基于双目视觉的运动物体实时跟踪与测距  

Real-Time Tracking and Measuring of Moving Objects Based on Binocular Vision

作　　者：祝琨[1] 杨唐文[1] 阮秋琦[1] 王红波[1] 韩建达[2] 

机构地区：[1]北京交通大学计算机与信息技术学院,北京100044 [2]中国科学院沈阳自动化研究所机器人学国家重点实验室,辽宁沈阳110016

出　　处：《机器人》2009年第4期327-334,共8页Robot

基　　金：北京交通大学科技基金资助项目(2007XM007);机器人学国家重点实验室基金资助项目(RLO200801)

摘　　要：利用双目视觉信息系统实现三维空间中运动物体实时跟踪与测距。当运动目标超出视野范围时,可通过控制摄像机云台转动搜索目标。此外,还研究了在摄像头运动情况下,无需重新标定,即可实现运动物体测距的算法。这里,自适应背景建模法与CamShift算法用于实现运动物体的辨识与跟踪。实验结果证明了所提出的算法能够有效地追踪物体,并同时准确地测量它的三维位置。A binocular camera information system is used to track and measure a moving object in real time in 3D space. As the moving object goes out of the field of view, pan＆tilt of cameras is turned to search the object. Furthermore, an arithmetic without recalibration for moving object distance measuring is studied, in the condition of camera moving. Herein adaptive background modelling algorithm and CamShift algorithm are used respectively to identify and track moving objects. Experimental results are presented to demonstrate that the proposed method can track the moving object effectively and get its 3D location accurately.

关 键 词：移动机器人 目标跟踪 三维测距 双目视觉系统 

分 类 号：TP319.41[自动化与计算机技术—计算机软件与理论]