基于角点增强改进的TLD目标跟踪算法

孙保基; 张葆; 宋策; 龙思源

doi:10.3788/YJYXS20163109.0921

您当前的位置：

首页 >

文章列表页 >

基于角点增强改进的TLD目标跟踪算法

显示与图像 | 更新时间：2020-07-07

- 基于角点增强改进的TLD目标跟踪算法
- Improved TLD object tracking algorithm based on corner reinforced
- 液晶与显示 2016年31卷第9期页码：921-928
- 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
- 作者简介：
  
  [ "孙保基(1993-),男,安徽宿州人,硕士研究生,主要从事图像处理与计算机视觉方面的研究。E-mail:sunbaoji15@163.com " ]
  张葆(1966-),男,吉林磐石人,研究员,博士生导师,主研航空光电成像技术。E-mail:cleresky@vip.sina.com
- 基金信息：
- DOI：10.3788/YJYXS20163109.0921
  中图分类号： TP391.4
- 收稿日期：2016-05-03，
  
  录用日期：2016-6-27，
  
  纸质出版日期：2016-09
- 稿件说明：
移动端阅览
孙保基, 张葆, 宋策, 等. 基于角点增强改进的TLD目标跟踪算法[J]. 液晶与显示, 2016,31(9):921-928.

Bao-ji SUN, Bao ZHANG, Ce SONG, et al. Improved TLD object tracking algorithm based on corner reinforced[J]. Chinese journal of liquid crystals and displays, 2016, 31(9): 921-928.
孙保基, 张葆, 宋策, 等. 基于角点增强改进的TLD目标跟踪算法[J]. 液晶与显示, 2016,31(9):921-928. DOI： 10.3788/YJYXS20163109.0921.

Bao-ji SUN, Bao ZHANG, Ce SONG, et al. Improved TLD object tracking algorithm based on corner reinforced[J]. Chinese journal of liquid crystals and displays, 2016, 31(9): 921-928. DOI： 10.3788/YJYXS20163109.0921.

摘要

针对TLD算法的特征点无法有效表述目标问题，提出了一种基于角点增强改进的TLD目标跟踪算法。改进算法在跟踪模块加入了对目标表述能力更强，具有光照不敏感性和旋转不变性的Shi-Tomas角点作为跟踪特征点。跟踪器运行时，在角点经光流法跟踪和双向误差检测后，利用剩余的稳定角点定位目标窗口。对照结果表明，改进算法在面对目标抖动和形变时可以稳定跟踪；有效抑制因跟踪平滑点造成的漂移现象；提高了跟踪的稳定性。针对TLD算法跟踪过程中因在线模板积累造成的计算量持续增大、实时性持续降低的问题，提出了一种依据相似度中值的模板判断删除机制。该删除机制在模板积累到设定阈值时运行，根据模板与当前目标的相似度，删除不再具备代表性的模板；调整模板空间并更新模板数目。实验表明，该删除机制在应对模板更新快、持续时间长的跟踪情景时有效降低算法计算量，实时性可提高约20%。

Abstract

A new object tracking algorithm has been proposed which is based on the corner reinforced for the problem of the feature points can't represent the target effectively in the TLD. The Shi-Tomas corner point has been added into the existing tracking points as the feature points

which is insensitive to the light change and invariant to rotation. When tracker is running

the target will be located by the reserved corner which is tracked by the optical flow and filtered by the error detector. The experiments show that the algorithm can track the target steadily when faced with the target shaking and appearance changing. And the algorithm can effectively prevent the drift tracking caused by the tracking smooth pixels and improve the tracking stability. In addition

a model deleting after judging system has been proposed for solving the problem of calculated amount increasing and real time decreasing caused by the online model accumulating during the TLD tracking. The system will delete the unrepresentative models according to the similarity of the target and the online model when the model quantity reaches to a certain amount. The experiments show that the system can decrease the computational complexity effectively

and increase the real time by 20% especially handling with the rapid model updating and longtime tracking circumstances.

关键词

Keywords

references

宋策.基于粒子滤波的目标跟踪技术研究[D].长春:中国科学院研究生院(长春光学精密机械与物理研究所),2014. http://cdmd.cnki.com.cn/article/cdmd-80139-1014263818.htm

SONG C.Research on target tracking based on particle filter[D].Changchun:University of Chinese Academy of Sciences (Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences),2014.(in Chinese)

YILMAZ A,JAVED O,SHAH M.Object tracking:a survey[J]. ACM Computing Surveys, 2006,38(4):13.

高文,朱明,贺柏根,等.目标跟踪技术综述[J].中国光学,2014,7(3):365-375.

GAO W,ZHU M,HE B G, et al. Overview of target tracking technology[J]. Chinese Optics, 2014,7(3):365-375.(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)

闫辉,许廷发,吴青青,等.多特征融合匹配的多目标跟踪[J].中国光学,2013,6(2):163-170.

YAN H,XU T F,WU Q Q, et al. Multi-object tracking based on multi-feature joint matching[J]. Chinese Optics, 2013,6(2):163-170.(in Chinese)

赵其杰,屠大维,高健,等.基于Kalman滤波的视觉预测目标跟踪及其应用[J].光学精密工程,2008,16(5):937-942.

ZHAO Q J,TU D W,GAO J, et al. Kalman filter based vision predicting and object tracking method and its application[J]. Optics and Precision Engineering ,2008,16(5):937-942.(in Chinese)

BABENKO B,YANG M H,BELONGIE S.Visual tracking with online multiple instance learning[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,Miami ,FL:IEEE,2009:983-990.

宋华军.基于支持向量机的目标跟踪技术研究[D].长春:中国科学院研究生院(长春光学精密机械与物理研究所),2006. http://cdmd.cnki.com.cn/article/cdmd-80139-2006179487.htm

SONG H J.Study on target tracking method based on SVM[D].Changchun:University of Chinese Academy of Sciences (Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences),2006.(in Chinese)

KALAL Z,MATAS J,MIKOLAJCZYK K.Online learning of robust object detectors during unstable tracking[C]// Proceedings of the IEEE 12th International Conference on Computer Vision Workshops ,Kyoto:IEEE,2009:1417-1424.

高帆,吴国平,刑晨,等.TLD目标跟踪算法研究[J].电视技术,2013,37(11):70-74,202.

GAO F,WU G P,XING C, et al. TLD Target tracking algorithm[J]. Video Engineering ,2013,37(11):70-74,202.(in Chinese)

王建刚,李醒飞,陈诚,等.改进TLD算法在光电跟踪中的应用[J].红外技术,2015,37(10):824-830.

WANG J G,LI X F,CHEN C, et al. Improved TLD approach applied in optoelectronic tracking[J]. Infrared Technology, 2015,37(10):824-830.(in Chinese)

KALAL Z,MIKOLAJCZYK K,MATAS J.Tracking-learning-detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence ,2012,34(7):1409-1422.

KALAL Z,MIKOLAJCZYK K,MATAS J.Forward-backward error:automatic detection of tracking failures[C]// Proceedings of the 20th International Conference on Pattern Recognition, Istanbul:IEEE,2010:2756-2759.

KALAL Z,MATAS J,MIKOLAJCZYK K.P-N learning:bootstrapping binary classifiers by structural constraints[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, San Francisco,CA:IEEE,2010:49-56.

谢凤英,赵丹培.Visual C++数字图象处理[M].北京:电子工业出版社,2008:9.

XIE F Y,ZHAO D P. Visual C++Digital Image Processing [M].BEIJING:Publishing House of Electronics Industry,2008:9.(in Chinese)

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

基于主干增强和特征重排的反无人机目标跟踪

基于卷积神经网络的响应自适应跟踪

多特征融合的多模板协同相关跟踪

基于压缩感知的变尺度目标跟踪技术

空间目标图像的天基动态识别