Obstacle information detection based on fusion of 3D LADAR and camera

Obstacle information detection is a prerequisite for safety and stability driving of unmanned platforms in the unknown environments. Joint detection of the obstacle using the 3D laser radar (3D LADAR) and camera is proposed in this paper. Firstly, joint calibration between 3D LADAR and camera based on the planar feature method is done. Secondly, laser point cloud is clustered to identify the number of obstacles, and then detected image is divided into the corresponding small regions according to the number of clusters. Finally, we use the image processing technology to detect the obstacle areas. The experiments in the real world are implemented to show that the proposed approach can effectively detect the size of the obstacles.