北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2006, Vol. 29 ›› Issue (s2): 164-167.doi: 10.13190/jbupt.2006s2.164.297

• 论文 • 上一篇    下一篇

一种自适应的运动目标实时跟踪算法

李 国1,2 , 张心珂2 , 杨国庆1 ,高庆吉2   

  1. 1. 南京航空航天大学 信息科学与技术学院, 南京 210016;
    2. 中国民用航空大学 机器人研究所,天津 300300
  • 收稿日期:2006-09-30 修回日期:1900-01-01 出版日期:2006-11-30 发布日期:2006-11-30
  • 通讯作者: 李 国

An Adaptive Real-Time Tracking Algorithm for Moving Object

LI Guo1,2, ZHANG Xin-ke2, YANG Guo-qing1, GAO Qing-ji2   

  1. 1 College of Information Science and Technolgy, NanJing University of Aeronautics and Astronautics,
    Nanjing 210016,China;
    2 Research Institute of Robotics, Civil Aviation University of China, Tianjin 300300, China
  • Received:2006-09-30 Revised:1900-01-01 Online:2006-11-30 Published:2006-11-30
  • Contact: LI Guo

摘要:

提出了一种基于均值平移(Mean Shift)和粒子滤波融合的自适应运动目标跟踪算法。该算法在处理来自PTZ (Pan/Tilt/Zoom)摄像机的视频图像时,自适应地更新直方图模板的特征信息,并结合Mean Shift算法来控制粒子滤波中粒子的产生,根据粒子的权值计算目标的位置。测试结果验证了该算法的实用性和有效性。

关键词: 运动目标, 粒子滤波, 均值平移, 自适应模板

Abstract:

An adaptive real-time tracking algorithm for moving object was presented based on mean shift and particle filter. In the method, video image from the PTZ camera was dealt, histogram character was updated adaptively, particle propagating was controlled by the mean shift algorithm ,based on the model-adaptive mode. Object position was calculated by particle weight. Finally the experiment results verified the practicability and efficiency of this algorithm.

Key words: moving object, particle filter, mean shift, adaptive model

中图分类号: