基于卷积神经网络的鲁棒高精度目标跟踪算法

doi:10.3969/j.issn.0372-2112.2018.09.007

电子学报 ›› 2018, Vol. 46 ›› Issue (9): 2087-2093.DOI: 10.3969/j.issn.0372-2112.2018.09.007

基于卷积神经网络的鲁棒高精度目标跟踪算法

李康, 李亚敏, 胡学敏, 邵芳

湖北大学计算机与信息工程学院, 湖北武汉 430062

收稿日期:2017-06-30 修回日期:2017-09-08 出版日期:2018-09-25
- 通讯作者:
- 李亚敏
- 作者简介:
- 李康男,1986年生于安徽亳州,现为湖北大学计算机与信息工程学院讲师,主要研究方法为目标跟踪和机器学习.E-mail:likang@hubu.edu.cn;胡学敏 男,1985年出生于湖南岳阳,现为湖北大学计算机与信息工程学院讲师,主要研究方向为图像处理,模式识别和计算机视觉.E-mail:huxuemin2003@163.com;邵芳女,1995年生于湖北荆门.现为湖北大学计算机与信息工程学院本科生,主要研究方向为图像处理.E-mail:notheve@163.com
- 基金资助:
- 湖北省自然科学基金 (No.2017CFB305）

A Robust and Accurate Object Tracking Algorithm Based on Convolutional Neural Network

LI Kang, LI Ya-min, HU Xue-min, SHAO Fang

School of Computer Science and Information Engineering, Hubei University, Wuhan, Hubei 430062, China

Received:2017-06-30 Revised:2017-09-08 Online:2018-09-25 Published:2018-09-25
- Corresponding author:
- LI Ya-min
- Supported by:
- Natural Science Foundation of Hubei Province, China (No.2017CFB305)

摘要/Abstract

摘要： 目标跟踪是计算机视觉中重要的研究领域之一.为了跟踪复杂场景中外观变化剧烈的目标，本文提出了一种基于卷积神经网络的目标跟踪算法.算法中的网络模型结构包括预训练的特征提取层和自适应更新的分类器层.在开始跟踪前，首先训练全连接层和分类器层的参数，以及目标的特征与位置之间的线性关系.其次，定义了评估跟踪结果可信度的标准.如果得到的跟踪结果的可信度较高，则根据跟踪结果的特征调整位置，提高跟踪结果的精确度.最后，在训练网络时，每次迭代都选择分类器得分的最高的负样本参与训练.该策略可以提高模型的分辨能力.在OTB50测试集中的实验结果表明，我们的算法取得了良好的跟踪结果.

关键词: 目标跟踪, 神经网络, 计算机视觉, 机器学习

Abstract: Object tracking is one of the most important area of computer vision. In order to track the object whose appearance changes dramatically in complex scene, we propose a tracking algorithm based on the convolutional neural network. The network of our tracker has two parts: the feature extraction layer and the adaptive classifier layer. At the beginning, we train a fully-connected layer, a softmax layer and the linear relationship between feature and position of these samples. Next, we define a reliability of the tracking result. If the result is reliable, we will adjust the result location according to its features. Finally, in the network training process, we select the negative samples with max classifying scores in each iteration. The strategy could improve distinguishability of our tracker. Experiments on OTB50 show that our tracker could achieve state-of-the-art performance.

Key words: object tracking, neural network, computer vision, machine learning

中图分类号:

TP391.41

李康, 李亚敏, 胡学敏, 等. 基于卷积神经网络的鲁棒高精度目标跟踪算法[J]. 电子学报, 2018, 46(9): 2087-2093.

LI Kang, LI Ya-min, HU Xue-min, et al. A Robust and Accurate Object Tracking Algorithm Based on Convolutional Neural Network[J]. Acta Electronica Sinica, 2018, 46(9): 2087-2093.

