The main sensors of self-driving cars are lidar and camera; this paper proposes a new fusion method. First, the image is de-distorted, and the noise points are filtered out by the KD-Tree algorithm. Finally, through coordinate transformation, the pixel RGB information of the image is assigned to the lidar point cloud, the fusion of lidar and image is realized. The kitti dataset is used in this experiment, use Yolov5 neural network for target detection and recognition, and the fusion effect is demonstrated by the ROS system; the results show that this method not only shortens the processing time by 62.36%, but also improves the sensor fusion accuracy by 18.89%.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2022) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: