A LiDAR-Camera Calibration and Sensor Fusion Method with Edge Effect Elimination

This article presents a multi-sensor calibration and fusion work for a dense reconstruction and colorization system. Unlike previous research, it explicitly addresses the edge effect in LiDAR measurements in both calibration and sensor fusion. The proposed calibration method can accomplish LiDAR-motor and LiDAR-camera calibration at a one-time experiment. The feature extraction algorithm for the point cloud automatically eliminates the edge-effect-induced noises in the calibration process. The experiment results show that it can achieve even better results than work without edge effect elimination in terms of calibration errors. As a result, it has achieved a calibration accuracy superior to previous researches. The system is testified by an indoor scene, which shows a desirable color point cloud generation with dense and clean data.