基于Point-to-Plane ICP的点云与影像数据自动配准

Automatic Registration of Point Cloud and Image Data Based on Point-to-Plane ICP

针对三维激光点云与二维影像数据的融合问题,采用了一种基于Point-to-Plane ICP的配准方法;该方法仅采用一块普通的平面黑白棋盘格作为标定板,能同时完成单目相机的标定与三维激光扫描仪和相机的联合标定,进而实现三维点云数据与二维影像数据的配准;与以往基于控制点或者边缘对应的配准方法不同,该方法使用RANSAC算法自动提取场景中的标定平面,通过优化点到平面的距离来求取两组数据的变换。实验结果表明,该配准方法减少了人工的干预,并获得了很高的精度。

A Point-to-Plane ICP based registration method is used for the fusion of 3 D laser point cloud and 2 D image data.The method uses only an ordinary flat black and white checkerboard as the calibration board. The calibration of the monocular cam-era combined with the calibration of the 3 D laser scanner and the camera can be completed at the same time,so as to realize the reg-istration of the 3 D point cloud data and the 2 D image data. Unlike the traditional registration methods based on the correspondenceof control points or edges,RANSAC algorithm is adopted to automatically extracts the calibration plane in the scene and achievesthe fusion of the two sets of data by optimizing the distance from the point to the plane. The experimental results show that the pro-posed method can reduce the manual operation and obtain high precision.