参考文献

[1] Yilmaz A,Javed O,Shah M.Object tracking:A survey[J].ACM Computing Surveys,2006,38(4):13.
[2] Yang H,Shao L,Zheng F,et al.Recent advances and trends in visual tracking:A review[J].Neurocomputing,2011,74(18):3823-3831.
[3] Wu Y,Lim J,Yang M H.Online object tracking:A benchmark[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Portland:IEEE Press,2013.2411-2418.
[4] Nam H,Han B.Learning multi-domain convolutional neural networks for visual tracking[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Las Vegas:IEEE Press,2016.4293-4302.
[5] Danelljan M,Robinson A,Khan F S,et al.Beyond correlation filters:Learning continuous convolution operators for visual tracking[A].European Conference on Computer Vision[C].Amsterdam:Springer International Publishing,2016.472-488.
[6] Wang L,Ouyang W,Wang X,et al.Visual tracking with fully convolutional networks[A].Proceedings of the IEEE International Conference on Computer Vision[C].Santiago:IEEE Press,2015.3119-3127.
[7] Qi Y,Zhang S,Qin L,et al.Hedged deep tracking[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Zurich:IEEE Press,2016.4303-4311.
[8] Hong S,You T,Kwak S,et al.Online tracking by learning discriminative saliency map with convolutional neural network[A].International Conference on Machine Learning[C].Lille,France:JMLR,2015.597-606.
[9] 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.
[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] 李康,何发智,潘一腾,等.基于簇相似的多分类器目标跟踪算法[J].电子学报,2016,44(4):821-825. LI Kang,HE Fa-zhi,PAN Yi-teng,SUN Hang.Multi-classifier object tracking based on cluster similarity[J].Acta Electronica Sinica,2016,44(4):821-825.(in Chinese)
[12] 李康,何发智,陈晓.基于MAP多子空间增量学习的目标跟踪算法[J].中国科学:信息科学,2016,4:005. LI Kang,HE Fa-zhi,CHEN Xiao.Visual tracking algorithm based on MAP multi-subspace incremental learning[J].Scientia Sinica Informationis,2016,4:005.(in Chinese)
[13] Li K,He F,Chen X.Real-time object tracking via compressive feature selection[J].Frontiers of Computer Science,2016,10(4):689-701.
[14] 王治丹,蒋建国,齐美彬.基于最大池图匹配的形变目标跟踪方法[J].电子学报,2017,45(3):704-711. WANG Zhi-dan,JIANG Jian-guo,QI Mei-bin.Deformable object tracking based on max-pooling graph matching[J].Acta Electronica Sinica,2017,45(3):704-711.(in Chinese)
[15] 宋涛,李鸥,刘广怡,等.基于改进协作目标外观模型的在线视觉跟踪[J].电子学报,2017,45(2):384-393. SONG Tao,LI Ou,LIU Guang-yi,CUI Hong-liang.Online visual tracking based on improved collaborative appearance model[J].Acta Electronica Sinica,2017,45(2):384-393.(in Chinese)
[16] Wang N,Shi J,Yeung D Y,et al.Understanding and diagnosing visual tracking systems[A].Proceedings of the IEEE International Conference on Computer Vision[C].Santiago:IEEE Press,2015.3101-3109.
[17] Krizhevsky A,Sutskever I,Hinton G E.Imagenet classification with deep convolutional neural networks[A].Advances in Neural Information Processing Systems[C].Lake Tahoe:Bartlett,2012.1097-1105.
[18] Girshick R,Donahue J,Darrell T,et al.Rich feature hierarchies for accurate object detection and semantic segmentation[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Columbus:IEEE Press,2014.580-587.
[19] Fan J,Xu W,Wu Y,et al.Human tracking using convolutional neural networks[J].IEEE Transactions on Neural Networks,2010,21(10):1610-1623.
[20] Zhang K,Zhang L,Yang M H.Real-time compressive tracking[A].European Conference on Computer Vision[C].Firenze:Springer Berlin Heidelberg,2012.864-877.
[21] Babenko B,Yang M H,Belongie S.Visual tracking with online multiple instance learning[A].IEEE Conference on Computer Vision and Pattern Recognition[C].Miami:IEEE Press,2009.983-990.
[22] Ross D A,Lim J,Lin R S,et al.Incremental learning for robust visual tracking[J].International Journal of Computer Vision,2008,77(1):125-141.

基于卷积神经网络的鲁棒高精度目标跟踪算法

A Robust and Accurate Object Tracking Algorithm Based on Convolutional Neural Network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	黄赟, 张帆, 郭威, 陈立, 羊光. 一种基于数据标准差的卷积神经网络量化方法[J]. 电子学报, 2023, 51(3): 639-647.
[2]	张金颢, 周恒, 张守龙, 蒋廷勇, 王胜涛, 刘真. 基于仿真及神经网络的大型电磁脉冲模拟器近区场计算[J]. 电子学报, 2023, 51(3): 712-719.
[3]	吕杭, 蒋明峰, 李杨, 张鞠成, 王志康. 基于混合时频域特征的卷积神经网络心律失常分类方法的研究[J]. 电子学报, 2023, 51(3): 701-711.
[4]	吴军君, 王涛, 王英楷, 王星辉. 基于迁移学习的LiPON制备工艺模拟优化[J]. 电子学报, 2023, 51(3): 687-693.
[5]	王子为, 鲁继文, 周杰. 基于自适应梯度优化的二值神经网络[J]. 电子学报, 2023, 51(2): 257-266.
[6]	杨利平, 侯振威, 辜小花, 郝峻永. 弱标签声音事件检测的空间-通道特征表征与自注意池化[J]. 电子学报, 2023, 51(2): 297-306.
[7]	许新征, 李杉. 基于特征膨胀卷积模块的轻量化技术研究[J]. 电子学报, 2023, 51(2): 355-364.
[8]	丁琪, 田萱, 孙国栋. 基于注意力增强的热点感知新闻推荐模型[J]. 电子学报, 2023, 51(1): 93-104.
[9]	李豪, 袁广林, 秦晓燕, 琚长瑞, 朱虹. 基于空间加权对数似然比相关滤波与Deep Snake的目标轮廓跟踪[J]. 电子学报, 2023, 51(1): 105-116.
[10]	王神龙, 雍宇, 吴晨睿. 基于伪孪生神经网络的低纹理工业零件6D位姿估计[J]. 电子学报, 2023, 51(1): 192-201.
[11]	王硕, 王坚, 王亚男, 宋亚飞. 一种基于特征融合的恶意代码快速检测方法[J]. 电子学报, 2023, 51(1): 57-66.
[12]	张永梅, 孙捷. 基于动静态特征双输入神经网络的咳嗽声诊断COVID-19算法[J]. 电子学报, 2023, 51(1): 202-212.
[13]	袁海英, 成君鹏, 曾智勇, 武延瑞. Mobile_BLNet：基于Big-Little Net的轻量级卷积神经网络优化设计[J]. 电子学报, 2023, 51(1): 180-191.
[14]	吴靖, 叶晓晶, 黄峰, 陈丽琼, 王志锋, 刘文犀. 基于深度学习的单帧图像超分辨率重建综述[J]. 电子学报, 2022, 50(9): 2265-2294.
[15]	毛国君, 王者浩, 黄山, 王翔. 基于剪边策略的图残差卷积深层网络模型[J]. 电子学报, 2022, 50(9): 2205-2214.