Please wait a minute...
浙江大学学报(理学版)  2022, Vol. 49 Issue (1): 10-18    DOI: 10.3785/j.issn.1008-9497.2022.01.002
图形模拟与目标跟踪     
基于事件相机的无人机目标跟踪算法
朱强,王超毅,张吉庆,尹宝才,魏小鹏(),杨鑫()
大连理工大学 计算机科学与技术学院,辽宁 大连 116000
UAV target tracking algorithm based on event camera
Qiang ZHU,Chaoyi WANG,Jiqing ZHANG,Baocai YIN,Xiaopeng WEI(),Xin YANG()
School of Computer Science and Technology,Dalian University of Technology,Dalian 116000,China
 全文: PDF(3961 KB)   HTML( 7 )
摘要:

无人机目标跟踪可应用于消防、军事等重要领域,已成为计算机视觉领域热门研究课题之一。现有的无人机目标跟踪算法大多基于传统RGB相机结合深度学习算法, 但此类算法一方面无法避免无人机机体抖动造成的运动模糊, 另一方面因传统RGB相机在低光照或过曝光场景下成像质量较差,难以跟踪目标,为此提出采用无人机搭载DAVIS事件相机的方法进行目标跟踪。设计了基于事件与灰度图的双模态融合跟踪网络,用Vicon运动捕捉系统制作了无人机视角下的目标跟踪Event-APS 28数据集,实现了在复杂光照场景下对目标物的有效跟踪。

关键词: 目标跟踪事件相机双模态融合跟踪网络    
Abstract:

The target tracking by unmanned aerial vehicle (UAV) has become a hot research topic in the field of computer vision.UAV target tracking can be applied to fire fighting, military and other important fields.At present, most UAV target tracking algorithms are based on traditional RGB cameras combined with deep learning algorithms. However, such methods are hard to deal with motion blur caused by UAV body jitter. Moreover, traditional RGB cameras have poor performance of imaging on low-illumination or over-exposure scenes. In order to solve the above problems, this paper presents an approach which takes the UAV with DAVIS event camera for target tracking, and designs a dual mode fusion tracking network based on events and grayscale, To better train the network, this paper employs the Vicon motion capture system to make the target tracking data sets:Event-APS 28 under the UAV perspective, ensuring that the target can be tracked effectively according to the image information even in complex lighting scenes.

Key words: target tracking    event camera    the dual-modal fusion tracking network
收稿日期: 2021-08-08 出版日期: 2022-01-18
CLC:  TP 391.41  
基金资助: 国家自然科学基金面上项目(61972067)
通讯作者: 魏小鹏,杨鑫     E-mail: xpwei@dlut.edu.cn;xinyang@dlut.edu.cn
作者简介: 朱强(1997—),ORCID: https://orcid.org/0000-0001-8014-3236,男,硕士,主要从事计算机图形学研究.
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
朱强
王超毅
张吉庆
尹宝才
魏小鹏
杨鑫

引用本文:

朱强,王超毅,张吉庆,尹宝才,魏小鹏,杨鑫. 基于事件相机的无人机目标跟踪算法[J]. 浙江大学学报(理学版), 2022, 49(1): 10-18.

Qiang ZHU,Chaoyi WANG,Jiqing ZHANG,Baocai YIN,Xiaopeng WEI,Xin YANG. UAV target tracking algorithm based on event camera. Journal of Zhejiang University (Science Edition), 2022, 49(1): 10-18.

