电子学报 ›› 2021, Vol. 49 ›› Issue (3): 550-558.DOI: 10.12263/DZXB.20200433

• 学术论文 • 上一篇    下一篇

运动模型引导的自适应核相关目标跟踪方法

陈丹, 姚伯羽   

  1. 西安理工大学自动化与信息工程学院, 陕西西安 710048
  • 收稿日期:2020-05-08 修回日期:2020-08-01 出版日期:2021-03-25 发布日期:2021-03-25
  • 作者简介:陈丹 女,1975年生于陕西汉中.博士,硕导.现为西安理工大学自动化与信息工程学院副教授.主要研究方向为现代信号处理,机器人目标识别与跟踪.E-mail:chdh@xaut.edu.cn;姚伯羽 男,1994年生于陕西西安.现为西安理工大学自动化与信息工程学院硕士研究生.主要研究方向为计算机视觉.E-mail:423058934@qq.com
  • 基金资助:
    国家自然科学基金(No.61671375);榆林市科技计划项目(No.2019-146);西安理工大学研究生竞赛培育项目(No.252051834)

Adaptive Response Kernel Correlation Target Tracking Method Guided by Motion Model

CHEN Dan, YAO Bo-yu   

  1. School of automation and Information Engineering, Xi'an University of Technology, Xi'an, Shaanxi 710048, China
  • Received:2020-05-08 Revised:2020-08-01 Online:2021-03-25 Published:2021-03-25

摘要: 针对小型移动机器人对人体目标快速运动或遮挡导致的跟踪准确率降低甚至跟踪失败问题,通过建立足部运动模型预测双脚位置信息,获得核相关滤波(KCF,Kernel Correlation Filter)目标检测区域,再结合输出响应峰值邻域相关检测,提出了运动模型引导的自适应核相关滤波算法.对实际拍摄的七组不同情况下的视频进行了足部目标跟踪实验,结果表明运动模型引导的自适应响应KCF算法平均跟踪准确率最高,且在短时间遮挡情况下的算法跟踪准确率也达到86%,明显高于自适应响应KCF、BACF (Background Aware Correlation Filters)以及SAMF (Scale Adaptive kernel correlation filters with Multiple Features)三种跟踪算法.最后在ROS (Robot Operating System)下将所提算法应用于Turtlebot机器人目标跟踪测试,成功克服了遮挡情况对足部跟踪带来的影响,验证了所提算法具有较强的鲁棒性和实时性.

 

关键词: 目标跟踪, 运动模型, 核相关滤波, 自适应输出响应, 遮挡检测, 机器人操作系统

Abstract: Aiming at the problem of low tracking accuracy and even tracking failure caused by fast motion or occlusion of human targets by small mobile robots,a foot motion model was established to predict the position information of feet,and the target detection region of kernel correlation filter (KCF) was obtained.In this paper,a motion model guided adaptive kernel correlation filtering algorithm is proposed by combining with the output response peak neighborhood correlation detection.Foot tracking experiments were carried out on seven groups of videos under different scenarios.The results show that the average tracking accuracy of the adaptive response KCF algorithm guided by the motion model is the highest,and the tracking precision rate of the algorithm reaches 86% in the case of short-term occlusion,which is significantly higher than that of the adaptive response KCF,BACF and SAMF algorithm.Finally,the proposed algorithm is applied to the target tracking test of a Turtlebot robot under the ROS (Robot Operating System),which successfully overcomes the influence of occlusion on feet tracking,and verifies that the proposed algorithm has strong robustness and real-time performance.

Key words: target tracking, motion model, kernel correlation filter, adaptive response, occlusion detection, ROS

中图分类号: