基于16线激光雷达和相机融合的三维目标检测

3D target detection based on fusion of 16-line lidar and camera

针对16线激光雷达点云过于稀疏而无法进行三维目标检测的问题,设计了一种基于激光雷达和相机融合的三维目标检测方法。首先,对激光雷达和相机进行联合标定,统一两者坐标系,并利用ROS进行时间同步,确保雷达和相机在相同的时间感知到相同的环境。然后,将相机采集的图像作为输入,送入YOLOv5算法中进行二维的目标检测,得到目标在图像中的二维检测框和类别。最后,利用联合标定过的激光雷达获取目标二维像素点对应的三维点云,通过这些雷达点云算出目标的三维坐标和三维边界框,从而得到三维目标检测结果。实验验证:通过搭建硬件平台并在实验场地测试后证明了该方法的可行性。

Aiming at the problem that the 16-line lidar point cloud is too sparse for 3D target detection,a 3D target detection method based on the fusion of lidar and camera is designed.First,the lidar and camera are jointly calibrated,the coordinate systems of the two are unified,and ROS is used for time synchronization to ensure that the lidar and the camera perceive the same environment at the same time.Then,the image collected by the camera is used as input and sent to the YOLOv5 algorithm for 2D target detection,and the 2D detection frame and category of the target in the image are obtained.Finally,the jointly calibrated lidar is used to obtain the 3D point cloud corresponding to the 2D pixel points of the target,and the 3D coordinates and 3D bounding box of the target are calculated through these lidar points,so as to obtain the 3D target detection result.Experimental verification is:The feasibility of the method is proved by building a hardware platform and testing it in the experimental site.