采用改进Mean Shift算法的移动机器人行人跟踪

王丽佳1; 贾松敏1; 王爽1; 李秀智1

doi:10.3788/OPE.20132109.2364

您当前的位置：

首页 >

文章列表页 >

采用改进Mean Shift算法的移动机器人行人跟踪

信息科学 | 更新时间：2020-08-12

- 采用改进Mean Shift算法的移动机器人行人跟踪
- Person tracking of a mobile robot using improved Mean Shift
- 光学精密工程 2013年21卷第9期页码：2364-2370
- 作者机构：
  
  1. 河北工业职业技术学院信息工程与自动化系,河北石家庄,05000
  2. 北京工业大学电子信息与控制工程学院，北京 100124
- 作者简介：
- 基金信息：
  
  江苏省普通高校研究生科研创新计划基金资助项目();国家教育部留学回国人员科研启动基金();北京市自然科学基金重点项目()
- DOI：10.3788/OPE.20132109.2364
  中图分类号： TP242.6;TP391.4
- 收稿日期：2013-03-12，
  
  修回日期：2013-05-03，
  
  网络出版日期：2013-09-30，
  
  纸质出版日期：2013-09-15
- 稿件说明：
移动端阅览
王丽佳贾松敏王爽李秀智. 采用改进Mean Shift算法的移动机器人行人跟踪[J]. 光学精密工程, 2013,21(9): 2364-2370

WANG Li-jia JIA Song-min WANG Shuang LI Xiu-zhi. Person tracking of a mobile robot using improved Mean Shift[J]. Editorial Office of Optics and Precision Engineering, 2013,21(9): 2364-2370
王丽佳贾松敏王爽李秀智. 采用改进Mean Shift算法的移动机器人行人跟踪[J]. 光学精密工程, 2013,21(9): 2364-2370 DOI： 10.3788/OPE.20132109.2364.

WANG Li-jia JIA Song-min WANG Shuang LI Xiu-zhi. Person tracking of a mobile robot using improved Mean Shift[J]. Editorial Office of Optics and Precision Engineering, 2013,21(9): 2364-2370 DOI： 10.3788/OPE.20132109.2364.

摘要

为了提高移动机器人目标跟踪系统在复杂环境中的跟踪性能，提出在双层定位机制下采用基于自适应核函数的Mean Shift算法实现目标跟踪。利用射频识别器件（RFID）检测携带标签的目标，实现外层粗定位并确定感兴趣区域（ROI）；在内层则根据对视差图的ROI的处理结果确定初始搜索窗口，然后应用基于自适应核函数的Mean Shift算法在从立体相机获得的左图中应用基于自适应核函数的Mean Shift算法实现对目标的精确定位。自适应核函数由目标的区域特征与Epanechnikov函数相融合构成，克服了目标边缘处背景像素对目标颜色概率分布的影响。与传统的Mean Shift算法相比，所提方法在同色背景干扰下仍能准确跟踪目标。另外，RFID限定了图像搜索范围，节省了运算开支，图像处理的平均时间为62.11 ms/frame，满足实时跟踪的要求。实验结果表明，该方法可实现移动机器人在同色背景干扰、遮挡、目标快速移动等情况下的目标跟踪。

Abstract

To improve the performance of person tracking for a mobile robot in complex environments

an adaptive kernel function based Mean Shift algorithm was proposed by using a coarse to fine localization mechanism. In the outer layer

a Radio Frequency Identification Device (RFID) was adopted to detect the person with an ID tag to determine the Region of Interest (ROI) coarsely. In the inner layer

the ROI of a disparity image was processed to estimate an initial searching window. Then

the adaptive kernel based Mean Shift algorithm was applied to location of the person precisely in the left image from a stereo camera. The adaptive kernel function was combined with the regional feature of person and the Epanechnikov function

which can reduce the effect of the background pixel on the targets color probability distribution. Compared with the traditional Mean Shift algorithm

the presented algorithm can track the target successfully when the background has the same color. Furthermore

the searching area is narrowed by the RFID

so that the computational cost is reduced. The average computing time is 62.11 ms/frame

which satisfies the requirements of real-time target tracking. The experimental results indicate that the proposed tracking method can complete the target tracking in a background with the same color

short-term occlusion

fast moving

and a sudden turn for a mobile robot.

