<-上一篇/Previous Article 下一篇/Next Article->

[1]刘侠,陶冶,邢春.统计差分与自启动的Camshift跟踪算法[J].智能系统学报,2011,6(4):355-359.
　LIU Xia,TAO Ye,XING Chun.An objective tracking Camshift algorithm based onautomatic startup and the statistical differential method[J].CAAI Transactions on Intelligent Systems,2011,6(4):355-359.

点击复制

统计差分与自启动的Camshift跟踪算法

PDF下载 HTML

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 6 期数: 2011年第4期页码: 355-359 栏目: 学术论文—智能系统出版日期: 2011-08-25

Title:: An objective tracking Camshift algorithm based onautomatic startup and the statistical differential method

文章编号:: 1673-4785(2011)04-0355-05

作者:: 刘侠，陶冶，邢春; 哈尔滨理工大学自动化学院，黑龙江哈尔滨 150080

Author(s):: LIU Xia, TAO Ye, XING Chun; School of Automation, Harbin University of Science and Technology, Harbin 150080, China

关键词:: CamShift算法; 背景统计; 目标跟踪; 目标掩膜

Keywords:: CamShift algorithm; background statistics; object tracking; object masking

分类号:: TP391

文献标志码:: A

摘要:: 传统的CamShift算法具有很好的实时性，但无法实现自启动，且在跟踪过程中易受噪声和颜色变化的干扰，针对这些缺点，提出了一种改进的Camshift算法.首先运用统计差分算法对移动目标快速定位，得到初始前景对象的区域，然后通过统计前景区域获取精确的目标颜色直方图模型并以此区域为原型计算肤色直方图的查询表.同时结合一些后续步骤成功实现了算法的自启动.最后用实验验证了借助统计差分与目标颜色直方图模型的CamShift算法能够实现自启动并提高Camshift算法跟踪时的准确度.

Abstract:: The camshift tracking algorithm has the advantage of better realtime characteristics, but it is not able to start up automatically and is susceptible to noise and color change in the tracking process. In order to solve these problems, an improved Camshift algorithm was proposed. First, a statistical difference algorithm was applied for rapid localization of mobile targets, and the initial foreground object’s region was obtained. Then, an accurate target color histogram model was obtained by gathering foreground area statistics. The query table of the skin color histogram was also calculated taking this area as a prototype. Meanwhile, an autostart algorithm was successfully achieved with several subsequent steps. Finally, the algorithm was validated through experiments. 

