作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (4): 190-192. doi: 10.3969/j.issn.1000-3428.2009.04.067

• 人工智能及识别技术 • 上一篇    下一篇

基于色斑联合推举的被遮挡运动目标跟踪

杨 伟,费耀平,李 敏   

  1. (中南大学信息科学与工程学院,长沙 410075)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-02-20 发布日期:2009-02-20

Moving Target Tracking Under Occlusion Based on Multi-hue-blob Voting

YANG Wei, FEI Yao-ping, LI Min   

  1. (School of Information Science and Engineering, Central South University, Changsha 410075)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-02-20 Published:2009-02-20

摘要: 针对非刚体运动目标被遮挡的跟踪问题,提出一种以目标色斑子块相关匹配联合推举的跟踪方法。利用目标色调特征的不变性对目标色斑进行自适应分块,通过子块的运动状态联合估计目标整体的运动状态,在遮挡发生时判断被遮挡子块,并将其排除到联合推举之外。实验结果表明,该方法可实现对非刚体目标遮挡下的有效跟踪。

关键词: 目标跟踪, 自适应分块, 非刚体, 遮挡

Abstract: A method based on multi-hue-blob correlation matching voting is proposed to solve target tracking problem of the moving target under occlusion efficiently in nonrigid target tracking. It uses invariable hue feature to achieve adaptive blobbing, and gains target’s motion state through combined estimate of blobs’ motion states. Experimental results indicate it is robust and has reliable performances under the case of heavy occlusion of the nonrigid moving object.

Key words: target tracking, adaptive blobbing, nonrigid target, occlusion

中图分类号: