南京大学学报(自然科学版) ›› 2019, Vol. 55 ›› Issue (1): 85–91.doi: 10.13232/j.cnki.jnju.2019.01.008

• • 上一篇    下一篇

基于稀疏模型和Gabor小波字典的跟踪算法

阚建飞,任永峰*,翟继友,董学育,霍 瑛   

  1. 南京工程学院能源研究院,南京,211167
  • 接受日期:2018-11-11 出版日期:2019-02-01 发布日期:2019-01-26
  • 通讯作者: 任永峰, E-mail:yfren@njit.edu.cn E-mail:yfren@njit.edu.cn
  • 基金资助:
    国家自然科学基金青年科学基金(61802174)

Tracking algorithm based on sparse model and Gabor wavelet dictionary

Kan Jianfei,Ren Yongfeng*,Zhai Jiyou,Dong Xueyu,Huo Ying   

  1. Energy Research Institute,Nanjing Institute of Technology,Nanjing,211167,China
  • Accepted:2018-11-11 Online:2019-02-01 Published:2019-01-26
  • Contact: Ren Yongfeng, E-mail:yfren@njit.edu.cn E-mail:yfren@njit.edu.cn

摘要: 基于稀疏表示理论的目标跟踪方法可以通过激活少量神经元完成目标的动态跟踪,但是要求在当前图像背景中的遮挡或者目标物的姿态变化是可以进行稀疏表示的小面积范围. 针对这一问题,基于Gabor函数和稀疏理论提出一种强鲁棒性的目标跟踪算法. 该算法首先使用目标模板在初始帧中创建Gabor字典,其次使用该字典对候选目标完成稀疏表示,最后通过对Gabor字典的更新完成目标跟踪. 实验结果表明了算法的有效性.

关键词: Gabor字典, 稀疏模型, 特征提取, 跟 踪

Abstract: Adaptive tracking methods are widely used for object tracking in computer vision. In the tracking process,the features used to describe the target and candidate target have a great impact on the target tracking results. It is difficult to find the features of the target,especially in the case of illumination change,morphological change,complex background and occlusion. Although some breakthroughs have been made in visual tracking in recent years,it is still a challenging problem to develop a robust tracker for the complex and changing scenes,such as illumination variation,partial occlusion,back ground clutter and pose changes. In this paper,aiming at the change of illumination and pose variation,a new algorithm based on sparse model and Gabor wavelet dictionary is proposed,which is based on the feature extraction of Gabor function. The method builds a Gabor dictionary through the target template in the initial frame,and then the candidate object is represented by the Gabor dictionary in the tracking process. As a result,a more generalized and high-effective tracker is constructed. Experimentally,our algorithm is shown to be able able to outperform state-of-the-art trackers on various benchmark videos. The results show the effectiveness of the tracking algorithm.

Key words: Gabor dictionary, sparse model, features extraction, tracking

中图分类号: 

  • TP391.1
[1] 王华剑,景占荣,羊 彦. 基于改进扩展卡尔曼粒子滤波的目标跟踪算法. 计算机应用研究,2011,28(5):1634-1643.(Wang H J,Jing Z R,Yang Y. Target tracking algorithm based on improved extend Kalman particle filter. Application Research of Computers,2011,28(5):1634-1643.)
  [2] 景 阳,刘 琳,曾昭龙等. 遮挡情况下的人体检测与跟踪. 科学技术与工程,2014,14(16):129-133,138.(Jing Y,Liu L,Zeng Z L,et al. Human detection and tracking under occlusion. Science Technology and Engineering,2014,14(16):129-133,138.)
[3] Hariharakrishnan K,Schonfeld D. Fast object tracking using adaptive block matching. IEEE Transactions on Multimedia,2005,7(5):853-859.
[4] Bai T X,Li Y F. Robust visual tracking with structured sparse representation appearance model. Pattern Recognition,2012,45(6):2390-2404.
[5] Donoho D L. Compressed sensing. IEEE Transactions on Information Theory,2006,52(4):1289-1306.
[6] ACˇehovin L,Leonardis A,Kristan M. Visual object tracking performance measures revisited. IEEE Transactions on Image Processing,2016,25(3):1261-1274.
[7] Kristan M,Leonardis A,Matas J,et al. The visual object tracking VOT2016 challenge results ∥ European Conference on Computer Vision - ECCV’16. Amsterdam,The Nether-lands:Springer International Publishing,2016:191-217.
[8] Hare S,Golodetz S,Saffari A,et al. Struck:Structured output tracking with kernels. IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(10):2096-2109.
[9] Zhang K H,Liu Q S,Wu Y,et al. Robust visual tracking via convolutional networks without training. IEEE Transactions on Image Processing,2016,25(4):1779-1792.
[10] Yong W,Hu S,Wu S. Object tracking based on Huber loss function. Visual Computer,2018(2):1-14.
[11] Mei X,Ling H B. Robust visual tracking using L1 minimization ∥ Proceedings of the 12th IEEE International Conference on Computer vision. Kyoto,Japan:IEEE,2009:1436-1443.
[12] 王然冉,张铁力,李晓佳. 基于粒子滤波器的人体目标跟踪. 计算机应用与软件,2008,25(12):247-248,251.(Wang R R,Zhang T L,Li X J. Human-body object tracking based on particle filter. Computer Applications and Software,2008,25(12):247-248,251.)
[13] Wang J Y,Chen X L,Gao W. Online selecting discriminative tracking features using particle filter ∥ Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego,CA,USA:IEEE,2015:1037-1042.
[14] Hartley R,Zisserman A. Multiple view geometry in computer vision. The 2nd Edition. Cambridge:Cambridge University Press,2003,560.
[15] Mei X,Ling H B. Robust visual tracking and vehicle classification via sparse representation. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(11):2259-2272.
[16] Danelljan M,Hager G,Khan F S,et al. Discriminative scale space tracking. IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(8):1561-1575.
[17] Wang D,Lu H C,Xiao Z Y,et al. Inverse sparse tracker with a locally weighted distance metric. IEEE Transactions on Image Processing,2015,24(9):2646-2657.
[1] 范 君, 业巧林, 业 宁. 基于线性鉴别的无参数局部保持投影算法[J]. 南京大学学报(自然科学版), 2019, 55(2): 211-220.
[2] 安 晶, 艾 萍, 徐 森, 刘 聪, 夏建生, 刘大琨. 一种基于一维卷积神经网络的旋转机械智能故障诊断方法[J]. 南京大学学报(自然科学版), 2019, 55(1): 133-142.
[3]  万鸣华1,2*,杨国为1,赖志辉2.  最大间距准则框架下的多流形局部图嵌入(MLGE/MMC)算法[J]. 南京大学学报(自然科学版), 2018, 54(2): 462-.
[4]  唐肝翌*,卢桂馥.  基于L1-范数的鲁棒稀疏的张量PCA人脸图像分析[J]. 南京大学学报(自然科学版), 2018, 54(1): 31-.
[5] 赵天青, 梁旭斌,许学忠*,蔡宗义,孙迪峰. EMD在目标声信号特征提取中的应用研究[J]. 南京大学学报(自然科学版), 2015, 51(7): 102-.
[6] 季舒瑶,袁飞*,程恩,陈柯宇. 基于声传播特性的供水管道泄漏检测与分类[J]. 南京大学学报(自然科学版), 2015, 51(7): 64-.
[7]  陈素根1,2, 吴小俊1*, 曹俊峰1
.  训练样本类内局部调整的人脸识别方法[J]. 南京大学学报(自然科学版), 2015, 51(1): 132-138.
[8] 陈 冲1,尤鸣宇1*,刘家铭1,王 铮1,李国正1,徐镶怀2,邱忠民2. 基于高频子带特征的咳嗽检测方法[J]. 南京大学学报(自然科学版), 2015, 51(1): 157-164.
[9]  李永忠**,王玉雷,刘真真
.  藏文印刷体字符识别技术研究*
[J]. 南京大学学报(自然科学版), 2012, 48(1): 55-62.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 胡 太, 杨 明. 结合目标检测的小目标语义分割算法[J]. 南京大学学报(自然科学版), 2019, 55(1): 73 -84 .
[2] 张霄涵,胡红钢*. 基于区块格的RFID双向认证协议[J]. 南京大学学报(自然科学版), 2019, 55(3): 458 -469 .
[3] 汪洋,陈泰格,陆晓凡,辛小燕,王坤,李茗,青钊,张英为,严晓敏,吴超,言方荣,张冰. COVID⁃19的临床和影像特征与试行指南的映证分析[J]. 南京大学学报(自然科学版), 2020, 56(3): 430 -436 .