采用地平面约束的双目PTZ主从跟踪方法

崔智高; 邓磊; 李艾华; 姜柯; 周杰

采用地平面约束的双目PTZ主从跟踪方法

崔智高^1,2,
邓磊¹,
李艾华²,
姜柯²,
周杰¹

1.
清华大学自动化系,北京 100084;
2.
第二炮兵工程大学 502教研室,陕西西安 710025

基金项目:

国家自然科学基金（61020106004，61021063，61225008）

详细信息

作者简介:
崔智高（1984－），男，博士生，主要从事双目视觉监控等方面的研究。Email：cuizg10@mails.tsinghua.edu.cn；周杰（1968－），男，教授，博士生导师，主要从事图像处理、双目视觉监控等方面的研究。Email：jzhou@mail.tsinghua.edu.cn

崔智高（1984－），男，博士生，主要从事双目视觉监控等方面的研究。Email：cuizg10@mails.tsinghua.edu.cn；周杰（1968－），男，教授，博士生导师，主要从事图像处理、双目视觉监控等方面的研究。Email：jzhou@mail.tsinghua.edu.cn

中图分类号: TP391.4

Master-slave tracking algorithm using dual PTZ cameras based on ground plane constraint

1.
Department of Automation,Tsinghua University,Beijing 100084,China;
2.
Faculty 502,The Second Artillery Engineering University,Xi'an 710025,China

摘要: 在视频监控领域，包含PTZ（pan-tilt-zoom）相机的双目主从系统可以同时获取跟踪目标的全景信息和高分辨率信息，因此得到了广泛研究与应用。针对智能视频监控的需求，提出了一种基于地平面约束的双目PTZ主从跟踪方法。该方法分为离线标定和在线实时跟踪两个阶段。离线阶段，利用两相机不同视角间的目标匹配关系计算地平面所诱导的单应矩阵，提出了一种从两相机同步视频流中自动估计单应矩阵的方法，该方法与传统方法相比具有不需要标定物和人工干预的优点，然后采用匹配特征点的方法估计相机的主点和等效焦距。在线实时跟踪时，通过单应变换建立主从相机之间的坐标关联，并利用离线阶段标定的主点和等效焦距估计从相机控制参数，从而实现主从跟踪。与其他算法相比，该方法可以应用于宽基线的情形，能够适应目标深度的变化，满足了实时性的要求。室内、外场景的多组实验验证了所提方法的有效性。
- 目标跟踪 /
- 主从跟踪 /
- 双目PTZ相机 /
- 地平面约束 /
- 单应矩阵
Abstract: Master-slave camera system had been widely used in visual surveillance domain because both large-view information and high-resolution local-view information of the tracked object could be obtained at the same time. For the widely demanding of intelligent visual surveillance, a master-salve tracking algorithm using dual PTZ(pan-tilt-zoom) cameras based on ground plane constraint was presented. The proposed method could be divided into two stages: offline calibration and online tracking. In offline stage, the view-to-view homography induced by the ground plane was computed using object matching from two cameras. A novel method that automatically estimated homography from synchronized video streams of the two cameras was presented. Compared with traditional methods, the proposed method didn't need calibration tools, and without the need of manual intervention. Then principal point and equivalent focal length were estimated using matched features. In online stage, the calculated homography transformation was used to establish coordinate correspondence of the master and slave cameras. With the estimated principal point and equivalent focal length, the control parameters of the salve camera could be computed. Therefore, master-salve tracking could be realized. Compared with other methods, the proposed algorithm can be used in the case of wide baseline. The proposed method can also adapt to the depth change of tracked target. Meanwhile, the requirements for real time master-salve tracking can be meeted. Quantitative results in the indoor and outdoor scene show good performance of the proposed approach.
- target tracking /
- master-slave tracking /
- dual PTZ(pan-tilt-zoom) cameras /
- ground plane constraint /
- homography

