基于全方位视觉的多目标跟踪技术
作者:
基金项目:

国家自然科学基金(61070134)


Multi-object Tracking Based on Omni-Directional Vision Sensor
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [11]
  • |
  • 相似文献 [20]
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    为了快速准确地检测并跟踪多目标对象, 提出了一种基于全方位视觉的多目标对象跟踪方法. 首先采用全方位视觉传感器(ODVS)实时地采集现场360°全景视频图像;接着融合运动历史图像算法(MHI)和运动能量算法(MEI)实现了快速高效的MHoEI(Motion History or Energy Images)自动跟踪算法, 对多目标对象进行检测和跟踪;最后, 本文采用面向对象技术融合目标对象进行匹配跟踪实验结果表明本文提出的方法能较好地跟踪多目标对象, 具有鲁棒性高、运算量小、便于硬件实现、高效等优点.

    Abstract:

    To rapidly and accurately detect and track multiple objects, this paper presents a method of multiple objects detection and tracking based on Omni-directional images. Firstly, getting 360-degree Omni-directional images by Omni-directional Vision Sensor in real time, which is an effective way to solve the problem of the multiple video source correlation. Secondly, the MHoEI algorithm that is created by combining Motion History Images algorithm with Motion Energy Images algorithm can efficiently detect and track multiple objects in this paper. Finally,the matching algorithm based on object oriented technology is proposed through fusing many properties of objects,which was used to identify different objects. The algorithm better solves the problem of data association of different properties. Experiment results show that the method in this paper can better track objects in complex background. The system has robustness, lightly computational load, high efficiency features.

    参考文献
    1 Stauffer C, Grimson W. Learning patterns of activity using realtime tracking, IEEE Trans. Pattern Anal. Mach. Intell.2000,22(8):747-767.
    2 Haritaoglu I, Harwood D, Davis LS. W4: real-timesurveillance of people and their activities, IEEE Trans.PatternAnal.Mach. Intell. 2000,22(8):809-830.
    3 Amer A, Dubois E, Mitiche A. Real-time system forhigh-level video representation: application to videosurveillance: Conference on Visual Communication andImage Processing (VCIP). Proc. of the SPIE InternationalSymposium on Electronic Imaging. Santa Clara, USA,January 2003, 5022: 530-541.
    4 Collins R, Lipton A, Kanada T, et al. A system for video surveillanceand monitoring:VSAM final report,Technical reportCMU-RI-TR-00-12, Carnegie Mellon University, May 2000.
    5 Lv F, Kang J, Nevatia R, Cohen I, Medioni G. Automatic tracking and labeling of human activities in a video sequence.Proc.of the 6th IEEE International Workshop on PerformanceEvaluation of Tracking and Surveillance(PETS04),Prague,Czech Republic,May2004.
    6 Davis J, Bobick A. The representation and recognition ofhuman movement using temporal templates. IEEE Transactionson Pattern Analysis and Machine Intelligence,2001,23(3):257-267.
    7 孟繁锋,屈桢深,曾庆双,等.isMHI 结合CAMSHIFT 的多目标跟踪.计算机应用,2009,29(6):1673-1676.
    8 GARY RB, JAMES WD. Motion segmentation and poserecognition with motion history gradients. Machine Visionand Application,2002,13(7):174-184.
    9 Aggarwal J, Cai Q. Human motion analysis: a review. IEEEWkshp.Nonrigid and Articulated Motion Workshop,1997:90-102.
    10 王亮亮,孙即祥,谭志国.基于人脸检测和CamShift 算法的人脸跟踪系统.微计算机应用,2008,29(2):14-17.
    11 汤一平,叶永杰,朱艺华,顾校凯.智能全方位视觉传感器及其应用研究.传感技术学报,2007,20(6):1316-1320.
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

严杭晨,汤一平,马宝庆,田旭园,吴立娟.基于全方位视觉的多目标跟踪技术.计算机系统应用,2013,22(9):185-190

复制
分享
文章指标
  • 点击次数:1661
  • 下载次数: 3962
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2013-03-12
  • 最后修改日期:2013-04-07
  • 在线发布日期: 2013-10-10
文章二维码
您是第11280383位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号