链接本文:

https://www.zjujournals.com/sci/CN/10.3785/j.issn.1008-9497.2022.01.002        https://www.zjujournals.com/sci/CN/Y2022/V49/I1/10

数据集事件大小/M帧数时长/s
EED113. 41797. 8
EV-IMO12-76 8001 920
Event-APS 2810 64394 8802 372
表1  Event-APS 28数据集与其他事件数据集对比
图1  Event-APS 28数据集部分展示
图2  事件与灰度图双模态融合目标跟踪网络架构
图3  FFM网络架构
图4  CA机制网络架构
图5  不同算法跟踪结果对比
图6  各算法对比实验
算法低光照/%运动模糊/%

过曝光/

%

正常 光照/

%

总体/

%

MIL31.23.91.88.35.1
3.78.23.915.210.4
TLD111.85.25.911.26.2
4.910.212.123.713.6
KCF54.318.414.825.916.8
7.228.625.535.523.9
Median Flow67.615.612.723.713.4
9.823.422.832.819.4
SiamFC711.428.219.435.823.6
15.440.124.553.534.8
CLNet818.247.428.742.236.1
25.878.245.364.152.1
ATOM929.848.842.659.848.7
49.880.865.180.965.7
PrDiMP1048.649.735.365.152.5
61.382.359.786.871.2
本文算法60.759.260.762.560.2
80.585.878.681.581.6
表2  SR和PR对比
1 周小龙, 刘倩倩, 产思贤, 等. 基于事件相机的视觉跟踪算法综述[J]. 小型微型计算机系统, 2020, 41(11): 2325-2332. DOI:10.3969/j.issn.1000-1220.2020.11.015
ZHOU X L, LIU Q Q, CHAN S X, et al. Event camera-based visual tracking algorithms: A survey[J]. Journal of Chinese Computer Systems, 2020, 41(11): 2325-2332. DOI:10.3969/j.issn.1000-1220. 2020. 11.015
doi: 10.3969/j.issn.1000-1220. 2020. 11.015
2 MOSSé Y P, LAUDENSLAGER M, LONGO L, et al. Identification of ALK as a major familial neuroblastoma predisposition gene[J]. Nature, 2008, 455(7215): 930-935. DOI:10.1038/nature07261
doi: 10.1038/nature07261
3 刘勇. 一种改进CSK目标跟踪算法[J]. 电子技术与软件工程, 2018 (7): 155-157.
LIU Y. An improved CSK target tracking algorithm[J]. Electronic Technology & Software Engineering, 2018 (7): 155-157.
4 周正松, 陈虹君, 周红. 基于多特征融合的尺度自适应KCF目标跟踪算法[J]. 四川大学学报(自然科学版), 2020, 57(4): 697-703. DOI:10.3969/j.issn. 0490-6756.2020.04.012
ZHOU Z S, CHEN H J, ZHOU H. Scale-adaptive kernel correlation filtering tracking algorithm based on multi-feature fusion[J]. Journal of Sichuan University(Natural Science Edition), 2020, 57(4): 697-703. DOI:10.3969/j.issn.0490-6756.2020.04.012
doi: 10.3969/j.issn.0490-6756.2020.04.012
5 BERTINETTO L, VALMADRE J, HENRIQUES J F, et al. Fully-convolutional siamese networks for object tracking[C]// European Conference on Computer Vision. Cham: Springer, 2016. doi:10.1007/978-3-319-48881-3_56
doi: 10.1007/978-3-319-48881-3_56
6 ZHENG W Y, ZHANG J. Feature extraction and image retrieval based on AlexNet[C]// Eighth International Conference on Digital Image Processing(ICDIP 2016). Bellingham: International Society for Optics and Photonics, 2016. doi:10.1117/12.2243849
doi: 10.1117/12.2243849
7 BO L, YAN J J, WEI W, et al. High performance visual tracking with siamese region proposal network[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2018. doi:10.1109/cvpr.2018.00935
