[1]董春燕,刘 怀,梁秦嘉.基于核相关滤波的TLD跟踪算法的研究[J].南京师范大学学报(工程技术版),2020,(04):037-43.[doi:10.3969/j.issn.1672-1292.2020.04.006]
 Dong Chunyan,Liu Huai,Liang Qinjia.TLD Object Tracking Algorithm Based on Kernelized Correlation Filter[J].Journal of Nanjing Normal University(Engineering and Technology),2020,(04):037-43.[doi:10.3969/j.issn.1672-1292.2020.04.006]
点击复制

基于核相关滤波的TLD跟踪算法的研究
分享到:

南京师范大学学报(工程技术版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2020年04期
页码:
037-43
栏目:
计算机科学与技术
出版日期:
2020-12-15

文章信息/Info

Title:
TLD Object Tracking Algorithm Based on Kernelized Correlation Filter
文章编号:
1672-1292(2020)04-0037-07
作者:
董春燕刘 怀梁秦嘉
南京师范大学电气与自动化工程学院,江苏 南京 210023
Author(s):
Dong ChunyanLiu HuaiLiang Qinjia
School of Electrical and Automation Engineering,Nanjing Normal University,Nanjing 210023,China
关键词:
目标跟踪核相关滤波特征融合遮挡检测
Keywords:
object trackingkernelized correlation filterfeature fusionocclusion detection
分类号:
TP391
DOI:
10.3969/j.issn.1672-1292.2020.04.006
文献标志码:
A
摘要:
对TLD跟踪算法进行改进,以提高在跟踪目标发生尺度变化或被遮挡时的跟踪性能. 首先使用KCF跟踪器替代TLD算法中原有的中值光流跟踪器,并在特征提取时增加目标的Lab颜色特征,在寻找目标位置时引入尺度估计,在模型更新阶段引入跟踪状态判别机制,通过设定跟踪器中输出响应最大值阈值、APCE阈值及检测器中随机蕨分类器阈值来判断跟踪器跟踪结果的可靠性,改善跟踪器在尺度变化、出现遮挡、光照变化等情况下的跟踪效果. 针对TLD算法中的检测器,为了减少大量无意义的窗口,提升算法在存在遮挡时的精确性,在检测之前使用Kalman滤波预估出目标位置,在预估位置周围使用改进的级联分类器更精准地定位目标,改进的级联分类器的前两级仍采用方差分类器和随机蕨分类器,第三级则采用改进的KCF跟踪器. 在OTB-50数据集上的实验结果分析表明,该算法跟踪性能优于其他算法,能够满足实时性.
Abstract:
TLD tracker is improved in this paper in order to increase the tracking performance when the tracked object scale is changing or in occlusion. For the tracker,KCF tracker is used to replace the original tracker of TLD algorithm,and KCF tracker is improved as follows:the Lab color feature of the object is added in feature extraction,scale estimation is introduced in search for the object location,in the model updating stage,and tracking state judgment mechanism is adopted to determine the reliability of the tracking results by setting the threshold value of maximum output response,the threshold value of APCE and the threshold value of the random fern classifier,so as to improve the tracking effect of the tracker under the conditions of scale variation,occlusion and illumination variation. For the detector,in order to reduce a large number of meaningless windows before detection,Kalman filter is used to estimate the object position,and then an improved cascade classifier is used to locate the object more accurately around the estimated position. Variance classifier and random fern classifier are still used in the first two levels of the improved cascade classifier,and the third level is classified by the improved KCF tracker. Qualitative and quantitative analysis of experimental results on the OTB-50 dataset show that the tracking performance of the proposed algorithm is superior to that of other tracking algorithms and that it can ensure the real-time performance.

