Research on target tracking based on improved optical flow method

LIU Xiangqian; YAN Juan; YANG Huibin; JIA Xiwei

Volume 35 Issue 3

Sep. 2021

Turn off MathJax

Article Contents

Article Navigation > Journal of Shanghai University of Engineering Science > 2021 > 35(3): 237-242

LIU Xiangqian, YAN Juan, YANG Huibin, JIA Xiwei. Research on target tracking based on improved optical flow method[J]. Journal of Shanghai University of Engineering Science, 2021, 35(3): 237-242.

Citation:

LIU Xiangqian, YAN Juan, YANG Huibin, JIA Xiwei. Research on target tracking based on improved optical flow method[J]. Journal of Shanghai University of Engineering Science, 2021, 35(3): 237-242.

LIU Xiangqian, YAN Juan, YANG Huibin, JIA Xiwei. Research on target tracking based on improved optical flow method[J]. Journal of Shanghai University of Engineering Science, 2021, 35(3): 237-242.

Citation:

LIU Xiangqian, YAN Juan, YANG Huibin, JIA Xiwei. Research on target tracking based on improved optical flow method[J]. Journal of Shanghai University of Engineering Science, 2021, 35(3): 237-242.

PDF( 1429 KB)

Research on target tracking based on improved optical flow method

School of Mechanical and Automotive Engineering, Shanghai University of Engineering Science, Shanghai 201620, China

Received Date: 2021-05-28
Publish Date: 2021-09-30

Abstract

Abstract

In view of the low operating efficiency of the optical flow method in tracking motion video sequences and the problem that the video sequence can only be processed in real time, designing a method based on optimized particle filter and optical flow method. Firstly, the algorithm strategy uses the method to find its target point, applys adaptive positioning to obtain the target central situations, and then processes the obtained video sequence with optical flow, then predicts the centroid movement information through particle filtering. Finally, in terms of the algorithm with the optical flow strategy, ViBe and YOLO algorithm, and track various video sequences and all kinds of objects in different situations. Test and simulation datas prove that the optimization strategy can not only effectively enhance the efficiency 13.7 percent and precision of target tracking 5.2 percent, but also demonstrate better anti-interference performance.
- target tracking,
- particle filter,
- low time complexity,
- optical flow method,
- video sequence,
- centroid positioning

FullText(HTML)

References(18)

References

[1]	XU C, WANG X X, DUAN S H, et al. Spatial-temporal constrained particle filter for cooperative target tracking[J] . Journal of network and computer applications,2021,176:102913.
[2]	GORJUP D, SLAVIC J, BABNIK A, et al. Still-camera multiview Spectral Optical Flow Imaging for 3D operating-deflection-shape identification[J] . Mechanical Systems & Signal Processing,2021,152:107456.
[3]	马晴. 基于深度学习和光流法的行人运动基本图获取方法[D]. 合肥: 中国科学技术大学, 2020.
[4]	潘宇巍, 何勇灵, 杨世春. 基于多级模糊控制的车辆目标跟踪研究[J] . 计算机仿真,2019(3):164 − 170.
[5]	熊炜, 王传胜, 李利荣, 等. 结合光流法和卡尔曼滤波的视频稳像算法[J] . 计算机工程与科学,2020(3):493 − 499. doi: 10.3969/j.issn.1007-130X.2020.03.015
[6]	陈玲, 李洁. 基于视觉传达的后继帧视频图像目标跟踪仿真[J] . 计算机仿真,2020(4):347 − 351.
[7]	李思嘉, 曹菲, 林浩申. 基于IMM-SCKF的海上机动目标跟踪算法研究[J] . 计算机仿真,2018(10):288 − 294. doi: 10.3969/j.issn.1006-9348.2018.10.058
[8]	蔡锦华, 祝义荣. 基于改进YOLOv3的目标跟踪算法研究[J] . 计算机仿真,2020(5):213 − 217, 321. doi: 10.3969/j.issn.1006-9348.2020.05.043
[9]	张保岗, 韩国栋, 汤先拓. 基于改进量子遗传算法的片上网络多目标映射技术[J] . 计算机应用与软件,2020(8):115 − 121. doi: 10.3969/j.issn.1000-386x.2020.08.021
[10]	庄博阳, 段建民, 郑榜贵, 等. 基于光流法的快速车道线识别算法研究[J] . 计算机测量与控制,2019(9):146 − 150.
[11]	刘夏轩德, 沈丹峰, 张旭祥, 等. 改进LK光流法在复杂环境中对移动小球目标追踪[J] . 计算机系统应用,2019(7):221 − 227.
[12]	张霞, 贺正然. 基于灰度与关键帧光流检测的视频异常判断[J] . 电子器件,2019(3):718 − 721. doi: 10.3969/j.issn.1005-9490.2019.03.035
[13]	梁硕. 基于背景减除法的运动目标检测与跟踪算法研究[D]. 西安: 西安石油大学, 2019.
[14]	怀天一. 基于信息几何的多传感器目标跟踪算法研究[D]. 杭州: 浙江大学, 2020.
[15]	张建丰. 运动图像目标跟踪优化仿真[J] . 计算机仿真,2017(6):256 − 259, 305. doi: 10.3969/j.issn.1006-9348.2017.06.055
[16]	韩玉兰, 韩崇昭, 薛丽. 多扩展目标混合粒子滤波器[J] . 控制工程,2019(6):1112 − 1117.
[17]	李献, 骆志伟, 于晋臣. MATLAB/Simulink系统仿真[M]. 北京: 清华大学出版社, 2017.
[18]	蔡瑞初, 谢伟浩, 郝志峰, 等. 基于多尺度时间递归神经网络的人群异常检测[J] . 软件学报,2015(11):2884 − 2896.