您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(工学版)》

山东大学学报(工学版) ›› 2015, Vol. 45 ›› Issue (2): 10-16.doi: 10.6040/j.issn.1672-3961.2.2014.069

• 机器学习与数据挖掘 • 上一篇    下一篇

多技术融合的Mean-Shift目标跟踪算法

郭志波, 董健, 庞成   

  1. 扬州大学信息工程学院, 江苏 扬州 225009
  • 收稿日期:2014-05-23 修回日期:2015-01-28 出版日期:2015-04-20 发布日期:2014-05-23
  • 作者简介:郭志波(1975-),男,江苏江都人,副教授,博士,主要研究方向为模式识别,机器学习,图像处理等.E-mail:zhibo_guo@163.com
  • 基金资助:
    教育部科学技术研究重点资助项目(311024);江苏省"六大人才高峰"资助项目(2013DZXX023)

A Mean-Shift target tracking algorithm fused multi technology

GUO Zhibo, DONG Jian, PANG Cheng   

  1. College of Information Engineer, Yangzhou University, Yangzhou 225009, Jiangsu, China
  • Received:2014-05-23 Revised:2015-01-28 Online:2015-04-20 Published:2014-05-23

摘要: 在研究经典算法的基础上,提出了一种多技术融合的Mean-Shift目标跟踪算法,有效地解决了经典Mean-Shift跟踪算法存在的缺陷。通过Kalman算法预测估计目标的中心位置,通过分块颜色直方图提取目标区域的空间信息进行,同时采用背景加权和核加权相结合的方式抑制背景像素对目标的干扰。在多个视频数据上的试验结果表明,研究方法有效地克服了经典的Mean-Shift目标跟踪算法对遮挡、背景像素敏感的问题,在复杂环境的背景下对运动目标跟踪更加准确。

关键词: Mean-Shift算法, Kalman预测器, 分块颜色直方图, 目标跟踪, 背景加权

Abstract: Based on the study of classic algorithm, a Mean-Shift target tracking algorithm fused multi-technology was proposed, and the defects of the classic Mean-Shift tracking algorithm were solved. The center position of target was estimated by the Kalman algorithm. The space information of the target area was extracted using the block color histogram. The combination approach of the background weighted and nuclear weighted was used to suppress the interference of background pixels on the target. The experiments resulted on several video data showed that the new method fused three kinds of technology effectively overcame the barrier and background pixel sensitive problem, and had more accurate tracking than classic Mean-Shift target tracking algorithm under complex environment.

Key words: Mean-Shift algorithm, background weighted, target tracking, block color histogram, Kalman predictor

中图分类号: 

