期刊文献+

复杂背景下多信息融合的粒子滤波跟踪算法 预览 被引量:14

A particle filter based tracking algorithm with cue fusion under complex background
在线阅读 下载PDF
分享 导出
摘要 将颜色和运动这两种信息融合进粒子滤波跟踪器,并提出分层抽样的方法,克服了利用单一信源所带来的跟踪不稳定问题;与典型的基于边缘特征或仅基于颜色信息的粒子滤波器相比,计算简单,并能够有效解决由于目标形状或颜色模糊而产生的跟踪困难问题。实验结果表明,该粒子滤波融合算法在复杂背景下能够稳健可靠的跟踪目标。 We present a particale filter(PF) approach to target tracking that fuses color and motion,in which a layered sampiing strategy is embedded. This method overcomes the instability brought by using a single measurement source. Compared with the traditional PF methods that merely use either edge features or color cues,this method has lower computational expense and effectively resolves the tracking difficulties due to imprecise silhouette and/or color models. The experimental results on real world videos show that the proposed method can robustly track moving objects of interest in complex background.
作者 胡昭华 宋耀良 梁德群 樊鑫 HU Zhao-hua, SONG Yao-liang , LIANG De-qun , FAN Xin (1. School of Electronic Engineering and Optoelectronic Technology ,Nanjing University of Science and Technology, Nanjing,Jiangsu 210094 ,China; 2. School of Information Engineering, Dalian Maritime University, Dalian, Liaoning 116026, China)
出处 《光电子.激光》 EI CAS 北大核心 2008年第5期 680-685,共6页 Journal of Optoelectronics·laser
基金 国家科技攻关计划子项目“基于图像处理技术的城市交通调度管理”资助项目(2004BA111B01)
关键词 目标跟踪 粒子滤波 信息融合 状态估计 object tracking particle filter cue fusion state estimation
作者简介 胡昭华(1981-),女,安徽省马鞍山市,博士研究生,主要研究方向为目标跟踪、粒子滤波、模式识别等.E-mail:zhaohuahu@163.com
  • 相关文献

参考文献15

  • 1Greiffenhagen M,Ramesh V,Comaniciu D,et al.Statistical modeling and performance characterization of a real-time dual camera surveillance system[C].In IEEE Conference on Computer Vision and Pattern Recognition,Hilton Head Island,South Carolina.2000,2:335-342. 被引量:1
  • 2王维雅,丁雪梅,黄向东,谭久彬,李海英.一种小目标快速识别与跟踪方法[J].光电子.激光,2007,18(1):121-124. 被引量:14
  • 3Black M,Jepson A.A Probabilistic framework for matching temporal trajectories:condensation-based recognition of gestures and expressions[C].European Conference on Computer Vision.1998:909-924. 被引量:1
  • 4Menser B,Brunig M.Face detection and tracking for video coding applications[C].Asilomar Conference on Signals,Systems,and Computers.2000,1:49-53. 被引量:1
  • 5Moeslund T,Granum E.A survey of computer vision-based human motion capture[J].Comput Vision and Image Understand.2001,81(3):231-268. 被引量:1
  • 6刘贵喜,范春宇,高恩克.基于粒子滤波与多特征融合的视频目标跟踪[J].光电子.激光,2007,18(9):1108-1111. 被引量:12
  • 7Sanjeev Arulampalam M,Simon Maskell,Gordon N.A tutorial on particle filters for online nonlinear/non-gaussian bayesial tracking[J].IEEE Transaction On Signal Processing.2002,50(2):174-188. 被引量:1
  • 8Isard M,Blake A.Condensation-conditional density propagation for visual tracking[J].International Journal of Comput Vision.1998,29(1):5-28. 被引量:1
  • 9张昊,黄战华,郁道银,蔡怀宇,刘正.二维图像序列中刚性目标的准确定位方法[J].光电子.激光,2005,16(1):102-104. 被引量:4
  • 10Shen C,Anton Van den Hengel,Dick A.Probabilistic multiple cue integration for particle filter based tracking[C].Proceedings of the Ⅶth Digital Image Computing:Techniques and Applications.2003,1:399-408. 被引量:1

二级参考文献34

  • 1张昊,黄战华,郁道银,蔡怀宇,刘正.二维图像序列中刚性目标的准确定位方法[J].光电子.激光,2005,16(1):102-104. 被引量:4
  • 2赵其杰,屠大维,高达明,王仁三,陈方泉.一种基于特征线条的虹膜跟踪实用方法[J].光电子.激光,2005,16(2):199-202. 被引量:3
  • 3Bab-Hadiashar A,Suter D. Robust optic flow computation[J]. Int J Comput Vision ,1998,29(1) :59-77.?A?A 被引量:1
  • 4Veltkamp Remco C,Hagedoorn Michiel. Shape similarity measures, properties and constructions [A]. In: Proc of the 4th Int Conf on Advances in Visual Information Systems[C]. 2000. 467-476. 被引量:1
  • 5Doucet Arnaud,Godsill Simon. On sequential Monte Carlo sampling methods for bayesian filtering[J]. Statistics and Computing, 2000, (10): 197-208. 被引量:1
  • 6Isard Michael, Blake Andrew. Condensation - conditional donsity propagation for visual tracking[J]. Int J Computer Vision,1998,29(1) :5-28. 被引量:1
  • 7Huang C T,Mitchell O R. A euclidean distance transform using grayscale morphology decomposition [ J]. IEEE Trans on Pattern Anal Machine Intell , 1994,4 ( 16 ): 443-448.?A 被引量:1
  • 8LUO Xian-long,PENG Jia-xiong. A modified method for detection and recognition of the infrared small target[J]. Joural of Huozhong University of Science and Technology (华中科技大学学报 ) , 2001,29(5) :71-73. (in Chinese). 被引量:1
  • 9WANG Ju, WU Si-liang. Research on location algorithm for vector miss distance measurement system[J]. Systems Engineering and Electronics(系统工程与电子技术) ,2004,26(11):1624-1627. (in Chinese). 被引量:1
  • 10Duven D, Devereux W. Multisensor miss distance measurement system for test & evaluation of high speed intercepts[A]. Proc. 49^th Annual Meeting on Future Global Navigation and Guidunoe [C] . 1993,767-776. 被引量:1

共引文献25

同被引文献130

引证文献14

二级引证文献66

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 意见反馈