参考文献/References:

[1] 卢湖川,李佩霞,王栋. 目标跟踪算法综述[J]. 模式识别与人工智能,2018,31(1):61-76.
[2]VOJIR T,NOSKOVA J,MATAS J. Robust scale-adaptive mean-shift for tracking[J]. Pattern Recognition Letters,2014,49:250-258.
[3]ZHANG T Z,XU C S,YANG M H. Learning multi-task correlation particle filters for visual tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2019,41(2):365-378.
[4]KALAL Z,MIKOLAJCZYK K,MATAS J. Tracking-learning-detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(7):1409-1422.
[5]HARE S,GOLODETZ S,SAFFARI A,et al. Struck:structured output tracking with kernels[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(10):2096-2109.
[6]LI H,LI Y,PORIKLI F. DeepTrack:learning discriminative feature representations online for robust visual tracking[J]. IEEE Transactions on Image Processing:A Publication of the IEEE Signal Processing Society,2016,25(4):1834-1848.
[7]CHOI J,KWON J,LEE K M. Real-time visual tracking by deep reinforced decision making[J]. Computer Vision and Image Understanding,2018,171:10-19.
[8]GORDON D,FARHADI A,FOX D. Re3:real-time recurrent regression networks for visual tracking of generic objects[J]. IEEE Robotics and Automation Letters,2018,3(2):788-795.
[9]WANG Q,ZHANG L,BERTINETTO L,et al. Fast online object tracking and segmentation:a unifying approach[C]//IEEE Conference on Computer Vision and Pattern Recognition. Los Angeles,USA:IEEE Computer Society,2019:1328-1338.
[10]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,2015,37(3):583-596.
[11]张惊雷,时鹏,温显斌. 基于尺度自适应均值偏移优化的TLD跟踪算法[J]. 控制与决策,2019,34(1):144-150.
[12]毛晓波,周晓东,刘艳红.基于FAST特征点改进的TLD目标跟踪算法[J]. 郑州大学学报(工学版),2018,39(2):1-5,17.
[13]王姣尧,侯志强,余旺盛,等. 采用核相关滤波的快速TLD视觉目标跟踪[J]. 中国图象图形学报,2018,23(11):1686-1696.
[14]常立博,杜慧敏,毛智礼,等. 面向视频监控的TLD改进目标跟踪算法研究[J]. 计算机工程与应用,2018,54(14):191-198.
[15]SINGH G,CHHABRA I. Effective and fast face recognition system using complementary OC-LBP and HOG feature descriptors with SVM classifier[J]. Journal of Information Technology Research,2018,11(1):91-110.
[16]BERTINETTO L,VALMADRE J,GOLODETZ S,et al. Staple:complementary learners for real-time tracking[C]//IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,USA:IEEE Computer Society,2016:1401-1409.
[17]WANG M,LIU Y,HUANG Z. Large margin object tracking with circulant feature maps[C]//IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,USA:IEEE Computer Society,2017:4800-4808.
[18]WU Y,LIM J,YANG M H. Object tracking benchmark[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1834-1848.

相似文献/References:

[1]沈世斌,谢 非,牛友臣,等.基于混合高斯模型优化的运动人体跟踪方法[J].南京师范大学学报(工程技术版),2019,19(01):051.[doi:10.3969/j.issn.1672-1292.2019.01.007]
 Shen Shibin,Xie Fei,Niu Youchen,et al.A Moving Human Body Tracking Method Based onOptimized Gaussian Mixture Model[J].Journal of Nanjing Normal University(Engineering and Technology),2019,19(04):051.[doi:10.3969/j.issn.1672-1292.2019.01.007]
[2]董春燕,刘 怀,梁秦嘉,等.自适应特征融合的核相关滤波目标跟踪算法研究[J].南京师范大学学报(工程技术版),2020,(03):050.[doi:10.3969/j.issn.1672-1292.2020.03.009]
 Dong Chunyan,Liu Huai,Liang Qinjia,et al.Kernelized Correlation Filter Tracking AlgorithmBased on Adaptive Feature Fusion[J].Journal of Nanjing Normal University(Engineering and Technology),2020,(04):050.[doi:10.3969/j.issn.1672-1292.2020.03.009]

备注/Memo

备注/Memo:
收稿日期:2020-03-06.
基金项目:国家自然科学基金项目(61603194).
通讯作者:刘怀,副教授,研究方向:数字图像处理、实时控制系统. E-mail:liuhuai@njnu.edu.cn
更新日期/Last Update: 2020-12-15