双目视觉的立体标定方法  

Stereo calibration method of binocular vision

作　　者：刘俸材[1] 谢明红[1] 王伟[1] 

机构地区：[1]华侨大学机电学院,福建泉州362021

出　　处：《计算机工程与设计》2011年第4期1508-1512,共5页Computer Engineering and Design

基　　金：泉州市科技基金项目(24200832)

摘　　要：为实现双目视觉系统的立体标定,分析了摄像机成像模型,并充分考虑了透镜的径向畸变和切向畸变,提出了一种新的立体标定算法。该算法利用张正友的灵活标定算法,初步求取摄像机的内参数,结合Brown算法并提取图像中角点的子像素级坐标,精确求取摄像机内参数和畸变向量。为方便后续的图像校正,基于前面的单个摄像机标定,通过计算空间中的景物点在左右摄像机成像平面上的位置关系,计算出双目视觉系统中两个摄像机之间的旋转矩阵R和平移向量T,从而实现了立体标定。实验结果表明,该算法能取得较高的精度,可以应用于双目视觉系统。In order to realize stereo calibration for binocular vision system,firstly,the image model of camera with a full consideration of radial distortion and tangential distortion of lens is analyzed,and a new method of stereo calibration is proposed.The flexible calibration method of Zhang Zhengyou is used to achieve the initial value of the camera＇s intrinsic parameters,and combined with the arithmetic of Brown,sub-pixel coordinates of images’corner point are extracted.Through this method,it calculates the intrinsic parameters and distortion vector accurately.Hereafter,based on the single camera calibration aforementioned and calculation of the position relationship of spatial point projected to the left and right camerasi’maging planes which helps to work out the rotation matrix R and translation vector T between the two cameras of the binocular stereo vision system,finally the stereo calibration is achieved.All experimental results show that this calculation method acquires higher precision,and can be applied to binocular system.

关 键 词：双目立体视觉 摄像机标定 摄像机模型 立体标定 角点提取 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]