关键词

Keywords

references

张娟, 潘建寿, 吴亚鹏, 等. 基于双目视觉的运动目标跟踪与测量[J]. 计算机工程与应用, 2009，45(25): 191-194.ZHANG J, PAN J SH, WU Y P, et al.. Tracking and measurement of moving object in binocular stereo vision [J]. Computer Engineering and Applications, 2009, 45(25): 191-194. (in Chinese)[2]TAKEMURA H, ZENTARO N, MIZOGUCHI H. Development of vision based person following module for mobile robots in/out door environment [C]. Proceedings of the 2009 IEEE International Conference on Robotics and Biomimetics, Guilin, China, December 19-23, 2009: 1675-1680.[3]沈志华, 赵英凯, 王晓荣, 等. 全自主机器人双目视觉运动检测研究[J]. 计算机测量与控制, 2006, 14(1): 26-27. SHEN ZH H, ZHAO Y K, WANG X R, et al.. Research on ios-vision system of autonomous mobile robot for motion detection [J]. Computer Measuement & Control, 2006, 14(1): 26-27. (in Chinese)[4]刘士荣, 孙凯, 张波涛, 等. 基于改进Camshift算法的移动机器人运动目标跟踪[J]. 华中科技大学学报：自然科学版, 2011,39 (S2): 223-226.LIU SH R, SUN K, ZHANG B T, et al.. Mobile robot moving object tracking based on modified Camshift algorithm [J]. J. Huazhong Univ. of Sci. & Tech.：Natural Science Edition, 2011, 39(S2): 223-226. (in Chinese)[5]薛陈, 朱明,陈爱华. 鲁棒的基于改进Mean-shift的目标跟踪[J]. 光学精密工程, 2010,18(1):234-239.XUE CH, ZHU M, CHEN AI H. Robust object tracking based on improved Mean-shift algorithm [J]. Opt. Precision Eng., 2010, 18(1):234-239. （in Chinese)[6]董蓉, 李勃, 陈启美. 基于SIFT特征的目标多自由度Mean Shift 跟踪算法[J]. 控制与决策, 2012, 27(3): 399-407.DONG R, LI B, CHEN Q M. Multi-degree-of-freedom mean-shift tracking algorithm based on SIFT feature [J]. Control and Decision, 2012, 27(3): 399-407. (in Chinese)[7]颜佳, 吴敏渊, 陈淑珍, 等. 跟踪窗口自适应的Mean Shift跟踪[J]. 光学精密工程, 2009, 17(10): 2606-2611. YAN J, WU M Y, CHEN SH ZH, et al.. Mean Shift tracking with adaptive tracking window [J]. Opt. Precision Eng., 2009, 17(10): 2606-2611. (in Chinese)[8]刘宁夏, 丁显廷, 孙明竹, 等. 基于Camshift与活动轮廓模型的细胞追踪方法 [C]. Proceeding of the 31st Chinese Control Conference, Hefei, China, July 25-27, 2012: 3656-3661.[9]LIU N X, DING X T, SUN M ZHU, et al.. Cells tracking based on Camshift and active contour model [C]. Proceeding of the 31st Chinese Control Conference, Hefei, China, July 25-27, 2012: 3656-3661. (in Chinese)[10]AMIR H M, ARASH A L. Applying mean shift, motion information and Kalman filtering approaches to object tracking [J]. ISA Transactions, 2012, 51: 485-497.[11]JIA S M, LIN W G, WANG K ZH, et al.. Network distributed multi-functional robotic system supporting the elderly and disabled people [J]. Journal of Intelligent and Robotic Systems, 2006, 45: 53-76.[12]OUADAH N, CADENAT V, LERASLE F, et al.. Multi-sensor-based control strategy for initiating and maintaining interaction between a robot and a human [J]. Advanced Robotics, 2011, 25: 1249-1270.[13]COMANICIU D, MEER P. Mean shift: a robust approach toward feature space analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603-619.[14]贾松敏, 王丽佳, 王爽, 等. 改进的步态光流图与视角相结合的身份识别 [J]. 光学精密工程, 2012，20(11): 2500-2507.JIA S M, WANG L J, WANG SH, et al.. Person identification combining modified gait flow image and view [J]. Opt. Precision Eng., 2012, 20(11): 2500-2507. (in Chinese)

浏览量

309

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于改进迭代无迹粒子滤波的同时定位与地图构建

基于偏振光传感器的移动机器人导航实验