期刊文献+

人类运动轨迹距离计算方法

Methods to calculate distances of human motion trajectories
下载PDF
导出
摘要 将场景中的人类行为以轨迹形式表征,引入五种计算方法衡量运动轨迹间距离,比较五种计算方法在计算轨迹间距离所耗的时间。利用多维标度技术(MDS)将五种方法得到的距离矩阵映射到二维空间中,通过人工标识MIT停车场行人路径数据,计算各类路径轨迹的类间、类内距离的均值和方差,衡量距离计算方法精度。并通过路径数据中的三类典型问题,比较计算方法在解决实际问题中的能力。实验表明,改进LCS应用于轨迹间距离计算,在时间消耗上最优,并且具有较高的精度,能很好的解决三类典型问题。 In this paper, the human motion in scene is characterized in the terms of trajectories, and five calculation methods are introduced to measure the distances between trajectories. The time consumption of five calculation methods during the calculation of the distances between trajectories is compared. MDS is adopted to map the matrix of distances got by five calculation methods into 2-D space to calculate the mean value and variance of inter-clusters and intra-clusters distances of each path trajectories, and to judge the accuracy of distance calculation methods by aritificial identification of pedestrian path data in MIT parkinglot. The capacities of these methods to solve the actual problems are compared by three kinds of repre-sentative problems existing in path data. The experiments indicate that LCS performs best in time consumption and accuracy. It can also solve the three representative problems very well.
作者 赵建军 陈滨
出处 《现代电子技术》 2012年第24期73-75,78,共4页 Modern Electronics Technique
关键词 人类运动路径 轨迹距离 多维标度技术 LCS human motion path trajectory distance multi-dimensional scaling technology LCS
  • 相关文献

参考文献9

  • 1DODGE Somayeh, WEIBEL Robert, LAUBE Partrick. Exploring movement similarity analysis of moving objects [J]. The SIGSPATIAL Special, 2009, 1(3): 11-16.
  • 2WANG Xiao-gang, MA T K, NG G W, et al. Trajectory analysis and semantic region modeling using a nonparametric Bayesian model [C]// Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK: IEEE, 2008: 1-8.
  • 3XIANG T, GONG S. Video behavior profiling for anomaly detection [J]. IEEE Transactions on PAMI, 2007, 30: 893-908.
  • 4BASHARAT A, GRITAI A, SHAH M. Learning object motion patterns for anomaly detection and improved object detection[C]// Proceedings of IEEE Int'l Conference on Computer Vision and Pattern Recognition. [S. l.]: IEEE, 2008 : 1-8.
  • 5ZHOU Y, YAN S, HUANG T. Detecting anomaly in videos from trajectory similarity analysis [C]//Proceedings of IEEE 2007 International Conference on Multimedia and Expo. [S. l.]: IEEE, 2007: 1087-1090.
  • 6PICIARELLI C, MICHELONI C. Trajectory-based anomalous event detection [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2008, 18 (11) : 1544-1554.
  • 7PICIARELLI C, FORESTI G L. Online trajectory clustering for anomalous events detection [J]. Pattern Recognition Letters, 2006, 27:1835-1842.
  • 8KANG H Y, KIM J S, LI K J. Similarity measures for trajectory of moving objects in cellular space [C]// Proceedings of the 2009 ACM Symposium on Applied Computing. [S. l. ]: ACM , 2007: 1325-1330.
  • 9BASHIR F, KHOKHAR A, SCHONFELD D. Object trajectory- based activity classilication and recognition using hidden markov models [J]. IEEE Transactions on Image Processing, 2007, 16(7): 1912-1919.

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