摘要
激光雷达和相机融合系统可感知环境的几何尺寸和颜色信息,在多个领域中得到了广泛应用。为了准确融合两种信息,提出了一种基于自然特征点的激光雷达和相机外部参数标定方法。首先,在激光雷达自校正的基础上,利用激光雷达数据的强度信息对点云以中心投影的方式生成灰度图。然后,通过尺度不变特征变换算法对投影生成的灰度图和相机图像进行特征点提取和匹配。最后,以同名特征点得到的信息建立标定数学模型,并进行数据优化,标定出三维激光雷达系统和相机系统的外部参数。实验结果表明,该方法计算的点云到图像像素点的重投影误差为2.3 pixel,验证了该位姿标定方法的有效性和准确性。
Lidar and camera fusion system can perceive the geometric size and color information of the environment,which has been widely used in many fields.In order to accurately fuse the two kinds of information,we propose a calibration method of external parameters of lidar and camera based on natural feature points.First,on the basis of lidar self correction,the gray image is generated by central projection of the point cloud using the intensity information of lidar data.Then,the scale-invariant feature transformation algorithm is used to extract and match the feature points of the gray image generated by projection and the camera image.Finally,the calibration mathematical model is established based on the information obtained from the feature points with the same name,and the data are optimized to calibrate the external parameters of the three-dimensional lidar system and camera system.The experimental results show that the re projection error from the point cloud to the image pixels calculated by this method is 2.3 pixel,which verifies the effectiveness and accuracy of the pose calibration method.
作者
宋文松
张宗华
高楠
孟召宗
Song Wensong;Zhang Zonghua;Gao Nan;Meng Zhaozong(School of Mechanical Engineering,Hebei University of Technology,Tianjin 300130,China)
出处
《激光与光电子学进展》
CSCD
北大核心
2022年第2期348-356,共9页
Laser & Optoelectronics Progress
基金
国家自然科学基金(51675160,52075147)
重大科学仪器设备开发重点专项(2017YFF0106404)。
关键词
机器视觉
激光雷达
强度信息
相机
外参标定
尺度不变特征变换
machine vision
lidar
intensity information
camera
external parameter calibration
scale-invariant feature transform
