计算机科学 ›› 2020, Vol. 47 ›› Issue (12): 169-176.doi: 10.11896/jsjkx.191000021

• 计算机图形学与多媒体 • 上一篇    下一篇

采用多相关滤波策略的鲁棒长时自适应目标跟踪

谭建豪, 殷旺, 刘力铭, 王耀南   

  1. 湖南大学电气与信息工程学院 长沙 410082
    机器人视觉感知与控制技术国家工程实验室 长沙 410082
  • 收稿日期:2019-10-04 修回日期:2020-03-07 出版日期:2020-12-15 发布日期:2020-12-17
  • 通讯作者: 殷旺(yinwang@hnu.edu.cn)
  • 作者简介:tanjianhao@hnu.edu.com
  • 基金资助:
    国家自然科学基金(61433016)

Robust Long-term Adaptive Object Tracking Based onMulti-correlation Filtering Strategy

TAN Jian-hao, YIN Wang, LIU Li-ming, WANG Yao-nan   

  1. College of Electrical and Information EngineeringHunan University Chansha 410082,China
    National Engineering Laboratory for Robot Visual Perception and Control Technology Changsha 410082,China
  • Received:2019-10-04 Revised:2020-03-07 Online:2020-12-15 Published:2020-12-17
  • About author:TAN Jian-hao ,born in 1962 Ph.D pro-fessor.His research interests include intelligent robot data mining pattern.recognitionsystem identification and image processing.
    YIN Wang ,born in 1995 master.His main research interests include machine vision robot Technology etc.

摘要: 传统相关滤波方法在目标运动模糊和光照变化上取得了一定的鲁棒效果但当目标存在形变、颜色变化、重度遮挡等干扰因素时难以实现跟踪鲁棒性差且当目标丢失后不能再恢复无法实现长时间跟踪.因此文中提出了一种鲁棒长时自适应目标跟踪算法.首先提出了一种特征互补策略将方向梯度直方图和全局颜色直方图的特征响应线性加权学习对颜色变化和形变都具有鲁棒性的相关滤波模型用以估计目标位移;然后仅提取目标前景HOG特征学习一个判别滤波器用以保持对目标外观的长期记忆使用该长期滤波器的输出响应来判别是否出现遮挡或跟踪失败采用在线SVM分类器对丢失目标进行再检测从而能够跟踪已丢失目标以实现长期跟踪;其次学习了以目标位置为中心的特征金字塔模型以预测尺度变化防止目标框漂移;最后在OTB目标跟踪基准数据集上对算法进行实验并与目前较为流行的目标跟踪算法进行对比进一步验证了所提算法的鲁棒性、准确性和优越性.

关键词: 长时目标跟踪, 颜色直方图, 相关滤波, SVM再检测器, 尺度自适应

Abstract: The traditional correlation filtering methods have recently achieved excellent performance and shown great robustness to exhibiting motion blur and illumination changes.Howeverit is difficult to achieve tracking when the object has interference factors such as deformationcolor changeand heavy occlusion.It shows poor robustness when the object is lost and cannot be recovered to achieve long-term tracking.Thereforthis paper proposes a robust long-term object tracking algorithm.Firsta feature complementation strategy is proposedwhich linearly weights the feature responses of the directional gradient histogram and the global color histogramand learns a correlation filtering model that is robust to color changes and deformations to estimate the target displacement.Thenthe object features are taken to learn a discriminant correlation filter to maintain long-term memory of object appearance.We use the output responses of this model to determine if tracking failure occurs.We use the online SVM classifier to re-detect the lost objectand retrack the lost target which can effectively recover the tracking target from failure to achieve long-term tracking.In additionthis paper learns a correlation filter over a feature pyramid centered at the estimated object position for predicting scale changes and further enhance robustness and accuracy.Finallythis paper compares the proposed algorithm with the state-of-the-art performance tracking algorithms on the online object tracking benchmark.The result shows that the proposed algorithm performs great robustness and accuracy.

