dense adjacent clusters[J]. Chemo- metrics and Intelligent Laboratory Systems, 2013,120 : 92-96.
  • 4Singh S, Awekar A. Incremental shared nearest neighbor density-based clustering[C]//Proceeding of the 22nd ACM Inter- national Conference on Information Knowledge Management. [S. 1. ]:ACM, 2013.,1533-1536.
  • 5许宏玮.基于密度与路径的谱聚类算法研究[D].广州:广东工业大学,2013.
  • 6Han J, Kanber M. Data mining, Concepts and techniques[M].北京:机械工业出版社,2000.
  • 7Chang H, Yeung D Y. Robust path-based spectral clustering[J]. Pattern Recognition, 2008,41(1):191-203.
  • 8Zahn C T. Graph-theoretical methods for detecting and describing gestalt clusters[J]. Computers, IEEE Transactions on 1971,100(1) :68-86.
  • 9Zhang K H, Zhang L, Yang M H. Real-time compressive tracking [C]//Proceedings of the 12th European Conference on Computer Vision. Berlin Heidelberg, Germany: Springer Verlag, 2012: 864-877. [DOI: 10.1007/978-3-642-33712-3_62].
  • 10Sevilla-Lara L, Learned-Miller E. Distribution fields for tracking [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington DC, United States: IEEE Computer Society, 2012: 1910-1917. [DOI: 10.1109/CVPR. 2012. 6247891].

共引文献6

同被引文献6

引证文献1

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