[1]	Mei Yuesong, Yu Jianqiao, Chen Xi. Moving object tracking in the moving background[J]. Infrared and Laser Engineering, 2011, 40(4): 757-761. (in Chinese)梅跃松, 于剑桥, 陈曦. 移动背景下的运动目标跟踪[J]. 红外与激光工程, 2011, 40(4): 757-761.
[2]	Mu Xinkan, Luo Haibo. Correlation tracking approach robust to the variation of image brightness[J]. Infrared and Laser Engineering, 2012, 41(1): 255-260. (in Chinese)穆欣侃, 罗海波. 一种对亮度变化鲁棒的相关跟踪方法[J]. 红外与激光工程, 2012, 41(1): 255-260.
[3]	Xue K, Liu Y, Ogunmakin G, et al. Panoramic gaussian mixture model and large-scale range background substraction method for PTZ camera-based surveillance systems[J]. Machine Vision and Application, 2012, 11(4): 1-16.
[4]	Varcheie P D Z, Bilodeau G A. People tracking using a network-based PTZ camera[J]. Machine Vision and Application, 2011, 22(4): 671-690.
[5]	Zhou X H, Collins R T, Kanade T, et al. A master-slave system to acquire biometric imagery of humans at a distance[C]//ACM SIGMM International Workshop on Video Surveillance, 2003: 113-120.
[6]	Horaud R, Knossow D, Michaelis M. Camera cooperation for achieving visual attention[J]. Machine Vision and Application, 2006, 16(6): 331-342.
[7]	Bimbo A D, Dini F, Lisanti G, et al. Exploiting distinctive visual landmark maps in pan-tilt-zoom camera networks[J]. Computer Vision and Image Understanding, 2010, 114(6): 611-623.
[8]	Li Y, Song L, Wang J. Automatic weak calibration of master-salve surveillance system based on mosaic image[C]//International Conference on Pattern Recognition Istanbul, Turkey, 2010: 1824-1827.
[9]	Hu W M, Hu M, Zhou X, et al. Principal axis-based correspondence between multiple cameras for people tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(4): 663-671.
[10]	Jiang Mingxin, Wang Hongyu, Liu Xiaokai. A multi-target tracking algorithm based on multiple cameras[J]. Acta Automatica Sinica, 2012, 38(4): 531-539. (in Chinese)姜明新, 王洪玉, 刘晓凯. 基于多相机的多目标跟踪算法[J]. 自动化学报, 2012, 38(4): 531-539.
[11]	Haritaoglu I, Harwood D, Davis L S. W4: real-time surveillance of people and their activities[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(8): 809-830.
[12]	Lowe D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2): 91-101.
[13]	Cao Yu, Feng Ying, Yang Yuntao, et al. Application of line estimation algorithm based on RANSAC in road points cloud optimization[J]. Infrared and Laser Engineering, 2012, 41(11): 3108-3112. (in Chinese)曹毓, 冯莹, 杨云涛, 等. RANSAC直线估计方法在路面三维点云优化中的应用[J]. 红外与激光工程, 2012, 41(11): 3108-3112.
[14]	Sinha S, Pollefeys M. Towards calibrating a pan-tilt-zoom cameras network[C]//European Conference on Computer Vision Workshop, 2004: 1-11.
[15]	Sinha S N, Pollefeys M. Pan-tilt-zoom camera calibration and high-resolution mosaic generation[J]. Computer Vision and Image Understanding, 2006, 103(3): 170-183.

