欢迎访问《图学学报》 分享到：
图学学报

图学学报

• 视觉与图像 • 上一篇    下一篇

主动漂移纠正跟踪算法研究

  

  1. 1. 沈阳大学信息与工程学院，辽宁 沈阳 110000；
    2. 东北大学信息科学与工程学院，辽宁 沈阳 110000
  • 出版日期:2017-04-30 发布日期:2017-04-28
  • 基金资助:
    国家自然科学基金项目(61503274)

A Template Tracking Algorithm Based on Active Drift Correction

  1. 1. College of Information Engineering, Shenyang University, Shenyang Liaoning 110000, China;
    2. College of Information Science and Engineering, Northeast University, Shenyang Liaoning 110000, China
  • Online:2017-04-30 Published:2017-04-28

PDF (PC)

157

摘要： 针对在视频图像目标跟踪应用中易发生模板漂移及算法计算效率低等问题，提出
了一种新的基于主动漂移纠正的跟踪算法。算法中引入了目标跟踪项和漂移纠正项。通过在传
统仿射配准算法中引入漂移纠正项，完善了传统目标跟踪表达式，解决了跟踪过程中经常出现
的模板漂移问题，增加了算法的鲁棒性。实验结果表明，所提出的算法具有更高的计算效率，
且具有更好的鲁棒性。

关键词: 目标跟踪, 主动漂移纠正, 仿射配准算法, 鲁棒性

Abstract: For the traditional problems of template drift and the low efficiency of algorithms in video
image target tracking applications, a new template tracking algorithm based on active drift correction
is proposed. The algorithm incorporates a drift correction term and a target tracking term. By adding a
drift correction term into the traditional affine image alignment algorithms, it improved the traditional
expression of target tracking and also it has solved the template drift problem and increased the
robustness of the algorithm. The results show that the new algorithm has higher operation efficiency
and better robustness.

Key words: target tracking, active drift correction, affine image alignment algorithm, robustness