Key words: Long-term object tracking, Color histogram, Correlation filter, SVM re-detector, Scale adaptation

中图分类号: 

  • TP391.41
[1] LIU D Q,LIU W J,FEI B W.Anti-jamming Matched TargetTracking Method with Foreground Constraints[J].Journal of Automation,2018,44(6):1138-1152.
[2] WANG Y N,LUO Q H,CHEN Y J.Multi-aircraft Visual-visual Tracking and Location System and Method for Rotor Flying Robot[J].Chinese Journal of Scientific Instrument,2018(2):1-10.
[3] YUAN J,YANG L,DONG X L.Mobile Robot Target Tracking Based on Online Classification of Motion Patterns[J].Chinese Journal of Scientific Instrument,2017,38(3):568-577.
[4] WU Y,LIM J,YANG M H.Online object tracking:A bench-mark[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2013:2411-2418.
[5] WU Y,LIM J,YANG M H.Online object tracking:A bench-mark[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2013:2411-2418.
[6] KRISTAN M,MATAS J,LEONARDIS A,et al.The visual object tracking vot2015 challenge results[C]//Proceedings of the IEEE International Conference on Computer Vision Workshops.2015:1-23.
[7] KRISTAN M,MATAS J,LEONARDIS A,et al.The visual object tracking VOT2015 challenge results[C]//In Proceedings of the 2015 IEEE International Conference on Computer Vision Workshops.Santiago,Chile:IEEE,2015:564-586.
[8] LIU C,ZHAO W,LIU P,et al.Selection,tracking and updatingof auxiliary targets in target tracking[J].Journal of Automation,2018,44(7):1195-1211.
[9] XIONG D,LU H M,XIAO J H,et al.Long time target tracking with scale and rotation adaptability[J].Journal of Automation,2019,45(2):289-304.
[10] BERTINETTO L,VALMADRE J,GOLODETZ S,et al.Sta-ple:Complementary learners for real-time tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:1401-1409.
[11] HADFIELD S J,BOWDEN R,LEBEDA K.The visual object tracking VOT2016 challenge results[J].Lecture Notes in Computer Science,2016,9914:777-823.
[12] MA C,YANG X,ZHANG C,et al.Long-term correlation trac-king[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2015:5388-5396.
[13] DANELLJAN M,BHAT G,SHAHBAZ K F,et al.Eco:Effi-cient convolution operators for tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:6638-6646.
[14] DANELLJAN M,HÄGER G,KHAN F,et al.Accurate scaleestimation for robust visual tracking[C]//British Machine Vision Conference.Nottingham,2014:1-5.
[15] BLASCHKO M B,LAMPERT C H.Learning to localize objects with structured output regression[C]//European Conference on Computer Vision.Springer,Berlin,Heidelberg,2008:2-15.
[16] HENRIQUES J F,CASEIRO R,MARTINS P,et al.High-speed tracking with kernelized correlation filters[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2014,37(3):583-596.
[17] HARE S,GOLODETZ S,SAFFARI A,et al.Struck:Structured output tracking with kernels[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,38(10):2096-2109.
[18] ZHANG J,MA S,SCLAROFF S.MEEM:robust tracking via multiple experts using entropy minimization[C]//European Conference on Computer Vision.Springer,Cham,2014:188-203.
[19] DANELLJAN M,HÄGER G,KHAN F,et al.Accurate scaleestimation for robust visual tracking[C]//British Machine Vision Conferenceö Nottingham.BMVA Press,2014:2014.
[20] BOLME D S,BEVERIDGE J R,DRAPER B A,et al.Visual object tracking using adaptive correlation filters[C]//2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.IEEE,2010:2544-2550.
[21] HENRIQUES J F,CASEIRO R,MARTINS P,et al.Exploiting the circulant structure of tracking-by-detection with kernels[C]//European Conference on Computer Vision.Springer,Berlin,Heidelberg,2012:702-715.
[22] DANELLJAN M,SHAHBAZ K F,FELSBERG M,et al.Adaptive color attributes for real-time visual tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2014:1090-1097.
[23] ZABIH R,WOODFILL J.Non-parametric local transforms forcomputing visual correspondence[C]//European Conference on Computer Vision.Springer,Berlin,Heidelberg,1994:151-158.
[24] MA C,HUANG J B,YANG X,et al.Adaptive correlation filters with long-term and short-term memory for object tracking[J].International Journal of Computer Vision,2018,126(8):771-796.
[25] CRAMMER K,DEKEL O,KESHET J,et al.Online passive-aggressive algorithms[J].Journal of Machine Learning Research,2006,7(Mar):551-585.
[26] HARE S,GOLODETZ S,SAFFARI A,et al.Struck:Structured output tracking with kernels[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,38(10):2096-2109.
[1] 程中建, 周双娥, 李康. 基于多尺度自适应权重的稀疏表示目标跟踪算法[J]. 计算机科学, 2020, 47(6A): 181-186.
[2] 喻露, 胡剑锋, 姚磊岳. 全局块与局部块协作的相关滤波目标跟踪算法[J]. 计算机科学, 2020, 47(6): 157-163.
[3] 马康, 娄静涛, 苏致远, 李永乐, 朱愿. 结合特征融合和尺度自适应的核相关滤波器目标跟踪算法研究[J]. 计算机科学, 2020, 47(11A): 224-230.
[4] 陈威, 李决龙, 邢建春, 杨启亮, 周启臻. 基于核相关滤波器和分层卷积特征的长时间目标跟踪[J]. 计算机科学, 2019, 46(9): 271-276.
[5] 李健鹏, 尚振宏, 刘辉. 融合多层卷积特征的相关滤波运动目标跟踪算法[J]. 计算机科学, 2019, 46(7): 252-257.
[6] 范蓉蓉, 樊佳庆, 刘青山. 实时高置信度更新补充学习跟踪[J]. 计算机科学, 2019, 46(3): 137-141.
[7] 赵广辉, 卓松, 徐晓龙. 基于卡尔曼滤波的多目标跟踪方法[J]. 计算机科学, 2018, 45(8): 253-257.
[8] 瞿中,赵从梅. 一种抗遮挡的自适应尺度目标跟踪算法[J]. 计算机科学, 2018, 45(4): 296-300.
[9] 贾娟娟, 贾富杰. 结合爬山法的模糊C均值彩色图像分割方法[J]. 计算机科学, 2018, 45(11A): 247-250.
[10] 黄健, 郭志波, 林科军. 一种基于核相关滤波的视觉跟踪算法[J]. 计算机科学, 2018, 45(11A): 230-233.
[11] 朱航江,朱帆,潘振福,朱永利. 运动状态与尺度估计的核相关目标跟踪方法[J]. 计算机科学, 2017, 44(Z11): 193-198.
[12] 袁大龙,纪庆革. 协同运动状态估计的多目标跟踪算法[J]. 计算机科学, 2017, 44(Z11): 154-159.
[13] 邹承明,薛栋,郭双双,赵广辉. 一种改进的图像相似度算法[J]. 计算机科学, 2016, 43(6): 72-76.
[14] 魏全禄,老松杨,白亮. 基于相关滤波器的视觉目标跟踪综述[J]. 计算机科学, 2016, 43(11): 1-5.
[15] 陈双叶,王善喜. PTZ摄像机跟踪运动目标的智能控制算法的研究[J]. 计算机科学, 2015, 42(Z11): 135-139.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 孙启,金燕,何琨,徐凌轩. 用于求解混合车辆路径问题的混合进化算法[J]. 计算机科学, 2018, 45(4): 76 -82 .
[2] 耿海军,施新刚,王之梁,尹霞,尹少平. 基于有向无环图的互联网域内节能路由算法[J]. 计算机科学, 2018, 45(4): 112 -116 .
[3] 侯彦娥,孔云峰,党兰学. 求解多车型校车路径问题的混合集合划分的GRASP算法[J]. 计算机科学, 2018, 45(4): 240 -246 .
[4] 蔡莉,梁宇,朱扬勇,何婧. 数据质量的历史沿革和发展趋势[J]. 计算机科学, 2018, 45(4): 1 -10 .
[5] 王正理,谢添,何琨,金燕. 考虑时间因素的0-1背包调度问题[J]. 计算机科学, 2018, 45(4): 53 -59 .
[6] 赵利博,刘奇,付方玲,何凌. 基于小波变换和倒谱分析的腭裂高鼻音等级自动识别[J]. 计算机科学, 2018, 45(4): 278 -284 .
[7] 李小薪,李晶晶,贺霖,刘志勇. 基于噪声空间结构嵌入和高维梯度方向嵌入的鲁棒人脸识别方法[J]. 计算机科学, 2018, 45(4): 285 -290 .
[8] 瞿中,赵从梅. 一种抗遮挡的自适应尺度目标跟踪算法[J]. 计算机科学, 2018, 45(4): 296 -300 .
[9] 杨沛安, 武杨, 苏莉娅, 刘宝旭. 网络空间威胁情报共享技术综述[J]. 计算机科学, 2018, 45(6): 9 -18 .
[10] 黄冬梅, 杜艳玲, 贺琪, 随宏运, 李瑶. 基于多属性最优化的海洋监测数据副本布局策略[J]. 计算机科学, 2018, 45(6): 72 -75 .