[1]	王向军, 朱慧. 利用国产FPGA的高帧率目标跟踪方法 . 红外与激光工程, 2023, 52(9): 20220905-1-20220905-12. doi: 10.3788/IRLA20220905
[2]	李博, 张心宇. 复杂场景下基于自适应特征融合的目标跟踪算法 . 红外与激光工程, 2022, 51(10): 20220013-1-20220013-11. doi: 10.3788/IRLA20220013
[3]	张景程, 乔新博, 赵永强. 红外偏振摄像机动目标检测跟踪系统（特邀） . 红外与激光工程, 2022, 51(4): 20220233-1-20220233-10. doi: 10.3788/IRLA20220233
[4]	许聪, 孙大鹰, 曹子奇, 李春琦, 顾文华. 采用自适应模糊优化的目标跟踪加速方案 . 红外与激光工程, 2022, 51(2): 20210864-1-20210864-8. doi: 10.3788/IRLA20210864
[5]	张宏伟, 李晓霞, 朱斌, 张杨. 基于孪生神经网络的两阶段目标跟踪方法 . 红外与激光工程, 2021, 50(9): 20200491-1-20200491-12. doi: 10.3788/IRLA20200491
[6]	陈法领, 丁庆海, 罗海波, 惠斌, 常铮, 刘云鹏. 采用时空上下文的抗遮挡实时目标跟踪 . 红外与激光工程, 2021, 50(1): 20200105-1-20200105-11. doi: 10.3788/IRLA20200105
[7]	王向军, 郭志翼, 王欢欢. 基于嵌入式平台的低时间复杂度目标跟踪算法 . 红外与激光工程, 2019, 48(12): 1226001-1226001(10). doi: 10.3788/IRLA201948.1226001
[8]	卢瑞涛, 任世杰, 申璐榕, 杨小冈. 基于稀疏表示多子模板的鲁棒目标跟踪算法 . 红外与激光工程, 2019, 48(3): 326003-0326003(8). doi: 10.3788/IRLA201948.0326003
[9]	曲蕴杰, 莫宏伟, 王常虹. 一种用于无人机的目标颜色核相关跟踪算法研究 . 红外与激光工程, 2018, 47(3): 326001-0326001(7). doi: 10.3788/IRLA201847.0326001
[10]	葛宝义, 左宪章, 胡永江, 张岩. 基于双步相关滤波的目标跟踪算法 . 红外与激光工程, 2018, 47(12): 1226004-1226004(10). doi: 10.3788/IRLA201847.1226004
[11]	罗海波, 许凌云, 惠斌, 常铮. 基于深度学习的目标跟踪方法研究现状与展望 . 红外与激光工程, 2017, 46(5): 502002-0502002(7). doi: 10.3788/IRLA201746.0502002
[12]	马俊凯, 罗海波, 常铮, 惠斌, 周晓丹, 侯德飞. 基于可变形模型的目标跟踪算法 . 红外与激光工程, 2017, 46(9): 928001-0928001(9). doi: 10.3788/IRLA201746.0928001
[13]	邢运龙, 李艾华, 崔智高, 方浩. 改进核相关滤波的运动目标跟踪算法 . 红外与激光工程, 2016, 45(S1): 214-221. doi: 10.3788/IRLA201645.S126004
[14]	邵春艳, 丁庆海, 罗海波, 李玉莲. 采用高维数据聚类的目标跟踪 . 红外与激光工程, 2016, 45(4): 428002-0428002(10). doi: 10.3788/IRLA201645.0428002
[15]	徐超, 高敏, 杨耀. 自调整分层卡尔曼粒子滤波的快速目标跟踪 . 红外与激光工程, 2015, 44(6): 1942-1949.
[16]	崔雄文, 吴钦章, 蒋平, 周进. 子空间模型下的仿射不变目标跟踪 . 红外与激光工程, 2015, 44(2): 769-774.
[17]	田立, 周付根, 孟偲. 基于嵌入式多核DSP 系统的并行粒子滤波目标跟踪 . 红外与激光工程, 2014, 43(7): 2354-2361.
[18]	黎志华, 李新国. 基于OpenCV的红外弱小运动目标检测与跟踪 . 红外与激光工程, 2013, 42(9): 2561-2565.
[19]	彭晨, 陈钱, 钱惟贤, 徐富元. 复杂地面场景下的红外运动目标跟踪 . 红外与激光工程, 2013, 42(6): 1410-1414.
[20]	崔智高, 李艾华, 姜柯, 苏延召, 金广智. 双目协同多分辨率主动跟踪方法 . 红外与激光工程, 2013, 42(12): 3509-3516.

点击查看大图

计量

文章访问数: 404
HTML全文浏览量: 100
PDF下载量: 137
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

采用地平面约束的双目PTZ主从跟踪方法

Master-slave tracking algorithm using dual PTZ cameras based on ground plane constraint

计量

采用地平面约束的双目PTZ主从跟踪方法

1. 清华大学自动化系,北京 100084;

2. 第二炮兵工程大学 502教研室,陕西西安 710025

English Abstract

Master-slave tracking algorithm using dual PTZ cameras based on ground plane constraint

1. Department of Automation,Tsinghua University,Beijing 100084,China;

2. Faculty 502,The Second Artillery Engineering University,Xi'an 710025,China

全文HTML

目录

留言板

采用地平面约束的双目PTZ主从跟踪方法

Master-slave tracking algorithm using dual PTZ cameras based on ground plane constraint

计量

出版历程

采用地平面约束的双目PTZ主从跟踪方法

1. 清华大学 自动化系,北京 100084; 2. 第二炮兵工程大学 502教研室,陕西 西安 710025

English Abstract

Master-slave tracking algorithm using dual PTZ cameras based on ground plane constraint

1. Department of Automation,Tsinghua University,Beijing 100084,China; 2. Faculty 502,The Second Artillery Engineering University,Xi'an 710025,China

全文HTML

目录

1. 清华大学自动化系,北京 100084;

2. 第二炮兵工程大学 502教研室,陕西西安 710025

1. Department of Automation,Tsinghua University,Beijing 100084,China;

2. Faculty 502,The Second Artillery Engineering University,Xi'an 710025,China