盲环境移动定位(Ⅱ):基于单目视觉系统的定位技术  

Mobile Positioning in Blind Environments(Ⅱ):Positioning Moving Targets Based on a Monocular Vision System

作　　者：吴立新[1] 王植[1,2] 孔祥勤[1] 

机构地区：[1]东北大学资源与土木工程学院,辽宁沈阳110819 [2]北京师范大学减灾与应急管理研究院,北京100875

出　　处：《东北大学学报（自然科学版）》2011年第5期736-739,744,共5页Journal of Northeastern University(Natural Science)

基　　金：国家杰出青年基金资助项目(50525414);国家自然科学基金资助项目(40901220);霍英东青年教师基金资助项目(122025);中央高校基本科研业务费专项资金资助项目(N100401009);河南理工大学矿山空间信息技术国家测绘局重点实验室开放基金项目(KLM200913)

摘　　要：针对盲环境中移动目标的定位需要,提出了一种基于序列编码图形路标和单目机器视觉系统的盲环境移动目标定位算法.内容包括:在序列编码图形路标模型的基础上,建立包含透镜畸变分析的摄像机针孔成像模型,在移动目标上安装经过标定的单目相机,实时采集包含编码图形路标的图像;根据解码得到的编码标志牌中心点的三维坐标,利用POSIT算法解算出相机的位姿参数,实现盲环境下移动目标的精确定位定姿.实验结果和精度分析表明,该研究为盲环境中移动目标快速精确的自主定位提供了一种可行的新方法.This paper proposes a mobile positioning algorithm for moving targets in a blind environment without satellite navigation signals and based on coded sequence pattern signposts and a monocular vision system.The coded sequence pattern signposts were fixed along the blind environment.The 3D coordinates of the center of each coded pattern identified a signpost.A pinhole imaging model was then established using a camera equipped with through-the-lens distortion analysis.A calibrated monocular vision system was mounted on the target object to capture images with coded patterns.Image processing methods were used to interpret the coded sequence pattern signposts,and to inquire about the coordinates of the center of each signpost according to its code pattern.The translation and rotation parameters of the monocular vision system were computed by using a POSIT algorithm so as to compute the pose of a moving target.Experiment results indicated that the proposed method delivers fast and accurate pose estimation for moving targets in a blind environment without GNSS signals.

关 键 词：盲环境 移动定位 序列编码图形 单目视觉 POSIT算法 

分 类 号：TG391.4[金属学及工艺—金属压力加工]