[1]谢维波,夏远祥,刘文.改进的核相关滤波目标跟踪算法[J].华侨大学学报(自然科学版),2017,38(3):379-384.[doi:10.11830/ISSN.1000-5013.201703017]
　XIE Weibo,XIA Yuanxiang,LIU Wen.Improved Object Tracking Algorithm Based on Kernelized Correlation Filter[J].Journal of Huaqiao University(Natural Science),2017,38(3):379-384.[doi:10.11830/ISSN.1000-5013.201703017]
点击复制

改进的核相关滤波目标跟踪算法()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第38卷
期数:
2017年第3期
页码:
379-384
栏目:
出版日期:
2017-05-20

文章信息/Info

Title:
Improved Object Tracking Algorithm Based on Kernelized Correlation Filter
文章编号:
1000-5013(2017)03-0379-06
作者:
谢维波;  夏远祥;  刘文
华侨大学 计算机科学与技术学院, 福建 厦门 361021
Author(s):
XIE Weibo;  XIA Yuanxiang;  LIU Wen
College of Computer Science and Technology, Huaqiao University, Xiamen 361021, China
关键词:
视频跟踪;  核相关滤波;  尺度计算;  角度计算;  遮挡检测
Keywords:
visual tracking;  kernelized correlation filter;  scale calculation;  angle calculation;  occlusion detection
分类号:
TP311
DOI:
10.11830/ISSN.1000-5013.201703017
文献标志码:
A
摘要:
提出一种基于核相关滤波的尺度和旋转自适应跟踪算法.首先,利用核相关滤波确定目标的中心位置;然后,使用特征点匹配的方式估计目标的尺度变化和旋转角度.在特征点匹配过程中,使用前、后两次光流匹配消除不稳定特征点;计算特征点对的权重分布,从而估计出目标的最佳尺度和角度;判断当前目标是否受到遮挡,进而使用更合理的方式更新特征点集和目标模型,进一步提高算法的鲁棒性.实验结果表明:文中算法不仅能在一定程度上处理目标外观变化问题,而且跟踪的实时性较好.
Abstract:
A scale and rotation adaptive tracking algorithm based on kernelized correlation filter is proposed. Firstly, the algorithm determines the center positon of the object via kernelized correlation filter. Then the algorithm estimates the scale changes and rotation angle of an object using keypoints matching. In the process of keypoints matching, the method eliminates unstable keypoints using forward and backward matching. The next, the algorithm estimates the optimal scale and angle by considering the weight of keypoints. At last, the method detects whether the target is occluded, and then update keypoints set and object model more reasonable, and hence improving the robustness of the algorithm. The results of experiments show that the proposed algorithm not only can handle changes of object’s appearance to a certain degree, but also have high tracking efficiency.

参考文献/References:

[1] BOLME D S,BEVERIDGE J R,DRAPER B A.Visual object tracking using adaptive correlation filters[C]//Proceedings of the International Conference on Computer Vision and Pattern Recognition.San Francisco:IEEE Press,2010:2544-2550.
[2] HENRIQUES J F,CASERIO R,MARTINS P.Exploiting the circulant structure of tracking-by-detection with kernels[C]//Proceedings of the Europe Conference on Computer Vision.Florence:Springer Press,2012:702-715.
[3] ZHANG Kaihua,ZHANG Lei,LIU Qingshan,et al.Fast visual tracking via dense spatio-temporal context learning[C]//Proceedings of the Europe Conference on Computer Vision.Zurich:Springer Press,2014:127-141.
[4] HENRIQUES J F,CASERIO 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.
[5] SEVILL L,MILLER E L.Distribution fields for tracking[C]//Proceedings of the International Conference on Computer Vision and Pattern Recognition.Providence:IEEE Press,2012:1910-1917.
[6] ZHANG Kaihua,ZHANG Lei,YANG Minghusan.Real-time compressive tracking[C]//Proceedings of the Europe Conference on Computer Vision.Florence:Springer Press,2012:864-877.
[7] WU Yi,LIM J,YANG M H.Online object tracking: A benchmark[C]//Proceedings of the International Conference on Computer Vision and Pattern Recognition.Portland:IEEE Press,2013:2411-2418.
[8] DANELJAN M,HAGER G,KHAN F S.Accurate scale estimation for robust visual tracking[C]//Proceedings of British Machine Vision Conference.Nottingham:Springer Press,2014:1-11.
[9] 钟必能,陈雁,沈映菊.在线机器学习跟踪算法的进展[J].华侨大学学报(自然科学版),2014,35(1):41-46.
[10] PATNAIK R,CASASENT D.Fast fft-based distortion-invariant kernel filters for general object recognition[C]//Proceedings of the International Society for Optics and Photonics.San Jose:IEEE Press,2009:725202.
[11] ROSTEN E,PORTER R,DRUMMOND T.Faster and better: A machine learning approach to corner detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(3):105-119.
[12] CALONDER M,LEPETIT V,STRECHA C,et al.BRIEF: Binary robust independent elementary features[C]//Proceedings of the Europe Conference on Computer Vision.Heraklion:Springer Press,2010:778-792.
[13] SHI Jianbo,TOMASI C.Good feature to track[C]//Proceedings of the International Conference on Computer Vision and Pattern Recognition.Seattle:IEEE Press,1994:21-23.

相似文献/References:

[1]谢维波,夏远祥,刘文.采用互补特征的核相关滤波目标跟踪算法[J].华侨大学学报(自然科学版),2018,39(3):429.[doi:10.11830/ISSN.1000-5013.201611030]
　XIE Weibo,XIA Yuanxiang,LIU Wen.Target Tracking Algorithm Using Complementary Features of Kernelized Correlation Filter[J].Journal of Huaqiao University(Natural Science),2018,39(3):429.[doi:10.11830/ISSN.1000-5013.201611030]

备注/Memo

备注/Memo:
收稿日期: 2016-05-14
通信作者: 谢维波(1964-),男,教授,博士,主要从事信号处理、视频图像分析的研究.E-mail:xwblxf@hqu.edu.cn.
基金项目: 国家自然科学基金资助项目(61271383); 华侨大学研究生科研创新能力培育计划资助项目(1400214007)
更新日期/Last Update: 2017-05-20