计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (34): 208-212.DOI: 10.3778/j.issn.1002-8331.2008.34.064

• 工程与应用 • 上一篇    下一篇

活动轮廓模型目标跟踪算法综述

董春利1,2,董育宁1,3,王 莉2   

  1. 1.南京邮电大学 通信与信息工程学院,南京 210003
    2.南京交通职业技术学院 信息工程系,南京 211188
    3.苏州大学 江苏省计算机信息处理技术重点实验室,江苏 苏州 215006
  • 收稿日期:2008-01-14 修回日期:2008-06-11 出版日期:2008-12-01 发布日期:2008-12-01
  • 通讯作者: 董春利

Survey of object tracking algorithms based on active contour models

DONG Chun-li1,2,DONG Yu-ning1,3,WANG Li2   

  1. 1.College of Communication and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
    2.Dept. of Information Engineering,Nanjing College of Transportation Vocational Technology,Nanjing 211188,China
    3.Jiangsu Provincial Key Lab of Computer Information Processing Technology,Soochow University,Suzhou,Jiangsu 215006,China
  • Received:2008-01-14 Revised:2008-06-11 Online:2008-12-01 Published:2008-12-01
  • Contact: DONG Chun-li

摘要: 目标跟踪是当前计算机视觉领域最活跃的研究主题。首先对基本的跟踪类型进行了介绍;然后讨论了基于活动轮廓模型的图像分割,重点分析了参数活动轮廓模型的梯度矢量流模型(Gradient Vector Flow,GVF),以及几何活动轮廓模型中的模型;并讨论了基于粒子滤波的目标跟踪算法的研究现状,最后展望了这一领域未来研究的热点。

关键词: 目标跟踪, 活动轮廓模型, 粒子滤波, 多目标跟踪

Abstract: Object tracking is hot topics currently in the fields of computer vision.Firstly,the basic tracking classification are introduced.Secondly,the image division based on active contour models is discussed with emphasis on analyzing the Gradient Vector Flow(GVF) of the parametric active contour models and Chan-Vese model of the geometric active contour models.Finally the present research situation of object tracking algorithms based on particle filter has been analyzed.Additionally,the trends in this fields are discussed.

Key words: object tracking, active contour models, particle filter, multiple object tracking