loading page

LiDAR--camera calibration based on back-projection of background point cloud
  • +2
  • Chengkai Chen,
  • Jinhui Lan,
  • Haoting Liu,
  • Shuai Chen,
  • Xiaohan Wang
Chengkai Chen
University of Science and Technology Beijing

Corresponding Author:[email protected]

Author Profile
Jinhui Lan
University of Science and Technology Beijing
Author Profile
Haoting Liu
University of Science and Technology Beijing
Author Profile
Shuai Chen
University of Science and Technology Beijing Shunde Graduate School
Author Profile
Xiaohan Wang
University of Science and Technology Beijing
Author Profile

Abstract

Calibrating the extrinsic parameters between LiDAR and the visible light camera is a challenging task, because accurate 3D feature points in sparse LiDAR point clouds are not easy to extract. In this paper, we propose a calibration method for estimating the extrinsic parameters of the LiDAR-camera system by considering the background point clouds. In this method, additional 3D points behind the calibration object are introduced to refine the 3D edges and feature points. We process each scanning line of LiDAR and back-projection points in turn to get accurate edge points and 3D feature poitns by a refining scheme, thus leading calibration results robust to sensor noise and distance variations. Our method has good effect for the LiDAR with poor horizontal angle resolution, which means our method performance on the LiDAR-camera system is more applicable for computer vision perception.