doi: 10.1109/cvpr.2018.00935
8 LIANG Z H, LIANG C J, ZHANG Y, et al. Tracking of moving target based on siammask for video SAR system[C]// 2019 IEEE International Conference on Signal, Information and Data Processing (ICSIDP). Piscataway: IEEE, 2019. doi:10.1109/icsidp47821.2019.9173432
doi: 10.1109/icsidp47821.2019.9173432
9 林靖豪. 用于视频问答的多级注意力循环神经网络算法研究[D]. 杭州: 浙江大学, 2019. doi:10.1007/s11063-019-10003-1
LIN J H. Multi-grained Hierarchical Attentional Recurrent Network for Video Question Answering[D]. Hangzhou: Zhejiang University, 2019. doi:10.1007/s11063-019-10003-1
doi: 10.1007/s11063-019-10003-1
10 ATTIA P, PHAN G Q, MAKER A V, et al. Autoimmunity correlates with tumor regression in patients with metastatic melanoma treated with anti-cytotoxic T-lymphocyte antigen-4[J]. Journal of Clinical Oncology, 2005, 23(25):6043-6053. DOI:10.1200/JCO.2005.06.205
doi: 10.1200/JCO.2005.06.205
11 考夫曼, 库尔茨, 阿姆伯格, 等. 基于事件相机的可变形对象跟踪CN111417983A[P]. 2020-07-24. doi:10.3969/j.issn.1000-7504.2008.06.001
KAUFMAN P, KURTZ D, AMBERG B, et al. Deformable Object Tracking Based on Event Camera CN111417983A[P]. 2020-07-24. doi:10.3969/j.issn.1000-7504.2008.06.001
doi: 10.3969/j.issn.1000-7504.2008.06.001
12 江盟, 刘舟, 余磊. 低维流形约束下的事件相机去噪算法[J]. 信号处理, 2019, 35(10): 1753-1761. DOI:10.16798/j.issn.1003-0530.2019.10.017
JIANG M, LIU Z, YU L. A denoising algorithm for event cameras based on low-dimensional manifold constraint[J]. Journal of Signal Processing, 2019, 35(10): 1753-1761. DOI:10.16798/j.issn.1003-0530. 2019.10.017
doi: 10.16798/j.issn.1003-0530. 2019.10.017
13 RUSSAKOVSKY O, DENG J, SU H, et al. ImageNet large scale visual recognition challenge[J]. International Journal of Computer Vision, 2015, 115(3): 211-252. DOI:10.1007/s11263-015-0816-y
doi: 10.1007/s11263-015-0816-y
14 王萌, 王策, 栗思思, 等. 深度学习模型融合正则化方法在高维数据特征筛选中的应用研究[J]. 中国卫生统计, 2021, 38(1): 73-75, 80. DOI:10.3969/j.issn. 1002-3674.2021.01.019
WANG M, WANG C, LI S S, et al. Application research of deep learning model fusion regularization method in high-dimensional data feature screening[J]. Chinese Journal of Health Statistics, 2021, 38(1): 73-75, 80. DOI:10.3969/j.issn.1002-3674.2021.01.019
doi: 10.3969/j.issn.1002-3674.2021.01.019
15 程建,周越,蔡念,等.基于粒子滤波的红外目标跟踪[J].红外与毫米波学报,2006,25(2):113-117. doi:10.3321/j.issn:1001-9014.2006.02.008
CHEN J,ZHOU Y,CAI N,et al. Infrared object tracking based on particle filters[J].Journal of Infrared Millimeter Waves,2006,25(2):113-117. doi:10.3321/j.issn:1001-9014.2006.02.008
doi: 10.3321/j.issn:1001-9014.2006.02.008
