一种新型的运动目标识别与跟踪算法研究

交通运输系统工程与信息 ›› 2008, Vol. 8 ›› Issue (4): 83-88 .

一种新型的运动目标识别与跟踪算法研究

高韬;刘正光^*;张军

天津大学电气与自动化工程学院，天津，300072

收稿日期:2008-01-28 修回日期:2008-05-25 出版日期:2008-08-25 发布日期:2008-08-25
通讯作者: 刘正光
作者简介:高韬（1981-），男，河北省石家庄市人，博士生。
基金资助:
天津市公安交通局科研基金（公科【2005】06）。

New Method for Moving Object Recognition and Tracking

GAO Tao;LIU Zheng-guang;ZHANG Jun

The Electrical Engineering and Automation College of Tianjin University，Tianjin，300072,China

Received:2008-01-28 Revised:2008-05-25 Online:2008-08-25 Published:2008-08-25
Contact: LIU Zheng-guang

摘要/Abstract

摘要： 提出了一种应用于智能交通监控系统的运动目标识别和跟踪方法。针对帧间差分提取运动目标的缺陷与不足，提出了一种基于冗余小波变换的运动目标识别算法，即直接在冗余小波变换域提取运动区域，从而检测出运动目标。对于检测出来的运动目标，本文对mean-shift算法进行了改进，采用自适应mean-shift算法，对目标进行跟踪。实验结果表明，本文提出的算法可以有效地提取运动目标，即使目标与背景具有较高的相似度，也可以较准确的提取出前景运动信息，效果要好于传统的帧差法；跟踪目标准确度高，不受目标大小变化的影响。本算法具较高的实用价值和应用前景。

关键词: 交通监控, 目标识别, 目标跟踪

Abstract: This paper presents a new method for moving object identification and tracking in the intelligent traffic monitoring system. For the shortcomings and deficiencies of the frame-subtraction method, we put forward a redundant wavelet transform based on the moving object recognition algorithm, which directly detects moving objects in the redundant wavelet transform domain. Subsequently, we use an improved adaptive mean-shift algorithm to track the object. Experimental results show that the proposed algorithm can effectively extract the moving object, even though the object is similar to the background. The segmentation results are more effective than the traditional frame-subtraction method, and the object tracking is accurate without the impact of changes in the size of the object. Therefore, the proposed algorithm has high practical value and prospects.

Key words: traffic monitoring, object recognition, object tracking

中图分类号:

UP391

高韬,刘正光,张军. 一种新型的运动目标识别与跟踪算法研究[J]. 交通运输系统工程与信息, 2008, 8(4): 83-88 .

GAO Tao,LIU Zheng-guang,ZHANG Jun. New Method for Moving Object Recognition and Tracking[J]. Journal of Transportation Systems Engineering and Information Technology, 2008, 8(4): 83-88 .

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/abstract/abstract17690.shtml

http://www.tseit.org.cn/CN/Y2008/V8/I4/83

参考文献

[1] 顾广华,崔冬.全局运动序列的视频对象分割算法[J].仪器仪表学报,2007,28(1),128-131.[GU Guang-Hua, CUI Dong. Video object segmentation algorithm for sequences with global motion[J]. Chinese Journal of Scientific Instrument,2007,28(1),128-131.]

[2] 胡建华,徐健健.交通监控系统中车辆和行人的检测与识别[J].电子测量技术,2007,30(1),16-17.[HU Jian-hua, XU Jian-jian. Detection and recognition of vehicle and pedestrian in intelligent traffic surveillance[J]. Electronic Measurement Technology,2007,30(1),16-17.]

[3] Horprasert T., Harwood D., Davis L.S.. A statistical approach for real-time robust background subtraction and shadow detection[C]. IEEE International Conference on Computer Vision, Frame-Rate Applications Workshop, 1999.

[4] 徐晓夏, 陈泉林.基于视频的车辆检测中阈值分割算法的改进[J].信息技术,2005(9),10-11.[XU Xiao-xia,CHEN Quan-lin. An improved algorithm for threshold segmentation in video-based vehicle detection[J].Information Technology,2005(9),10-11.]

[5] 朱彤,袁志业,周殉.一种基于直方图的车辆检测方法[J].交通标准化,2006(11),184-186.[ZHU Tong, YUAN Zhi-ye, ZHOU Xun. A method for vehicle detection based on histogram[J]. Communications Standardization, 2006(1),184-186.]

[6] Thomas M.,.King N.N. Automatic segmentation of moving objects for video object plane generation[J]. IEEE Trans on Circuits and System for Video Technology, 1998, 8(5), 525-538.

[7] Wren C., et al. PFinder: real-time tracking of the human body[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 1997,19(7), 780-785.

[8] 高韬,于明.基于冗余小波变换的运动估计及其DSP实现[C].NCIG,2006,153-156.[GAO Tao,YU Ming. Motion estimation based on redundant wavelet transform and its DSP implementation[C]. NCIG,2006,153-156.]

[9] Daubechies I, Sweldens W. Factoring wavelet transforms into lifting steps[R]. Technical report, Bell Laboratories, Lucent Technologies, 1998.

[10] Calderbank A R, Daubechies I, Sweldens W. Yeo B L. Wavelet transforms that map integers to integers[R].Technical report, Department of Mathematics, Princeton University, 1996

[11] Otsu N..A threshold selection method from gray-level histogram[J].IEEE Trans.SMC.1978, 8:62-66.

[12] 高韬,于明.背景渐变的视频对象分割及其DSP实现[J].电视技术,2006(07),84-86.[GAOTao, YU Ming. Segmentation of the moving object in video sequence of background changes gradually and its dsp implement.Video Engineering.2006 (07),84-86.]

[13] Comaniciu D, Ramesh V, Meer P. Real-time tracking of non—rigid objects using mean shift[C]//.Proc of the IEEE Conference on Computer Vision and Pattern Recognition, Hilton Head Island, South Carolina, 2000, II, l42-l49.

[14] Comaniciu D, Ramesh V. Mean shift and optimal prediction for efficient object tracking[C]. Proc of the IEEE International Conference on Image Processing, Vancouver, Canada, 2000, III, 70-73.

[15] 彭宁嵩,杨杰,刘志,张风超.Mean-Shift跟踪算法中核函数窗宽的自动选取[J].软件学报,2005,16(9),1542-1550.[PENG Ning-Song, YANG Jie, LIU Zhi, ZHANG Feng-Chao. Automatic selection of kernel-bandwidth for mean-shift object tracking[J].Journal of Software,2005, 16(9),1542-1550.]