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

• 人工智能及识别技术 • 上一篇    下一篇

基于粒子滤波和在线学习的目标跟踪

刘海龙,胡福乔,赵宇明   

  1. (上海交通大学自动化系系统控制与信息处理教育部重点实验室,上海 200240)
  • 收稿日期:2012-10-07 出版日期:2013-10-15 发布日期:2013-10-14
  • 作者简介:刘海龙(1987-),男,硕士研究生,主研方向:模式识别,图像处理;胡福乔、赵宇明,副教授
  • 基金资助:

    国家自然科学基金资助项目“多视角下的多类型目标识别与行为分析”(61175009)

Object Tracking Based on Particle Filtering and Online Learning

LIU Hai-long, HU Fu-qiao, ZHAO Yu-ming   

  1. (Key Laboratory of System Control and Information Processing, Ministry of Education, Department of Automation, Shanghai Jiaotong University, Shanghai 200240, China)
  • Received:2012-10-07 Online:2013-10-15 Published:2013-10-14

摘要:

针对粒子滤波跟踪丢失目标后较难恢复的问题,提出一种基于粒子滤波和在线学习的目标跟踪方法。使用粒子滤波有效的跟踪结果作为正训练样本不断更新样本库,将随机蕨作为分类器检测目标位置,当分类器和粒子滤波的检测结果存在较大差异时,重新初始化粒子滤波器。在线学习采用二维二值特征,具有计算简单、尺度不变和光照不变的特点。实验结果证明,该方法的跟踪结果优于传统的粒子滤波,能够准确地跟踪到被遮挡和消失再出现的目标。

关键词: 粒子滤波, 在线学习, 随机蕨, 目标跟踪, 二维二值模式, 巴氏距离

Abstract:

For the problem that the tracker is hard to be resumed when particle filtering fails to track the target, this paper introduces a method that combines particle filtering with online learning. It uses the validated result of particle filtering as positive sample to update the training set. It uses random ferns as classifier to detect object. When there is a big difference between two results, the particle filter will be reinitialized. Two bit binary pattern is used as the online learning feature. It is easy to be computed, and has invariance to illumination and scale. Experimental result proves that this method has better tracking result than particle filtering and it can track the sheltered and disappeared target.

Key words: particle filtering, online learning, random ferns, object tracking, two dimensional binary pattern, Bhattacharyya distance

中图分类号: