Multi-channel video target tracking algorithm based on feature constraint and optical flow field model

Bo ZHANG; Hui LONG; Gang LIU

doi:10.37188/CJLCD.2021-0113

您当前的位置：

首页 >

文章列表页 >

Multi-channel video target tracking algorithm based on feature constraint and optical flow field model

Target Detection and Tracking | 更新时间：2021-11-08

- Multi-channel video target tracking algorithm based on feature constraint and optical flow field model
- Chinese Journal of Liquid Crystals and Displays Vol. 36, Issue 11, Pages: 1554-1564(2021)
- 作者机构：
  
  1.长沙师范学院信息科学与工程学院, 湖南长沙 410100
  2.中南大学物理与电子学院, 湖南长沙 410083
- 作者简介：
- 基金信息：
  
  Youth Fund of National Natural Science Foundation of China(41904127);Research Project of Teaching Reform in Hunan Province Universities(HNJG-2021-1195);Ministry of Education China University Industry-University-Pesearch Innovation Fund(2020JTA05028);Ministry of Education Cooperative Education Project(201901014024);Key Scientific Research Project of Hunan Provincial Department of Education(20A036)
- DOI：10.37188/CJLCD.2021-0113
  CLC： TP391.41
- Received：23 April 2021，
  
  Revised：24 May 2021，
  
  Published：2021-11
- 稿件说明：
移动端阅览
Bo ZHANG, Hui LONG, Gang LIU. Multi-channel video target tracking algorithm based on feature constraint and optical flow field model[J]. Chinese journal of liquid crystals and displays, 2021, 36(11): 1554-1564.
DOI：

Bo ZHANG, Hui LONG, Gang LIU. Multi-channel video target tracking algorithm based on feature constraint and optical flow field model[J]. Chinese journal of liquid crystals and displays, 2021, 36(11): 1554-1564. DOI： 10.37188/CJLCD.2021-0113.

摘要

现有的目标跟踪算法中网格信息的冗杂度较高，增加了跟踪算法的复杂程度，使得跟踪时计算量过大，影响了跟踪效果。为此，以降低跟踪过程复杂度为首要目标，本文在考虑特征约束的情况下，利用光流场模型实现对多通道视频目标的有效跟踪。通过设定视频随机变量控制光流场数值模型的变化控制偏差并提取出多通道视频的归一化特征，引入全局池化处理方式解决传统方法多维度特征计算繁杂的问题。首先，根据图像像素强度与时间参数间的数值关系，控制光流场模型均值的偏差，得到视频特征；然后，利用激活函数将多通道参数数值映射处理为固定参数，构建多通道特征约束关系控制特征参数发生升维，减少数据计算维数；最后，划分视频图像中的背景及目标区域，通过构建像素运动模型完成对视频目标的跟踪。在设置实验仪器初始参数后，在数据库中随机选取视频片段作为实验对象，并通过设定不同帧图像的分辨率将其模拟为多通道视频信息，对传统算法与本文算法的性能展开测试，结果表明：本文算法不仅可以保证跟踪精确度高于0.90，还能将成功追踪率控制在96%以上，且其跟踪过程耗时最多仅需5.5 s。本文算法能够用复杂度较低的计算过程实现对多通道视频目标的跟踪，具有较低的计算量。

Abstract

The miscellaneity of grid information in the existing target tracking algorithm is high

which increases the complexity of the tracking algorithm and makes the calculation amount of tracking too large

which affects the tracking effect. Therefore

the primary goal is reducing the complexity of the tracking process. In this paper

under the condition of considering the feature constraints

the optical flow field model is used to achieve the effective tracking of multi-channel video targets. The video random variable is set to control the variation of the numerical model of optical flow field

so as to control the deviation and extract the normalized features of multi-channel video

and the global pooling processing method is introduced to solve the problem of complex calculation of multi-dimensional features in the traditional method. Firstly

according to the numerical relationship between image pixel intensity and time parameters

the deviation of the mean value of optical flow field model is controlled to obtain video features. Then

the activation function is used to map the multi-channel parameters into fixed parameters

and the multi-channel feature constraint relationship is constructed to control the increasing dimension of the feature parameters and reduce the calculation dimension of the data. Finally

the background and target areas in the video image are divided

and the video target is tracked by constructing the pixel motion model. After setting the initial parameters of the experimental instrument

the video clips are randomly selected from the database as the experimental object

and the image resolution of different frames is set to simulate them as multi-channel video information. The performance tests of the traditional algorithm and the proposed algorithm are carried out

and the results show that the algorithm in this paper can not only guarantee the tracking accuracy higher than 0.90

but also control the successful tracking rate above 96%

and the tracking process takes only 5.5 s at most. The algorithm in this paper can track multi-channel video target with low computational complexity and has low computational cost.

关键词

Keywords

references

胡正平 , 尹艳华 , 顾健新 . 位置-尺度异空间协调的多特征选择相关滤波目标跟踪算法 . 信号处理 , 2019 . 35 ( 12 ): 1979 - 1989 . https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201912007.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201912007.htm .

Z P HU , Y H YIN , J X GU . Multi-feature selection correlated filtering target tracking algorithms based on location-scale different space coordination . Journal of Signal Processing , 2019 . 35 ( 12 ): 1979 - 1989 . https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201912007.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201912007.htm .

刘威 , 靳宝 , 周璇 , 等 . 基于特征融合及自适应模型更新的相关滤波目标跟踪算法 . 智能系统学报 , 2020 . 15 ( 4 ): 714 - 721 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT202004014.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT202004014.htm .

W LIU , B JIN , X ZHOU , 等 . Correlation filter target tracking algorithm based on feature fusion and adaptive model updating . CAAI Transactions on Intelligent Systems , 2020 . 15 ( 4 ): 714 - 721 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT202004014.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT202004014.htm .

欧伟奇 , 尹辉 , 许宏丽 , 等 . 一种基于Multi-Egocentric视频运动轨迹重建的多目标跟踪算法 . 智能系统学报 , 2019 . 14 ( 2 ): 246 - 253 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT201902007.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT201902007.htm .

W Q OU , H YIN , H L XU , 等 . A multi-object tracking algorithm based on trajectory reconstruction on multi-egocentric video . CAAI Transactions on Intelligent Systems , 2019 . 14 ( 2 ): 246 - 253 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT201902007.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZNXT201902007.htm .

韩晓微 , 王雨薇 , 谢英红 , 等 . 基于双相关滤波器的多通道尺度自适应目标跟踪 . 仪器仪表学报 , 2019 . 40 ( 11 ): 73 - 81 . https://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201911010.htm https://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201911010.htm .

X W HAN , Y W WANG , Y H XIE , 等 . Multi-channel scale adaptive target tracking based on double correlation filter . Chinese Journal of Scientific Instrument , 2019 . 40 ( 11 ): 73 - 81 . https://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201911010.htm https://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201911010.htm .

张晓丽 , 张龙信 , 肖满生 , 等 . 快速多域卷积神经网络和光流法融合的目标跟踪 . 计算机工程与科学 , 2020 . 42 ( 12 ): 2217 - 2222 . DOI: 10.3969/j.issn.1007-130X.2020.12.015 http://doi.org/10.3969/j.issn.1007-130X.2020.12.015 .

X L ZHANG , L X ZHANG , M S XIAO , 等 . Target tracking by deep fusion of fast multi-domain convolutional neural network and optical flow method . Computer Engineering and Science , 2020 . 42 ( 12 ): 2217 - 2222 . DOI: 10.3969/j.issn.1007-130X.2020.12.015 http://doi.org/10.3969/j.issn.1007-130X.2020.12.015 .

王殿伟 , 许春香 , 刘颖 . 基于多特征融合的核相关滤波目标跟踪算法 . 计算机工程与设计 , 2019 . 40 ( 12 ): 3463 - 3468 . https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201912017.htm https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201912017.htm .

D W WANG , C X XU , Y LIU . Kernelized correlation filter for target tracking with multi-feature fusion . Computer Engineering and Design , 2019 . 40 ( 12 ): 3463 - 3468 . https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201912017.htm https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201912017.htm .

孙彦景 , 石韫开 , 云霄 , 等 . 基于多层卷积特征的自适应决策融合目标跟踪算法 . 电子与信息学报 , 2019 . 41 ( 10 ): 2464 - 2470 . DOI: 10.11999/JEIT180971 http://doi.org/10.11999/JEIT180971 .

Y J SUN , Y K SHI , X YUN , 等 . Adaptive strategy fusion target tracking based on multi-layer convolutional features . Journal of Electronics & Information Technology , 2019 . 41 ( 10 ): 2464 - 2470 . DOI: 10.11999/JEIT180971 http://doi.org/10.11999/JEIT180971 .

王殿伟 , 方浩宇 , 刘颖 , 等 . 一种基于改进RT-MDNet的全景视频目标跟踪算法 . 哈尔滨工业大学学报 , 2020 . 52 ( 10 ): 152 - 160, 174 . https://www.cnki.com.cn/Article/CJFDTOTAL-HEBX202010020.htm https://www.cnki.com.cn/Article/CJFDTOTAL-HEBX202010020.htm .

D W WANG , H Y FANG , Y LIU , 等 . Improved RT-MDNet for panoramic video target tracking . Journal of Harbin Institute of Technology , 2020 . 52 ( 10 ): 152 - 160, 174 . https://www.cnki.com.cn/Article/CJFDTOTAL-HEBX202010020.htm https://www.cnki.com.cn/Article/CJFDTOTAL-HEBX202010020.htm .

刘芳 , 王洪娟 , 黄光伟 , 等 . 基于自适应深度网络的无人机目标跟踪算法 . 航空学报 , 2019 . 40 ( 3 ): 322332 https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201903037.htm https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201903037.htm .

F LIU , H J WANG , G W HUANG , 等 . UAV target tracking algorithm based on adaptive depth network . Acta Aeronautica et Astronautica Sinica , 2019 . 40 ( 3 ): 322332 https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201903037.htm https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201903037.htm .

何嘉 , 奚峥皓 , 阚秀 . 基于地面估算与目标位置估计的多目标跟踪算法 . 电子技术应用 , 2020 . 46 ( 8 ): 112 - 116 . https://www.cnki.com.cn/Article/CJFDTOTAL-DZJY202008038.htm https://www.cnki.com.cn/Article/CJFDTOTAL-DZJY202008038.htm .

J HE , Z H XI , X KAN . Multi-target tracking algorithm based on ground projection and target position estimation . Application of Electronic Technique , 2020 . 46 ( 8 ): 112 - 116 . https://www.cnki.com.cn/Article/CJFDTOTAL-DZJY202008038.htm https://www.cnki.com.cn/Article/CJFDTOTAL-DZJY202008038.htm .

储珺 , 危振 , 缪君 , 等 . 基于遮挡检测和多块位置信息融合的分块目标跟踪算法 . 模式识别与人工智能 , 2020 . 33 ( 1 ): 59 - 65 . https://www.cnki.com.cn/Article/CJFDTOTAL-MSSB202001010.htm https://www.cnki.com.cn/Article/CJFDTOTAL-MSSB202001010.htm .

J CHU , Z WEI , J MIAO , 等 . Block target tracking based on occlusion detection and multi-block position information fusion . Pattern Recognition and Artificial Intelligence , 2020 . 33 ( 1 ): 59 - 65 . https://www.cnki.com.cn/Article/CJFDTOTAL-MSSB202001010.htm https://www.cnki.com.cn/Article/CJFDTOTAL-MSSB202001010.htm .

刘晓悦 , 王云明 . 基于HOG-SVM的改进跟踪-学习-检测算法的目标跟踪方法 . 科学技术与工程 , 2019 . 19 ( 27 ): 266 - 271 . DOI: 10.3969/j.issn.1671-1815.2019.27.038 http://doi.org/10.3969/j.issn.1671-1815.2019.27.038 .

X Y LIU , Y M WANG . Improved tracking-learning-deterction target tracking method based on HOG-SVM . Science Technology and Engineering , 2019 . 19 ( 27 ): 266 - 271 . DOI: 10.3969/j.issn.1671-1815.2019.27.038 http://doi.org/10.3969/j.issn.1671-1815.2019.27.038 .

宋建锋 , 苗启广 , 申猛 , 等 . 多特征融合的相关滤波红外单目标跟踪算法 . 西安电子科技大学学报 , 2019 . 46 ( 5 ): 142 - 147 . https://www.cnki.com.cn/Article/CJFDTOTAL-XDKD201905023.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XDKD201905023.htm .

J F SONG , Q G MIAO , M SHEN , 等 . Algorithm for tracking an infrared single target based on correlation filtering with multi-feature fusion . Journal of Xidian University , 2019 . 46 ( 5 ): 142 - 147 . https://www.cnki.com.cn/Article/CJFDTOTAL-XDKD201905023.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XDKD201905023.htm .

王译萱 , 吴小俊 . 基于深度特征与局部约束掩膜的相关滤波跟踪算法 . 数据采集与处理 , 2019 . 34 ( 5 ): 808 - 818 . https://www.cnki.com.cn/Article/CJFDTOTAL-SJCJ201905005.htm https://www.cnki.com.cn/Article/CJFDTOTAL-SJCJ201905005.htm .

Y X WANG , X J WU . Correlation filter tracking algorithm based on deep features with local con-strained mask . Journal of Data Acquisition and Processing , 2019 . 34 ( 5 ): 808 - 818 . https://www.cnki.com.cn/Article/CJFDTOTAL-SJCJ201905005.htm https://www.cnki.com.cn/Article/CJFDTOTAL-SJCJ201905005.htm .

黄树成 , 张瑜 , 张天柱 , 等 . 基于条件随机场的深度相关滤波目标跟踪算法 . 软件学报 , 2019 . 30 ( 4 ): 927 - 940 . https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201904006.htm https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201904006.htm .

S C HUANG , Y ZHANG , T Z ZHANG , 等 . Improved deep correlation filters via conditional random field . Journal of Software , 2019 . 30 ( 4 ): 927 - 940 . https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201904006.htm https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201904006.htm .

江南 . 基于动态自适应相关滤波的复杂场景目标跟踪算法 . 电子测量与仪器学报 , 2019 . 33 ( 2 ): 140 - 147 . https://www.cnki.com.cn/Article/CJFDTOTAL-DZIY201902020.htm https://www.cnki.com.cn/Article/CJFDTOTAL-DZIY201902020.htm .

N JIANG . Visual target tracking algorithm based on dynamic adaptive filtering . Journal of Electronic Measurement and Instrumentation , 2019 . 33 ( 2 ): 140 - 147 . https://www.cnki.com.cn/Article/CJFDTOTAL-DZIY201902020.htm https://www.cnki.com.cn/Article/CJFDTOTAL-DZIY201902020.htm .

刘波 , 许廷发 , 李相民 , 等 . 自适应上下文感知相关滤波跟踪 . 中国光学 , 2019 . 12 ( 2 ): 265 - 273 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201902007.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201902007.htm .

B LIU , T F XU , X M LI , 等 . Adaptive context-aware correlation filter tracking . Chinese Optics , 2019 . 12 ( 2 ): 265 - 273 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201902007.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201902007.htm .

宋俊芳 , 王菽裕 , 薛茹 , 等 . 基于数据关联融合KCF与Kalman滤波的车辆多目标跟踪 . 科学技术与工程 , 2020 . 20 ( 31 ): 12927 - 12933 . DOI: 10.3969/j.issn.1671-1815.2020.31.037 http://doi.org/10.3969/j.issn.1671-1815.2020.31.037 .

J F SONG , S Y WANG , R XUE , 等 . Multi-vehicle tracking combing KCF and Kalman based on data association . Science Technology and Engineering , 2020 . 20 ( 31 ): 12927 - 12933 . DOI: 10.3969/j.issn.1671-1815.2020.31.037 http://doi.org/10.3969/j.issn.1671-1815.2020.31.037 .

张博言 , 钟勇 , 李振东 . 基于动态模式和卷积特征的单目标跟踪算法 . 西北工业大学学报 , 2019 . 37 ( 6 ): 1310 - 1319 . https://www.cnki.com.cn/Article/CJFDTOTAL-XBGD201906027.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XBGD201906027.htm .

B Y ZHANG , Y ZHONG , Z D LI . A visual object tracking algorithm based on dynamics pattern and convolutional feature . Journal of Northwestern Polytechnical University , 2019 . 37 ( 6 ): 1310 - 1319 . https://www.cnki.com.cn/Article/CJFDTOTAL-XBGD201906027.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XBGD201906027.htm .

张能波 , 苏振斌 , 谢维信 . Lasso约束下融和光流信息的DCF目标跟踪算法 . 信号处理 , 2019 . 35 ( 5 ): 911 - 918 . https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201905024.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201905024.htm .

N B ZHANG , Z B SU , W X XIE . DCF visual object tracking algorithm with lasso constraints and fusion optical flow . Journal of Signal Processing , 2019 . 35 ( 5 ): 911 - 918 . https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201905024.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XXCN201905024.htm .

茅正冲 , 陈海东 . 基于核相关滤波的长期目标跟踪算法 . 激光与光电子学进展 , 2019 . 56 ( 1 ): 010702 https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201901011.htm https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201901011.htm .

Z C MAO , H D CHEN . Long-term object tracking algorithm based on kernelized correlation filter . Laser & Optoelectronics Progress , 2019 . 56 ( 1 ): 010702 https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201901011.htm https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201901011.htm .

吴言枫 , 王延杰 , 孙海江 , 等 . 复杂动背景下的"低小慢"目标检测技术 . 中国光学 , 2019 . 12 ( 4 ): 854 - 866 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201904015.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201904015.htm .

Y F WU , Y J WANG , H J SUN , 等 . LSS-target detection in complex sky backgrounds . Chinese Optics , 2019 . 12 ( 4 ): 854 - 866 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201904015.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201904015.htm .

郭利 , 周盛宗 . 基于通道权重融合特征的目标跟踪算法 . 计算机系统应用 , 2020 . 29 ( 12 ): 178 - 186 . https://www.cnki.com.cn/Article/CJFDTOTAL-XTYY202012024.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XTYY202012024.htm .

L GUO , S Z ZHOU . Object tracking algorithm with fusion feature based on channel weight . Computer Systems & Applications , 2020 . 29 ( 12 ): 178 - 186 . https://www.cnki.com.cn/Article/CJFDTOTAL-XTYY202012024.htm https://www.cnki.com.cn/Article/CJFDTOTAL-XTYY202012024.htm .

姜可孟 , 曾聪文 , 江泽涛 . 基于光照不变特征的无模式跟踪算法 . 计算机工程与设计 , 2019 . 40 ( 1 ): 161 - 166 . https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201901026.htm https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201901026.htm .

K M MENG , C W ZENG , Z T JIANG . Model-free tracking algorithm based on illumination invariant features . Computer Engineering and Design , 2019 . 40 ( 1 ): 161 - 166 . https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201901026.htm https://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ201901026.htm .

张玲 , 田鹏 , 王溢 , 等 . 基于多空间显著特征的目标跟踪算法 . 计算机技术与发展 , 2019 . 29 ( 11 ): 71 - 76 . DOI: 10.3969/j.issn.1673-629X.2019.11.015 http://doi.org/10.3969/j.issn.1673-629X.2019.11.015 .

L ZHANG , P TIAN , Y WANG , 等 . Object tracking algorithm based on discriminative feature in multiple sub-spaces . Computer Technology and Development , 2019 . 29 ( 11 ): 71 - 76 . DOI: 10.3969/j.issn.1673-629X.2019.11.015 http://doi.org/10.3969/j.issn.1673-629X.2019.11.015 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Improved target tracking algorithm based on Swin-Transformer

Advances in twin network research in visual tracking technology

UAV target tracking with dual response anomaly constraint

Target tracking method based on improved ECO-HC

Related Author

Bo Zhang

Hui LONG

Gang LIU

LIU Shi

ZHU Ming

HE Zemin

ZENG Juntao

YUAN Baoxi

Related Institution

School of Physics and Electronics， Central South University

College of information Science and Engineering， Changsha Normal University

University of Chinese Academy of Sciences

Changchun Institute of Optics， Fine Mechanics and Physics， Chinese Academy of Sciences

School of Artificial Intelligence， Optics and Electronics， Northwestern Polytechnical University

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China 130033 Postal code：130033
Tel：0431-86176059 Email：yjxs@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