基于颜色和边缘直方图的多目标跟踪

张雷; 王延杰; 孙宏海; 姚志军; 何舒文

doi:10.3788/YJYXS20163106.0595

您当前的位置：

首页 >

文章列表页 >

基于颜色和边缘直方图的多目标跟踪

显示与图像 | 更新时间：2020-08-12

- 基于颜色和边缘直方图的多目标跟踪
- Color and edge histogram based multi-target tracking
- 液晶与显示 2016年31卷第6期页码：595-603
- 作者机构：
  
  1. 中国科学院长春光学精密机械与物理研究所,吉林长春,130033
  2. 中国科学院大学, 北京 100049
  3. 北方自动控制技术研究所,山西太原,030006
- 作者简介：
- 基金信息：
  
  国家863高技术研究发展计划资助项目(No.2014AA7031010B)()
- DOI：10.3788/YJYXS20163106.0595
  中图分类号： TP391
- 收稿日期：2015-07-22，
  
  修回日期：2015-08-23，
  
  纸质出版日期：2016-06-05
- 稿件说明：
移动端阅览
张雷, 王延杰, 孙宏海等. 基于颜色和边缘直方图的多目标跟踪[J]. 液晶与显示, 2016,31(6): 595-603

ZHANG Lei, WANG Yan-jie, SUN Hong-hai etc. Color and edge histogram based multi-target tracking[J]. , 2016,31(6): 595-603
张雷, 王延杰, 孙宏海等. 基于颜色和边缘直方图的多目标跟踪[J]. 液晶与显示, 2016,31(6): 595-603 DOI： 10.3788/YJYXS20163106.0595.

ZHANG Lei, WANG Yan-jie, SUN Hong-hai etc. Color and edge histogram based multi-target tracking[J]. , 2016,31(6): 595-603 DOI： 10.3788/YJYXS20163106.0595.

摘要

为了解决基于颜色直方图的多目标跟踪方法对复杂场景适应能力差

容易丢失目标的问题

文中提出一种将颜色直方图与边缘方向直方图相结合的多目标跟踪方法。该方法首先采用一种分块连通域标记方法进行多目标提取

并获得目标的颜色、边缘特征;然后融合目标颜色与边缘两种特征来描述目标的外观模型;最后对跟踪过程中的目标模板进行更新。实验结果表明

该方法对于目标在尺度、光照、姿态发生变化以及目标发生旋转情况下能够实现目标的稳定跟踪

具有很强的鲁棒性。实验中对3组挑战性的视频序列进行了测试

目标数目选定为2个

目标窗口大小为64 pixels 64 pixels的情况下

本文方法跟踪速度最高可达20 fps

基本上可以满足实时性的跟踪需求。

Abstract

In order to solve the problem that the color-histogram based multi-target tracking is poorly adaptive to the complex scene and easy to lose the target

a multi-target tracking which combines the color-histogram and edge-direction-histogram is proposed in this paper. Firstly

the algorithm adopts a cell-based connected component labeling algorithm to extract the multiple targets and obtains the color and edge features of the targets. Then we combine the color and edge features to describe the target appearance models. Finally

the target models are updated in the tracking process. The experimental results show that the algorithm can complete the stable tracking when the target undergoes the scale variation

illumination changes

pose variation and rotation. And the algorithm has a strong robustness. We perform our algorithm on 3 challenging video sequences in the experiment

when the targets number is chosen to 2 and the target window size is 64 pixels64 pixels

the maximum speed of our approach can reach 20 fps and it can mainly satisfy the real-time tracking requirement.

关键词

Keywords

references

程帅,孙俊喜,曹永刚,等.增量深度学习目标跟踪[J].光学精密工程,2015,23(4):1161-1170. CHENG S, SUN J X, CAO Y G, et al. Target tracking based on incremental deep learning[J]. Opt. Precision Eng., 2015, 23(4):1161-1170. (in Chinese)

李静宇,王延杰.基于子空间的目标跟踪算法研究[J].液晶与显示,2014,29(4):617-622. LI J Y, WANG Y J. Subspace based target tracking algorithm[J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(4):617-622. (in Chinese)

郭敬明,何昕,魏仲慧.基于在线支持向量机的Mean Shift彩色图像跟踪[J].液晶与显示,2014,29(1):120-128. GUO J M, HE X, WEI Z H. New mean shift tracking for color image based on online support vector machine[J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(1):120-128. (in Chinese)

张雷,王延杰,何舒文.基于相位一致性的实时压缩跟踪方法[J].光子学报,2014,43(8):0810003. ZHANG L, WANG Y J, HE S W. Real-time compressive tracking method based on phase congruency[J]. Acta Photonica Sinica, 2014, 43(8):0810003. (in Chinese)

BIRCHFIELD S T, RANGARAJAN SR. Spatiograms versus histograms for region-based tracking[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, CA, USA:IEEE, 2005, 2:1158-1163.

COMANICIU D, RAMESH V, MEER P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5):564-577.

ZHU W Y, LEVINSON S E. Edge orientation-based multi-view object recognition[C]//Proceedings of the 15th IEEE International Conference on Pattern Recognition, Barcelona:IEEE, 2000, 1:936-939.

BRADSKI G R. Real time face and object tracking as a component of a perceptual user interface[C]//Proceedings of the 4th IEEE Workshop on Applications of Computer Vision, Princeton, NJ:IEEE, 1998:214-219.

COMANICIU D, RAMESH V, MEER P. Real-time tracking of non-rigid objects using mean shift[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Hilton Head Island, SC:IEEE, 2000, 2:142-149.

ZHAO Q, TAO H. A motion observable representation using color correlogram and its applications to tracking[J]. Computer Vision and Image Understanding, 2009, 113(2):273-290.

NUMMIARO K, KOLLER-MEIER E, GOOL L V. An adaptive color-based particle filter[J]. Image and Vision Computing, 2003, 21(1):99-110.

GU Q Y, TAKAKI T, ISHII I. 2000 fps multi-object tracking based on color histogram[C]//Proceedings of SPIE 8437, Real-Time Image and Video Processing, Brussels, Belgium:SPIE, 2012, 8437:84370E-1-84370E-13.

ROSENFELD A, PFALTZ J L. Sequential operations in digital picture processing[J]. Journal of the ACM, 1966, 13(4):471-494.

HARALICK R M. Some neighborhood operations[M]//Real Time/Parallel Computing Image Analysis. New York:Plenum Press, 1981:11-35.

WU K, OTOO E, SUZUKI K. Two strategies to speed up connected component labeling algorithms[R]. Technical Report 59102. Berkeley:Lawrence Berkeley National Lab, 2005.

HE L F, CHAO Y Y, SUZUKI K, et al. Fast connected-component labeling[J]. Pattern Recognition, 2008, 42(9):1977-1987.

浏览量

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YOLOv5和重识别的行人多目标跟踪方法