  • TP391.4
[1] FUKUNAGA K, HOSTETER L D. The estimation of the gradient of a density function with applications in pattern recognition[J]. IEEE Trans Information Theory, 1975, 21:32-40.
[2] CHENG Yizong. Mean-shift, mode seeking, and clustering[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 1995, 17(8):790-799.
[3] COMANICIU D, RANMESH V, MEER P. Kemel based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5):564-577.
[4] COMANICIU D, RANMESH V, MEER P. Mean-shift:a robust approach towards festure space analysis[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2002, 24(5):603-619.
[5] 陈少华, 闫钧华. 基于Kalman预测和自适应模板的目标相关跟踪研究[J]. 电子设计工程, 2011, 19(23):189-192. CHEN Shaohua, YAN Junhua. Correlation-based target tracking algorithm based on Kalman prediction and adaptive template[J]. Electronic Design Engineering, 2011, 19(23):189-192.
[6] 党晓军, 尹俊文. 一种基于模板匹配的运动目标跟踪方法[J]. 计算机应用, 2010, 46(5):173-176. DANG Xiaojun, YIN Junwen. Template matching based moving object tracking method[J]. Computer Engineer and Applications, 2010, 46(5):173-176.
[7] ROBERT Collins. Mean-shift blob tracking through scale space[C]//Proceedings of the 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New Jersey, USA: IEEE Press, 2003:234-240.
[8] 彭宁嵩, 杨杰,刘志,等. Mean-shift跟踪算法中核函数窗口宽的自动选取[J].软件学报, 2005, 16(9):1542-1550. PENG Ningsong, YANG Jie, LIU Zhi, et al. Automatic selection of kernel-band width for mean-shift object tracking[J]. Journal of Software, 2005, 16(9):1542-1550.
[9] 贾静平, 柴艳妹. 一种健壮的目标多自由度Mean-Shift序列图像跟踪算法[J]. 中国图像图形学报, 2006, 11(5):707-713. JIA Jingping, CHAI Yanmei. Robust tracking of objects in image sequences using multiple degrees of freedom mean-shift algorithm[J]. Journal of Chinese Image and Graphical, 2006, 11(5):707-713.
[10] 胡铟, 杨静宇. 基于分块颜色直方图的Mean-Shift跟踪算法[J]. 系统仿真学报, 2009,21(10):2936-2939. HU Yin, YANG Jingyu. Tracking algorithm based on block color histogram and mean-shift[J]. Journal of System Simulation, 2009, 21(10):2936-2939.
[11] 张玲,蒋大永. 基于Mean-Shift的改进目标跟踪算法[J]. 计算机应用, 2008, 28(12):3120-3123. ZHANG Ling, JIANG Dayong. Improved target tracking algorithm based on mean-shift[J]. Journal of Computer Application, 2008, 28(12):3120-3123.
[12] BEYAN C, TEMIZEL A. Adaptive mean-shift for automated multi object tracking[J]. Computer Vision, IET, 2013, 6(1):1-12.
[13] CHEN Xiaohui, ZHANG Mengjiao, RUAN Kai, et al. Improved mean-shift target tracking based on self-organizing maps[J]. Signal, Image and Video Processing, 2014, 8(1):103-112.
[14] DENG Yu, LIU Fang, SU Guangda. Mean-shift tracker with chaotic artificial bee colony and space variant resolution[J]. International Journal for Light and Electron Optics, 2014, 125(16):4572-4577.
[15] LANG Fengkai, YANG Jie, LI Deren, et al. Mean-Shift based speckle filtering of polarimetric SAR data[J]. Geoscience and Remote Sensing, 2014, 52(7):4440-4454.
[16] PHADKE G, VELMURUGAN R. Improved mean-shift for multi-target tracking[C]//Procedding of 2013 IEEE International Workshop on Performance Evaluation of Tracking and Surveillance. New Jersey, USA:IEEE Press, 2013:37-44.
[17] SHOU Jen, CHANG Chien, YANG Minshen, et al. On mean-shift-based clustering for circular data[J]. Soft Computing, 2012, 16(6):1043-1060.
[18] TOMAS Vojir, JANA Noskova, JIRI Matas. Robust scale-adaptive mean-shift for tracking[J]. Image Analysis, Lecture Notes in Computer Science, 2013, 7944:652-663.
[19] SIDRAM M H, BHAJANTRI N U. Enhancement of mean-shift tracking through joint histogram of color and color coherence vector[C]//Proceedings of the Second International Conference on Soft Computing for Problem Solving.[S.l.]:Springer India, 2014:547-555.
[20] HATI K K, VARDHANAN A V. Review and improvement areas of Mean-Shift tracking algorithm[C]//Procedding of the 18th IEEE International Symposium on Consumer Electronics. New Jersey, USA: IEEE Press, 2014:22-25.
[21] KALMAN R E. A new approach to linear filtering and prediction problems[J]. Journal of Basix Engineering, 1960, 82(1):35-45.
[1] 马帅依凡,赵子健. 基于人工标记的手术导航仪[J]. 山东大学学报(工学版), 2017, 47(3): 63-68.
[2] 王海军,葛红娟,张圣燕. 基于L1范数和最小软阈值均方的目标跟踪算法[J]. 山东大学学报(工学版), 2016, 46(3): 14-22.
[3] 葛凯蓉, 常发亮, 董文会. 基于局部敏感直方图的稀疏表达跟踪算法[J]. 山东大学学报(工学版), 2014, 44(5): 14-19.
[4] 邱晓欣1,2,张文强1,2*,秦晋贤1,2,杜正阳1,2,张德峰1,2. 恶劣环境下多目标实时跟踪算法研究[J]. 山东大学学报(工学版), 2014, 44(2): 21-27.
[5] 李武,侯志强*,魏国剑,余旺盛. 跟踪框自适应的尺度变化目标跟踪算法[J]. 山东大学学报(工学版), 2014, 44(2): 28-34.
[6] 方 挺,杨 忠,沈春林 . 无人机编队视频序列中的多目标精确跟踪[J]. 山东大学学报(工学版), 2008, 38(4): 22-26 .
[7] 马丽,常发亮,乔谊正 . 基于遗传算法和粒子滤波器的非刚性目标跟踪[J]. 山东大学学报(工学版), 2006, 36(3): 26-29 .
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!