16 JANG E, GU S X, POOLE B. Categorical reparameterization with gumbel-softmax[C]// The fifth International Conference on Learning Representations. Toulon:ICLR, 2017.
17 高帆, 吴国平, 刑晨, 等. TLD目标跟踪算法研究[J]. 电视技术, 2013, 37(11): 70-74, 202. DOI:10.3969/j.issn.1002-8692.2013.11.019
GAO F, WU G P, XING C, et al. TLD target tracking algorithm[J]. Video Engineering, 2013, 37(11): 70-74, 202. DOI:10.3969/j.issn.1002-8692. 2013.11.019
doi: 10.3969/j.issn.1002-8692. 2013.11.019
18 LI Q, GE X S, WANG G. An improved TLD tracking method using compressive sensing[C]// Proceedings of the 2016 International Conference on Artificial Intelligence: Technologies & Applications.Paris: Atlantis Press, 2016. doi:10.2991/icaita-16.2016.64
doi: 10.2991/icaita-16.2016.64
19 MEADER J L, HART F L. CLNET:A computer model for tracking movement, decay, and concentrations throughout water distribution systems[C]// Critical Water Issues & Computer Applications. Reston: ASCE, 2015.
20 CHOI S, LEE J, LEE Y, et al. Robust long-term object tracking via improved discriminative model prediction[J]. ECCV 2020 Workshops. Cham: Springer, 2020. DOI:10.1007/978-3-030-68238-5_40
doi: 10.1007/978-3-030-68238-5_40
21 王海涛, 王荣耀, 王文皞, 等. 目标跟踪综述[J]. 计算机测量与控制, 2020, 28(4): 1-6,21. DOI:10. 16526/j.cnki.11-4762/tp.2020.04.001
WANG H T, WANG R Y, WANG W H, et al. A survey on recent advance and trends in object tracking[J]. Computer Measurement & Control, 2020, 28(4): 1-6,21. DOI:10.16526/j.cnki.11-4762/tp.2020.04.001
doi: 10.16526/j.cnki.11-4762/tp.2020.04.001
[1] 方于华,叶枫. MFDC-Net:一种融合多尺度特征和注意力机制的乳腺癌病理图像分类算法[J]. 浙江大学学报(理学版), 2023, 50(4): 455-464.
[2] 虞瑞麒,刘玉华,沈禧龙,翟如钰,张翔,周志光. 表征学习驱动的多重网络图采样[J]. 浙江大学学报(理学版), 2022, 49(3): 271-279.
[3] 祝锦泰,叶继华,郭凤,江蕗,江爱文. FSAGN: 一种自主选择关键帧的表情识别方法[J]. 浙江大学学报(理学版), 2022, 49(2): 141-150.
[4] 钟颖,王松,吴浩,程泽鹏,李学俊. 基于SEMMA的网络安全事件可视探索[J]. 浙江大学学报(理学版), 2022, 49(2): 131-140.
[5] 杨猛,丁曙,马云涛,谢佳翊,段瑞枫. 基于纹理特征的小麦锈病动态模拟方法[J]. 浙江大学学报(理学版), 2022, 49(1): 1-9.
[6] 余鹏, 刘兰, 蔡韵, 何煜, 张松海. 基于单目摄像头的自主健身监测系统[J]. 浙江大学学报(理学版), 2021, 48(5): 521-530.
[7] 傅汝佳, 冼楚华, 李桂清, 万隽杰, 曹铖, 杨存义, 高月芳. 面向表型精确鉴定的豆株快速三维重建[J]. 浙江大学学报(理学版), 2021, 48(5): 531-539.
[8] 桂志强, 姚裕友, 张高峰, 徐本柱, 郑利平. 3D-power图的快速生成方法[J]. 浙江大学学报(理学版), 2021, 48(4): 410-417.
[9] 徐敏, 王科, 戴浩然, 罗晓博, 余炜伦, 陶煜波, 林海. 基于电子病历的乳腺癌群组与治疗方案可视分析[J]. 浙江大学学报(理学版), 2021, 48(4): 391-401.
[10] 邹北骥, 杨文君, 刘姝, 姜灵子. 面向自然场景图像的三阶段文字识别框架[J]. 浙江大学学报(理学版), 2021, 48(1): 1-8.
[11] 陈园琼, 邹北骥, 张美华, 廖望旻, 黄嘉儿, 朱承璋. 医学影像处理的深度学习可解释性研究进展[J]. 浙江大学学报(理学版), 2021, 48(1): 18-29.
[12] 邓惠俊. 排序支持的交互数据分类算法及其应用[J]. 浙江大学学报(理学版), 2021, 48(1): 9-17.
[13] 李华飙, 侯小刚, 王婷婷, 赵海英. 基于规则学习的传统纹样统一生成模式研究[J]. 浙江大学学报(理学版), 2020, 47(6): 669-676.
[14] 檀结庆, 曹宁宁. 一种四边形网格上的Midedge细分格式[J]. 浙江大学学报(理学版), 2019, 46(2): 154-163.