参考文献/References:: ［1］杨振兴. 基于DM642的Camshift目标跟踪算法研究［D］. 南京：南京理工大学， 2010： 1523.
YANG Zhenxing. Research of Camshift tracking algorithm based on DM642［D］.Nanjing: Nanjing University of Science, 2010: 1523. 
［2］COMANICIU D, RAMESH V, MEER R. Realtime tracking of nonrigid objects using mean shift［C］//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Dublin, Ireland, 2000, 2: 142149.
［3］BRADSKI G R. Computer vision face tracking for use in aperceptual user interface［J］. Intel Technology Journal, 1998, 2: 214219.
［4］COMANICIU D, RAMESH V, MEER P. Kernel based object tracking［J］. IEEE Trans on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564575.
?［5］董蓓,谢勤岚,贾茜.基于CamShift算法的运动目标实时跟踪系统［J］.自动化技术， 2008, 19: 116118.
DONG Bei, XIE Qinlan, JIA Xi. Realtime tracking system of moving object based on Camshift algorithm［J］. Automation Technology, 2008, 19: 116118.
［6］刘雪,常发亮,王华杰.基于改进Camshift 算法的视频对象跟踪方法［J］. 微计算机信息, 2007(23): 297299.
LIU Xue,CHANG Faliang,WANG Huajie.Video object tracking method based on improved Camshift algorithm［J］. Computer Information, 2007(23): 297299.
［7］ LIU Qiang, CAI Canhui, KING N N, LI Hongliang. CamShift based realtime multiple faces match tracking［C］//Proceedings of 2007 International Symposium on Intelligent Signal Processing and Communication Systems. Hong Kong: Chinese Univ of Hong Kong， 2007: 726729.
［8］WANG Zhaowen, YANG Xiaokang, XU Yi, YU Songyu. CamShift guided particle filter for visual tracking［C］//IEEE Xplore, 2007: 301306.
［9］朱胜利.Mean Shift及相关算法在视频跟踪中的研究［D］.杭州：浙江大学， 2006: 4354.
?ZHU Shengli. Research of Mean Shift and related algorithms in the video tracking［D］.Hangzhou: Zhejiang University, 2006: 4354. 
［10］魏旭宾,王向军.基于累积差分图像和高斯背景统计模型的烟尘目标检测［J］.宇航计测技术， 2007（2）： 5961.
WEI Xubin,WANG Xiangjun. Smoke target detection based on the cumulative difference image and Gaussian background statistical model［J］. Metrology and Measurement, 2007(2): 5961.
［11］NOBUYUKI OTSU.A threshold selection method from gray level histogram［J］. IEEE Transactions on Systems, Man and Cybernetics, 2008, 9(1): 6265.
［12］吴勃，许文芳，陈虹丽.神经滑模控制在机器人轨迹跟踪中的应用［J］.电机与控制学报, 2009, 13(增1)： 99105.
WU Bo, XU Wenfang, CHEN Hongli. Application of neural networks slidingmode control in tracking control of robot manipulators［J］.Electric Machines and Control, 2009, 13(suppl 1)：99105.
［13］汪语哲, 史小平, 朱胤. 抑制闪烁噪声的SMMIUKF目标跟踪算法［J］. 哈尔滨理工大学学报, 2010, 15(5): 6975.
?WANG Yuzhe, SHI Xiaoping, ZHU Yin. A SMMIUKF tracking method against target glint［J］. Journal of Harbin University of Science and Technology, 2010, 15(5): 6975.
［14］蔡春波，张华军，张义顺，刘立君. 基于视觉传感器焊缝跟踪轨迹自示教方法研究［J］. 哈尔滨理工大学学报,2009, 14(2): 9597.
CAI Chunbo, ZHANG Huajun, ZHANG Yishun, LIU Lijun. Investigation on seam tacking path selfteaching method based on visual sensor［J］. Journal of Harbin University of Science and Technology, 2009, 14(2): 9597.
［15］张丽丽，王建民. 太阳位置光电模拟信号检测跟踪的实现［J］. 自动化技术与应用, 2010, 29(9): 4043.
ZHANG Lili，WANG Jianmin. Implementation of the sun’s position photoelectricity analog signal detecting and tracking［J］. Techniques of Automation and Applications, 2010, 29(9): 4043.［16］白一鸣，孟宪尧. 基于FPGA的太阳能自动跟踪控制系统［J］. 自动化技术与应用, 2009, 28(9): 2629. BAI Yiming， MENG Xianyao. A FPGAbased tracking system of the solar panel［J］. Techniques of Automation and Applications, 2009, 28(9): 2629.

备注/Memo

收稿日期：2010-07-18.
基金项目：黑龙江省自然科学基金资助项目(F2007-13).
通信作者：陶冶.E-mail:taoye8882@163.com.
作者简介：刘侠，男，1975年生，硕士生导师，主要研究方向为图像处理与模式识别.
陶冶，男，1986年生，硕士研究生，主要研究方向为人脸识别.
邢春，男，1986年生，硕士研究生，主要研究方向为生物特征加密与图像检索.

更新日期/Last Update: 2011-09-30

统计差分与自启动的Camshift跟踪算法 PDF下载HTML

备注/Memo

统计差分与自启动的Camshift跟踪算法

PDF下载 HTML