Straight line detection algorithm of truck point cloud based on statistical line chart
Ming Wei, Jun‐Guo Lu, Siyou Luo, Qing‐Hao Zhang, Lunkai Zhang · 2022 34th Chinese Control and Decision Conference (CCDC) · 2022
Line feature extraction has always been an important part of point cloud processing. In the task of cement automatic loading, the line feature of truck side point cloud plays an important role in model reconstruction and dimension calculation. This paper introduces a line detection algorithm for truck side edge based on statistical line graph. Firstly, we correct the horizontal posture of the truck based on the geometric characteristics of the truck. The straight lines of front board, rear gate and side gates of the truck after correction are approximately parallel to a coordinate axis. Then, using the edge line detection algorithm of statistical broken line graph proposed in this paper, we draw the coordinate value-density distribution diagram, so as to determine the edge position of the truck side. and finally, calculate the edge straight line of the truck side point cloud. The algorithm proposed in this paper can effectively eliminate the interference of foreign bodies on the truck side and accurately extract the truck side point cloud. The simulation and real experiments verify the robustness and accuracy of the algorithm.